Tino Sanandaji, a scholar at Sweden’s Research Institute of Industrial Economics, observes that Sweden has now surpassed the United States in hours worked per working age adult. In recent decades, we’ve tended to think of the U.S. as a society in which labor market regulations are relatively lax and work incentives are relatively strong. Now, however, it seems that Sweden is a country that is at least somewhat friendlier to work, despite the fact that it continues to have a much larger public sector. Tino raises a few important points:
In an influential study, Edward Prescott pointed out that until around 1970 Europeans worked as much as or more than Americans. Hours worked in the United States remains roughly the same while declining in most European countries. In Sweden most of this decline came following the recession starting in 1991, with hours worked dropping and not returning to prior levels even after the recession ended.
In recent years, Sweden has cut taxes and dramatically reduced the generosity of unemployment insurance and other programs.
Hours worked per working age adult is estimated by dividing total hours worked in the economy with the 15-64 population, the potential labor force. It is a better measure than hours worked per worker, since it also takes into account the employment rate. …
Much of the increase in hours worked in Sweden is not higher employment rates, but more hours worked per employee due to lower absenteeism in sick-leave following reform of sick-leave benefits. …
Among the 21 OECD countries with available data regarding hours worked per working age adult, the top five countries are Switzerland, Iceland, New Zealand, Australia and Canada.
Earlier this summer, Tino discussed Sweden’s high level of labor force participation in the context of its ongoing difficulties with integrating non-European-origin migrants in an article for NRO:
[Sweden] takes in more immigrants relative to its population than the U.S. did at the peak of the transatlantic migration. Sweden has about 9 million inhabitants and last year took in almost 90,000 immigrants, excluding Swedes returning from abroad. Non-Western immigrants were 1 percent of Sweden’s population in 1980 and have since increased to 10 percent of the population.
Today, 60 percent of total welfare payouts in Sweden go to immigrants. Problems such as child poverty, which the welfare state was supposed to have solved, are reemerging as a consequence of immigration. Second-generation immigrants born in Sweden remain less likely to work or graduate from college than the children of natives are.
The only reason the welfare state remains solvent is that an astonishing 85 percent of working-age native Swedes work and pay taxes, far above the European average of 70 percent. By contrast, only half of non-Western immigrants work. [Emphasis added]
Though support for increasing labor force participation isn’t universal in American public life — at least some social conservatives are wary of increasing labor force participation of mothers with children younger than six, which currently stands at 65 percent, and there is a current on the Marxist and post-Marxist left which sees declining labor force participation as the labor market position of less-skilled workers deteriorates to be a good thing, provided it is accompanied by an increase in unconditional transfers — it is very widespread. Back in 2010, Lane Kenworthy wrote on the surprisingly limited role of wages in raising households incomes in the bottom tenth of the income distribution:
The principal way economic growth is expected to trickle down to the poor is via rising earnings. But in most of the countries in which bottom-decile incomes increased, the source of that rise was govern- ment transfers.
This should not be too surprising. In most rich countries, 20-35 percent of all households have no earnings, and some of these are in the bottom decile of the posttransfer-posttax income distribution. As of the early to mid-2000s, the share of bottom-income-decile house- holds with zero earnings was 40 percent in Finland and Norway, 55 percent in France and the Netherlands, 60 percent in Sweden, and 75 percent in Ireland (my calculations from LIS data). Some of these are elderly households with savings and pensions as the main source of income. Others are households with working-age adults whose chief income source is government transfers such as social assistance or unemployment, sickness, or disability compensation.
Over the past few decades, rising incomes at the low end of the distribution have been a product mainly of policy arrangements and choices that determine net government transfers, rather than of the rate of economic growth (Kenworthy 2011). [Emphasis added]
One instrument for raising household incomes at the low end of the distribution is to increase labor force participation. The problem, however, is that this is easier said than done, particularly with individuals who have been disconnected from the labor force for long periods of time and have thus seen their skills and social networks atrophy. The “low-hanging fruit” might be to increase work hours among the employed, a strategy that would presumably involve improving work incentives, e.g., lowering the implicit marginal tax rates faced by low-income households.
In recent years, Sweden has adopted an Earned Income Tax Credit (EITC), which is, like its U.S. counterpart, designed to encourage labor force participation. Yet the Swedish EITC works very differently from the U.S. EITC. Tino kindly explained some of the differences to me, drawing on published sources. For example, the Swedish EITC is tied to individuals rather than households and the size of the wage subsidy is unrelated to the number of dependents. It is also far more expansive than the U.S. EITC, as it reaches its maximum level at 1.4 times the median income and it never phases out. The key difference between the Swedish and the U.S. EITC is that pensions and most transfers, including unemployment insurance, are taxed in Sweden, and so the goal of the Swedish EITC is in effect to cut taxes on wage income while leaving taxes on pensions and most transfers unchanged, thus making work a more attractive option. As such, it is better understood as a tax cut for middle-income households rather than a measure aimed primarily at the poor. The evidence regarding the impact of the Swedish EITC on labor supply is inconclusive.
I’d be very interested to see if Robert Stein’s proposal for a greatly expanded child credit would increase earnings among middle-income parents of minor children. (It’s not entirely clear that it would. Stein’s expanded child credit might instead encourage some parents to become full-time caregivers, but it’s not obvious to me that this would outweigh its other virtues.)