Mexico’s tequila crisis–hangover or hair of the dog? Country risk and exchange rate regimes

Mexico’s tequila crisis–hangover or hair of the dog? Country risk and exchange rate regimes

Goldberg, Cathy S

This paper examines the importance of economic factors in a time-varying beta model of country risk. We examine measures of Mexico’s inflation, exchange rate, trade and monetary policy in the model but find that exchange rates are the only economic variable that matters for variations in beta. We investigate the role that changes in exchange rate regimes, fixed to floating, play in affecting the time variation of beta for Mexico. Our results show that a switch from a managed to a flexible exchange rate regime leads to a rise in the level and volatility of Mexico’s country beta over time.

INTRODUCTION

The World Capital Asset Pricing Model (CAPM) is a standard approach for estimating the risk of a country’s financial markets with respect to a world market index. A large body of literature has established the importance of a country’s economic and political variables in explaining cross-sectional variations in beta measures across countries. This literature begins with Harvey (1991) who investigates global CAPM regressions and continues through Ferson and Harvey (1999) who investigate multi-factor model regressions across 18 countries.

Another strand of this literature recognizes that a country’s beta with respect to the world index is likely to change for countries that liberalize their previously closed financial markets to foreign investors. It is likely that the beta exhibits a structural break for a liberalizing country, with beta changing to reflect the higher correlations of local financial markets returns to movements in the world index after liberalization. Much empirical work has been done on exploring the effects of liberalization on country risk as measured by its beta, and dating liberalization incidents based on shifts in estimated betas. Bekaert and Campbell (1995) estimate timevarying degrees of integration across a variety of liberalizing countries. More recently, Goldberg and Delgado (2001) date liberalization episodes using panels of stock returns for each country. If country betas change systematically across liberalization episodes, as all the literature suggests, it seems reasonable to believe that a single country’s beta may fluctuate with changes in that country’s other important economic variables. Gangemi, Brooks, and Faff (2000) explore the notion that economic risk factors may change a country’s beta over time, even if the country’s degree of integration into world markets remains essentially unchanged. They focus on whether increasing government debt burdens are responsible for shifts in country risk for Australia. Their results are somewhat surprising; the only risk factor that seems important for changing Australia’s beta with respect to the world index are surprises associated with changes in Australia’s exchange rate.

Our paper investigates the importance of economic factors in a time-varying beta model of country risk for Mexico, a developing economy with financial markets that have been integrated with world markets for at least the past decade. We find that exchange rate surprises are the main determinants of how Mexico’s beta varies over the period 1990 2000. We also find that the type of exchange rate regime, managed rates versus floating rates, affects the influence of exchange rate surprises on Mexico’s country risk. Mexico’s beta trends lower over the managed rate period, 1990 1994, but shifts up and becomes more volatile in the floating rate period, 1995 – 1999.

We provide an overview of Mexico experiences with financial market liberalization, exchange rate regimes, and correlation structure with world and US capital markets. We also provide a discussion of the expected effects of changes in exchange rate regimes on country risk. The next section introduces the data to be used and the paper’s estimation strategy for time-varying country risk. We then present our results on time-varying beta estimates for Mexico and summarize the consequences for exchange rate regimes and country risk. Finally, we conclude the paper and make suggestions for future research.

EXCHANGE RATE REGIMES AND COUNTRY RISK: MEXICO’S EXCHANGE RATE EXPERIENCE

In 1981 the Mexican Country Fund was issued on the NYSE. Prior to this time the market was effectively closed to foreign investors. In May 1989, the Mexican stock market was made fully investable with the exception of a few key areas. The official IFC liberalization date for Mexico occurs during 1989. Bekaert and Harvey (1995) find their measure of the degree of integration for Mexico’s financial markets peaks between mid-1988 and the end of 1989. In contrast, Goldberg and Delgado(2000), using return series on a panel of individual Mexican stocks, identify June 1987 as the breakpoint for liberalization of the Mexican stock market.

