College of Economics and Management, Huazhong Agricultural University, Wuhan 430070, China

1 Introduction

Strategy is the combination of the goal that the company strives for and the way to achieve it[1], which relates to the survival and development of the enterprise. With the continuous change of the external environment, the concept of strategy is becoming more and more perfect, and more and more emphases are placed on response, competitiveness and risk. Different strategies will bring different risks to enterprises, and the existence of risks will restrict the development of enterprises. This also means that in the face of today’s rapidly changing environment, enterprises must choose the right strategy, in order to remain invincible.

In the process of economic development, each industry will form a set of traditional strategic models[2], the development of enterprises under this traditional strategic model will not have great ups and downs, maintaining the "golden mean" state. Of course, there will also be some enterprises carrying the banner of "revitalization" to make bold reform, gradually deviating from the traditional industry strategy, namely strategic deviance[3]. Strategic deviance may promote enterprises to grasp the competitive advantage to develop vigorously, improve performance, enhance enterprise value, and also may cause enterprises to decline and even fall into bankruptcy because they can not bear the huge risk. It is precisely because the strategic deviance will produce two different results to the enterprise value[4], we must pay attention to the extent to which the enterprise strategy deviates from the traditional industry strategy, in order to avoid the risk, and promote the enterprise value. And agriculture, as the primary industry of our country, has a very significant impact on people’s production and life. Only when agricultural enterprises have good development can the whole society be more quickly towards prosperity.

At present, scholars are generally concerned about the direct relationship between strategic deviance and enterprise value, and play less attention to indirect impact of the two, especially the impact on enterprise value through some kind of channels. The only research results suggest that there are three main transmission paths between strategic deviance and enterprise value: risk, information asymmetry and innovation input. Among them, the asymmetric information is mainly combined with business earnings management for study[5-6], the risk is focused on the information quality of the enterprise[7-9], and innovation input is to emphasize the improvement and enhancement of the original production system[10]. In fact, the transmission path of strategic deviance includes not only the above three ways, but also the customer relationship policy. As customer relationship management is a key link in the process of production and operation, if there is a great degree of deviation in the strategy, the enterprise customer policy will be adjusted accordingly, and such adjustment will inevitably affect the enterprise’s customer concentration. Generally speaking, customer concentration refers to the degree of concentration of key customers for an enterprise as a supplier. Existing research shows that customer concentration will affect the performance of the business[11-13], thereby affecting the value of the enterprise. Therefore, this paper speculates that strategic deviance will affect the customer concentration, and then affect the enterprise value.

In conclusion, this paper intends to study the impact of strategic deviance on enterprise value from the perspective of strategy, and to explore whether customer concentration in plays a mediating effect in it based on the customer concentration. The possible contributions of this article are: (i) It expands the relevant literature of the influencing factors of customer concentration, and conducts empirical research from the meso-level of strategy combined with micro-enterprises, which provides a new perspective for the research of customer concentration. (ii) It enriches strategic research. Nowadays, the study of strategic deviance is still in the preliminary stage, and the scope of the study has great limitations. This paper studies the mediating effect of customer concentration between the strategic deviance and enterprise value, and broadens the scope of the study. (iii) It provides some practical and feasible empirical evidence for agricultural enterprises to enhance their own value. The customer is the root of the enterprise, and it is necessary to optimize customer concentration through the adjustment of its own strategy to enhance the value of agricultural enterprises, which provides a new management idea for agricultural enterprises.

2 Theoretical analysis and research hypothesis

2.1StrategicdevianceandenterprisevalueEnterprise strategy is the overall planning of development of the enterprise, and also the basis for the enterprise to live. Under the impact of macro-economic environment, institutional environment and other factors, each industry will form a strategic model suitable for the development of the industry, and this strategic model is relatively conservative, once the enterprises adopt this traditional strategic model, the economy of enterprises will not be too volatile[14]. But this does not mean that all enterprises are willing to adopt the traditional strategic model. Once the enterprise has chosen the strategy which is different from the traditional strategy model, it forms the strategic deviance, and this deviance has two sides[3-4, 15].

