Kenneth A. Lewis

Human rights and the distribution of U.S. foreign aid

In contrast to the findings of other studies, we conclude that human rights play a significant and substantive role in determining the distribution of U.S. foreign aid. We find that the foreign aid program relates aid to the need of recipient nations, rewards nations for furthering human rights, does not discriminate on the basis of race or religion, and responds to national security interests of the U.S. The finding that the program does what most people assert it should do provides a new explanation for the rigidity of distributions over time.

A New Design for Automatic Fiscal Policy

We offer a new design for automatic fiscal policy that can strengthen its role as a complement to counter-cyclical monetary policy, and analyse it in the Fair macroeconometric model that has been estimated using quarterly data for the US economy. Our automatic fiscal policy consists of the triggering of a transfer (or income tax rebate) whenever real GDP is at least X% below normal, the amount of the transfer varying with the size of the GDP gap. The size of the transfer is set with the sole purpose of effectively combating a recession. By contrast, the magnitudes of current automatic stabilizers are unintended by-products of setting the ratio of tax revenue to GDP, the degree of progressivity of the tax system and the level of unemployment benefits. We generate a severe recession using historical data and simulate the impact of our automatic fiscal policy. We assume that the Federal Reserve adheres to a counter-cyclical monetary policy governed by the interest rate (Taylor) rule estimated from historical data. We find that the interest rate rule alone mitigates the severe recession only modestly, whereas our automatic fiscal policy (together with the interest rate rule) substantially reduces the severity of the recession while generating only a relatively small rise in the government debt/GDP ratio. Copyright 2002 by Blackwell Publishers Ltd.

Cultural and Institutional Determinants of Economic Growth: A Cross-Section Analysis

Economic growth rates for the period 1968–1987 are analyzed for ninety countries. Culture, political and economic arrangements, and personal freedoms are statistically significant determinants of growth. Personal freedom is shown to be a normal good whose demand might be affected by cultural influences. Democracies raise personal freedoms,ceteris paribus, and, consequently, grow more quickly than non-democratic regimes. Evidence is found for the convergence hypothesis; other things equal, lower income countries grow more rapidly than higher income countries.

Math—time capital matters: A cross-country analysis

Abstract We compute and compare the rate of accumulation of one component of educational capital — “math-time capital” — through eighth grade in a sample of countries that includes Japan and the US, and measure the impact on math achievement of raising the rate of accumulation of US math-time capital. We examine the 1982 IEA math exam administered to 13-year olds on which the US mean score was substantially below Japan. We find that Japan is the highest math-time country — the Japanese 8th grader has accumulated 30% more math time than the American 8th grader — and that math-time capital matters for math achievement. We estimate that if the US lengthened its school year by three weeks and assigned required summer math homework, it would close 26% of the gap between the US and Japanese mean scores, and move the US from 13th to 8th in score rank.

A median-voter model of economic regulation

A basic median-voter model is developed and extended to analyze issues of economic regulation and public policy outcomes. The model is used to generate comparative static results relating changes in public-policy outcomes to changes in relative group sizes, total population, information costs, and population heterogeneity. The model is also used to explore the issue of optimal group size — the size of the special-interest group that maximizes the groups per capita public policy gains. Comparative static analysis reveals how optimal group size and gains per capita are affected by changes in population heterogeneity, the size of the total population, and relative knowledge levels.

Did the 2008 Rebate Fail? A Response to Taylor and Feldstein

Did the 2008 rebate fail to stimulate consumer spending? In their influential American Economic Review articles, John Taylor and Martin Feldstein each claim that Bureau of Economic Analysis (BEA) aggregate time series data show that the 2008 rebate failed. Reexamining the BEA data, we find that the data instead show there is a high probability that the rebate stimulated consumption. Moreover, the hypothesis that a rebate has half the impact of ordinary disposable income cannot be rejected. Thus, we find that analysis of the BEA aggregate time series data is consistent with the conclusion from the microdata studies that the 2008 rebate stimulated consumer spending.

The Impact of Converting to a Consumption Tax When Saving Propensities Vary: An Empirical Analysis

It has been recognized that conversion of an income tax to a consumption tax can increase aggregate saving even if each household maintains a constant propensity to save. The reason is heterogeneity: the variation in the propensity to save among households. How much of an increase in saving is an empirical question. Using the best available (but not wholly adequate) U.S. data, we estimate that the increase may be as much as 10% of saving. New data stratified by age would be necessary to obtain a more reliable estimate.

A Temporary Tax Rebate in a Recession: Is it Effective and Safe?

Is a temporary tax rebate effective and safe as an antirecession policy? Simulations with an empirically tested macroeconometric model are used to estimate the impact of the actual one-time 2001 tax rebate in the United States and of a hypothetical rebate twice as large, injected four times (quarterly). The results of the simulations are interpreted in light of two important recent empirical studies of the spending of the 2001 rebate by households. We estimate that a rebate equal to three percent of quarterly GDP (roughly the size of the 2001 rebate) repeated four times (quarterly) would reduce the unemployment rate at the end of a year by one percentage point (for example, from six percent to five percent). As along as a tax rebate is detriggered when the recession ends, its use during a recession does not pose a significant debt or inflation problem.

The Consumption Tax and Transitional Relief

Perhaps the most serious ethical and political obstacle to converting an income tax to a consumption tax is that older persons who have accumulated wealth after paying income tax will be taxed again under the new consumption tax if they spend down their wealth. A crucial question that arises is whether the provision of transitional relief would temporarily reduce the impact of tax conversion on capital accumulation. Assuming perfect foresight along the transition path and an empirically plausible range of parameter values, we find that the magnitude of the temporary effect depends crucially on the value of the intertemporal elasticity of substitution in the utility function.

A phased increase in the U.S. investment rate: Sacrifice times, T-year gains, and investment rate returns

Abstract We simulate a phased increase in the U.S. investment rate using a translog production function with technical progress (disembodied and/or embodied). We assume there will be an absorption lag implying that factors are underutilized during the transition to a higher investment rate. We find that the “sacrifice time” (the time that elapses until consumption surpasses the value it would have had under the initial investment rate) is roughly nine years. Across alternative specifications, phase-in periods, and absorption lags, the sacrifice time varies from seven to 13 years, and is insensitive to the percentage increase in the investment rate. With a three-year phase-in of a 20 percent increase in the investment rate with a one-year absorption lag, the average “ecade gain” in output (the percentage gain at the end of a decade) is roughly 4 percent; the decade gain in consumption, 0 percent; the five-decade gain, 10 percent in output and 6 percent in consumption; and the “investment rate return” (the internal rate of return on a permanent increase in the investment rate), 13 percent.