Public Economics

November 5-6, 2015
Raj Chetty of Harvard University and Mark Duggan of Stanford University, Organizers

Florian Scheuer, Stanford University and NBER, and Iván Werning, MIT and NBER

The Taxation of Superstars (NBER Working Paper No. 21323)

How are optimal taxes affected by the presence of superstar phenomena at the top of the earnings distribution? To answer this question, Scheuer and Werning extend the Mirrlees model to incorporate an assignment problem in the labor market that generates superstar effects. Perhaps surprisingly, rather than providing a rationale for higher taxes, they show that superstar effects provide a force for lower marginal taxes, conditional on the observed distribution of earnings. Superstar effects make the earnings schedule convex, which increases the responsiveness of individual earnings to tax changes. The researchers show that various common elasticity measures are not sufficient statistics and must be adjusted upwards in optimal tax formulas. Finally, they study a comparative static that does not keep the observed earnings distribution fixed: when superstar technologies are introduced, inequality increases but the authors obtain a neutrality result, finding tax rates at the top unaltered.


Benjamin B. Lockwood, Harvard University; Charles G. Nathanson, Northwestern University; and Glen Weyl, Microsoft Corporation

Taxation and the Allocation of Talent

In this paper, Lockwood, Nathanson, and Weyl study how taxation affects the allocation of talented individuals across occupations by blunting material incentives and thus relatively magnifying the non-pecuniary benefits of pursuing a "calling." Estimates from the literature suggest that high-paying professions pursued by these individuals have negative externalities while low-paying professions have positive externalities. As a result, a calibrated model indicates that total wealth is maximized by subsidies on middle class incomes and realistic tax rates on the rich. This result is robust to many uncertain features of the environment, though depends crucially on externality estimates and substitution patterns across professions, both of which deserve greater empirical study.


Daniel K. Fetter, Wellesley College and NBER, and Lee Lockwood, Northwestern University and NBER

Government Old-Age Support and Labor Supply: Evidence from the Old Age Assistance Program

A major source of the expansion of governments over the last several decades has been their role in operating social insurance programs. In this paper, Fetter and Lockwood investigate the Old Age Assistance Program (OAA), a means-tested and state-administered pension program created by the Social Security Act of 1935. Using newly available complete-count Census data from 1940, the researchers exploit the large differences in OAA programs across states and the rules that governed eligibility for OAA within states to estimate the labor supply effects of OAA. They find that OAA led to a 5.7 percentage point reduction in labor force participation among men aged 65-74, relative to a base of roughly 50 percent participation. Estimating a standard model of life cycle labor supply, the authors find that the majority of the OAA-induced reduction in late-life labor supply was due to income effects despite the high implicit tax rates on work imposed by OAA's earnings test. Simulations of the model suggest that Social Security had large effects on retirement once it started making payments, and that substitution effects from Social Security's earnings test were a larger share of the total than they were for OAA.


Alexander M. Gelber, University of California at Berkeley and NBER; Timothy J. Moore, George Washington University and NBER; and Alexander Strand, Social Security Administration

The Effect of Disability Insurance Payments on Beneficiaries' Earnings

A crucial issue in studying social insurance programs is whether they affect work decisions through income or substitution effects. The answer is largely unknown for U.S. Social Security Disability Insurance (DI), which is one of the largest social insurance programs in the U.S. The formula linking DI payments to past earnings has discontinuous changes in the marginal replacement rate that allow for a regression kink design to estimate the effect of payment size on earnings. Using Social Security Administration data on all new DI beneficiaries from 2001 to 2007, Gelber, Moore, and Strand document a robust income effect of DI payments on earnings. The researchers' preferred estimate is that an increase in DI payments of one dollar causes an average decrease in beneficiaries' earnings of twenty cents. This suggests that the income effect represents an important factor in driving DI-induced reductions in earnings.


