posts: 37823
Data license: CC-BY
This data as json
id | title | slug | type | status | content | archieml | archieml_update_statistics | published_at | updated_at | gdocSuccessorId | authors | excerpt | created_at_in_wordpress | updated_at_in_wordpress | featured_image | formattingOptions | markdown | wpApiSnapshot |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
37823 | Time Use | time-use | page | publish | <!-- wp:paragraph --> <p>Time is the ultimate limited resource. Every single one of us has exactly the same “time budget”: 24 hours per day, 365 days per year – 8,760 hours <a href="https://owid.cloud/life-expectancy" data-type="page" data-id="1670">every year of our lives</a>.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>How do we actually spend our time? There are many commonalities across the world: we all sleep, work, eat, and enjoy leisure time. But there are also important differences in the freedom people have to spend time on the things they value most. Studying how people across the world spend their time provides an important perspective for understanding living conditions, economic opportunities, and general well-being.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Here we present the data on time use. We explore how it differs across countries and over time and how these differences matter for people’s lives.</p> <!-- /wp:paragraph --> <!-- wp:owid/summary --> <!-- wp:list --> <ul><li><a href="https://ourworldindata.org/time-use#how-do-people-across-the-world-spend-their-time-and-what-does-this-tell-us-about-living-conditions" data-type="URL" data-id="https://ourworldindata.org/time-use#how-do-people-across-the-world-spend-their-time-and-what-does-this-tell-us-about-living-conditions">How people spend their time is similar in many ways across countries: we all sleep, work, eat, and enjoy leisure. But there are also important differences.</a></li><li><a href="https://ourworldindata.org/time-use#going-beyond-averages-the-gender-gap-in-leisure-time" data-type="URL" data-id="https://ourworldindata.org/time-use#going-beyond-averages-the-gender-gap-in-leisure-time">There is a gender gap in leisure time: in all countries surveyed the average leisure time for men is higher than for women.</a></li><li><a href="https://ourworldindata.org/time-use#who-do-we-spend-time-with-across-our-lifetime" data-type="URL" data-id="https://ourworldindata.org/time-use#who-do-we-spend-time-with-across-our-lifetime">Who we spend time with evolves across our lifetimes. As we enter older age we spend an increasing amount of time alone.</a></li><li><a href="https://ourworldindata.org/time-use#are-parents-spending-less-time-with-their-kids" data-type="URL" data-id="https://ourworldindata.org/time-use#are-parents-spending-less-time-with-their-kids">The amount of time parents spend with their kids has been increasing.</a></li><li><a href="https://ourworldindata.org/time-use#are-we-working-more-than-ever" data-type="URL" data-id="https://ourworldindata.org/time-use#are-we-working-more-than-ever">Working hours have decreased dramatically in the last 150 years for workers in many countries.</a></li><li><a href="https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours" data-type="URL" data-id="https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours">But there are still large differences between countries: workers in poorer countries tend to work much more than workers in richer countries.</a></li><li><a href="https://ourworldindata.org/time-use#how-are-working-hours-measured-and-what-can-we-learn-from-the-data" data-type="URL" data-id="https://ourworldindata.org/time-use#how-are-working-hours-measured-and-what-can-we-learn-from-the-data">The primary way to measure working hours is with surveys, but the data can have limitations that are important to understand.</a></li></ul> <!-- /wp:list --> <!-- /wp:owid/summary --> <!-- wp:heading --> <h2>Time use from the perspective of an average day</h2> <!-- /wp:heading --> <!-- wp:heading {"level":3} --> <h3>How do people across the world spend their time and what does this tell us about living conditions?</h3> <!-- /wp:heading --> <!-- wp-block-tombstone 37838 --> <!-- wp:paragraph --> <p>Sleep, work, eat, leisure – at a high level most of us spend time on similar activities. But just how similar are the daily activities of people across the world? This is something worth considering, not just to serve our curiosity but because differences in the way we spend time give us meaningful perspectives on living conditions, economic opportunities and general well-being.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Here we take a look at the data. We explore some of the key patterns that emerge from cross-country time use surveys, and then dig deeper to understand how these differences matter for people’s well-being.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>Daily activities: similarities and differences across countries</h4> <!-- /wp:heading --> <!-- wp:columns --> <div class="wp-block-columns"><!-- wp:column --> <div class="wp-block-column"><!-- wp:paragraph --> <p>In the chart here we compare average time spent across a number of common activities. The data comes from the OECD and brings together estimates from time diaries where respondents are asked to record the sequence of what they did over a specific day, as well as from general questionnaires where respondents are asked to recall the amount of time spent on different activities on a specific day in the previous week.{ref}The ‘time-diary method’ is generally more reliable and allows a richer analysis of routines, because it measures not only aggregate times but also sequences and clock-times. Time-diary data is less common, but it is available for some countries from the <a rel="noreferrer noopener" href="https://www.timeuse.org/mtus" target="_blank">Multinational Time Use Study</a>. We explore time-use ‘tempograms’ from the MTUS in a forthcoming companion post.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The first thing that jumps out from this chart is that there are indeed many similarities across countries.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>This is not surprising – most of us try to split our days into “work, rest and fun”, and so there are some predictable patterns. We spend the most time working and sleeping; and paid work, housework, leisure, eating and sleeping take together 80-90% of the 1440 minutes that we all have available every day.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>But if we look closely, we also see some important differences. Consider sleeping, for example. From this sample of countries, South Koreans sleep the least – averaging 7 hours and 51 minutes of sleep every day. In India and the US, at the other end of the spectrum, people sleep an hour more on average.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Work is another important activity where we see large differences. Countries are sorted by paid work hours in the chart – from highest to lowest. In China and Mexico people spend, on an average day, almost twice as much time on paid work as people in Italy and France do. This is a general pattern: <a rel="noreferrer noopener" href="https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours" data-type="URL" data-id="https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours" target="_blank">People in richer countries can afford to work less</a>. Keep in mind that this chart shows the average for all people in the working age bracket, from 15 to 64 years, whether they are actually employed or not.{ref}Because these estimates include people who are not employed they are much lower than the estimates of working hours <em>per</em> <em>worker</em> <a rel="noreferrer noopener" target="_blank" href="https://ourworldindata.org/time-use#time-spent-working">we present elsewhere</a>. The estimates also differ because of differences in the sources: time-use surveys compared to labor force surveys and national accounts data.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Differences in <a rel="noreferrer noopener" href="https://ourworldindata.org/age-structure" target="_blank">demographics</a>, <a rel="noreferrer noopener" href="https://ourworldindata.org/global-education" target="_blank">education</a> and <a rel="noreferrer noopener" href="https://ourworldindata.org/economic-growth" target="_blank">economic prosperity</a> all contribute to these inequalities in work and time use. But what’s clear in the chart here is that there are also some differences in time use that are not well explained by economic or demographic differences. In the UK, for example, people spend more time working than in France; but in both countries people report spending a similar amount of time on leisure activities.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Cultural differences are likely to play a role here. The French seem to spend much more time eating than the British – and in this respect the data actually goes in line with stereotypes about food culture. People in France, Greece, Italy and Spain report spending more time eating than people in most other European countries. The country where people spend the least time eating and drinking is the USA (63 minutes).</p> <!-- /wp:paragraph --></div> <!-- /wp:column --> <!-- wp:column --> <div class="wp-block-column"><!-- wp:heading {"level":6} --> <h6>How do people spend their time?{ref}OECD (2020) <a rel="noreferrer noopener" href="https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE" target="_blank">Time Use Database.</a>{/ref}</h6> <!-- /wp:heading --> <!-- wp:image {"id":37744,"width":764,"height":574,"sizeSlug":"large"} --> <figure class="wp-block-image size-large is-resized"><img src="https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD-732x550.png" alt="" class="wp-image-37744" width="764" height="574"/></figure> <!-- /wp:image --> <!-- wp:paragraph --> <p><a href="https://owid.cloud/app/uploads/2020/12/Time-Use-in-OECD-Countries-OECD.xlsx" data-type="URL" data-id="https://owid.cloud/app/uploads/2020/12/Time-Use-in-OECD-Countries-OECD.xlsx">Download the underlying data for this chart (.xlsx)</a></p> <!-- /wp:paragraph --></div> <!-- /wp:column --></div> <!-- /wp:columns --> <!-- wp:heading {"level":4} --> <h4>Going beyond averages: The gender gap in leisure time</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>Going beyond national averages reveals important within-country inequalities. The gender gap in leisure time, for example, is a key dimension along which large inequalities exist.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The chart here relies on the same time-use data described above, but shows total leisure time for men and women separately. Time for men is shown on the horizontal axis, while time for women appears on the vertical axis. The dotted diagonal line denotes ‘gender parity’, so the further away a country is from the diagonal line, the larger the difference between men and women.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>As we can see, in all countries the average leisure time for men is higher than for women – all bubbles are below the diagonal line – but in some countries the gaps are much larger. In Norway the difference is very small, while in Portugal men report almost 50% more leisure time than women.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>A key factor driving these differences in leisure time is the gender gap in unpaid work. As we explain in detail in a <a rel="noreferrer noopener" href="https://ourworldindata.org/female-labor-supply#informal-work-unpaid-care-work" target="_blank">companion post</a>, women are responsible for a disproportionate amount of unpaid work, and have less leisure as a result.{ref}If you want to dig deeper you can explore gender differences across all other activities directly from our source, via the <a rel="noreferrer noopener" href="https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE" target="_blank">OECD Data Portal</a>. And you can read more about within-country inequalities in time use along other dimensions, such as income and education, in <a rel="noreferrer noopener" href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=The+Middle+Class+Time+Squeeze%2C+Economic+Studies+at+Brookings%2C+Sawhill+and+Guyot+%282020%29&btnG=" target="_blank">this Brookings Paper</a>, where the authors focus on the ‘middle class time squeeze’ in the US. See: Sawhill, I. V., & Guyot, K. (2020). The Middle Class Time Squeeze. Economic Studies at Brookings. Brookings Institution.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/minutes-spent-on-leisure?tab=chart&stackMode=absolute&time=latest&country=&region=World" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>Why should we care about differences in time use?</h4> <!-- /wp:heading --> <!-- wp:columns --> <div class="wp-block-columns"><!-- wp:column --> <div class="wp-block-column"><!-- wp:paragraph --> <p>Every single one of us has the same “time budget”: 24 hours per day and 365 days per year. But of course not all of us can choose to spend time on the activities that we enjoy most. Differences in our freedom to allocate time to the things we enjoy is the main reason why time-use data is important for studying living conditions.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In the UK, researchers from the <a rel="noreferrer noopener" href="https://www.timeuse.org/" target="_blank">Centre for Time Use Research</a> linked time-use diaries with the respondents’ assessments of enjoyment, on a scale from 1 to 7, to better understand the connection between time use and well-being. The chart here, which we’ve adapted from the book ‘What We Really Do All Day’<em>, </em>by professors Jonathan Gershuny and Oriel Sullivan, shows the results. The estimates correspond to average reported levels of enjoyment for each activity, with confidence intervals.{ref}The underlying data comes from time-use diaries where respondents are asked to record the sequence of what they do over a specific day, and how much they enjoy each ‘episode’ (i.e. what they do) on a scale from 1 to 7. All episodes reported are then coded and grouped into similar activities. To arrive at the mean enjoyment scores, the authors multiply the duration of each episode where the activity category concerned is the primary activity recorded, by the enjoyment level to arrive at the total enjoyment score for that episode. Then they sum these total enjoyment scores for each category of activity across the day, and finally divide these daily enjoyment total scores for each activity by the amount of time devoted to the activity. In this way, they arrive at an appropriately weighted mean enjoyment level for each activity across all those who engage in it. For more details see Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>We see that the most enjoyed activities involve rest or leisure activities such as eating out, sleeping, going to sports events, playing computer games or attending cultural performances. The activities receiving the lowest ratings include doing school homework, looking for a job, or doing housework.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The activity where people show the greatest variation in enjoyment is working a “Second Job”. This likely reflects the difference between people who work a second job because they want to, and those who work a second job because they <em>have</em> <em>to</em>.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>So what do we learn from this?</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>First, we learn that the enjoyment of activities is, at least to some degree, predictable and stable. This means we can take activity groups and make meaningful comparisons across groups of people. Economists, for example, <a rel="noreferrer noopener" href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Ramey%2C+V.+A.%2C+%26+Francis%2C+N.+%282009%29.+A+century+of+work+and+leisure.+American+Economic+Journal%3A+Macroeconomics%2C+1%282%29%2C+189-224.&btnG=" target="_blank">often classify</a> any activity with an enjoyment level below work as a “non-leisure activity”, in order to measure trends in leisure across people and time.{ref}You find a very clear and complete explanation of this in Ramey, V. A., & Francis, N. (2009). A century of work and leisure. American Economic Journal: Macroeconomics, 1(2), 189-224.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>But beyond this, and more importantly, this confirms that time-use is informative about well-being.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The fact that there is a very clear and predictable pattern in the enjoyment of activities suggests that differences in time use do, indeed, give us meaningful perspectives on living conditions and economic opportunities. In countries where people do more paid and unpaid work, and have less time for leisure, their enjoyment – and <a rel="noreferrer noopener" href="https://ourworldindata.org/happiness-and-life-satisfaction" target="_blank">happiness and life satisfaction</a> – levels are likely to be lower.</p> <!-- /wp:paragraph --></div> <!-- /wp:column --> <!-- wp:column --> <div class="wp-block-column"><!-- wp:heading {"level":6} --> <h6>How do people rate the enjoyment of different activities?{ref}Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref}</h6> <!-- /wp:heading --> <!-- wp:image {"id":37746,"width":615,"height":930,"sizeSlug":"large"} --> <figure class="wp-block-image size-large is-resized"><img src="https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-364x550.png" alt="" class="wp-image-37746" width="615" height="930"/></figure> <!-- /wp:image --></div> <!-- /wp:column --></div> <!-- /wp:columns --> <!-- wp:paragraph --> <p></p> <!-- /wp:paragraph --> <!-- wp:heading --> <h2>Time use from the perspective of the life cycle</h2> <!-- /wp:heading --> <!-- wp:heading {"level":3} --> <h3>Who do we spend time with across our lifetime?</h3> <!-- /wp:heading --> <!-- wp-block-tombstone 37832 --> <!-- wp:paragraph --> <p>As we go through life we build personal relationships with different people – family, friends, coworkers, partners. These relationships, which are <a href="https://ourworldindata.org/social-connections-and-loneliness#social-connections" target="_blank" rel="noreferrer noopener">deeply important to all of us</a>, evolve with time. As we grow older we build new relationships while others transform or fade, and towards the end of life many of us spend a lot of time alone.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Taking the big picture over the entire life course: Who do we actually spend our time with?</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>From adolescence to old age: who do we spend our time with?</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>To understand how social connections evolve throughout our lives we can look at survey data on <em>how much</em> time people spend with others, and <em>who</em> that time is spent with.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The chart here shows the amount of time that people in the US report spending in the company of others, based on their age. The data comes from time-use surveys, where people are asked to list all the activities that they perform over a full day, and the people who were there during each activity. We currently only have data with this granularity for the US – time-use surveys are common across many countries, but what is special about the US is that respondents of the American Time Use Survey are asked to list everyone who was present for each activity.</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/time-spent-with-relationships-by-age-us?tab=chart&stackMode=absolute&region=World" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:paragraph --> <p>The numbers in this chart are based on averages for a cross-section of the American society – people are only interviewed once, but we have brought together a decade of surveys, tabulating the average amount of time that survey respondents of different ages report spending with other people<em>.</em>{ref}When interpreting this chart it’s important to bear in mind that the relationships used to categorise people are not exhaustive (i.e., survey respondents could also list being with people who didn’t fit any of the listed categories, or for whom a relationship was unclear or unknown – we do not count these instances in the estimates). Additionally, time spent with multiple people can be counted more than once; so attending a party with friends and your spouse, for example, would show up for both “friends” and “partner” in our estimates. The implication is that companion categories cannot be stacked to add up total time spent in the company of others.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":5} --> <h5>Who we spend our time with changes a lot over the course of life</h5> <!-- /wp:heading --> <!-- wp:paragraph --> <p>When we’re young – particularly in our teens – we spend a lot of our time with friends, parents, siblings and extended family. As we enter our 20s, time with friends, siblings and parents starts to drop off quickly. Instead, we start spending an increasing amount of time with partners and children. (The chart shows an average across Americans, so for those that have children the time spent with children is even higher, since the average is pulled down by those without children.)</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>As the chart shows, this continues throughout our 30s, 40s and 50s – over this period of their life, Americans spend much of their time with partners, children and, unsurprisingly, co-workers.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>For those 60 and older, we see a significant drop-off in time spent with co-workers. This makes sense, considering many people in the US <a href="https://ourworldindata.org/grapher/average-effective-retirement-men" data-type="URL" data-id="https://ourworldindata.org/grapher/average-effective-retirement-men" target="_blank" rel="noreferrer noopener">enter retirement in their mid 60s</a>. We see that this time is partly displaced by more time with partners.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In terms of the diversity of interactions, this chart suggests that the number of people with whom we interact is highest around 40, but then things change substantially after that. And this is perhaps the most conspicuous trend in the chart: above 40, people spend an increasing amount of time alone.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>Older people spend a lot of time alone</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>Older people spend a large amount of time alone and it is understandable why – time spent alone increases with age because this is when health typically deteriorates and people lose relatives and friends.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Indeed, many people who are older than 60 live alone as this chart shows clearly: living alone is particularly common for older adults. Today nearly 4 out of every 10 Americans who are older than 89 years old live alone.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Another interesting point here is that the share of people across all age groups who live alone has been increasing over time. This is part of a more general global trend – if you want to read more about the global ‘rise of living alone’, we provide a detailed account of this trend across countries in a <a rel="noreferrer noopener" href="http://ourworldindata.org/living-alone" target="_blank">companion post</a>.</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/percentage-of-americans-living-alone-by-age" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>Older people spend more time alone, but this doesn’t necessarily mean they’re lonely</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The data shows that as we become older we tend to spend more and more time alone. What’s more, the data also shows that older people today spend more time alone than older people did in the past.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>We might think older people are therefore more lonely – but this is not necessarily the case.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Spending time alone is not the same as feeling lonely. This is a point that is well recognised by researchers, and one which has been confirmed empirically across countries. Surveys that ask people about living arrangements, time use, and feelings of loneliness find that solitude, by itself, is not a good predictor of loneliness. You can read our overview of the evidence in <a rel="noreferrer noopener" href="http://ourworldindata.org/lonely-not-alone" target="_blank">this post</a>.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>So, what about loneliness? If we focus on self-reported loneliness, there is <a rel="noreferrer noopener" href="https://ourworldindata.org/loneliness-epidemic#loneliness-trends-over-time-are-people-lonelier-today-than-in-the-past" target="_blank">little evidence</a> of an upward trend over time in the US; and importantly, it’s <a rel="noreferrer noopener" href="https://ourworldindata.org/social-connections-and-loneliness#do-we-become-lonelier-as-we-get-older" target="_blank">not the case</a> that loneliness keeps going up as we become older. In fact a <a rel="noreferrer noopener" href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Are+U.S.+older+adults+getting+lonelier%3F%3A+Age%2C+period%2C+and+cohort+differences.+Psychology+and+Aging.&btnG=" target="_blank">recent study</a> based on surveys that track the same individuals over time found that after age 50 – which is the earliest age of participants in the analysis – loneliness tended to decrease, until about 75, after which it began to increase again.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Taking the evidence together, the message is not that we should be sad about the prospect of aging, but rather that we should recognise the fact that social connections are complex. We often tend to look at the amount of time spent with others as a marker of social well-being; but the quality of time spent with others, and our expectations, matter even more for our feelings of connection and loneliness.</p> <!-- /wp:paragraph --> <!-- wp:owid/additional-information --> <!-- wp:heading {"level":3} --> <h3><strong>A note about the interpretation of age trends</strong></h3> <!-- /wp:heading --> <!-- wp:columns {"className":"is-style-sticky-right"} --> <div class="wp-block-columns is-style-sticky-right"><!-- wp:column --> <div class="wp-block-column"><!-- wp:paragraph --> <p>In the chart where we plot the amount of time that people in the US report spending in the company of others, it’s important to keep in mind that we are taking a look at a cross-section of society. This means that we are actually seeing the result of two underlying trends.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>On one hand, we see the effect of aging on social connections (we relate to different people and reallocate time as we go through different stages of life), but we also see the effect of cohort trends (compared to people in the past, today’s older generations in the US tend to be <a rel="noreferrer noopener" href="https://ourworldindata.org/grapher/life-expectancy?tab=chart&country=JPN~GBR~IND~ETH~OWID_WRL~KOR~ZAF~USA&region=World" target="_blank">healthier</a> and <a href="https://ourworldindata.org/grapher/average-real-gdp-per-capita-across-countries-and-regions?tab=chart&country=Western%20Offshoots~Western%20Europe~Western%20Asia~Eastern%20Europe~OWID_WRL~Latin%20America~East%20Asia~Africa~USA">richer</a>, and might also have different expectations, preferences and opportunities).</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Disentangling these two effects is difficult, so it is important to keep in mind that at least some of the age gradients we observe might be partly explained by cohort changes, rather than life-cycle trends. </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>This is why it’s important to rely not only on cross-sectional data, but also on surveys that track the same individuals over time.</p> <!-- /wp:paragraph --></div> <!-- /wp:column --> <!-- wp:column --> <div class="wp-block-column"><!-- wp:html /--></div> <!-- /wp:column --></div> <!-- /wp:columns --> <!-- /wp:owid/additional-information --> <!-- wp:paragraph --> <p></p> <!-- /wp:paragraph --> <!-- wp:heading --> <h2>Parents’ time with children</h2> <!-- /wp:heading --> <!-- wp:heading {"level":3} --> <h3>Are parents spending less time with their kids?</h3> <!-- /wp:heading --> <!-- wp-block-tombstone 37843 --> <!-- wp:paragraph --> <p>Over the last 50 years many countries have seen large changes in family structures and the institution of marriage. These changes – which include a <a href="https://ourworldindata.org/marriages-and-divorces#single-parenting-is-common-and-in-many-countries-it-has-increased-in-recent-decades" target="_blank" rel="noreferrer noopener">rise in single-parenting</a> and a large increase in the <a href="https://ourworldindata.org/female-labor-supply#female-participation-in-labor-markets-grew-remarkably-in-the-20th-century" target="_blank" rel="noreferrer noopener">share of women working outside the home</a> – have made some people worry that children might be getting ‘short-changed’, because parents are not spending as much time with them.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In 1999, for example, a <a href="https://clintonwhitehouse4.archives.gov/media/pdf/famfinal.pdf" target="_blank" rel="noreferrer noopener">report</a> from the Council of Economic Advisors in the US analyzed trends over the second half of the 20th century and concluded: <em>“The increase in hours mothers spend in paid work, combined with the shift toward single-parent families, resulted in families on average experiencing a decrease of 22 hours a week (14 percent) in parental time available outside of paid work that they could spend with their children.”</em></p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The line of thought behind these concerns is that changes to the structure of families and work have meant that children spend less time with parents, because parents – particularly mothers – spend less time at home.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Here we review the evidence and show that this reasoning is flawed. As we explain, in the US and many other rich countries parents spend more time with their kids today than 50 years ago. Equating ‘mother time at home' with 'children’s time with parents’ is a huge and unhelpful oversimplification.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>Mothers and fathers spend an increasing amount of time with kids: evidence from time-use surveys</h4> <!-- /wp:heading --> <!-- wp:columns --> <div class="wp-block-columns"><!-- wp:column --> <div class="wp-block-column"><!-- wp:paragraph --> <p>The chart here shows the time that mothers and fathers spend with their children. This is measured using time-use diaries where parents record the amount of time that they spend on various activities, including child care.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>These estimates, which are sourced from a <a rel="noreferrer noopener" href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=" target="_blank">paper</a> published in 2016 by sociologists Giulia Dotti Sani and Judith Treas, are disaggregated by education levels and are adjusted to account for demographic differences between countries. We explain <a href="https://ourworldindata.org/parents-time-with-kids#digging-deeper-raw-vs-population-adjusted-trends-in-time-use" data-type="URL" data-id="https://ourworldindata.org/parents-time-with-kids#digging-deeper-raw-vs-population-adjusted-trends-in-time-use">below</a> in more detail why these adjustments are important.{ref}The full reference for the source paper is: Dotti Sani, G. M., & Treas, J. (2016). <a rel="noreferrer noopener" href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=" target="_blank">Educational gradients in parents' child‐care time across countries</a>, 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096. {/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>As we can see here, there has been a clear increase in the amount of time parents spend with their children over the last 50 years. This is true for both fathers and mothers, and holds across almost all education groups and countries. The two exceptions are France, where mothers’ time has declined (from a very comparatively high level); and Slovenia where it has remained roughly constant among non-university-educated parents.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In terms of within-country inequalities we also see two clear patterns. First, in all countries mothers spend more time in child care activities than fathers. The differences are large and persistent across education groups. In some countries, such as Canada, France and the US, this gender gap has shrunk; while in other countries such as Denmark and Spain, the gap has widened.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Second, in all countries there is a positive ‘educational gradient’, meaning more educated parents tend to spend more time with their children. In many countries this gradient increased, and nowhere did it decline.</p> <!-- /wp:paragraph --></div> <!-- /wp:column --> <!-- wp:column --> <div class="wp-block-column"><!-- wp:heading {"level":6} --> <h6>How much time do parents spend with their children per day?{ref}Dotti Sani, G. M., & Treas, J. (2016). <a rel="noreferrer noopener" href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=" target="_blank">Educational gradients in parents' child‐care time across countries</a>, 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref}</h6> <!-- /wp:heading --> <!-- wp:image {"id":37745,"width":747,"height":600,"sizeSlug":"large"} --> <figure class="wp-block-image size-large is-resized"><img src="https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2-685x550.png" alt="" class="wp-image-37745" width="747" height="600"/></figure> <!-- /wp:image --></div> <!-- /wp:column --></div> <!-- /wp:columns --> <!-- wp:owid/additional-information --> <!-- wp:heading {"level":3} --> <h3>Digging Deeper: Raw vs Population-adjusted Trends in Time Use</h3> <!-- /wp:heading --> <!-- wp:columns {"className":"is-style-sticky-right"} --> <div class="wp-block-columns is-style-sticky-right"><!-- wp:column --> <div class="wp-block-column"><!-- wp:paragraph --> <p>The main chart in this post, where we show the amount of time that fathers and mothers spend with their children, plots ‘population-adjusted trends’. This means that the estimates are not what you directly observe in the surveys, but rather the output of a statistical model that takes these surveys as an input but corrects for differences in key factors such as age, partnership status, number of children and employment. </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>This is why the trends look so smooth – rather than plotting the raw survey averages for each year, the chart shows ‘modelled estimates’ after holding compositional factors constant. </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>These adjustments are helpful for the analysis, because there are large differences between population groups within countries (e.g. single parents are not able to spend as much time with their children); and since the relative size of these groups is changing, the aggregate trend that we observe is affected by these compositional changes (e.g. in some countries single-parenting is increasing, while in others it is decreasing).</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>So how much difference do these “adjustments” make? After all, we are also interested in the <em>observed</em> amount of time that parents spend with their children, and not only the values that are adjusted for population changes. What would the trends look like if we just plotted the raw survey estimates?</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The next chart shows this with data for the US. As we can see, the trend goes in the same direction for both fathers and mothers.</p> <!-- /wp:paragraph --></div> <!-- /wp:column --> <!-- wp:column --> <div class="wp-block-column"><!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/us-child-care-time-sex" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --></div> <!-- /wp:column --></div> <!-- /wp:columns --> <!-- /wp:owid/additional-information --> <!-- wp:heading {"level":4} --> <h4>Parents’ time with children, as reported by children, has also increased</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The analysis above looked at time-use diaries from <em>parents.</em> But children may experience time and attention differently from adults. Is there evidence that children also feel they are getting more time with their parents? </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In a <a href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=" target="_blank" rel="noreferrer noopener">study</a> published in 2001 in the journal <em>Demography</em>, John Sandberg and Sandra Hofferth analyzed two surveys with data from child time-use diaries in the US, and found that children reported spending about 4.3 more hours per week with mothers, and 3 more hours per week with fathers in 1997 versus 1981.{ref}The estimates from Sandberg and Hofferth correspond to weekly hours among children ages 3-12. The estimates cover both ‘engaged time’ and other activities where parents are not engaged but children say they are present. When looking jointly at both parents, Sandberg and Hofferth find that time spent with either parent remains roughly unchanged, which suggests part of the increase corresponds to an intra-household reallocation of time. In the paper you find more details and discussion, including a detailed analysis of trends for more narrow population groups and possible demographic and behavioral drivers: Sandberg, J. F., & Hofferth, S. L. (2001). <a href="https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=" target="_blank" rel="noreferrer noopener">Changes in children’s time with parents: United States, 1981–1997</a>. Demography, 38(3), 423-436. {/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The data is sparse because time-use diaries for children are not very common; but the evidence that is available is consistent with what we’ve seen above: Children in the US agree that they’re spending more time with their parents than in the past.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>Social norms affect how we allocate time</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The amount of time that parents spend with their children is determined by many factors, and working hours outside the home are only one of them. Choices, parenting norms, family size, and relationship ideals – for men and women, parents and children – matter a lot.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Several academic studies have dissected the data from the US in an attempt to disentangle the relative importance of different underlying factors. The conclusion from these studies is that the reason why we see an increase in the amount of time that American parents spend with kids is that many families, particularly those who are well-off, have been able to undergo changes in their routines and the allocation of tasks and time within the household, in order to spend more time with their children. Parents have been able to undergo ‘behavioral changes’ that have more than compensated for ‘structural changes’ that could have pushed in the opposite direction.{ref}For an overview of the literature see (i) Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents' child‐care time across countries, 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096. (ii) Sayer, L. C., Bianchi, S. M., & Robinson, J. P. (2004). Are parents investing less in children? Trends in mothers’ and fathers’ time with children. American journal of sociology, 110(1), 1-43. (iii) Sandberg, J. F., & Hofferth, S. L. (2001). Changes in children’s time with parents: United States, 1981–1997. Demography, 38(3), 423-436.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Employment, and in particular <a rel="noreferrer noopener" href="https://ourworldindata.org/working-hours" target="_blank">working hours</a> outside the home, determine a range of possible arrangements, but individual preferences and social conventions determine the actual allocation of time within that range of feasible options. </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Preferences and social conventions are dynamic and change with time. There is a large academic literature that emphasizes the importance of parent-child interactions for children’s development, and it seems natural to allow for the possibility that in recent decades parenting norms have undergone considerable change in response to this scientific evidence.{ref}The parenting guide from the National Health Insurance in the UK, for example, says <em>“Parents should make the time to play with a first or only child. And while brothers and sisters are natural playmates, parents can also play an active role in siblings' games...If you're pressed for time as a parent, it's a good idea to find ways to involve your child in what you're doing – even the housework.</em>” You find a review of the academic literature on the importance of parent-child interactions for children’s development in Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents' child‐care time across countries, 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>Why should we care?</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The available data shows that in many countries parents spend more time with their kids than ever before, and this is partly due to changes in social conventions and economic progress, especially declining working hours.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>There are large differences between countries, and also large inequalities across different population groups within each country; but overall the trends tend to go in the same direction.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Of course, there’s a lot that is not being captured by these aggregate statistics. There are important distributional issues, and there is also clearly more to parents’ and children’s welfare than ‘total time spent together’. But despite these limitations, the available research and data still offers a clear lesson: contrary to what some people fear, it’s not the case that children in rich countries are being systematically ‘shortchanged’ by widespread changes in family structures. </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Parents are spending more time with their kids than they used to, and this matters because parent-child interactions are important for childhood development.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p></p> <!-- /wp:paragraph --> <!-- wp:heading --> <h2>Time spent working</h2> <!-- /wp:heading --> <!-- wp:heading {"level":3} --> <h3>Are we working more than ever?</h3> <!-- /wp:heading --> <!-- wp-block-tombstone 37821 --> <!-- wp:paragraph --> <p>In today’s hustle and bustle world, it’s easy to assume that we are all, by and large, working more than ever. But is that really the case?</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>As we explain in detail below, the research on the history of working hours shows that this is not the case.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The available data shows that in the 19th century people across the world used to work extremely long hours, but in the last 150 years working hours have decreased substantially, particularly in today’s richest countries.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>Working hours per worker have declined after the Industrial Revolution</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The chart here shows average working hours since 1870 for a selection of countries that industrialized early. You can add or remove countries by clicking Add country on the chart.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>We show annual totals, so the trends account for changes in both the length of working days as well as the number of days worked through the year. The data comes from research by the economic historians Michael Huberman and Chris Minns, who have brought together evidence from historical records, National Accounts data, and other sources.{ref}In this chart we have taken the original data published by <a rel="noreferrer noopener" href="https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!" target="_blank">Huberman & Minns (2007)</a> and extended coverage using an updated vintage of the Penn World Table (PWT), which is in turn based on the same underlying source that Huberman and Minns used for all data since 1950, the Total Economy Database. You can find more details and links to our sources in the ‘Sources’ tab of the chart.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The chart shows that average working hours declined dramatically for workers in early-industrialized economies over the last 150 years. In 1870, workers in most of these countries worked more than 3,000 hours annually — equivalent to a grueling 60–70 hours each week for 50 weeks per year.