Companies may have stronger competitive advantages as a result of adopting a strategy that deviates from the industry. This competitive advantage not only leads to a significant increase in the market share of enterprises, but also promotes various aspects of enterprise supervision to be more perfect. This leads to the business boom, the enterprise’s management performance enhancement, the staff’s work efficiency enhancement, closer cooperation with the customer, the brand effect enhancement and obvious enterprise’s value enhancement[16-17], but it may cause the enterprise to fall into the bigger risk. The current situation faced by enterprises has not been experienced by the enterprises in the same industry, so it is impossible to emulate and they can only rely on their own to overcome. The existing resources of the enterprise may not meet the needs of the existing strategy of the enterprise[18]. At the same time, the experts and scholars of the industry only stay in the study of the traditional strategic model, so that the enterprises can not obtain any professional advice. And because of the strategic deviation of the enterprise, the stakeholders can not understand and evaluate the enterprise value according to their own ability level and existing knowledge[6], and gradually lose trust in the enterprise, which leads to the decline of the enterprise’s public trust, and affects the reputation and social image of the enterprise.

In the current economic globalization, the economic situation faced by enterprises is not optimistic, the competition between agricultural enterprises is very fierce, so the competitive advantage due to strategic deviance may not be enough to make up for the risk caused by the loss of enterprises. Therefore, according to the above analysis, this paper puts forward the following hypothesis:

H1The bigger the strategic deviance of enterprise, the smaller the enterprise value.

2.2Strategicdeviance,customerconcentrationandenterprisevalueIf the enterprise strategy greatly deviates from the same industry, the policy measures taken by the enterprise for the whole development will be different from those by the traditional strategic model enterprises of the same industry. The policy of the enterprise involves the product development, the production, the sale, the financing and so on. In today’s society of excess capacity and extreme imbalance between supply and demand, the maintenance of customer relationship has become a very important factor restricting the development of enterprises. Therefore, once the enterprise’s strategic deviation degree is big, the enterprise certainly will pay more attention to the customer relationship management, and will take the corresponding measures, such as offering rebate to customers[19], increasing exclusive investment including purchasing specific equipment for key customers, employing professional staff, and so on[20]to keep key customers, so as to enhance the stability of key customers and increase the degree of customer concentration. Therefore, this paper puts forward following theoretical hypothesis:

H2The greater the strategic deviance, the greater the customer concentration for enterprises.

Domestic and foreign scholars mainly focus on the studies of impact of customer concentration on various aspects of enterprises, such as cost and income[21-23], accounting conservatism[24], cash holdings decision[25-28], innovation[29-31], corporate tax avoidance[32-34], earnings management and earnings forecasting[35-36]. From the existing research results, customer concentration will affect the business performance, financial behavior and accounting behavior, leading to changes in the enterprise value; whether the impact on the enterprise value is positive or negative, there are two possibilities.

Theoretically, the higher the customer concentration, the closer the relationship between the enterprise and the customer[19], the enterprise can obtain the customer information in a timely and accurate manner, improve the understanding of customer needs and preferences, produce marketable products to meet customer needs, and improve customer satisfaction, in order to retain customers[37]. It can also help to stabilize the supply chain, save a large sales cost, management costs and other costs[21], while ensuring timely cash recovery, reducing environmental uncertainty and corporate risk[38-39], and enhancing the value of enterprises.

But from a practical point of view, the current market is extremely unbalanced between supply and demand, leading to a weak position of the enterprise as a supplier in front of customers. In order to retain key customers, enterprises are forced to or actively put themselves in a weak position of bargaining power, which also means that if there is high customer concentration, enterprises have to make concessions in the negotiations to achieve cooperation. At the same time, in order to stabilize the customer relationship, the enterprise will also make exclusive investment[40], and the enterprise will invest a large amount of money in the early stage to purchase specific assets and conduct special R & D and so on, which makes the cash flow of the enterprise slow, and increases enterprise risk[41]. The loss of key customers can lead to unrecoverable investments with significant losses and adverse social impacts. When the cost is controlled, the profit of the enterprise will be reduced and the enterprise value will be reduced. According to the analysis, the following hypothesis is put forward:

H3The higher the customer concentration, the smaller the enterprise value.

