Content area
Full Text
(ProQuest: ... denotes formulae omitted.)
Recently, evidence about the effect of supply and demand on bonds has challenged researchers and bond managers to rethink the determinants of the term structure of interest rates because standard term structure models assume perfect markets without accounting for supply and demand. A well-known observation is the U.S. Treasury's buyback program of2000-2001. Greenwood and Vayanos [2010] find that after the Treasury buyback plan of long-term bonds was announced in January 2000, the spread between the 20and 5-year spot rates decreased from 26 basis points to -39 basis points in three weeks (a total decrease of 65 basis points). Other observations include the effect of the U.K. pension reform of 2004 on U.K. government bond yields (Greenwood and Vayanos [2010]), and the effect of international capital flows on U.S. interest rates and bond returns (Warnock and Warnock [2009]; Sierra [2010]).
An important practical question to a global bond manager is how broad such a problem exists and how to account for the effects to better manage a bond portfolio. In this article, we introduce a simple preferredhabitat model to distinguish roles of supply and demand from that of alternative investment opportunities in bond pricing. Moreover, we implement the model and provide empirical evidence to show the substantial economic impact from supply and demand.
The supply and demand factor is clearly much more of an issue in emerging markets than in developed markets. To simplify our analysis, we focus on the Chinese bond market, which is large and receiving increasing attention from bond portfolio managers around the world. The Chinese government maintains a persistent preferredhabitat demand effect by setting an official term structure for alternative investments on bank loans. In our model, following Vayanos and Vila [2009], we assume that there are two types of investors in the government bond market: preferred-habitat investors and arbitrageurs. We also assume there are assets with similar maturities as government bonds and the yields on these assets serve as return benchmarks for government bonds. Arbitrageurs seek maximal expected return with given risk by holding only bonds (in other words, we assume arbitrageurs can't arbitrage the yield difference between bonds and other assets with similar maturities for institutional reasons, which is approximately true in our setting...