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>But we see that today those extreme working hours have been roughly cut in half. In Germany, for example, annual working hours decreased by nearly 60% — from 3,284 hours in 1870 to 1,354 hours in 2017 — and in the UK the decrease was around 40%. Before this revolution in working hours people worked as many hours between January and July as we work today in an entire year.{ref}A key point to keep in mind when interpreting these trends is that they refer to working hours <em>per worker</em>, which is different from working hours <em>per person. </em>The per person measure corresponds to working hours per worker multiplied by the employment rate. Hence, changes in employment patterns — such as the historical rise of female participation in paid employment in these countries — mean that changes in hours per worker do not translate directly into changes in hours per person.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>For many countries in the chart we don’t have long-run series going back to the 19th century. But we do have evidence from other historical records from 1870–1900 that in many of those countries workers also used to work extremely long hours.{ref}A <a href="https://www.econstor.eu/handle/10419/67824" data-type="URL" data-id="https://www.econstor.eu/handle/10419/67824" target="_blank" rel="noreferrer noopener">study by Michael Huberman and Frank Lewis</a> reconstructed estimates of working hours in 1870 and 1900 for 48 countries across six continents using data from worker records kept by individual business establishments. They drew from a collection of records published by the US Department of Labor in 1900, and found substantial variation, but very high working hours for many non-industrialized countries. They found for example that in 1870, Colombia, Uruguay and Brazil had similar average working hours per worker as the US. The full reference of the paper is Huberman, M., & Lewis, F. D. (2007). Bend it like Beckham: Hours and wages across forty-eight countries in 1900 (No. 1229). Queen's Economics Department Working Paper.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>For those countries with long-run data in this chart we can see three distinct periods: From 1870–1913 there was a relatively slow decline; then from 1913–1938 the decline in hours steepened in the midst of the powerful sociopolitical, technological, and economic changes that took shape with World War I, the Great Depression, and the lead-up to World War II; and then after an uptick in hours during and just after World War II, the decline in hours continued for many countries, albeit at a slower pace and with large differences between countries.{ref}The increase in hours between 1938 and 1950 in the chart for some countries is due in part to the uptick during and just after World War II, but also plausibly due in part to differences in the source data and methodology.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/annual-working-hours-per-worker" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>In recent decades working hours have continued to decline in many countries, but there are large differences between countries</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>Zooming in to the last 70 years and looking at other countries beyond those who industrialized early, the data reveals a continued decline in working hours for many countries but also large differences between countries.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In the chart here we zoom in to the period since 1950 and we change the selection of countries to highlight some of the diversity in trends.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>For some countries, such as Germany, working hours have continued their steep historical decline; while for other countries, such as the US, the decline has leveled off in recent decades.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In some countries we see an inverted U-shaped pattern. In South Korea, for example, hours rose dramatically between 1950 and 1980 before falling again since the mid 1980s. And in other countries we see no recent declines — in China, for example, hours actually rose in the 1990s and early 2000s before leveling off in recent years.</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/annual-working-hours-per-worker?tab=chart&stackMode=absolute&time=1950..latest&country=DEU~USA~BRA~CHN~KOR~IND&region=World" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>Shorter work days, but also more holidays and vacations</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The decline in annual working hours described above has come from fewer working hours each day, as well as fewer working days each week and fewer working weeks in the year.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In a <a rel="noreferrer noopener" href="https://www.jstor.org/stable/10.1086/209954" target="_blank">paper</a> analyzing historical data for the US, the economist Dora Costa summarizes the evidence:{ref}Costa, D. L. (2000). <a rel="noreferrer noopener" href="https://www.jstor.org/stable/10.1086/209954?seq=1" data-type="URL" data-id="https://www.jstor.org/stable/10.1086/209954?seq=1" target="_blank">The Wage and the Length of the Work Day: From the 1890s to 1991.</a> <em>Journal of Labor Economics</em>, 18(1).{/ref}</p> <!-- /wp:paragraph --> <!-- wp:quote --> <blockquote class="wp-block-quote"><p><em>“The length of the work day fell sharply between the 1880s, when the typical worker labored 10 hours a day, 6 days a week, and 1920, when his counterpart worked an 8-hour day, 6 days a week. By 1940 the typical work schedule was 8 hours a day, 5 days a week. Although further reductions in work time largely took the form of increases in vacations, holidays, sick days, personal leave, and earlier retirement, time diary studies suggest that the work day has continued to trend downward less than 8 hours a day.”</em></p></blockquote> <!-- /wp:quote --> <!-- wp:paragraph --> <p>As Costa notes, workers had regular days off each week: one day off (usually Sunday) from at least the 1880s until around the 1940s, when two days off became more typical.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In addition to regular days off each week, workers across early-industrialized countries had days off from work for vacations and national holidays. This is shown in the chart here, which again relies on research from Huberman and Minns. The chart shows that days of vacation and holidays increased over the period from 1870–2000. The Netherlands is a stark example — workers there saw an increase from four days off for vacations and holidays in 1870 to almost 38 days off in 2000.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The declines in the length of the work day and the number of working days have been driven by several factors, including <a rel="noreferrer noopener" href="https://ourworldindata.org/rich-poor-working-hours" data-type="URL" data-id="https://ourworldindata.org/rich-poor-working-hours" target="_blank">increases in productivity</a> and the adoption of regulations that limit working hours. We discuss these and other key drivers behind working hours trends across countries and time in a series of forthcoming posts.{ref}In our <a href="https://ourworldindata.org/rich-poor-working-hours" data-type="URL" data-id="https://ourworldindata.org/rich-poor-working-hours" target="_blank" rel="noreferrer noopener">first post in the series</a>, we discuss how increases in labor productivity have driven a rise in incomes and a decrease in working hours.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/days-of-vacation-and-holidays" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>Why should we care?</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The evidence presented here comes from decades of work from economic historians and other researchers. Of course, the data is not perfect — as we explain in a forthcoming post, <a href="https://ourworldindata.org/measure-working-hours" data-type="URL" data-id="https://ourworldindata.org/measure-working-hours" target="_blank" rel="noreferrer noopener">measuring working hours with accuracy is difficult</a>, and surveys and historical records have limitations, so estimates of working hours spanning centuries necessarily come with a margin of error. But for any given country, the changes across time are much larger than the error margins at any point in time: The average worker in a rich country today really does work many fewer hours than the average worker 150 years ago.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>As the economists Diane Coyle and Leonard Nakamura explain, the study of working hours is crucial not only to measure <a rel="noreferrer noopener" href="https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable" target="_blank">macroeconomic productivity</a>, but also to measure economic well-being beyond economic <em>output</em>. A more holistic <a rel="noreferrer noopener" href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602" target="_blank">framework for measuring ‘progress’</a> needs to consider changes in how people are allowed to allocate their time over multiple activities, among which paid work is only one.{ref}Coyle, D. and Nakamura, L. I. (2019). <a rel="noreferrer noopener" href="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602" target="_blank" data-type="URL" data-id="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602">Toward a Framework for Time Use, Welfare, and Household Centric Economic Measurement.</a> Federal Reserve Bank of Philadelphia Working Paper No. 19-11.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The available evidence shows that, rather than working more than ever, workers in many countries today work <em>much less </em>than in the past 150 years. There are huge inequalities within and across countries, but substantial progress has been made.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":3} --> <h3>Do workers in richer countries work longer hours?</h3> <!-- /wp:heading --> <!-- wp-block-tombstone 37897 --> <!-- wp:paragraph --> <p>Economic prosperity in different places across our world today is <a rel="noreferrer noopener" href="https://ourworldindata.org/global-economic-inequality" target="_blank">vastly unequal</a>. People in Switzerland, one of the richest countries in the world, have an average income that is more than <em>20-times higher</em> than that of people in Cambodia.{ref}We chose Cambodia and Switzerland here because they both also have working hours data available, but the difference in average income can be even more extreme. For instance, people in Qatar have an <a rel="noreferrer noopener" href="https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT" data-type="URL" target="_blank">average income that is <em>1</em></a><a rel="noreferrer noopener" href="https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT" data-type="URL" target="_blank"><em>1</em></a><a href="https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT" data-type="URL" target="_blank" rel="noreferrer noopener"><em>7</em></a><a rel="noreferrer noopener" href="https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT" data-type="URL" target="_blank"><em>-times higher</em></a> than that of people in the Central African Republic.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>These differences refer to GDP per capita measured in international-$ and account for price differences between countries to enable comparisons. You can read more about this <a rel="noreferrer noopener" href="https://ourworldindata.org/what-are-ppps" target="_blank">here</a>.{/ref} Life in these two countries can <a rel="noreferrer noopener" href="https://www.gapminder.org/dollar-street/?max=2755&countries=kh%2Cch&media=image&min=59&topic=homes&zoom=3" target="_blank">look starkly different</a>.{ref}But life can also look similar, as you see in the pictures of the <a rel="noreferrer noopener" href="https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=homes&zoom=3" data-type="URL" data-id="https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=homes&zoom=3" target="_blank">homes</a>, <a rel="noreferrer noopener" href="https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=computers&zoom=3" data-type="URL" data-id="https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=computers&zoom=3" target="_blank">computers</a>, and <a rel="noreferrer noopener" href="https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=phones&zoom=3" data-type="URL" data-id="https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=phones&zoom=3" target="_blank">phones</a> of people on similar income levels in the two countries.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>When considering such differences in prosperity, a natural question is: who works more, people in richer countries like Switzerland or in poorer ones like Cambodia?</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Looking at the available data, the answer is clear: workers in poorer countries actually tend to work more, and sometimes <em>much</em> more.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>We see that in the chart here, with GDP per capita on the horizontal axis and annual working hours per worker on the vertical axis. </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Countries like Cambodia (which is the country in the very top-left corner) or Myanmar have some of the lowest GDP per capita but highest working hours. In Cambodia the average worker puts in 2,456 hours each year, nearly 900 more hours than in Switzerland (1,590 hours) at the bottom-right of the chart. The extra 900 hours for Cambodian workers means longer work days and many fewer days off.</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita-pwt?tab=chart&stackMode=absolute&time=2019..latest&country=&region=World" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>Working hours tend to decrease as countries become richer</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>There is a link between national income and average working hours, not only across countries at a given point in time — as shown in the chart above — but also for individual countries <em>over time.</em> </p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Since the Industrial Revolution people in many countries have <a rel="noreferrer noopener" href="https://ourworldindata.org/breaking-the-malthusian-trap" target="_blank">become richer</a>, and <a href="https://ourworldindata.org/working-more-than-ever" data-type="URL" data-id="https://ourworldindata.org/working-more-than-ever" target="_blank" rel="noreferrer noopener">working hours have decreased dramatically</a> over these last 150 years.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In the chart here we show this association between incomes and working hours over time, country by country. It is the same chart as above, except now countries’ single data points have become lines, connecting observations over time from 1950 until today.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The four highlighted countries exemplify how working hours have decreased at the same time that average incomes have increased. Germany, for example, moved far to the right as its GDP per capita increased more than 10-fold (from $5,227 to $51,191), and far to the bottom as working hours decreased by nearly half (from 2,428 hours to 1,386 hours each year).{ref}These trends in GDP per capita are measured in constant international-$ and account for inflation to enable comparisons over time and between countries. You can read more about this <a rel="noreferrer noopener" href="https://ourworldindata.org/economic-growth" target="_blank">here</a>.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>This makes sense: as people's incomes rise they can afford more of the things they enjoy, including more leisure and less time spent working.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>You can explore this association for other countries by clicking “Select countries” on the chart.</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita?tab=chart&xScale=log&stackMode=absolute&time=1950..latest&country=BRA~USA~DEU~TWN&region=World" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>People are able to work less when they work in more productive economies</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The key driver of rising national incomes and decreasing working hours is productivity growth.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Productivity refers to the rate at which inputs are turned into outputs. To understand working hours the key metric is <em>labor</em> productivity: the economic return for one hour of work.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>At the most concrete level, labor productivity captures things like the number of breads that a baker bakes in an hour, or the number of cars factory workers assemble in an hour. At the most comprehensive level, it relates the total output of the economy (GDP) to the total labor input (total annual hours worked), giving us the aggregate measure of labor productivity, GDP per hour of work.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Higher labor productivity is associated with fewer working hours, as shown in the chart here with labor productivity on the horizontal axis and annual working hours on the vertical axis. The chart currently shows data for the latest available year, but you can explore this relationship over time since 1950 by using the blue time slider at the bottom of the chart.</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/productivity-vs-annual-hours-worked?tab=chart&country=&region=World" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:paragraph --> <p>We see that the same richer countries with lower working hours we noted before — like Germany and Switzerland — have very high labor productivity (69 and 83 $/h, respectively). If workers can produce more with each hour of work, it becomes possible for them to work less.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Though this doesn’t necessarily mean they <em>actually do</em> work less — workers in the US and Singapore, for instance, work many more hours than their counterparts in countries with similar productivity.{ref}We explore the differences in working hours between similar, highly productive countries — and also the differences within those countries — in forthcoming posts.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In contrast, the countries toward the top-left of this chart have far lower labor productivity — Cambodia, for example, is at only 3$/h — and thus workers there need to work many more hours to compensate.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>At the heart of the link between productivity, incomes, and working hours is technological innovation</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>Technological innovation — defined broadly here to include physical machines as well as ideas, knowledge, and processes — makes it possible for each worker to become much more productive. And increases in productivity in turn help drive both increases in incomes and decreases in working hours.{ref}For a discussion of how technology drives productivity growth and a rise in incomes (economic growth), see Romer, P. (1990) <a rel="noreferrer noopener" href="https://www.journals.uchicago.edu/doi/abs/10.1086/261725" target="_blank">Endogenous Technological Change.</a> <em>Journal of Political Economy.</em> For a discussion of the relationship between productivity growth, economic growth, and working hours, see Boppart, T. and P. Krusell (2020) <a rel="noreferrer noopener" href="https://www.journals.uchicago.edu/doi/full/10.1086/704071?casa_token=gHAZpu6iXwkAAAAA:WUwiHwVFoOWNFdnfKenDUm9yOtgcjKxwZpohoDcmZk8ZMwMASp86fPHmmd3-r8NLJk-9UKorX7o" data-type="URL" data-id="https://www.journals.uchicago.edu/doi/full/10.1086/704071?casa_token=gHAZpu6iXwkAAAAA:WUwiHwVFoOWNFdnfKenDUm9yOtgcjKxwZpohoDcmZk8ZMwMASp86fPHmmd3-r8NLJk-9UKorX7o" target="_blank">Labor Supply in the Past, Present, and Future: A Balanced-Growth Perspective</a>. <em>Journal of Political Economy.</em>{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>A prime example of how tech innovation drives productivity growth is agriculture. As we show in detail in <a href="https://ourworldindata.org/crop-yields" target="_blank" rel="noreferrer noopener">our entry on Crop Yields</a>, innovations like better machinery, crop varieties, fertilizers, and land management have enabled farmers to be <em>much more</em> productive. In the US, for example, farm production per labor hour increased nearly 16-fold from 1948–2011.{ref}See Figure 18 on p. 28 of Wang et al (2015) <a href="https://www.ers.usda.gov/webdocs/publications/45387/53417_err189.pdf?v=6052.7" target="_blank" rel="noreferrer noopener">Agricultural Productivity Growth in the United States: Measurement, Trends, and Drivers</a>. <em>USDA Economic Research Report 189.</em>{/ref} This increased productivity enables us to feed a <a href="https://ourworldindata.org/grapher/population?country=~OWID_WRL" target="_blank" rel="noreferrer noopener">rapidly growing population</a>, even while the <a href="https://ourworldindata.org/grapher/employment-by-economic-sector?stackMode=relative" target="_blank" rel="noreferrer noopener">fraction of people working in agriculture</a> is smaller than ever.{ref}The transition of employment out of agriculture to other economic sectors as countries become richer is known as ‘structural transformation’. You can read more about this in our post <a href="https://ourworldindata.org/structural-transformation-and-deindustrialization-evidence-from-todays-rich-countries" target="_blank" rel="noreferrer noopener"><em>Structural transformation: how did today’s rich countries become ‘deindustrialized’?</em></a>{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The chart here shows the growth in labor productivity, not just for agriculture but for the entire economy. The technological, economic, and social structures in richer countries have enabled workers there to produce more while working less.</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~TWN~GBR~USA~CHE&region=World" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:paragraph --> <p>Besides tech innovation, there is evidence that working fewer hours can itself keep productivity higher, making the link between working hours and productivity self-reinforcing. For example, economist <a rel="noreferrer noopener" href="https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext" target="_blank">John Pencavel (2015) studied</a> munitions workers in war-time Britain and found that their productivity stayed high up to a certain threshold of hours, but declined markedly above that threshold.{ref}Pencavel, J. (2015) <a rel="noreferrer noopener" href="https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext" target="_blank">The productivity of working hours</a>. <em>The Economic Journal.</em>{/ref} We’ve probably all experienced the drop in productivity that comes at the end of a very long day of work.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>What we learn from this</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The data show that it is workers in poorer countries who tend to work more, and sometimes <em>a lot</em> more, than those in richer countries.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>This has large implications for the way we think about the economic progress made in the last two centuries and the nature of inequality between countries today.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>It means that residents of today’s poorer countries like Cambodia and Myanmar — and also of today’s richer countries in the past when they were poor — are not just <em>consumption</em> poor, often unable to afford necessities like food and medicine. It means they are also <em>leisure</em> poor: because productivity is low and they must work so much just to scrape by, they can’t afford to spend much time improving their condition, becoming educated, or simply enjoying leisure time.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>That people in poorer countries work so much more than in richer countries shows that differences in prosperity are not due to differences in work ethic — they are largely due to differences in circumstance and opportunity. As we ask in <a rel="noreferrer noopener" href="https://ourworldindata.org/talent-is-everywhere-opportunity-is-not" target="_blank">another post</a>, “what would have been the chances for Steve Jobs if he was born in the Central African Republic?” No matter how hard he worked or how smart he was, it is difficult to imagine that Steve Jobs would’ve been able to realize his potential with such a steep mountain of inequality to climb.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>We also see what <em>the world </em><a rel="noreferrer noopener" href="https://www.imf.org/en/Publications/WP/Issues/2018/12/07/Invisible-Geniuses-Could-the-Knowledge-Frontier-Advance-Faster-46383" target="_blank">misses out on</a> when exceptionally talented people, including all the brilliant but underprivileged people in today’s poorest countries, don’t have the opportunity to realize their potential.{ref}Agarwal, R. and Gaule, P. (2020) <a rel="noreferrer noopener" href="https://www.aeaweb.org/articles?id=10.1257/aeri.20190457&&from=f" target="_blank">Invisible Geniuses: Could the Knowledge Frontier Advance Faster?</a><em> American Economic Review: Insights.</em>{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Finding ways to raise productivity is therefore not just key to increasing production, but also to the reduction in working hours that is necessary for a society to flourish.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":3} --> <h3>How are working hours measured and what can we learn from the data?</h3> <!-- /wp:heading --> <!-- wp-block-tombstone 38252 --> <!-- wp:paragraph --> <p>Work is a central part of our lives. It is something we do almost every day, for much of the day, for decades on end. Because it is so central, looking closely at how much time we spend working can tell us a lot about our lives and the societies we live in.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The data on working hours shows, for example, that rather than working more than ever — as is so commonly believed — people in many countries today <a href="https://ourworldindata.org/working-more-than-ever" target="_blank" rel="noreferrer noopener">work <em>much less</em> than in the past 150 years</a>.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Working less means being able to spend time becoming <a rel="noreferrer noopener" href="https://ourworldindata.org/global-education" target="_blank">more educated</a>, or simply <a rel="noreferrer noopener" href="https://ourworldindata.org/time-use-living-conditions" target="_blank">enjoying more leisure time</a>. This is substantial progress, but there is still <a rel="noreferrer noopener" href="https://ourworldindata.org/rich-poor-working-hours" data-type="URL" data-id="https://ourworldindata.org/rich-poor-working-hours" target="_blank">huge inequality across countries</a>, and progress still to make.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>To understand these changes in societies and people’s lives over time, and the substantial differences we see in the world today, it is crucial to measure and study how much time people spend working.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>How are working hours actually measured? Where does the data come from, and how can researchers reconstruct long-run trends?</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Here we provide an overview of the main data sources, compare the data, and explain the relevant differences and measurement limitations.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>How are working hours measured?</h4> <!-- /wp:heading --> <!-- wp:heading {"level":5} --> <h5>Surveys</h5> <!-- /wp:heading --> <!-- wp:paragraph --> <p>Surveys are the primary way to collect data on working hours. They are typically conducted by national statistical agencies and come in three main types: labor force surveys, establishment surveys, and time use surveys. These surveys all provide an important perspective on working hours, but there are some key differences.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Labor force surveys collect data on employment status and time spent working by asking individual workers themselves. Of the survey types, these provide the most comprehensive perspective, covering hours <em>actually</em> worked in all economic sectors as part of both formal and <a href="https://ourworldindata.org/grapher/informal-employment-of-total-non-agricultural-employment" target="_blank" rel="noreferrer noopener">informal employment</a>, full-time and part-time, as well as self-employment and unpaid family work.{ref}Hours actually worked means hours spent directly on work and excludes things like annual leave, sick leave, public holidays, meal breaks, and commuting time. Unpaid family work in this case generally includes market-oriented work, such as for the family business, but not other unpaid work at home such as childcare, cooking, and cleaning. Since the latter type of unpaid work is typically performed by women, this has large implications for understanding gender differences in labor. We discuss these issues as part of <a href="https://ourworldindata.org/female-labor-supply#definitions-measurement" target="_blank" rel="noreferrer noopener">our entry on Women’s Employment</a>.{/ref} But labor force surveys only cover residents of a country above a certain age (usually 15), which depending on the country might exclude a non-trivial number of workers.{ref}Only covering resident workers means that any <a href="https://ec.europa.eu/eurostat/cache/digpub/eumove/bloc-2c.html?lang=en#:~:text=In%202019%2C%20the%20largest%20number,and%20Belgium%20(50%20000)." target="_blank" rel="noreferrer noopener">cross-border workers</a> are excluded. Only covering workers above a certain age means that any child laborers are excluded. While the incidence of child labor has been going down over time, especially in high-income countries, there are still an <a href="https://ourworldindata.org/child-labor" target="_blank" rel="noreferrer noopener">estimated 265 million working children</a> in the world (almost 17% of the worldwide child population).{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Establishment surveys collect data on employment and working hours as reported by employers.{ref}Employers include businesses, non-profits, some government agencies, and other organizations that pay a wage.{/ref} But because hours are reported by employers, these surveys often only cover paid or contractual hours and exclude self-employment, informal work, and some smaller firms.{ref}Unlike hours actually worked, paid or contractual hours typically include some time <em>not</em> spent working, such as during sick leave, and fail to include time spent working that wasn't paid or planned, such as overtime.{/ref} On the other hand, establishment surveys provide more detail on the industry of work than other surveys, and are more consistent with how GDP is measured, making them useful for <a rel="noreferrer noopener" href="https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~CHE~TWN~GBR~USA&region=World" target="_blank">studying labor productivity</a>.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Time use surveys collect data on how individuals spend their time — <a href="https://ourworldindata.org/time-use-living-conditions#daily-activities-similarities-and-differences-across-countries" data-type="URL" data-id="https://ourworldindata.org/time-use-living-conditions#daily-activities-similarities-and-differences-across-countries" target="_blank" rel="noreferrer noopener">down to the minute</a> — across a number of activities in a typical day, including paid work.{ref}Activities also include unpaid household work, school, leisure time, eating, and sleeping.{/ref} This level of granularity provides a useful complement to the other surveys, but as a trade-off time use surveys sample fewer people and are conducted less frequently and by fewer countries.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":5} --> <h5>National accounts</h5> <!-- /wp:heading --> <!-- wp:paragraph --> <p>To get the most comprehensive perspective on working hours possible, many countries aggregate data from these surveys with data from other sources — such as censuses, tax records, and social security registers — in an economic measurement framework called national accounts.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>National accounts, and the surveys they rely on, are standardized to a degree across countries, which can facilitate international comparisons.{ref}By organizations such as the <a rel="noreferrer noopener" href="https://unstats.un.org/unsd/nationalaccount/sna.asp" target="_blank">United Nations</a>, <a href="https://ilostat.ilo.org/resources/concepts-and-definitions/description-hours-of-work/" data-type="URL" data-id="https://ilostat.ilo.org/resources/concepts-and-definitions/description-hours-of-work/" target="_blank" rel="noreferrer noopener">International Labor Organization (ILO)</a>, <a rel="noreferrer noopener" href="https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en" target="_blank">OECD</a>, and <a rel="noreferrer noopener" href="https://ec.europa.eu/eurostat/web/microdata/european-union-labour-force-survey" target="_blank">Eurostat</a>.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>But these comparisons often have limitations because many countries still implement the methods in different ways. For instance, countries might bring together different data in their national accounts, or aggregate it differently. And many countries don’t have the capacity to conduct comprehensive surveys of their labor force and produce national accounts-based statistics, giving a more limited view of work there.{ref}For further discussion of different sources and their comparability, see the methods guides of the <a rel="noreferrer noopener" href="https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en" target="_blank">OECD</a> and the <a rel="noreferrer noopener" href="https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite" target="_blank">Total Economy Database</a> and the work of <a href="https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344" data-type="URL" data-id="https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344" target="_blank" rel="noreferrer noopener">Bick, Brüggemann, and Fuchs-Schündeln (2019)</a>.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>How do researchers reconstruct long-run historical trends?</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>Comprehensive, cross-country data on working hours just isn’t available for the years before the mid 20th century. But researchers like <a href="http://www.sciencedirect.com/science/article/pii/S0014498307000058" target="_blank" rel="noreferrer noopener">Huberman and Minns (2007)</a>{ref}Huberman, M. and Minns, C. (2007) <a href="http://www.sciencedirect.com/science/article/pii/S0014498307000058" target="_blank" rel="noreferrer noopener">The times they are not changin’: Days and hours of work in Old and New Worlds, 1870–2000.</a> <em>Explorations in Economic History.</em>{/ref} have been able to fill some of the gap by reconstructing long-run trends for a selection of countries. How do they do it?</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Through often painstaking effort, researchers have been able to find and piece together the relevant historical records that do exist. In the work of Huberman and Minns, one of the key sources for historical data on many countries is a <a href="https://catalog.hathitrust.org/Record/008420895" target="_blank" rel="noreferrer noopener">report from the US Department of Labor</a> published in 1900.{ref}U.S. Department of Labor (1900) <a href="https://catalog.hathitrust.org/Record/008420895" target="_blank" rel="noreferrer noopener">Fifteenth Annual Report of the Commissioner of Labor: Wages in Commercial Countries. 2 vols.</a> Washington, DC.{/ref} The report compiled the records of many thousands of workers across numerous sectors from establishment surveys in 88 countries and territories. To reconstruct the trends in later years, Huberman and Minns pulled together data from the International Labor Organization, the work of peer researchers, and other sources.{ref}The original sources are: 1870–1913: Huberman (2004) [in turn relying on the US Department of Labor Fifteenth Annual Report, 1900]; 1929–1938: International Labor Organization (1934–39), except for Canada (Ostry and Zaidi, 1972), U.S. (Jones, 1963; Owen, 1988), and Australia (Butlin, 1977); 1950–2000: University of Groningen and the Conference Board GGDC Total Economy Database (2005).{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>This was an impressive feat of reconstruction, but historical records like this do have limitations. For instance, as exhaustive as they were, the establishment-level records used by Huberman and Minns still excluded agricultural work, part-time work, and many smaller firms.</p> <!-- /wp:paragraph --> <!-- wp:heading {"level":4} --> <h4>How does the data from different sources compare?</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The work by Huberman and Minns is an important example of how researchers often combine and adjust underlying sources to produce one-off cross-country estimates. Another important study is the one of Bick, Brüggemann, and Fuchs-Schündeln (2019),{ref}Bick, A., Brüggemann, B., and Fuchs-Schündeln, N. (2019) <a rel="noreferrer noopener" href="https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344" target="_blank">Hours Worked in Europe and the United States: New Data, New Answers.</a> <em>The Scandinavian Journal of Economics.</em>{/ref} who further standardized labor force surveys to enhance comparability for a selection of countries.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Besides these one-off estimates, several international organizations and research centers aggregate the working hours estimates published by national statistical agencies into cross-country datasets. The two most important datasets come from the <a rel="noreferrer noopener" href="https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#" target="_blank">OECD</a> and the <a rel="noreferrer noopener" href="https://www.rug.nl/ggdc/productivity/pwt/?lang=en" target="_blank">Penn World Table</a> (PWT). These both draw on national accounts estimates when available, but they can differ in the other sources they use and their method of aggregation.{ref}PWT sources its working hours data from <a rel="noreferrer noopener" href="https://www.conference-board.org/data/economydatabase/total-economy-database-productivity" target="_blank">The Conference Board’s Total Economy Database</a> (TED). For more details on the underlying sources, see the <a rel="noreferrer noopener" href="https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite" target="_blank">TED guide</a> and the <a rel="noreferrer noopener" href="https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#" target="_blank">OECD database</a>.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In the chart you can compare annual working hours data from these four datasets. The data is shown one country at a time — with France currently selected. You can look at other countries by clicking ‘Change country’ on the chart, but note that not all sources publish data for every country.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>As expected, there are differences between the sources. In 2000, for instance, Bick et al. estimates 1,642 hours of work for French workers, OECD estimates 1,558 hours, PWT estimates 1,550 hours, and Huberman and Minns estimates 1,443 hours. These differences are due to the use of different underlying sources and methods. Bick et al. use only labor force surveys; the others all rely primarily on national accounts data, but which nonetheless still have differences.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>It’s also clear that these differences between sources are quite small when compared to the huge changes over the longer run. The difference between sources in 2000 is at most 200 hours, while the historical data from Huberman and Minns shows that from 1870 to 2000 annual working hours in France decreased by <em>1,725 hours</em> (from 3,168 to 1,443 hours).</p> <!-- /wp:paragraph --> <!-- wp:html --> <iframe src="https://ourworldindata.org/grapher/compare-sources-working-hours" loading="lazy" style="width: 100%; height: 600px; border: 0px none;"></iframe> <!-- /wp:html --> <!-- wp:heading {"level":4} --> <h4>What does this tell us about the study of working hours?</h4> <!-- /wp:heading --> <!-- wp:paragraph --> <p>The analysis here shows that working hours data can have limitations — due to differences in the sources or the way the method is implemented — but that what these matter for our interpretation of the data depends on the context.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In a context where precise comparisons of similar countries is important, smaller differences between sources can really matter. This is why to compare recent working hours levels in the US and Europe, Bick et al. used only labor force surveys, which they standardized even further to maximize cross-country comparability. But as a trade-off, it was only possible to look at a small selection of richer countries.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>In a context where we want to focus on a larger scale — such as the <a rel="noreferrer noopener" href="https://ourworldindata.org/working-more-than-ever" target="_blank">long-run historical trends</a> we see in the chart — the limitations of the sources are not large enough to undermine our conclusions.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>Large international datasets like PWT do not have the highest levels of cross-country comparability, but they allow us to look at many more countries across the world and uncover broad and important trends, such as the <a rel="noreferrer noopener" href="https://ourworldindata.org/rich-poor-working-hours" data-type="URL" data-id="https://ourworldindata.org/rich-poor-working-hours" target="_blank">large differences in working hours between the richest and poorest countries</a>.{ref}We gain further confidence in these conclusions when they are echoed by research that focuses only on more standardized, comparable sources for a necessarily smaller set of countries, as in the work by <a rel="noreferrer noopener" href="https://www.aeaweb.org/articles?id=10.1257/aer.20151720" target="_blank">Bick, Fuchs-Schündeln, and Lagakos (2018)</a>.{/ref}</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>PWT and OECD are also useful in contexts where we want an exhaustive picture of the trends in individual countries, since they are often based on national accounts that bring together data from many sources to give a comprehensive perspective on working hours.</p> <!-- /wp:paragraph --> <!-- wp:paragraph --> <p>The data on working hours isn’t perfect, and it’s important to understand the limitations, but it can still tell us a lot about our lives and the world.</p> <!-- /wp:paragraph --> <!-- wp:heading --> <h2>Data Sources</h2> <!-- /wp:heading --> <!-- wp:heading {"level":4} --> <h4>OECD Time Use Database and Gender Data Portal</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Time spent on various activities per day, by sex</li><li><strong>Geographical coverage:</strong> OECD countries plus China, India, and South Africa</li><li><strong>Time span:</strong> from 1998 onwards</li><li><strong>Available at:</strong> <a rel="noreferrer noopener" href="https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE" target="_blank">Time Use Database</a>; <a rel="noreferrer noopener" href="https://www.oecd.org/gender/data/balancingpaidworkunpaidworkandleisure.htm" target="_blank">Gender Data Portal</a></li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Centre for Time Use Research</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Time spent on various activities per day; self-reported levels of enjoyment</li><li><strong>Geographical coverage:</strong> 30 countries across the world</li><li><strong>Time span:</strong> from 1964 onwards</li><li><strong>Available at:</strong> <a rel="noreferrer noopener" href="https://www.timeuse.org/" target="_blank">https://www.timeuse.org/</a><ul><li>Gershuny, J., and Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.