The existing research shows that the deviance of enterprise strategy will affect the enterprise value. According to the above analysis, the strategic deviance of the enterprise will affect the internal business decision-making, and the customer relationship is an important part of the internal business decision-making. Therefore, the enterprise will pay more attention to customer relationship management and maintenance, and adjust the relevant policies to retain key customers, which leads to the increase of customer concentration, and the increase of customer concentration has two sides to the enterprise value. In other words, customer concentration may be a mediator in the relationship between strategic deviance and enterprise value, and it may be a partial mediator. This is because strategy has an impact on all aspects of business management, customer is only a part of business management, customer concentration may be just one of the transmission paths. Therefore, this paper puts forward the following hypothesis:

H4Customer concentration plays a partial mediating role between strategic deviance and enterprise value.

3 Research design

3.1SelectionofsamplesanddatasourcesTaking the data of A-share agricultural listed companies from 2011 to 2016 as the sample, after excluding ST, PT and other companies with incomplete data, 334 samples are obtained, including 34 samples in 2011, 58 samples in 2012, 61 samples in 2013, 50 samples in 2014, 65 samples in 2015, and 66 samples in 2016. To control the impact of extreme values, all variables are winsorized at the level of 1%-99%. The data in this paper mainly come from CSMAR database, and Stata software is used for data processing.

3.2Definitionofvariables

3.2.1Measurement of enterprise value. This paper uses Tobin’s Q to measure the enterprise value, enterprise value=enterprise market value/total assets.

3.2.2Measurement of strategic deviance. At present, the most common method to measure the deviation degree between enterprise strategy and general industry strategy is to calculate the deviance degree of strategy[3, 7]. This paper mainly uses for reference the calculation method of Ye Kangtaoetal.[6-7], which is adjusted according to China’s national conditions. The advertising and publicity inputs (sales costs/operating income), R & D inputs (net intangible assets/operating income), capital intensity (net fixed assets/number of employees), management expense inputs (management costs/operating income) and financial levers[(short-term borrowing + long-term borrowing + payable bonds)/equity book value] are calculated respectively. Then these six indices were subtracted from the average of this index in agriculture in the same year, respectively, and then divided by the standard deviation, and the absolute value is taken after normalization. And finally the six indices are averaged to obtain the strategic deviance ds.

3.2.3Measurement of customer concentration. There are two main ways to measure the customer concentration: the proportion of the top five customers’ operating income to total operating income and the Herfindahl index of the proportion of the top five customers’ sales. In this paper, the proportion of the top five customers’ operating income to total operating income (cc) is used for empirical study, and the Herfindahl index (hhi) of the proportion of the top five customers’ sales is used to verify in the robustness test.

3.2.4Control variables. This paper mainly controls the enterprise scale, equity concentration and so on, the control variables are detailed in Table 1.

Table1Definitionofvariables

VariabletypeVariableabbreviationVariablenameCalculationstepsExplainedvariabletqTobinsQMarketvalue/totalassetsExplanatoryvariabledsStrategicdeviancedegreeTomeasurethedegreeofdeviationbetweenenterprisestrategyandgeneralindustrystrategyccCustomerconcentrationTheproportionofthetopfivecustomersoperatingincometototaloperatingincomeControlvariablelevFinancialleversTotalliabilities/totalassetszsEquityconcentrationindex1(%)1%ofequityconcentrationindexbmBook?to?marketvalueratioNetassets/totalenterprisemarketvaluelbInterestcoverageratio(Netprofit+incometaxexpense+financialexpense)/financialexpensegrownGrowthGrowthrateofmainbusinessrevenueinthepastthreeyearssizeEnterprisesizeLn(totalassets)

