Economic Effect of China’s Rural Financial Market Growth during 1952-2013

,

College of Economics and Management, Southwest University, Chongqing 400715, China

1 Introduction

Since New China was founded 60 years ago, China’s rural financial market has developed by leaps and bounds, and especially after the reform and opening up in 1978, the growth is significantly faster. From the market growth scale, China’s rural financial supply can still far from meet the demand, but it shows a fast growing trend. On the surface, the growth scale of rural financial market continues to expand, and it is bound to improve the shortage of funds in rural areas, thus promoting the development of rural economy and increasing farmers’ income. But in fact, this conclusion requires strict conditions, that is, rural financial market can efficiently allocate financial resources. The so-called effectiveness of financial resource allocation means that if the financial resources can be reallocated to decrease social welfare, then the allocation state of financial resources is effective before reallocation (Besley and Coate, 1994). Under the premise of effective allocation, financial resources can be allocated to the farmers or rural enterprises with the best investment opportunities, in order to truly promote the production and investment, and promote rural economic development. Under the premise of ineffective allocation, even though the rural financial market scale continues to grow, the financial resources may fail to have a positive impact on output and income due to abuse, and even may mislead farmers and enterprises to adopt inappropriate capital-intensive technology, resulting in the deterioration of the rural economy[3]. A lot of literature provides some reference for us to research the economic effect of rural financial market growth, but there are still some drawbacks to be redressed. First of all, many scholars have realized that there is something wrong with China’s rural financial market efficiency, but it lacks rigorous theoretical explanation and empirical research to support. Secondly, although some scholars, such as Yao Yaojun (2004)[2], Wen Taoetal. (2004, 2005)[1], Zhu Xi, Li Zinai (2006)[3], Ran Guanghe, Li Jing and Wen Tao (2008)[1], have conducted the empirical studies, their analysis period is 1952-2002 or 1978-2006, and the selection of empirical indicators also need to be improved. In fact, since the 1980s, with the rise of township enterprises and expansion of farmers’ autonomy in management, the loans for township enterprises and farmers have continued to increase, becoming an important part of rural loans, and if it is ignored, there will be misleading results to underestimate the real economic effects of rural loans. In addition, it is meaningless to some extent in the existing studies which use the ratio of township enterprises’ loans to rural loans as a rural financial structure indicator. The difficulty in getting loans for farmers is one of the most prominent problems restricting rural economic development at present, while it is easier for township enterprises to get loans, so increasing the availability of loans to farmers will do good to the rural economy. For these reasons, this paper not only sorts out the theories and improves the selection of empirical indicators and sample period, but also uses VAR model and cointegration analysis developed by Johansen (1988, 1991)[6-7]for empirical test of the economic effect of China’s rural financial market growth.

2 Empirical design

In order to avoid the "spurious regression" phenomenon in model, we first use ADF unit root test method developed by Dickey & Fuller (1981) to test the stationarity of variable, and conduct differential treatment on the non-stationary variable to make it become stationary time series. Cointegration theory is an important method to analyze non-stationary time series. Engle and Granger (1987)[5]point out that if two or more series are individually integrated (in the time series sense) but some linear combination of them has a lower order of integration, then the series are said to be cointegrated. In this paper, we will use the cointegration test method developed by Johansen (1988, 1991)[6-7]to test the cointegration relationship between variables. After obtaining the results of cointegration test, if there is cointegration relationship, we will use vector error correction model (VECM) to reveal the short-term dynamic relationship between these variables; if there is no cointegration relationship, we will use Granger causality test to judge whether there is causal relationship between these variables. The effectiveness of Granger causality test depends on the determination of optimal lag. If the lag is determined at random, it will lead to erroneous test results. In this study, we use Eviews software for operation of the empirical results. The variables and data involved mainly include rural economic indicators and rural financial market indicators during 1952-2013. Rural economic indicators include rural GDP, total rural population and rural fixed assets investment. Rural financial market indicators include balance of deposits in rural areas, and balance of loans in rural areas. The demographic factors can not be ignored in the analysis of rural economic development, and based on the above total quantity indicators, we employ the rural average of relevant indicators in actual analysis. Rural per capita GDP is obtained after we divide total rural GDP by average rural population in different years.

3 Empirical test results and analysis

3.1UnitroottestIn this study, we use ADF unit root test method to test each variable in order to determine the stationarity of variable. The lag order is selected in accordance with AIC. We first take the logarithm of rural per capita GDP, denoted byRGDP. It is found from test thatRGDP,KL,FL,FEandFCare all non-stationary variables. We conduct differential treatment on the non-stationary variable, and the results are shown in Table 1.RGDP,KL,FL,FEandFCdenote the first-order differential value of the relevant variables, respectively. As can be seen from Table 1, all the series after processing are stationary at the 1% significance level, soRGDP,KL,FL,FEandFCare all integrated of order 1.

