Chaotic tatonnement

Summary Debreu's theorem on excess demand functions is used to demonstrate the possibilities of ergodic and topological chaos in a discrete-time tatonnement process with only two goods. The result is in sharp contrast with the well-known result of Arrow and Hurwicz on system stability in a continuous time model of price adjustment with two commodities.We would like to thank Professors Jess Benhabib, Richard Day, John Guckenheimer, Philip Holmes, Nicholas Kiefer and Tapan Mitra.     

Accounting variables as determinants of systematic risk in Malaysian common stocks

The objective of this paper is to examine the relationship between financial accounting variables and systematic securities risk in a small and developing capital market, namely the Kuala Lumpur Stock Exchange. Factor analysis was used to group and identify the financial variables into independent dimensions. The possible bias due to multicollinearity between accounting data was ameliorated by selecting one representative variable from each factor profile of the firm. Evidence has shown that financial ratios/profitability ratio and, to a certain extent, activity ratio are important determinants of the systematic risk of a common stock. Contrary to most of the reported findings in other markets, the results show a negative relationship between leverage ratio and systematic risk, for which we have no explanation.     

The intertemporal volatility structure of Euro CD rates

In this paper we examine the intertemporal volatility structure of Eurocurrency rates of five different maturities ranging from seven days to twelve months for six Euro CD currency denominations spanning the 1986–1992 period. the analysis used the common ARCH-feature testing methodology recently developed by Engle and Kozicki (1993). First, the results indicate presence of ARCH effects in the Eurocurrency rate series. This result suggests that modelling of Eurocurrency rates requires the inclusion of time-varying risk premia. Second, our evidence reveals that short- and long-term Eurocurrency rate series have the same volatility process. the results point out that a common time-varying volatility process characterises most Eurocurrency rate series across maturities and currency denominations. Hence, the common ARCH results imply that a common time-varying variance model would be the appropriate specification of the conditional heterscedasticity for most Eurocurrency rates.     

A pitchfork bifurcation in the tatonnement process★

Summary . This note extends the example of Gale (1963) by considering the continuous time tatonnement process for a class of two agent, two commodity exchange economies, parametrized by a number μ∈(0,1). We demonstrate that as the parameter passes a threshold value μ^* the unique, globally stable competitive equilibrium loses local stability while two new locally stable equilibria appear. Intuitively, as μ increases the income effect become increasingly more important relative to substitution effect, and eventually overwhelms the latter. As the parameter μ approaches 1, the economy tends to the example considered by Gale, as does the limiting behavior of the tatonnement. Received: February 28, 1996; revised version August 5, 1996     

EXPLAINING PREMIUMS AND DISCOUNTS ON CLOSED-END EQUALITY COUNTRY FUNDS

Over the past decade or so, the surge of interest among U.S. investors in international investing has led to the creation of numerous foreign equity country funds. Like U.S. closed-end mutual funds, the prices of such closed-end country funds fluctuate widely in relation to their underlying net asset values (NAVs).
In this paper, the authors summarize the major findings of their recent study of the performance of 28 country funds relative to their NAVs over the period 1978–1995. While 20 of the 28 funds traded at average discounts to their net asset values, the discounts for the country funds were smaller than those of the average U.S. fund, and over a quarter of the funds sold at premiums.
In an attempt to explain such premiums or discounts, the authors examined primarily three factors: (1) the sensitivity of country-fund returns (relative to that of local market indices) to U.S. returns; (2) the possible effects of local government investment restrictions; and (3) the impact of exchange rate changes. Although most of the eight funds that traded at average premiums represented countries with significant restrictions on capital flows and foreign ownership, there were also a number of funds with similar restrictions trading at significant discounts. In exploring the reasons for such discounts, the authors noted that the returns to the country funds were "surprisingly sensitive" to U.S. market conditions, thus reducing the extent of their diversification benefits for U.S. investors. The article also raises the possibility that if such country funds are not "priced at the margin" by globally diversified investors, U.S investors' "country-risk sentiments" could cause such funds to trade at discounts.