3.3ModeldesignIn order to test the relationship among strategic deviance, customer concentration and enterprise value, this paper builds the following model by using the mediating effect test method of Wen Zhonglin and Ye Baojuan for reference[42]:

tq=α1ds+αicontrol+ε2

(1)

cc=β1ds+βicontrol+ε3

(2)

tq=γ1ds+γ2cc+γicontrol+ε4

(3)

tq=ρ1cc+ρicontrol+ε5

(4)

The model (1) is used to test whether there is a significant relationship between enterprise strategic deviance (ds) and enterprise value (tq), that is, to verifyH1aandH1b, and if is significant, the model (2) will be used for test. The model (2) verifies the relationship between the degree of customer concentration (cc) and the strategic deviance (ds) of the enterprise, that is,H2is verified, and ifβ1is significant, the model (3) will be used for test. If bothγandγ2are significant in the test results of model (3), it indicates that the degree of customer concentration (cc) plays a partial mediating effect, andH4is verified. Ifγ1is not significant in the test results, butγ2is significant, then it plays a full mediating effect. The model (4) is used to examine the relationship between customer concentration (cc) and enterprise value (tq), that is,H3aandH3bare verified.

4 Empirical analysis

4.1DescriptiveanalysisThe results of descriptive analysis are shown in Table 2, from which we can see that the average enterprise value (tq) is 19.520, and the variance is 19.758, showing that the value of agribusiness is quite different. The average strategic deviance (ds) is 0.420, the median is 0.348 8, the variance is 0.264, while the average customer concentration (cc) is 0.276, and the variance is 0.187, suggesting that the strategic deviation and the customer concentration of enterprises in the same industry are obviously different, and because the average of the strategic deviance (ds) is greater than the median, the strategic deviance of more than half of the enterprises is large. Similarly, there are also great deviances in the size of enterprises, financial levers, and so on, so this paper takes them as control variables to control.

4.2CorrelationanalysisTable 3 shows the results of correlation analysis, from which we can see that the enterprise value (tq) is significantly positively correlated with firm growth (grown) and financial leverage (lev), and negatively correlated with enterprise strategic deviance (ds), customer concentration (cc), interest coverage ratio (lb), book-to-market value ratio (bm), equity concentration index 1 (%) (zs), enterprise size (size), which is basically consistent with our hypothesis. The degree of customer concentration (cc) has also a significant positive correlation with the deviance of enterprise strategy (ds), and the degree of customer concentration (cc) has a significant negative correlation with enterprise value (tq), which provides a partial data support for further explaining that the degree of customer concentration has a partial mediating effect between the deviance of enterprise strategy and enterprise value.

Table2Resultsofdescriptiveanalysis

VariablesObsMinMaxMidMeanStd．Dev．tq3343．076182．30014．35519．52019．758ds3340．1261．8700．34880．4200．264cc3340．00560．75250．22320．2760．187lb334-258．6271．72．36031．55551．550bm334-0．01481．5460．34990．4150．295zs334069．9723．371924．62017．490grown334-0．4713．0210．09340．1560．428size33419．3725．2521．814121．9301．212lev3340．94112．802．03872．6972．015

Table3Resultsofcorrelationanalysis

tqdscclbbmzsgrownsizelevtq1ds-0．0300?(0．0797)1cc-0．0861???(0．000)0．0880???(0．000)1lb-0．0332?(0．0518)-0．0235(0．168)0．00370(0．829)1bm-0．5645???(0．000)-0．0639???(0．0002)-0．2167???(0．000)0．00460(0．789)1zs-0．1178???(0．000)0．00610(0．721)-0．0975???(0．000)-0．0118(0．488)0．2648???(0．000)1grown0．1015???(0．000)-0．00160(0．925)0．0200(0．242)0．0297?(0．0815)-0．0809???(0．000)-0．0823???(0．000)1size-0．4680???(0．000)-0．0259(0．130)-0．2621???(0．000)-0．0002(0．000)0．5576???(0．000)0．2710???(0．000)0．0461???(0．007)1

(To be continued)

(Continued)

tqdscclbbmzsgrownsizelevlev0．3855???(0．000)0．0576???(0．001)0．0798???(0．000)-0．1096???(0．000)-0．1136???(0．000)-0．1140???(0．000)0．00380(0．823)-0．3155???(0．000)1

Note:*,**and***indicate significant levels of 10%, 5% and 1%, respectively.

