So we work harder in Canada, eh?

In a recent blog post, Noah Smith points to some graphs posted by my old friend David Andolfatto. David's graphs show a widening gap between Canadian and US labour force participation rates, with the Canadian rate now outstripping the US rate by some margin.

David - having learnt the hard way what happens to people who give their opinions too freely - is careful not to speculate on what might be behind these trends. Noah Smith has no such reservations. His interpretation of these graphs is that "young and prime-age Canadians work more than do their American counterparts."

In fact, the graphs show no such thing. Let me just list all of the things wrong with seeing a difference in labour force participation rates and concluding that Canadians "work more."

First, the labour force participation rate is equal to the percentage of people employed plus the percentage of people unemployed. A higher labour force participation rate does not necessarily mean that more people are employed; it could mean that more people are unemployed.

"Unemployment" is a word with a precise technical meaning. Typically a person is considered to be unemployed if they are available, and looking, for work. However the exact definition of "available for work" and "looking for work" varies from country to country. Thus unemployment rates (and hence participation rates) are not always strictly comparable.

However, even if the Canadian employment rate was higher than the US employment rate, one still could not conclude that people are working harder. The total amount of work in the economy is equal to the number of people working times the average hours worked. The average hours that people work in Canada are lower than the average hours in the US. There are lots of reasons why this might be so:

• In the US, some employers provide health insurance, which creates large fixed costs per worker. For any firm providing health insurance for its workers, it is cheaper to hire fewer workers and have them work long hours as opposed to hiring more workers and having them work fewer hours.
• In the US, illegal immigration has brought large numbers of unskilled workers into the country, driving down the cost of domestic help. I don't know of any paper that proves this is a reason why Americans work longer hours, but it is an explanation that gets thrown around by Canadian economists over drinks.
• In the US, there is less of a social safety net, and less security of unemployment, due to lower rates of unionization/different labour laws. If I had no job security, I'd work longer hours too.
• Quebec. For some reason the Quebecois work shorter hours than other Canadians, pulling down the Canadian average.
• In Canada, taxes are, for the most part, individually based. In a Canadian couple, the "secondary earner" typically faces a lower marginal tax rate than the "primary earner", which creates an incentive for both partners to participate in the labour market. In the US, the tax unit is the married couple. In the US, it makes sense for whoever has the higher wage rate to increase their labour effort (payroll taxes may also matter here too).

There are other reasons to be careful about drawing any conclusions about how hard people work from data on labour force participation rates: the very high incarceration rates in the US have a measurable impact on unemployment rates, especially for young males. Canada and the US differ in terms of participation in higher education, too, which also impacts unemployment and employment rates. 

I wish I could point to a nice reference for all of this, but I can't. This 2009 OECD study gives a general intro to the subject of hours worked (here), and Craig Riddell wrote some nice papers a while ago that pick apart a bunch of differences between the Canadian and US labour markets: e.g. here. 

Comments

[Brenb]

Frances, you are stretching a bit to make this point.

First , there is no need to include "even if" the employment rate in Canada is higher. It IS higher for every demographic group except the elderly, and is over 3 percentage points higher in the aggregate.

Second, in 2011, the average employed person worked 34 hours fewer hours in Canada than the U.S. (about 2% less), essentially a week's worth of work. http://www.bls.gov/ilc/intl_gdp_capita_gdp_hour.htm#table07
Given the gap in employment rates, Canada is working more hours per capita.

Finally, the high incarceration rate in the U.S. probably LOWERS their measured unemployment rate by removing a chunk of the male population with less than a high school degree.

[Brenb]

I agree with you however that Noah Smith knows nothing about Canada!

[Frances Woolley]

Brenb - "he high incarceration rate in the U.S. probably LOWERS their measured unemployment rate by removing a chunk of the male population with less than a high school degree"

Yes, it does. That's the point. Maybe one reason why Canada's labour force participation rate is higher is that our low skill poor people with mental health issues are out there looking for work instead of being behind bars.

"there is no need to include "even if" the employment rate in Canada is higher"

There is a need if one is writing quickly in moments stolen from one's day job and can't be bothered to look up the employment numbers.

[Phil Koop]

Who is this David "Aldofatto" of whom you speak?

[FR Woolley]

Oops. Thanks, Phil.

[Min]

Maybe Canadians work smarter, eh? :)

[Noah Smith]

Frances, look carefully at the numbers. The difference in hours worked does not make up for the difference in LFPR combined with the difference in unemployment. You are right that if unemployment is measured differently in the two countries, then this data will be skewed. But I also suspect that hours worked data are much *more* mismeasured, especially because it's from the OECD dataset, which is notoriously unreliable.