Table1ADFtestresultsofvariables

VariableTesttype(C,T)LagorderTestvalueCriticalvalueAICvalueRGDPC,T1-1.8841151%(-4.1314)-2.298320△RGDPC,T3-4.3450081%(-4.1420)-2.224682KLC,T4-3.9768671%(-4.1420)-0.450856△KL0,01-3.9843251%(-2.6055)-0.236048FL0,012.8089871%(-2.6048)-1.964813△FLC,T1-5.4060711%(-4.1348)-1.985357FEC,02-2.0445381%(-3.5547)-0.811384△FE0,04-3.2340881%(-2.6081)-1.012154FCC,05-1.8468761%(-3.5625)-4.743479△FC0,06-2.6530701%(-2.6100)-4.818395

Note: C and T represent constant term and trend term, respectively.

3.2CointegrationtestThe above variables are all integrated of order 1, and it means that there may be one kind of stationary linear combination for these variables, thereby reflecting that there may be a long-term stable cointegration relationship between variables. Therefore, we can use Johansen test to judge whether there is cointegration relationship between them, and further determine the symbol relationship between the relevant variables. Johansen cointegration test is a test method based on vector autoregression (VAR) model, and we must determine the structure of VAR model at first before performing test. From the unit root test, it is found that amongRGDP,KL,FL,FEandFCtime series, two have constant term and linear trend term, and accordingly, cointegration equation also contains constant term and trend term[4, 8], so we choose the cointegration equation containing constant term and trend term for testing. In order to maintain a reasonable degree of freedom and eliminate autocorrelation of the error term, the optimal lag of VAR model is 2 according to AIC, SC, LR and Q statistic, and the residual series are stationary. Based on the optimal VAR model, the Johansen cointegration test results obtained are shown in Table 2. From Table 2, it can be found that during 1952-2013, there was only a cointegration relationship among five variables (RGDP,KL,FL,FEandFC) at the 5% significance level. According to VECM, we get the equilibrium vector as follows:

ξ’= (1.000，0.618，-1.414，0.718，-1.291)

Then the cointegration equation between the variables is as follows:

RGDP=-0.618KL+1.414FL-0.718FE+1.291FC+4.724

(0.06603)(0.18691)(0.13468) (0.00771)

[9.36767] [-7.56772][-5.33524] [-0.49191]

The cointegration equation shows that during 1952-2013, there was a long-run equilibrium relationship among the above five variables. It can be found that during 1952-2013, there was a negative relationship between rural per capita GDP and rural capital formation rate (KL) or rural financial market growth efficiency (FE), while there was a positive relationship between rural per capita GDP and rural financial market growth scale (FL) and rural financial market growth structure (FC). This shows that during 1952-2008, China’s rural financial market growth scale and farmer-biased credit structure of financial markets helped to promote rural economic growth in general. This finding proves that the capital distribution of China’s rural financial market was effective in general during 1952-2013. Such effectiveness can be attributed to the fact that the rural economic entities, such as the farmers who have independent management rights and enterprises, improved efficiency in the use of funds. With the development of Chinese rural economy, rural capital formation (KL) and rural financial market efficiency (FE) do not become favorable factors for the rural economic growth, possibly because Chinese rural capital formation is mainly exogenous. Since 1952, the rural fixed assets investment has been basically dominated by the government, with serious blind and repeated construction. Especially since the 1990s, more and more farmers have become migrant workers, and farmers’ enthusiasm for agricultural production and investment in agriculture has sharply dropped, resulting in low rural fixed assets investment rate and rural capital formation rate, so that rural capital formation does not become an endogenous factor to drive rural economic growth. At the same time, the rural financial market efficiency (FE) (namely rural loan-to-deposit ratio) was always in the process of decline during 1952-2013. Before the reform and opening up, due to the state’s top priority to heavy and chemical industry, finance was invested with "transfer" functions, and considerable financial resources were transferred from countryside to cities to support industry, resulting in a significant decline in rural loans. Since the rural financial market reform in the mid-1990s, the rural financial institutions with significantly enhanced profit-seeking motive, have lent credit funds to township enterprises and the well-off households. We omit the rural capital formation factors, and perform the cointegration test of rural per capita GDP and the variables related to rural financial market growth. It is found that there is a long-run stable equilibrium relationship betweenRGDPand the variables related to rural financial market growth. We perform the cointegration test ofRGDPandFL,RGDPandFE,RGDPandFC, respectively, and find that there is a long-term equilibrium relationship between various groups of variables, indicating that during 1952-2008, there was a stable relationship between rural economy and various variables related to rural financial market growth.

Table2Cointegrationtestresultsofruraleconomicgrowthandruralfinancialmarketgrowthduring1952-2013

Nullhypothesis:numberofcointegrationvectorsEigenvalueTracestatistic(Pvalue)λ-maxstatistic(Pvalue)00.51774594.84494(0.0171)40.11049(0.0309)1atmost0.33676354.73444(0.2305)22.58425(0.4488)2atmost0.22780832.15019(0.3804)25.82321(0.7037)3atmost0.16277217.93150(0.3485)19.38704(0.6433)

Note: indicates that the null hypothesis is rejected at the 5% significance level.

