Xiaofang ZOU，Xueqin JIANG

1.School of Economics，Yangtze University，Jingzhou 434023，China；2.Research Center for Rural Development in Hubei Province，Jingzhou 434023，China

The domestic and foreign scholars have been concerned about the issues concerning the allocation efficiency of rural financial resources.Begano（1993）［1］points out that the government's agricultural loans solve the problem of shortage of agricultural funds to some extent，but the government's subsidized loans do not significantly improve the output of agriculture.Bencivenga＆＆Smith（1991）［2］and Merton＆Bodie（1995）［3］study the allocation efficiency of rural financial resources from the angles of financial markets and commodity market efficiency.The domestic scholar Zhang Jie believes that the low-income developing country governments often provide low interest loans to stimulate the development of agriculture，but the input and output do not produce satisfactory results.

The studies of Dong Wenjie et al.show that there is low efficiency in China's agricultural loans and financial support for agriculture during the agricultural transformation process［4］.Based on DEA method，Li Jigang analyzes the allocation efficiency of rural financial resources［5］.Based on principal component analysis，Tang Qingsheng et al.build the evaluation indicator system for the allocation of rural financial resources［6］.From the relevant studies，we should not only depend on the financial support for agriculture but also focus on the improvement of the allocation efficiency of rural financial resources，to accelerate the development of rural areas.

Therefore，taking Hubei Province for example，we carry out the empirical research of the allocation efficiency of rural financial resources from the perspective of agricultural input and output relationship，and use the modern rural financial development theory to set forth some policy recommendations on how to build a new rural financial resource allocation system.

1 Research methods and data sources

From the perspective of relationship between agricultural input and output，this article researches the allocation efficiency of rural financial resources in Hubei Province.The capital input of rural financial resources is mainly from financial support for agriculture in rural areas，agricultural credit input of financial institutions，and township enterprise loans of financial institutions.The output is mainly measured by total agricultural output value.

Taking into account the impact of population base，the rural population is divided by these indicators in this study to get the per capita input and output indicators for analysis.The capital input of rural financial resources is mainly from financial support for agriculture in rural areas，agricultural credit input of financial institutions，and township enterprise loans of financial institutions.The output is mainly measured by total agricultural output value.

Thus，we can build the multiple linear regression model with rural per capita output（GDP）as the dependent variable，per capita financial support for agriculture（CZZN），per capita agricultural loans（NDK）and per capita township enterprise loans（LTE）as the independent variables.

GDP＝F（CZZN，NDK，LTE）

The logarithmic transformation of data does not change the variable cointegration，but it can make its trend linearized，and eliminate the heteroscedasticity between the time series.Therefore，we take the logarithm of rural per capita output，per capita financial support for agriculture，per capita agricultural loans and per capita township enterprise loans，denoted by ln GDP，ln CZZN，ln NDK，ln LTE，respectively.

Using the data in the period 1988－2012 as the sample set，the raw data are from Hubei Statistical Yearbook，HubeiRural Statistical Yearbook and Hubei Financial Operation Report.All data are under logarithmic treatment，for the purpose of reducing the volatility of the data.Eviews6.0 software is used for the data processing.

2 Results and analysis

2.1 Variable stationarity testWhen performing the time series analysis，the time series is traditionally required to be stationary，that is，there must be no random trends or deterministic trends，or else there will be＂spurious regression＂.However，the time series is generally non-stationary in the real economy.The study uses ADF unit root test method to test the stationarity of four variables，and in the test process，the lag is determined according to AIC selection criteria.The results are shown in Table 1.

The original time series are non-stationary series at the 5%significance level，and in order to avoid spurious regression，here we can not directly use the traditional method of least squares.The first-order difference series of the original series are stationary at the 5%significance level，and all of them are integrated of 1 series，denoted as I（1），so we can first perform difference and then perform regression on the variables.

However，the long-term information of total amount will be lost，so the cointegration test is used to solve this problem.

Table 1 ADF unit root test results of variables

2.2 Cointegration testStationary test results show that LN（GDP），LN（CZZN），LN（NDK）and LN（LTE）are all integrated of1 series，meeting the cointegration premise，so we can use cointegration test to determine whether there is a long-term stable relationship between them.

According to minimization principle of AIC and SC criterion，we determine the lag P as 1，and in this condition，the fitting effect of this model is the best.The lag P＝1 is also relatively reasonable to the year data in this article.

Based on the characteristics of the original data，when testing the number of cointegration vector，the cointegration equation contains only the intercept.The test results are shown in Table 2.

Table 2 Cointegration test results

The test results in Table 2 show that there is one statistic greater than the critical value at the 5%level，so there is one hypothesis to be rejected，that is，there is a cointegration equation accordingly，namely there is a long-term cointegration relationship between the rural per capita output and some variables（per capita financial support for agriculture，per capita agricultural loans，and per capita township enterprise loans）.