</li></ul></li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>US Bureau of Labor Statistics</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Time spent with others per day</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> from 2003 onwards</li><li><strong>Available at:</strong> <a rel="noreferrer noopener" href="https://www.bls.gov/tus/datafiles-0319.htm" target="_blank">https://www.bls.gov/tus/datafiles-0319.htm</a></li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>IPUMS USA</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Percentage of Americans living alone, by age</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> from 1900 onwards</li><li><strong>Available at:</strong> <a rel="noreferrer noopener" href="https://ipums.org/projects/ipums-usa/d010.v8.0" target="_blank">https://ipums.org/projects/ipums-usa/d010.v8.0</a></li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Dotti Sani and Treas (2016)</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Time parents spent with children per day</li><li><strong>Geographical coverage:</strong> United States and select countries in Europe</li><li><strong>Time span:</strong> 1965–2012</li><li><strong>Available at:</strong> Dotti Sani, G. M., and Treas, J. (2016). <a rel="noreferrer noopener" href="https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Dotti+Sani+and+Treas+2016+Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965%E2%80%932012&btnG=" target="_blank">Educational gradients in parents’ child‐care time across countries, 1965–2012</a>. Journal of Marriage and Family.</li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Sayer, Bianchi, and Robinson (2004)</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Time parents spent on primary childcare activities per day</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> 1965–1998</li><li><strong>Available at:</strong> Sayer, L., Bianchi, S., and Robinson, J. (2004). <a href="https://www.journals.uchicago.edu/doi/pdf/10.1086/386270" target="_blank" rel="noreferrer noopener">Are Parents Investing Less in Children? Trends in Mothers’ and Fathers’ Time with Children</a>. American Journal of Sociology.</li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Huberman and Minns (2007)</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Annual hours of full-time production workers (male and female) in non-agricultural activities; days off from work for vacations and holidays</li><li><strong>Geographical coverage:</strong> United States, Australia, Canada, and select countries in Europe</li><li><strong>Time span:</strong> 1870–2000</li><li><strong>Available at:</strong> Huberman, M. and Minns, C. (2007). <a rel="noreferrer noopener" href="https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!" target="_blank">The times they are not changin’: Days and hours of work in Old and New Worlds, 1870–2000.</a> Explorations in Economic History.</li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Penn World Table</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Average annual hours worked by persons engaged; number of persons engaged; real and PPP-adjusted GDP in US millions of dollars</li><li><strong>Geographical coverage:</strong> Countries across the world</li><li><strong>Time span:</strong> from 1950 onwards</li><li><strong>Available at:</strong> <a rel="noreferrer noopener" href="https://www.rug.nl/ggdc/productivity/pwt/" target="_blank">https://www.rug.nl/ggdc/productivity/pwt/</a><ul><li>Feenstra, R. C., Inklaar, R., and Timmer, M.P. (2015). <a rel="noreferrer noopener" href="https://www.aeaweb.org/articles?id=10.1257/aer.20130954" target="_blank">The Next Generation of the Penn World Table.</a> American Economic Review.</li></ul></li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Eurostat</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Time use survey data</li><li><strong>Geographical coverage:</strong> Europe</li><li><strong>Time span:</strong> from 1998 onwards</li><li><strong>Available at:</strong> <a rel="noreferrer noopener" href="https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#" target="_blank">https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#</a></li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Ramey and Francis (2009)</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Self-reported enjoyment of various activities; time spent on various activities, by sex and age; days of work lost to sickness</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> 1900–2005</li><li><strong>Available at:</strong> Ramey, V. A., and Francis, N. (2009). <a rel="noreferrer noopener" href="https://www.aeaweb.org/articles?id=10.1257/mac.1.2.189" target="_blank">A century of work and leisure</a>. American Economic Journal: Macroeconomics.</li></ul> <!-- /wp:list --> <!-- wp:heading {"level":4} --> <h4>Costa (2000)</h4> <!-- /wp:heading --> <!-- wp:list --> <ul><li><strong>Data:</strong> Working hours, by sex; number of working days; wages</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> 1890–1991</li><li><strong>Available at:</strong> Costa, D. L. (2000). <a rel="noreferrer noopener" href="https://www.jstor.org/stable/10.1086/209954?seq=1" target="_blank">The Wage and the Length of the Work Day: From the 1890s to 1991.</a> Journal of Labor Economics.</li></ul> <!-- /wp:list --> | { "id": "wp-37823", "slug": "time-use", "content": { "toc": [], "body": [ { "type": "text", "value": [ { "text": "Time is the ultimate limited resource. Every single one of us has exactly the same \u201ctime budget\u201d: 24 hours per day, 365 days per year \u2013 8,760 hours ", "spanType": "span-simple-text" }, { "url": "https://owid.cloud/life-expectancy", "children": [ { "text": "every year of our lives", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "How do we actually spend our time? There are many commonalities across the world: we all sleep, work, eat, and enjoy leisure time. But there are also important differences in the freedom people have to spend time on the things they value most. Studying how people across the world spend their time provides an important perspective for understanding living conditions, economic opportunities, and general well-being.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Here we present the data on time use. We explore how it differs across countries and over time and how these differences matter for people\u2019s lives.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "entry-summary", "items": [ { "slug": "how-do-people-across-the-world-spend-their-time-and-what-does-this-tell-us-about-living-conditions", "text": "How people spend their time is similar in many ways across countries: we all sleep, work, eat, and enjoy leisure. But there are also important differences." }, { "slug": "going-beyond-averages-the-gender-gap-in-leisure-time", "text": "There is a gender gap in leisure time: in all countries surveyed the average leisure time for men is higher than for women." }, { "slug": "who-do-we-spend-time-with-across-our-lifetime", "text": "Who we spend time with evolves across our lifetimes. As we enter older age we spend an increasing amount of time alone." }, { "slug": "are-parents-spending-less-time-with-their-kids", "text": "The amount of time parents spend with their kids has been increasing." }, { "slug": "are-we-working-more-than-ever", "text": "Working hours have decreased dramatically in the last 150 years for workers in many countries." }, { "slug": "do-workers-in-richer-countries-work-longer-hours", "text": "But there are still large differences between countries: workers in poorer countries tend to work much more than workers in richer countries." }, { "slug": "how-are-working-hours-measured-and-what-can-we-learn-from-the-data", "text": "The primary way to measure working hours is with surveys, but the data can have limitations that are important to understand." } ], "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Time use from the perspective of an average day", "spanType": "span-simple-text" } ], "type": "heading", "level": 1, "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "How do people across the world spend their time\u00a0and what does this tell us about living conditions?", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Sleep, work, eat, leisure \u2013 at a high level most of us spend time on similar activities. But just how similar are the daily activities of people across the world? This is something worth considering, not just to serve our curiosity but because differences in the way we spend time give us meaningful perspectives on living conditions, economic opportunities and general well-being.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Here we take a look at the data. We explore some of the key patterns that emerge from cross-country time use surveys, and then dig deeper to understand how these differences matter for people\u2019s well-being.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Daily activities: similarities and differences across countries", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "In the chart here we compare average time spent across a number of common activities. The data comes from the OECD and brings together estimates from time diaries where respondents are asked to record the sequence of what they did over a specific day, as well as from general questionnaires where respondents are asked to recall the amount of time spent on different activities on a specific day in the previous week.{ref}The \u2018time-diary method\u2019 is generally more reliable and allows a richer analysis of routines, because it measures not only aggregate times but also sequences and clock-times. Time-diary data is less common, but it is available for some countries from the ", "spanType": "span-simple-text" }, { "url": "https://www.timeuse.org/mtus", "children": [ { "text": "Multinational Time Use Study", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". We explore time-use \u2018tempograms\u2019 from the MTUS in a forthcoming companion post.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The first thing that jumps out from this chart is that there are indeed many similarities across countries.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "This is not surprising \u2013 most of us try to split our days into \u201cwork, rest and fun\u201d, and so there are some predictable patterns. We spend the most time working and sleeping; and paid work, housework, leisure, eating and sleeping take together 80-90% of the 1440 minutes that we all have available every day.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "But if we look closely, we also see some important differences. Consider sleeping, for example. From this sample of countries, South Koreans sleep the least \u2013 averaging 7 hours and 51 minutes of sleep every day. In India and the US, at the other end of the spectrum, people sleep an hour more on average.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Work is another important activity where we see large differences. Countries are sorted by paid work hours in the chart \u2013 from highest to lowest. In China and Mexico people spend, on an average day, almost twice as much time on paid work as people in Italy and France do. This is a general pattern: ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours", "children": [ { "text": "People in richer countries can afford to work less", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". Keep in mind that this chart shows the average for all people in the working age bracket, from 15 to 64 years, whether they are actually employed or not.{ref}Because these estimates include people who are not employed they are much lower than the estimates of working hours\u00a0", "spanType": "span-simple-text" }, { "children": [ { "text": "per", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "children": [ { "text": "worker", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "url": "https://ourworldindata.org/time-use#time-spent-working", "children": [ { "text": "we present elsewhere", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". The estimates also differ because of differences in the sources: time-use surveys compared to labor force surveys and national accounts data.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Differences in ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/age-structure", "children": [ { "text": "demographics", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/global-education", "children": [ { "text": "education", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/economic-growth", "children": [ { "text": "economic prosperity", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " all contribute to these inequalities in work and time use. But what\u2019s clear in the chart here is that there are also some differences in time use that are not well explained by economic or demographic differences. In the UK, for example, people spend more time working than in France; but in both countries people report spending a similar amount of time on leisure activities.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Cultural differences are likely to play a role here. The French seem to spend much more time eating than the British \u2013 and in this respect the data actually goes in line with stereotypes about food culture. People in France, Greece, Italy and Spain report spending more time eating than people in most other European countries. The country where people spend the least time eating and drinking is the USA (63 minutes).", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "How do people spend their time?{ref}OECD (2020) ", "spanType": "span-simple-text" }, { "url": "https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE", "children": [ { "text": "Time Use Database.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": "{/ref}", "spanType": "span-simple-text" } ], "type": "heading", "level": 5, "parseErrors": [] }, { "alt": "", "size": "wide", "type": "image", "filename": "Time-Use-by-Country-OECD.png", "parseErrors": [] }, { "type": "text", "value": [ { "url": "https://owid.cloud/app/uploads/2020/12/Time-Use-in-OECD-Countries-OECD.xlsx", "children": [ { "text": "Download the underlying data for this chart (.xlsx)", "spanType": "span-simple-text" } ], "spanType": "span-link" } ], "parseErrors": [] }, { "text": [ { "text": "Going beyond averages: The gender gap in leisure time", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Going beyond national averages reveals important within-country inequalities. The gender gap in leisure time, for example, is a key dimension along which large inequalities exist.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The chart here relies on the same time-use data described above, but shows total leisure time for men and women separately. Time for men is shown on the horizontal axis, while time for women appears on the vertical axis. The dotted diagonal line denotes \u2018gender parity\u2019, so the further away a country is from the diagonal line, the larger the difference between men and women.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "As we can see, in all countries the average leisure time for men is higher than for women \u2013 all bubbles are below the diagonal line \u2013 but in some countries the gaps are much larger. In Norway the difference is very small, while in Portugal men report almost 50% more leisure time than women.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "A key factor driving these differences in leisure time is the gender gap in unpaid work. As we explain in detail in a ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/female-labor-supply#informal-work-unpaid-care-work", "children": [ { "text": "companion post", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", women are responsible for a disproportionate amount of unpaid work, and have less leisure as a result.{ref}If you want to dig deeper you can explore gender differences across all other activities directly from our source, via the ", "spanType": "span-simple-text" }, { "url": "https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE", "children": [ { "text": "OECD Data Portal", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". And you can read more about within-country inequalities in time use along other dimensions, such as income and education, in ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=The+Middle+Class+Time+Squeeze%2C+Economic+Studies+at+Brookings%2C+Sawhill+and+Guyot+%282020%29&btnG=", "children": [ { "text": "this Brookings Paper", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", where the authors focus on the \u2018middle class time squeeze\u2019 in the US. See: Sawhill, I. V., & Guyot, K. (2020). The Middle Class Time Squeeze. Economic Studies at Brookings. Brookings Institution.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/minutes-spent-on-leisure?tab=chart&stackMode=absolute&time=latest&country=®ion=World", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "Why should we care about differences in time use?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Every single one of us has the same \u201ctime budget\u201d: 24 hours per day and 365 days per year. But of course not all of us can choose to spend time on the activities that we enjoy most. Differences in our freedom to allocate time to the things we enjoy is the main reason why time-use data is important for studying living conditions.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In the UK, researchers from the ", "spanType": "span-simple-text" }, { "url": "https://www.timeuse.org/", "children": [ { "text": "Centre for Time Use Research", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " linked time-use diaries with the respondents\u2019 assessments of enjoyment, on a scale from 1 to 7, to better understand the connection between time use and well-being. The chart here, which we\u2019ve adapted from the book \u2018What We Really Do All Day\u2019", "spanType": "span-simple-text" }, { "children": [ { "text": ", ", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "by professors Jonathan Gershuny and Oriel Sullivan, shows the results. The estimates correspond to average reported levels of enjoyment for each activity, with confidence intervals.{ref}The underlying data comes from time-use diaries where respondents are asked to record the sequence of what they do over a specific day, and how much they enjoy each \u2018episode\u2019 (i.e. what they do) on a scale from 1 to 7. All episodes reported are then coded and grouped into similar activities. To arrive at the mean enjoyment scores, the authors multiply the duration of each episode where the activity category concerned is the primary activity recorded, by the enjoyment level to arrive at the total enjoyment score for that episode. Then they sum these total enjoyment scores for each category of activity across the day, and finally divide these daily enjoyment total scores for each activity by the amount of time devoted to the activity. In this way, they arrive at an appropriately weighted mean enjoyment level for each activity across all those who engage in it. For more details see Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "We see that the most enjoyed activities involve rest or leisure activities such as eating out, sleeping, going to sports events, playing computer games or attending cultural performances. The activities receiving the lowest ratings include doing school homework, looking for a job, or doing housework.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The activity where people show the greatest variation in enjoyment is working a \u201cSecond Job\u201d. This likely reflects the difference between people who work a second job because they want to, and those who work a second job because they ", "spanType": "span-simple-text" }, { "children": [ { "text": "have", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "children": [ { "text": "to", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": ".", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "So what do we learn from this?", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "First, we learn that the enjoyment of activities is, at least to some degree, predictable and stable. This means we can take activity groups and make meaningful comparisons across groups of people. Economists, for example, ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Ramey%2C+V.+A.%2C+%26+Francis%2C+N.+%282009%29.+A+century+of+work+and+leisure.+American+Economic+Journal%3A+Macroeconomics%2C+1%282%29%2C+189-224.&btnG=", "children": [ { "text": "often classify", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " any activity with an enjoyment level below work as a \u201cnon-leisure activity\u201d, in order to measure trends in leisure across people and time.{ref}You find a very clear and complete explanation of this in Ramey, V. A., & Francis, N. (2009). A century of work and leisure. American Economic Journal: Macroeconomics, 1(2), 189-224.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "But beyond this, and more importantly, this confirms that time-use is informative about well-being.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The fact that there is a very clear and predictable pattern in the enjoyment of activities suggests that differences in time use do, indeed, give us meaningful perspectives on living conditions and economic opportunities. In countries where people do more paid and unpaid work, and have less time for leisure, their enjoyment \u2013 and ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/happiness-and-life-satisfaction", "children": [ { "text": "happiness and life satisfaction", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " \u2013 levels are likely to be lower.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "How do people rate the enjoyment of different activities?{ref}Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref}", "spanType": "span-simple-text" } ], "type": "heading", "level": 5, "parseErrors": [] }, { "alt": "", "size": "wide", "type": "image", "filename": "Enjoyment-Level-of-Time-Use.png", "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Time use from the perspective of the life cycle", "spanType": "span-simple-text" } ], "type": "heading", "level": 1, "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Who do we spend time with across our lifetime?", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "text", "value": [ { "text": "As we go through life we build personal relationships with different people \u2013 family, friends, coworkers, partners. These relationships, which are ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/social-connections-and-loneliness#social-connections", "children": [ { "text": "deeply important to all of us", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", evolve with time. As we grow older we build new relationships while others transform or fade, and towards the end of life many of us spend a lot of time alone.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Taking the big picture over the entire life course: Who do we actually spend our time with?", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "From adolescence to old age: who do we spend our time with?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "To understand how social connections evolve throughout our lives we can look at survey data on ", "spanType": "span-simple-text" }, { "children": [ { "text": "how much", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " time people spend with others, and ", "spanType": "span-simple-text" }, { "children": [ { "text": "who", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " that time is spent with.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The chart here shows the amount of time that people in the US report spending in the company of others, based on their age. The data comes from time-use surveys, where people are asked to list all the activities that they perform over a full day, and the people who were there during each activity. We currently only have data with this granularity for the US \u2013 time-use surveys are common across many countries, but what is special about the US is that respondents of the American Time Use Survey are asked to list everyone who was present for each activity.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/time-spent-with-relationships-by-age-us?tab=chart&stackMode=absolute®ion=World", "type": "chart", "parseErrors": [] }, { "type": "text", "value": [ { "text": "The numbers in this chart are based on averages for a cross-section of the American society \u2013 people are only interviewed once, but we have brought together a decade of surveys, tabulating the average amount of time that survey respondents of different ages report spending with other people", "spanType": "span-simple-text" }, { "children": [ { "text": ".", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "{ref}When interpreting this chart it\u2019s important to bear in mind that the relationships used to categorise people are not exhaustive (i.e., survey respondents could also list being with people who didn\u2019t fit any of the listed categories, or for whom a relationship was unclear or unknown \u2013 we do not count these instances in the estimates). Additionally, time spent with multiple people can be counted more than once; so attending a party with friends and your spouse, for example, would show up for both \u201cfriends\u201d and \u201cpartner\u201d in our estimates. The implication is that companion categories cannot be stacked to add up total time spent in the company of others.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Who we spend our time with changes a lot over the course of life", "spanType": "span-simple-text" } ], "type": "heading", "level": 4, "parseErrors": [] }, { "type": "text", "value": [ { "text": "When we\u2019re young \u2013 particularly in our teens \u2013 we spend a lot of our time with friends, parents, siblings and extended family. As we enter our 20s, time with friends, siblings and parents starts to drop off quickly. Instead, we start spending an increasing amount of time with partners and children. (The chart shows an average across Americans, so for those that have children the time spent with children is even higher, since the average is pulled down by those without children.)", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "As the chart shows, this continues throughout our 30s, 40s and 50s \u2013 over this period of their life, Americans spend much of their time with partners, children and, unsurprisingly, co-workers.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "For those 60 and older, we see a significant drop-off in time spent with co-workers. This makes sense, considering many people in the US ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/average-effective-retirement-men", "children": [ { "text": "enter retirement in their mid 60s", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". We see that this time is partly displaced by more time with partners.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In terms of the diversity of interactions, this chart suggests that the number of people with whom we interact is highest around 40, but then things change substantially after that. And this is perhaps the most conspicuous trend in the chart: above 40, people spend an increasing amount of time alone.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Older people spend a lot of time alone", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Older people spend a large amount of time alone and it is understandable why \u2013 time spent alone increases with age because this is when health typically deteriorates and people lose relatives and friends.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Indeed, many people who are older than 60 live alone as this chart shows clearly: living alone is particularly common for older adults. Today nearly 4 out of every 10 Americans who are older than 89 years old live alone.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Another interesting point here is that the share of people across all age groups who live alone has been increasing over time. This is part of a more general global trend \u2013 if you want to read more about the global \u2018rise of living alone\u2019, we provide a detailed account of this trend across countries in a ", "spanType": "span-simple-text" }, { "url": "http://ourworldindata.org/living-alone", "children": [ { "text": "companion post", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/percentage-of-americans-living-alone-by-age", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "Older people spend more time alone, but this doesn\u2019t necessarily mean they\u2019re lonely", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The data shows that as we become older we tend to spend more and more time alone. What\u2019s more, the data also shows that older people today spend more time alone than older people did in the past.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "We might think older people are therefore more lonely \u2013 but this is not necessarily the case.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Spending time alone is not the same as feeling lonely. This is a point that is well recognised by researchers, and one which has been confirmed empirically across countries. Surveys that ask people about living arrangements, time use, and feelings of loneliness find that solitude, by itself, is not a good predictor of loneliness. You can read our overview of the evidence in ", "spanType": "span-simple-text" }, { "url": "http://ourworldindata.org/lonely-not-alone", "children": [ { "text": "this post", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "So, what about loneliness? If we focus on self-reported loneliness, there is ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/loneliness-epidemic#loneliness-trends-over-time-are-people-lonelier-today-than-in-the-past", "children": [ { "text": "little evidence", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " of an upward trend over time in the US; and importantly, it\u2019s ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/social-connections-and-loneliness#do-we-become-lonelier-as-we-get-older", "children": [ { "text": "not the case", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " that loneliness keeps going up as we become older. In fact a ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Are+U.S.+older+adults+getting+lonelier%3F%3A+Age%2C+period%2C+and+cohort+differences.+Psychology+and+Aging.&btnG=", "children": [ { "text": "recent study", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " based on surveys that track the same individuals over time found that after age 50 \u2013 which is the earliest age of participants in the analysis \u2013 loneliness tended to decrease, until about 75, after which it began to increase again.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Taking the evidence together, the message is not that we should be sad about the prospect of aging, but rather that we should recognise the fact that social connections are complex. We often tend to look at the amount of time spent with others as a marker of social well-being; but the quality of time spent with others, and our expectations, matter even more for our feelings of connection and loneliness.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "gray-section", "items": [ { "text": [ { "text": "Additional information", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "expandable-paragraph", "items": [ { "type": "text", "value": [ { "text": "In the chart where we plot the amount of time that people in the US report spending in the company of others, it\u2019s important to keep in mind that we are taking a look at a cross-section of society. This means that we are actually seeing the result of two underlying trends.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "On one hand, we see the effect of aging on social connections (we relate to different people and reallocate time as we go through different stages of life), but we also see the effect of cohort trends (compared to people in the past, today\u2019s older generations in the US tend to be ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/life-expectancy?tab=chart&country=JPN~GBR~IND~ETH~OWID_WRL~KOR~ZAF~USA®ion=World", "children": [ { "text": "healthier", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/average-real-gdp-per-capita-across-countries-and-regions?tab=chart&country=Western%20Offshoots~Western%20Europe~Western%20Asia~Eastern%20Europe~OWID_WRL~Latin%20America~East%20Asia~Africa~USA", "children": [ { "text": "richer", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", and might also have different expectations, preferences and opportunities).", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Disentangling these two effects is difficult, so it is important to keep in mind that at least some of the age gradients we observe might be partly explained by cohort changes, rather than life-cycle trends.\u00a0", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "This is why it\u2019s important to rely not only on cross-sectional data, but also on surveys that track the same individuals over time.", "spanType": "span-simple-text" } ], "parseErrors": [] } ], "parseErrors": [] } ], "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Parents\u2019 time with children", "spanType": "span-simple-text" } ], "type": "heading", "level": 1, "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Are parents spending less time with their kids?", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Over the last 50 years many countries have seen large changes in family structures and the institution of marriage. These changes \u2013 which include a ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/marriages-and-divorces#single-parenting-is-common-and-in-many-countries-it-has-increased-in-recent-decades", "children": [ { "text": "rise in single-parenting", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and a large increase in the ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/female-labor-supply#female-participation-in-labor-markets-grew-remarkably-in-the-20th-century", "children": [ { "text": "share of women working outside the home", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " \u2013 have made some people worry that children might be getting \u2018short-changed\u2019, because parents are not spending as much time with them.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In 1999, for example, a ", "spanType": "span-simple-text" }, { "url": "https://clintonwhitehouse4.archives.gov/media/pdf/famfinal.pdf", "children": [ { "text": "report", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " from the Council of Economic Advisors in the US analyzed trends over the second half of the 20th century and concluded: ", "spanType": "span-simple-text" }, { "children": [ { "text": "\u201cThe increase in hours mothers spend in paid work, combined with the shift toward single-parent families, resulted in families on average experiencing a decrease of 22 hours a week (14 percent) in parental time available outside of paid work that they could spend with their children.\u201d", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The line of thought behind these concerns is that changes to the structure of families and work have meant that children spend less time with parents, because parents \u2013 particularly mothers \u2013 spend less time at home.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Here we review the evidence and show that this reasoning is flawed. As we explain, in the US and many other rich countries parents spend more time with their kids today than 50 years ago. Equating \u2018mother time at home' with 'children\u2019s time with parents\u2019 is a huge and unhelpful oversimplification.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Mothers and fathers spend an increasing amount of time with kids: evidence from time-use surveys", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The chart here shows the time that mothers and fathers spend with their children. This is measured using time-use diaries where parents record the amount of time that they spend on various activities, including child care.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "These estimates, which are sourced from a ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=", "children": [ { "text": "paper", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " published in 2016 by sociologists Giulia Dotti Sani and Judith Treas, are disaggregated by education levels and are adjusted to account for demographic differences between countries. We explain ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/parents-time-with-kids#digging-deeper-raw-vs-population-adjusted-trends-in-time-use", "children": [ { "text": "below", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " in more detail why these adjustments are important.{ref}The full reference for the source paper is: Dotti Sani, G. M., & Treas, J. (2016). ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=", "children": [ { "text": "Educational gradients in parents' child\u2010care time across countries", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096. {/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "As we can see here, there has been a clear increase in the amount of time parents spend with their children over the last 50 years. This is true for both fathers and mothers, and holds across almost all education groups and countries. The two exceptions are France, where mothers\u2019 time has declined (from a very comparatively high level); and Slovenia where it has remained roughly constant among non-university-educated parents.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In terms of within-country inequalities we also see two clear patterns. First, in all countries mothers spend more time in child care activities than fathers. The differences are large and persistent across education groups. In some countries, such as Canada, France and the US, this gender gap has shrunk; while in other countries such as Denmark and Spain, the gap has widened.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Second, in all countries there is a positive \u2018educational gradient\u2019, meaning more educated parents tend to spend more time with their children. In many countries this gradient increased, and nowhere did it decline.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "How much time do parents spend with their children per day?{ref}Dotti Sani, G. M., & Treas, J. (2016). ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=", "children": [ { "text": "Educational gradients in parents' child\u2010care time across countries", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref}", "spanType": "span-simple-text" } ], "type": "heading", "level": 5, "parseErrors": [] }, { "alt": "", "size": "wide", "type": "image", "filename": "Parent-time-spent-with-children-2.png", "parseErrors": [] }, { "type": "gray-section", "items": [ { "text": [ { "text": "Additional information", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "expandable-paragraph", "items": [ { "type": "text", "value": [ { "text": "The main chart in this post, where we show the amount of time that fathers and mothers spend with their children, plots \u2018population-adjusted trends\u2019. This means that the estimates are not what you directly observe in the surveys, but rather the output of a statistical model that takes these surveys as an input but corrects for differences in key factors such as age, partnership status, number of children and employment.\u00a0", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "This is why the trends look so smooth \u2013 rather than plotting the raw survey averages for each year, the chart shows \u2018modelled estimates\u2019 after holding compositional factors constant.\u00a0", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "These adjustments are helpful for the analysis, because there are large differences between population groups within countries (e.g. single parents are not able to spend as much time with their children); and since the relative size of these groups is changing, the aggregate trend that we observe is affected by these compositional changes (e.g. in some countries single-parenting is increasing, while in others it is decreasing).", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "So how much difference do these \u201cadjustments\u201d make? After all, we are also interested in the ", "spanType": "span-simple-text" }, { "children": [ { "text": "observed", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " amount of time that parents spend with their children, and not only the values that are adjusted for population changes. What would the trends look like if we just plotted the raw survey estimates?", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The next chart shows this with data for the US. As we can see, the trend goes in the same direction for both fathers and mothers.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/us-child-care-time-sex", "type": "chart", "parseErrors": [] } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Parents\u2019 time with children, as reported by children, has also increased", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The analysis above looked at time-use diaries from ", "spanType": "span-simple-text" }, { "children": [ { "text": "parents.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " But children may experience time and attention differently from adults. Is there evidence that children also feel they are getting more time with their parents?\u00a0", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In a ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=", "children": [ { "text": "study", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " published in 2001 in the journal ", "spanType": "span-simple-text" }, { "children": [ { "text": "Demography", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": ", John Sandberg and Sandra Hofferth analyzed two surveys with data from child time-use diaries in the US, and found that children reported spending about 4.3 more hours per week with mothers, and 3 more hours per week with fathers in 1997 versus 1981.{ref}The estimates from Sandberg and Hofferth correspond to weekly hours among children ages 3-12. The estimates cover both \u2018engaged time\u2019 and other activities where parents are not engaged but children say they are present. When looking jointly at both parents, Sandberg and Hofferth find that time spent with either parent remains roughly unchanged, which suggests part of the increase corresponds to an intra-household reallocation of time. In the paper you find more details and discussion, including a detailed analysis of trends for more narrow population groups and possible demographic and behavioral drivers: Sandberg, J. F., & Hofferth, S. L. (2001). ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=", "children": [ { "text": "Changes in children\u2019s time with parents: United States, 1981\u20131997", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". Demography, 38(3), 423-436. {/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The data is sparse because time-use diaries for children are not very common; but the evidence that is available is consistent with what we\u2019ve seen above: Children in the US agree that they\u2019re spending more time with their parents than in the past.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Social norms affect how we allocate time", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The amount of time that parents spend with their children is determined by many factors, and working hours outside the home are only one of them. Choices, parenting norms, family size, and relationship ideals \u2013 for men and women, parents and children \u2013 matter a lot.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Several academic studies have dissected the data from the US in an attempt to disentangle the relative importance of different underlying factors. The conclusion from these studies is that the reason why we see an increase in the amount of time that American parents spend with kids is that many families, particularly those who are well-off, have been able to undergo changes in their routines and the allocation of tasks and time within the household, in order to spend more time with their children. Parents have been able to undergo \u2018behavioral changes\u2019 that have more than compensated for \u2018structural changes\u2019 that could have pushed in the opposite direction.{ref}For an overview of the literature see (i) Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents' child\u2010care time across countries, 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096. (ii) Sayer, L. C., Bianchi, S. M., & Robinson, J. P. (2004). Are parents investing less in children? Trends in mothers\u2019 and fathers\u2019 time with children. American journal of sociology, 110(1), 1-43. (iii) Sandberg, J. F., & Hofferth, S. L. (2001). Changes in children\u2019s time with parents: United States, 1981\u20131997. Demography, 38(3), 423-436.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Employment, and in particular ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/working-hours", "children": [ { "text": "working hours", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " outside the home, determine a range of possible arrangements, but individual preferences and social conventions determine the actual allocation of time within that range of feasible options.\u00a0", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Preferences and social conventions are dynamic and change with time. There is a large academic literature that emphasizes the importance of parent-child interactions for children\u2019s development, and it seems natural to allow for the possibility that in recent decades parenting norms have undergone considerable change in response to this scientific evidence.