The liquidity trap: Japan, 1996-2001 versus the United States, 1933-1940

Keynes' "liquidity trap" rarely occurs. But when it does, it has a tremendously adverse effect on the economy concerned. Such was the case of the United States in the 1930s and now that of contemporary Japan. In a liquidity trap, monetary policy pushes the money interest rate to the zero level while expanding the money supply (M1) at a faster rate than nominal GDP. Conventional theory explains this phenomenon as the result of money demand that becomes infinitely interest-elastic at the zero rate, rendering ineffective the rapidly expanding money supply established by the monetary authorities. In this paper, we show that the liquidity trap is a multifaceted phenomenon not limited to the money market. It involves the bank loan market, the bank deposit market, and the bond market interacting together. Of these, the most important is the bank loan market and the least important is the bank deposit market, whose deposit supply becomes horizontal at the zero rate. They are met by relatively interest-inelastic bank loan demand and bank deposit demand. Hence, the causality is completely reversed from the conventional understanding. We give empirical evidence in support of our theory based on data from the United States, 1933-1940 and Japan, 1996-2001. Far apart in time and space, the two cases are remarkably alike and, hence, provide strong supporting evidence.