A number of empirical studies have identified an important role of banks for real economic activity. Samolyk (1994) examines the relationship between banking conditions and economic performance at the U.S. state level and shows how regional banking conditions can affect local economic activity by impacting on a region's ability to fund local investments. Peek and Rosengren (2000) focus on real estate lending of Japanese banks in U.S. states following the Japanese banking crisis.
They find that reduced real estate lending by Japanese banks was not compensated by increased lending of domestic banks and had a significant and sizeable impact on real construction projects. Harrison et al. (1999) analyze the relationship between economic growth and cost of bank monitoring using U.S. state level data. They find an inverse, albeit small, relationship between state per-capita income and the cost of banking suggesting a feedback between real and financial development. Gambacorta and Mistrulli (2004) analyze a sample of Italian banks and show that bank capital matters in the propagation of different types of shocks, owing to the existence of regulatory capital constraints and imperfections in the market for bank fund-raising.
Despite the rich empirical literature on the role of banks, there is only a small number of theoretical macroeconomic models linking the banking sector to the macroeconomy. Friedman (1991) notes: "Traditionally, most economists have regarded the fact that banks hold capital as at best a macroeconomic irrelevance and at worst a pedagogical inconvenience." The lack of a tractable macroeconomic model that provides a meaningful role for banks in intermediation has left several important questions unanswered: Do variations in banks' net worth significantly affect macroeconomic outcomes? Does incorporating banks into a standard business cycle model change the propagation of business cycle shocks? Are there implications for the conduct of monetary policy arising from an explicit focus on bank loan supply?