Stock prices, foreign reserves, and regime collapse

Abstract

Using a regime collapse model, this paper analyzes the impact of foreign financial disturbances in the foreign exchange market on the economy under the assumption of perfect foresight. When there are foreign financial disturbances and the amount of foreign exchange reserves reaches the threshold, the government contracts the domestic credit so as to prevent an additional decrease in foreign reserves. The results show that the relative scale of the threshold for foreign reserves influences the timing of the regime collapse, the extent of domestic credit contraction and the dynamic adjustment of the economy.

Keywords:

• foreign financial disturbances
• regime collapse
• dynamic adjustment

JEL Classification:

• F41

Acknowledgment

The authors are grateful to two anonymous referees and the editor for their invaluable comments. All remaining errors are however our own.

Notes

1. Blanchard (1981) assumes that domestic stocks and domestic bonds are perfect substitutes in the closed-economy, while Chu and Tsaur (1987), and Gavin (1989) assume that domestic stocks, domestic bonds, and foreign bonds are perfect substitutes in the open-economy.

2. The goods market could be set as following Dornbusch (1976), which implies that excess demand in the goods market raises the price level. Nevertheless, the conclusion remains the same in the long run under the assumption of a slowly adjusting price level.

3. See Bhandari (1982) for a relevant definition of .

4. The stock return rate in Equation (3(3) ) is , which can be obtained as follows. Suppose Q, Y, P, Q^e, i, and w represent the stock price, aggregate output, the price level, the expected stock price, the domestic nominal interest rate (or, the rate of return on bonds), and the ratio of dividends per domestic stock share to aggregate output, respectively. In addition, one unit of domestic currency can buy units of stocks and acquire wPY/Q dividends. Moreover, 1/Q units of domestic stock shares are expected to obtain Q^e/Q units of domestic currency over the trade. Therefore, the expected rate of return of one unit of domestic currency in buying domestic stocks, γ^s, can be derived as Assume that people have perfect expectations that the expected rate of change in stock prices equals the rate of change in practice, i.e. (Q^e− Q)/Q = , the Taylor expansion of γ^s will result in . Since bonds and stocks are assumed to be perfect substitutes in this paper, their return rate will be equal in equilibrium, as shown in Equation (3(3) )

5. Assume that Q is the buy price, Q^e is the expected sell price, and is the fluctuation in the stock price. Also assume that dividends are a constant proportion (j′) of domestic output (PY), where P is the domestic price level. The investment in the domestic stock market using the domestic currency will bring residents who are resident in the home country the following rate of return: Under the assumption of rational expectations and by applying a Taylor series, the rate of return for domestic stocks is rewritten as , where the first item is the capital gain/loss and the second item is dividends

6. Here we exclude the J-curve effect since the Marshall–Lerner condition may not hold in the short run, i.e. φ < 0. See Driskill (1981) and Bhandari (1983).

7. The moment T is nonpositive in situations (I) and (III). The results are available from the authors upon request.

8. Even though there are two types of diagrams for the economy before the regime collapse, depending on the relative sizes of k and , the model developed in this paper predicts the same dynamic adjustments before and after the regime collapse. That is, the dynamic adjustments are independent of the relative sizes of k and .

9. In the literature on regime collapse, such as Flood and Garber (1984), residents with a forward-looking prediction engage in a panic buying of foreign reserves at the instant when they predict the occurrence of a depreciating domestic currency resulting from the continuous depletion of foreign reserves and a further collapse of the fixed exchange rate regime with its replacement by a flexible exchange rate regime. The phenomenon is the speculative attack. However, the model in this paper assumes a fixed exchange rate system before and after regime collapse, and therefore no speculative attacks occur at the instant of the regime collapse. This is in concert with Keister (2009) who stated that “a currency market with a fixed exchange rate regime can be viewed, as in Obstfeld (1996), as a coordination game. This game often has multiple equilibria, one in which speculators attack the currency and a devaluation occurs and another in which no one attacks and no devaluation occurs.”