4.3RegressionresultsandanalysisTable 4 shows the regression results of strategic deviance, customer concentration and enterprise value. As can be seen from Table 4, according to the results of model (1), under the condition of controlling the same industry and other conditions, there is a significant negative correlation between the enterprise value (tq) and the strategic deviance (ds) under the significant level of 5%, and the coefficient is -214.04, which shows that the greater the strategic deviance of agribusiness, the smaller the value of enterprise, thenH1is valid. From the regression results of the model (2), it is found that the explained variable is customer concentration (cc), the explanatory variable is strategic deviance (ds), under the control of the same other variables, the strategic deviance (ds) has a significant positive relationship with the customer concentration (cc), and the coefficient is 0.049, which shows that the greater the strategic deviance of agribusiness, the greater the customer concentration, validating the hypothesisH2. Finally, look at the model (3), in the regression results, the coefficient before the strategic deviance (ds) is -198.11, and the coefficient before the customer concentration (cc) is-326.988, and both of them are significant at the level of 5%, which shows that the customer concentration plays a partial mediating effect between the strategic deviance and the enterprise value, that is to say, the hypothesisH4is tenable.

Table4Strategicdeviance,customerconcentrationandenterprisevalue

Step1Model(1)Explainedvariable：tqStep2Model(2)Explainedvariable：ccStep3Model(3)Explainedvariable：tqds-214．040??(-2．28)ds0．049??(2．14)ds-198．111??(-2．17)bm-1414．136???(-13．39)bm-0．066???(-2．82)cc-326．988??(-2．42)zs5．899???(4．48)zs-0．0003(-0．83)bm-1435．576???(-13．62)grown179．894???(4．71)grown0．0096(0．94)zs5．815???(4．44)size-121．573???(-4．55)size-0．0317???(-4．90)grown183．030???(-4．94)lev145．3776???(10．79)lev-0．00032(-0．09)size-131．952???(-4．94)cons4503．433???(8．07)cons0．9638???(6．94)lev145．2736???(11．12)N334N334cons4818．568???(8．52)R20．466R20．0865N334F158．94F7．67R20．4697F153．28

Note: The figures in brackets are thet-test values;*,**,***indicate a significant level of 10%, 5%, 1%, respectively. The same as below.

According to the model (4), the regression results of customer concentration and enterprise value are obtained (Table 5). From the results, we can see that the regression coefficient of enterprise value (tq) and customer concentration (cc) is -347.131, which is significant at the level of 5%, indicating that the higher the customer concentration, the smaller the enterprise value, that is to say, the hypothesisH3is proved.

Table5Regressionresultsoftherelationshipbetweencustomerconcentrationandagriculturalenterprisevalue

Explainedvariable：tqcc-347．131??(-2．51)grown183．329???(4．84)

(To be continued)

(Continued)

Explainedvariable：tqbm-1413．416???(-13．44)size-134．724???(-5．03)zs5．805???(4．42)lev143．657???(10．97)cons4813．995???(8．44)N334R20．4670F167．84

4.4Robustnesstest

4.4.1Changing the measurement method of customer concentration. As there are two measurement methods of customer concentration, the previous section has adopted customer concentration (cc), namely the proportion of the top five customers’ operating income to the total operating income for regression. In order to make the conclusion of the study more robust, the Herfindahl index (hhi) of the sales proportion of the top five customers is used to test the robustness. The results of the robustness tests are shown in Table 6 and Table 7.

