Search ETDs:
Every bank run need not cause a currency crisis. models of twin crisis with imperfect information
Solomon, Raphael Haim Reuven

2003, Doctor of Philosophy, Ohio State University, Economics.
In the 1970's, the “twin crisis” pattern emerged: currency crises often followed bank runs. Many observers developed deterministic twin crisis models: every bank run causes a currency crisis. These models contrast with the historical record: even if several banks fail, the currency peg may survive. A satisfactory theory of such phenomena allows for endogenous twin crises. This dissertation consists of two endogenous twin crisis models and an empirical examination of these models. The first chapter, Anatomy of a Twin Crisis, models twin crisis in a one-bank setting with risk-averse domestic depositors and risk-neutral foreign depositors. Some domestic depositors (“impatient”) require immediate liquidity. Other depositors demand immediate liquidity when fearing bank failure. Depositors use imperfectly correlated sunspot variables when deciding to run. Domestic depositors observe one sunspot variable; foreigners observe another. A net foreign reserve drain may cause a currency crisis. Foreign depositors contribute to this drain through currency conversions; domestic depositors exacerbate the drain by running, forcing a bank bailout. When foreign and domestic depositors run on the bank, currency crisis results; a run by foreign or domestic depositors alone need not provoke a crisis. The second chapter, When Bad Things Happen to Good Banks, emphasizes systemic aspects of twin crises. This model features multiple ex-ante identical banks and depositors who observe sunspots specific to their bank and country of residence. After calibrating to Turkish data, the model computes “systemic risk” - the probability that sufficient banks fail, causing a currency crisis. Systemic risk may be fundamental or self-fulfilling. As the fraction of impatient depositors increases, the banks' emphasis shifts from serving depositors to providing consumption insurance. Also, almost all systemic risk is self-fulfilling. The model predicts that increasing interbank deposits encourages banks to consider systemic impacts of their actions. The third chapter, Diamond-Dybvig Theory Passes a Simple Test, examines withdrawal decisions by domestic and foreign investors in detail. When depositors “run,” they choose short-maturity deposits over long-maturity ones. If many depositors run, banks face “liquidity shocks.” This chapter demonstrates a relationship between interest rates offered by banks cognizant of the possibility of runs and the liquidity shocks they face.
Nelson Mark (Advisor)
144 p.

Recommended Citations

Hide/Show APA Citation

Solomon, R. (2003). Every bank run need not cause a currency crisis. models of twin crisis with imperfect information. (Electronic Thesis or Dissertation). Retrieved from https://etd.ohiolink.edu/

Hide/Show MLA Citation

Solomon, Raphael. "Every bank run need not cause a currency crisis. models of twin crisis with imperfect information." Electronic Thesis or Dissertation. Ohio State University, 2003. OhioLINK Electronic Theses and Dissertations Center. 21 Apr 2015.

Hide/Show Chicago Citation

Solomon, Raphael "Every bank run need not cause a currency crisis. models of twin crisis with imperfect information." Electronic Thesis or Dissertation. Ohio State University, 2003. https://etd.ohiolink.edu/

Files

osu1054309457.pdf (776 KB) View|Download