40 hours. That’s what we think of as a typical workweek. 8 hours per day. 5 days per week. Perhaps the widespread practice of working from home during the pandemic (as well as the abnormal schedule changes for those unable to work from home), has led some to rethink the nature of the workweek. But the truth is that the workweek has always been evolving.
Take this chart, for example. It comes from Our World in Data (be sure to read their excellent related essay as well), and the historical data comes from a paper by Huberman and Minns. I’ve singled out 4 countries, but you can add others at the OWiD link.
The historical declines are dramatic. This is especially true in Sweden. The average Swedish worker labored for over 3,400 hours per year in 1870. Today, that’s down to 1,600 hours. In other words, the typical Swede works less than half as many hours as her historical counterpart. Wow! The decline for the US is not quite as dramatic, but still astonishing: a US worker today labors for only about 57% of the hours of his 1870 predecessor.
It’s tempting to focus on the differences across countries today: the average worker in the US works about 250 hours more than the average French worker. That’s 6 weeks of vacation! And as recently as 1980, the US and France were roughly equal on this measure. We might also wonder why these historical changes happened. For a very brief introduction to the research, I recommend the last section of this essay by Robert Whaples.
But still, the historical declines are dramatic, even if we in the US haven’t seen much improvement in the past generation (and those poor Swedes, working 100 hours per year more than 40 years ago).
I think another natural question to ask is whether GDP data is distorted, at least as a measure of well being, given these differences in working hours. The answer is partially. Let’s look at the data!
Here’s a chart I created which plots both GDP per hour worked (from the OECD) and GDP per Capita (from the World Bank). Both figures are for 2019, and both are adjusted for purchasing power parity. So it should be roughly comparable. I’ve also removed a few countries where their GDP might be distorted due to the country’s “tax haven” status, such as Ireland.
So what does this chart tell us? Let’s focus on the US to understand it. When measure by GDP per Capita, the US is the second country on the list, such behind oil-rich Norway. But in terms of GDP per Hour Worked, the US falls slightly to fourth place — Denmark is now clearly ahead of us, and Belgium is just a little higher.
Also note that while the US still ranks very high, the gap with some countries is notably closed. For example, as measured by GDP per Capita, the US is 32% above France. As measured by GDP per Hour Worked, the gap is just 6.3%. Remember, those French workers take an extra 6 weeks off. Their enjoyment of that leisure is not counted in GDP. Similarly Sweden: the US is 17% above without adjusting for hours of work, but only 2.7% above with the adjustment. The Swedes also enjoy their extra weeks of vacation.
However, it’s not just the US where GDP per Capita overstates how well off we are. You can use the trend line I’ve plotted to on the chart to roughly say that countries above the line have their GDP overstated without adjusting for hours of work. That includes not just the US, but also Australia, Canada, South Korea, and New Zealand. While “the overworked American,” is something of a cliche, there is some truth to it, though we are certainly not alone. Similarly, countries like France, Belgium, and Spain aren’t getting their full due as measured purely by GDP per Capita.
Look more closely at France. As measured by GDP per Capita, they are just below Canada and just above the UK. But when we factor in their shorter hours of work, they are 28% great than Canada and 16% ahead of the UK. So much for the lazy Frenchman! French workers may enjoy their extra time off from work (and no doubt this is in large part due to the legal 35-hour workweek in France, so it may not be what every French worker wants), but they are still highly productive workers.
But the bottom line here, I think, is that GDP per Capita is a good rough estimate of well being, even without adjusting for hours worked. For some countries, such as the US or France, it may not paint a complete picture. But perhaps for many Americans today, work itself is partially a form of leisure (not all Americans of course, but many). I am probably a biased source though, as I take a few moments away from my Spring Break (woo-hoo!) to… write more about economics.
Very interesting, thanks.
At my old company we had a number of European ex-pats. I got the impression that life is truly mellower on the Continent. You have a smaller house and smaller car than in US, but no worries about healthcare or paying for college (provided you pass the entrance exams), or pension, and much more vacation time.
(I don’t know if any of that has changed in last five years; and Europe for some reason is way behind in vaccinations)
Interesting article Jeremy thanks. I really like your stuff on twitter. The comments by Scott regarding Europe are still valid I believe. GDP per Capita as a measure is understandable as it is something we can fairly easily count but I’m not sure that for many in the UK or the US (I stand to be corrected of course) it tells us much about the average/median person or household and their well being. I think this is partly because of income and wealth inequality but other factors too.
Thanks for the comment. I absolutely agree that we should look at multiple measures of income and well-being, including median income. But doing that for the OECD won’t change the overall picture much: the UK is right around the overall OECD median income, and the US is still in the top 5.
Click to access 45ae3dae-en.pdf
I also added a chart on Twitter with median income: