The Financial Accelerator Effect : Multiresolution Evidence from the C&I Business Loan Market

Aggregate empirical evidence on the financial accelerator effect is highly debated. In this study, we apply multiresolution analysis, based on the continuous wavelet transform, to investigate the relationship between aggregate net worth and credit market variables in the commercial and industrial (C&I) business loan market. We find that nonprice terms, represented by bank lending standards, are better indicators of business lending of commercial banks than interest rate spread on bank loans. Furthermore, when we examine whether C&I loan market variables are related to changes in aggregate net worth across frequencies and over time, we show that the financial accelerator effect is both time- and frequency-dependent. The relationships are significantly related at lower business cycle frequencies, 4 to 8-10 years, and their statistical significance is concentrated in the post-1990s period. We claim that the shift of the business cycle frequency range towards lower frequencies and the deregulation of credit markets in the 1990s support the interpretation of the financial accelerator as a longer run mechanism operating in financial deregulated markets characterized by asymmetric information