The impact of treasury bond futures prices on the transfer discount rate of bank acceptance bills
Abstract: As a financial derivative, treasury bond futures occupy an important position in China's interest rate market-oriented reform, but researchers lack discussion on the externality impact of treasury bond futures prices. Based on the two-way fixed-effect model, this paper empirically tests the impact of treasury bond futures price on a specific market interest rate, the banker's acceptance bill rediscount rate. The results show that the price of treasury bond futures has the ability to discover the price of the banker's acceptance bill to the discount rate. The regulatory effect of treasury bond futures prices on the transfer discount rate of bank acceptance bills is mainly achieved by adjusting investment yields and economic expectations. Further analysis shows that there is a maturity heterogeneity in the impact of treasury bond futures prices on the rediscount rate of bank acceptance bills, and the change of the price of treasury bond futures with a shorter maturity is more significant than that of treasury bond futures with a longer maturity on the rediscount rate of bank acceptance bills. The above results show that the government should strengthen market supervision and optimize macroeconomic policies to promote the positive interaction between the treasury bond futures market and the interbank market.
Keywords: Treasury bond futures; Banker's Acceptance; rediscount rates; Return on investment
I. Introduction
Treasury bond futures are an interest rate derivative whose price can reflect the market's expectations of future interest rates and can also be used to hedge interest rate risk. The discount rate of bank acceptance bills is an important interest rate in the interbank market, reflecting the supply and demand of funds between banks. In April 2019, the China Securities Regulatory Commission (CSRC) and the International Monetary Fund (IMF) jointly organized a seminar on treasury bond futures, which once again emphasized the role of treasury bond futures in the implementation of monetary policy, the construction of interest rate curves, and the management of interest rate risks, and pointed out the importance of the improvement of China's treasury bond futures market to help the implementation of national macro-financial policies and the two-way opening of the financial market. Studying the impact of treasury bond futures on the transfer discount rate of bank acceptance bills can help us understand the interaction between the treasury bond futures market and the interbank market, as well as the transmission mechanism of treasury bond futures prices on the interbank cost of funds.
Treasury bond futures are futures contracts with treasury bonds as the underlying asset, in which the buyer and seller agree to trade a certain amount of a certain treasury bond at a certain price at a specific date in the future. Treasury bond futures are a type of interest rate futures and are a kind of advanced financial derivative instruments. Under the interest rate marketization, the spot interest rate of treasury bonds is affected by many factors such as the level of market interest rates, the supply and demand of treasury bond funds, and monetary policy, and the fluctuation range is large. Investors can not only hedge and avoid interest rate risks through treasury bond futures, but also make full use of its price discovery function to accurately and timely capture the possible price changes of treasury bonds. Today, China's treasury bond futures products have four varieties: two-year, five-year, ten-year, and 30-year, basically forming a treasury bond futures product system covering short, medium and long-term. To sum up, the changes in various aspects of treasury bond futures will directly reflect the supply and demand of funds and the trading activity of the treasury bond spot market, and have a significant relationship with the fluctuation of market interest rates with treasury bond yields as the benchmark interest rate.
In order to straighten out the relationship between the treasury bond futures market and the interbank market, this paper uses a two-way fixed-effect model to test the impact of treasury bond futures prices on a specific market interest rate, the banker's acceptance bill transfer discount rate. Firstly, from the perspective of theoretical analysis, this paper proposes two possible influencing mechanisms, namely the investment yield mechanism and the macroeconomic mechanism, and puts forward the theoretical hypothesis that the price of treasury bond futures has the ability to discover the price of the bank acceptance bill to discount rate. Secondly, the 3-month, 6-month, and 12-month bank acceptance bill transfer discount rate was selected as the explanatory variable, and the daily opening price of the treasury bond futures contract was used as the core explanatory variable and the appropriate control variable, and the two-way fixed-effect model was used to perform benchmark regression for the sample interval from December 5, 2018 to February 17, 2023, to empirically test the significance of the impact of the treasury bond futures price on the bank acceptance bill transfer discount rate. The price of treasury bond futures has the ability to discover the price of bank acceptance bills to the discount rate; Furthermore, the robustness test of the original sample interval is carried out by discussing the endogeneity, excluding the impact of disasters, adding new control variables, and replacing the core explanatory variables, which also confirms the price discovery ability of the treasury bond futures price to the bank acceptance bill discount rate, and the above conclusion is robust. Finally, this paper further expands the analysis of the investment yield mechanism and the macroeconomic mechanism, which provides reliable evidence for the empirical effect of treasury bond futures prices on the transfer discount rate of bank acceptance bills, and reveals the internal influence mechanism behind the benchmark regression results.
This paper connects the treasury bond futures market and the interbank market, and supplements the discussion of the relationship between the two markets. Many foreign scholars have conducted research on the price discovery and price fluctuation relationship between spot and futures of treasury bonds. In terms of the relationship between price discovery, Chari et al. (1990) concluded through theoretical analysis that the game between hedgers and speculators in the futures market promotes the price discovery of the spot market in the futures market. On this basis, subsequent scholars empirically test the price discovery function. For example, Harvey (1996) and Miyanoya et al. (1999) demonstrated that the price of Treasury futures reacts faster than the price of Treasury bonds to new information, reflecting the price discovery function of Treasury bond futures. In terms of price volatility, Engle (1998) found that the volatility and implied volatility of treasury bond futures and spot are related to the macroeconomic environment. Due to the inevitability of the development of treasury bond futures in China's financial opening up, domestic scholars have also conducted in-depth research on it. Zhang et al. (2019) and Zhang Maojun et al. (2019) both studied the cross-market information transmission mechanism of China's treasury bond futures and spot market, enriching the research content of treasury bond futures. Zhang Zongxin and Zhang Xiuxiu (2019) focused on whether the introduction of treasury bond futures contracts can play a stabilizing effect in the spot market, and found that China's treasury bond futures have achieved the function of suppressing the volatility of the spot market and can reduce the volatility impact of the financial cycle. The above literature focuses on the linkage between the treasury bond futures market and the spot market, but lacks discussion on the externalities of the treasury bond futures market to the interbank market. This paper complements the study of the interaction between the treasury bond futures market and the interbank market.
This article focuses on a specific market interest rate and complements the description of changes in the discount rate for bank acceptance bills. Interest rates are the price of money in the financial markets, and there are many factors that lead to their movements. For example, Zhang Xueying (2012) found that various market interest rates of different maturities are affected by the adjustment of the reserve ratio, and Qiang et al. (2018) revealed that the policy benchmark interest rate, liquidity factor, and risk premium factor are important variables that determine the interest rate of each period in the bond market. For example, Zhang Xueying and He Feiping (2014) have concluded that the changes in the market repo rate are significantly guided by the central bank's repo operation rate, both in the long and short term. Liu et al. (2018) conclude that benchmark interest rates, market premiums, and policy premiums are important external factors influencing bank interest rate decisions. Kacperczyk and Schnabl (2010) concluded that the three main factors influencing the change of interest rates in the commercial bank bill market are the substitution of other sources of financing in the market, adverse selection and the institutional constraints faced by money market funds. Retkwa (1982) also paid attention to the commercial bill market and concluded that there was a positive correlation between the commercial bill market interest rate and the liquidity and benchmark interest rate of commercial banks. The above-mentioned literature may start from the changes of several types of market interest rates at a macro level, or study a more specific type of market interest rate, but few literatures directly study the changes in the transfer discount rate of bank acceptance bills. This article complements the discussion of this interest rate change.
