نرخ رسمی چین و بازدهی اوراق قرضه

Recent research shows that bond yields are influenced by monetary policy decisions. To learn how this works in a bond market that differs significantly from those in the US and Europe, we model Chinese bond yields using the one-year deposit interest rate as a state variable. We also include the spread between the one-year market interest rate and the one-year deposit interest rate as another factor. The model is developed in an affine framework and closed-form solutions are obtained. We then test the model empirically with a Markov Chain Monte Carlo simulation procedure. The results show that the new model that incorporates the official rate in China characterizes the changing shape of the yield curve well.

Early yield curve models focus on modeling the short-term interest rate as a stochastic process. Having specified the short-term interest rate, the medium- and long-term interest rates can then be seen as functions of the short rate. One-factor models such as the Vasicek (1977) model, the Ho and Lee (1986) model, the Cox et al. (1985) model, and the Hull and White (1990) model all define interest rate movements in terms of the dynamics of the short rate. Later models include other factors that may influence the yield curve. For instance, the Brennan and Schwartz (1979) model uses both the short- and long-term interest rates to define the yield curve. Heath et al. (1992) also show that it is possible to use the information contained in the whole yield curve to model the evolution of interest rates. Besides parametric models such as those mentioned above, there are also studies that model the term structure using nonparametric approaches (e.g. Gómez-Valle and Martínez-Rodríguez, 2009). There is a growing literature that uses economic variables to model the yield curve (e.g. Ang and Piazzesi, 2003, Gallmeyer et al., 2005, Duffee, 2006, Ang et al., 2008 and Spencer and Liu, 2009). Recently, a number of studies have focused on the relationship between monetary policy and the yield curve (e.g. Piazzesi, 2005). In the US and Europe, monetary policy is carried out either through a specific interest rate, such as the federal funds rate, or by adjusting money supply. Monetary policy in this framework focuses on the short-term interest rate, while the medium- and long-term interest rates are set by the market. It is thus natural to assume that monetary policy directly influences the short-term interest rate and thereby affects the whole yield curve. Different approaches to yield curve modeling have therefore been developed based on this standard monetary policy framework. The new literature that brings macroeconomic and finance perspectives together is of significant interest to researchers, policy makers, and market participants. However, so far, studies that take monetary policy into consideration when modeling the yield curve have almost exclusively focused on the US and, to some extent, Europe. Even though these studies help us understand the underlying processes in the US and European bond markets, they are difficult to apply to countries with monetary systems that differ significantly from those of the US and Europe. For example, while the US central bank uses the target interest rate to set monetary policy, China’s central bank, the People’s Bank of China (PBC), uses a set of official interest rates among which the one-year deposit interest rate is arguably the most commonly used benchmark. One of the main novelties of this paper is that it takes the institutional features in China into account when modeling domestic bond yields. We apply a multi-factor model that incorporates the one-year deposit interest rate to bond yields with maturities ranging from one to five years. The deposit rate is directly specified by the central bank and has a direct impact on Chinese market rates (Fan and Johansson, 2009). The market interest rates are assumed to be influenced by the one-year deposit rate and the spread between the one-year market rate and the one-year deposit rate. The deposit rate is specified with a stochastic jump process. This paper thus differs from related studies in that it includes the one-year deposit rate and the spread between the market rate and the official rate as state variables rather than a typical target rate. The jump process for the deposit rate also differs from those of earlier studies (e.g. Piazzesi, 2005). We assume that the jump size follows a stochastic process. This approach allows us to capture some of the unique features in China’s monetary policy conducted through official rates. To model the bond yields, we use an affine approach (see Duffie and Kan, 1996). Then, we derive a closed-form solution for the model and use a Markov Chain Monte Carlo (MCMC) procedure to estimate its parameters. It is shown that the model captures the movements in yields with different maturities well. The rest of the paper is organized as follows: Section 2 discusses related literature on the relationship between monetary policy and bond yields. It also brings up some of the unique features in China’s monetary policy and their effects on bond yields. Section 3 introduces the new model. Section 4 briefly explains the estimation procedure and then discusses the data and the empirical results from the estimation. Finally, Section 5 concludes the paper.

Several recent research articles focus on how monetary policy influences bond yields in the US and Europe. We take a similar macro-finance approach when we model bond yields in China. While the so-called CHIBOR and SHIBOR were created to enhance the pricing mechanisms in the domestic bond market and to improve the monetary transmission system, a number of alternative official rates are still used as benchmarks by market participants and analysts. Among the many different official rates in China, the one-year deposit rate is arguably the most commonly used benchmark rate. Due to the importance of the one-year deposit rate, we argue that it is suitable to include it as a state variable in a model for bond yields in China. We also include a second state variable, the spread between the one-year market rate and the one-year deposit rate. This variable reflects the effect of other economic variables that influence market interest rates. We use an affine framework to price bonds and derive a model of the yield curve that includes the state variables. Using monthly yield data for different maturities and the one-year deposit rate during the period 1998–2007, we estimate the model with MCMC simulation. We show that the estimated bond yields fit the bond yields observed in the market well and that the model captures the main features of the bond yields at different maturities in terms of mean, standard deviation, skewness and kurtosis. The model captures the movements in the yield curve during periods of rising, falling and stable interest rates. The estimates of the model parameters also show how the jump risk in the official interest rate and the risk of the spread between the one-year market rate and the one-year official rate influence the excess returns of bonds. Finally, a comparison with the standard Vasicek, 1977 and Cox et al., 1985 models shows that the new model performs well with lower pricing errors across all maturities. This study is an attempt to improve the understanding of how monetary policy influences bond yields in China. As seen in recent studies on the US and European yield curves, an approach that includes monetary policy when modeling the yield curve can have positive results. The results in this study suggest that a macro-finance approach for analyzing bond yields also works in a setting that differs significantly from that of the US and Europe. Given these initial encouraging results, a number of venues for future research open up, including using alternative economic variables such as inflation and other measures of monetary policy as explanatory variables when analyzing China’s bond market.