Date of Degree


Document Type


Degree Name





Tao Wang

Committee Members

Wim P.M. Vijverberg

Thom B. Thurston

Subject Categories

Econometrics | Finance | Growth and Development | International Economics | Macroeconomics


Exchange rate risk exposure, Euro, Rolling regression, Multinational Corporation, Currency revluation, Output


Exchange rate movements are widely believed to be a major source of uncertainty at both micro- and macro-economic levels. At the microeconomic level, corporate managers controlling risk and investors seeking to hedge their portfolios are both obviously interested in estimates of a firm’s exchange rate exposure ̶ i.e., how much of the value of a firm will be affected by exchange rate movements. At the macroeconomic level, a currency appreciation may have a contractionary effect on a country's output.

Chapter 1 tests whether the launch of the euro around 1999 significantly reduced stock market volatility and exchange rate risk exposure to eurozone firms. The experiment analyzes monthly stock returns of 6574 nonfinancial firms from eleven eurozone countries, six non-euro European countries, and two countries outside of Europe (the United States and Japan) over the period of 1993-2011. I find that, near the introduction of the euro, eurozone firms had a larger decrease in market beta compared to other regions. Surprisingly, however, this decrease was only a temporary effect. Leading up to the crisis of 2008, the eurozone market beta had already shot up back to its previous level in the pre-euro period. Relative to other regions, the eurozone firms that experienced less increase in the exchange rate risk exposure after the euro's initial adoption also experienced a larger increase in exchange rate risk exposure during the crisis period. In addition, I examine the determinants of firm-level exchange exposure in a dynamic context for eurozone firms. The results indicate that exposure is correlated with components such as firm size, foreign sales, foreign assets, debt to total assets ratio, but all these relationships vary across different periods.

Chapter 2 investigates several existing exchange rate risk exposure issues in the context of US multinational firms with a new five-factor Fama-French model. A portfolio-level analysis is employed to eliminate the key source of time-variant factors that may exist in the first- and second-regression approaches employed by the existing articles. I find that (1) the relationship between the foreign sales ratio and exchange exposure may not be linear ̶ i.e., a firm with a higher foreign sales ratio is not necessarily exposed more to the exchange rate risk, and (2) a broad exchange rate index captures the exchange rate changes more comprehensively in estimation ̶ i.e., a firm could have statistically significant exposure to some currency even when it has no foreign sales in the country with that currency. In addition, new firm-level exchange exposure evidences are found on the effects of lagged period, return horizon, and large volatile period.

Chapter 3 seeks to determine whether a negative long-run effect of real appreciation on output can be identified once accounted for the following factors: (1) reverse causation from output to the real effective exchange rate, (2) spurious correlation with third factors such as capital account shocks, and (3) temporary contractionary effects of appreciation. Six vector autoregression (VAR) models are built to identify the sources of shocks and to control for important external shocks. Based on the results obtained from the VAR models, the response of output to a real appreciation is based not only on one single predominant channel but on several different channels, such as inflation, capital account, and money supply. A finding that real appreciation has led to economic contraction in China does not mean that the authorities should simply discourage real exchange rate appreciation.