This paper further examines the direct impact of treasury bond futures prices on the rediscount rate of bank acceptance bills, while the existing literature only reveals the interest rate changes caused by treasury bond futures. Wu Xiaoqiu and Ying Zhanyu (2003) pointed out that treasury bond futures essentially take treasury bonds as the carrier and interest rates as the trading object, and the root cause of their emergence lies in the market's demand for avoiding interest rate risks, and the process of interest rate marketization provides the basic conditions for the rebirth of treasury bond futures. The introduction of treasury bond futures can promote the process of interest rate liberalization, and domestic scholars have explored the reasons and mechanisms for this. In terms of reasons, Hu Yuyue et al. (2012) pointed out that due to the complete competition characteristics of the treasury bond futures market, the market benchmark interest rate can have sufficient market credibility to maintain the value of the spot market. Chen Han (2014) pointed out that treasury bond futures promote the smooth progress of interest rate marketization reform for three reasons: promote the formation of a benchmark interest rate system, improve the information efficiency and pricing efficiency of the bond market, and assist financial institutions in managing interest rate risk. He Ping (2017) found that the introduction of formal trading of treasury bond futures has a significant calming effect on the volatility of the interest rate market and increases the stability of the market. In terms of mechanism, Zeng Yun et al. (2019) revealed that the listing of treasury bond futures has led to the bond market's increased sensitivity to changes in the repo market interest rate, and the interest rate transmission of monetary policy has been smoother. However, the above-mentioned existing literature focuses on the impact of treasury bond futures on the overall interest rate market, and few literature focuses on the impact of treasury bond futures on the rediscount rate. Therefore, this paper complements the direct study of the impact of treasury bond futures prices on the discount rate of bank acceptance bills.
The subsequent structure of this paper is arranged as follows: the second part is the theoretical hypothesis; The third part is the study design; The fourth part is the empirical results; The fifth part is extended analysis; Finally, the conclusion and enlightenment of this paper.
2. Theoretical hypotheses
Treasury futures price refers to the price of Treasury futures contracts traded on futures exchanges. Treasury futures prices are affected by a variety of factors, including the real interest rate of Treasury bonds, market interest rates, economic data, monetary policy, and investors' expectations. Changes in Treasury futures prices can reflect the market's expectations for future Treasury spot prices and interest rate movements, helping investors manage interest rate risk or insulate their portfolios from price fluctuations through strategies such as hedging. The discount rate of bank acceptance bill transfer refers to the discount interest charged by the counterparty and the annualized interest rate of the coupon amount when the bank or other financial institution holding the bill transfers the right to endorse the bill to other banks or financial institutions to obtain funds before the bank acceptance bill expires. This interest rate reflects the cost of financing the bill in the secondary market, as well as the market's assessment of the risk and liquidity of the bill.
The impact of treasury bond futures prices on the transfer discount rate of bank acceptance bills is supported by the following two theories. On the one hand, according to the asset allocation theory, investors allocate the funds in their portfolios to different asset classes according to their risk appetite and expected returns in investment management to achieve the best balance between risk and return. The theory is that by making a reasonable allocation between different asset classes (e.g., stocks, bonds, cash, etc.), the risk of the portfolio can be minimized while obtaining a reasonable expected return. Asset allocation theory plays an important role in wealth management and investment strategies, guiding investors on how to adjust their portfolios in different market environments to achieve long-term financial goals. The literature shows that if the scope of asset allocation is extended from domestic to international, investors can diversify their risks and obtain higher returns, thus achieving the goal of effective outward expansion of the border (Levy and Sarnat, 1970). On the other hand, according to the theory of rational expectations, people make decisions based on their accurate and consistent expectations of how the economic situation will change in the future, which means that people make predictions based on all available information and experience, rather than based on random or irrational factors. For example, studies have shown that interaction between individual investors and the exchange of information between neighbors have a positive impact on the investment behavior of individuals and households (Shiller and Pound, 1986).
Therefore, from the asset allocation theory, it can be seen that commercial banks will allocate treasury bond futures in their portfolios to increase investment yields, and the increase in the price of treasury bond futures indicates that the number of suppliers of bank acceptance bills for rediscount will decrease, thereby reducing the interest rate of bank acceptance bills. The increase in the price of treasury bond futures reflects the sluggish market sentiment and forms a downward pressure on the macroeconomic sentiment index.
Based on the above two intrinsic influencing mechanisms, this paper proposes hypothesis 1: the price of treasury bond futures has the ability to discover the price of the discount rate of bank acceptance bills.
The bank's return on investment refers to the ratio of the bank's total return on a particular investment project or asset to the amount invested. This ratio reflects the profitability of the bank's asset allocation and investment decisions. By calculating the return on investment, banks can evaluate the investment benefits and provide an important reference for future investment decisions.
The existing literature points out that commercial banks can use treasury bond futures to increase the return on investment of assets. Treasury bond futures provide a new possibility for commercial banks' portfolio investment, and the bank's investment managers can optimize asset allocation and replace the treasury bonds originally planned to be purchased by purchasing treasury bond futures, while the high leverage brought about by the margin system of treasury bond futures trading enables them to have a larger amount of treasury bond positions with less capital, and more of the remaining funds can be used to invest in other financial bonds or corporate bonds with higher returns, so as to achieve higher investment returns with fixed capital (Zhou Bing and Chen Yanglong, 2013). With the gradual deepening of the gradual interest rate market-oriented reform since the mid-90s of the 20th century, although a series of supporting reform measures have effectively reduced financing costs and improved the efficiency of capital allocation, after China's economy has entered the new normal, the economic growth rate has slowed down significantly, the banking industry is facing the problem of intensified competition in the financial market, and the narrowing of deposit and loan spreads has led to the loss of banks' profit sources, and banks have to develop investment business with higher capital use efficiency to survive in an increasingly severe market competition environment (Jiang Hai et al., 2018)。 As mentioned above, in the case of fixed investment funds, commercial banks can provide more sufficient funds for investment in other high-yield products by holding a certain amount of treasury bond futures, thereby increasing the bank's investment yield. As a result, when the price of treasury bond futures rises, the market value of the treasury bond futures contracts previously purchased by commercial banks in their portfolios in order to increase investment yields increases, and there is no need to rediscount them to inject more funds into other commercial banks or discount institutions, and the number of suppliers of bank acceptance bills for rediscount decreases, and the rediscount rate of bank acceptance bills decreases when the number of demand side of bills remains unchanged.
Therefore, this paper proposes hypothesis 2: the ability of treasury bond futures price to find the discount rate of bank acceptance bills is mainly realized through the return on investment.
There are cyclical changes in the long-term operation of the economy, which can last for decades or even centuries. The macroeconomic prosperity index is an indicator to measure the degree of economic prosperity, which can be used to determine what stage the current economy is in the cycle of operation, whether it is an expansion stage or a contraction stage, and also predict the future economic development trend, whether it is rising, declining or basically stable. By observing the time lag relationship between the changes in the economic indicators and the fluctuations of the business cycle, the economic indicators can be divided into three categories: leading indicators, consistent indicators and lagging indicators. This article mainly discusses consistent indicators, which have an important role in reflecting the current state and trend of the economy, and can help us better understand the fundamental trend of the current economy.
The rise in Treasury futures prices reflects increased demand for Treasuries as safe-haven assets, with the macroeconomic sentiment index declining due to economic slowdown or increased uncertainty. Relevant studies show that the fluctuation of the rediscount rate is greatly affected by the macroeconomic situation, and this effect shows a certain cyclical nature in the long run. According to the above analysis, when the macroeconomic situation is unstable, market participants have pessimistic expectations for the future economic trend, and the demand for bills by enterprises decreases, resulting in a decrease in the acceptance and direct discount of bills, and the source of bills is relatively insufficient (Wang Banxing, 2010). In this case, the activity of rediscount transactions decreases, and the interest rate of bill rediscount shows a downward trend. On the contrary, the decline in the price of treasury bond futures contracts reflects the moderate growth of the macro economy, and people are full of confidence in the future macro economy, and the demand for bills by enterprises will be relatively strong, which has led to an increase in the acceptance and direct discount of bills, providing a relatively sufficient source of bills for the rediscount market. In this case, the rediscount transaction is relatively active, and the rediscount rate rises accordingly.
Therefore, this paper proposes hypothesis 3: the price discovery ability of treasury bond futures price to the discount rate of bank acceptance bills is mainly realized through economic expectations.
3. Research design
Data source
The data used in this paper, such as the bank acceptance bill discount rate, the daily opening price of treasury bond futures, and the 7-day pledged repo fixing rate, are from the Wind database and CSMAR database. Due to the successful listing of 2-year treasury bond futures on the China Financial Futures Exchange in August 2018, the time period of the sample is from December 5, 2018 to February 17, 2023, based on the availability of bank acceptance bill transfer discount rate data. The data such as the transfer discount rate of bank acceptance bills, the daily opening price of treasury bond futures, and the 7-day pledged repo fixing rate are all panel data of the trading day, with a total of 1046 trading days.