Our interest is not in when the liberalization of Mexico’s financial markets occurred per se, but rather on how a newly integrated country’s covariance risk with the world market index is affected by a change in its exchange rate regime. We chose January 1990 through September 1999 as our sample period – a period in which Mexico had moved towards integrated financial markets but which also includes a major change in Mexico’s exchange rate regime.

The early part of our sample is Mexico’s managed exchange rate period; a period in which Mexico begins with a dual exchange rate system, abandoned in 1991, through a pegged rate, to a crawling peg, and finally to a widening band regime which collapsed in the “Tequila Crisis” of December 1994. The Tequila crisis marked a decisive defeat for Mexico’s attempts at managed exchange rate regimes. Post crisis, Mexico moves to a floating rate regime for the peso with limited interventions by the central bank into the foreign exchange markets.

Exchange Rate Regimes and Country Risk

When financial markets are completely integrated, assets with identical risk should earn identical returns across, as well as within, markets. In our model, risk refers to the country’s exposure to a common world factor, the Morgan Stanley World Index (MSWI). Changes in the degree of capital market integration for a country should cause shifts in the betas of its market index or its individual stocks with respect to the world index. Identifying such shifts has formed the basis for dating liberalization episodes in a variety of ways.

Changes in a country’s exchange rate regime should have similar effects on a country’s beta measures. Managed exchange rate regimes seek to limit the volatility of a country’s exchange rate using international reserves and/or capital controls. As a result the US dollar denominated returns for a country will reflect mostly movements in local currency denominated returns. These returns are likely to be heavily dependent on local economic conditions. As a result, the country’s US dollar returns are likely to move differently from those of the world index, making the country a diversification play so long as the managed rate regime survives. We expect that the country’s beta in a global CAPM regression will be low but highly sensitive to changes in the exchange rate, particularly large changes that may indicate fragility in the managed regime.

A collapse in a managed exchange rate regime to floating regime means US dollar returns will reflect both local currency returns, and economic conditions, and unanticipated exchange rate changes. As a result, we expect a country’s US dollar return to move more closely with returns on the world index. Flexible exchange rates represent another stage in the integration of a country’s capital markets into world markets. Under a flexible regime, the country’s exchange rate should respond to capital flows, resulting from return differentials, with more rapid and sizable adjustments. As a result we expect changes in the correlation structure between the country’s returns in local currency and US dollars. It is expected that a country’s US dollar returns will be more volatile and more highly correlated with returns on the world index than under a managed regime. We expect the result of a switch from managed to flexible exchange rates to be a rise in the country’s beta and an increase in the variability of its beta over time. Table 1 demonstrates that during our managed exchange rate period, Mexico’s local currency returns exhibit essentially the same variability as its US dollar return. In contrast, during the later floating exchange rate period, Mexico’s US dollar returns are almost twice as variable as its local currency returns. As expected, the correlation between the US dollar and local currency returns is also lower in the floating exchange rate period at 0.9610 versus 0.9924 in the earlier fixed rate period. Finally, the correlations of Mexico’s returns, both US dollar and in pesos, to the returns on the Morgan Stanley World Index (MSWI) and the S&P 500 increase in the floating rate period, suggesting that Mexico’s beta changed systematically across exchange rate regimes.

MODELING TIME-VARYING COUNTRY RISK

Selection of Economic Variables In TimeVarying Country Beta Models

Macroeconomic influences should play a major role in altering country risk, particularly in emerging economies. Thus, we argue that timevarying country betas may be explained by constructing a model that incorporates macroeconomic variables that impact country risk for Mexico. Our choice of relevant variables is influenced by past research in this area. Most relevant to our research is a paper by Gangemi et. al. (2000) who model country risk for Australia. They include measures for the exchange rate, inflation, economic activity, national debt, monetary policy, government borrowing, interest rates and commodity prices. Other work by Erb et. al. (1996) and Bekaert et. al. (1996) also rely on macroeconomic variables to explain country risk. These papers utilize macroeconomic variables which capture political risk, inflation, exchange rate volatility, GDP, size of the trade sector, indebtedness of the country, current account balance and the trade balance.