Hugh Macartney, Duke University and NBER; Robert McMillan, University of Toronto and NBER; and Uros Petronijevic, University of Toronto

Education Production and Incentives

The substantial 'value-added' literature that seeks to measure the overall impact of teachers on student achievement does not distinguish between teacher effects that are invariant to prevailing incentives and those that are responsive to them. In contrast, Macartney, McMillan, and Petronijevic develop an empirical approach that, for the first time, allows them to separate out incentive-varying teacher effort from incentive-invariant teacher ability, and further, to explore whether the effects of effort and ability persist differentially. The researchers' strategy exploits exogenous variation in the incentive strength of a well-known federal accountability scheme, along with rich administrative data covering all public school students in North Carolina. These in hand, they first separately identify teacher effort and teacher ability to determine their relative magnitudes contemporaneously, finding that a one standard deviation increase in teacher ability is equivalent to 20 percent of a standard deviation increase in student test scores, while an analogous change in teacher effort accounts for between 3 and 5 percent of such an increase. The authors then use prior incentive strength to reject the hypothesis that the persistence of teacher ability and effort is similar. To interpret this reduced-form result, a complementary structural estimation procedure reveals that effort affects future scores less than ability. From a policy perspective, their results indicate that incentives matter when measuring teacher value-added. The analysis also has implications for the cost effectiveness of sharpening incentives relative to altering the distribution of teacher ability across classrooms and schools.


Joshua Rauh, Stanford University and NBER, and Xavier Giroud, MIT and NBER

State Taxation and the Reallocation of Business Activity: Evidence from Establishment Level Data (NBER Working Paper No. 21534)

In a sample of over 27 million establishments of U.S. firms with activities in more than one state, Giroud and Rauh estimate the impact of state business taxation on business activity. Only firms organized as subchapter C corporations are subject to the corporate tax code, whereas the income of partnerships, sole-proprietorships, and S corporations is passed through annually to the firm's owners and taxed at individual rates. For C corporations, both employment at existing establishments (intensive margin) and the number of establishments in the state (extensive margin) have corporate tax elasticities of -0.4. Pass-through entities, which serve as a control group for the corporate tax reforms, respond only to the personal tax code, with tax elasticities of -0.2 to -0.3. Around half of the effects are driven by reallocation of productive resources to other states where the treated firms have establishments. Capital shows similar patterns but is 36% less elastic than labor. A narrative approach confirms that the results are robust and strongest in the sample of tax changes that were implemented due to inherited budget deficits, long-run goals, or cross-state variation caused by Federal tax reforms.

Stefanie Stantcheva, Harvard University and NBER; Ufuk Akcigit, University of Chicago and NBER; and Salomé Baslandze, University of Pennsylvania

Taxation and the International Mobility of Inventors (NBER Working Paper No. 21024)

This paper studies the effect of top tax rates on inventors' international mobility since 1977. Akcigit, Baslandze, and Stantcheva put special emphasis on "superstar" inventors, those with the most abundant and most valuable patents. The researchers use panel data on inventors from the United States and European Patent Offices to track inventors' locations over time and combine it with international effective top tax rate data. They construct a detailed set of proxies for inventors' counterfactual incomes in each possible destination country including, among others, measures of patent quality and technological fit with each potential destination. Exploiting the differential impact of changes in the top tax rate on inventors of different qualities, the authors find that superstar top 1% inventors are significantly affected by top tax rates when deciding where to locate. The elasticity to the net-of-tax rate of the number of domestic superstar inventors is relatively small, around 0.03, while the elasticity of the number of foreign superstar inventors is around 1. Inventors who work in multinational companies are more likely to take advantage of tax differentials. On the other hand, if the company of an inventor has a higher share of its research activity in a given country, the inventor is less sensitive to the tax rate in that country.