Variable selection
(1) Explanatory variables
The main explanatory variable in this paper is the discount rate of bank acceptance bills. In this paper, the discount rates for bank acceptance bills with a maturity of 3 months (R003), 6 months (R006) and 12 months (R0012) are used in this paper 1 , which is consistent with the practice of Zhang Xueying (2012). To a certain extent, the discount rate of bank acceptance bills can reflect the demand for credit and the delivery situation. When credit demand is insufficient and credit supply is weak, the rediscount rate tends to be lower. On the other hand, a higher rediscount rate may reflect better credit demand and delivery.
(2) Core explanatory variables
Consistent with the approach of Wang Jinzhong and Hu Xiaofan (2015) and Cong Yingnan et al. (2023), this paper selects the daily closing price (ClosePrice) and daily settlement price (SettlePrice) of the main two-year, five-year, and ten-year treasury bond futures contracts as the core explanatory variables. The Low Price is used as the core explanatory variable 2 .
(3) Control variables
In order to remove the exogenous influencing factors common to the explanatory variable and the explanatory variable, this paper selects the repo market interest rate (FR), the weekly effect variable ( ), the rise and fall of the Treasury bond futures contract measured by the closing price (Change1), the rise and fall of the Treasury bond futures contract measured by the settlement price (Change2), The six control variables 3 of the Treasury bond futures contract measured by the closing price (ChangeRatio1) and the Treasury bond futures contract measured by the settlement price (ChangeRatio2) are described as follows:
1. Repo market rate (FR). The repo rate in China's interbank bond market covers overnight to 6 months, of which the 7-day pledged repo rate is one of the benchmark interest rates reflecting the financing cost of market transactions, and it is also an important operational target of China's monetary policy.
2. Week-effect variable ( ). In order to eliminate this exogenous effect and obtain robust measurement results, this paper uses four dummy variables to represent Monday to Thursday: 1 when the trading day falls on Monday, and 0 at other times; 1 on Tuesday and 0 on other days; 1 when the trading day is on Wednesday and 0 at other times; 1 on Thursday and 0 on other days.
3. The change in the Treasury futures contract measured at the closing price (Change1) and the change in the Treasury futures contract measured at the settlement price (Change2). The calculation formula of Change1 is today's closing price - yesterday's settlement price, and the calculation formula of Change2 is today's settlement price - the previous day's settlement price, both of which reflect investors' profit expectations to a certain extent in the form of prices.
4. The change in the Treasury futures contract measured at the closing price (ChangeRatio1) and the change in the Treasury futures contract measured at the settlement price (ChangeRatio2). ChangeRatio1 is calculated from (Today's Close - Yesterday's Close)/Yesterday's Close and ChangeRatio2 is calculated from (Today's Close - Yesterday's Close)/Yesterday's Settlement Price, both of which also reflect market expectations to some extent as percentages, except that the prices used as benchmarks are different.
(5) Mechanism variables
1. Average return on investment (ROI). The quarterly investment return rate of each listed company in the monetary and financial services industry is derived from the market capitalization weighted average, which reflects the average rate of return obtained by each listed company in a specific quarter, and can more comprehensively reflect the investment performance of the entire industry. Changes in ROI can reflect the profitability and investment efficiency of different companies in the industry, which helps investors evaluate the overall investment potential of the industry.
2. China's Macroeconomic Sentiment Index Consensus Index (MSI). The monthly indicator is based on the results of a regular survey of entrepreneurs, which synthesizes entrepreneurs' assessments and expectations of their business operations and the macroeconomic environment. It can reveal the production and operation status of enterprises, as well as the overall performance of the economy, and help predict future trends of the economy. The composite index is the result of seasonally adjusting and re-normalizing individual indicators.
Table 1 describes the variable types, symbols, and definitions in this article.
Table 1 Variable definitions
Variable type
| 变量名 | Variable definitions
|
Explanatory variables
| R003 | The discount rate for the transfer of bank acceptance bills with a maturity of 3 months
|
| R006 | The discount rate for the transfer of bank acceptance bills with a maturity of 6 months
|
| R0012 | The rate at which the banker's acceptance bill is converted to a discount rate with a maturity of 12 months
|
Core explanatory variables
| OpenPrice | The daily opening price of the Treasury futures contract
|
| HighPrice | The daily high price of a Treasury futures contract
|
| LowPrice | The lowest price of the day for a Treasury futures contract
|
| ClosePrice | The daily closing price of the Treasury futures contract
|
| SettlePrice | The daily settlement price of the Treasury futures contract
|
Control variables
| FR | The repo market interest rate is represented by the 7-day interbank pledged repo weighted rate
|
|
| For the week-effect variable, 1 is taken for the trading day and 0 for the rest of the day
|
| Change1 | The rise and fall of a Treasury futures contract measured by its closing price
|
| Change2 | The rise and fall of a Treasury futures contract measured by the settlement price
|
| ChangeRatio1 | The rise and fall of a Treasury futures contract measured by its closing price
|
| ChangeRatio2 | The rise and fall of a Treasury futures contract measured at the settlement price
|
| CAR | Capital adequacy ratio of China's commercial banks
|
| RLD | Loan-to-deposit ratio of commercial banks in China, loan balance/deposit balance*100%
|
| RML | Medium- and long-term loan ratio, medium- and long-term loan balance of Chinese financial institutions/domestic deposit balance of Chinese financial institutions*100%
|
Mechanism variables
| ROI | The average return on investment in the monetary financial services sector weighted by market capitalization
|
| MSI | China's macroeconomic sentiment index is a consistent indicator
|
Descriptive statistics of variables
Table 2 provides descriptive statistics on the main variables, and it can be seen that the mean values of R003, R006 and R0012 are 2.329, 2.406 and 2.467 respectively, indicating that the average rediscount rates of bank acceptance bills with a maturity of 3 months, 6 months and 12 months during the sample period are 2.329%, 2.406% and 2.467%, with the mean increasing in turn, while the standard deviation decreases in turn, which is in line with the financial market law of higher yield and less volatility with a longer term. The mean values of OpenPrice, HighPrice, LowPrice, ClosePrice, and SettlePrice are 99.948 yuan, 100.023 yuan, 99.874 yuan, 99.948 yuan, and 99.950 yuan, respectively, and the difference between the average of each price is small, and the standard deviation is roughly the same, and the minimum value of each price of treasury bond futures is 95.170 yuan, and the maximum value is 105.200 yuan, which is within a reasonable range.
Table 2 Variable description statistics
变量 | 观测值 | 均值 | 标准差 | 最小值 | 最大值 |
Explanatory variables
| | | | | |
R003 | 9162 | 2.329 | 0.768 | 0.007 | 4.000 |
R006 | 9162 | 2.406 | 0.675 | 0.080 | 3.829 |
R0012 | 9045 | 2.467 | 0.627 | 0.834 | 3.591 |
Core explanatory variables
| | | | | |
OpenPrice | 9162 | 99.948 | 1.358 | 95.290 | 105.145 |
HighPrice | 9162 | 100.023 | 1.348 | 95.435 | 105.200 |
LowPrice | 9162 | 99.874 | 1.366 | 95.170 | 104.890 |
ClosePrice | 9162 | 99.948 | 1.356 | 95.225 | 105.195 |
SettlePrice | 9162 | 99.950 | 1.354 | 95.250 | 105.105 |
Control variables
| | | | | |
FR | 9153 | 2.314 | 0.483 | 1.353 | 5.922 |
Change1 | 9162 | 0.003 | 0.175 | -1.270 | 1.375 |
Change2 | 9162 | 0.005 | 0.144 | -0.920 | 1.420 |
ChangeRatio1 | 9162 | 0.000 | 0.001 | -0.011 | 0.013 |
ChangeRatio2 | 9162 | 0.000 | 0.001 | -0.009 | 0.014 |
CAR | 9162 | 14.62 | 0.333 | 14.12 | 15.17 |
RLD | 9162 | 76.42 | 2.278 | 72.22 | 79.69 |
RML | 9162 | 1.204 | 0.0310 | 1.145 | 1.269 |
Mechanism variables
| | | | | |
ROI | 3249 | 0.0268 | 0.027 | 0.007 | 0.125 |
MSI | 3249 | 98.687 | 6.791 | 86.369 | 116.522 |
Model settings
In this paper, we set up a two-way fixed-effect model as follows:
(1)
In Eq. (1), the subscript represents the Treasury futures contract with different maturities, and the observation date. is the explanatory variable, which indicates the discount rate of bank acceptance bills; Among the core explanatory variables, it is the daily opening price of Treasury bond futures; Among the control variables, the repo market interest rate factor is the weekly effect variable, which is the rise and fall of the Treasury bond futures contract measured by the closing price, the rise and fall of the Treasury bond futures contract measured by the settlement price, the rise and fall of the Treasury bond futures contract measured by the closing price, and the rise and fall of the Treasury bond futures contract measured by the settlement price. The coefficient of interest in this article is , which represents the average percentage change in the discount rate of bank acceptance bills caused by every one yuan change in the opening price of treasury bond futures.