{ref}The parenting guide from the National Health Insurance in the UK, for example, says ", "spanType": "span-simple-text" }, { "children": [ { "text": "\u201cParents should make the time to play with a first or only child. And while brothers and sisters are natural playmates, parents can also play an active role in siblings' games...If you're pressed for time as a parent, it's a good idea to find ways to involve your child in what you're doing \u2013 even the housework.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "\u201d You find a review of the academic literature on the importance of parent-child interactions for children\u2019s development in Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents' child\u2010care time across countries, 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Why should we care?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The available data shows that in many countries parents spend more time with their kids than ever before, and this is partly due to changes in social conventions and economic progress, especially declining working hours.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "There are large differences between countries, and also large inequalities across different population groups within each country; but overall the trends tend to go in the same direction.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Of course, there\u2019s a lot that is not being captured by these aggregate statistics. There are important distributional issues, and there is also clearly more to parents\u2019 and children\u2019s welfare than \u2018total time spent together\u2019. But despite these limitations, the available research and data still offers a clear lesson: contrary to what some people fear, it\u2019s not the case that children in rich countries are being systematically \u2018shortchanged\u2019 by widespread changes in family structures.\u00a0", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Parents are spending more time with their kids than they used to, and this matters because parent-child interactions are important for childhood development.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Time spent working", "spanType": "span-simple-text" } ], "type": "heading", "level": 1, "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Are we working more than ever?", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "text", "value": [ { "text": "In today\u2019s hustle and bustle world, it\u2019s easy to assume that we are all, by and large, working more than ever. But is that really the case?", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "As we explain in detail below, the research on the history of working hours shows that this is not the case.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The available data shows that in the 19th century people across the world used to work extremely long hours, but in the last 150 years working hours have decreased substantially, particularly in today\u2019s richest countries.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Working hours per worker have declined after the Industrial Revolution", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The chart here shows average working hours since 1870 for a selection of countries that industrialized early. You can add or remove countries by clicking Add country on the chart.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "We show annual totals, so the trends account for changes in both the length of working days as well as the number of days worked through the year. The data comes from research by the economic historians Michael Huberman and Chris Minns, who have brought together evidence from historical records, National Accounts data, and other sources.{ref}In this chart we have taken the original data published by ", "spanType": "span-simple-text" }, { "url": "https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!", "children": [ { "text": "Huberman & Minns (2007)", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and extended coverage using an updated vintage of the Penn World Table (PWT), which is in turn based on the same underlying source that Huberman and Minns used for all data since 1950, the Total Economy Database. You can find more details and links to our sources in the \u2018Sources\u2019 tab of the chart.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The chart shows that average working hours declined dramatically for workers in early-industrialized economies over the last 150 years. In 1870, workers in most of these countries worked more than 3,000 hours annually \u2014 equivalent to a grueling 60\u201370 hours each week for 50 weeks per year.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "But we see that today those extreme working hours have been roughly cut in half. In Germany, for example, annual working hours decreased by nearly 60% \u2014 from 3,284 hours in 1870 to 1,354 hours in 2017 \u2014 and in the UK the decrease was around 40%. Before this revolution in working hours people worked as many hours between January and July as we work today in an entire year.{ref}A key point to keep in mind when interpreting these trends is that they refer to working hours\u00a0", "spanType": "span-simple-text" }, { "children": [ { "text": "per worker", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": ", which is different from working hours\u00a0", "spanType": "span-simple-text" }, { "children": [ { "text": "per person.\u00a0", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "The per person measure corresponds to working hours per worker multiplied by the employment rate. Hence, changes in employment patterns \u2014 such as the historical rise of female participation in paid employment in these countries \u2014 mean that changes in hours per worker do not translate directly into changes in hours per person.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "For many countries in the chart we don\u2019t have long-run series going back to the 19th century. But we do have evidence from other historical records from 1870\u20131900 that in many of those countries workers also used to work extremely long hours.{ref}A ", "spanType": "span-simple-text" }, { "url": "https://www.econstor.eu/handle/10419/67824", "children": [ { "text": "study by Michael Huberman and Frank Lewis", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " reconstructed estimates of working hours in 1870 and 1900 for 48 countries across six continents using data from worker records kept by individual business establishments. They drew from a collection of records published by the US Department of Labor in 1900, and found substantial variation, but very high working hours for many non-industrialized countries. They found for example that in 1870, Colombia, Uruguay and Brazil had similar average working hours per worker as the US. The full reference of the paper is Huberman, M., & Lewis, F. D. (2007). Bend it like Beckham: Hours and wages across forty-eight countries in 1900 (No. 1229). Queen's Economics Department Working Paper.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "For those countries with long-run data in this chart we can see three distinct periods: From 1870\u20131913 there was a relatively slow decline; then from 1913\u20131938 the decline in hours steepened in the midst of the powerful sociopolitical, technological, and economic changes that took shape with World War I, the Great Depression, and the lead-up to World War II; and then after an uptick in hours during and just after World War II, the decline in hours continued for many countries, albeit at a slower pace and with large differences between countries.{ref}The increase in hours between 1938 and 1950 in the chart for some countries is due in part to the uptick during and just after World War II, but also plausibly due in part to differences in the source data and methodology.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/annual-working-hours-per-worker", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "In recent decades working hours have continued to decline in many countries, but there are large differences between countries", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Zooming in to the last 70 years and looking at other countries beyond those who industrialized early, the data reveals a continued decline in working hours for many countries but also large differences between countries.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In the chart here we zoom in to the period since 1950 and we change the selection of countries to highlight some of the diversity in trends.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "For some countries, such as Germany, working hours have continued their steep historical decline; while for other countries, such as the US, the decline has leveled off in recent decades.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In some countries we see an inverted U-shaped pattern. In South Korea, for example, hours rose dramatically between 1950 and 1980 before falling again since the mid 1980s. And in other countries we see no recent declines \u2014 in China, for example, hours actually rose in the 1990s and early 2000s before leveling off in recent years.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/annual-working-hours-per-worker?tab=chart&stackMode=absolute&time=1950..latest&country=DEU~USA~BRA~CHN~KOR~IND®ion=World", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "Shorter work days, but also more holidays and vacations", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The decline in annual working hours described above has come from fewer working hours each day, as well as fewer working days each week and fewer working weeks in the year.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In a ", "spanType": "span-simple-text" }, { "url": "https://www.jstor.org/stable/10.1086/209954", "children": [ { "text": "paper", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " analyzing historical data for the US, the economist Dora Costa summarizes the evidence:{ref}Costa, D. L. (2000). ", "spanType": "span-simple-text" }, { "url": "https://www.jstor.org/stable/10.1086/209954?seq=1", "children": [ { "text": "The Wage and the Length of the Work Day: From the 1890s to 1991.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "children": [ { "text": "Journal of Labor Economics", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": ", 18(1).{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "type": "text", "value": [ { "children": [ { "text": "\u201cThe length of the work day fell sharply between the 1880s, when the typical worker labored 10 hours a day, 6 days a week, and 1920, when his counterpart worked an 8-hour day, 6 days a week. By 1940 the typical work schedule was 8 hours a day, 5 days a week. Although further reductions in work time largely took the form of increases in vacations, holidays, sick days, personal leave, and earlier retirement, time diary studies suggest that the work day has continued to trend downward less than 8 hours a day.\u201d", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "parseErrors": [] } ], "type": "blockquote", "parseErrors": [] }, { "type": "text", "value": [ { "text": "As Costa notes, workers had regular days off each week: one day off (usually Sunday) from at least the 1880s until around the 1940s, when two days off became more typical.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In addition to regular days off each week, workers across early-industrialized countries had days off from work for vacations and national holidays. This is shown in the chart here, which again relies on research from Huberman and Minns. The chart shows that days of vacation and holidays increased over the period from 1870\u20132000. The Netherlands is a stark example \u2014 workers there saw an increase from four days off for vacations and holidays in 1870 to almost 38 days off in 2000.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The declines in the length of the work day and the number of working days have been driven by several factors, including ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/rich-poor-working-hours", "children": [ { "text": "increases in productivity", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and the adoption of regulations that limit working hours. We discuss these and other key drivers behind working hours trends across countries and time in a series of forthcoming posts.{ref}In our ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/rich-poor-working-hours", "children": [ { "text": "first post in the series", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", we discuss how increases in labor productivity have driven a rise in incomes and a decrease in working hours.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/days-of-vacation-and-holidays", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "Why should we care?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The evidence presented here comes from decades of work from economic historians and other researchers. Of course, the data is not perfect \u2014 as we explain in a forthcoming post, ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/measure-working-hours", "children": [ { "text": "measuring working hours with accuracy is difficult", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", and surveys and historical records have limitations, so estimates of working hours spanning centuries necessarily come with a margin of error. But for any given country, the changes across time are much larger than the error margins at any point in time: The average worker in a rich country today really does work many fewer hours than the average worker 150 years ago.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "As the economists Diane Coyle and Leonard Nakamura explain, the study of working hours is crucial not only to measure ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable", "children": [ { "text": "macroeconomic productivity", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", but also to measure economic well-being beyond economic ", "spanType": "span-simple-text" }, { "children": [ { "text": "output", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": ". A more holistic ", "spanType": "span-simple-text" }, { "url": "https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602", "children": [ { "text": "framework for measuring \u2018progress\u2019", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " needs to consider changes in how people are allowed to allocate their time over multiple activities, among which paid work is only one.{ref}Coyle, D. and Nakamura, L. I. (2019). ", "spanType": "span-simple-text" }, { "url": "https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602", "children": [ { "text": "Toward a Framework for Time Use, Welfare, and Household Centric Economic Measurement.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " Federal Reserve Bank of Philadelphia Working Paper No. 19-11.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The available evidence shows that, rather than working more than ever, workers in many countries today work ", "spanType": "span-simple-text" }, { "children": [ { "text": "much less ", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "than in the past 150 years. There are huge inequalities within and across countries, but substantial progress has been made.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "Do workers in richer countries work longer hours?", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Economic prosperity in different places across our world today is ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/global-economic-inequality", "children": [ { "text": "vastly unequal", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". People in Switzerland, one of the richest countries in the world, have an average income that is more than ", "spanType": "span-simple-text" }, { "children": [ { "text": "20-times higher", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " than that of people in Cambodia.{ref}We chose Cambodia and Switzerland here because they both also have working hours data available, but the difference in average income can be even more extreme. For instance, people in Qatar have an ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT", "children": [ { "text": "average income that is ", "spanType": "span-simple-text" }, { "children": [ { "text": "1", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "spanType": "span-link" }, { "url": "https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT", "children": [ { "children": [ { "text": "1", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "spanType": "span-link" }, { "url": "https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT", "children": [ { "children": [ { "text": "7", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "spanType": "span-link" }, { "url": "https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT", "children": [ { "children": [ { "text": "-times higher", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "spanType": "span-link" }, { "text": " than that of people in the Central African Republic.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "These differences refer to GDP per capita measured in international-$ and account for price differences between countries to enable comparisons. You can read more about this ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/what-are-ppps", "children": [ { "text": "here", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{/ref} Life in these two countries can ", "spanType": "span-simple-text" }, { "url": "https://www.gapminder.org/dollar-street/?max=2755&countries=kh%2Cch&media=image&min=59&topic=homes&zoom=3", "children": [ { "text": "look starkly different", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{ref}But life can also look similar, as you see in the pictures of the ", "spanType": "span-simple-text" }, { "url": "https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=homes&zoom=3", "children": [ { "text": "homes", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", ", "spanType": "span-simple-text" }, { "url": "https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=computers&zoom=3", "children": [ { "text": "computers", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", and ", "spanType": "span-simple-text" }, { "url": "https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=phones&zoom=3", "children": [ { "text": "phones", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " of people on similar income levels in the two countries.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "When considering such differences in prosperity, a natural question is: who works more, people in richer countries like Switzerland or in poorer ones like Cambodia?", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Looking at the available data, the answer is clear: workers in poorer countries actually tend to work more, and sometimes ", "spanType": "span-simple-text" }, { "children": [ { "text": "much", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " more.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "We see that in the chart here, with GDP per capita on the horizontal axis and annual working hours per worker on the vertical axis.\u00a0", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Countries like Cambodia (which is the country in the very top-left corner) or Myanmar have some of the lowest GDP per capita but highest working hours. In Cambodia the average worker puts in 2,456 hours each year, nearly 900 more hours than in Switzerland (1,590 hours) at the\u00a0bottom-right of the chart. The extra 900 hours for Cambodian workers means longer work days and many fewer days off.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita-pwt?tab=chart&stackMode=absolute&time=2019..latest&country=®ion=World", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "Working hours tend to decrease as countries become richer", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "There is a link between national income and average working hours, not only across countries at a given point in time \u2014 as shown in the chart above \u2014 but also for individual countries ", "spanType": "span-simple-text" }, { "children": [ { "text": "over time.", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Since the Industrial Revolution people in many countries have ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/breaking-the-malthusian-trap", "children": [ { "text": "become richer", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", and ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/working-more-than-ever", "children": [ { "text": "working hours have decreased dramatically", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " over these last 150 years.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In the chart here we show this association between incomes and working hours over time, country by country. It is the same chart as above, except now countries\u2019 single data points have become lines, connecting observations over time from 1950 until today.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The four highlighted countries exemplify how working hours have decreased at the same time that average incomes have increased. Germany, for example, moved far to the right as its GDP per capita increased more than 10-fold (from $5,227 to $51,191), and far to the bottom as working hours decreased by nearly half (from 2,428 hours to 1,386 hours each year).{ref}These trends in GDP per capita are measured in constant international-$ and account for inflation to enable comparisons over time and between countries. You can read more about this ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/economic-growth", "children": [ { "text": "here", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "This makes sense: as people's incomes rise they can afford more of the things they enjoy, including more leisure and less time spent working.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "You can explore this association for other countries by clicking \u201cSelect countries\u201d on the chart.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita?tab=chart&xScale=log&stackMode=absolute&time=1950..latest&country=BRA~USA~DEU~TWN®ion=World", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "People are able to work less when they work in more productive economies", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The key driver of rising national incomes and decreasing working hours is productivity growth.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Productivity refers to the rate at which inputs are turned into outputs. To understand working hours the key metric is ", "spanType": "span-simple-text" }, { "children": [ { "text": "labor", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " productivity: the economic return for one hour of work.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "At the most concrete level, labor productivity captures things like the number of breads that a baker bakes in an hour, or the number of cars factory workers assemble in an hour. At the most comprehensive level, it relates the total output of the economy (GDP) to the total labor input (total annual hours worked), giving us the aggregate measure of labor productivity, GDP per hour of work.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Higher labor productivity is associated with fewer working hours, as shown in the chart here with labor productivity on the horizontal axis and annual working hours on the vertical axis. The chart currently shows data for the latest available year, but you can explore this relationship over time since 1950 by using the blue time slider at the bottom of the chart.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/productivity-vs-annual-hours-worked?tab=chart&country=®ion=World", "type": "chart", "parseErrors": [] }, { "type": "text", "value": [ { "text": "We see that the same richer countries with lower working hours we noted before \u2014 like Germany and Switzerland \u2014 have very high labor productivity (69 and 83 $/h, respectively). If workers can produce more with each hour of work, it becomes possible for them to work less.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Though this doesn\u2019t necessarily mean they ", "spanType": "span-simple-text" }, { "children": [ { "text": "actually do", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " work less \u2014 workers in the US and Singapore, for instance, work many more hours than their counterparts in countries with similar productivity.{ref}We explore the differences in working hours between similar, highly productive countries \u2014 and also the differences within those countries \u2014 in forthcoming posts.{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In contrast, the countries toward the top-left of this chart have far lower labor productivity \u2014 Cambodia, for example, is at only 3$/h \u2014 and thus workers there need to work many more hours to compensate.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "At the heart of the link between productivity, incomes, and working hours is technological innovation", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Technological innovation \u2014 defined broadly here to include physical machines as well as ideas, knowledge, and processes \u2014 makes it possible for each worker to become much more productive. And increases in productivity in turn help drive both increases in incomes and decreases in working hours.{ref}For a discussion of how technology drives productivity growth and a rise in incomes (economic growth), see Romer, P. (1990) ", "spanType": "span-simple-text" }, { "url": "https://www.journals.uchicago.edu/doi/abs/10.1086/261725", "children": [ { "text": "Endogenous Technological Change.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "children": [ { "text": "Journal of Political Economy.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " For a discussion of the relationship between productivity growth, economic growth, and working hours, see Boppart, T. and P. Krusell (2020) ", "spanType": "span-simple-text" }, { "url": "https://www.journals.uchicago.edu/doi/full/10.1086/704071?casa_token=gHAZpu6iXwkAAAAA:WUwiHwVFoOWNFdnfKenDUm9yOtgcjKxwZpohoDcmZk8ZMwMASp86fPHmmd3-r8NLJk-9UKorX7o", "children": [ { "text": "Labor Supply in the Past, Present, and Future: A Balanced-Growth Perspective", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". ", "spanType": "span-simple-text" }, { "children": [ { "text": "Journal of Political Economy.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "A prime example of how tech innovation drives productivity growth is agriculture. As we show in detail in ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/crop-yields", "children": [ { "text": "our entry on Crop Yields", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", innovations like better machinery, crop varieties, fertilizers, and land management have enabled farmers to be ", "spanType": "span-simple-text" }, { "children": [ { "text": "much more", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " productive. In the US, for example, farm production per labor hour increased nearly 16-fold from 1948\u20132011.{ref}See Figure 18 on p. 28 of Wang et al (2015) ", "spanType": "span-simple-text" }, { "url": "https://www.ers.usda.gov/webdocs/publications/45387/53417_err189.pdf?v=6052.7", "children": [ { "text": "Agricultural Productivity Growth in the United States: Measurement, Trends, and Drivers", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". ", "spanType": "span-simple-text" }, { "children": [ { "text": "USDA Economic Research Report 189.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "{/ref} This increased productivity enables us to feed a ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/population?country=~OWID_WRL", "children": [ { "text": "rapidly growing population", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", even while the ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/employment-by-economic-sector?stackMode=relative", "children": [ { "text": "fraction of people working in agriculture", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " is smaller than ever.{ref}The transition of employment out of agriculture to other economic sectors as countries become richer is known as \u2018structural transformation\u2019. You can read more about this in our post ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/structural-transformation-and-deindustrialization-evidence-from-todays-rich-countries", "children": [ { "children": [ { "text": "Structural transformation: how did today\u2019s rich countries become \u2018deindustrialized\u2019?", "spanType": "span-simple-text" } ], "spanType": "span-italic" } ], "spanType": "span-link" }, { "text": "{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The chart here shows the growth in labor productivity, not just for agriculture but for the entire economy. The technological, economic, and social structures in richer countries have enabled workers there to produce more while working less.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~TWN~GBR~USA~CHE®ion=World", "type": "chart", "parseErrors": [] }, { "type": "text", "value": [ { "text": "Besides tech innovation, there is evidence that working fewer hours can itself keep productivity higher, making the link between working hours and productivity self-reinforcing. For example, economist ", "spanType": "span-simple-text" }, { "url": "https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext", "children": [ { "text": "John Pencavel (2015) studied", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " munitions workers in war-time Britain and found that their productivity stayed high up to a certain threshold of hours, but declined markedly above that threshold.{ref}Pencavel, J. (2015) ", "spanType": "span-simple-text" }, { "url": "https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext", "children": [ { "text": "The productivity of working hours", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". ", "spanType": "span-simple-text" }, { "children": [ { "text": "The Economic Journal.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "{/ref} We\u2019ve probably all experienced the drop in productivity that comes at the end of a very long day of work.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "What we learn from this", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The data show that it is workers in poorer countries who tend to work more, and sometimes ", "spanType": "span-simple-text" }, { "children": [ { "text": "a lot", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " more, than those in richer countries.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "This has large implications for the way we think about the economic progress made in the last two centuries and the nature of inequality between countries today.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "It means that residents of today\u2019s poorer countries like Cambodia and Myanmar \u2014 and also of today\u2019s richer countries in the past when they were poor \u2014 are not just ", "spanType": "span-simple-text" }, { "children": [ { "text": "consumption", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " poor, often unable to afford necessities like food and medicine. It means they are also ", "spanType": "span-simple-text" }, { "children": [ { "text": "leisure", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " poor: because productivity is low and they must work so much just to scrape by, they can\u2019t afford to spend much time improving their condition, becoming educated, or simply enjoying leisure time.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "That people in poorer countries work so much more than in richer countries shows that differences in prosperity are not due to differences in work ethic \u2014 they are largely due to differences in circumstance and opportunity. As we ask in ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/talent-is-everywhere-opportunity-is-not", "children": [ { "text": "another post", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", \u201cwhat would have been the chances for Steve Jobs if he was born in the Central African Republic?\u201d No matter how hard he worked or how smart he was, it is difficult to imagine that Steve Jobs would\u2019ve been able to realize his potential with such a steep mountain of inequality to climb.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "We also see what ", "spanType": "span-simple-text" }, { "children": [ { "text": "the world ", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "url": "https://www.imf.org/en/Publications/WP/Issues/2018/12/07/Invisible-Geniuses-Could-the-Knowledge-Frontier-Advance-Faster-46383", "children": [ { "text": "misses out on", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " when exceptionally talented people, including all the brilliant but underprivileged people in today\u2019s poorest countries, don\u2019t have the opportunity to realize their potential.{ref}Agarwal, R. and Gaule, P. (2020) ", "spanType": "span-simple-text" }, { "url": "https://www.aeaweb.org/articles?id=10.1257/aeri.20190457&&from=f", "children": [ { "text": "Invisible Geniuses: Could the Knowledge Frontier Advance Faster?", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "children": [ { "text": " American Economic Review: Insights.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Finding ways to raise productivity is therefore not just key to increasing production, but also to the reduction in working hours that is necessary for a society to flourish.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "How are working hours measured and what can we learn from the data?", "spanType": "span-simple-text" } ], "type": "heading", "level": 2, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Work is a central part of our lives. It is something we do almost every day, for much of the day, for decades on end. Because it is so central, looking closely at how much time we spend working can tell us a lot about our lives and the societies we live in.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The data on working hours shows, for example, that rather than working more than ever \u2014\u00a0as is so commonly believed \u2014 people in many countries today ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/working-more-than-ever", "children": [ { "text": "work ", "spanType": "span-simple-text" }, { "children": [ { "text": "much less", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " than in the past 150 years", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Working less means being able to spend time becoming ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/global-education", "children": [ { "text": "more educated", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", or simply ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/time-use-living-conditions", "children": [ { "text": "enjoying more leisure time", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". This is substantial progress, but there is still ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/rich-poor-working-hours", "children": [ { "text": "huge inequality across countries", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", and progress still to make.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "To understand these changes in societies and people\u2019s lives over time, and the substantial differences we see in the world today, it is crucial to measure and study how much time people spend working.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "How are working hours actually measured? Where does the data come from, and how can researchers reconstruct long-run trends?", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Here we provide an overview of the main data sources, compare the data, and explain the relevant differences and measurement limitations.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "How are working hours measured?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "text": [ { "text": "Surveys", "spanType": "span-simple-text" } ], "type": "heading", "level": 4, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Surveys are the primary way to collect data on working hours. They are typically conducted by national statistical agencies and come in three main types: labor force surveys, establishment surveys, and time use surveys. These surveys all provide an important perspective on working hours, but there are some key differences.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Labor force surveys collect data on employment status and time spent working by asking individual workers themselves. Of the survey types, these provide the most comprehensive perspective, covering hours ", "spanType": "span-simple-text" }, { "children": [ { "text": "actually", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " worked in all economic sectors as part of both formal and ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/informal-employment-of-total-non-agricultural-employment", "children": [ { "text": "informal employment", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", full-time and part-time, as well as self-employment and unpaid family work.{ref}Hours actually worked means hours spent directly on work and excludes things like annual leave, sick leave, public holidays, meal breaks, and commuting time. Unpaid family work in this case generally includes market-oriented work, such as for the family business, but not other unpaid work at home such as childcare, cooking, and cleaning. Since the latter type of unpaid work is typically performed by women, this has large implications for understanding gender differences in labor. We discuss these issues as part of ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/female-labor-supply#definitions-measurement", "children": [ { "text": "our entry on Women\u2019s Employment", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{/ref} But labor force surveys only cover residents of a country above a certain age (usually 15), which depending on the country might exclude a non-trivial number of workers.{ref}Only covering resident workers means that any ", "spanType": "span-simple-text" }, { "url": "https://ec.europa.eu/eurostat/cache/digpub/eumove/bloc-2c.html?lang=en#:~:text=In%202019%2C%20the%20largest%20number,and%20Belgium%20(50%20000).", "children": [ { "text": "cross-border workers", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " are excluded. Only covering workers above a certain age means that any child laborers are excluded. While the incidence of child labor has been going down over time, especially in high-income countries, there are still an ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/child-labor", "children": [ { "text": "estimated 265 million working children", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " in the world (almost 17% of the worldwide child population).{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Establishment surveys collect data on employment and working hours as reported by employers.{ref}Employers include businesses, non-profits, some government agencies, and other organizations that pay a wage.{/ref} But because hours are reported by employers, these surveys often only cover paid or contractual hours and exclude self-employment, informal work, and some smaller firms.{ref}Unlike hours actually worked, paid or contractual hours typically include some time ", "spanType": "span-simple-text" }, { "children": [ { "text": "not", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " spent working, such as during sick leave, and fail to include time spent working that wasn't paid or planned, such as overtime.{/ref} On the other hand, establishment surveys provide more detail on the industry of work than other surveys, and are more consistent with how GDP is measured, making them useful for ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~CHE~TWN~GBR~USA®ion=World", "children": [ { "text": "studying labor productivity", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Time use surveys collect data on how individuals spend their time \u2014 ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/time-use-living-conditions#daily-activities-similarities-and-differences-across-countries", "children": [ { "text": "down to the minute", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " \u2014 across a number of activities in a typical day, including paid work.{ref}Activities also include unpaid household work, school, leisure time, eating, and sleeping.{/ref} This level of granularity provides a useful complement to the other surveys, but as a trade-off time use surveys sample fewer people and are conducted less frequently and by fewer countries.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "National accounts", "spanType": "span-simple-text" } ], "type": "heading", "level": 4, "parseErrors": [] }, { "type": "text", "value": [ { "text": "To get the most comprehensive perspective on working hours possible, many countries aggregate data from these surveys with data from other sources \u2014 such as censuses, tax records, and social security registers \u2014 in an economic measurement framework called national accounts.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "National accounts, and the surveys they rely on, are standardized to a degree across countries, which can facilitate international comparisons.{ref}By organizations such as the ", "spanType": "span-simple-text" }, { "url": "https://unstats.un.org/unsd/nationalaccount/sna.asp", "children": [ { "text": "United Nations", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", ", "spanType": "span-simple-text" }, { "url": "https://ilostat.ilo.org/resources/concepts-and-definitions/description-hours-of-work/", "children": [ { "text": "International Labor Organization (ILO)", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", ", "spanType": "span-simple-text" }, { "url": "https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en", "children": [ { "text": "OECD", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ", and ", "spanType": "span-simple-text" }, { "url": "https://ec.europa.eu/eurostat/web/microdata/european-union-labour-force-survey", "children": [ { "text": "Eurostat", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "But these comparisons often have limitations because many countries still implement the methods in different ways. For instance, countries might bring together different data in their national accounts, or aggregate it differently. And many countries don\u2019t have the capacity to conduct comprehensive surveys of their labor force and produce national accounts-based statistics, giving a more limited view of work there.{ref}For further discussion of different sources and their comparability, see the methods guides of the ", "spanType": "span-simple-text" }, { "url": "https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en", "children": [ { "text": "OECD", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and the ", "spanType": "span-simple-text" }, { "url": "https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite", "children": [ { "text": "Total Economy Database", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and the work of ", "spanType": "span-simple-text" }, { "url": "https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344", "children": [ { "text": "Bick, Br\u00fcggemann, and Fuchs-Sch\u00fcndeln (2019)", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "How do researchers reconstruct long-run historical trends?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "Comprehensive, cross-country data on working hours just isn\u2019t available for the years before the mid 20th century. But researchers like ", "spanType": "span-simple-text" }, { "url": "http://www.sciencedirect.com/science/article/pii/S0014498307000058", "children": [ { "text": "Huberman and Minns (2007)", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": "{ref}Huberman, M. and Minns, C. (2007) ", "spanType": "span-simple-text" }, { "url": "http://www.sciencedirect.com/science/article/pii/S0014498307000058", "children": [ { "text": "The times they are not changin\u2019: Days and hours of work in Old and New Worlds, 1870\u20132000.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "children": [ { "text": "Explorations in Economic History.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "{/ref} have been able to fill some of the gap by reconstructing long-run trends for a selection of countries. How do they do it?", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Through often painstaking effort, researchers have been able to find and piece together the relevant historical records that do exist. In the work of Huberman and Minns, one of the key sources for historical data on many countries is a ", "spanType": "span-simple-text" }, { "url": "https://catalog.hathitrust.org/Record/008420895", "children": [ { "text": "report from the US Department of Labor", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " published in 1900.