We chose the following set of macroeconomic variable for our analysis presented in table 2.

We obtained monthly macroeconomic data from the IFS database and use the monthly percentage changes in our variables computed as the log difference in our data between the current and previous month data. CPI or the index of consumer prices is our inflation measure, EXR is the Mexican peso to US dollar exchange rate and TRADE is defined as net exports. Monetary policy is captured by the remaining two variables; MBASE, reserve money, and IRES, international reserves for Mexico.

ARIMA Analysis

We use the unanticipated components or residuals from an ARIMA analysis as the relevant variables in analyzing Mexico’s monthly stock returns. The reason for this is twofold. First, if markets are efficient, then changes in stock market returns should only be explained by the unanticipated components of the macroeconomic variables used. Singh (1993,1995) uses this approach when examining announcements pertaining to macroeconomic variables and their impact on the stock market, exchange rates and interest rates. Second, using the raw data for our analysis would most likely result in a high degree of multicollinearity as macroeconomic variables tend to exhibit strong correlations with each other as the result of common trends.

First, it is interesting to note that the level of systematic risk in Mexico increased substantially after the tequila crisis, due in large part to the switch from a managed exchange rate regime to a floating one.

There are 2 outliers that occur in our post sample analysis. In October 1997, the country beta for Mexico increased to 2.31. We link this outlier to the Asian crisis which directly impacted Mexico’s exchange rate and hence the beta value. In this month, the exchange rate of the peso to the US dollar increased by 8%. A second outlier was identified in August 1998 where beta was 2.69. This was most likely caused by the crisis in Russia. In this month, Mexico’s peso/US dollar exchange rate increased by 11%. It is apparent that significant economic events that impacted Mexico’s exchange rate were a major cause for the shift in Mexico’s country risk, particularly in the second half of the 1990’s.

CONCLUSION The World Capital Asset Pricing Model (CAPM) is a standard approach for estimating the risk of a country’s financial markets with respect to a world market index. A large body of literature has established the importance of a country’s economic and political variables in explaining cross-sectional variations in beta measures across countries.

Our paper has investigated the importance of economic factors in a time-varying beta model of country risk for Mexico, a developing economy with financial markets that have been integrated with world markets for at least the past decade. We find that exchange rate surprises are the main determinants of how Mexico’s beta varies over the period 1990 2000. We also find that the type of exchange rate regime, managed rates versus floating rates, affects the influence of exchange rate surprises on Mexico’s country risk. Mexico’s beta trends lower over the managed rate period, 1990 1994, but shifts up and becomes more volatile in the floating rate period, 1995 – 1999.

REFERENCES

Abell, J., And Krueger, T. 1989. Macroeconomic Influences On Beta. Journal of Economics And Business 41: 185-193.

Bekaert, Geert, and Harvey, Campbell R., 1995, Time-Varying World Market Integration, Journal ofFinance 50, no. 2: 403 -441.

Braun, Phillip, Daniel Nelson, And Alain Sunier, 1995, Good News, Bad News, Volatility And Betas, Journal ofFinance 50, 1575-1604. Campbell, John Y., 1987, Stock Returns And The Tenn Structure, Journal of Financial Economics 18,373-400.

Chen, Nai-Fu, Richard Roll, And Stephen A. Ross, 1986, Economic Forces And The Stock Market, Journal Of Business 59, 383-403.

Connor, Gregory, And Robert A. Korajczyk, 1988, Risk And Return In An Equilibrium APT: Applications Of A New Test Methodology, Journal of Financial Economics 21, 255-289. Elton, Edwin J., Martin J. Gruber, And Christopher R. Blake, 1995, Fundamental Economic Variables, Expected Returns, And Bond Fund Performance, Journal of Finance 50, 1229-1256.

Erb C., Harvey, C. And Viskanta, T. 1994. Forecasting International Equity Correlations. Financial Analysts Journal. NovemberDecember: 32-45.

Erb, C. B., Harvey, C., And Viskanta, T., 1996a. Political Risk, Economic Risk And Financial Risk. Financial Analysts Journal 52: 28-46.

