This paper reports tests of hypotheses that a variety of interest rates and other measures from financial markets in countries belonging to the European Monetary Union (EMU) were converging prior to the introduction of the euro in January 1999. We expected to find convergence because of i) removal of national barriers to flows of funds, ii) explicit and market-driven harmonization of regulation and supervisory standards, iii) coordinated macroeconomic policies, iv) privatization of state enterprises, and v) fiscal redistribution of resources. The first series of tests (sigma-tests) are that standard deviations and/or coefficients of variation of cross-sections of national measures are diminishing over time, relative to a group of non-EMU countries. Evidence of convergence was found for inflation rates, short- and long-term nominal interest rates, and ex post real short-term rates, but not for real per capita GDP. The second series concerned levels and trends in interbank claims and noninterest income at banks. These measures are believed to be larger and growing more rapidly when banks are attempting to escape binding national regulations. Interbank claims were larger at EMU banks than at banks in other countries, but had no interpretable trends. The ratios of noninterest income to total bank income and assets were found to have positive trends. The third series of tests used a statistical cost accounting model estimated for twelve countries to examine whether marginal costs of liabilities and revenues from assets were tending toward equality, as might be expected in an efficient unified economy. Within the EMU, significant differences across six major countries were observed for 1994, 1995, and 1996, but not in 1997 or 1998. Convergence seemed to be being achieved.
Read full abstract