The Monetary Transmission Mechanism, Non-residential Fixed Investment and Housing

We specify a vector autoregression (VAR) model for the U.S. for 1980–2008 to investigate the statistical causal relationships between private non-residential fixed investment, the effective Federal funds rate, personal consumption expenditures, nonfinancial corporate profits, and the nonfinancial corporate credit market debt to test the validity of macroeconomic relationships in a macro model. The VAR utilizes the Toda-Yamamote procedure to test for Granger causality. Our preliminary results show that the transmission mechanism does not work as expected; we find that fixed investment depends on the level of demand in the economy and profits but not on the interest rate. This casts doubt on the usual assumptions about how the monetary transmission mechanism is expected to work. The second part of the paper investigates the effects of the change in the monetary regime towards low and stable interest rates, a policy pursued by the U.S. Fed since the beginning of the 1990s. We find that the new monetary policy regime has the following effects: (1) our VAR model does not support the hypothesis that low interest rates lead to higher fixed nonresidential investment; (2) low interest rates led to a search for higher yields through increasing risk, and (3) they led to an increase in the demand for securitized assets, especially mortgage-backed securities, which eventually resulted in a housing bubble. The overall results therefore raise doubts about the effectiveness of low interest rates as a policy regime designed as a component of a counter-cyclical policy. Copyright International Atlantic Economic Society 2013