Erb, C. B., Harvey C. R., And Viskanta, T. E. 1996b. Expected Returns And Volatility In 135 Countries. Journal of Portfolio Management Spring: 46-58.

Fama, Eugene F., 1990, Stock Returns, Expected Returns, And Real Activity. Journal of Finance 45, 1089-1108.

Fama, Eugene F., And Kenneth R. French, 1993, Common Risk Factors In The Returns On Stocks And Bonds, Journal of Financial Economics 33, 3-56.

Ferson, Wayne E., And Campbell R. Harvey, 1999, Conditioning variables and the Cross Section of Stock Returns, Journal ofFinance 54, no. 4, 1325-1359.

Ferson, Wayne E., And Campbell R. Harvey, 1993, The Risk And Predictability Of International Equity Returns, Review of Financial Studies 6, 527-566.

Ferson, Wayne E., And Campbell R. Harvey, 1991, The Variation Of Economic Risk Premiums, Journal of Political Economy 99, 385-415.

Ferson, Wayne, And Robert Korajczyk, 1995, Do Arbitrage Pricing Models Explain The Predictability Of Stock Returns? Journal of Business 68, 309-349.

——- 1994a, Sources Of Risk And Expected Returns In Global Equity Markets, Journal of Banking And Finance 18, 775-803.

Gangemi, A.M.,Robert D. Brooks And Robert W. Faff, 2000, Modeling Australia’s Country Risk: A Country Beta Approach, Journal of Economics and Business, 252,259-276.

Goldberg C.S. and Francisco Delgado, 2001, Integration of Emerging Markets: An Analysis of Individual Stocks, Working Paper.

Ghysels, Eric, 1993, A Time Series Model With Periodic Stochastic Regime Switches, Working Paper, University Of Montreal.

Groenewold, N., And Fraser, P., 1997, Share Prices And Macro Economic Factors. Journal of Business Finance And Accounting 24: 13671383.

Harvey, Campbell R., 1989, Time Varying Conditional Covariances In Tests Of Asset Pricing Models, Journal of Financial Economics 24, 289-318.

Harvey, Campbell R., 1991, The World Price of Covariance Risk, Journal of Finance 46, no. 1, 111-157.

Jagannathan, Ravi, And Zhenwu Wang, 1996, The Conditional CAPM And The Cross Section Of Expected Returns, Journal of Finance 51, 354.

Kothari, S. P., Jay Shanken, And Richard G. Sloan, 1995, Another Look At The Cross Section Of Expected Stock Returns, Journal of Finance 50, 185-224.

Lehmann, Bruce N., And David M. Modest, 1988, The Empirical Foundations Of The Arbitrage Pricing Theory, Journal of Financial Economics 21, 213-254.

Naranjo, A., And Protopapadakis, A., 1997, Financial Market Integration Tests: An Investigation Using US Equity Markets. Journal of International Financial Markets, Institutions And Money 7:93- 135.

Obstfeld, Maurice, 1994, Risk Taking, Global Diversification And Growth, American Economic Review 84,1310-1329.

Shanken, Jay, 1992, On The Estimation Of Beta Pricing Models, Review of Financial Studies 5, 1-34.

Sharpe, William. F., 1964, Capital Asset Prices: A Theory Of Market Equilibrium Under Conditions Of Risk, Journal of Finance 19, 425442.

Singh, R.A, (1993), “Response of Stock Prices to Money Supply Announcements: Australian Evidence”, Accounting and Finance 33, 43-60.

Singh, R.A, (1995), “Response of Financial Markets to Announcements of the Australian Current Account Balance”, Accounting and Finance 35, 155-174.

Solnik, Bruno, 1983, International Arbitrage Pricing Theory, Journal ofFinance 38, 449-457.

Cathy S. Goldberg (goldberg@usfca.edu)

University of San Francisco

John M. Veitch (veitchj@usfca.edu)

University of San Francisco

Copyright St. Louis University, John Cook School of Business, Boeing Institute of International Business Fall 2002

Provided by ProQuest Information and Learning Company. All rights Reserved