{ref}U.S. Department of Labor (1900) ", "spanType": "span-simple-text" }, { "url": "https://catalog.hathitrust.org/Record/008420895", "children": [ { "text": "Fifteenth Annual Report of the Commissioner of Labor: Wages in Commercial Countries. 2 vols.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " Washington, DC.{/ref} The report compiled the records of many thousands of workers across numerous sectors from establishment surveys in 88 countries and territories. To reconstruct the trends in later years, Huberman and Minns pulled together data from the International Labor Organization, the work of peer researchers, and other sources.{ref}The original sources are: 1870\u20131913: Huberman (2004) [in turn relying on the US Department of Labor Fifteenth Annual Report, 1900]; 1929\u20131938: International Labor Organization (1934\u201339), except for Canada (Ostry and Zaidi, 1972), U.S. (Jones, 1963; Owen, 1988), and Australia (Butlin, 1977); 1950\u20132000: University of Groningen and the Conference Board GGDC Total Economy Database (2005).{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "This was an impressive feat of reconstruction, but historical records like this do have limitations. For instance, as exhaustive as they were, the establishment-level records used by Huberman and Minns still excluded agricultural work, part-time work, and many smaller firms.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "text": [ { "text": "How does the data from different sources compare?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The work by Huberman and Minns is an important example of how researchers often combine and adjust underlying sources to produce one-off cross-country estimates. Another important study is the one of Bick, Br\u00fcggemann, and Fuchs-Sch\u00fcndeln (2019),{ref}Bick, A., Br\u00fcggemann, B., and Fuchs-Sch\u00fcndeln, N. (2019) ", "spanType": "span-simple-text" }, { "url": "https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344", "children": [ { "text": "Hours Worked in Europe and the United States: New Data, New Answers.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "children": [ { "text": "The Scandinavian Journal of Economics.", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": "{/ref} who further standardized labor force surveys to enhance comparability for a selection of countries.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Besides these one-off estimates, several international organizations and research centers aggregate the working hours estimates published by national statistical agencies into cross-country datasets. The two most important datasets come from the ", "spanType": "span-simple-text" }, { "url": "https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#", "children": [ { "text": "OECD", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and the ", "spanType": "span-simple-text" }, { "url": "https://www.rug.nl/ggdc/productivity/pwt/?lang=en", "children": [ { "text": "Penn World Table", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " (PWT). These both draw on national accounts estimates when available, but they can differ in the other sources they use and their method of aggregation.{ref}PWT sources its working hours data from ", "spanType": "span-simple-text" }, { "url": "https://www.conference-board.org/data/economydatabase/total-economy-database-productivity", "children": [ { "text": "The Conference Board\u2019s Total Economy Database", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " (TED). For more details on the underlying sources, see the ", "spanType": "span-simple-text" }, { "url": "https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite", "children": [ { "text": "TED guide", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " and the ", "spanType": "span-simple-text" }, { "url": "https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#", "children": [ { "text": "OECD database", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In the chart you can compare annual working hours data from these four datasets. The data is shown one country at a time \u2014 with France currently selected. You can look at other countries by clicking \u2018Change country\u2019 on the chart, but note that not all sources publish data for every country.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "As expected, there are differences between the sources. In 2000, for instance, Bick et al. estimates 1,642 hours of work for French workers, OECD estimates 1,558 hours, PWT estimates 1,550 hours, and Huberman and Minns estimates 1,443 hours. These differences are due to the use of different underlying sources and methods. Bick et al. use only labor force surveys; the others all rely primarily on national accounts data, but which nonetheless still have differences.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "It\u2019s also clear that these differences between sources are quite small when compared to the huge changes over the longer run. The difference between sources in 2000 is at most 200 hours, while the historical data from Huberman and Minns shows that from 1870 to 2000 annual working hours in France decreased by ", "spanType": "span-simple-text" }, { "children": [ { "text": "1,725 hours", "spanType": "span-simple-text" } ], "spanType": "span-italic" }, { "text": " (from 3,168 to 1,443 hours).", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "url": "https://ourworldindata.org/grapher/compare-sources-working-hours", "type": "chart", "parseErrors": [] }, { "text": [ { "text": "What does this tell us about the study of working hours?", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "text", "value": [ { "text": "The analysis here shows that working hours data can have limitations \u2014 due to differences in the sources or the way the method is implemented \u2014 but that what these matter for our interpretation of the data depends on the context.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In a context where precise comparisons of similar countries is important, smaller differences between sources can really matter. This is why to compare recent working hours levels in the US and Europe, Bick et al. used only labor force surveys, which they standardized even further to maximize cross-country comparability. But as a trade-off, it was only possible to look at a small selection of richer countries.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "In a context where we want to focus on a larger scale \u2014 such as the ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/working-more-than-ever", "children": [ { "text": "long-run historical trends", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " we see in the chart \u2014 the limitations of the sources are not large enough to undermine our conclusions.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "Large international datasets like PWT do not have the highest levels of cross-country comparability, but they allow us to look at many more countries across the world and uncover broad and important trends, such as the ", "spanType": "span-simple-text" }, { "url": "https://ourworldindata.org/rich-poor-working-hours", "children": [ { "text": "large differences in working hours between the richest and poorest countries", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{ref}We gain further confidence in these conclusions when they are echoed by research that focuses only on more standardized, comparable sources for a necessarily smaller set of countries, as in the work by ", "spanType": "span-simple-text" }, { "url": "https://www.aeaweb.org/articles?id=10.1257/aer.20151720", "children": [ { "text": "Bick, Fuchs-Sch\u00fcndeln, and Lagakos (2018)", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ".{/ref}", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "PWT and OECD are also useful in contexts where we want an exhaustive picture of the trends in individual countries, since they are often based on national accounts that bring together data from many sources to give a comprehensive perspective on working hours.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "text": "The data on working hours isn\u2019t perfect, and it\u2019s important to understand the limitations, but it can still tell us a lot about our lives and the world.", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "Data Sources", "spanType": "span-simple-text" } ], "type": "heading", "level": 1, "parseErrors": [] }, { "type": "horizontal-rule", "parseErrors": [] }, { "text": [ { "text": "OECD Time Use Database and Gender Data Portal", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Time spent on various activities per day, by sex", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " OECD countries plus China, India, and South Africa", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " from 1998 onwards", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "url": "https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE", "children": [ { "text": "Time Use Database", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": "; ", "spanType": "span-simple-text" }, { "url": "https://www.oecd.org/gender/data/balancingpaidworkunpaidworkandleisure.htm", "children": [ { "text": "Gender Data Portal", "spanType": "span-simple-text" } ], "spanType": "span-link" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Centre for Time Use Research", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Time spent on various activities per day; self-reported levels of enjoyment", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " 30 countries across the world", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " from 1964 onwards", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "url": "https://www.timeuse.org/", "children": [ { "text": "https://www.timeuse.org/", "spanType": "span-simple-text" } ], "spanType": "span-link" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "US Bureau of Labor Statistics", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Time spent with others per day", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " United States", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " from 2003 onwards", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "url": "https://www.bls.gov/tus/datafiles-0319.htm", "children": [ { "text": "https://www.bls.gov/tus/datafiles-0319.htm", "spanType": "span-simple-text" } ], "spanType": "span-link" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "IPUMS USA", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Percentage of Americans living alone, by age", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " United States", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " from 1900 onwards", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "url": "https://ipums.org/projects/ipums-usa/d010.v8.0", "children": [ { "text": "https://ipums.org/projects/ipums-usa/d010.v8.0", "spanType": "span-simple-text" } ], "spanType": "span-link" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Dotti Sani and Treas (2016)", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Time parents spent with children per day", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " United States and select countries in Europe", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " 1965\u20132012", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Dotti Sani, G. M., and Treas, J. (2016). ", "spanType": "span-simple-text" }, { "url": "https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Dotti+Sani+and+Treas+2016+Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965%E2%80%932012&btnG=", "children": [ { "text": "Educational gradients in parents\u2019 child\u2010care time across countries, 1965\u20132012", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". Journal of Marriage and Family.", "spanType": "span-simple-text" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Sayer, Bianchi, and Robinson (2004)", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Time parents spent on primary childcare activities per day", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " United States", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " 1965\u20131998", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Sayer, L., Bianchi, S., and Robinson, J. (2004). ", "spanType": "span-simple-text" }, { "url": "https://www.journals.uchicago.edu/doi/pdf/10.1086/386270", "children": [ { "text": "Are Parents Investing Less in Children? Trends in Mothers\u2019 and Fathers\u2019 Time with Children", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". American Journal of Sociology.", "spanType": "span-simple-text" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Huberman and Minns (2007)", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Annual hours of full-time production workers (male and female) in non-agricultural activities; days off from work for vacations and holidays", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " United States, Australia, Canada, and select countries in Europe", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " 1870\u20132000", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Huberman, M. and Minns, C. (2007). ", "spanType": "span-simple-text" }, { "url": "https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!", "children": [ { "text": "The times they are not changin\u2019: Days and hours of work in Old and New Worlds, 1870\u20132000.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " Explorations in Economic History.", "spanType": "span-simple-text" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Penn World Table", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Average annual hours worked by persons engaged; number of persons engaged; real and PPP-adjusted GDP in US millions of dollars", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Countries across the world", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " from 1950 onwards", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "url": "https://www.rug.nl/ggdc/productivity/pwt/", "children": [ { "text": "https://www.rug.nl/ggdc/productivity/pwt/", "spanType": "span-simple-text" } ], "spanType": "span-link" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Eurostat", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Time use survey data", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Europe", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " from 1998 onwards", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "url": "https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#", "children": [ { "text": "https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#", "spanType": "span-simple-text" } ], "spanType": "span-link" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Ramey and Francis (2009)", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Self-reported enjoyment of various activities; time spent on various activities, by sex and age; days of work lost to sickness", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " United States", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " 1900\u20132005", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Ramey, V. A., and Francis, N. (2009). ", "spanType": "span-simple-text" }, { "url": "https://www.aeaweb.org/articles?id=10.1257/mac.1.2.189", "children": [ { "text": "A century of work and leisure", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": ". American Economic Journal: Macroeconomics.", "spanType": "span-simple-text" } ], "parseErrors": [] } ], "parseErrors": [] }, { "text": [ { "text": "Costa (2000)", "spanType": "span-simple-text" } ], "type": "heading", "level": 3, "parseErrors": [] }, { "type": "list", "items": [ { "type": "text", "value": [ { "children": [ { "text": "Data:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Working hours, by sex; number of working days; wages", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Geographical coverage:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " United States", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Time span:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " 1890\u20131991", "spanType": "span-simple-text" } ], "parseErrors": [] }, { "type": "text", "value": [ { "children": [ { "text": "Available at:", "spanType": "span-simple-text" } ], "spanType": "span-bold" }, { "text": " Costa, D. L. (2000). ", "spanType": "span-simple-text" }, { "url": "https://www.jstor.org/stable/10.1086/209954?seq=1", "children": [ { "text": "The Wage and the Length of the Work Day: From the 1890s to 1991.", "spanType": "span-simple-text" } ], "spanType": "span-link" }, { "text": " Journal of Labor Economics.", "spanType": "span-simple-text" } ], "parseErrors": [] } ], "parseErrors": [] } ], "type": "linear-topic-page", "title": "Time Use", "authors": [ "Esteban Ortiz-Ospina", "Charlie Giattino", "Max Roser" ], "excerpt": "How do people across the world spend their time? How do daily activities differ across countries, and how do these differences matter for people\u2019s lives? Explore data and research on time use.", "dateline": "November 29, 2020", "subtitle": "How do people across the world spend their time? How do daily activities differ across countries, and how do these differences matter for people\u2019s lives? Explore data and research on time use.", "sticky-nav": [], "sidebar-toc": true, "featured-image": "thumbnailTime-Use-by-Country-OECD.png" }, "createdAt": "2020-12-01T15:07:17.000Z", "published": false, "updatedAt": "2023-01-16T21:06:15.000Z", "revisionId": null, "publishedAt": "2020-11-29T21:12:40.000Z", "relatedCharts": [], "publicationContext": "listed" } |
{ "errors": [ { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag image" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag image" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag image" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag quote" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected elements in list item", "details": "Found 1 elements" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected elements in list item", "details": "Found 1 elements" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" }, { "name": "unexpected wp component tag", "details": "Found unhandled wp:comment tag list" } ], "numBlocks": 228, "numErrors": 17, "wpTagCounts": { "html": 13, "list": 12, "image": 3, "quote": 1, "column": 10, "columns": 5, "heading": 52, "paragraph": 151, "owid/summary": 1, "owid/additional-information": 2 }, "htmlTagCounts": { "p": 152, "h2": 5, "h3": 8, "h4": 33, "h5": 3, "h6": 3, "ul": 14, "div": 15, "figure": 3, "iframe": 12, "blockquote": 1 } } |
2020-11-29 21:12:40 | 2024-02-16 14:22:41 | [ "Esteban Ortiz-Ospina", "Charlie Giattino", "Max Roser" ] |
How do people across the world spend their time? How do daily activities differ across countries, and how do these differences matter for people’s lives? Explore data and research on time use. | 2020-12-01 15:07:17 | 2023-01-16 21:06:15 | https://ourworldindata.org/wp-content/uploads/2020/12/thumbnailTime-Use-by-Country-OECD.png | {} |
Time is the ultimate limited resource. Every single one of us has exactly the same “time budget”: 24 hours per day, 365 days per year – 8,760 hours [every year of our lives](https://owid.cloud/life-expectancy). How do we actually spend our time? There are many commonalities across the world: we all sleep, work, eat, and enjoy leisure time. But there are also important differences in the freedom people have to spend time on the things they value most. Studying how people across the world spend their time provides an important perspective for understanding living conditions, economic opportunities, and general well-being. Here we present the data on time use. We explore how it differs across countries and over time and how these differences matter for people’s lives. --- # Time use from the perspective of an average day --- ## How do people across the world spend their time and what does this tell us about living conditions? Sleep, work, eat, leisure – at a high level most of us spend time on similar activities. But just how similar are the daily activities of people across the world? This is something worth considering, not just to serve our curiosity but because differences in the way we spend time give us meaningful perspectives on living conditions, economic opportunities and general well-being. Here we take a look at the data. We explore some of the key patterns that emerge from cross-country time use surveys, and then dig deeper to understand how these differences matter for people’s well-being. ### Daily activities: similarities and differences across countries In the chart here we compare average time spent across a number of common activities. The data comes from the OECD and brings together estimates from time diaries where respondents are asked to record the sequence of what they did over a specific day, as well as from general questionnaires where respondents are asked to recall the amount of time spent on different activities on a specific day in the previous week.{ref}The ‘time-diary method’ is generally more reliable and allows a richer analysis of routines, because it measures not only aggregate times but also sequences and clock-times. Time-diary data is less common, but it is available for some countries from the [Multinational Time Use Study](https://www.timeuse.org/mtus). We explore time-use ‘tempograms’ from the MTUS in a forthcoming companion post.{/ref} The first thing that jumps out from this chart is that there are indeed many similarities across countries. This is not surprising – most of us try to split our days into “work, rest and fun”, and so there are some predictable patterns. We spend the most time working and sleeping; and paid work, housework, leisure, eating and sleeping take together 80-90% of the 1440 minutes that we all have available every day. But if we look closely, we also see some important differences. Consider sleeping, for example. From this sample of countries, South Koreans sleep the least – averaging 7 hours and 51 minutes of sleep every day. In India and the US, at the other end of the spectrum, people sleep an hour more on average. Work is another important activity where we see large differences. Countries are sorted by paid work hours in the chart – from highest to lowest. In China and Mexico people spend, on an average day, almost twice as much time on paid work as people in Italy and France do. This is a general pattern: [People in richer countries can afford to work less](https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours). Keep in mind that this chart shows the average for all people in the working age bracket, from 15 to 64 years, whether they are actually employed or not.{ref}Because these estimates include people who are not employed they are much lower than the estimates of working hours _per__worker_[we present elsewhere](https://ourworldindata.org/time-use#time-spent-working). The estimates also differ because of differences in the sources: time-use surveys compared to labor force surveys and national accounts data.{/ref} Differences in [demographics](https://ourworldindata.org/age-structure), [education](https://ourworldindata.org/global-education) and [economic prosperity](https://ourworldindata.org/economic-growth) all contribute to these inequalities in work and time use. But what’s clear in the chart here is that there are also some differences in time use that are not well explained by economic or demographic differences. In the UK, for example, people spend more time working than in France; but in both countries people report spending a similar amount of time on leisure activities. Cultural differences are likely to play a role here. The French seem to spend much more time eating than the British – and in this respect the data actually goes in line with stereotypes about food culture. People in France, Greece, Italy and Spain report spending more time eating than people in most other European countries. The country where people spend the least time eating and drinking is the USA (63 minutes). ##### How do people spend their time?{ref}OECD (2020) [Time Use Database.](https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE){/ref} <Image filename="Time-Use-by-Country-OECD.png" alt=""/> [Download the underlying data for this chart (.xlsx)](https://owid.cloud/app/uploads/2020/12/Time-Use-in-OECD-Countries-OECD.xlsx) ### Going beyond averages: The gender gap in leisure time Going beyond national averages reveals important within-country inequalities. The gender gap in leisure time, for example, is a key dimension along which large inequalities exist. The chart here relies on the same time-use data described above, but shows total leisure time for men and women separately. Time for men is shown on the horizontal axis, while time for women appears on the vertical axis. The dotted diagonal line denotes ‘gender parity’, so the further away a country is from the diagonal line, the larger the difference between men and women. As we can see, in all countries the average leisure time for men is higher than for women – all bubbles are below the diagonal line – but in some countries the gaps are much larger. In Norway the difference is very small, while in Portugal men report almost 50% more leisure time than women. A key factor driving these differences in leisure time is the gender gap in unpaid work. As we explain in detail in a [companion post](https://ourworldindata.org/female-labor-supply#informal-work-unpaid-care-work), women are responsible for a disproportionate amount of unpaid work, and have less leisure as a result.{ref}If you want to dig deeper you can explore gender differences across all other activities directly from our source, via the [OECD Data Portal](https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE). And you can read more about within-country inequalities in time use along other dimensions, such as income and education, in [this Brookings Paper](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=The+Middle+Class+Time+Squeeze%2C+Economic+Studies+at+Brookings%2C+Sawhill+and+Guyot+%282020%29&btnG=), where the authors focus on the ‘middle class time squeeze’ in the US. See: Sawhill, I. V., & Guyot, K. (2020). The Middle Class Time Squeeze. Economic Studies at Brookings. Brookings Institution.{/ref} <Chart url="https://ourworldindata.org/grapher/minutes-spent-on-leisure?tab=chart&stackMode=absolute&time=latest&country=®ion=World"/> ### Why should we care about differences in time use? Every single one of us has the same “time budget”: 24 hours per day and 365 days per year. But of course not all of us can choose to spend time on the activities that we enjoy most. Differences in our freedom to allocate time to the things we enjoy is the main reason why time-use data is important for studying living conditions. In the UK, researchers from the [Centre for Time Use Research](https://www.timeuse.org/) linked time-use diaries with the respondents’ assessments of enjoyment, on a scale from 1 to 7, to better understand the connection between time use and well-being. The chart here, which we’ve adapted from the book ‘What We Really Do All Day’_, _by professors Jonathan Gershuny and Oriel Sullivan, shows the results. The estimates correspond to average reported levels of enjoyment for each activity, with confidence intervals.{ref}The underlying data comes from time-use diaries where respondents are asked to record the sequence of what they do over a specific day, and how much they enjoy each ‘episode’ (i.e. what they do) on a scale from 1 to 7. All episodes reported are then coded and grouped into similar activities. To arrive at the mean enjoyment scores, the authors multiply the duration of each episode where the activity category concerned is the primary activity recorded, by the enjoyment level to arrive at the total enjoyment score for that episode. Then they sum these total enjoyment scores for each category of activity across the day, and finally divide these daily enjoyment total scores for each activity by the amount of time devoted to the activity. In this way, they arrive at an appropriately weighted mean enjoyment level for each activity across all those who engage in it. For more details see Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref} We see that the most enjoyed activities involve rest or leisure activities such as eating out, sleeping, going to sports events, playing computer games or attending cultural performances. The activities receiving the lowest ratings include doing school homework, looking for a job, or doing housework. The activity where people show the greatest variation in enjoyment is working a “Second Job”. This likely reflects the difference between people who work a second job because they want to, and those who work a second job because they _have__to_. So what do we learn from this? First, we learn that the enjoyment of activities is, at least to some degree, predictable and stable. This means we can take activity groups and make meaningful comparisons across groups of people. Economists, for example, [often classify](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Ramey%2C+V.+A.%2C+%26+Francis%2C+N.+%282009%29.+A+century+of+work+and+leisure.+American+Economic+Journal%3A+Macroeconomics%2C+1%282%29%2C+189-224.&btnG=) any activity with an enjoyment level below work as a “non-leisure activity”, in order to measure trends in leisure across people and time.{ref}You find a very clear and complete explanation of this in Ramey, V. A., & Francis, N. (2009). A century of work and leisure. American Economic Journal: Macroeconomics, 1(2), 189-224.{/ref} But beyond this, and more importantly, this confirms that time-use is informative about well-being. The fact that there is a very clear and predictable pattern in the enjoyment of activities suggests that differences in time use do, indeed, give us meaningful perspectives on living conditions and economic opportunities. In countries where people do more paid and unpaid work, and have less time for leisure, their enjoyment – and [happiness and life satisfaction](https://ourworldindata.org/happiness-and-life-satisfaction) – levels are likely to be lower. ##### How do people rate the enjoyment of different activities?{ref}Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref} <Image filename="Enjoyment-Level-of-Time-Use.png" alt=""/> --- # Time use from the perspective of the life cycle --- ## Who do we spend time with across our lifetime? As we go through life we build personal relationships with different people – family, friends, coworkers, partners. These relationships, which are [deeply important to all of us](https://ourworldindata.org/social-connections-and-loneliness#social-connections), evolve with time. As we grow older we build new relationships while others transform or fade, and towards the end of life many of us spend a lot of time alone. Taking the big picture over the entire life course: Who do we actually spend our time with? ### From adolescence to old age: who do we spend our time with? To understand how social connections evolve throughout our lives we can look at survey data on _how much_ time people spend with others, and _who_ that time is spent with. The chart here shows the amount of time that people in the US report spending in the company of others, based on their age. The data comes from time-use surveys, where people are asked to list all the activities that they perform over a full day, and the people who were there during each activity. We currently only have data with this granularity for the US – time-use surveys are common across many countries, but what is special about the US is that respondents of the American Time Use Survey are asked to list everyone who was present for each activity. <Chart url="https://ourworldindata.org/grapher/time-spent-with-relationships-by-age-us?tab=chart&stackMode=absolute®ion=World"/> The numbers in this chart are based on averages for a cross-section of the American society – people are only interviewed once, but we have brought together a decade of surveys, tabulating the average amount of time that survey respondents of different ages report spending with other people_._{ref}When interpreting this chart it’s important to bear in mind that the relationships used to categorise people are not exhaustive (i.e., survey respondents could also list being with people who didn’t fit any of the listed categories, or for whom a relationship was unclear or unknown – we do not count these instances in the estimates). Additionally, time spent with multiple people can be counted more than once; so attending a party with friends and your spouse, for example, would show up for both “friends” and “partner” in our estimates. The implication is that companion categories cannot be stacked to add up total time spent in the company of others.{/ref} #### Who we spend our time with changes a lot over the course of life When we’re young – particularly in our teens – we spend a lot of our time with friends, parents, siblings and extended family. As we enter our 20s, time with friends, siblings and parents starts to drop off quickly. Instead, we start spending an increasing amount of time with partners and children. (The chart shows an average across Americans, so for those that have children the time spent with children is even higher, since the average is pulled down by those without children.) As the chart shows, this continues throughout our 30s, 40s and 50s – over this period of their life, Americans spend much of their time with partners, children and, unsurprisingly, co-workers. For those 60 and older, we see a significant drop-off in time spent with co-workers. This makes sense, considering many people in the US [enter retirement in their mid 60s](https://ourworldindata.org/grapher/average-effective-retirement-men). We see that this time is partly displaced by more time with partners. In terms of the diversity of interactions, this chart suggests that the number of people with whom we interact is highest around 40, but then things change substantially after that. And this is perhaps the most conspicuous trend in the chart: above 40, people spend an increasing amount of time alone. ### Older people spend a lot of time alone Older people spend a large amount of time alone and it is understandable why – time spent alone increases with age because this is when health typically deteriorates and people lose relatives and friends. Indeed, many people who are older than 60 live alone as this chart shows clearly: living alone is particularly common for older adults. Today nearly 4 out of every 10 Americans who are older than 89 years old live alone. Another interesting point here is that the share of people across all age groups who live alone has been increasing over time. This is part of a more general global trend – if you want to read more about the global ‘rise of living alone’, we provide a detailed account of this trend across countries in a [companion post](http://ourworldindata.org/living-alone). <Chart url="https://ourworldindata.org/grapher/percentage-of-americans-living-alone-by-age"/> ### Older people spend more time alone, but this doesn’t necessarily mean they’re lonely The data shows that as we become older we tend to spend more and more time alone. What’s more, the data also shows that older people today spend more time alone than older people did in the past. We might think older people are therefore more lonely – but this is not necessarily the case. Spending time alone is not the same as feeling lonely. This is a point that is well recognised by researchers, and one which has been confirmed empirically across countries. Surveys that ask people about living arrangements, time use, and feelings of loneliness find that solitude, by itself, is not a good predictor of loneliness. You can read our overview of the evidence in [this post](http://ourworldindata.org/lonely-not-alone). So, what about loneliness? If we focus on self-reported loneliness, there is [little evidence](https://ourworldindata.org/loneliness-epidemic#loneliness-trends-over-time-are-people-lonelier-today-than-in-the-past) of an upward trend over time in the US; and importantly, it’s [not the case](https://ourworldindata.org/social-connections-and-loneliness#do-we-become-lonelier-as-we-get-older) that loneliness keeps going up as we become older. In fact a [recent study](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Are+U.S.+older+adults+getting+lonelier%3F%3A+Age%2C+period%2C+and+cohort+differences.+Psychology+and+Aging.&btnG=) based on surveys that track the same individuals over time found that after age 50 – which is the earliest age of participants in the analysis – loneliness tended to decrease, until about 75, after which it began to increase again. Taking the evidence together, the message is not that we should be sad about the prospect of aging, but rather that we should recognise the fact that social connections are complex. We often tend to look at the amount of time spent with others as a marker of social well-being; but the quality of time spent with others, and our expectations, matter even more for our feelings of connection and loneliness. ## Additional information In the chart where we plot the amount of time that people in the US report spending in the company of others, it’s important to keep in mind that we are taking a look at a cross-section of society. This means that we are actually seeing the result of two underlying trends. On one hand, we see the effect of aging on social connections (we relate to different people and reallocate time as we go through different stages of life), but we also see the effect of cohort trends (compared to people in the past, today’s older generations in the US tend to be [healthier](https://ourworldindata.org/grapher/life-expectancy?tab=chart&country=JPN~GBR~IND~ETH~OWID_WRL~KOR~ZAF~USA®ion=World) and [richer](https://ourworldindata.org/grapher/average-real-gdp-per-capita-across-countries-and-regions?tab=chart&country=Western%20Offshoots~Western%20Europe~Western%20Asia~Eastern%20Europe~OWID_WRL~Latin%20America~East%20Asia~Africa~USA), and might also have different expectations, preferences and opportunities). Disentangling these two effects is difficult, so it is important to keep in mind that at least some of the age gradients we observe might be partly explained by cohort changes, rather than life-cycle trends. This is why it’s important to rely not only on cross-sectional data, but also on surveys that track the same individuals over time. --- # Parents’ time with children --- ## Are parents spending less time with their kids? Over the last 50 years many countries have seen large changes in family structures and the institution of marriage. These changes – which include a [rise in single-parenting](https://ourworldindata.org/marriages-and-divorces#single-parenting-is-common-and-in-many-countries-it-has-increased-in-recent-decades) and a large increase in the [share of women working outside the home](https://ourworldindata.org/female-labor-supply#female-participation-in-labor-markets-grew-remarkably-in-the-20th-century) – have made some people worry that children might be getting ‘short-changed’, because parents are not spending as much time with them. In 1999, for example, a [report](https://clintonwhitehouse4.archives.gov/media/pdf/famfinal.pdf) from the Council of Economic Advisors in the US analyzed trends over the second half of the 20th century and concluded: _“The increase in hours mothers spend in paid work, combined with the shift toward single-parent families, resulted in families on average experiencing a decrease of 22 hours a week (14 percent) in parental time available outside of paid work that they could spend with their children.”_ The line of thought behind these concerns is that changes to the structure of families and work have meant that children spend less time with parents, because parents – particularly mothers – spend less time at home. Here we review the evidence and show that this reasoning is flawed. As we explain, in the US and many other rich countries parents spend more time with their kids today than 50 years ago. Equating ‘mother time at home' with 'children’s time with parents’ is a huge and unhelpful oversimplification. ### Mothers and fathers spend an increasing amount of time with kids: evidence from time-use surveys The chart here shows the time that mothers and fathers spend with their children. This is measured using time-use diaries where parents record the amount of time that they spend on various activities, including child care. These estimates, which are sourced from a [paper](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=) published in 2016 by sociologists Giulia Dotti Sani and Judith Treas, are disaggregated by education levels and are adjusted to account for demographic differences between countries. We explain [below](https://ourworldindata.org/parents-time-with-kids#digging-deeper-raw-vs-population-adjusted-trends-in-time-use) in more detail why these adjustments are important.{ref}The full reference for the source paper is: Dotti Sani, G. M., & Treas, J. (2016). [Educational gradients in parents' child‐care time across countries](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=), 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096. {/ref} As we can see here, there has been a clear increase in the amount of time parents spend with their children over the last 50 years. This is true for both fathers and mothers, and holds across almost all education groups and countries. The two exceptions are France, where mothers’ time has declined (from a very comparatively high level); and Slovenia where it has remained roughly constant among non-university-educated parents. In terms of within-country inequalities we also see two clear patterns. First, in all countries mothers spend more time in child care activities than fathers. The differences are large and persistent across education groups. In some countries, such as Canada, France and the US, this gender gap has shrunk; while in other countries such as Denmark and Spain, the gap has widened. Second, in all countries there is a positive ‘educational gradient’, meaning more educated parents tend to spend more time with their children. In many countries this gradient increased, and nowhere did it decline. ##### How much time do parents spend with their children per day?{ref}Dotti Sani, G. M., & Treas, J. (2016). [Educational gradients in parents' child‐care time across countries](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=), 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref} <Image filename="Parent-time-spent-with-children-2.png" alt=""/> ## Additional information The main chart in this post, where we show the amount of time that fathers and mothers spend with their children, plots ‘population-adjusted trends’. This means that the estimates are not what you directly observe in the surveys, but rather the output of a statistical model that takes these surveys as an input but corrects for differences in key factors such as age, partnership status, number of children and employment. This is why the trends look so smooth – rather than plotting the raw survey averages for each year, the chart shows ‘modelled estimates’ after holding compositional factors constant. These adjustments are helpful for the analysis, because there are large differences between population groups within countries (e.g. single parents are not able to spend as much time with their children); and since the relative size of these groups is changing, the aggregate trend that we observe is affected by these compositional changes (e.g. in some countries single-parenting is increasing, while in others it is decreasing). So how much difference do these “adjustments” make? After all, we are also interested in the _observed_ amount of time that parents spend with their children, and not only the values that are adjusted for population changes. What would the trends look like if we just plotted the raw survey estimates? The next chart shows this with data for the US. As we can see, the trend goes in the same direction for both fathers and mothers. <Chart url="https://ourworldindata.org/grapher/us-child-care-time-sex"/> ### Parents’ time with children, as reported by children, has also increased The analysis above looked at time-use diaries from _parents._ But children may experience time and attention differently from adults. Is there evidence that children also feel they are getting more time with their parents? In a [study](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=) published in 2001 in the journal _Demography_, John Sandberg and Sandra Hofferth analyzed two surveys with data from child time-use diaries in the US, and found that children reported spending about 4.3 more hours per week with mothers, and 3 more hours per week with fathers in 1997 versus 1981.