Fourth, empirical results
(1) Benchmark regression results
In this paper, we use a two-way fixed-effect model to estimate the impact of treasury bond futures price on the rediscount rate of bank acceptance bills, and obtain that the price of treasury bond futures contracts has the ability to discover the price of bank acceptance bills to discount rates. Table 3 reports the benchmark regression results of the impact of Treasury futures prices on the rediscount rate of bank acceptance bills. Column (1) estimates the impact of the opening price of treasury bond futures on the discount rate of bank acceptance bills with a maturity of 3 months. The results show that the opening price of treasury bond futures increased by 1 yuan, and the discount rate of bank acceptance bills with a maturity of 3 months decreased by 0.279 percentage points on average, which was significant at the level of 1%. From the perspective of economic significance, the change in the three-month bank acceptance bill rediscount rate brought about by the increase in the price of treasury bond futures accounted for 12.0% (0.279/2.329) of the average three-month bank acceptance bill rediscount rate. In the real futures market, the average monthly value of the opening price of treasury bond futures increased by 0.0319 yuan on average, and the discount rate of bank acceptance bills decreased by 0.009 percentage points (0.0319*0.279) on average. Column (2) is the estimate with the control variable added, and the results are still robust. Column (3) replaces the explanatory variable with the 6-month bank acceptance bill discount rate, and the results show that at the 1% significance level, the daily opening price of treasury bond futures increases by 1 yuan, and the 6-month bank acceptance bill discount rate decreases by 0.262 percentage points on average. Column (5) shows the impact of the daily opening price of treasury bond futures on the rediscount rate of 12-month bank acceptance bills, indicating that the daily opening price of treasury bond futures increased by 1 yuan, and the rediscount rate of 12-month bank acceptance bills decreased by 0.252 percentage points on average, which was significant at the level of 1%.
Table 3 Impact of Treasury Bond Futures Price on the Transfer Discount Rate of Banker's Acceptance Bills: Benchmark Regression
| (1) | (2) | (3) | (4) | (5) | (6) |
| R003 | R003 | R006 | R006 | R0012 | R0012 |
OpenPrice | -0.279*** | -0.201*** | -0.262*** | -0.184*** | -0.252*** | -0.176*** |
| (0.003) | (0.004) | (0.003) | (0.004) | (0.002) | (0.003) |
FR | | 0.556*** | | 0.557*** | | 0.545*** |
| | (0.019) | | (0.017) | | (0.016) |
d1t | | -0.032* | | -0.027* | | -0.022 |
| | (0.018) | | (0.015) | | (0.014) |
d2t | | -0.061*** | | -0.057*** | | -0.051*** |
| | (0.017) | | (0.015) | | (0.014) |
d3t | | -0 .047*** | | -0.045*** | | -0.034** |
| | (0.017) | | (0.015) | | (0.014) |
d4t | | -0.030* | | -0.028* | | -0.024* |
| | (0.017) | | (0.015) | | (0.014) |
Change1 | | -0.095* | | -0.134*** | | -0.164*** |
| | (0.050) | | (0.045) | | (0.044) |
Change2 | | -1.947 | | -1.190 | | 0.021 |
| | (2.378) | | (2.070) | | (1.872) |
ChangeRatio1 | | -12.466 | | -5.772 | | -4.145 |
| | (9.637) | | (7.898) | | (7.275) |
ChangeRatio2 | | 207.490 | | 131.337 | | 15.464 |
| | (235.103) | | (204.731) | | (185.243) |
Constant | 30.218*** | 21.175*** | 28.597*** | 19.519*** | 27.678*** | 18.778*** |
| (0.302) | (0.479) | (0.255) | (0.399) | (0.223) | (0.352) |
| | | | | | |
Observations | 3,054 | 3,051 | 9,162 | 9,153 | 9,045 | 9,036 |
R-squared | 0.479 | 0.488 | 0.408 | 0.526 | 0.405 | 0.537 |
Bond FE | YES | YES | YES | YES | YES | YES |
Year FE | YES | YES | YES | YES | YES | YES |
Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the robust standard errors clustered to the Treasury bond futures level are in parentheses.
(2) The results of the robustness test
1. Endogenous discussion
Considering the possible inverse causal relationship between the price of treasury bond futures and the transfer discount rate of bank acceptance bills, the first-order lag term (OpenPrice_Lag1) of the daily opening price of treasury bond futures (OpenPrice) is used in this paper, which can weaken the reverse causal relationship to a certain extent. The results in Table 4 show that the price index of treasury bond futures with a lag period also has the ability to discover the price of the discount rate of bank acceptance bills, which is basically consistent with the analysis results of the benchmark regression model.
Table 4 Introducing the endogeneity test of the first-order hysteresis term
| (1) | (2) | (3) | (4) | (5) | (6) |
| R003 | R003 | R006 | R006 | R0012 | R0012 |
OpenPrice_Lag1 | -0.274*** | -0.262*** | -0.244*** | -0.234*** | -0.219*** | -0.211*** |
| (0.022) | (0.023) | (0.018) | (0.019) | (0.017) | (0.018) |
FR | | 0.152*** | | 0.140*** | | 0.111*** |
| | (0.024) | | (0.024) | | (0.016) |
d2t | | -0.098*** | | -0.065*** | | -0.039*** |
| | (0.018) | | (0.011) | | (0.007) |
d3t | | -0.077*** | | -0.056*** | | -0.037*** |
| | (0.015) | | (0.011) | | (0.008) |
d4t | | -0.041*** | | -0.032*** | | -0.027*** |
| | (0.011) | | (0.008) | | (0.006) |
Change1 | | -0.112 | | -0.047 | | 0.011 |
| | (0.144) | | (0.098) | | (0.065) |
Change2 | | -2.131 | | -0.599 | | -0.023 |
| | (2.390) | | (1.756) | | (1.179) |
ChangeRatio1 | | 8.793 | | 4.270 | | -2.432 |
| | (12.170) | | (8.783) | | (6.490) |
ChangeRatio2 | | 203.877 | | 46.833 | | -8.983 |
| | (237.397) | | (173.064) | | (115.446) |
Constant | 29.684*** | 28.202*** | 26.721*** | 25.436*** | 24.311*** | 23.355*** |
| (2.242) | (2.269) | (1.831) | (1.885) | (1.699) | (1.769) |
| | | | | | |
Observations | 2,499 | 2,490 | 2,499 | 2,490 | 2,499 | 2,490 |
R-squared | 0.834 | 0.842 | 0.878 | 0.887 | 0.921 | 0.928 |
Bond FE | YES | YES | YES | YES | YES | YES |
Year FE | YES | YES | YES | YES | YES | YES |
Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the robust standard errors clustered to the Treasury bond futures level are in parentheses.
