Guest Post by Arpit Gupta: Labor Supply and Taxes

by Arpit Gupta

Editor’s note: I’ve asked my Economics 21 colleague Arpit Gupta to summarize some of the latest research concerning labor supply and taxes.

While many recent discussions of tax policy are motivated by using taxation for the purposes of economic stimulus or for addressing income inequality, the really important long-term impacts of taxation center on labor supply. Higher taxes induce people to cut back on hours worked (what economists refer to as the intensive margin), as well as to drop out of the workforce entirely (the extensive margin). Figuring out how much taxes induce household labor behavior along these dimensions has been a major area of economic research, and the implications for government policy are sizable.

Edward Prescott for instance has argued that high labor supply elasticities — greater responsiveness of household labor supply to taxes — can explain the US-Western European income gap. Though productivity per hour figures between countries like Germany or France and America are comparable, American supply something like 50% more labor during prime working years (interestingly, and counter to stereotype, Europeans don’t seem to spend this additional time in leisure, but rather in non-market production such as in the household). The implication is that lower American marginal tax rates support higher rates of taxable work, and so appreciably higher living standards.

Other economists believe in lower estimates of labor supply elasticities. Peter Diamond and Emmanuel Saez, instance, implicitly relied on low labor supply estimates in a recent paper in which they called for top marginal tax rates as high as 76%.

Broadly, macroeconomists like Ed Prescott, studying the behavior of economies in the aggregate, tend to prefer large labor supply estimates; while microeconomists like Saez and Diamond, studying groups of individuals, tend to support smaller estimates. This empirical dispute has fairly far-reaching consequences on the structure of government policy. If Diamond and Saez are right, then even large hikes in taxation, though they may affect after-tax income, will not impact labor supply or GDP. If Prescott are company are right, then hikes to marginal tax rates will hit labor supply, and through that GDP, and so damage a key source of competitive advantage for the US economy.

One resolution to this puzzle has come from Richard Rogerson, who has emphasized the difference between measuring the intensive and extensive margin. His argument is that people finding smaller labor supply estimates have focused on the response of individuals already in the workforce. But there are many reasons to think that people who are already working have limited ways, at least over the course of a few years, to alter their hours. On the other hand, people tend to have far greater discretion over their lifetime to choose whether to work or not. For instance, mothers in particular tend to face a sizable tradeoff between working or not, as do the elderly. Failing to properly account for this extensive margin decision would lead to the impression that higher taxes had little impact on labor supply; when in fact taxes might have a large impact through the channel of entry or exit into the labor force.  

In a new paper, Raj Chetty and co-authors provide estimates of this extensive margin:

Macroeconomic calibrations imply much larger labor supply elasticities than microeconometric studies. The most well known explanation for this divergence is that indivisible labor generates extensive margin responses that are not captured in micro studies of hours choices… We find that micro estimates are consistent with macro evidence on the steady-state (Hicksian) elasticities relevant for cross-country comparisons. However, micro estimates of extensive-margin elasticities are an order of magnitude smaller than the values needed to explain business cycle fluctuations in aggregate hours. Hence, indivisible labor supply does not explain the large gap between micro and macro estimates of intertemporal substitution (Frisch) elasticities. Our synthesis of the micro evidence points to Hicksian elasticities of 0.3 on the intensive and 0.25 on the extensive margin and Frisch elasticities of 0.5 on the intensive and 0.25 on the extensive margin. (Emphasis added)

Chetty’s et al. suggests the the labor supply elasticity on the extensive may not be as large as Rogerson and some colleagues had hoped, but remains sizable. A 1% increase in taxes is associated with a .55% reduction in hours. The implication is that the entry or exit into the workforce does seem to be highly sensitive to taxes. The authors continue:

These findings indicate that labor supply responses to taxation could indeed explain much of the variation in hours of work across countries with different tax systems.

The authors are careful to note that this does not necessarily imply that cross-country differences in income are due only to differences in tax structure — only that this is a reasonable hypothesis given the data.

This work complements another recent paper by Michael Keane, referenced recently by Tyler Cowen. Keane finds: 

I survey the male and female labor supply literatures, focusing on implications for effects of wages and taxes… The literature is characterized by considerable controversy over the responsiveness of labor supply to changes in wages and taxes. At least for males, it is fair to say that most economists believe labor supply elasticities are small. But a sizeable minority of studies that I examine obtain large values. Hence, there is no clear consensus on this point. In fact, a simple average of Hicks elasticities across all the studies I examine is 0.30. Several simulation studies have shown that such a value is large enough to generate large welfare costs of income taxation.

For males, I conclude that two factors drive many of the differences in results across studies. One factor is use of direct vs. ratio wage measures, with studies that use the former tending to find larger elasticties. Another factor is the failure of most studies to account for human capital returns to work experience. I argue that this may lead to downward bias in elasticity estimates. In a model that includes human capital, I show how even modest elasticities – as conventionally measured – can be consistent with large welfare costs of taxation.

For women, in contrast, it is fair to say that most studies find large labor supply elasticities, especially on the participation margin. In particular, I find that estimates of “long run” labor supply elasticities – by which I mean estimates that allow for dynamic effects of wages on fertility, marriage, education and work experience – are generally quite large. (Emphasis added)

Keane discusses the mix of estimates other researchers have found in estimating labor supply elasticities. For women, the conclusion seems to be that the decision to supply labor is highly responsible to tax rates — an excellent reason to remove marriage penalties on labor supply at the very least. It might even be economically ideal for women to face a different marginal income schedule entirely, though this idea is of course much more controversial.

Keane notes that while many studies do have smaller estimates of the elasticity of labor supply, particularly on the intensive margin, studies on average find a sizable role for labor elasticities. This becomes particularly important when factoring in other various long-run decisions like education investment, family formation, and entrepreneurship. A number of studies have found that the decision to become an entrepreneur or seek further education is highly dependent on financial returns, which are affected by taxation. Factoring in these dynamic responses over time is difficult, but it suggests that the various long-run welfare costs of taxation might be high.

In the midst of continuing economic woes, there is little appetite to raise taxes on anything but a tiny fraction of Americans. Yet, eventually, politicians will face difficult fiscal choices. Raising taxes may avoid difficult cuts to social spending programs; yet there is a sizable body of research that suggests higher taxes may have long-run consequences on the long-run success of the economy — which will in turn make it more difficult to balance Debt/GDP. This may still be a choice worth making. Yet it is important to keep in mind the tradeoffs that higher taxes imply. At the very least, it’s a reason to pursue tax reform that lowered marginal tax rates by broadening the tax base, which would improve economic efficiency with few economic costs.