3.3GrangercausalitytestSince there was no long-term cointegration relationship betweenRGDPandFL,FEorFCduring 1978-2013, we use Granger causality test to analyze the relationship between these variables. Table 3 shows the Granger causality test results of rural economic growth and rural financial market growth during 1978-2013. (i) Since the reform and opening up, the large-scale growth of China’s rural financial market has not yet become the Granger cause of rural economic growth; on the contrary, at the 10% significance level, rural economic growth Granger causes the large-scale growth of China’s rural financial market. That is to say, since the reform and opening up, rural economy has promoted the quantitative growth of rural financial market, indicating that the theoretical "rural economy determining rural finance" has been practically proved in China. However, in turn, the large-scale growth of rural financial market does not effectively promote the rural economic growth under considerable non-farm conversion of rural financial resources, indicating that the theoretical effect of rural finance on rural economy is not obvious in China. (ii) The improvement of rural financial market efficiency (FE) does not promote rural economic growth in the short term; in turn, rural economic growth Granger causes the improvement of rural financial market efficiency. (iii) The improvement of rural financial market growth structure (FC) does not Granger cause rural economic growth in the short term, and in turn, rural economic growth does not effectively support the improvement of rural-credit-oriented rural financial market structure (FC).

Table3Grangercausalitytestofruraleconomicgrowthandruralfinancialmarketgrowthduring1978-2013

VariablesNullhypothesisOptimallagFstatisticProbabilityΔFLΔFLdoesnotGrangercauseΔRGDP11.837760.18687ΔRGDPdoesnotGrangercauseΔFL14.042020.05486ΔFEΔFEdoesnotGrangercauseΔRRGDP21.184800.32379ΔRGDPdoesnotGrangercauseΔFE24.276030.02637ΔFCΔFCdoesnotGrangercauseΔRGDP20.443090.64741ΔRGDPdoesnotGrangercauseΔFC20.563260.57701

4 Conclusions and recommendations

4.1ConclusionsThrough study, it is found that since 1952, there has been a long-run equilibrium relationship between China’s rural financial market growth and rural economic growth, the government-led rural financial market growth has effectively supported rural economic growth, and increasing the farmers’ financing ratio has always helped to boost long-term growth of the rural economy,but there is a certain lag. However, dominated by market mechanism from 1978, there is only one-way support relationship: rural economic growth brings about quantitative growth of rural financial market. The empirical results exactly reveal the fact that China’s rural financial market growth is not in coordination with rural economic development.

4.2Recommendations(i) It is necessary to simplify the rural financial market access procedures, reduce access risk under controllable risk, actively nurture and develop rural micro-finance institutions, vigorously develop rural financial services, and promote the quantitative growth of rural financial market. (ii) There is an urgent need to accelerate rural financial legislation, improve the rural financial trading rules, foster rural financial culture, establish incentive mechanisms for rural credit, reduce risk costs of rural financial services, improve the enthusiasm of financial institutions for supporting agriculture, and develop rural securities, insurance, futures, to optimize market structure and disperse agricultural loan risk. (iii) We must make a strict distinction between fiscal and financial support for rural economic development in terms of boundaries and functions, and establish the mechanism of rural public finance and financial support for agriculture, to promote the coordinated development of rural public goods and private goods. (iv) It is necessary to use preferential fiscal, taxation and financial policies, and establish the incentive system for rural financial institutions to support agriculture, in order to increase benefit for farmers and prevent the non-farm trend and urbanization tendency of rural capital. (v) It is also necessary to speed up the innovation of rural informal finance and private financial regulation, and well handle the relationship between rural exogenous finance and endogenous finance. (vi) The government has to protect the active role of market mechanism in promoting growth of rural financial market, and establish effective rural public financial system and fiscal incentives, to promote the rapid development of rural finance and prevent non-farm trend of rural finance. (vii) It is necessary to positively develop rural economy, gradually eliminate the comparative advantage gap between urban and rural areas, improve farmers’ enthusiasm for investment in agriculture, foster effective rural financial needs, encourage and guide urban financial market to drive the rural financial market growth.

[1] WEN T, RAN GH, XIONG DP. Financial development and the income growth of farmer in China[J]. Economic Research Journal, 2005(9):30-43. (in Chinese).

[2] YAO YJ. An empirical analysis of rural financial development and rural economic growth [J]. Economic Science, 2004(5):24-31.(in Chinese).

[3] ZHU X, LI ZN. Analysis on the efficiency of rural credit in China since the reform[J]. Management World, 2006(7):68-76 . (in Chinese).

[4] ZHAO H, PAN SF. The approach to dealing with intercept and trend in practical cointegration analysis[J]. The Journal of Quantitative & Technical Economics, 2004(1):106-109. (in Chinese).

[5] ENGLE RF, GRANGER CWJ. Long-run economic relationships，readings in cointegration[M]. Oxford University Press, Oxford, 1991.

[6] JOHANSEN S. Statistical analysis of co-integrating vectors[J]. Journal of Economic Dynamics and Control, 1998(12): 231-54.

[7] JOHANSEN S. Estimation and hypothesis testing of cointegration vectors in gaussian vector autoregressive models[J]. Econometrica, 1991, 59: 1511-1580.

[8] JOHASEN S. The role of the constant and linear terms in integration analysis of nonstationary variables[J]. Econometric Reviews, 1994, 13: 5-572.