2. Adjust the sample period
Since 2020, international public health crises represented by the new crown pneumonia epidemic have erupted from time to time, which has not only brought huge losses to the development of the real economy, but also seriously impacted the financial market. Compared with positive shocks, the financial sector is more vulnerable to spillovers from other industries under negative shocks, due to the unique industry characteristics of the financial industry itself, which makes itself more vulnerable due to its highly indebted operating model (Li et al., 2022). Therefore, under the negative impact of the new crown pneumonia epidemic, the impact of the price of treasury bond futures on the discount rate of bank acceptance bills may be different from that before the epidemic. In this paper, the sample period is adjusted, and the samples before the disaster impact (before January 1, 2020) are selected for regression to exclude the impact of the disaster impact on the price of treasury bond futures under all samples on the bank acceptance bill transfer discount rate, and the regression results are shown in Table 5.
Table 5 Robustness test for excluding disaster shocks
| (1) | (2) | (3) | (4) | (5) | (6) |
| R003 | R003 | R006 | R006 | R0012 | R0012 |
OpenPrice | -0.012*** | -0.010*** | -0.011*** | -0.009*** | -0.011*** | -0.009*** |
| (0.003) | (0.003) | (0.003) | (0.003) | (0.003) | (0.003) |
FR | | 0.162*** | | 0.170*** | | 0.152*** |
| | (0.013) | | (0.012) | | (0.011) |
d1t | | -0.037*** | | -0.034*** | | -0.023** |
| | (0.014) | | (0.012) | | (0.012) |
d2t | | -0.067*** | | -0.064*** | | -0.048*** |
| | (0.015) | | (0.014) | | (0.013) |
d3t | | -0.057*** | | -0.055*** | | -0.034*** |
| | (0.015) | | (0.014) | | (0.013) |
d4t | | -0.025* | | -0.024* | | -0.021* |
| | (0.014) | | (0.013) | | (0.012) |
Change1 | | -0.040** | | -0.049*** | | -0.055*** |
| | (0.019) | | (0.018) | | (0.019) |
Change2 | | 0.834 | | 0.209 | | -0.204 |
| | (2.142) | | (1.862) | | (1.802) |
ChangeRatio1 | | -0.065 | | 2.083 | | 1.763 |
| | (6.156) | | (6.008) | | (6.038) |
ChangeRatio2 | | -69.450 | | -10.385 | | 30.130 |
| | (211.392) | | (184.230) | | (178.657) |
Constant | 4.061*** | 3.437*** | 4.039*** | 3.374*** | 4.116*** | 3.498*** |
| (0.326) | (0.317) | (0.296) | (0.278) | (0.273) | (0.257) |
| | | | | | |
Observations | 2,358 | 2,358 | 2,358 | 2,358 | 2,358 | 2,358 |
R-squared | 0.674 | 0.702 | 0.692 | 0.727 | 0.704 | 0.732 |
Bond FE | YES | YES | YES | YES | YES | YES |
Year FE | YES | YES | YES | YES | YES | YES |
Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the robust standard errors clustered to the Treasury bond futures level are in parentheses.
Columns (1), (3) and (5) respectively estimate the impact of the opening price of treasury bond futures on the transfer discount rate of bank acceptance bills with a maturity of 3 months, 6 months and 12 months after excluding the epidemic, and the results show that the opening price of treasury bond futures increases by 1 yuan, and the transfer discount rates of bank acceptance bills with a maturity of 3 months, 6 months and 12 months decrease by 0.012, 0.011 and 0.011 percentage points on average, respectively, and the regression coefficient size has changed compared with the benchmark regression results, but the significance level has not changed. The direction and significance of the regression coefficients after adding the control variables in columns (2), (4) and (6) were the same as before, and the results were still robust.
3. Introduce new control variables
Considering the credit attributes of the bill market, the discount rate of bank acceptance bills will also be affected by the size of the credit. Under normal circumstances, when the credit scale is relatively sufficient, the commercial bank can expand its credit business as the discounter of the bank acceptance bill transfer discount business, so as to make full use of the credit scale, and the fund supply for bank acceptance bill discount in the rediscount market is relatively sufficient, and the transfer discount interest rate is relatively low; When the scale of credit is relatively tight, because the profits of commercial banks investing in general credit business are higher than those of bill assets, the funds that commercial banks are willing to invest in the business of rediscounting bank acceptance bills will be relatively reduced, and the rediscount interest rate is relatively high. The existing indicators for measuring the scale of credit mainly include the capital adequacy ratio, the deposit-loan ratio and the medium and long-term loan ratio. Therefore, on the basis of benchmark regression, new control variables affecting the banker's acceptance bill conversion discount rate, such as capital adequacy ratio (CAR), loan-to-deposit ratio (RLD), and medium and long-term loan ratio (RML), are introduced, and the regression results are shown in Table 6.
Table 6 Robustness test with the addition of new control variables
| (1) | (2) | (3) | (4) | (5) | (6) |
| R003 | R003 | R006 | R006 | R0012 | R0012 |
OpenPrice | -0.279*** | -0.077*** | -0.262*** | -0.069*** | -0.252*** | -0.060*** |
| (0.003) | (0.003) | (0.003) | (0.002) | (0.002) | (0.002) |
FR | | 0.084*** | | 0.126*** | | 0.117*** |
| | (0.014) | | (0.009) | | (0.006) |
d1t | | -0.029** | | -0.023** | | -0.010 |
| | (0.012) | | (0.009) | | (0.007) |
d2t | | -0.030** | | -0.026*** | | -0.013** |
| | (0.013) | | (0.009) | | (0.007) |
d3t | | -0.028** | | -0.027*** | | -0.014** |
| | (0.011) | | (0.009) | | (0.007) |
d4t | | -0.019 | | -0.016* | | -0.011 |
| | (0.013) | | (0.010) | | (0.007) |
Change1 | | 0.376*** | | 0.306*** | | 0.189*** |
| | (0.032) | | (0.024) | | (0.017) |
Change2 | | -1.361 | | -0.651 | | 0.517 |
| | (1.264) | | (0.986) | | (0.803) |
ChangeRatio1 | | -29.016*** | | -21.907*** | | -16.285*** |
| | (7.275) | | (5.322) | | (4.169) |
ChangeRatio2 | | 121.988 | | 52.585 | | -55.671 |
| | (125.352) | | (97.478) | | (79.461) |
CAR | | -1.538*** | | -1.497*** | | -1.443*** |
| | (0.043) | | (0.032) | | (0.022) |
RLD | | -0.054*** | | -0.039*** | | -0.060*** |
| | (0.008) | | (0.006) | | (0.004) |
RML | | 4.994*** | | 5.201*** | | 7.536*** |
| | (0.383) | | (0.295) | | (0.191) |
Constant | 30.216*** | 30.429*** | 28.595*** | 27.627*** | 27.665*** | 24.816*** |
| (0.309) | (0.574) | (0.262) | (0.416) | (0.235) | (0.275) |
| | | | | | |
Observations | 9,162 | 9,153 | 9,162 | 9,153 | 9,045 | 9,036 |
R-squared | 0.410 | 0.783 | 0.409 | 0.841 | 0.406 | 0.897 |
County FE | YES | YES | YES | YES | YES | YES |
Year FE | YES | YES | YES | YES | YES | YES |
Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the robust standard errors clustered to the Treasury bond futures level are in parentheses.
Columns (2), (4) and (6) estimate the impact of the opening price of treasury bond futures on the rediscount rate of bank acceptance bills with a maturity of 3 months, 6 months and 12 months, respectively, and the regression results show that the opening price of treasury bond futures increases by 1 yuan, and the rediscount rates of bank acceptance bills with a maturity of 3 months, 6 months and 12 months decrease by 0.077, 0.069 and 0.060 percentage points on average, respectively. Although the regression coefficient is lower than that of the benchmark regression after adding the control variables, it is still significant at the level of 1%, so the conclusion of this paper is robust.
4. Replace core explanatory variables
In the above study, OpenPrice is selected as the core explanatory variable in the study of the impact of treasury bond futures price on the transfer discount rate of bank acceptance bills. In this paper, the core explanatory variables are replaced with HighPrice, LowPrice, ClosePrice, and SettlePrice respectively for robustness checking, as shown in Table 7.