By OLS estimation，the derived cointegration equation is as follows：

The coefficient of determination and adjusted coefficient of determination of regression equation（1）are 0.903 5 and 0.898 3，respectively，and t value is significant，passing t test，indicating that the goodness of fit of the model is good.

Letε＝LN（GDP）－4.218 1－0.155 5LN（CZZN）－0.471 8LN（NDK）－0.032 9LN（LTE）and perform the stationarity test on the seriesε.The ADF test value is－2.302 1，and 5%critical value is－1.957 2.The results show that at the 5%significance level，is the stationary series，and the value fluctuates up and down around zero.

This proves that there is a unique long-term cointegration relationship between the time series（rural per capita output，per capita financial support for agriculture，per capita agricultural loans and per capita township enterprise loans）.

Based on the cointegration equation（1），we can draw the conclusion that there is a long-term positive relationship between agricultural output and some variables（financial support for agriculture，agricultural loans，township enterprise loans），and the coefficient of elasticity of financial support for agriculture and agricultural loans to agricultural output is high.

For each additional one yuan in the financial support for agriculture，the agricultural output will grow by 0.155 5 yuan，and for each additional one yuan in the agricultural loans，the agricultural out put will grow by 0.471 8 yuan.The coefficient of elasticity of township enterprise loans to agricultural output is low.For each additional one yuan in township enterprise loans，it will bring 0.032 9 yuan of income for the agricultural output.

The error correction model（ECM）is built to examine the short-term dynamic impact of per capita financial support for agriculture，per capita agricultural loans and per capita township enterprise loans on the per capita agricultural output，as well as the adjustment of per capita output on its own long－run equilibrium trend deviation.

According to the selected variables and the OLS estimate correction，the error correction model can be obtained as follows：

As can be seen from the error correction equation（2），the error correction coefficient is less than zero（－0.0109），fully consistent with the reverse correction mechanism.And in an average sense，the degree of adjustment of the non-equilibrium error of GDP，financial support for agriculture，agricultural loans and township enterprise loans in the previous year on the non-equilibrium trend deviation in the current year is 1.09%.

2.3 Granger causality testOn the basis of the cointegration test，we carry out the Granger causality test on the relationship between the variables，to determine whether there is a causal equilibrium relationship between the variables.This study performs test in the optimal lag 2，and the results are shown in Table 3.

Table 3 Granger causality test results

From Table 3，it can be found that the financial support for agriculture and township enterprise loans do not Granger cause the increase in agricultural output.Similarly，the increase in agricultural output does not Granger cause financial support for agriculture and township enterprise loans.

That is to say，the increase in financial support for agriculture and township enterprise loans does not cause changes in the added value of agricultural output，and the former will not be adjusted accordingly with the changes in the latter.

Agricultural loans do not Granger cause the increase in agricultural output，but the increase in agricultural output does Granger cause the increase in agricultural loans.It can be understood as the fact that the increase in agricultural loans input does not directly bring the increased agricultural output，but the increased agricultural output will cause an increase in agricultural loans.

The economic theory based on the resource allocation efficiency believes that in a capital market with high capital allocation efficiency，the increase in the profits of one industry may cause more money to flow into this industry，that is，the industry profit growth can lead to investment growth accordingly.In fact，there is no such relationship between agricultural output and the variables of financial support for agriculture，agricultural loans，and township enterprise loans in Hubei Province.

This shows that the resource allocation efficiency of agricultural loans and financial support for agriculture in Hubei Province is low.

3 Conclusions and recommendations

3.1 ConclusionsFirstly，there is a long-term cointegration relationship between agricultural output and the variables of financial support for agriculture，agricultural loans and township enterprise loans.In the main factors affecting the increase in agricultural output，agricultural loans make the greatest contribution（47.18%），followed by financial support for agriculture（15.5%）and township enterprise loans（3.29%）.

This shows that in the allocation of financial resources in rural areas，in order to promote agricultural production and farmers' income，the role played by the government and financial institutions in directly providing input for the farmers is much higher than the role played by the credit funds in supporting the rural economic organizations.

This conclusion is consistent with the fact that the key to current China's rural economic development lies in the support of government spending and financial institutions.The key to improving the the allocation efficiency of rural financial resources lies in the increase of quantity and optimization of structure of input of government spending and financial institutions to＂agriculture，farmers and countryside＂.

Secondly，in the short term，the contribution of agricultural loans to output is large，and for each additional one percentage point in the short-term agricultural loans，the out put will increase by 0.2769 percentage points，19.49%less than the rate of contribution of short-term agricultural loans to agricultural output.

In the short term，the contribution of financial support for agriculture and township enterprise loans to output is negative，indicating that although the financial support for agriculture and township enterprise loans are increased，the per capita agricultural output is reduced.

The reasons for this conclusion may be as follows：

（i）The utilization efficiency of agricultural loans is low and it does not produce the expected results.