Table6Robustnesstestofstrategicdeviance,customerconcentration(hhi)andenterprisevalue

Step1Model(1)Explainedvariable：tqStep2Model(2)Explainedvariable：hhiStep3Model(3)Explainedvariable：tqds-214．040??(-2．28)ds240．155???(2．74)ds-175．2715?(-1．96)bm-1414．136???(-13．39)bm434．068???(4．43)hhi-0．161??(-6．75)zs5．899???(4．48)zs4．703???(3．37)bm-1484．207???(-15．19)grown179．894???(4．71)grown37．780(0．90)zs5．140???(4．25)size-121．573???(-4．55)size-56．330??(-2．27)grown173．795???(4．62)lev145．3776???(10．79)lev19．809(1．48)size-112．480???(-4．41)cons4503．433???(8．07)cons245．08???(4．69)lev142．180???(11．03)N334N334cons4103．88???(7．62)R20．466R20．0274N334F158．94F4．29R20．492F194．04

Table7Robustnesstestoftherelationshipbetweencustomerconcentrationandagriculturalenterprisevalue

Explainedvariable：tqhhi-0．165???(-6．79)grown173．774???(4．64)bm-1464．734???(-15．04)size-114．197???(-4．46)zs5．121???(4．22)lev140．694???(10．90)cons4075．133???(7．45)N334R20．4893F215．06

From the table, we can see that the regression results of Huffindale index (hhi) of customer concentration are basically the same as the regression results of customer concentration (cc).

4.4.2Changing the measurement method of strategic deviance. Using the research methods of Tangetal.[3]and Ye Kangtaoetal.[7]for reference, the advertising and publicity inputs calculated by sales costs and the R & D inputs calculated by the net intangible assets are excluded, because the two have big problems, and they cannot very accurately indicate the cost of advertising and publicity inputs and R & D inputs. According to the previous method, the remaining four dimensions are recalculated to obtain the new strategic deviance (nds), which is then re-introduced into the validation hypothesis. Through the correlation analysis between the four-dimensional strategic deviance (nds) and the six-dimensional strategic deviance (ds), it is found that the correlation coefficient is 0.816 2, and is significant at the level of 1%. The test results of the four-dimensional strategy variance (nds) brought into the model are consistent with the six-dimensional results, and the regression results are shown in Table 8.

Table8Robustnesstestofstrategicdeviance(nds),customerconcentrationandagriculturalenterprisevalue

Step1Model(1)Explainedvariable：tqStep2Model(2)Explainedvariable：ccStep3Model(3)Explainedvariable：tqds-397．60???(-4．44)ds0．049??(2．14)ds-375．446???(-4．34)bm-1403．798???(-13．44)bm-0．066???(-2．82)cc-281．076??(-2．16)zs5．957???(4．58)zs-0．001(-0．83)bm-1423．015???(-13．67)grown181．220???(4．80)grown0．096(0．94)zs5．882???(4．54)

(To be continued)

(Continued)

Step1Model(1)Explainedvariable：tqStep2Model(2)Explainedvariable：ccStep3Model(3)Explainedvariable：tqsize-117．783???(-4．47)size-0．032???(-4．90)grown183．841???(4．88)lev142．818???(10．92)lev-0．001(-0．09)size-126．897???(-4．81)cons4467．554???(8．11)cons0．964???(6．94)lev142．885???(11．23)N334N334cons4740．656???(7．62)R20．475R20．087N334F162．46F7．67R20．477F154．14

5 Conclusions

Based on the samples of A-share agricultural listed companies from 2011 to 2016, this paper discusses the relationship among strategic deviance, customer concentration and enterprise value. It is found that the greater the extent to which the agribusiness strategy deviates from the traditional strategic model of industry, the higher the customer concentration of enterprises, but due to the poor bargaining power of the enterprise in negotiating with the customers and increasing exclusive investment to maintain the customer relationship, the enterprise value is reduced.

The enterprise strategy is the "lighthouse" for the business development, once the degree of strategic deviation is high, the risk faced by the enterprise will be high, so the managers of agricultural enterprises must be more careful to determine the enterprise’s own strategy. It is necessary to make decisions according to the environment of the enterprise and the existing resources and conditions of the enterprise, pay attention to the impact of the strategic adjustment on the customer relationship, and take appropriate measures to avoid the problems caused by the high customer concentration, so as to effectively enhance the enterprise value.

[1] POTER M, CHEN XY(Translator). Competitive advantage[M].Huaxia Publishing House, 1997. (in Chinese).