Table 7 Robustness test for substitution of core explanatory variables
Panel A: The highest price of the day
|
VARIABLES | R003 | R003 | R006 | R006 | R0012 | R0012 |
HighPrice | -0.283*** | -0.204*** | -0.266*** | -0.187*** | -0.256*** | -0.179*** |
| (0.003) | (0.004) | (0.003) | (0.004) | (0.002) | (0.003) |
FR | | 0.548*** | | 0.550*** | | 0.538*** |
| | (0.017) | | (0.014) | | (0.014) |
d1t | | -0.037** | | -0.031** | | -0.024* |
| | (0.017) | | (0.014) | | (0.014) |
d2t | | -0.065*** | | -0.059*** | | -0.052*** |
| | (0.017) | | (0.015) | | (0.014) |
d3t | | -0.050*** | | -0.048*** | | -0.036** |
| | (0.018) | | (0.016) | | (0.015) |
d4t | | -0.034* | | -0.030* | | -0.026* |
| | (0.018) | | (0.016) | | (0.015) |
Change1 | | -0.082* | | -0.120*** | | -0.151*** |
| | (0.046) | | (0.042) | | (0.042) |
Change2 | | -1.558 | | -0.790 | | 0.288 |
| | (2.359) | | (2.114) | | (1.923) |
ChangeRatio1 | | -4.568 | | 0.860 | | 2.459 |
| | (11.553) | | (9.877) | | (9.373) |
ChangeRatio2 | | 166.281 | | 89.749 | | -13.059 |
| | (231.549) | | (207.689) | | (189.115) |
Constant | 30.645*** | 21.543*** | 29.008*** | 19.861*** | 28.086*** | 19.134*** |
| (0.316) | (0.471) | (0.269) | (0.395) | (0.240) | (0.354) |
Observations | 9,162 | 9,153 | 9,162 | 9,153 | 9,045 | 9,036 |
R-squared | 0.413 | 0.500 | 0.413 | 0.527 | 0.411 | 0.538 |
County/Year FE | YES | YES | YES | YES | YES | YES |
Panel B: The lowest price of the day
|
LowPrice | -0.275*** | -0.198*** | -0.258*** | -0.181*** | -0.247*** | -0.172*** |
| (0.003) | (0.004) | (0.003) | (0.004) | (0.002) | (0.003) |
FR | | 0.560*** | | 0.562*** | | 0.550*** |
| | (0.017) | | (0.014) | | (0.014) |
d1t | | -0.040** | | -0.033** | | -0.026* |
| | (0.017) | | (0.014) | | (0.014) |
d2t | | -0.066*** | | -0.060*** | | -0.053*** |
| | (0.017) | | (0.015) | | (0.014) |
d3t | | -0.053*** | | -0.050*** | | -0.038** |
| | (0.018) | | (0.016) | | (0.015) |
d4t | | -0.033* | | -0.030* | | -0.025* |
| | (0.018) | | (0.016) | | (0.015) |
Change1 | | -0.096** | | -0.133*** | | -0.165*** |
| | (0.048) | | (0.044) | | (0.044) |
Change2 | | -0.831 | | -0.130 | | 0.907 |
| | (2.260) | | (2.023) | | (1.838) |
ChangeRatio1 | | -5.297 | | 0.208 | | 1.968 |
| | (11.724) | | (10.041) | | (9.563) |
ChangeRatio2 | | 96.449 | | 26.326 | | -72.515 |
| | (221.733) | | (198.688) | | (180.587) |
Constant | 29.747*** | 20.820*** | 28.135*** | 19.177*** | 27.170*** | 18.390*** |
| (0.302) | (0.448) | (0.256) | (0.375) | (0.230) | (0.339) |
Observations | 9,162 | 9,153 | 9,162 | 9,153 | 9,045 | 9,036 |
R-squared | 0.405 | 0.498 | 0.404 | 0.524 | 0.399 | 0.534 |
County/Year FE | YES | YES | YES | YES | YES | YES |
Panel C: Daily closing price
|
ClosePrice | -0.279*** | -0.202*** | -0.262*** | -0.184*** | -0.252*** | -0.176*** |
| (0.003) | (0.004) | (0.003) | (0.004) | (0.002) | (0.003) |
FR | | 0.554*** | | 0.556*** | | 0.544*** |
| | (0.017) | | (0.014) | | (0.014) |
d1t | | -0.039** | | -0.032** | | -0.025* |
| | (0.017) | | (0.014) | | (0.014) |
d2t | | -0.065*** | | -0.059*** | | -0.052*** |
| | (0.017) | | (0.015) | | (0.014) |
d3t | | -0.052*** | | -0.049*** | | -0.037** |
| | (0.018) | | (0.016) | | (0.015) |
d4t | | -0.034* | | -0.031* | | -0.026* |
| | (0.018) | | (0.016) | | (0.015) |
Change1 | | -0.079* | | -0.118*** | | -0.150*** |
| | (0.047) | | (0.043) | | (0.043) |
Change2 | | -1.210 | | -0.473 | | 0.587 |
| | (2.292) | | (2.052) | | (1.865) |
ChangeRatio1 | | 4.192 | | 8.871 | | 10.150 |
| | (11.629) | | (9.960) | | (9.463) |
ChangeRatio2 | | 129.077 | | 55.812 | | -44.985 |
| | (224.977) | | (201.644) | | (183.392) |
Constant | 30.198*** | 21.225*** | 28.572*** | 19.565*** | 27.613*** | 18.800*** |
| (0.307) | (0.458) | (0.261) | (0.384) | (0.234) | (0.346) |
Observations | 9,162 | 9,153 | 9,162 | 9,153 | 9,045 | 9,036 |
R-squared | 0.409 | 0.499 | 0.408 | 0.526 | 0.404 | 0.536 |
County/Year FE | YES | YES | YES | YES | YES | YES |
Panel D: Daily settlement price
|
SettlePrice | -0.278*** | -0.202*** | -0.261*** | -0.184*** | -0.251*** | -0.176*** |
| (0.003) | (0.004) | (0.003) | (0.004) | (0.002) | (0.003) |
FR | | 0.554*** | | 0.556*** | | 0.544*** |
| | (0.017) | | (0.014) | | (0.014) |
d1t | | -0.039** | | -0.032** | | -0.025* |
| | (0.017) | | (0.014) | | (0.014) |
d2t | | -0.065*** | | -0.059*** | | -0.052*** |
| | (0.017) | | (0.015) | | (0.014) |
d3t | | -0.052*** | | -0.049*** | | -0.037** |
| | (0.018) | | (0.016) | | (0.015) |
d4t | | -0.034* | | -0.031* | | -0.026* |
| | (0.018) | | (0.016) | | (0.015) |
Change1 | | -0.281*** | | -0.302*** | | -0.325*** |
| | (0.046) | | (0.042) | | (0.042) |
Change2 | | -1.008 | | -0.288 | | 0.763 |
| | (2.292) | | (2.052) | | (1.865) |
ChangeRatio1 | | 4.192 | | 8.871 | | 10.150 |
| | (11.629) | | (9.960) | | (9.463) |
ChangeRatio2 | | 129.077 | | 55.812 | | -44.985 |
| | (224.977) | | (201.644) | | (183.392) |
Constant | 30.158*** | 21.225*** | 28.520*** | 19.565*** | 27.547*** | 18.800*** |
| (0.306) | (0.458) | (0.260) | (0.384) | (0.233) | (0.346) |
Observations | 9,162 | 9,153 | 9,162 | 9,153 | 9,045 | 9,036 |
R-squared | 0.408 | 0.499 | 0.406 | 0.526 | 0.402 | 0.536 |
County/Year FE | YES | YES | YES | YES | YES | YES |
Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the robust standard errors clustered to the Treasury bond futures level are in parentheses.
As can be seen from Table 7, the regression results of the robustness test are basically consistent with the benchmark regression, that is, after replacing the core explanatory variables, the causal relationship between the price of the treasury bond futures contract and the bank acceptance bill transfer discount rate is still valid, and the sign and significance level of the regression coefficient are roughly in line with expectations. Therefore, the conclusions of this paper are consistent.
