UC San Diego

Official dollarization is rare; yet, many countries' banking systems are characterized by high levels of unofficial dollarization. This dissertation provides a political-economic argument for why governments allow these deposits to exist at all. I explain the timing and level of regulations over resident foreign currency (FC) deposit accounts as well as why some countries that chose partial liberalization tend to remain there and test that explanation on a new dataset I created coding IMF reports on exchange-rate controls. Similar to other forms of monetary delegation, unrestricted accounts can serve as a commitment device for credible macroeconomic policy, contingent on two factors. Politicians must be responsive to a broad audience; and, they must be credible in their protection of property rights. Thus, unrestricted accounts are most likely when: politicians are responsive to a broad population, have experienced macroeconomic instability, and are respectful of their citizens' property rights. Responsiveness is important because restricted deposits are unlikely to be useful as a commitment device. When responsiveness is broad group, politicians are more likely to provide public goods rather than targeted, private goods. Allowing FC deposits with restrictions resembles a private good, in that politicians choose which groups are allowed access to the accounts, while unrestricted deposits resemble non-excludeable public goods. Unrestricted deposits should be more likely in those countries where politicians are responsive to a broad population and face incentives to provide public goods. Second, unofficial dollarization is a strategic situation. If property rights are weak, citizens should anticipate that their deposits would be at risk, making individuals less likely to use FC accounts. Governments should anticipate this as well; if not credible on property rights, they should be more likely to adopt restricted deposits. I test this argument on a new dataset of regulations of resident FC accounts in a sample of developed and developing countries I coded from the IMF's Annual Report on Exchange Arrangements and Exchange Restrictions. I use both binary cross-sectional time- series analysis and a competing risks model. I find support for my central argument, although its effectiveness is limited to periods of moderately levels of macroeconomic instability