[Frances Woolley]

Noah - thanks for the link on your blog and for taking the time to respond. Yes, I should probably have read your post more carefully before getting into a twitter fight!

I'm still not convinced that Canadians are working more than Americans. There are just too many quirks in the data. E.g., look at the data for women. In Canada employed women are typically take a year of maternity/parental leave, and are counted as being employed while they're on leave (lots of men take parental leave also, but they aren't entitled to quite a full year). If that's on average one year of leave per employed woman, that's enough to make a difference in the participation rate numbers. (As you know, there is far less maternity/parental leave in the US, and I don't know if people on leave are counted as employed on your side of the border.) I would strongly suspect that people on maternity/parental leave don't figure into annual hours/worker calculations however.

Also if we're talking about comparing work effort across specific age groups, it's hours in those age groups that matter, not overall hours.

Plus, as you say, the OECD hours data is problematic, and the OECD explicitly cautions against using the data for using x-country comparisons.

"The concept used is the total number of hours worked over the year divided by the average number of people in employment. The data are intended for comparisons of trends over time; they are unsuitable for comparisons of the level of average annual hours of work for a given year, because of differences in their sources. Part-time workers are covered as well as full-time workers.

But your basic point, that tax cuts don't necessarily increase work effort - I completely agree. Income effects can easily outweigh substitution effects - and indeed a tax cut might not decrease marginal tax rates at all. See, e.g., this post http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/04/the-impact-of-tax-cuts-on-government-revenues.html.

By the way we don't have a nationalized health care system - we have government run health insurance, but much provision is through the private sector - i.e. doctors are typically self-employed professionals.

[Brenb]

Frances,

The U.S. labour force participation rate is calculated as a percent of non-institutional population - those incarcerated are not included.

As for "can't be bothered to look up employment numbers"... well la-dee-da

[Mike Sax]

"In Canada employed women are typically take a year of maternity/parental leave, and are counted as being employed while they're on leave (lots of men take parental leave also, but they aren't entitled to quite a full year). If that's on average one year of leave per employed woman, that's enough to make a difference in the participation rate numbers. (As you know, there is far less maternity/parental leave in the US, and I don't know if people on leave are counted as employed on your side of the border.) I would strongly suspect that people on maternity/parental leave don't figure into annual hours/worker calculations however."

Geez Frances. I may not know for sure if Canadians work better but the sure have it better than we Americans. For me, there are just so many questions begged in this whole enterprise of Prescott's. I mean it's assumed that the goal is to have the citizens that put in the longest hours. Maybe in Europe they have such great benefits, they can maintain the same standard of living working much fewer hours. Hardly seems like a reason to want to go out and cut taxes so we can go back to working harder.

[FR Woolley]

Mike - agreed. The first of Nick's posts on micro-foundations pretty much exactly sums up my views about the purpose of economics: ultimately, the best econ policies are the ones that facilitate people living good lives. Not obvious at all that good life=more time in paid employment/self-employment.

[genauer]

Well, according to some statistics,

the Greeks are working a lot harder than the Germans http://www.bbc.co.uk/news/magazine-17155304

just that it somehow doesn' t show up as GDP generated.

But since Noah Smith is Krugtron, he will always be "just right about nearly everything" : - )

just not anything in real life

[Neil]

"In a Canadian couple, the "secondary earner" typically faces a lower marginal tax rate than the "primary earner", which creates an incentive for both partners to participate in the labour market."

I'm wondering if there's much research on this, particularly generational changes. I often see it bandied about, but in the subset of the population I interact with regularly (highly-educated thirty-somethings in Alberta), spouses - barring temporary interruptions like school or mat leave - are almost always in the same tax bracket, or at least close enough together that the total tax burden isn't going to be substantially different, even if one has jumped up to the next marginal rate. The idea of a primary and secondary earner seems old fashioned from this perspective. I'm curious whether this is true of a wider segment of the population, or just my insulated bubble.

[Frances Woolley]

Neil, in Alberta the provincial tax is pretty close to flat rate, so this is less of an issue there than in other provinces.

I don't know if the primary/secondary earner distinction is disappearing but I would agree that it's a less clear distinction than it was 30 or 40 years ago. Women's and men's earnings start diverging in their 30s and 40s - this is what studies that track, say, lawyers over time find (Philip Cohen at Family Inequality had a nice post on that the other day) but a chunk of the divergence in average earnings is caused by a few men earning much more than everyone else - the gap in median earnings is less pronounced.

[Frances Woolley]

BrenB "well la-dee-da"

No complaints about the free pizza.