Alternative Hypotheses About the Excess Return on Dollar Assets, 1980–84—eduardo r. borensztein (pages 29–59)

From late 1980 through 1984, assets denominated in U.S. dollars exhibited average annual returns 12–18 percent higher than those on similar assets denominated in other major currencies. To be consistent with the existence of efficient financial markets, this large and unprecedented margin requires some explanation.

This paper suggests that the large return differentials can be attributed to a systematic difference between expected and actual exchange rate changes. It empirically investigates two hypotheses, both of which are based on the idea that, for a prolonged period, there was a small but significant probability of a large depreciation of the dollar. Because this depreciation did not take place, the actual returns on dollar assets were persistently higher than the expected returns on these assets. The two models studied are a “peso-problem” model, in which the possibility of a change in the policy regime is the source of the discrepancy between actual and expected exchange rate changes; and a speculative bubble model, in which the exchange rate is temporarily driven away from its fundamental value by market expectations, but in which the possibility of a burst in the bubble and of a collapse to the fundamental value remains present.

Although not necessarily excluding other explanations, the empirical tests in this paper show the plausibility of both alternative hypotheses. Moreover, both models could contribute to the explanation. The sudden depreciation of the dollar in March 1985 had the appearance of a bursting bubble. Another abrupt fall of the dollar in September-October 1985, after the launching of a plan of coordinated intervention and other policy measures by the Group of Five countries, signaled an important change in policy regime—a situation of the type described by the peso-problem model.

Measurement and Alleviation of Poverty, with an Application to the Effects of Macroeconomic Adjustment—ss.m. ravi kanbur (pages 60–85)

This paper reviews recent developments in the measurement of poverty, analyzes alternative strategies for poverty alleviation, and suggests a methodology for evaluating the impact of macroeconomic adjustment on poverty.

The paper begins with the conceptual basis of alternative poverty lines and alternative indices. It then introduces the concept of “crossover time,” the time that it would take the average poor person to cross the poverty line if his income grew at the per capita growth rate experienced in the past. For reasonable estimates this crossover time can exceed twenty years, which perhaps explains the concern with redistribution rather than a complete reliance on “trickle-down” effects.

Regarding redistribution, the paper considers the question of targeting public expenditure on the poor. If large-scale means testing of individual households is administratively infeasible or too costly, policy must rely on instruments that measure the incomes of broadly defined groups. Although leakage is inevitable, how can it be minimized? The paper develops a methodology to answer this question and derives budgetary rules for poverty alleviation that rely only on information readily available in surveys of household income and expenditure.

The paper also analyzes macroeconomic adjustment and poverty. The bottom line for most policies is to switch expenditure toward the traded goods sector and to reduce overall expenditure. The paper derives rules for when such a switch will increase or decrease poverty as well as for deciding which broad categories of income should bear the burden of reduction.

Finally, the paper suggests areas for further research, especially through an application of the methodology developed in it to actual case studies.

Protectionism in a Framework with Intra-Industry Trade: Tariffs, Quotas, Retaliation, and Welfare Losses—daniel gros (pages 86–114)

This paper uses Krugman’s (1980) model of trade with product differentiation and monopolistic competition to examine the effects of various protectionist measures with and without retaliation. The main results are as follows.

Diversity (the number of products available to consumers) will not be affected by any ad valorem tariff.

An export tariff has the same effect on exports, imports, and the terms of trade as an import tariff with the same tariff rate.

A quota is equivalent to a (uniform ad valorem) tariff only if firms cannot price-discriminate between markets; that is, if the market for import licenses is competitive. In this case it is possible to calculate the tariff rates that are equivalent to certain quotas. These equivalent tariff rates seem substantial; under reasonable parameter assumptions, a quota that reduces the market share of imports by, say, 50 percent would be equivalent to a tariff of 47 percent.

A tariff war in which each country retaliates by imposing its own optimal tariff, taking the tariff of the other country as given, leads to a stable equilibrium in which each country imposes a tariff that is smaller than the tariff it would impose without retaliation.

The welfare losses from protectionism are likely to be substantial. Under reasonable parameter assumptions, a tariff war between two countries of equal size causes a welfare loss for consumers in both countries equivalent to a drop in national income of about 4 percent. “Voluntary” export restraint agreements are also likely to involve substantial welfare losses; under reasonable parameter assumptions, a voluntary export restraint agreement that limits imports into the home country by only 10 percent causes a welfare loss for consumers in the home country equivalent to a drop in national income of about 2 percent.

The Growth of Government: A Public Choice Perspective—dennis c. mueller (pages 115–49)

The size of the government sector has grown relative to the private sector throughout the developed democratic countries of the world. This growth has been particularly pronounced since World War II. Although government size has grown in every country, a wide range of relative sizes existed at the beginning of the 1980s. This paper examines both the causes of the growth in government size and the disparity of relative government size among countries.

The focus is on explanations of government size that stem from the literature on public choice. The various explanations of government size and growth are grouped into five main categories: (1) government as a provider of public goods and eliminator of externalities, (2) government as a redistributor of income and wealth, (3) interest groups as inducers of government growth, (4) the government bureaucracy as inducer of growth, and (5) fiscal illusion as enhancer of government growth.

These five explanations for the growth and relative size of government are in turn divided into two distinctive conceptualizations of the state. The first sees the state as implementer of the preferences of citizens. These might be the preferences of the median voter as revealed by the use of majority rule, or those of interest groups as revealed through a more complicated interaction of interest groups and electoral politics. The second sees the state, as personified by political leaders and the public bureaucracy, as able to dictate political outcomes, in part because of the impotence of citizens that is brought about by fiscal illusion.

Although none of the five explanations for the growth of government is found to have overwhelming logical and empirical support, some merit is found in each.

The Cost of Trade Constraints: The Case of Japanese Automobile Exports to the United States—Charles collyns and steven dunaway (pages 150–75)

This paper assesses the extent to which restraints on exports of Japanese automobiles to the United States affected car sales and prices in the period 1981-84. The estimates are obtained using a model of the U.S. automobile sector that explicitly allows for the effects of the quotas on the average quality of cars purchased. By distinguishing between pure price effects and quality effects, it is possible to assess the welfare costs and income transfers resulting from the imposition of the restraints.

The results presented in the paper suggest that the restraints had a substantial impact. The average transactions price for all new automobiles in the United States increased by nearly 50 percent over the period 1981-84, compared with an increase of 27 percent which would have been expected in the absence of the quotas. The higher prices resulted from a combination of “pure” price increases and improvements in quality. Sales were reduced by 4 million units during 1981-84 as a result of the quotas, while the value of expenditures was raised by $5¼ billion as increases in prices more than offset the effects of the restraints on the number of cars sold. At the same time, the quotas served to increase the market share of the U.S. industry, and expenditures on U.S. automobiles were $17½ billion higher than they otherwise would have been. The rise in new automobile prices (adjusted to exclude price increases owing to changes in quality) induced by the export restraints is estimated to have cost purchasers nearly $17 billion during the four-year period. Of this increase in purchasers’ costs, $6–12 billion represented a transfer to the U.S. automobile industry, and the remaining $5–11 billion is accounted for by a transfer to foreign producers and a deadweight loss to purchasers.