{ref}The estimates from Sandberg and Hofferth correspond to weekly hours among children ages 3-12. The estimates cover both ‘engaged time’ and other activities where parents are not engaged but children say they are present. When looking jointly at both parents, Sandberg and Hofferth find that time spent with either parent remains roughly unchanged, which suggests part of the increase corresponds to an intra-household reallocation of time. In the paper you find more details and discussion, including a detailed analysis of trends for more narrow population groups and possible demographic and behavioral drivers: Sandberg, J. F., & Hofferth, S. L. (2001). [Changes in children’s time with parents: United States, 1981–1997](https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=). Demography, 38(3), 423-436. {/ref} The data is sparse because time-use diaries for children are not very common; but the evidence that is available is consistent with what we’ve seen above: Children in the US agree that they’re spending more time with their parents than in the past. ### Social norms affect how we allocate time The amount of time that parents spend with their children is determined by many factors, and working hours outside the home are only one of them. Choices, parenting norms, family size, and relationship ideals – for men and women, parents and children – matter a lot. Several academic studies have dissected the data from the US in an attempt to disentangle the relative importance of different underlying factors. The conclusion from these studies is that the reason why we see an increase in the amount of time that American parents spend with kids is that many families, particularly those who are well-off, have been able to undergo changes in their routines and the allocation of tasks and time within the household, in order to spend more time with their children. Parents have been able to undergo ‘behavioral changes’ that have more than compensated for ‘structural changes’ that could have pushed in the opposite direction.{ref}For an overview of the literature see (i) Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents' child‐care time across countries, 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096. (ii) Sayer, L. C., Bianchi, S. M., & Robinson, J. P. (2004). Are parents investing less in children? Trends in mothers’ and fathers’ time with children. American journal of sociology, 110(1), 1-43. (iii) Sandberg, J. F., & Hofferth, S. L. (2001). Changes in children’s time with parents: United States, 1981–1997. Demography, 38(3), 423-436.{/ref} Employment, and in particular [working hours](https://ourworldindata.org/working-hours) outside the home, determine a range of possible arrangements, but individual preferences and social conventions determine the actual allocation of time within that range of feasible options. Preferences and social conventions are dynamic and change with time. There is a large academic literature that emphasizes the importance of parent-child interactions for children’s development, and it seems natural to allow for the possibility that in recent decades parenting norms have undergone considerable change in response to this scientific evidence.{ref}The parenting guide from the National Health Insurance in the UK, for example, says _“Parents should make the time to play with a first or only child. And while brothers and sisters are natural playmates, parents can also play an active role in siblings' games...If you're pressed for time as a parent, it's a good idea to find ways to involve your child in what you're doing – even the housework._” You find a review of the academic literature on the importance of parent-child interactions for children’s development in Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents' child‐care time across countries, 1965–2012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref} ### Why should we care? The available data shows that in many countries parents spend more time with their kids than ever before, and this is partly due to changes in social conventions and economic progress, especially declining working hours. There are large differences between countries, and also large inequalities across different population groups within each country; but overall the trends tend to go in the same direction. Of course, there’s a lot that is not being captured by these aggregate statistics. There are important distributional issues, and there is also clearly more to parents’ and children’s welfare than ‘total time spent together’. But despite these limitations, the available research and data still offers a clear lesson: contrary to what some people fear, it’s not the case that children in rich countries are being systematically ‘shortchanged’ by widespread changes in family structures. Parents are spending more time with their kids than they used to, and this matters because parent-child interactions are important for childhood development. --- # Time spent working --- ## Are we working more than ever? In today’s hustle and bustle world, it’s easy to assume that we are all, by and large, working more than ever. But is that really the case? As we explain in detail below, the research on the history of working hours shows that this is not the case. The available data shows that in the 19th century people across the world used to work extremely long hours, but in the last 150 years working hours have decreased substantially, particularly in today’s richest countries. ### Working hours per worker have declined after the Industrial Revolution The chart here shows average working hours since 1870 for a selection of countries that industrialized early. You can add or remove countries by clicking Add country on the chart. We show annual totals, so the trends account for changes in both the length of working days as well as the number of days worked through the year. The data comes from research by the economic historians Michael Huberman and Chris Minns, who have brought together evidence from historical records, National Accounts data, and other sources.{ref}In this chart we have taken the original data published by [Huberman & Minns (2007)](https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!) and extended coverage using an updated vintage of the Penn World Table (PWT), which is in turn based on the same underlying source that Huberman and Minns used for all data since 1950, the Total Economy Database. You can find more details and links to our sources in the ‘Sources’ tab of the chart.{/ref} The chart shows that average working hours declined dramatically for workers in early-industrialized economies over the last 150 years. In 1870, workers in most of these countries worked more than 3,000 hours annually — equivalent to a grueling 60–70 hours each week for 50 weeks per year. But we see that today those extreme working hours have been roughly cut in half. In Germany, for example, annual working hours decreased by nearly 60% — from 3,284 hours in 1870 to 1,354 hours in 2017 — and in the UK the decrease was around 40%. Before this revolution in working hours people worked as many hours between January and July as we work today in an entire year.{ref}A key point to keep in mind when interpreting these trends is that they refer to working hours _per worker_, which is different from working hours _per person. _The per person measure corresponds to working hours per worker multiplied by the employment rate. Hence, changes in employment patterns — such as the historical rise of female participation in paid employment in these countries — mean that changes in hours per worker do not translate directly into changes in hours per person.{/ref} For many countries in the chart we don’t have long-run series going back to the 19th century. But we do have evidence from other historical records from 1870–1900 that in many of those countries workers also used to work extremely long hours.{ref}A [study by Michael Huberman and Frank Lewis](https://www.econstor.eu/handle/10419/67824) reconstructed estimates of working hours in 1870 and 1900 for 48 countries across six continents using data from worker records kept by individual business establishments. They drew from a collection of records published by the US Department of Labor in 1900, and found substantial variation, but very high working hours for many non-industrialized countries. They found for example that in 1870, Colombia, Uruguay and Brazil had similar average working hours per worker as the US. The full reference of the paper is Huberman, M., & Lewis, F. D. (2007). Bend it like Beckham: Hours and wages across forty-eight countries in 1900 (No. 1229). Queen's Economics Department Working Paper.{/ref} For those countries with long-run data in this chart we can see three distinct periods: From 1870–1913 there was a relatively slow decline; then from 1913–1938 the decline in hours steepened in the midst of the powerful sociopolitical, technological, and economic changes that took shape with World War I, the Great Depression, and the lead-up to World War II; and then after an uptick in hours during and just after World War II, the decline in hours continued for many countries, albeit at a slower pace and with large differences between countries.{ref}The increase in hours between 1938 and 1950 in the chart for some countries is due in part to the uptick during and just after World War II, but also plausibly due in part to differences in the source data and methodology.{/ref} <Chart url="https://ourworldindata.org/grapher/annual-working-hours-per-worker"/> ### In recent decades working hours have continued to decline in many countries, but there are large differences between countries Zooming in to the last 70 years and looking at other countries beyond those who industrialized early, the data reveals a continued decline in working hours for many countries but also large differences between countries. In the chart here we zoom in to the period since 1950 and we change the selection of countries to highlight some of the diversity in trends. For some countries, such as Germany, working hours have continued their steep historical decline; while for other countries, such as the US, the decline has leveled off in recent decades. In some countries we see an inverted U-shaped pattern. In South Korea, for example, hours rose dramatically between 1950 and 1980 before falling again since the mid 1980s. And in other countries we see no recent declines — in China, for example, hours actually rose in the 1990s and early 2000s before leveling off in recent years. <Chart url="https://ourworldindata.org/grapher/annual-working-hours-per-worker?tab=chart&stackMode=absolute&time=1950..latest&country=DEU~USA~BRA~CHN~KOR~IND®ion=World"/> ### Shorter work days, but also more holidays and vacations The decline in annual working hours described above has come from fewer working hours each day, as well as fewer working days each week and fewer working weeks in the year. In a [paper](https://www.jstor.org/stable/10.1086/209954) analyzing historical data for the US, the economist Dora Costa summarizes the evidence:{ref}Costa, D. L. (2000). [The Wage and the Length of the Work Day: From the 1890s to 1991.](https://www.jstor.org/stable/10.1086/209954?seq=1)_Journal of Labor Economics_, 18(1).{/ref} -- undefined As Costa notes, workers had regular days off each week: one day off (usually Sunday) from at least the 1880s until around the 1940s, when two days off became more typical. In addition to regular days off each week, workers across early-industrialized countries had days off from work for vacations and national holidays. This is shown in the chart here, which again relies on research from Huberman and Minns. The chart shows that days of vacation and holidays increased over the period from 1870–2000. The Netherlands is a stark example — workers there saw an increase from four days off for vacations and holidays in 1870 to almost 38 days off in 2000. The declines in the length of the work day and the number of working days have been driven by several factors, including [increases in productivity](https://ourworldindata.org/rich-poor-working-hours) and the adoption of regulations that limit working hours. We discuss these and other key drivers behind working hours trends across countries and time in a series of forthcoming posts.{ref}In our [first post in the series](https://ourworldindata.org/rich-poor-working-hours), we discuss how increases in labor productivity have driven a rise in incomes and a decrease in working hours.{/ref} <Chart url="https://ourworldindata.org/grapher/days-of-vacation-and-holidays"/> ### Why should we care? The evidence presented here comes from decades of work from economic historians and other researchers. Of course, the data is not perfect — as we explain in a forthcoming post, [measuring working hours with accuracy is difficult](https://ourworldindata.org/measure-working-hours), and surveys and historical records have limitations, so estimates of working hours spanning centuries necessarily come with a margin of error. But for any given country, the changes across time are much larger than the error margins at any point in time: The average worker in a rich country today really does work many fewer hours than the average worker 150 years ago. As the economists Diane Coyle and Leonard Nakamura explain, the study of working hours is crucial not only to measure [macroeconomic productivity](https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable), but also to measure economic well-being beyond economic _output_. A more holistic [framework for measuring ‘progress’](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602) needs to consider changes in how people are allowed to allocate their time over multiple activities, among which paid work is only one.{ref}Coyle, D. and Nakamura, L. I. (2019). [Toward a Framework for Time Use, Welfare, and Household Centric Economic Measurement.](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602) Federal Reserve Bank of Philadelphia Working Paper No. 19-11.{/ref} The available evidence shows that, rather than working more than ever, workers in many countries today work _much less _than in the past 150 years. There are huge inequalities within and across countries, but substantial progress has been made. ## Do workers in richer countries work longer hours? Economic prosperity in different places across our world today is [vastly unequal](https://ourworldindata.org/global-economic-inequality). People in Switzerland, one of the richest countries in the world, have an average income that is more than _20-times higher_ than that of people in Cambodia.{ref}We chose Cambodia and Switzerland here because they both also have working hours data available, but the difference in average income can be even more extreme. For instance, people in Qatar have an [average income that is _1_](https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT)[_1_](https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT)[_7_](https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT)[_-times higher_](https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT) than that of people in the Central African Republic. These differences refer to GDP per capita measured in international-$ and account for price differences between countries to enable comparisons. You can read more about this [here](https://ourworldindata.org/what-are-ppps).{/ref} Life in these two countries can [look starkly different](https://www.gapminder.org/dollar-street/?max=2755&countries=kh%2Cch&media=image&min=59&topic=homes&zoom=3).{ref}But life can also look similar, as you see in the pictures of the [homes](https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=homes&zoom=3), [computers](https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=computers&zoom=3), and [phones](https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=phones&zoom=3) of people on similar income levels in the two countries.{/ref} When considering such differences in prosperity, a natural question is: who works more, people in richer countries like Switzerland or in poorer ones like Cambodia? Looking at the available data, the answer is clear: workers in poorer countries actually tend to work more, and sometimes _much_ more. We see that in the chart here, with GDP per capita on the horizontal axis and annual working hours per worker on the vertical axis. Countries like Cambodia (which is the country in the very top-left corner) or Myanmar have some of the lowest GDP per capita but highest working hours. In Cambodia the average worker puts in 2,456 hours each year, nearly 900 more hours than in Switzerland (1,590 hours) at the bottom-right of the chart. The extra 900 hours for Cambodian workers means longer work days and many fewer days off. <Chart url="https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita-pwt?tab=chart&stackMode=absolute&time=2019..latest&country=®ion=World"/> ### Working hours tend to decrease as countries become richer There is a link between national income and average working hours, not only across countries at a given point in time — as shown in the chart above — but also for individual countries _over time._ Since the Industrial Revolution people in many countries have [become richer](https://ourworldindata.org/breaking-the-malthusian-trap), and [working hours have decreased dramatically](https://ourworldindata.org/working-more-than-ever) over these last 150 years. In the chart here we show this association between incomes and working hours over time, country by country. It is the same chart as above, except now countries’ single data points have become lines, connecting observations over time from 1950 until today. The four highlighted countries exemplify how working hours have decreased at the same time that average incomes have increased. Germany, for example, moved far to the right as its GDP per capita increased more than 10-fold (from $5,227 to $51,191), and far to the bottom as working hours decreased by nearly half (from 2,428 hours to 1,386 hours each year).{ref}These trends in GDP per capita are measured in constant international-$ and account for inflation to enable comparisons over time and between countries. You can read more about this [here](https://ourworldindata.org/economic-growth).{/ref} This makes sense: as people's incomes rise they can afford more of the things they enjoy, including more leisure and less time spent working. You can explore this association for other countries by clicking “Select countries” on the chart. <Chart url="https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita?tab=chart&xScale=log&stackMode=absolute&time=1950..latest&country=BRA~USA~DEU~TWN®ion=World"/> ### People are able to work less when they work in more productive economies The key driver of rising national incomes and decreasing working hours is productivity growth. Productivity refers to the rate at which inputs are turned into outputs. To understand working hours the key metric is _labor_ productivity: the economic return for one hour of work. At the most concrete level, labor productivity captures things like the number of breads that a baker bakes in an hour, or the number of cars factory workers assemble in an hour. At the most comprehensive level, it relates the total output of the economy (GDP) to the total labor input (total annual hours worked), giving us the aggregate measure of labor productivity, GDP per hour of work. Higher labor productivity is associated with fewer working hours, as shown in the chart here with labor productivity on the horizontal axis and annual working hours on the vertical axis. The chart currently shows data for the latest available year, but you can explore this relationship over time since 1950 by using the blue time slider at the bottom of the chart. <Chart url="https://ourworldindata.org/grapher/productivity-vs-annual-hours-worked?tab=chart&country=®ion=World"/> We see that the same richer countries with lower working hours we noted before — like Germany and Switzerland — have very high labor productivity (69 and 83 $/h, respectively). If workers can produce more with each hour of work, it becomes possible for them to work less. Though this doesn’t necessarily mean they _actually do_ work less — workers in the US and Singapore, for instance, work many more hours than their counterparts in countries with similar productivity.{ref}We explore the differences in working hours between similar, highly productive countries — and also the differences within those countries — in forthcoming posts.{/ref} In contrast, the countries toward the top-left of this chart have far lower labor productivity — Cambodia, for example, is at only 3$/h — and thus workers there need to work many more hours to compensate. ### At the heart of the link between productivity, incomes, and working hours is technological innovation Technological innovation — defined broadly here to include physical machines as well as ideas, knowledge, and processes — makes it possible for each worker to become much more productive. And increases in productivity in turn help drive both increases in incomes and decreases in working hours.{ref}For a discussion of how technology drives productivity growth and a rise in incomes (economic growth), see Romer, P. (1990) [Endogenous Technological Change.](https://www.journals.uchicago.edu/doi/abs/10.1086/261725)_Journal of Political Economy._ For a discussion of the relationship between productivity growth, economic growth, and working hours, see Boppart, T. and P. Krusell (2020) [Labor Supply in the Past, Present, and Future: A Balanced-Growth Perspective](https://www.journals.uchicago.edu/doi/full/10.1086/704071?casa_token=gHAZpu6iXwkAAAAA:WUwiHwVFoOWNFdnfKenDUm9yOtgcjKxwZpohoDcmZk8ZMwMASp86fPHmmd3-r8NLJk-9UKorX7o). _Journal of Political Economy._{/ref} A prime example of how tech innovation drives productivity growth is agriculture. As we show in detail in [our entry on Crop Yields](https://ourworldindata.org/crop-yields), innovations like better machinery, crop varieties, fertilizers, and land management have enabled farmers to be _much more_ productive. In the US, for example, farm production per labor hour increased nearly 16-fold from 1948–2011.{ref}See Figure 18 on p. 28 of Wang et al (2015) [Agricultural Productivity Growth in the United States: Measurement, Trends, and Drivers](https://www.ers.usda.gov/webdocs/publications/45387/53417_err189.pdf?v=6052.7). _USDA Economic Research Report 189._{/ref} This increased productivity enables us to feed a [rapidly growing population](https://ourworldindata.org/grapher/population?country=~OWID_WRL), even while the [fraction of people working in agriculture](https://ourworldindata.org/grapher/employment-by-economic-sector?stackMode=relative) is smaller than ever.{ref}The transition of employment out of agriculture to other economic sectors as countries become richer is known as ‘structural transformation’. You can read more about this in our post [_Structural transformation: how did today’s rich countries become ‘deindustrialized’?_](https://ourworldindata.org/structural-transformation-and-deindustrialization-evidence-from-todays-rich-countries){/ref} The chart here shows the growth in labor productivity, not just for agriculture but for the entire economy. The technological, economic, and social structures in richer countries have enabled workers there to produce more while working less. <Chart url="https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~TWN~GBR~USA~CHE®ion=World"/> Besides tech innovation, there is evidence that working fewer hours can itself keep productivity higher, making the link between working hours and productivity self-reinforcing. For example, economist [John Pencavel (2015) studied](https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext) munitions workers in war-time Britain and found that their productivity stayed high up to a certain threshold of hours, but declined markedly above that threshold.{ref}Pencavel, J. (2015) [The productivity of working hours](https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext). _The Economic Journal._{/ref} We’ve probably all experienced the drop in productivity that comes at the end of a very long day of work. ### What we learn from this The data show that it is workers in poorer countries who tend to work more, and sometimes _a lot_ more, than those in richer countries. This has large implications for the way we think about the economic progress made in the last two centuries and the nature of inequality between countries today. It means that residents of today’s poorer countries like Cambodia and Myanmar — and also of today’s richer countries in the past when they were poor — are not just _consumption_ poor, often unable to afford necessities like food and medicine. It means they are also _leisure_ poor: because productivity is low and they must work so much just to scrape by, they can’t afford to spend much time improving their condition, becoming educated, or simply enjoying leisure time. That people in poorer countries work so much more than in richer countries shows that differences in prosperity are not due to differences in work ethic — they are largely due to differences in circumstance and opportunity. As we ask in [another post](https://ourworldindata.org/talent-is-everywhere-opportunity-is-not), “what would have been the chances for Steve Jobs if he was born in the Central African Republic?” No matter how hard he worked or how smart he was, it is difficult to imagine that Steve Jobs would’ve been able to realize his potential with such a steep mountain of inequality to climb. We also see what _the world _[misses out on](https://www.imf.org/en/Publications/WP/Issues/2018/12/07/Invisible-Geniuses-Could-the-Knowledge-Frontier-Advance-Faster-46383) when exceptionally talented people, including all the brilliant but underprivileged people in today’s poorest countries, don’t have the opportunity to realize their potential.{ref}Agarwal, R. and Gaule, P. (2020) [Invisible Geniuses: Could the Knowledge Frontier Advance Faster?](https://www.aeaweb.org/articles?id=10.1257/aeri.20190457&&from=f)_ American Economic Review: Insights._{/ref} Finding ways to raise productivity is therefore not just key to increasing production, but also to the reduction in working hours that is necessary for a society to flourish. ## How are working hours measured and what can we learn from the data? Work is a central part of our lives. It is something we do almost every day, for much of the day, for decades on end. Because it is so central, looking closely at how much time we spend working can tell us a lot about our lives and the societies we live in. The data on working hours shows, for example, that rather than working more than ever — as is so commonly believed — people in many countries today [work _much less_ than in the past 150 years](https://ourworldindata.org/working-more-than-ever). Working less means being able to spend time becoming [more educated](https://ourworldindata.org/global-education), or simply [enjoying more leisure time](https://ourworldindata.org/time-use-living-conditions). This is substantial progress, but there is still [huge inequality across countries](https://ourworldindata.org/rich-poor-working-hours), and progress still to make. To understand these changes in societies and people’s lives over time, and the substantial differences we see in the world today, it is crucial to measure and study how much time people spend working. How are working hours actually measured? Where does the data come from, and how can researchers reconstruct long-run trends? Here we provide an overview of the main data sources, compare the data, and explain the relevant differences and measurement limitations. ### How are working hours measured? #### Surveys Surveys are the primary way to collect data on working hours. They are typically conducted by national statistical agencies and come in three main types: labor force surveys, establishment surveys, and time use surveys. These surveys all provide an important perspective on working hours, but there are some key differences. Labor force surveys collect data on employment status and time spent working by asking individual workers themselves. Of the survey types, these provide the most comprehensive perspective, covering hours _actually_ worked in all economic sectors as part of both formal and [informal employment](https://ourworldindata.org/grapher/informal-employment-of-total-non-agricultural-employment), full-time and part-time, as well as self-employment and unpaid family work.{ref}Hours actually worked means hours spent directly on work and excludes things like annual leave, sick leave, public holidays, meal breaks, and commuting time. Unpaid family work in this case generally includes market-oriented work, such as for the family business, but not other unpaid work at home such as childcare, cooking, and cleaning. Since the latter type of unpaid work is typically performed by women, this has large implications for understanding gender differences in labor. We discuss these issues as part of [our entry on Women’s Employment](https://ourworldindata.org/female-labor-supply#definitions-measurement).{/ref} But labor force surveys only cover residents of a country above a certain age (usually 15), which depending on the country might exclude a non-trivial number of workers.{ref}Only covering resident workers means that any [cross-border workers](https://ec.europa.eu/eurostat/cache/digpub/eumove/bloc-2c.html?lang=en#:~:text=In%202019%2C%20the%20largest%20number,and%20Belgium%20(50%20000).) are excluded. Only covering workers above a certain age means that any child laborers are excluded. While the incidence of child labor has been going down over time, especially in high-income countries, there are still an [estimated 265 million working children](https://ourworldindata.org/child-labor) in the world (almost 17% of the worldwide child population).{/ref} Establishment surveys collect data on employment and working hours as reported by employers.{ref}Employers include businesses, non-profits, some government agencies, and other organizations that pay a wage.{/ref} But because hours are reported by employers, these surveys often only cover paid or contractual hours and exclude self-employment, informal work, and some smaller firms.{ref}Unlike hours actually worked, paid or contractual hours typically include some time _not_ spent working, such as during sick leave, and fail to include time spent working that wasn't paid or planned, such as overtime.{/ref} On the other hand, establishment surveys provide more detail on the industry of work than other surveys, and are more consistent with how GDP is measured, making them useful for [studying labor productivity](https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~CHE~TWN~GBR~USA®ion=World). Time use surveys collect data on how individuals spend their time — [down to the minute](https://ourworldindata.org/time-use-living-conditions#daily-activities-similarities-and-differences-across-countries) — across a number of activities in a typical day, including paid work.{ref}Activities also include unpaid household work, school, leisure time, eating, and sleeping.{/ref} This level of granularity provides a useful complement to the other surveys, but as a trade-off time use surveys sample fewer people and are conducted less frequently and by fewer countries. #### National accounts To get the most comprehensive perspective on working hours possible, many countries aggregate data from these surveys with data from other sources — such as censuses, tax records, and social security registers — in an economic measurement framework called national accounts. National accounts, and the surveys they rely on, are standardized to a degree across countries, which can facilitate international comparisons.{ref}By organizations such as the [United Nations](https://unstats.un.org/unsd/nationalaccount/sna.asp), [International Labor Organization (ILO)](https://ilostat.ilo.org/resources/concepts-and-definitions/description-hours-of-work/), [OECD](https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en), and [Eurostat](https://ec.europa.eu/eurostat/web/microdata/european-union-labour-force-survey).{/ref} But these comparisons often have limitations because many countries still implement the methods in different ways. For instance, countries might bring together different data in their national accounts, or aggregate it differently. And many countries don’t have the capacity to conduct comprehensive surveys of their labor force and produce national accounts-based statistics, giving a more limited view of work there.{ref}For further discussion of different sources and their comparability, see the methods guides of the [OECD](https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en) and the [Total Economy Database](https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite) and the work of [Bick, Brüggemann, and Fuchs-Schündeln (2019)](https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344).{/ref} ### How do researchers reconstruct long-run historical trends? Comprehensive, cross-country data on working hours just isn’t available for the years before the mid 20th century. But researchers like [Huberman and Minns (2007)](http://www.sciencedirect.com/science/article/pii/S0014498307000058){ref}Huberman, M. and Minns, C. (2007) [The times they are not changin’: Days and hours of work in Old and New Worlds, 1870–2000.](http://www.sciencedirect.com/science/article/pii/S0014498307000058)_Explorations in Economic History._{/ref} have been able to fill some of the gap by reconstructing long-run trends for a selection of countries. How do they do it? Through often painstaking effort, researchers have been able to find and piece together the relevant historical records that do exist. In the work of Huberman and Minns, one of the key sources for historical data on many countries is a [report from the US Department of Labor](https://catalog.hathitrust.org/Record/008420895) published in 1900.{ref}U.S. Department of Labor (1900) [Fifteenth Annual Report of the Commissioner of Labor: Wages in Commercial Countries. 2 vols.](https://catalog.hathitrust.org/Record/008420895) Washington, DC.{/ref} The report compiled the records of many thousands of workers across numerous sectors from establishment surveys in 88 countries and territories. To reconstruct the trends in later years, Huberman and Minns pulled together data from the International Labor Organization, the work of peer researchers, and other sources.{ref}The original sources are: 1870–1913: Huberman (2004) [in turn relying on the US Department of Labor Fifteenth Annual Report, 1900]; 1929–1938: International Labor Organization (1934–39), except for Canada (Ostry and Zaidi, 1972), U.S. (Jones, 1963; Owen, 1988), and Australia (Butlin, 1977); 1950–2000: University of Groningen and the Conference Board GGDC Total Economy Database (2005).{/ref} This was an impressive feat of reconstruction, but historical records like this do have limitations. For instance, as exhaustive as they were, the establishment-level records used by Huberman and Minns still excluded agricultural work, part-time work, and many smaller firms. ### How does the data from different sources compare? The work by Huberman and Minns is an important example of how researchers often combine and adjust underlying sources to produce one-off cross-country estimates. Another important study is the one of Bick, Brüggemann, and Fuchs-Schündeln (2019),{ref}Bick, A., Brüggemann, B., and Fuchs-Schündeln, N. (2019) [Hours Worked in Europe and the United States: New Data, New Answers.](https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344)_The Scandinavian Journal of Economics._{/ref} who further standardized labor force surveys to enhance comparability for a selection of countries. Besides these one-off estimates, several international organizations and research centers aggregate the working hours estimates published by national statistical agencies into cross-country datasets. The two most important datasets come from the [OECD](https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#) and the [Penn World Table](https://www.rug.nl/ggdc/productivity/pwt/?lang=en) (PWT). These both draw on national accounts estimates when available, but they can differ in the other sources they use and their method of aggregation.{ref}PWT sources its working hours data from [The Conference Board’s Total Economy Database](https://www.conference-board.org/data/economydatabase/total-economy-database-productivity) (TED). For more details on the underlying sources, see the [TED guide](https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite) and the [OECD database](https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#).{/ref} In the chart you can compare annual working hours data from these four datasets. The data is shown one country at a time — with France currently selected. You can look at other countries by clicking ‘Change country’ on the chart, but note that not all sources publish data for every country. As expected, there are differences between the sources. In 2000, for instance, Bick et al. estimates 1,642 hours of work for French workers, OECD estimates 1,558 hours, PWT estimates 1,550 hours, and Huberman and Minns estimates 1,443 hours. These differences are due to the use of different underlying sources and methods. Bick et al. use only labor force surveys; the others all rely primarily on national accounts data, but which nonetheless still have differences. It’s also clear that these differences between sources are quite small when compared to the huge changes over the longer run. The difference between sources in 2000 is at most 200 hours, while the historical data from Huberman and Minns shows that from 1870 to 2000 annual working hours in France decreased by _1,725 hours_ (from 3,168 to 1,443 hours). <Chart url="https://ourworldindata.org/grapher/compare-sources-working-hours"/> ### What does this tell us about the study of working hours? The analysis here shows that working hours data can have limitations — due to differences in the sources or the way the method is implemented — but that what these matter for our interpretation of the data depends on the context. In a context where precise comparisons of similar countries is important, smaller differences between sources can really matter. This is why to compare recent working hours levels in the US and Europe, Bick et al. used only labor force surveys, which they standardized even further to maximize cross-country comparability. But as a trade-off, it was only possible to look at a small selection of richer countries. In a context where we want to focus on a larger scale — such as the [long-run historical trends](https://ourworldindata.org/working-more-than-ever) we see in the chart — the limitations of the sources are not large enough to undermine our conclusions. Large international datasets like PWT do not have the highest levels of cross-country comparability, but they allow us to look at many more countries across the world and uncover broad and important trends, such as the [large differences in working hours between the richest and poorest countries](https://ourworldindata.org/rich-poor-working-hours).{ref}We gain further confidence in these conclusions when they are echoed by research that focuses only on more standardized, comparable sources for a necessarily smaller set of countries, as in the work by [Bick, Fuchs-Schündeln, and Lagakos (2018)](https://www.aeaweb.org/articles?id=10.1257/aer.20151720).{/ref} PWT and OECD are also useful in contexts where we want an exhaustive picture of the trends in individual countries, since they are often based on national accounts that bring together data from many sources to give a comprehensive perspective on working hours. The data on working hours isn’t perfect, and it’s important to understand the limitations, but it can still tell us a lot about our lives and the world. --- # Data Sources --- ### OECD Time Use Database and Gender Data Portal * **Data:** Time spent on various activities per day, by sex * **Geographical coverage:** OECD countries plus China, India, and South Africa * **Time span:** from 1998 onwards * **Available at:**[Time Use Database](https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE); [Gender Data Portal](https://www.oecd.org/gender/data/balancingpaidworkunpaidworkandleisure.htm) ### Centre for Time Use Research * **Data:** Time spent on various activities per day; self-reported levels of enjoyment * **Geographical coverage:** 30 countries across the world * **Time span:** from 1964 onwards * **Available at:**[https://www.timeuse.org/](https://www.timeuse.org/) ### US Bureau of Labor Statistics * **Data:** Time spent with others per day * **Geographical coverage:** United States * **Time span:** from 2003 onwards * **Available at:**[https://www.bls.gov/tus/datafiles-0319.htm](https://www.bls.gov/tus/datafiles-0319.htm) ### IPUMS USA * **Data:** Percentage of Americans living alone, by age * **Geographical coverage:** United States * **Time span:** from 1900 onwards * **Available at:**[https://ipums.org/projects/ipums-usa/d010.v8.0](https://ipums.org/projects/ipums-usa/d010.v8.0) ### Dotti Sani and Treas (2016) * **Data:** Time parents spent with children per day * **Geographical coverage:** United States and select countries in Europe * **Time span:** 1965–2012 * **Available at:** Dotti Sani, G. M., and Treas, J. (2016). [Educational gradients in parents’ child‐care time across countries, 1965–2012](https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Dotti+Sani+and+Treas+2016+Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965%E2%80%932012&btnG=). Journal of Marriage and Family. ### Sayer, Bianchi, and Robinson (2004) * **Data:** Time parents spent on primary childcare activities per day * **Geographical coverage:** United States * **Time span:** 1965–1998 * **Available at:** Sayer, L., Bianchi, S., and Robinson, J. (2004). [Are Parents Investing Less in Children? Trends in Mothers’ and Fathers’ Time with Children](https://www.journals.uchicago.edu/doi/pdf/10.1086/386270). American Journal of Sociology. ### Huberman and Minns (2007) * **Data:** Annual hours of full-time production workers (male and female) in non-agricultural activities; days off from work for vacations and holidays * **Geographical coverage:** United States, Australia, Canada, and select countries in Europe * **Time span:** 1870–2000 * **Available at:** Huberman, M. and Minns, C. (2007). [The times they are not changin’: Days and hours of work in Old and New Worlds, 1870–2000.](https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!) Explorations in Economic History. ### Penn World Table * **Data:** Average annual hours worked by persons engaged; number of persons engaged; real and PPP-adjusted GDP in US millions of dollars * **Geographical coverage:** Countries across the world * **Time span:** from 1950 onwards * **Available at:**[https://www.rug.nl/ggdc/productivity/pwt/](https://www.rug.nl/ggdc/productivity/pwt/) ### Eurostat * **Data:** Time use survey data * **Geographical coverage:** Europe * **Time span:** from 1998 onwards * **Available at:**[https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#](https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#) ### Ramey and Francis (2009) * **Data:** Self-reported enjoyment of various activities; time spent on various activities, by sex and age; days of work lost to sickness * **Geographical coverage:** United States * **Time span:** 1900–2005 * **Available at:** Ramey, V. A., and Francis, N. (2009). [A century of work and leisure](https://www.aeaweb.org/articles?id=10.1257/mac.1.2.189). American Economic Journal: Macroeconomics. ### Costa (2000) * **Data:** Working hours, by sex; number of working days; wages * **Geographical coverage:** United States * **Time span:** 1890–1991 * **Available at:** Costa, D. L. (2000). [The Wage and the Length of the Work Day: From the 1890s to 1991.](https://www.jstor.org/stable/10.1086/209954?seq=1) Journal of Labor Economics. | { "id": 37823, "date": "2020-11-29T21:12:40", "guid": { "rendered": "https://owid.cloud/?page_id=37823" }, "link": "https://owid.cloud/time-use", "meta": { "owid_publication_context_meta_field": [], "owid_key_performance_indicators_meta_field": { "raw": "Time is the ultimate limited resource", "rendered": "<p>Time is the ultimate limited resource</p>\n" } }, "slug": "time-use", "tags": [], "type": "page", "title": { "rendered": "Time Use" }, "_links": { "self": [ { "href": "https://owid.cloud/wp-json/wp/v2/pages/37823" } ], "about": [ { "href": "https://owid.cloud/wp-json/wp/v2/types/page" } ], "author": [ { "href": "https://owid.cloud/wp-json/wp/v2/users/10", "embeddable": true } ], "curies": [ { "href": "https://api.w.org/{rel}", "name": "wp", "templated": true } ], "replies": [ { "href": "https://owid.cloud/wp-json/wp/v2/comments?post=37823", "embeddable": true } ], "wp:term": [ { "href": "https://owid.cloud/wp-json/wp/v2/categories?post=37823", "taxonomy": "category", "embeddable": true }, { "href": "https://owid.cloud/wp-json/wp/v2/tags?post=37823", "taxonomy": "post_tag", "embeddable": true } ], "collection": [ { "href": "https://owid.cloud/wp-json/wp/v2/pages" } ], "wp:attachment": [ { "href": "https://owid.cloud/wp-json/wp/v2/media?parent=37823" } ], "version-history": [ { "href": "https://owid.cloud/wp-json/wp/v2/pages/37823/revisions", "count": 29 } ], "wp:featuredmedia": [ { "href": "https://owid.cloud/wp-json/wp/v2/media/38050", "embeddable": true } ], "predecessor-version": [ { "id": 49487, "href": "https://owid.cloud/wp-json/wp/v2/pages/37823/revisions/49487" } ] }, "author": 10, "parent": 0, "status": "publish", "content": { "rendered": "\n<p>Time is the ultimate limited resource. Every single one of us has exactly the same \u201ctime budget\u201d: 24 hours per day, 365 days per year \u2013 8,760 hours <a href=\"https://owid.cloud/life-expectancy\" data-type=\"page\" data-id=\"1670\">every year of our lives</a>.