China's Financial Markets, The Benchmark Interest Rate Selection Problem

Our country is developing the economy fast, and the financial market needs a benchmark rate to provide a reference to the other interest rates. As a pricing standard for other interest rates, the benchmark rate has to be capable to reflect the situation of the demand and the supply of the money in the market, and it could affect other interest rates effectively. Also, the benchmark rate needs system stability and can be used as an instrument to control the market. If monetary policies want to be made and operated reasonably, it becomes important to make it clear that which one is the benchmark. It is also helpful to judge whether the money is used suitably or the financial assets' prices proper. For now, we are obliged to develop the future market and the financial derivatives market, which both call for a benchmark interest rate.According to the international experience, the development process needs carrying out in steps. The benchmark rate is chosen from the existing interest rates, and which one can take on the responsibility is crucial to the success of the reform. In the background of this, the paper compares the existing rates, concludes the characteristics of each one, tries to find out the benchmark, and digs hard to find out the reasonable method to testing the benchmark rates.It includes seven chapters: Introduction tells you why we choose to study the benchmark rates and the point of it The second part is about other analysts' research on this field, and followed by the detailed description of existing interest rates system in our country. Here we introduce the history, the development and the state for now of the interest rates used in our money market. Meanwhile, the advantages and disadvantages of them are discussed. The fourth part is about benchmark rates in the international financial market. We hope to find something meaningful to the choice of the benchmark for our country. Then, we give the method used in this paper, including main theories and the model. Working steps and the results can be found in this chapter, too. We apply for nonparametric density to estimate the distribution function, and in Hilbert spaces to calculate the mean-variance frontier. After comparison, the result, which interest rate stands for the standard, comes out. At the end of the paper are conclusive discussion and political advices.The main contribution of us is the method, used to examining benchmark rates status, has a theory base. Most analysis before us generalized the characteristics of benchmark from the purely empirical matter, and it somehow makes the research lack of support from the theory. We begin with Beta pricing theory to find out prerequisites of the benchmark rates. If an interest rate is the benchmark, it should be in mean-variance frontier. So we calculate out the frontier spaces, and test if the interest rates are mean-variance frontier efficient. Another contribution is that we use nonparametric method to estimate the density function. It is a better estimation when the total distribution is unknown. Opposing to classical parametric estimation, nonparametric method dose not require the variances only belonging to the total distribution. As a result, the estimation is more efficient.