(3) Heterogeneity analysis results
The subject matter of the two-year, five-year, and ten-year treasury bond futures contracts is different, namely notional short-term treasury bonds with a face value of RMB 2 million, notional medium-term treasury bonds with a face value of RMB 1 million, and notional long-term treasury bonds with a face value of RMB 1 million. Therefore, there may be heterogeneity in the impact of the price of treasury bond futures contracts of different maturities on the discount rate of bank acceptance bills. In this paper, the heterogeneity analysis of the interaction terms is carried out according to the different maturities of the treasury bond futures contract, and the regression results are shown in Table 8.
Table 8 Heterogeneity analysis of Treasury bond futures contract duration
| (1) | (2) | (3) | (4) | (5) | (6) |
| R003 | R003 | R006 | R006 | R0012 | R0012 |
OpenPrice*term | -0.540*** | -0.403*** | -0.516*** | -0.374*** | -0.503*** | -0.364*** |
| (0.022) | (0.022) | (0.019) | (0.020) | (0.017) | (0.018) |
OpenPrice | -0.284*** | -0.220*** | -0.266*** | -0.201*** | -0.255*** | -0.191*** |
| (0.004) | (0.005) | (0.003) | (0.004) | (0.003) | (0.004) |
term | 54.608*** | 40.712*** | 52.091*** | 37.822*** | 50.824*** | 36.755*** |
| (2.183) | (2.232) | (1.909) | (1.963) | (1.692) | (1.761) |
FR | | 0.458*** | | 0.466*** | | 0.457*** |
| | (0.020) | | (0.017) | | (0.015) |
d1t | | -0.034* | | -0.028* | | -0.022 |
| | (0.018) | | (0.015) | | (0.014) |
d2t | | -0.056*** | | -0.052*** | | -0.046*** |
| | (0.018) | | (0.015) | | (0.014) |
d3t | | -0.039** | | -0.038** | | -0.028* |
| | (0.019) | | (0.016) | | (0.014) |
d4t | | -0.025 | | -0.024* | | -0.021 |
| | (0.017) | | (0.014) | | (0.014) |
Change1 | | 0.037 | | -0.014 | | -0.045 |
| | (0.045) | | (0.042) | | (0.041) |
Change2 | | -3.318 | | -2.380 | | -1.241 |
| | (2.443) | | (2.098) | | (1.835) |
ChangeRatio1 | | -15.610 | | -8.297 | | -6.377 |
| | (10.865) | | (8.910) | | (8.118) |
ChangeRatio2 | | 334.869 | | 241.277 | | 132.253 |
| | (241.809) | | (207.830) | | (181.706) |
Constant | 30.585*** | 23.191*** | 28.902*** | 21.350*** | 27.920*** | 20.494*** |
| (0.384) | (0.548) | (0.323) | (0.455) | (0.279) | (0.388) |
| | | | | | |
Observations | 9,162 | 9,153 | 9,162 | 9,153 | 9,045 | 9,036 |
R-squared | 0.479 | 0.537 | 0.490 | 0.567 | 0.496 | 0.582 |
Bond FE | YES | YES | YES | YES | YES | YES |
Year FE | YES | YES | YES | YES | YES | YES |
Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the robust standard errors clustered to the Treasury bond futures level are in parentheses.
The heterogeneity analysis of the interaction terms of the two-year, five-year, and ten-year treasury bond futures contracts shows that when the treasury bond futures contract is a two-year contract, the grouping variable value is 1, which represents the short-term treasury bond futures contract, and when the treasury bond futures contract is a five-year or ten-year contract, the value is 0, which represents the long-term treasury bond futures contract. The empirical test results in columns (1), (3) and (5) of Table 4 show that for every 1 yuan increase in the price of short-term treasury bond futures contracts, the average percentage reduction of the transfer discount rate of bank acceptance bills with a maturity of 3 months, 6 months and 12 months is 0.540, 0.516 and 0.503 percentage points higher than that of long-term treasury bond futures contracts, which is significantly higher than that of long-term treasury bond futures contracts, which is significantly at the level of 1%. This indicates that the change in the price of treasury bond futures with a shorter maturity has a greater impact on the discount rate of banker's acceptance bills than that with a longer maturity, and the results are heterogeneous. However, the price of treasury bond futures still has the ability to detect the price of the banker's acceptance bill to the discount rate, which indicates that the conclusions of this paper are robust.
5. Mechanism analysis
The benchmark regression obtains that the price of treasury bond futures has the ability to discover the price of the bank acceptance bill to the discount rate. However, we are not yet clear about the mechanism by which the price of treasury bond futures affects the rate of rediscount on bank acceptance bills. This paper discusses the influencing mechanism, i.e., the return on investment mechanism and the macroeconomic mechanism, from the perspectives of portfolio theory and rational expectation theory.
The investment yield mechanism proposes a direction in which the price of treasury bond futures affects the discount rate of bank acceptance bills. According to the second hypothesis of the theoretical hypothesis, commercial banks holding a certain amount of treasury bond futures can free up sufficient funds through the high leverage brought by their margin system to other products with higher returns, thereby increasing the return on investment. Therefore, when the market price of the Chinese bond futures contract in the portfolio of commercial banks increases, the bank's investment yield increases, and there is no need to rediscount to inject more funds, the supply of bank acceptance bills by commercial banks decreases, and the rediscount rate of bank acceptance bills decreases under the condition that the demand for bills remains unchanged.
The other mechanism of influence is from the perspective of macroeconomic operation. From the third hypothesis of the theoretical hypothesis, it can be seen that when the price of treasury bond futures rises, it reflects the decline of the macroeconomic sentiment index due to the economic slowdown or the increase in uncertainty, the negative sentiment of the market, the relatively small trading volume in the market, the small demand for enterprises to open bills, and the insufficient supply of bank acceptance bill assets in the bill market, which promotes the decline in the discount rate of bank acceptance bills.
This paper attempts to examine the impact of treasury bond futures prices on the investment yield mechanism and macroeconomic mechanism.
Table 9 reports on the impact of Treasury futures prices on the investment yield mechanism and the macroeconomic mechanism. As can be seen from columns (1) and (2) of Table 9, OpenPrice's estimation coefficient is significantly positive regardless of whether the control variables are added, indicating that the increase in the price of treasury bond futures contracts can increase the bank's investment yield, and then pull down the banker's acceptance bill transfer discount rate, and hypothesis 2 is verified. As can be seen from columns (3) and (4) of Table 9, the estimation coefficient of OpenPrice is significantly negative at the level of 1%, regardless of whether the control variables are added, indicating that the increase in the daily opening price of treasury bond futures indicates a decrease in the macroeconomic sentiment index, thereby promoting the downward trend of the discount rate of bank acceptance bills, and hypothesis 3 is verified. The theoretical way of the increase in the opening price of treasury bond futures on a daily basis leads to a decrease in the macroeconomic sentiment index is as follows: the increase in the opening price of treasury bond futures on a daily basis reflects the market's concern about the future economic situation, and investors transfer funds to relatively safe assets such as treasury bonds, resulting in a decline in the price of other assets such as the stock market, thereby affecting the macroeconomic sentiment index.
Table 9 The impact of treasury bond futures prices on the rediscount rate of bank acceptance bills: mechanism analysis
| (1) | (2) | (3) | (4) |
| ROI | MSI |
OpenPrice | 0.003*** | 0.003** | -2.077*** | -1.198*** |
| (0.000) | (0.000) | -0.315 | -0.213 |
FR | | -0.008 | | 5.755*** |
| | (0.003) | | -1.17 |
d1t | | -0.011** | | 1.318 |
| | (0.003) | | -1.153 |
d2t | | -0.019* | | -7.567*** |
| | (0.005) | | -0.629 |
d3t | | -0.006* | | 2.283** |
| | (0.002) | | -0.789 |
d4t | | -0.016* | | -1.764 |
| | (0.005) | | -1.476 |
Change1 | | -0.009 | | -0.412 |
| | (0.013) | | -5.244 |
Change2 | | 1.631** | | 274.635* |
| | (0.348) | | -118.949 |
ChangeRatio1 | | 3.583 | | 3,251.305*** |
| | (2.195) | | -612.173 |
ChangeRatio2 | | -161.824** | | -31,204.034** |
| | (34.576) | | -11,508.72 |
Constant | -0.300*** | -0.267** | 306.588*** | 206.042*** |
| (0.026) | (0.037) | -31.535 | -21.691 |
| | | | |
Observations | 153 | 153 | 297 | 297 |
R-squared | 0.412 | 0.510 | 0.255 | 0.532 |
Bond FE | YES | YES | YES | YES |
Year FE | YES | YES | YES | YES |
Note: *, **, and *** represent the significance levels of 10%, 5%, and 1%, respectively, and the robust standard errors clustered to the Treasury bond futures level are in parentheses.