</p>\n\n\n\n<p>How do we actually spend our time? There are many commonalities across the world: we all sleep, work, eat, and enjoy leisure time. But there are also important differences in the freedom people have to spend time on the things they value most. Studying how people across the world spend their time provides an important perspective for understanding living conditions, economic opportunities, and general well-being.</p>\n\n\n\n<p>Here we present the data on time use. We explore how it differs across countries and over time and how these differences matter for people\u2019s lives.</p>\n\n\n\t<div class=\"wp-block-owid-summary\">\n\t\t<h2>Summary</h2>\n\t\t\n\n<ul><li><a href=\"https://ourworldindata.org/time-use#how-do-people-across-the-world-spend-their-time-and-what-does-this-tell-us-about-living-conditions\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#how-do-people-across-the-world-spend-their-time-and-what-does-this-tell-us-about-living-conditions\">How people spend their time is similar in many ways across countries: we all sleep, work, eat, and enjoy leisure. But there are also important differences.</a></li><li><a href=\"https://ourworldindata.org/time-use#going-beyond-averages-the-gender-gap-in-leisure-time\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#going-beyond-averages-the-gender-gap-in-leisure-time\">There is a gender gap in leisure time: in all countries surveyed the average leisure time for men is higher than for women.</a></li><li><a href=\"https://ourworldindata.org/time-use#who-do-we-spend-time-with-across-our-lifetime\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#who-do-we-spend-time-with-across-our-lifetime\">Who we spend time with evolves across our lifetimes. As we enter older age we spend an increasing amount of time alone.</a></li><li><a href=\"https://ourworldindata.org/time-use#are-parents-spending-less-time-with-their-kids\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#are-parents-spending-less-time-with-their-kids\">The amount of time parents spend with their kids has been increasing.</a></li><li><a href=\"https://ourworldindata.org/time-use#are-we-working-more-than-ever\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#are-we-working-more-than-ever\">Working hours have decreased dramatically in the last 150 years for workers in many countries.</a></li><li><a href=\"https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours\">But there are still large differences between countries: workers in poorer countries tend to work much more than workers in richer countries.</a></li><li><a href=\"https://ourworldindata.org/time-use#how-are-working-hours-measured-and-what-can-we-learn-from-the-data\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#how-are-working-hours-measured-and-what-can-we-learn-from-the-data\">The primary way to measure working hours is with surveys, but the data can have limitations that are important to understand.</a></li></ul>\n\n\n\t</div>\n\n\n<h2>Time use from the perspective of an average day</h2>\n\n\n\n<h3>How do people across the world spend their time and what does this tell us about living conditions?</h3>\n\n\n\n<p>Sleep, work, eat, leisure \u2013 at a high level most of us spend time on similar activities. But just how similar are the daily activities of people across the world? This is something worth considering, not just to serve our curiosity but because differences in the way we spend time give us meaningful perspectives on living conditions, economic opportunities and general well-being.</p>\n\n\n\n<p>Here we take a look at the data. We explore some of the key patterns that emerge from cross-country time use surveys, and then dig deeper to understand how these differences matter for people\u2019s well-being.</p>\n\n\n\n<h4>Daily activities: similarities and differences across countries</h4>\n\n\n\n<div class=\"wp-block-columns\">\n<div class=\"wp-block-column\">\n<p>In the chart here we compare average time spent across a number of common activities. The data comes from the OECD and brings together estimates from time diaries where respondents are asked to record the sequence of what they did over a specific day, as well as from general questionnaires where respondents are asked to recall the amount of time spent on different activities on a specific day in the previous week.{ref}The \u2018time-diary method\u2019 is generally more reliable and allows a richer analysis of routines, because it measures not only aggregate times but also sequences and clock-times. Time-diary data is less common, but it is available for some countries from the <a rel=\"noreferrer noopener\" href=\"https://www.timeuse.org/mtus\" target=\"_blank\">Multinational Time Use Study</a>. We explore time-use \u2018tempograms\u2019 from the MTUS in a forthcoming companion post.{/ref}</p>\n\n\n\n<p>The first thing that jumps out from this chart is that there are indeed many similarities across countries.</p>\n\n\n\n<p>This is not surprising \u2013 most of us try to split our days into \u201cwork, rest and fun\u201d, and so there are some predictable patterns. We spend the most time working and sleeping; and paid work, housework, leisure, eating and sleeping take together 80-90% of the 1440 minutes that we all have available every day.</p>\n\n\n\n<p>But if we look closely, we also see some important differences. Consider sleeping, for example. From this sample of countries, South Koreans sleep the least \u2013 averaging 7 hours and 51 minutes of sleep every day. In India and the US, at the other end of the spectrum, people sleep an hour more on average.</p>\n\n\n\n<p>Work is another important activity where we see large differences. Countries are sorted by paid work hours in the chart \u2013 from highest to lowest. In China and Mexico people spend, on an average day, almost twice as much time on paid work as people in Italy and France do. This is a general pattern: <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use#do-workers-in-richer-countries-work-longer-hours\" target=\"_blank\">People in richer countries can afford to work less</a>. Keep in mind that this chart shows the average for all people in the working age bracket, from 15 to 64 years, whether they are actually employed or not.{ref}Because these estimates include people who are not employed they are much lower than the estimates of working hours <em>per</em> <em>worker</em> <a rel=\"noreferrer noopener\" target=\"_blank\" href=\"https://ourworldindata.org/time-use#time-spent-working\">we present elsewhere</a>. The estimates also differ because of differences in the sources: time-use surveys compared to labor force surveys and national accounts data.{/ref}</p>\n\n\n\n<p>Differences in <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/age-structure\" target=\"_blank\">demographics</a>, <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/global-education\" target=\"_blank\">education</a> and <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/economic-growth\" target=\"_blank\">economic prosperity</a> all contribute to these inequalities in work and time use. But what\u2019s clear in the chart here is that there are also some differences in time use that are not well explained by economic or demographic differences. In the UK, for example, people spend more time working than in France; but in both countries people report spending a similar amount of time on leisure activities.</p>\n\n\n\n<p>Cultural differences are likely to play a role here. The French seem to spend much more time eating than the British \u2013 and in this respect the data actually goes in line with stereotypes about food culture. People in France, Greece, Italy and Spain report spending more time eating than people in most other European countries. The country where people spend the least time eating and drinking is the USA (63 minutes).</p>\n</div>\n\n\n\n<div class=\"wp-block-column\">\n<h6>How do people spend their time?{ref}OECD (2020) <a rel=\"noreferrer noopener\" href=\"https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE\" target=\"_blank\">Time Use Database.</a>{/ref}</h6>\n\n\n\n<figure class=\"wp-block-image size-large is-resized\"><img loading=\"lazy\" src=\"https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD-732x550.png\" alt=\"\" class=\"wp-image-37744\" width=\"764\" height=\"574\" srcset=\"https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD-732x550.png 732w, https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD-400x300.png 400w, https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD-150x113.png 150w, https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD-768x577.png 768w, https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD-1536x1154.png 1536w, https://owid.cloud/app/uploads/2020/11/Time-Use-by-Country-OECD.png 1752w\" sizes=\"(max-width: 764px) 100vw, 764px\" /></figure>\n\n\n\n<p><a href=\"https://owid.cloud/app/uploads/2020/12/Time-Use-in-OECD-Countries-OECD.xlsx\" data-type=\"URL\" data-id=\"https://owid.cloud/app/uploads/2020/12/Time-Use-in-OECD-Countries-OECD.xlsx\">Download the underlying data for this chart (.xlsx)</a></p>\n</div>\n</div>\n\n\n\n<h4>Going beyond averages: The gender gap in leisure time</h4>\n\n\n\n<p>Going beyond national averages reveals important within-country inequalities. The gender gap in leisure time, for example, is a key dimension along which large inequalities exist.</p>\n\n\n\n<p>The chart here relies on the same time-use data described above, but shows total leisure time for men and women separately. Time for men is shown on the horizontal axis, while time for women appears on the vertical axis. The dotted diagonal line denotes \u2018gender parity\u2019, so the further away a country is from the diagonal line, the larger the difference between men and women.</p>\n\n\n\n<p>As we can see, in all countries the average leisure time for men is higher than for women \u2013 all bubbles are below the diagonal line \u2013 but in some countries the gaps are much larger. In Norway the difference is very small, while in Portugal men report almost 50% more leisure time than women.</p>\n\n\n\n<p>A key factor driving these differences in leisure time is the gender gap in unpaid work. As we explain in detail in a <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/female-labor-supply#informal-work-unpaid-care-work\" target=\"_blank\">companion post</a>, women are responsible for a disproportionate amount of unpaid work, and have less leisure as a result.{ref}If you want to dig deeper you can explore gender differences across all other activities directly from our source, via the <a rel=\"noreferrer noopener\" href=\"https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE\" target=\"_blank\">OECD Data Portal</a>. And you can read more about within-country inequalities in time use along other dimensions, such as income and education, in <a rel=\"noreferrer noopener\" href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=The+Middle+Class+Time+Squeeze%2C+Economic+Studies+at+Brookings%2C+Sawhill+and+Guyot+%282020%29&btnG=\" target=\"_blank\">this Brookings Paper</a>, where the authors focus on the \u2018middle class time squeeze\u2019 in the US. See: Sawhill, I. V., & Guyot, K. (2020). The Middle Class Time Squeeze. Economic Studies at Brookings. Brookings Institution.{/ref}</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/minutes-spent-on-leisure?tab=chart&stackMode=absolute&time=latest&country=&region=World\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>Why should we care about differences in time use?</h4>\n\n\n\n<div class=\"wp-block-columns\">\n<div class=\"wp-block-column\">\n<p>Every single one of us has the same \u201ctime budget\u201d: 24 hours per day and 365 days per year. But of course not all of us can choose to spend time on the activities that we enjoy most. Differences in our freedom to allocate time to the things we enjoy is the main reason why time-use data is important for studying living conditions.</p>\n\n\n\n<p>In the UK, researchers from the <a rel=\"noreferrer noopener\" href=\"https://www.timeuse.org/\" target=\"_blank\">Centre for Time Use Research</a> linked time-use diaries with the respondents\u2019 assessments of enjoyment, on a scale from 1 to 7, to better understand the connection between time use and well-being. The chart here, which we\u2019ve adapted from the book \u2018What We Really Do All Day\u2019<em>, </em>by professors Jonathan Gershuny and Oriel Sullivan, shows the results. The estimates correspond to average reported levels of enjoyment for each activity, with confidence intervals.{ref}The underlying data comes from time-use diaries where respondents are asked to record the sequence of what they do over a specific day, and how much they enjoy each \u2018episode\u2019 (i.e. what they do) on a scale from 1 to 7. All episodes reported are then coded and grouped into similar activities. To arrive at the mean enjoyment scores, the authors multiply the duration of each episode where the activity category concerned is the primary activity recorded, by the enjoyment level to arrive at the total enjoyment score for that episode. Then they sum these total enjoyment scores for each category of activity across the day, and finally divide these daily enjoyment total scores for each activity by the amount of time devoted to the activity. In this way, they arrive at an appropriately weighted mean enjoyment level for each activity across all those who engage in it. For more details see Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref}</p>\n\n\n\n<p>We see that the most enjoyed activities involve rest or leisure activities such as eating out, sleeping, going to sports events, playing computer games or attending cultural performances. The activities receiving the lowest ratings include doing school homework, looking for a job, or doing housework.</p>\n\n\n\n<p>The activity where people show the greatest variation in enjoyment is working a \u201cSecond Job\u201d. This likely reflects the difference between people who work a second job because they want to, and those who work a second job because they <em>have</em> <em>to</em>.</p>\n\n\n\n<p>So what do we learn from this?</p>\n\n\n\n<p>First, we learn that the enjoyment of activities is, at least to some degree, predictable and stable. This means we can take activity groups and make meaningful comparisons across groups of people. Economists, for example, <a rel=\"noreferrer noopener\" href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Ramey%2C+V.+A.%2C+%26+Francis%2C+N.+%282009%29.+A+century+of+work+and+leisure.+American+Economic+Journal%3A+Macroeconomics%2C+1%282%29%2C+189-224.&btnG=\" target=\"_blank\">often classify</a> any activity with an enjoyment level below work as a \u201cnon-leisure activity\u201d, in order to measure trends in leisure across people and time.{ref}You find a very clear and complete explanation of this in Ramey, V. A., & Francis, N. (2009). A century of work and leisure. American Economic Journal: Macroeconomics, 1(2), 189-224.{/ref}</p>\n\n\n\n<p>But beyond this, and more importantly, this confirms that time-use is informative about well-being.</p>\n\n\n\n<p>The fact that there is a very clear and predictable pattern in the enjoyment of activities suggests that differences in time use do, indeed, give us meaningful perspectives on living conditions and economic opportunities. In countries where people do more paid and unpaid work, and have less time for leisure, their enjoyment \u2013 and <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/happiness-and-life-satisfaction\" target=\"_blank\">happiness and life satisfaction</a> \u2013 levels are likely to be lower.</p>\n</div>\n\n\n\n<div class=\"wp-block-column\">\n<h6>How do people rate the enjoyment of different activities?{ref}Gershuny, J., & Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.{/ref}</h6>\n\n\n\n<figure class=\"wp-block-image size-large is-resized\"><img loading=\"lazy\" src=\"https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-364x550.png\" alt=\"\" class=\"wp-image-37746\" width=\"615\" height=\"930\" srcset=\"https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-364x550.png 364w, https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-265x400.png 265w, https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-99x150.png 99w, https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-768x1160.png 768w, https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-1017x1536.png 1017w, https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use-1356x2048.png 1356w, https://owid.cloud/app/uploads/2020/11/Enjoyment-Level-of-Time-Use.png 1630w\" sizes=\"(max-width: 615px) 100vw, 615px\" /></figure>\n</div>\n</div>\n\n\n\n<p></p>\n\n\n\n<h2>Time use from the perspective of the life cycle</h2>\n\n\n\n<h3>Who do we spend time with across our lifetime?</h3>\n\n\n\n<p>As we go through life we build personal relationships with different people \u2013 family, friends, coworkers, partners. These relationships, which are <a href=\"https://ourworldindata.org/social-connections-and-loneliness#social-connections\" target=\"_blank\" rel=\"noreferrer noopener\">deeply important to all of us</a>, evolve with time. As we grow older we build new relationships while others transform or fade, and towards the end of life many of us spend a lot of time alone.</p>\n\n\n\n<p>Taking the big picture over the entire life course: Who do we actually spend our time with?</p>\n\n\n\n<h4>From adolescence to old age: who do we spend our time with?</h4>\n\n\n\n<p>To understand how social connections evolve throughout our lives we can look at survey data on <em>how much</em> time people spend with others, and <em>who</em> that time is spent with.</p>\n\n\n\n<p>The chart here shows the amount of time that people in the US report spending in the company of others, based on their age. The data comes from time-use surveys, where people are asked to list all the activities that they perform over a full day, and the people who were there during each activity. We currently only have data with this granularity for the US \u2013 time-use surveys are common across many countries, but what is special about the US is that respondents of the American Time Use Survey are asked to list everyone who was present for each activity.</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/time-spent-with-relationships-by-age-us?tab=chart&stackMode=absolute&region=World\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<p>The numbers in this chart are based on averages for a cross-section of the American society \u2013 people are only interviewed once, but we have brought together a decade of surveys, tabulating the average amount of time that survey respondents of different ages report spending with other people<em>.</em>{ref}When interpreting this chart it\u2019s important to bear in mind that the relationships used to categorise people are not exhaustive (i.e., survey respondents could also list being with people who didn\u2019t fit any of the listed categories, or for whom a relationship was unclear or unknown \u2013 we do not count these instances in the estimates). Additionally, time spent with multiple people can be counted more than once; so attending a party with friends and your spouse, for example, would show up for both \u201cfriends\u201d and \u201cpartner\u201d in our estimates. The implication is that companion categories cannot be stacked to add up total time spent in the company of others.{/ref}</p>\n\n\n\n<h5>Who we spend our time with changes a lot over the course of life</h5>\n\n\n\n<p>When we\u2019re young \u2013 particularly in our teens \u2013 we spend a lot of our time with friends, parents, siblings and extended family. As we enter our 20s, time with friends, siblings and parents starts to drop off quickly. Instead, we start spending an increasing amount of time with partners and children. (The chart shows an average across Americans, so for those that have children the time spent with children is even higher, since the average is pulled down by those without children.)</p>\n\n\n\n<p>As the chart shows, this continues throughout our 30s, 40s and 50s \u2013 over this period of their life, Americans spend much of their time with partners, children and, unsurprisingly, co-workers.</p>\n\n\n\n<p>For those 60 and older, we see a significant drop-off in time spent with co-workers. This makes sense, considering many people in the US <a href=\"https://ourworldindata.org/grapher/average-effective-retirement-men\" data-type=\"URL\" data-id=\"https://ourworldindata.org/grapher/average-effective-retirement-men\" target=\"_blank\" rel=\"noreferrer noopener\">enter retirement in their mid 60s</a>. We see that this time is partly displaced by more time with partners.</p>\n\n\n\n<p>In terms of the diversity of interactions, this chart suggests that the number of people with whom we interact is highest around 40, but then things change substantially after that. And this is perhaps the most conspicuous trend in the chart: above 40, people spend an increasing amount of time alone.</p>\n\n\n\n<h4>Older people spend a lot of time alone</h4>\n\n\n\n<p>Older people spend a large amount of time alone and it is understandable why \u2013 time spent alone increases with age because this is when health typically deteriorates and people lose relatives and friends.</p>\n\n\n\n<p>Indeed, many people who are older than 60 live alone as this chart shows clearly: living alone is particularly common for older adults. Today nearly 4 out of every 10 Americans who are older than 89 years old live alone.</p>\n\n\n\n<p>Another interesting point here is that the share of people across all age groups who live alone has been increasing over time. This is part of a more general global trend \u2013 if you want to read more about the global \u2018rise of living alone\u2019, we provide a detailed account of this trend across countries in a <a rel=\"noreferrer noopener\" href=\"http://ourworldindata.org/living-alone\" target=\"_blank\">companion post</a>.</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/percentage-of-americans-living-alone-by-age\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>Older people spend more time alone, but this doesn\u2019t necessarily mean they\u2019re lonely</h4>\n\n\n\n<p>The data shows that as we become older we tend to spend more and more time alone. What\u2019s more, the data also shows that older people today spend more time alone than older people did in the past.</p>\n\n\n\n<p>We might think older people are therefore more lonely \u2013 but this is not necessarily the case.</p>\n\n\n\n<p>Spending time alone is not the same as feeling lonely. This is a point that is well recognised by researchers, and one which has been confirmed empirically across countries. Surveys that ask people about living arrangements, time use, and feelings of loneliness find that solitude, by itself, is not a good predictor of loneliness. You can read our overview of the evidence in <a rel=\"noreferrer noopener\" href=\"http://ourworldindata.org/lonely-not-alone\" target=\"_blank\">this post</a>.</p>\n\n\n\n<p>So, what about loneliness? If we focus on self-reported loneliness, there is <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/loneliness-epidemic#loneliness-trends-over-time-are-people-lonelier-today-than-in-the-past\" target=\"_blank\">little evidence</a> of an upward trend over time in the US; and importantly, it\u2019s <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/social-connections-and-loneliness#do-we-become-lonelier-as-we-get-older\" target=\"_blank\">not the case</a> that loneliness keeps going up as we become older. In fact a <a rel=\"noreferrer noopener\" href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Are+U.S.+older+adults+getting+lonelier%3F%3A+Age%2C+period%2C+and+cohort+differences.+Psychology+and+Aging.&btnG=\" target=\"_blank\">recent study</a> based on surveys that track the same individuals over time found that after age 50 \u2013 which is the earliest age of participants in the analysis \u2013 loneliness tended to decrease, until about 75, after which it began to increase again.</p>\n\n\n\n<p>Taking the evidence together, the message is not that we should be sad about the prospect of aging, but rather that we should recognise the fact that social connections are complex. We often tend to look at the amount of time spent with others as a marker of social well-being; but the quality of time spent with others, and our expectations, matter even more for our feelings of connection and loneliness.</p>\n\n\n\t<block type=\"additional-information\" default-open=\"false\">\n\t\t<content>\n\n<h3><strong>A note about the interpretation of age trends</strong></h3>\n\n\n\n<div class=\"wp-block-columns is-style-sticky-right\">\n<div class=\"wp-block-column\">\n<p>In the chart where we plot the amount of time that people in the US report spending in the company of others, it\u2019s important to keep in mind that we are taking a look at a cross-section of society. This means that we are actually seeing the result of two underlying trends.</p>\n\n\n\n<p>On one hand, we see the effect of aging on social connections (we relate to different people and reallocate time as we go through different stages of life), but we also see the effect of cohort trends (compared to people in the past, today\u2019s older generations in the US tend to be <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/grapher/life-expectancy?tab=chart&country=JPN~GBR~IND~ETH~OWID_WRL~KOR~ZAF~USA&region=World\" target=\"_blank\">healthier</a> and <a href=\"https://ourworldindata.org/grapher/average-real-gdp-per-capita-across-countries-and-regions?tab=chart&country=Western%20Offshoots~Western%20Europe~Western%20Asia~Eastern%20Europe~OWID_WRL~Latin%20America~East%20Asia~Africa~USA\">richer</a>, and might also have different expectations, preferences and opportunities).</p>\n\n\n\n<p>Disentangling these two effects is difficult, so it is important to keep in mind that at least some of the age gradients we observe might be partly explained by cohort changes, rather than life-cycle trends. </p>\n\n\n\n<p>This is why it\u2019s important to rely not only on cross-sectional data, but also on surveys that track the same individuals over time.</p>\n</div>\n\n\n\n<div class=\"wp-block-column\"></div>\n</div>\n\n</content>\n\t</block>\n\n\n<p></p>\n\n\n\n<h2>Parents\u2019 time with children</h2>\n\n\n\n<h3>Are parents spending less time with their kids?</h3>\n\n\n\n<p>Over the last 50 years many countries have seen large changes in family structures and the institution of marriage. These changes \u2013 which include a <a href=\"https://ourworldindata.org/marriages-and-divorces#single-parenting-is-common-and-in-many-countries-it-has-increased-in-recent-decades\" target=\"_blank\" rel=\"noreferrer noopener\">rise in single-parenting</a> and a large increase in the <a href=\"https://ourworldindata.org/female-labor-supply#female-participation-in-labor-markets-grew-remarkably-in-the-20th-century\" target=\"_blank\" rel=\"noreferrer noopener\">share of women working outside the home</a> \u2013 have made some people worry that children might be getting \u2018short-changed\u2019, because parents are not spending as much time with them.</p>\n\n\n\n<p>In 1999, for example, a <a href=\"https://clintonwhitehouse4.archives.gov/media/pdf/famfinal.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">report</a> from the Council of Economic Advisors in the US analyzed trends over the second half of the 20th century and concluded: <em>\u201cThe increase in hours mothers spend in paid work, combined with the shift toward single-parent families, resulted in families on average experiencing a decrease of 22 hours a week (14 percent) in parental time available outside of paid work that they could spend with their children.\u201d</em></p>\n\n\n\n<p>The line of thought behind these concerns is that changes to the structure of families and work have meant that children spend less time with parents, because parents \u2013 particularly mothers \u2013 spend less time at home.</p>\n\n\n\n<p>Here we review the evidence and show that this reasoning is flawed. As we explain, in the US and many other rich countries parents spend more time with their kids today than 50 years ago. Equating \u2018mother time at home’ with ‘children\u2019s time with parents\u2019 is a huge and unhelpful oversimplification.</p>\n\n\n\n<h4>Mothers and fathers spend an increasing amount of time with kids: evidence from time-use surveys</h4>\n\n\n\n<div class=\"wp-block-columns\">\n<div class=\"wp-block-column\">\n<p>The chart here shows the time that mothers and fathers spend with their children. This is measured using time-use diaries where parents record the amount of time that they spend on various activities, including child care.</p>\n\n\n\n<p>These estimates, which are sourced from a <a rel=\"noreferrer noopener\" href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=\" target=\"_blank\">paper</a> published in 2016 by sociologists Giulia Dotti Sani and Judith Treas, are disaggregated by education levels and are adjusted to account for demographic differences between countries. We explain <a href=\"https://ourworldindata.org/parents-time-with-kids#digging-deeper-raw-vs-population-adjusted-trends-in-time-use\" data-type=\"URL\" data-id=\"https://ourworldindata.org/parents-time-with-kids#digging-deeper-raw-vs-population-adjusted-trends-in-time-use\">below</a> in more detail why these adjustments are important.{ref}The full reference for the source paper is: Dotti Sani, G. M., & Treas, J. (2016). <a rel=\"noreferrer noopener\" href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=\" target=\"_blank\">Educational gradients in parents’ child\u2010care time across countries</a>, 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096. {/ref}</p>\n\n\n\n<p>As we can see here, there has been a clear increase in the amount of time parents spend with their children over the last 50 years. This is true for both fathers and mothers, and holds across almost all education groups and countries. The two exceptions are France, where mothers\u2019 time has declined (from a very comparatively high level); and Slovenia where it has remained roughly constant among non-university-educated parents.</p>\n\n\n\n<p>In terms of within-country inequalities we also see two clear patterns. First, in all countries mothers spend more time in child care activities than fathers. The differences are large and persistent across education groups. In some countries, such as Canada, France and the US, this gender gap has shrunk; while in other countries such as Denmark and Spain, the gap has widened.</p>\n\n\n\n<p>Second, in all countries there is a positive \u2018educational gradient\u2019, meaning more educated parents tend to spend more time with their children. In many countries this gradient increased, and nowhere did it decline.</p>\n</div>\n\n\n\n<div class=\"wp-block-column\">\n<h6>How much time do parents spend with their children per day?{ref}Dotti Sani, G. M., & Treas, J. (2016). <a rel=\"noreferrer noopener\" href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965-2012%3A+Educational+Gradients+in+Parents%27+Child-Care+Time&btnG=\" target=\"_blank\">Educational gradients in parents’ child\u2010care time across countries</a>, 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref}</h6>\n\n\n\n<figure class=\"wp-block-image size-large is-resized\"><img loading=\"lazy\" src=\"https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2-685x550.png\" alt=\"\" class=\"wp-image-37745\" width=\"747\" height=\"600\" srcset=\"https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2-685x550.png 685w, https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2-400x321.png 400w, https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2-150x120.png 150w, https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2-768x616.png 768w, https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2-1536x1233.png 1536w, https://owid.cloud/app/uploads/2020/11/Parent-time-spent-with-children-2.png 1812w\" sizes=\"(max-width: 747px) 100vw, 747px\" /></figure>\n</div>\n</div>\n\n\n\t<block type=\"additional-information\" default-open=\"false\">\n\t\t<content>\n\n<h3>Digging Deeper: Raw vs Population-adjusted Trends in Time Use</h3>\n\n\n\n<div class=\"wp-block-columns is-style-sticky-right\">\n<div class=\"wp-block-column\">\n<p>The main chart in this post, where we show the amount of time that fathers and mothers spend with their children, plots \u2018population-adjusted trends\u2019. This means that the estimates are not what you directly observe in the surveys, but rather the output of a statistical model that takes these surveys as an input but corrects for differences in key factors such as age, partnership status, number of children and employment. </p>\n\n\n\n<p>This is why the trends look so smooth \u2013 rather than plotting the raw survey averages for each year, the chart shows \u2018modelled estimates\u2019 after holding compositional factors constant. </p>\n\n\n\n<p>These adjustments are helpful for the analysis, because there are large differences between population groups within countries (e.g. single parents are not able to spend as much time with their children); and since the relative size of these groups is changing, the aggregate trend that we observe is affected by these compositional changes (e.g. in some countries single-parenting is increasing, while in others it is decreasing).</p>\n\n\n\n<p>So how much difference do these \u201cadjustments\u201d make? After all, we are also interested in the <em>observed</em> amount of time that parents spend with their children, and not only the values that are adjusted for population changes. What would the trends look like if we just plotted the raw survey estimates?</p>\n\n\n\n<p>The next chart shows this with data for the US. As we can see, the trend goes in the same direction for both fathers and mothers.</p>\n</div>\n\n\n\n<div class=\"wp-block-column\">\n<iframe src=\"https://ourworldindata.org/grapher/us-child-care-time-sex\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n</div>\n</div>\n\n</content>\n\t</block>\n\n\n<h4>Parents\u2019 time with children, as reported by children, has also increased</h4>\n\n\n\n<p>The analysis above looked at time-use diaries from <em>parents.</em> But children may experience time and attention differently from adults. Is there evidence that children also feel they are getting more time with their parents? </p>\n\n\n\n<p>In a <a href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=\" target=\"_blank\" rel=\"noreferrer noopener\">study</a> published in 2001 in the journal <em>Demography</em>, John Sandberg and Sandra Hofferth analyzed two surveys with data from child time-use diaries in the US, and found that children reported spending about 4.3 more hours per week with mothers, and 3 more hours per week with fathers in 1997 versus 1981.{ref}The estimates from Sandberg and Hofferth correspond to weekly hours among children ages 3-12. The estimates cover both \u2018engaged time\u2019 and other activities where parents are not engaged but children say they are present. When looking jointly at both parents, Sandberg and Hofferth find that time spent with either parent remains roughly unchanged, which suggests part of the increase corresponds to an intra-household reallocation of time. In the paper you find more details and discussion, including a detailed analysis of trends for more narrow population groups and possible demographic and behavioral drivers: Sandberg, J. F., & Hofferth, S. L. (2001). <a href=\"https://scholar.google.com/scholar?hl=es&as_sdt=0%2C5&q=Sandberg%2C+J.+F.%2C+%26+Hofferth%2C+S.+L.+%282001%29.+Changes+in+children%E2%80%99s+time+with+parents%3A+United+States%2C+1981%E2%80%931997.+Demography&btnG=\" target=\"_blank\" rel=\"noreferrer noopener\">Changes in children\u2019s time with parents: United States, 1981\u20131997</a>. Demography, 38(3), 423-436. {/ref}</p>\n\n\n\n<p>The data is sparse because time-use diaries for children are not very common; but the evidence that is available is consistent with what we\u2019ve seen above: Children in the US agree that they\u2019re spending more time with their parents than in the past.</p>\n\n\n\n<h4>Social norms affect how we allocate time</h4>\n\n\n\n<p>The amount of time that parents spend with their children is determined by many factors, and working hours outside the home are only one of them. Choices, parenting norms, family size, and relationship ideals \u2013 for men and women, parents and children \u2013 matter a lot.</p>\n\n\n\n<p>Several academic studies have dissected the data from the US in an attempt to disentangle the relative importance of different underlying factors. The conclusion from these studies is that the reason why we see an increase in the amount of time that American parents spend with kids is that many families, particularly those who are well-off, have been able to undergo changes in their routines and the allocation of tasks and time within the household, in order to spend more time with their children. Parents have been able to undergo \u2018behavioral changes\u2019 that have more than compensated for \u2018structural changes\u2019 that could have pushed in the opposite direction.{ref}For an overview of the literature see (i) Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents’ child\u2010care time across countries, 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096. (ii) Sayer, L. C., Bianchi, S. M., & Robinson, J. P. (2004). Are parents investing less in children? Trends in mothers\u2019 and fathers\u2019 time with children. American journal of sociology, 110(1), 1-43. (iii) Sandberg, J. F., & Hofferth, S. L. (2001). Changes in children\u2019s time with parents: United States, 1981\u20131997. Demography, 38(3), 423-436.{/ref}</p>\n\n\n\n<p>Employment, and in particular <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/working-hours\" target=\"_blank\">working hours</a> outside the home, determine a range of possible arrangements, but individual preferences and social conventions determine the actual allocation of time within that range of feasible options. </p>\n\n\n\n<p>Preferences and social conventions are dynamic and change with time. There is a large academic literature that emphasizes the importance of parent-child interactions for children\u2019s development, and it seems natural to allow for the possibility that in recent decades parenting norms have undergone considerable change in response to this scientific evidence.{ref}The parenting guide from the National Health Insurance in the UK, for example, says <em>\u201cParents should make the time to play with a first or only child. And while brothers and sisters are natural playmates, parents can also play an active role in siblings’ games…If you’re pressed for time as a parent, it’s a good idea to find ways to involve your child in what you’re doing \u2013 even the housework.</em>\u201d You find a review of the academic literature on the importance of parent-child interactions for children\u2019s development in Dotti Sani, G. M., & Treas, J. (2016). Educational gradients in parents’ child\u2010care time across countries, 1965\u20132012. Journal of Marriage and Family, 78(4), 1083-1096.{/ref}</p>\n\n\n\n<h4>Why should we care?</h4>\n\n\n\n<p>The available data shows that in many countries parents spend more time with their kids than ever before, and this is partly due to changes in social conventions and economic progress, especially declining working hours.</p>\n\n\n\n<p>There are large differences between countries, and also large inequalities across different population groups within each country; but overall the trends tend to go in the same direction.</p>\n\n\n\n<p>Of course, there\u2019s a lot that is not being captured by these aggregate statistics. There are important distributional issues, and there is also clearly more to parents\u2019 and children\u2019s welfare than \u2018total time spent together\u2019. But despite these limitations, the available research and data still offers a clear lesson: contrary to what some people fear, it\u2019s not the case that children in rich countries are being systematically \u2018shortchanged\u2019 by widespread changes in family structures. </p>\n\n\n\n<p>Parents are spending more time with their kids than they used to, and this matters because parent-child interactions are important for childhood development.</p>\n\n\n\n<p></p>\n\n\n\n<h2>Time spent working</h2>\n\n\n\n<h3>Are we working more than ever?</h3>\n\n\n\n<p>In today\u2019s hustle and bustle world, it\u2019s easy to assume that we are all, by and large, working more than ever. But is that really the case?</p>\n\n\n\n<p>As we explain in detail below, the research on the history of working hours shows that this is not the case.</p>\n\n\n\n<p>The available data shows that in the 19th century people across the world used to work extremely long hours, but in the last 150 years working hours have decreased substantially, particularly in today\u2019s richest countries.</p>\n\n\n\n<h4>Working hours per worker have declined after the Industrial Revolution</h4>\n\n\n\n<p>The chart here shows average working hours since 1870 for a selection of countries that industrialized early. You can add or remove countries by clicking Add country on the chart.</p>\n\n\n\n<p>We show annual totals, so the trends account for changes in both the length of working days as well as the number of days worked through the year. The data comes from research by the economic historians Michael Huberman and Chris Minns, who have brought together evidence from historical records, National Accounts data, and other sources.{ref}In this chart we have taken the original data published by <a rel=\"noreferrer noopener\" href=\"https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!\" target=\"_blank\">Huberman & Minns (2007)</a> and extended coverage using an updated vintage of the Penn World Table (PWT), which is in turn based on the same underlying source that Huberman and Minns used for all data since 1950, the Total Economy Database. You can find more details and links to our sources in the \u2018Sources\u2019 tab of the chart.{/ref}</p>\n\n\n\n<p>The chart shows that average working hours declined dramatically for workers in early-industrialized economies over the last 150 years. In 1870, workers in most of these countries worked more than 3,000 hours annually \u2014 equivalent to a grueling 60\u201370 hours each week for 50 weeks per year.</p>\n\n\n\n<p>But we see that today those extreme working hours have been roughly cut in half. In Germany, for example, annual working hours decreased by nearly 60% \u2014 from 3,284 hours in 1870 to 1,354 hours in 2017 \u2014 and in the UK the decrease was around 40%. Before this revolution in working hours people worked as many hours between January and July as we work today in an entire year.{ref}A key point to keep in mind when interpreting these trends is that they refer to working hours <em>per worker</em>, which is different from working hours <em>per person. </em>The per person measure corresponds to working hours per worker multiplied by the employment rate. Hence, changes in employment patterns \u2014 such as the historical rise of female participation in paid employment in these countries \u2014 mean that changes in hours per worker do not translate directly into changes in hours per person.{/ref}</p>\n\n\n\n<p>For many countries in the chart we don\u2019t have long-run series going back to the 19th century. But we do have evidence from other historical records from 1870\u20131900 that in many of those countries workers also used to work extremely long hours.{ref}A <a href=\"https://www.econstor.eu/handle/10419/67824\" data-type=\"URL\" data-id=\"https://www.econstor.eu/handle/10419/67824\" target=\"_blank\" rel=\"noreferrer noopener\">study by Michael Huberman and Frank Lewis</a> reconstructed estimates of working hours in 1870 and 1900 for 48 countries across six continents using data from worker records kept by individual business establishments. They drew from a collection of records published by the US Department of Labor in 1900, and found substantial variation, but very high working hours for many non-industrialized countries. They found for example that in 1870, Colombia, Uruguay and Brazil had similar average working hours per worker as the US. The full reference of the paper is Huberman, M., & Lewis, F. D. (2007). Bend it like Beckham: Hours and wages across forty-eight countries in 1900 (No. 1229). Queen’s Economics Department Working Paper.{/ref}</p>\n\n\n\n<p>For those countries with long-run data in this chart we can see three distinct periods: From 1870\u20131913 there was a relatively slow decline; then from 1913\u20131938 the decline in hours steepened in the midst of the powerful sociopolitical, technological, and economic changes that took shape with World War I, the Great Depression, and the lead-up to World War II; and then after an uptick in hours during and just after World War II, the decline in hours continued for many countries, albeit at a slower pace and with large differences between countries.{ref}The increase in hours between 1938 and 1950 in the chart for some countries is due in part to the uptick during and just after World War II, but also plausibly due in part to differences in the source data and methodology.