（ii）The financial support for agriculture is mainly used for construction of water conservancy facility and roads，and the development of agriculture is promoted by constantly improving the rural infrastructure.The infrastructure input is large，and the payback period is long，so the impact of financial support for agriculture on the increase in agricultural output is negative in the short term but positive in the long term.

Thirdly，the reverse flow of rural financial resources is obvious in Hubei Province.The deposits absorbed by Postal Savings Bank and rural credit cooperatives mostly flow out to economically developed regions.This results in the shortage of capital for supporting agriculture in Hubei Province and low allocation efficiency of rural financial resources.

3.2 Recommendations

3.2.1 Improving the main function of rural finance and playing the role of financial institutions in supporting agriculture.It is necessary to establish the long-term agriculture-supporting mechanism of Agricultural Bank to make it provide services for＂agriculture，farmers and countryside＂and county economy；expand the agriculture-supporting service scope of Agricultural Development Bank to strengthen its policy banking functions；give full play to the role of the main force of rural credit cooperatives to enhance the level of financial services for the county's economic development；build the capital backflow mechanism of Postal Savings Bank，set up the rural consumption and rural microfinance business to nurture the county's economic development and give full play to its role in supporting the new rural construction.

3.2.2 Guiding and regulating the private finance to expand financing channels.In rural areas，taking into account the risk prevention and the pursuit of profit，the formal finance is unwilling to provide loans to farmers，and the private finance can transfer and spread the bank's credit risk to provide another financing path for the rural businesses and farmers，and effectively fill the gaps in rural financial needs.

It is necessary to guide and regulate the private lending，open the rural private financial markets，and actively explore the＂sunshine＂mode of private financial financing，to promote the rural economic development and optimize the rural social credit environment.

3.2.3 Accelerating the rural financial service innovation.It is necessary to innovate upon the derivative products of financial support for agriculture，especially agricultural insurance，agricultural credit insurance，agricultural futures and options，to effectively disperse and avoid all kinds of agricultural credit risk；expand the rural credit service field of financial institutions，and increase variety of businesses，to meet farmers and rural enterprises' diversified needs for financial services.

3.2.4 Improving the rural financial system and optimizing the allocation of financial resources.It is necessary to promote the formation and development of diversified pattern of rural financial institutions，and make the financial institutions actively carry out rational and orderly full competition.

Meanwhile，the local governments should vigorously promote the building of rural credit system，and implement the farmer credit information sharing，to achieve the connection between rural credit assessment work and credit system，and ease the rural information asymmetry.

3.2.5 Striving to improve and optimize the rural financial ecological environment.The key to new rural construction is to establish the long-term funding gathering mechanism，and its core is to optimize the rural financial ecological environment.Favorable financial ecological environment is conducive to improving the allocation efficiency of financial resources，rationalizing the operational mechanism of rural funds，guiding the back flow of capital，guarding against the financial risks，and improving the investment environment in rural areas，to truly form the＂depression＂effect of capital accumulation.

［1］Pagano.Financial market and growth：An overview［J］.European Economic Review，1993（7）613－622.

［2］Bencivenga Smith.The order of economic liberalization financial control in the transition to a market economy［M］.John Hopk ins University Press，1991.

［3］Walter Erwin Diewert，Douglas W Caves，Laurits R Christensen.The economic theory of index numbers and the measurement of input，output and productivity［J］.Econometrics，1982，50（6）：1393－1414.

［4］DONGWJ，WEN T.An empirical study on financial and monetary coordination and the development of agriculture economy in China［J］.Financial Theory and Practice，2010（10）30－34.（in Chinese）.

［5］LIJG.Analysis on rural financial resource allocation's efficiency based on DEA［J］.Finance and Accounting Monthly，2010（7）40－43.（in Chinese）.

［6］TANG QS，ZHOU MY.An empirical study on the efficiency of rural financial resources allocation in China's western regions——An analysis based on principle component method［J］.Journal of Yunnan Finance and Trade Institute，2009，25（4）116－122.（in Chinese）.

［7］REN F，YIN YP，CHANGM.Rural financial resource allocation efficiency and empirical studies：A case of Shanxi Province［J］.Financial Theory and Practice，2009（8）：58－62.（in Chinese）.

［8］LIMX，XIANGZD.Empirical analysis on rural financial resource allocation efficiency in middle China［J］.Journal of Agrotechnical Economics，2011（7）：75－81.（in Chinese）.

［9］YANG JL.Analysis on factors influencing the rapid development of＂People Finance＂in rural China［J］.Inquiry into Economic Issues，2007（5）98－100.（in Chinese）.

［10］LI JG，CHEN T.Empirical analysis on efficiency of rural financial resources allocation in Xinjiang［J］.Xinjiang Agricultural Sciences，2006，43（5）：446－450.（in Chinese）.