Pablo Fajgelbaum, University of California, Los Angeles and NBER; Eduardo Morales, Princeton University and NBER; Juan Carlos Suarez Serrato, Duke University and NBER; and Owen M. Zidar, University of Chicago and NBER

State Taxes and Spatial Misallocation

Fajgelbaum, Morales, Suarez Serrato, and Zidar study the impact of the state tax distribution on aggregate real income, welfare, and the distribution of economic activity across U.S. states. The researchers build a spatial general-equilibrium model in which state governments use tax revenue to finance public services, and estimate firm and worker mobility elasticities and preferences for public services. They find that a revenue-neutral tax harmonization leads to a welfare gain of 0.7% and a similar increase in real GDP. Additionally, state tax cuts lower own-state economic activity and impact other states differently due to trade linkages. General-equilibrium effects on prices and public services largely drive these results.


Lorenz Kueng, Northwestern University and NBER

Explaining Consumption Excess Sensitivity with Near-Rationality: Evidence from Large Predetermined Payments

Using new transaction data Kueng shows that consumption is excessively sensitive to large, predetermined, regular, and salient payments from the Alaska Permanent Fund, with a large average marginal propensity to consume (MPC) of 25% for nondurables and services and 60% for total expenditures. However, this deviation from the standard inter-temporal consumption model is concentrated among households for whom the loss from failing to smooth consumption is small in terms of equivalent variation. In particular, the MPC is increasing in household income but decreasing in the size of the loss. As a result, statistically significant excess sensitivity in response to these large payments is consistent with households following near-rational alternative consumption plans. For macroeconomic policies, such as an economic stimulus program, these near-rational alternatives might be the more relevant behavior than the standard consumption model. Repeating the analysis with the Consumer Expenditure Survey (CEX) shows that credit constraints also predict larger MPCs for lower-income households, who are underrepresented in the transaction data. Finally, measurement error and composition effects fully explain the attenuated average MPC in the CEX.


Marco Di Maggio, Columbia University, and Amir Kermani, University of California, Berkeley

The Importance of Unemployment Insurance as an Automatic Stabilizer

Di Maggio and Kermani assess the extent to which unemployment insurance (UI) serves as an automatic stabilizer to mitigate the economy's sensitivity to shocks. Using a local labor market design based on heterogeneity in local benefit "generosity" (defined as the percentage of household income recovered by the unemployment benefit), the researchers estimate that a one standard deviation increase in generosity attenuates the effect of adverse shocks on employment growth by 12% and on earnings growth by 18%. Consistent with the hypothesis that this effect derives from the local demand channel, the authors find that consumption is less responsive to local labor demand shocks in counties with more generous benefits. Moreover, the average wage growth of employed workers is less elastic to local labor shocks when benefits are more generous. The authors' analysis finds that the local fiscal multiplier of unemployment insurance expenditure is approximately 1.2-1.8. Overall, their results suggest that UI has a beneficial effect on the economy by decreasing its sensitivity to shocks.


Pascal Michaillat, London School of Economics, and Emmanuel Saez, University of California at Berkeley and NBER

The Optimal Use of Government Purchases for Macroeconomic Stabilization (NBER Working Paper No. 21322)

This paper extends Samuelson's theory of optimal government purchases by accounting for the contribution of government purchases to macroeconomic stabilization. Using a matching model of the macroeconomy, Michaillat and Saez derive a sufficient-statistics formula for optimal government purchases. The formula implies that the deviation of optimal government purchases from the Samuelson level is proportional to the elasticity of substitution between government and personal consumption times the government-purchases multiplier times the deviation of the unemployment rate from its efficient level. Hence, with a positive multiplier, optimal government purchases are above the Samuelson level when unemployment is inefficiently high and below it when unemployment is inefficiently low. The researchers calibrate the formula to U.S. data. A first implication is that U.S. government purchases are optimal with a small multiplier of 0.04; if the multiplier is larger, U.S. government purchases are not countercyclical enough. Another implication is that optimal government purchases should increase during recessions. With a multiplier of 0.5 the optimal government purchases-output ratio increases from 16.6% to 20.0% when the unemployment rate rises from the U.S. average of 5.9% to 9%. With multipliers higher than 0.5 the optimal ratio increases less because fewer government purchases are required to fill the unemployment gap: with a multiplier of 2 the optimal ratio only increases from 16.6% to 17.6%; this is the same increase as with a multiplier of 0.07.