October 30, 2014
Nathaniel Hendren, Harvard University and NBER
This paper develops a tractable method for resolving the equity-efficiency tradeoff that modifies the Kaldor-Hicks compensation principle to account for the distortionary cost of redistribution. Weighting measures of individual surplus by the inequality deflator corresponds to searching for local Pareto improvements by making transfers through the income tax schedule. Empirical evidence consistently suggests redistribution from rich to poor is more costly than redistribution from poor to rich. As a result, the inequality deflator weights surplus accruing to the poor more so than to the rich. Regardless of one's own social preferences, surplus to the poor can hypothetically be turned into more surplus to everyone through reductions in distortionary taxation. Hendren estimates the deflator using existing estimates of the response to taxation, combined with a new estimation of the joint distribution of taxable income and marginal tax rates. The researcher shows adjusting for increased income inequality lowers the rate of U.S. economic growth since 1980 by roughly 15-20%, implying a social cost of increased income inequality in the U.S. of roughly $400 billion. Adjusting for differences in income inequality across countries, the U.S. is poorer than countries like Austria and the Netherlands, despite having higher national income per capita. Hendren concludes by providing an empirical framework for characterizing the existence of local Pareto improvements from government policy changes.
Benjamin Lockwood, Harvard University, and
Matthew Weinzierl, Harvard University and NBER
Calculating the welfare implications of changes to economic policy or shocks to the economy requires economists to decide on a normative criterion. One way to make that decision is to elicit the relevant moral criteria from real-world policy choices, converting a normative decision into a positive inference exercise as in, for example, the recent surge of so-called "inverse-optimum" research. Lockwood and Weinzierl find that capitalizing on the potential of this approach is not as straightforward as we might hope. The researchers perform the inverse-optimum inference on U.S. tax policy from 1979 through 2010 and identify two broad explanations for its evolution. These explanations, however, either undermine the reliability of the inference exercise's conclusions or challenge conventional assumptions upon which economists routinely rely when performing welfare evaluations. The authors emphasize the need for better evidence on society's positive and normative judgments in order to resolve the questions these findings raise.
Adnan Khan, London School of Economics;
Asim Ijaz Khwaja, Harvard University and NBER; and
Ben Olken, MIT and NBER
Despite the importance of performance-based payment systems, there is little rigorous evidence on their impact in a civil service. In tax, high-powered incentives for tax collectors could increase revenues, but come at a high political cost if collectors exert excessive pressure on taxpayers. Khan, Khwaja, and Olken report the results of the first large-scale, randomized field experiment designed to investigate these issues. Working with the entire property tax department in Punjab, Pakistan, the researchers experimentally allocated collectors in the entire 482 property tax units into one of three performance-pay schemes or a control group. They find that incentivized units experience an average of 9 percent higher revenue growth than controls, and the effects persist over the two years of the experiment. Yet these units do not report greater taxpayer discontent. The performance pay scheme that rewarded purely on revenue collection does better, increasing revenue growth by 13 percent, with relatively little penalty for customer satisfaction and assessment accuracy compared to the two other schemes that explicitly rewarded these dimensions. Digging deeper into the findings reveals an interesting heterogeneity: most taxpayers in incentivized areas do not get reassessed or pay higher taxes, but do report higher level of bribes and instances of corruption. In contrast, conditional on their property being reassessed, taxpayers in incentivized areas pay higher tax and report less corruption. These results are consistent with a simple model of collusion between collector and taxpayer, in which taxpayers in incentivized areas either have to pay higher bribes to avoid being reassessed, or pay substantially higher taxes if the bargaining process breaks down. The authors' results suggest that while incentives for tax collectors can substantially boost revenue collected and lead to little overall discontent, they do empower the tax collector in a manner that can both lead to increased revenue and rents at the expense of the taxpayer.
Mikhail Golosov, Princeton University and NBER;
Aleh Tsyvinski, Yale University and NBER; and
Nicolas Werquin, Yale University
Golosov, Tsyvinski, and Werquin develop a general method to study the effects of non-linear taxation in dynamic settings using variational arguments. The researchers propose a sufficient condition on individual demand that allows them to derive the effects of perturbations of the tax system in terms of intuitive parameters, such as the labor and capital income elasticities and the hazard rates of the income distributions. The authors first derive general theoretical formulas that characterize the welfare effects of local tax reforms and, in particular, the optimal tax system, potentially restricted within certain classes (e.g., age-independent, linear, separable). Second, they apply these formulas to various specific settings. In particular, they decompose the gains arising from each element of tax reform, starting from a simple baseline system, as the available tax instruments become more sophisticated. They show that the design of tax systems obeys a common general principle, namely that more sophisticated tax instruments (e.g., age-dependent, non-linear, non-separable) allow the government to better fine-tune the tax rates by targeting higher distortions to the segments of the population whose behavior responds relatively little to those taxes.
Marika Cabral and Michael Geruso, University of Texas, Austin and NBER, and
Neale Mahoney, University of Chicago and NBER
The debate over privatizing Medicare stems from a fundamental disagreement about whether privatization would primarily generate consumer surplus for individuals or producer surplus for insurance companies and health care providers. This paper investigates this question by studying an existing form of privatized Medicare called Medicare Advantage (MA). Using difference-in-differences variation brought about by payment floors established by the 2000 Benefits Improvement and Protection Act, Cabral, Geruso, and Mahoney find that for each dollar in increased capitation payments, MA insurers reduced premiums to individuals by 45 cents and increased the actuarial value of benefits by 8 cents. Using administrative data on the near-universe of Medicare beneficiaries, the researchers show that advantageous selection into MA cannot explain this incomplete pass-through. Instead, their evidence suggests that insurer market power is an important determinant of the division of surplus, with premium pass-through rates of 13% in the least competitive markets and 74% in the markets with the most competition.
Bhaven Sampat, Columbia University and NBER, and
Heidi Williams, MIT and NBER
In markets where innovation is cumulative -- in the sense that discoveries are inputs into follow-on discoveries -- optimal patent policy in part depends on how patents on existing technologies affect follow-on innovation. In this paper, Sampat and Williams investigate whether patents on one specific technology -- human genes -- have affected follow-on scientific research and product development. Using administrative data on successful and unsuccessful patent applications submitted to the US Patent and Trademark Office, the researchers link the exact gene sequences claimed in each patent application with data measuring gene-related scientific research (publications) and commercial investments (clinical development). They document that patented genes are positively selected relative to non-patented genes based on pre-application measures of scientific and commercial value, thus motivating two quasi-experimental approaches. First, the authors present a simple comparison of follow-on innovation across genes claimed in successful versus unsuccessful patent applications. Second, they use the "leniency" of the assigned patent examiner as an instrumental variable for whether the patent application was granted a patent. Both approaches suggest that -- on average -- gene patents have had no effect on follow-on innovation. The researchers' confidence intervals on these estimates are meaningfully precise: they are able to reject the effect sizes estimated in past papers investigating the effect of non-patent forms of intellectual property on follow-on innovation.