[2] DIMAGGIO PJ, POWELL WW.The iron cage revisited institutional isomorphism and collective rationality in organizational fields[J].American Sociological Review，1983，48(2)：147-160.

[3] TANG J, CROSSAN M, ROWE WG. Dominant CEO, deviant strategy, and extreme performance：The moderating role of a powerful board[J].Journal of Management Studies，2011，48 (7) ：1479-1503.

[4] YIN ZP, ZHANG ZG. Manager tenure, internal control and strategic differences[J].China Soft Science，2016, 31(12)：132-143. (in Chinese).

[5] CAO Y. Will strategic differences exacerbate corporate earnings management [J].Journal of the Postgraduate of Zhongnan University of Economics and Law，2014, 31(6)：83-90. (in Chinese).

[6] YE KT, DONG XY, CUI YJ. Enterprise strategic positioning and accounting earnings management behavior choice[J].Accounting Research，2015, 36(10)：23-29, 96. (in Chinese).

[7] YE KT, ZHANG SS, ZHANG YX. The correlation between enterprise strategic difference and accounting information value[J].Accounting Research，2014,35(5)：44-51, 94. (in Chinese).

[8] SUI Y. The correlation between enterprise strategic difference and accounting information value[J]. Money China, 2016, 34(13): 264. (in Chinese).

[9] MA Y, ZHANG W. The correlation between enterprise strategic differences, information disclosure quality and accounting information value-data from the listed companies in Shenzhen from 2010 to 2014[J]. Communication of Finance and Accounting, 2016, 37(36):13-17. (in Chinese).

[10] ZENG FR, LI JJ. The analysis of the impact of competition strategy difference on performance[J]. Communication of Finance and Accounting, 2016, 37(29):35-37. (in Chinese).

[11] GOSMAN ML, KOHLBECK MJ.Effects of the existence and identity of major customers on supplier profitability is Wal-Mart different [J].Journal of Management Accounting Research，2009，21(1)：179-201.

[12] NASER A, FATEMEL A. The customer relationship management process: its measurement and impact on performance[J].Uncertain Supply Chain Management，2015，1(3)：43-50.

[13] RADWAN EL SHOGHARI, KASSEM ABDALLAH. The impact of supply chain management on customer service (A case study of Lebanon)[J].Management，2016，6(2)：46-54.

[14] LI ZG, SHI XW. Strategic differences, management characteristics and bank loan covenants——based on the perspective of risk taking[J]. Journal of Zhongnan University of Economics and Law, 2016, 59(2): 68-77, 159. (in Chinese).

[15] GELETKANYCZ MA, HAMBRICK DC. The external ties of top executives：implications for strategic choice and performance[J]. Administrative Science Quarterly，1997, 42(4)：654-81.

[16] GOLL I，JOHNSON NB，RASHEED AA.Knowledge capability，strategic change，and firm performance: the moderating role of the environment[J].Management Decision，2007，45(2)：161-179.

[17] CHEN S, XIAO XX, YANG Y,etal. CEO power, strategic deviation and firm performance：Moderating effects of environmental dynamics[J].Finance and Trade Research, 2014, 25(1): 7-16. (in Chinese).

[18] CHEN S, SHU Q, YANG Y. The influence of Environmental uncertainty on the relationship between strategic change and performance[J].Systems Engineering，2012，30(9)：1-8. (in Chinese).

[19] ZHANG M, MA LJ, ZHANG S. Supplier-customer relationship and auditor selection[J]. Accounting Research, 2012, 33(12):81-86, 95. (in Chinese).

[20] WANG XY, YU CQ. Proprietary costs and voluntary information disclosure: analysis based on customer Information disclosure[J]. Journal of Finance and Economics, 2014, 40(12): 27-38. (in Chinese).

[21] PANOS N, PATATOUKAS. Customer-base concentration: Implications for firm performance and capital markets[J].The Accounting Review，2012，87(2)：363-392.

[22] WANG XY, WANG P, ZHANG JP. Customer concentration and audit fee： customer risk or integration[J]. Audit Research, 2014, 30(6): 72-82. (in Chinese).

