تاثیر سیاست های پولی در اوراق قرضه را برمی گرداند : رویکرد بازارهای تقسیم بندی شده

This paper assesses the contribution of monetary policy to the dynamics of bond real returns. We assume that the monetary authority controls the short-term nominal interest rate. We then model exogenously the joint dynamics of the aggregate endowment and the monetary policy variable, and determine bond real returns endogenously. Market segmentation is introduced by permanently excluding a fraction of households from financial markets. When markets are segmented, monetary policy has a liquidity effect on the participants’ consumption and marginal utility, on the stochastic discount factor, and on real returns. Data on bond returns strongly favor the segmented markets model over the full participation model. For maturities up to 2 years, the segmented markets model is able to replicate the sign and the size of the impulse response of bond returns to monetary policy shocks, it correctly predicts the sign of their autocorrelation, and it closely matches their volatility as a function of maturity.

This paper assesses the contribution of monetary policy to the dynamics of bond real returns. We assume that the monetary authority controls the short-term nominal interest rate. We then model exogenously the joint dynamics of the aggregate endowment and the monetary policy variable, and determine bond real returns endogenously. We adopt a heterogenous agents variant of the limited participation framework, the segmented markets model, previously studied by Alvarez and Atkeson, 1996, Alvarez et al., 2001 and Occhino, 2004, and Lahiri, Singh, and Vegh (2007). The central feature is that a set of households are permanently excluded from financial markets. In the full participation version of the model, real returns are determined by the marginal utility of the representative household, and, therefore, by the aggregate consumption and endowment. Hence, monetary policy affects real returns through its effect on the aggregate endowment. When markets are segmented, however, monetary policy has an additional liquidity effect. Changes in the stance of monetary policy affect the distribution of cash balances and consumption expenditures across households. An increase in interest rates induces traders to hold more bonds, to lower their holdings of cash balances, and to reduce their purchases of consumption goods. The traders’ marginal utility of consumption rises, lowering the stochastic discount factor, and increasing expected real returns. The smaller the economic weight of traders in the economy, the larger this liquidity effect of monetary policy on bond real returns. We take the full participation and segmented markets models to the data. Three empirical dimensions are explored: the response of bond returns to nominal interest rate shocks; the autocorrelation of bond returns; and the term structure of volatility. The evidence strongly favors the segmented markets model in each case. The full participation model has incorrect predictions about the impact effect of monetary policy, with real returns rising after an increase in interest rates. Real returns fall in the segmented markets model and closely track the impulse responses in the data thereafter. The segmented markets model also matches the declining positive autocorrelations and increasing volatilities of bond returns as time to maturity increases. The full participation model has negative autocorrelations and can only match the higher volatilities of longer term bond returns by overstating short-term bond volatility. The paper is organized as follows: Section 2 describes the economy and defines the equilibrium; Section 3 explains the numerical solution method; Section 4 presents and comments on the empirical results; Section 5 concludes.

In a segmented markets model, we have been able to account for the contribution of monetary policy to bond real returns. Data on Treasury bond returns strongly favor the segmented markets model over the full participation model. For maturities up to 2 years, the segmented markets model is able to replicate the sign and the size of the impulse response of bond returns to monetary policy shocks, it correctly predicts the sign of their autocorrelation, and it closely matches their volatility along the yield curve. In future work, we plan to study the effect of endogenizing production. With real sector shocks, we hope to explain the impact of segmented markets on long term bonds and equities.