{/ref}</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/annual-working-hours-per-worker\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>In recent decades working hours have continued to decline in many countries, but there are large differences between countries</h4>\n\n\n\n<p>Zooming in to the last 70 years and looking at other countries beyond those who industrialized early, the data reveals a continued decline in working hours for many countries but also large differences between countries.</p>\n\n\n\n<p>In the chart here we zoom in to the period since 1950 and we change the selection of countries to highlight some of the diversity in trends.</p>\n\n\n\n<p>For some countries, such as Germany, working hours have continued their steep historical decline; while for other countries, such as the US, the decline has leveled off in recent decades.</p>\n\n\n\n<p>In some countries we see an inverted U-shaped pattern. In South Korea, for example, hours rose dramatically between 1950 and 1980 before falling again since the mid 1980s. And in other countries we see no recent declines \u2014 in China, for example, hours actually rose in the 1990s and early 2000s before leveling off in recent years.</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/annual-working-hours-per-worker?tab=chart&stackMode=absolute&time=1950..latest&country=DEU~USA~BRA~CHN~KOR~IND&region=World\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>Shorter work days, but also more holidays and vacations</h4>\n\n\n\n<p>The decline in annual working hours described above has come from fewer working hours each day, as well as fewer working days each week and fewer working weeks in the year.</p>\n\n\n\n<p>In a <a rel=\"noreferrer noopener\" href=\"https://www.jstor.org/stable/10.1086/209954\" target=\"_blank\">paper</a> analyzing historical data for the US, the economist Dora Costa summarizes the evidence:{ref}Costa, D. L. (2000). <a rel=\"noreferrer noopener\" href=\"https://www.jstor.org/stable/10.1086/209954?seq=1\" data-type=\"URL\" data-id=\"https://www.jstor.org/stable/10.1086/209954?seq=1\" target=\"_blank\">The Wage and the Length of the Work Day: From the 1890s to 1991.</a> <em>Journal of Labor Economics</em>, 18(1).{/ref}</p>\n\n\n\n<blockquote class=\"wp-block-quote\"><p><em>\u201cThe length of the work day fell sharply between the 1880s, when the typical worker labored 10 hours a day, 6 days a week, and 1920, when his counterpart worked an 8-hour day, 6 days a week. By 1940 the typical work schedule was 8 hours a day, 5 days a week. Although further reductions in work time largely took the form of increases in vacations, holidays, sick days, personal leave, and earlier retirement, time diary studies suggest that the work day has continued to trend downward less than 8 hours a day.\u201d</em></p></blockquote>\n\n\n\n<p>As Costa notes, workers had regular days off each week: one day off (usually Sunday) from at least the 1880s until around the 1940s, when two days off became more typical.</p>\n\n\n\n<p>In addition to regular days off each week, workers across early-industrialized countries had days off from work for vacations and national holidays. This is shown in the chart here, which again relies on research from Huberman and Minns. The chart shows that days of vacation and holidays increased over the period from 1870\u20132000. The Netherlands is a stark example \u2014 workers there saw an increase from four days off for vacations and holidays in 1870 to almost 38 days off in 2000.</p>\n\n\n\n<p>The declines in the length of the work day and the number of working days have been driven by several factors, including <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/rich-poor-working-hours\" data-type=\"URL\" data-id=\"https://ourworldindata.org/rich-poor-working-hours\" target=\"_blank\">increases in productivity</a> and the adoption of regulations that limit working hours. We discuss these and other key drivers behind working hours trends across countries and time in a series of forthcoming posts.{ref}In our <a href=\"https://ourworldindata.org/rich-poor-working-hours\" data-type=\"URL\" data-id=\"https://ourworldindata.org/rich-poor-working-hours\" target=\"_blank\" rel=\"noreferrer noopener\">first post in the series</a>, we discuss how increases in labor productivity have driven a rise in incomes and a decrease in working hours.{/ref}</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/days-of-vacation-and-holidays\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>Why should we care?</h4>\n\n\n\n<p>The evidence presented here comes from decades of work from economic historians and other researchers. Of course, the data is not perfect \u2014 as we explain in a forthcoming post, <a href=\"https://ourworldindata.org/measure-working-hours\" data-type=\"URL\" data-id=\"https://ourworldindata.org/measure-working-hours\" target=\"_blank\" rel=\"noreferrer noopener\">measuring working hours with accuracy is difficult</a>, and surveys and historical records have limitations, so estimates of working hours spanning centuries necessarily come with a margin of error. But for any given country, the changes across time are much larger than the error margins at any point in time: The average worker in a rich country today really does work many fewer hours than the average worker 150 years ago.</p>\n\n\n\n<p>As the economists Diane Coyle and Leonard Nakamura explain, the study of working hours is crucial not only to measure <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable\" target=\"_blank\">macroeconomic productivity</a>, but also to measure economic well-being beyond economic <em>output</em>. A more holistic <a rel=\"noreferrer noopener\" href=\"https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602\" target=\"_blank\">framework for measuring \u2018progress\u2019</a> needs to consider changes in how people are allowed to allocate their time over multiple activities, among which paid work is only one.{ref}Coyle, D. and Nakamura, L. I. (2019). <a rel=\"noreferrer noopener\" href=\"https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602\" target=\"_blank\" data-type=\"URL\" data-id=\"https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3333602\">Toward a Framework for Time Use, Welfare, and Household Centric Economic Measurement.</a> Federal Reserve Bank of Philadelphia Working Paper No. 19-11.{/ref}</p>\n\n\n\n<p>The available evidence shows that, rather than working more than ever, workers in many countries today work <em>much less </em>than in the past 150 years. There are huge inequalities within and across countries, but substantial progress has been made.</p>\n\n\n\n<h3>Do workers in richer countries work longer hours?</h3>\n\n\n\n<p>Economic prosperity in different places across our world today is <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/global-economic-inequality\" target=\"_blank\">vastly unequal</a>. People in Switzerland, one of the richest countries in the world, have an average income that is more than <em>20-times higher</em> than that of people in Cambodia.{ref}We chose Cambodia and Switzerland here because they both also have working hours data available, but the difference in average income can be even more extreme. For instance, people in Qatar have an <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT\" data-type=\"URL\" target=\"_blank\">average income that is <em>1</em></a><a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT\" data-type=\"URL\" target=\"_blank\"><em>1</em></a><a href=\"https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT\" data-type=\"URL\" target=\"_blank\" rel=\"noreferrer noopener\"><em>7</em></a><a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/grapher/real-gdp-per-capita-pennwt?tab=chart&country=CAF~KHM~CHE~QAT\" data-type=\"URL\" target=\"_blank\"><em>-times higher</em></a> than that of people in the Central African Republic.</p>\n\n\n\n<p>These differences refer to GDP per capita measured in international-$ and account for price differences between countries to enable comparisons. You can read more about this <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/what-are-ppps\" target=\"_blank\">here</a>.{/ref} Life in these two countries can <a rel=\"noreferrer noopener\" href=\"https://www.gapminder.org/dollar-street/?max=2755&countries=kh%2Cch&media=image&min=59&topic=homes&zoom=3\" target=\"_blank\">look starkly different</a>.{ref}But life can also look similar, as you see in the pictures of the <a rel=\"noreferrer noopener\" href=\"https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=homes&zoom=3\" data-type=\"URL\" data-id=\"https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=homes&zoom=3\" target=\"_blank\">homes</a>, <a rel=\"noreferrer noopener\" href=\"https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=computers&zoom=3\" data-type=\"URL\" data-id=\"https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=computers&zoom=3\" target=\"_blank\">computers</a>, and <a rel=\"noreferrer noopener\" href=\"https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=phones&zoom=3\" data-type=\"URL\" data-id=\"https://www.gapminder.org/dollar-street/?max=2754&countries=kh%2Cch&media=image&min=1357&topic=phones&zoom=3\" target=\"_blank\">phones</a> of people on similar income levels in the two countries.{/ref}</p>\n\n\n\n<p>When considering such differences in prosperity, a natural question is: who works more, people in richer countries like Switzerland or in poorer ones like Cambodia?</p>\n\n\n\n<p>Looking at the available data, the answer is clear: workers in poorer countries actually tend to work more, and sometimes <em>much</em> more.</p>\n\n\n\n<p>We see that in the chart here, with GDP per capita on the horizontal axis and annual working hours per worker on the vertical axis. </p>\n\n\n\n<p>Countries like Cambodia (which is the country in the very top-left corner) or Myanmar have some of the lowest GDP per capita but highest working hours. In Cambodia the average worker puts in 2,456 hours each year, nearly 900 more hours than in Switzerland (1,590 hours) at the bottom-right of the chart. The extra 900 hours for Cambodian workers means longer work days and many fewer days off.</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita-pwt?tab=chart&stackMode=absolute&time=2019..latest&country=&region=World\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>Working hours tend to decrease as countries become richer</h4>\n\n\n\n<p>There is a link between national income and average working hours, not only across countries at a given point in time \u2014 as shown in the chart above \u2014 but also for individual countries <em>over time.</em> </p>\n\n\n\n<p>Since the Industrial Revolution people in many countries have <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/breaking-the-malthusian-trap\" target=\"_blank\">become richer</a>, and <a href=\"https://ourworldindata.org/working-more-than-ever\" data-type=\"URL\" data-id=\"https://ourworldindata.org/working-more-than-ever\" target=\"_blank\" rel=\"noreferrer noopener\">working hours have decreased dramatically</a> over these last 150 years.</p>\n\n\n\n<p>In the chart here we show this association between incomes and working hours over time, country by country. It is the same chart as above, except now countries\u2019 single data points have become lines, connecting observations over time from 1950 until today.</p>\n\n\n\n<p>The four highlighted countries exemplify how working hours have decreased at the same time that average incomes have increased. Germany, for example, moved far to the right as its GDP per capita increased more than 10-fold (from $5,227 to $51,191), and far to the bottom as working hours decreased by nearly half (from 2,428 hours to 1,386 hours each year).{ref}These trends in GDP per capita are measured in constant international-$ and account for inflation to enable comparisons over time and between countries. You can read more about this <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/economic-growth\" target=\"_blank\">here</a>.{/ref}</p>\n\n\n\n<p>This makes sense: as people’s incomes rise they can afford more of the things they enjoy, including more leisure and less time spent working.</p>\n\n\n\n<p>You can explore this association for other countries by clicking \u201cSelect countries\u201d on the chart.</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/annual-working-hours-vs-gdp-per-capita?tab=chart&xScale=log&stackMode=absolute&time=1950..latest&country=BRA~USA~DEU~TWN&region=World\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>People are able to work less when they work in more productive economies</h4>\n\n\n\n<p>The key driver of rising national incomes and decreasing working hours is productivity growth.</p>\n\n\n\n<p>Productivity refers to the rate at which inputs are turned into outputs. To understand working hours the key metric is <em>labor</em> productivity: the economic return for one hour of work.</p>\n\n\n\n<p>At the most concrete level, labor productivity captures things like the number of breads that a baker bakes in an hour, or the number of cars factory workers assemble in an hour. At the most comprehensive level, it relates the total output of the economy (GDP) to the total labor input (total annual hours worked), giving us the aggregate measure of labor productivity, GDP per hour of work.</p>\n\n\n\n<p>Higher labor productivity is associated with fewer working hours, as shown in the chart here with labor productivity on the horizontal axis and annual working hours on the vertical axis. The chart currently shows data for the latest available year, but you can explore this relationship over time since 1950 by using the blue time slider at the bottom of the chart.</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/productivity-vs-annual-hours-worked?tab=chart&country=&region=World\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<p>We see that the same richer countries with lower working hours we noted before \u2014 like Germany and Switzerland \u2014 have very high labor productivity (69 and 83 $/h, respectively). If workers can produce more with each hour of work, it becomes possible for them to work less.</p>\n\n\n\n<p>Though this doesn\u2019t necessarily mean they <em>actually do</em> work less \u2014 workers in the US and Singapore, for instance, work many more hours than their counterparts in countries with similar productivity.{ref}We explore the differences in working hours between similar, highly productive countries \u2014 and also the differences within those countries \u2014 in forthcoming posts.{/ref}</p>\n\n\n\n<p>In contrast, the countries toward the top-left of this chart have far lower labor productivity \u2014 Cambodia, for example, is at only 3$/h \u2014 and thus workers there need to work many more hours to compensate.</p>\n\n\n\n<h4>At the heart of the link between productivity, incomes, and working hours is technological innovation</h4>\n\n\n\n<p>Technological innovation \u2014 defined broadly here to include physical machines as well as ideas, knowledge, and processes \u2014 makes it possible for each worker to become much more productive. And increases in productivity in turn help drive both increases in incomes and decreases in working hours.{ref}For a discussion of how technology drives productivity growth and a rise in incomes (economic growth), see Romer, P. (1990) <a rel=\"noreferrer noopener\" href=\"https://www.journals.uchicago.edu/doi/abs/10.1086/261725\" target=\"_blank\">Endogenous Technological Change.</a> <em>Journal of Political Economy.</em> For a discussion of the relationship between productivity growth, economic growth, and working hours, see Boppart, T. and P. Krusell (2020) <a rel=\"noreferrer noopener\" href=\"https://www.journals.uchicago.edu/doi/full/10.1086/704071?casa_token=gHAZpu6iXwkAAAAA:WUwiHwVFoOWNFdnfKenDUm9yOtgcjKxwZpohoDcmZk8ZMwMASp86fPHmmd3-r8NLJk-9UKorX7o\" data-type=\"URL\" data-id=\"https://www.journals.uchicago.edu/doi/full/10.1086/704071?casa_token=gHAZpu6iXwkAAAAA:WUwiHwVFoOWNFdnfKenDUm9yOtgcjKxwZpohoDcmZk8ZMwMASp86fPHmmd3-r8NLJk-9UKorX7o\" target=\"_blank\">Labor Supply in the Past, Present, and Future: A Balanced-Growth Perspective</a>. <em>Journal of Political Economy.</em>{/ref}</p>\n\n\n\n<p>A prime example of how tech innovation drives productivity growth is agriculture. As we show in detail in <a href=\"https://ourworldindata.org/crop-yields\" target=\"_blank\" rel=\"noreferrer noopener\">our entry on Crop Yields</a>, innovations like better machinery, crop varieties, fertilizers, and land management have enabled farmers to be <em>much more</em> productive. In the US, for example, farm production per labor hour increased nearly 16-fold from 1948\u20132011.{ref}See Figure 18 on p. 28 of Wang et al (2015) <a href=\"https://www.ers.usda.gov/webdocs/publications/45387/53417_err189.pdf?v=6052.7\" target=\"_blank\" rel=\"noreferrer noopener\">Agricultural Productivity Growth in the United States: Measurement, Trends, and Drivers</a>. <em>USDA Economic Research Report 189.</em>{/ref} This increased productivity enables us to feed a <a href=\"https://ourworldindata.org/grapher/population?country=~OWID_WRL\" target=\"_blank\" rel=\"noreferrer noopener\">rapidly growing population</a>, even while the <a href=\"https://ourworldindata.org/grapher/employment-by-economic-sector?stackMode=relative\" target=\"_blank\" rel=\"noreferrer noopener\">fraction of people working in agriculture</a> is smaller than ever.{ref}The transition of employment out of agriculture to other economic sectors as countries become richer is known as \u2018structural transformation\u2019. You can read more about this in our post <a href=\"https://ourworldindata.org/structural-transformation-and-deindustrialization-evidence-from-todays-rich-countries\" target=\"_blank\" rel=\"noreferrer noopener\"><em>Structural transformation: how did today\u2019s rich countries become \u2018deindustrialized\u2019?</em></a>{/ref}</p>\n\n\n\n<p>The chart here shows the growth in labor productivity, not just for agriculture but for the entire economy. The technological, economic, and social structures in richer countries have enabled workers there to produce more while working less.</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~TWN~GBR~USA~CHE&region=World\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<p>Besides tech innovation, there is evidence that working fewer hours can itself keep productivity higher, making the link between working hours and productivity self-reinforcing. For example, economist <a rel=\"noreferrer noopener\" href=\"https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext\" target=\"_blank\">John Pencavel (2015) studied</a> munitions workers in war-time Britain and found that their productivity stayed high up to a certain threshold of hours, but declined markedly above that threshold.{ref}Pencavel, J. (2015) <a rel=\"noreferrer noopener\" href=\"https://academic.oup.com/ej/article-abstract/125/589/2052/5078088?redirectedFrom=fulltext\" target=\"_blank\">The productivity of working hours</a>. <em>The Economic Journal.</em>{/ref} We\u2019ve probably all experienced the drop in productivity that comes at the end of a very long day of work.</p>\n\n\n\n<h4>What we learn from this</h4>\n\n\n\n<p>The data show that it is workers in poorer countries who tend to work more, and sometimes <em>a lot</em> more, than those in richer countries.</p>\n\n\n\n<p>This has large implications for the way we think about the economic progress made in the last two centuries and the nature of inequality between countries today.</p>\n\n\n\n<p>It means that residents of today\u2019s poorer countries like Cambodia and Myanmar \u2014 and also of today\u2019s richer countries in the past when they were poor \u2014 are not just <em>consumption</em> poor, often unable to afford necessities like food and medicine. It means they are also <em>leisure</em> poor: because productivity is low and they must work so much just to scrape by, they can\u2019t afford to spend much time improving their condition, becoming educated, or simply enjoying leisure time.</p>\n\n\n\n<p>That people in poorer countries work so much more than in richer countries shows that differences in prosperity are not due to differences in work ethic \u2014 they are largely due to differences in circumstance and opportunity. As we ask in <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/talent-is-everywhere-opportunity-is-not\" target=\"_blank\">another post</a>, \u201cwhat would have been the chances for Steve Jobs if he was born in the Central African Republic?\u201d No matter how hard he worked or how smart he was, it is difficult to imagine that Steve Jobs would\u2019ve been able to realize his potential with such a steep mountain of inequality to climb.</p>\n\n\n\n<p>We also see what <em>the world </em><a rel=\"noreferrer noopener\" href=\"https://www.imf.org/en/Publications/WP/Issues/2018/12/07/Invisible-Geniuses-Could-the-Knowledge-Frontier-Advance-Faster-46383\" target=\"_blank\">misses out on</a> when exceptionally talented people, including all the brilliant but underprivileged people in today\u2019s poorest countries, don\u2019t have the opportunity to realize their potential.{ref}Agarwal, R. and Gaule, P. (2020) <a rel=\"noreferrer noopener\" href=\"https://www.aeaweb.org/articles?id=10.1257/aeri.20190457&&from=f\" target=\"_blank\">Invisible Geniuses: Could the Knowledge Frontier Advance Faster?</a><em> American Economic Review: Insights.</em>{/ref}</p>\n\n\n\n<p>Finding ways to raise productivity is therefore not just key to increasing production, but also to the reduction in working hours that is necessary for a society to flourish.</p>\n\n\n\n<h3>How are working hours measured and what can we learn from the data?</h3>\n\n\n\n<p>Work is a central part of our lives. It is something we do almost every day, for much of the day, for decades on end. Because it is so central, looking closely at how much time we spend working can tell us a lot about our lives and the societies we live in.</p>\n\n\n\n<p>The data on working hours shows, for example, that rather than working more than ever \u2014 as is so commonly believed \u2014 people in many countries today <a href=\"https://ourworldindata.org/working-more-than-ever\" target=\"_blank\" rel=\"noreferrer noopener\">work <em>much less</em> than in the past 150 years</a>.</p>\n\n\n\n<p>Working less means being able to spend time becoming <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/global-education\" target=\"_blank\">more educated</a>, or simply <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/time-use-living-conditions\" target=\"_blank\">enjoying more leisure time</a>. This is substantial progress, but there is still <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/rich-poor-working-hours\" data-type=\"URL\" data-id=\"https://ourworldindata.org/rich-poor-working-hours\" target=\"_blank\">huge inequality across countries</a>, and progress still to make.</p>\n\n\n\n<p>To understand these changes in societies and people\u2019s lives over time, and the substantial differences we see in the world today, it is crucial to measure and study how much time people spend working.</p>\n\n\n\n<p>How are working hours actually measured? Where does the data come from, and how can researchers reconstruct long-run trends?</p>\n\n\n\n<p>Here we provide an overview of the main data sources, compare the data, and explain the relevant differences and measurement limitations.</p>\n\n\n\n<h4>How are working hours measured?</h4>\n\n\n\n<h5>Surveys</h5>\n\n\n\n<p>Surveys are the primary way to collect data on working hours. They are typically conducted by national statistical agencies and come in three main types: labor force surveys, establishment surveys, and time use surveys. These surveys all provide an important perspective on working hours, but there are some key differences.</p>\n\n\n\n<p>Labor force surveys collect data on employment status and time spent working by asking individual workers themselves. Of the survey types, these provide the most comprehensive perspective, covering hours <em>actually</em> worked in all economic sectors as part of both formal and <a href=\"https://ourworldindata.org/grapher/informal-employment-of-total-non-agricultural-employment\" target=\"_blank\" rel=\"noreferrer noopener\">informal employment</a>, full-time and part-time, as well as self-employment and unpaid family work.{ref}Hours actually worked means hours spent directly on work and excludes things like annual leave, sick leave, public holidays, meal breaks, and commuting time. Unpaid family work in this case generally includes market-oriented work, such as for the family business, but not other unpaid work at home such as childcare, cooking, and cleaning. Since the latter type of unpaid work is typically performed by women, this has large implications for understanding gender differences in labor. We discuss these issues as part of <a href=\"https://ourworldindata.org/female-labor-supply#definitions-measurement\" target=\"_blank\" rel=\"noreferrer noopener\">our entry on Women\u2019s Employment</a>.{/ref} But labor force surveys only cover residents of a country above a certain age (usually 15), which depending on the country might exclude a non-trivial number of workers.{ref}Only covering resident workers means that any <a href=\"https://ec.europa.eu/eurostat/cache/digpub/eumove/bloc-2c.html?lang=en#:~:text=In%202019%2C%20the%20largest%20number,and%20Belgium%20(50%20000).\" target=\"_blank\" rel=\"noreferrer noopener\">cross-border workers</a> are excluded. Only covering workers above a certain age means that any child laborers are excluded. While the incidence of child labor has been going down over time, especially in high-income countries, there are still an <a href=\"https://ourworldindata.org/child-labor\" target=\"_blank\" rel=\"noreferrer noopener\">estimated 265 million working children</a> in the world (almost 17% of the worldwide child population).{/ref}</p>\n\n\n\n<p>Establishment surveys collect data on employment and working hours as reported by employers.{ref}Employers include businesses, non-profits, some government agencies, and other organizations that pay a wage.{/ref} But because hours are reported by employers, these surveys often only cover paid or contractual hours and exclude self-employment, informal work, and some smaller firms.{ref}Unlike hours actually worked, paid or contractual hours typically include some time <em>not</em> spent working, such as during sick leave, and fail to include time spent working that wasn’t paid or planned, such as overtime.{/ref} On the other hand, establishment surveys provide more detail on the industry of work than other surveys, and are more consistent with how GDP is measured, making them useful for <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/grapher/labor-productivity-per-hour-pennworldtable?tab=chart&time=1950..2017&country=AUS~BEL~BRA~KHM~CHL~CHN~DEU~IND~ZAF~KOR~CHE~TWN~GBR~USA&region=World\" target=\"_blank\">studying labor productivity</a>.</p>\n\n\n\n<p>Time use surveys collect data on how individuals spend their time \u2014 <a href=\"https://ourworldindata.org/time-use-living-conditions#daily-activities-similarities-and-differences-across-countries\" data-type=\"URL\" data-id=\"https://ourworldindata.org/time-use-living-conditions#daily-activities-similarities-and-differences-across-countries\" target=\"_blank\" rel=\"noreferrer noopener\">down to the minute</a> \u2014 across a number of activities in a typical day, including paid work.{ref}Activities also include unpaid household work, school, leisure time, eating, and sleeping.{/ref} This level of granularity provides a useful complement to the other surveys, but as a trade-off time use surveys sample fewer people and are conducted less frequently and by fewer countries.</p>\n\n\n\n<h5>National accounts</h5>\n\n\n\n<p>To get the most comprehensive perspective on working hours possible, many countries aggregate data from these surveys with data from other sources \u2014 such as censuses, tax records, and social security registers \u2014 in an economic measurement framework called national accounts.</p>\n\n\n\n<p>National accounts, and the surveys they rely on, are standardized to a degree across countries, which can facilitate international comparisons.{ref}By organizations such as the <a rel=\"noreferrer noopener\" href=\"https://unstats.un.org/unsd/nationalaccount/sna.asp\" target=\"_blank\">United Nations</a>, <a href=\"https://ilostat.ilo.org/resources/concepts-and-definitions/description-hours-of-work/\" data-type=\"URL\" data-id=\"https://ilostat.ilo.org/resources/concepts-and-definitions/description-hours-of-work/\" target=\"_blank\" rel=\"noreferrer noopener\">International Labor Organization (ILO)</a>, <a rel=\"noreferrer noopener\" href=\"https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en\" target=\"_blank\">OECD</a>, and <a rel=\"noreferrer noopener\" href=\"https://ec.europa.eu/eurostat/web/microdata/european-union-labour-force-survey\" target=\"_blank\">Eurostat</a>.{/ref}</p>\n\n\n\n<p>But these comparisons often have limitations because many countries still implement the methods in different ways. For instance, countries might bring together different data in their national accounts, or aggregate it differently. And many countries don\u2019t have the capacity to conduct comprehensive surveys of their labor force and produce national accounts-based statistics, giving a more limited view of work there.{ref}For further discussion of different sources and their comparability, see the methods guides of the <a rel=\"noreferrer noopener\" href=\"https://www.oecd-ilibrary.org/sites/33bc1355-en/index.html?itemId=/content/component/33bc1355-en\" target=\"_blank\">OECD</a> and the <a rel=\"noreferrer noopener\" href=\"https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite\" target=\"_blank\">Total Economy Database</a> and the work of <a href=\"https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344\" data-type=\"URL\" data-id=\"https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344\" target=\"_blank\" rel=\"noreferrer noopener\">Bick, Br\u00fcggemann, and Fuchs-Sch\u00fcndeln (2019)</a>.{/ref}</p>\n\n\n\n<h4>How do researchers reconstruct long-run historical trends?</h4>\n\n\n\n<p>Comprehensive, cross-country data on working hours just isn\u2019t available for the years before the mid 20th century. But researchers like <a href=\"http://www.sciencedirect.com/science/article/pii/S0014498307000058\" target=\"_blank\" rel=\"noreferrer noopener\">Huberman and Minns (2007)</a>{ref}Huberman, M. and Minns, C. (2007) <a href=\"http://www.sciencedirect.com/science/article/pii/S0014498307000058\" target=\"_blank\" rel=\"noreferrer noopener\">The times they are not changin\u2019: Days and hours of work in Old and New Worlds, 1870\u20132000.</a> <em>Explorations in Economic History.</em>{/ref} have been able to fill some of the gap by reconstructing long-run trends for a selection of countries. How do they do it?</p>\n\n\n\n<p>Through often painstaking effort, researchers have been able to find and piece together the relevant historical records that do exist. In the work of Huberman and Minns, one of the key sources for historical data on many countries is a <a href=\"https://catalog.hathitrust.org/Record/008420895\" target=\"_blank\" rel=\"noreferrer noopener\">report from the US Department of Labor</a> published in 1900.{ref}U.S. Department of Labor (1900) <a href=\"https://catalog.hathitrust.org/Record/008420895\" target=\"_blank\" rel=\"noreferrer noopener\">Fifteenth Annual Report of the Commissioner of Labor: Wages in Commercial Countries. 2 vols.</a> Washington, DC.{/ref} The report compiled the records of many thousands of workers across numerous sectors from establishment surveys in 88 countries and territories. To reconstruct the trends in later years, Huberman and Minns pulled together data from the International Labor Organization, the work of peer researchers, and other sources.{ref}The original sources are: 1870\u20131913: Huberman (2004) [in turn relying on the US Department of Labor Fifteenth Annual Report, 1900]; 1929\u20131938: International Labor Organization (1934\u201339), except for Canada (Ostry and Zaidi, 1972), U.S. (Jones, 1963; Owen, 1988), and Australia (Butlin, 1977); 1950\u20132000: University of Groningen and the Conference Board GGDC Total Economy Database (2005).{/ref}</p>\n\n\n\n<p>This was an impressive feat of reconstruction, but historical records like this do have limitations. For instance, as exhaustive as they were, the establishment-level records used by Huberman and Minns still excluded agricultural work, part-time work, and many smaller firms.</p>\n\n\n\n<h4>How does the data from different sources compare?</h4>\n\n\n\n<p>The work by Huberman and Minns is an important example of how researchers often combine and adjust underlying sources to produce one-off cross-country estimates. Another important study is the one of Bick, Br\u00fcggemann, and Fuchs-Sch\u00fcndeln (2019),{ref}Bick, A., Br\u00fcggemann, B., and Fuchs-Sch\u00fcndeln, N. (2019) <a rel=\"noreferrer noopener\" href=\"https://onlinelibrary.wiley.com/doi/abs/10.1111/sjoe.12344\" target=\"_blank\">Hours Worked in Europe and the United States: New Data, New Answers.</a> <em>The Scandinavian Journal of Economics.</em>{/ref} who further standardized labor force surveys to enhance comparability for a selection of countries.</p>\n\n\n\n<p>Besides these one-off estimates, several international organizations and research centers aggregate the working hours estimates published by national statistical agencies into cross-country datasets. The two most important datasets come from the <a rel=\"noreferrer noopener\" href=\"https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#\" target=\"_blank\">OECD</a> and the <a rel=\"noreferrer noopener\" href=\"https://www.rug.nl/ggdc/productivity/pwt/?lang=en\" target=\"_blank\">Penn World Table</a> (PWT). These both draw on national accounts estimates when available, but they can differ in the other sources they use and their method of aggregation.{ref}PWT sources its working hours data from <a rel=\"noreferrer noopener\" href=\"https://www.conference-board.org/data/economydatabase/total-economy-database-productivity\" target=\"_blank\">The Conference Board\u2019s Total Economy Database</a> (TED). For more details on the underlying sources, see the <a rel=\"noreferrer noopener\" href=\"https://www.conference-board.org/retrievefile.cfm?filename=TED_SMDetailed_nov2017.pdf&type=subsite\" target=\"_blank\">TED guide</a> and the <a rel=\"noreferrer noopener\" href=\"https://stats.oecd.org/Index.aspx?DataSetCode=ANHRS#\" target=\"_blank\">OECD database</a>.{/ref}</p>\n\n\n\n<p>In the chart you can compare annual working hours data from these four datasets. The data is shown one country at a time \u2014 with France currently selected. You can look at other countries by clicking \u2018Change country\u2019 on the chart, but note that not all sources publish data for every country.</p>\n\n\n\n<p>As expected, there are differences between the sources. In 2000, for instance, Bick et al. estimates 1,642 hours of work for French workers, OECD estimates 1,558 hours, PWT estimates 1,550 hours, and Huberman and Minns estimates 1,443 hours. These differences are due to the use of different underlying sources and methods. Bick et al. use only labor force surveys; the others all rely primarily on national accounts data, but which nonetheless still have differences.</p>\n\n\n\n<p>It\u2019s also clear that these differences between sources are quite small when compared to the huge changes over the longer run. The difference between sources in 2000 is at most 200 hours, while the historical data from Huberman and Minns shows that from 1870 to 2000 annual working hours in France decreased by <em>1,725 hours</em> (from 3,168 to 1,443 hours).</p>\n\n\n\n<iframe src=\"https://ourworldindata.org/grapher/compare-sources-working-hours\" loading=\"lazy\" style=\"width: 100%; height: 600px; border: 0px none;\"></iframe>\n\n\n\n<h4>What does this tell us about the study of working hours?</h4>\n\n\n\n<p>The analysis here shows that working hours data can have limitations \u2014 due to differences in the sources or the way the method is implemented \u2014 but that what these matter for our interpretation of the data depends on the context.</p>\n\n\n\n<p>In a context where precise comparisons of similar countries is important, smaller differences between sources can really matter. This is why to compare recent working hours levels in the US and Europe, Bick et al. used only labor force surveys, which they standardized even further to maximize cross-country comparability. But as a trade-off, it was only possible to look at a small selection of richer countries.</p>\n\n\n\n<p>In a context where we want to focus on a larger scale \u2014 such as the <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/working-more-than-ever\" target=\"_blank\">long-run historical trends</a> we see in the chart \u2014 the limitations of the sources are not large enough to undermine our conclusions.</p>\n\n\n\n<p>Large international datasets like PWT do not have the highest levels of cross-country comparability, but they allow us to look at many more countries across the world and uncover broad and important trends, such as the <a rel=\"noreferrer noopener\" href=\"https://ourworldindata.org/rich-poor-working-hours\" data-type=\"URL\" data-id=\"https://ourworldindata.org/rich-poor-working-hours\" target=\"_blank\">large differences in working hours between the richest and poorest countries</a>.{ref}We gain further confidence in these conclusions when they are echoed by research that focuses only on more standardized, comparable sources for a necessarily smaller set of countries, as in the work by <a rel=\"noreferrer noopener\" href=\"https://www.aeaweb.org/articles?id=10.1257/aer.20151720\" target=\"_blank\">Bick, Fuchs-Sch\u00fcndeln, and Lagakos (2018)</a>.{/ref}</p>\n\n\n\n<p>PWT and OECD are also useful in contexts where we want an exhaustive picture of the trends in individual countries, since they are often based on national accounts that bring together data from many sources to give a comprehensive perspective on working hours.</p>\n\n\n\n<p>The data on working hours isn\u2019t perfect, and it\u2019s important to understand the limitations, but it can still tell us a lot about our lives and the world.</p>\n\n\n\n<h2>Data Sources</h2>\n\n\n\n<h4>OECD Time Use Database and Gender Data Portal</h4>\n\n\n\n<ul><li><strong>Data:</strong> Time spent on various activities per day, by sex</li><li><strong>Geographical coverage:</strong> OECD countries plus China, India, and South Africa</li><li><strong>Time span:</strong> from 1998 onwards</li><li><strong>Available at:</strong> <a rel=\"noreferrer noopener\" href=\"https://stats.oecd.org/Index.aspx?DataSetCode=TIME_USE\" target=\"_blank\">Time Use Database</a>; <a rel=\"noreferrer noopener\" href=\"https://www.oecd.org/gender/data/balancingpaidworkunpaidworkandleisure.htm\" target=\"_blank\">Gender Data Portal</a></li></ul>\n\n\n\n<h4>Centre for Time Use Research</h4>\n\n\n\n<ul><li><strong>Data:</strong> Time spent on various activities per day; self-reported levels of enjoyment</li><li><strong>Geographical coverage:</strong> 30 countries across the world</li><li><strong>Time span:</strong> from 1964 onwards</li><li><strong>Available at:</strong> <a rel=\"noreferrer noopener\" href=\"https://www.timeuse.org/\" target=\"_blank\">https://www.timeuse.org/</a><ul><li>Gershuny, J., and Sullivan, O. (2019). What We Really Do All Day: Insights from the Centre for Time Use Research. Penguin UK.</li></ul></li></ul>\n\n\n\n<h4>US Bureau of Labor Statistics</h4>\n\n\n\n<ul><li><strong>Data:</strong> Time spent with others per day</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> from 2003 onwards</li><li><strong>Available at:</strong> <a rel=\"noreferrer noopener\" href=\"https://www.bls.gov/tus/datafiles-0319.htm\" target=\"_blank\">https://www.bls.gov/tus/datafiles-0319.htm</a></li></ul>\n\n\n\n<h4>IPUMS USA</h4>\n\n\n\n<ul><li><strong>Data:</strong> Percentage of Americans living alone, by age</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> from 1900 onwards</li><li><strong>Available at:</strong> <a rel=\"noreferrer noopener\" href=\"https://ipums.org/projects/ipums-usa/d010.v8.0\" target=\"_blank\">https://ipums.org/projects/ipums-usa/d010.v8.0</a></li></ul>\n\n\n\n<h4>Dotti Sani and Treas (2016)</h4>\n\n\n\n<ul><li><strong>Data:</strong> Time parents spent with children per day</li><li><strong>Geographical coverage:</strong> United States and select countries in Europe</li><li><strong>Time span:</strong> 1965\u20132012</li><li><strong>Available at:</strong> Dotti Sani, G. M., and Treas, J. (2016). <a rel=\"noreferrer noopener\" href=\"https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Dotti+Sani+and+Treas+2016+Educational+Gradients+in+Parents%27+Child-Care+Time+Across+Countries%2C+1965%E2%80%932012&btnG=\" target=\"_blank\">Educational gradients in parents\u2019 child\u2010care time across countries, 1965\u20132012</a>. Journal of Marriage and Family.</li></ul>\n\n\n\n<h4>Sayer, Bianchi, and Robinson (2004)</h4>\n\n\n\n<ul><li><strong>Data:</strong> Time parents spent on primary childcare activities per day</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> 1965\u20131998</li><li><strong>Available at:</strong> Sayer, L., Bianchi, S., and Robinson, J. (2004). <a href=\"https://www.journals.uchicago.edu/doi/pdf/10.1086/386270\" target=\"_blank\" rel=\"noreferrer noopener\">Are Parents Investing Less in Children? Trends in Mothers\u2019 and Fathers\u2019 Time with Children</a>. American Journal of Sociology.</li></ul>\n\n\n\n<h4>Huberman and Minns (2007)</h4>\n\n\n\n<ul><li><strong>Data:</strong> Annual hours of full-time production workers (male and female) in non-agricultural activities; days off from work for vacations and holidays</li><li><strong>Geographical coverage:</strong> United States, Australia, Canada, and select countries in Europe</li><li><strong>Time span:</strong> 1870\u20132000</li><li><strong>Available at:</strong> Huberman, M. and Minns, C. (2007). <a rel=\"noreferrer noopener\" href=\"https://www.sciencedirect.com/science/article/abs/pii/S0014498307000058#!\" target=\"_blank\">The times they are not changin\u2019: Days and hours of work in Old and New Worlds, 1870\u20132000.</a> Explorations in Economic History.</li></ul>\n\n\n\n<h4>Penn World Table</h4>\n\n\n\n<ul><li><strong>Data:</strong> Average annual hours worked by persons engaged; number of persons engaged; real and PPP-adjusted GDP in US millions of dollars</li><li><strong>Geographical coverage:</strong> Countries across the world</li><li><strong>Time span:</strong> from 1950 onwards</li><li><strong>Available at:</strong> <a rel=\"noreferrer noopener\" href=\"https://www.rug.nl/ggdc/productivity/pwt/\" target=\"_blank\">https://www.rug.nl/ggdc/productivity/pwt/</a><ul><li>Feenstra, R. C., Inklaar, R., and Timmer, M.P. (2015). <a rel=\"noreferrer noopener\" href=\"https://www.aeaweb.org/articles?id=10.1257/aer.20130954\" target=\"_blank\">The Next Generation of the Penn World Table.</a> American Economic Review.</li></ul></li></ul>\n\n\n\n<h4>Eurostat</h4>\n\n\n\n<ul><li><strong>Data:</strong> Time use survey data</li><li><strong>Geographical coverage:</strong> Europe</li><li><strong>Time span:</strong> from 1998 onwards</li><li><strong>Available at:</strong> <a rel=\"noreferrer noopener\" href=\"https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#\" target=\"_blank\">https://ec.europa.eu/eurostat/data/database?node_code=tus_00npaywork#</a></li></ul>\n\n\n\n<h4>Ramey and Francis (2009)</h4>\n\n\n\n<ul><li><strong>Data:</strong> Self-reported enjoyment of various activities; time spent on various activities, by sex and age; days of work lost to sickness</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> 1900\u20132005</li><li><strong>Available at:</strong> Ramey, V. A., and Francis, N. (2009). <a rel=\"noreferrer noopener\" href=\"https://www.aeaweb.org/articles?id=10.1257/mac.1.2.189\" target=\"_blank\">A century of work and leisure</a>. American Economic Journal: Macroeconomics.</li></ul>\n\n\n\n<h4>Costa (2000)</h4>\n\n\n\n<ul><li><strong>Data:</strong> Working hours, by sex; number of working days; wages</li><li><strong>Geographical coverage:</strong> United States</li><li><strong>Time span:</strong> 1890\u20131991</li><li><strong>Available at:</strong> Costa, D. L. (2000). <a rel=\"noreferrer noopener\" href=\"https://www.jstor.org/stable/10.1086/209954?seq=1\" target=\"_blank\">The Wage and the Length of the Work Day: From the 1890s to 1991.</a> Journal of Labor Economics.</li></ul>\n", "protected": false }, "excerpt": { "rendered": "How do people across the world spend their time? How do daily activities differ across countries, and how do these differences matter for people\u2019s lives? Explore data and research on time use.", "protected": false }, "date_gmt": "2020-11-29T21:12:40", "modified": "2023-01-16T21:06:15", "template": "", "categories": [ 52, 236 ], "menu_order": 112, "ping_status": "closed", "authors_name": [ "Esteban Ortiz-Ospina", "Charlie Giattino", "Max Roser" ], "modified_gmt": "2023-01-16T21:06:15", "comment_status": "closed", "featured_media": 38050, "featured_media_paths": { "thumbnail": "/app/uploads/2020/12/thumbnailTime-Use-by-Country-OECD-150x79.png", "medium_large": "/app/uploads/2020/12/thumbnailTime-Use-by-Country-OECD-768x402.png" } } |