Revisiting the tale of two interest rates with endogenous asset market segmentation

Abstract

We develop a monetary model that is unique in its ability to deliver a negative correlation between aggregate consumption growth and short-term real interest rates consistent with U.S. data. The essential ingredient to this success is endogenous asset market segmentation permitting the extent of household participation in asset markets to vary smoothly with changes in aggregate conditions. Households in our model incur fixed transactions costs when exchanging bonds and money and, as a result, carry money balances in excess of current spending to limit the frequency of such trades. While we impose no stickiness at the microeconomic level in either prices or portfolio adjustment, our model drives gradual adjustment in the aggregate price level following a monetary shock and thus persistent non-neutralities. In our model, households can alter the timing of their trading activities in response to changes in both individual and aggregate states. We show that this added flexibility relative to fixed segmentation models can substantially reinforce the sluggishness in aggregate price adjustment following a monetary shock, and it can transform dramatic, transitory changes in real and nominal interest rates into more moderate and persistent liquidity effects. When we extend our setting to consider production, we find that small changes in household participation rates add substantial persistence to movements in inflation, and they deliver persistence in real interest rates that is otherwise absent. These changes are also critically important to our model's success in reproducing the empirical correlation between aggregate consumption growth and real interest rates; when they are suppressed, the success is lost.