NBER Publications by Harris Dellas

We show that international trade in goods is the main determinant of international equity portfolios and offers a compelling -- theoretically and empirically -- resolution of the portfolio home bias puzzle. The model implies that investors can achieve full international risk diversification if the share of wealth invested in foreign equity matches their country's degree of openness (the imports to GDP share). The empirical evidence strongly supports this implication.

This paper demonstrates that disturbances to supplies or demands for internationally traded goods affect exchange-rates differently than do disturbances in markets for nontraded goods. The paper develops a stochastic two-country equilibrium model of exchange rates, asset prices, and goods prices, with two internationally traded goods and a nontraded good in each country. Optimal portfolios differ across countries because of differences in consumption bundles. Changes in exchange-rates, asset prices, and goods prices occur in response to underlying disturbances to supplies and demands for goods. We examine the ways in which responses of the exchange-rate are related to parameters of tastes and production shares, and we discuss conditions under which these exchange-rate responses are "large"...

This paper examines the endogenous implementation of capital controls in the context of a fixed exchange rate regime. It is shown that if there exists a non-zero probability that the policymaker's response to a speculative attack on official foreign reserves will be the introduction of controls, such an attack may occur even when current and expected monetary policy is consistent with a permanently viable, control-free fixed exchange rate regime. Consequently, capital controls may be the outcome of self- fulfilling expectations rather than the result of imprudent economic policies.

The paper examines a simple model in which exogenous political risk creates uncertainty about tariffs. The model predicts a relation between consumption and tariffs that differs radically from that implied by models without asset markets or political risk. Given the probability distribution of tariffs, domestic consumption and utility (ex post) are lower in states of the world with a domestic tariff and no foreign tariff than with a foreign tariff and no domestic tariff.This conclusion emerges despite the fact that the opposite would be obtained in the absence of asset markets. So economists should not be surprised if observed relations between consumption and tariffs differ from the predictions of static theory in either time-series or cross-sections.