See how they run: Linkages between national elections and the behavior of international banking flows in developing countries

This dissertation studies 115 national elections in 35 developing countries over a fifteen-year period from 1985 to 1999. Pooled time-series data on five classifications of banking flows are considered, as well as country specific data originating from banks in 14 leading developed countries. Relying on data published by the Bank for International Settlements, I find the presence of an “election effect,” a risk that may cause international banks to reduce credit exposures to a given country in the period before a national election. Such actions may contribute to the potential for ex-ante and ex-post electron liquidity crises. Credit squeezing behavior by bankers before, during and after national elections may be an attempt to reduce heightened default and renegotiation risks, a financial counterbalance to the behavior of opportunistic and/or partisan politicians seeking reelection. A framework is developed to show that there is an intersection between political business cycle theory and research into banking and currency crises. The focus of the framework is on the behavioral aspects of risk decisions rather than measurement of asset price changes. The framework, as it applies to transaction volumes and liquidity, can also be applied to other asset markets, including bonds, equities and foreign direct investments.; National elections may be a country's most important validation of democracy and the rule of law. In open elections, citizens choose from a competitive roster of candidates, thereby formalizing principal - agent relationships. In emerging markets, not all elections are competitive, a factor accounted for in this research. Elections imply uncertainty because they may bring about policy evolution in a host of areas, both political and economic in nature. I argue that bankers believe in the existence of opaque political actions, including those designed solely to ensure reelection. These actions may damage a country's development prospects. Furthermore, elections may affect a country's ability and willingness to honor international debt contracts. A variety of political control variables are used to test the importance of election rules, fraud, executive power and political parties.; Distribution of pre-election goodies may cause a country's balance of payments position to deteriorate as external debt increases, particularly short-term obligations. Further, in the period before an election, international reserves may silently fall as governments expend hard currency holdings, an attempt to maintain inflated local currency values. A financial crisis in one developing country may also affect the supply and price of credit in other countries, thereby disrupting international financial markets. This research controls for pre-existing banking and currency crises in order to isolate the election effects on international banking flows. The concept of a liquidity event, as it pertains to international banking claims is introduced and tested. This dissertation postulates that any election effect is supply-side driven, rather than representing reduced demand by developing country borrowers.