6. Conclusions and Implications
This paper empirically examines the impact of treasury bond futures prices on the rediscount rate of bank acceptance bills by collecting panel data on the trading day frequencies of bank acceptance bill transfer discount rate, treasury bond futures daily opening price, and 7-day pledged repo fixing rate between December 5, 2018 and February 17, 2023, and empirically tests the impact of treasury bond futures price on bank acceptance bill transfer discount rate by using a two-way fixed-effect model. The conclusion is still valid after the robustness test of adding new control variables and replacing the core explanatory variables. (2) As important intermediary variables, the investment yield and macroeconomic sentiment index are important transmission mechanisms for the price discovery ability of treasury bond futures prices to the rediscount rate of bank acceptance bills. (3) The impact of treasury bond futures price on the rediscount rate of bank acceptance bills is heterogeneous, and compared with the longer-term treasury bond futures contracts, the change of the short-term treasury bond futures price has a more significant impact on the rediscount rate of banker's acceptance bills.
Based on the above analysis and conclusions, this paper puts forward the following policy recommendations:
First, strengthen market supervision. In view of the ability of treasury bond futures prices to discover the price of bank acceptance bills to discount rates, the regulator should strengthen the supervision of the treasury bond futures market. This includes improving market transparency, ensuring the fairness and effectiveness of transactions, and preventing market manipulation and insider trading. Regulators should also regularly release Treasury futures market data and analysis reports to help investors better understand the dynamics of the Treasury futures market and the interbank market.
Second, we need to optimize macroeconomic policies. Investment yield and macroeconomic sentiment index are important intermediary variables for the price of treasury bond futures to affect the discount rate of bank acceptance bills, and the government should take these factors into account when formulating macroeconomic policies. For example, the government can affect the rate of return on investment by adjusting fiscal and monetary policies, or the macroeconomic sentiment index through economic stimulus or austerity measures. In this way, the signals of the treasury bond futures market can be used more effectively to promote the steady growth of the economy.
Third, refine deadline management. There is maturity heterogeneity in the impact of treasury bond futures price on the discount rate of bank acceptance bills. Relevant financial institutions should take into account the impact of changes in the price of treasury bond futures of different maturities when formulating interest rate policies. For example, different risk management strategies and interest rate policies can be set for Treasury futures of different maturities to more accurately reflect market conditions and risk levels.
Fourth, raise the risk awareness of market participants. Market participants should raise awareness of the risks associated with changes in the price of Treasury futures. Regulators and financial institutions can raise investors' awareness of risk through educational programs and public information campaigns. In addition, financial institutions should provide appropriate risk management tools and advisory services to help investors conduct effective risk assessment and management. Financial market entities should strengthen their awareness of risk management, pay attention to market monitoring and information disclosure, and avoid risk exposure caused by fluctuations in the price of treasury bond futures.
References:
CHEN Han. China Finance,2014(4):83-85.
Cong Yingnan,Liu Yixin,Yang Dasen. Research on the interdependence structure of China's financial markets based on Copula method[J].Journal of Financial Economics,2023,38(2):51-65.)
HE Ping,LIU Zehao,FAN Zhongjie. An empirical test of the impact of treasury bond futures trading on the volatility of interest rate market[J].Journal of Tsinghua University(Natural Science Edition),2017,57(5):544-549.)
HU Yuyue,GUO Chenkai,CAO Feilong. China Finance,2012,(8):62-63.
JIANG Hai,ZHANG Xiaolin,CHEN Chuanglian. Capital buffer behavior of commercial banks in the process of interest rate liberalization[J].China Industrial Economics,2018(11):61-78.)
Li Zheng, Shi Qing, Wen Bohui, et al. Finance & Trade Economics,2022,43(9):53-68.)
Liu Mingkang, Huang Jia, Jun Jun. Bank interest rate decision and internal fund transfer pricing: Experience from China's interest rate market-oriented reform[J].Economic Research Journal,2018,53(6):4-20.)
QIANG Jing,HOU Xin,FAN Longzhen. The formation mechanism of benchmark interest rate, expected inflation rate and market interest rate term structure[J].Economic Research Journal,2018,53(4):92-107.)
WANG Banxing. Market-oriented Transformation of Discount Interest Rate of Commercial Bank Bills: Research on Pricing Model Based on Shibor[J].Financial Forum,2010,15(12):24-29.)
WANG Jinzhong,HU Xiaofan. Research on the market effectiveness of Chinese government bond futures[J].Economic Review,2015,(6):55-68.)
Wu Xiaoqiu,Ying Zhanyu. Some issues on the re-establishment of treasury bond futures[J].Finance & Trade Economics,2003(10):35-42+96.
ZENG Yun,HUO Da,YUAN Shaofeng. Does Treasury Bond Futures Promote the Transmission of Monetary Policy Interest Rates?——Based on the Perspective of the Linkage of Treasury Bond Futures, Spot and Repo Market[J].Financial Review,2019,11(6):98-108+123.)
ZHANG Jinfan, TANG Yingwei, GANG Jianhua, et al. Price discovery in China's interest rate market: A study on the spot, futures and interest rate swap markets of treasury bonds[J].Journal of Financial Research,2019(1):19-34.)
ZHANG Maojun,GUO Mengfei,LI Hao. Research on cross-market behavior of information transmission: Based on the spillover effect of treasury bond futures and spot[J].Finance and Economics,2019(1):20-27.)
ZHANG Xueying. Research on the impact of reserve requirement ratio adjustment on market interest rate[J].Journal of Quantitative and Technical Economics,2012,29(12):136-146.)
ZHANG Xueying,HE Feiping. The impact of central bank repo operation on money market interest rate: Theoretical model and empirical test[J].Journal of Financial Research,2014(3):41-53.)
ZHANG Zongxin,ZHANG Xiuxiu. Can the introduction of treasury bond futures contract play a stabilizing effect in the spot market?—— based on the research perspective of China's financial cycle[J].Journal of Financial Research,2019(6):58-75.)
ZHOU Bing,CHEN Yanglong. Research on the core function of treasury bond futures and empirical test: Based on the observation of simulated trading of treasury bond futures in China[J].Journal of Financial Research,2013(04):24-28.)
ZHU Lirong,SU Xin,ZHOU Yong. Research on intertemporal arbitrage based on China's futures market[J].Operations Research and Management,2015,24(3):179-188.)
Chari, V. V., Jagannathan, R., & Jones, L. (1990). Price stability and futures trading in commodities. The Quarterly Journal of Economics, 105(2), 527-534.
Engle, R. (1998). Macroeconomic Announcements and Volatility of Treasury Futures. Department of Economics, UC San Diego.
Harvey, N. (1996). The market for futures contracts on Canadian bankers' acceptances. Bank of Canada Review, 1996, 19-36.
Johnston, E. T., Kracaw, W. A., & McConnell, J. J. (1991). Day-of-the-week effects in financial futures: An analysis of GNMA, T-bond, T-note, and T-bill contracts. Journal of Financial and Quantitative Analysis, 26(1), 23-44.
Kacperczyk, M., & Schnabl, P. (2010). When safe proved risky: commercial paper during the financial crisis of 2007–2009. Journal of Economic Perspectives, 24(1), 29-50.
Levy, H., & Sarnat, M. (1970). International diversification of investment portfolios. The American Economic Review, 60(4), 668-675.
Miyanoya, A., Inoue, H., & Higo, H. (1999). Microstructure and liquidity in the Japanese government securities market (No. 99-1). Bank of Japan Working Paper.
Retkwa, R. (1982). Why foreign borrowers love America's commercial paper market. Euromoney, 6, 151-157.
Shiller, R. J., & Pound, J. (1986). Survey Evidence on Diffusion of Investment Among Institutional Investors. NBER Working Paper, (w1851).