[23] SHI XL, SHI ZY. Customer concentration, enterprise-customer relationship and revenue[J]. Money China, 2015, 33(12): 349-350. (in Chinese).

[24] HUI KW, KLASA S, YEUNG PE. Corporate suppliers and customers and accounting conservatism[J].Journal of Accounting and Economics，2012，53(1-2)：115-135.

[25] ITZKOWITZ J. Customers and cash：How relationships affect suppliers’ cash holdings[J].Journal of Corporate Finance，2013，19(1)：159-180.

[26] ZHAO XY, BAO Q. Whether the relationship between supplier and customer affects the level of enterprise cash holding——based on the empirical analysis of panel data of manufacturing listed companies[J]. Journal of Jiangxi University of Finance and Economics, 2014, 16(5): 41-48. (in Chinese).

[27] ZHANG ZH, CHEN J. The influence of customer concentration on the level of corporate cash holding——based on the empirical analysis of listed companies in A-share manufacturing industry[J]. Finance and Trade Research, 2015, 26(5): 148-156. (in Chinese).

[28] WANG XY. Customer concentration, proprietary investment and corporate operating fund whitewash[J]. Friends of Accounting, 2016, 34(15):2-8. (in Chinese).

[29] JOHNSON WC, KANG J, YI S. The certification role of large customers in the new issues market[J].Financial Management，2010，39(4)：1425-1474.

[30] XU H, LIN ZG, RUI C. Customer relationship and R&D investment decisions[J]. Collected Essays on Finance and Economics, 2016,32 (1):47-56. (in Chinese).

[31] ZHENG DP, ZHANG D. Impact of vertical relationship concentration and manager governance mechanism on technological innovation performance[J]. Science of Science and Management of S.& T., 2016, 37(5): 85-95. (in Chinese).

[32] CEN L, EDWARD LM, ZHANG LD,etal. Customer-supplier relationships and corporate tax avoidance[J].Journal of Financial Economics，2017, 123(2): 377-394.

[33] HUANG HH，LOBO GJ，WANG C，etal. Customer concentration and corporate tax avoidance[J].Journal of Banking and Finance，2016，72(12)：184-200.

[34] LI XL, FANG SY, ZHANG L. CEO power, customer concentration and corporate tax avoidance intensity[J]. Modern Finance and Economics-Journal of Tianjin University of Finance and Economics, 2016, 36(9): 100-113. (in Chinese).

[35] FANG HX, ZHANG Y. Supplier/customer relationship trading, earnings management and auditor decision making[J]. Accounting Research, 2016, 37(1):79-86, 96. (in Chinese).

[36] WANG XY, PENG X. Does stable relationship with customers improve the analyst forecasts about suppliers？[J]. Journal of Financial Research, 2016, 37(5):156-172. (in Chinese).

[37] ABBAS K, FATEMEH S. Multidimensional appraisal of customer relationship management: integrating balanced scorecard and multi criteria decision making approaches[J].Information Systems and e-Business Management，2016，14(2)：217-251.

[38] MAHESH S, DEBMALYA M, AJAI SG. Buyer-supplier partnership quality and supply chain performance：Moderating role of risks， and environmental uncertainty[J].European Management Journal，2011，29(4)：260-271.

[39] CHU J, FANG JX. Does customer concentration raise or reduce firms’ crash risk[J]. Economic Theory and Business Management, 2016, 36(7):44-57. (in Chinese).

[40] CHEN J, ZHANG ZH. Environment uncertainty, customer concentration and investment efficiency[J]. Collected Essays On Finance and Economics, 2016, 32(4):54-61. (in Chinese).

[41] DAN D, J SCOTT JUDD, MATTHEW W,etal. Customer concentration risk and the cost of equity capital[J].Journal of Accounting and Economics，2016，61(1)：23-48.

[42] WEN ZL, YE BJ. Analyses of mediating effects： The development of methods and models[J]. Advances in Psychological Science, 2014, 22(5): 731-745. (in Chinese).