How does subsidy change a firm's market power? The case of China's rice processing industry

Dai, Jiawu; Li, Xun Article How does subsidy change a firm's market power? The case of China's rice processing industry Journal of Applied Economics Provided in Cooperation with: University of CEMA, Buenos Aires Suggested Citation: Dai, Jiawu; Li, Xun (2020) : How does subsidy change a firm's market power? The case of China's rice processing industry, Journal of Applied Economics, ISSN 1667-6726, Taylor & Francis, Abingdon, Vol. 23, Iss. 1, pp. 372-384, https://doi.org/10.1080/15140326.2020.1776976 This Version is available at: https://hdl.handle.net/10419/314097 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Journal of Applied Economics ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/recs20 How does subsidy change a firm’s market power? The case of China’s rice processing industry Jiawu Dai & Xun Li To cite this article: Jiawu Dai & Xun Li (2020) How does subsidy change a firm’s market power? The case of China’s rice processing industry, Journal of Applied Economics, 23:1, 372-384, DOI: 10.1080/15140326.2020.1776976 To link to this article: https://doi.org/10.1080/15140326.2020.1776976 © 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. Published online: 18 Jun 2020. Submit your article to this journal Article views: 4001 View related articles View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=recs20 How does subsidy change a firm’s market power? The case of China’s rice processing industry Jiawu Dai a and Xun Li b a Business College & Center of Large Country Economy Research, Hunan Normal University, Changsha, Hunan Province, P.R. China; b College of Economics and Management, Wuhan University, Wuhan, Hubei Province, P. R. China ABSTRACT The effect of subsidy on firms’ market power is controversial and unclear. In this article, we investigate such effect through an unbalanced panel data at firm level. Empirical results indicate that subsidy weakens the market power of firms subsidized. We then verify our hypothesis for this result that striving for subsidy through building or keeping relationship with governments will lead to higher administration and selling expense, and therefore lower market power, given that the rice processing industry is relatively competitive due to its low entry barrier and high homogenous product. Compared with non-state-owned enterprises, stateowned enterprises are found to be weaker in market power, to be higher in administration expense and to be lower in selling expense, which are well consistent with China’s reality. Finally, robustness test consolidates our conclusions. ARTICLE HISTORY Received 24 February 2020 Accepted 27 May 2020 KEYWORDS Subsidy; market power; SOEs; non-SOEs 1. Introduction Social stability and economic growth are generally the two main goals for transition economies. To achieve these objectives, governments from countries in transition usually provide various assistance to intervene economic activities. A widely used one is subsidy (Frye & Shleifer, 1997), which is regarded to be necessary, especially on protecting infant industries or vulnerable groups. In China, for example, the government spends an enormous sum of money to subsidize enterprises each year, with aims to prompt employment, innovation and development, especially for some industries with disadvantage or strategic significance, such as agriculture, food and high-tech sectors (Dang & Motohashi, 2015; Jaumandreu, 2005; Yi, Sun, & Zhou, 2015). The subsidy for China’s grain processing mainly targets at promoting the level of industrialization and scale development of the enterprises, which are regarded as essential cornerstone for rural employment and food security (Wu & Xu, 2017). The National Office of Comprehensive Development of Agriculture (NOCDA), a specialized agency established under the Ministry of Finance (MOF) of China, undertakes the work of authorizing subsidy policies related to agribusiness. For instance, the Guidance on Subsidizing Programs of CONTACT Xun Li [email protected] College of Economics and Management, Wuhan University, Wuhan, Hubei Province 430072, P.R. China JOURNAL OF APPLIED ECONOMICS 2020, VOL. 23, NO. 1, 372–384 https://doi.org/10.1080/15140326.2020.1776976 © 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/ licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Comprehensive Development and Industrialized Operation of Agriculture in 2009 (No. 2008–208 Document of NOCDA) explicitly stipulates the detailed plan of subsidy on agribusiness, including the principles, subsidizing range and targets, application requirements and other items. It also regulates the spending orientations of the subsidy fund on production workshop, equipment, infrastructure of water and electricity, road, quality inspection, environmental protection, etc. 1 The effect of subsidies in China and other countries is discussed and confirmed by the literature. For instance, subsidy is demonstrated to be affirmative in increasing innovation output for Chinese manufacturers (Chen & Zhu, 2008), or in facilitating innovation performance only if the subsidy income is used to promote human capital (An, Zhou, & Pi, 2009). Moreover, Broekel (2015) found that subsidies for R&D cooperation contributed to stimulate the regional innovation efficiency in Germany. Pechrová (2015) manifested a positive and statistically significant impact of subsidies on the technical efficiency for Czech farms. Despite Shepherd (1972) declared that political elements including all kinds of subsidies are factors of importance on determining market power, whether subsidy strengthens firms’ market power is still controversial. On the one side, a positive relationship between subsidy and market power may originate from the following aspects. First, subsidy tends to induce rent-seeking and unfair competition, resulting in the change of firms’ market power. Specifically, subsidy is regarded as a recognized signal by the government, or a symbol of good relationship with the government in a political-led society. By virtue of either or both the enterprises subsidized are relatively easier to obtain financing and bank loan (Feldman & Kelley, 2006; Kleer, 2010), and some other invisible advantages. Second, the positive effect from subsidy on innovation may also bring a positive relationship between subsidy and market power, given innovation is proved as an important source of monopoly (Aghion & Howitt, 1992; Klette & Griliches, 2000; Liu & Huang, 2016; Zhang & Jia, 2011). On the other side, enterprises subsidized by the government are supposed to face higher cost and therefore lower market power. First, it is common that many enterprises in China strive to acquire more subsidy through building and managing relationship with government. It spends a lot of resources which would be used to improve enterprises’ performance. This process increases their production and management cost (Zhao, Wang, Yang, & Cao, 2015). Yu, Hui, and Pan (2010) proved this point that fiscal subsidy on China’s private enterprises connected to local governments will generate a negative effect on their performance. From another perspective, Liang, Li, and Lv (2012) found that subsidy policy in under-developed regions tends to attract firms with low efficiency to enter the local markets. Furthermore, to obtain more subsidy, managers always make some unpractical decisions, e.g., financial fraud, excess employment or production. Second, subsidy would reduce the incentive of enterprises subsidized on raising efficiency as they can enjoy a steady and sometimes large benefits easily. In terms of viability, Lin (2012) pointed out that huge subsidy has to be paid to enterprises when the government forces to develop the sectors violating factor endowment advantage, while those enterprises being short of viability would be hard to achieve international competitiveness when they lose subsidy. Similarly, Huang, Song, and Zhu (2015) also 1 Please see the official website of the MOF of China for detailed information. http://nfb.mof.gov.cn/zhengwuxinxi/ zhengcefabu/xiangmuguanlilei/200812/t20081202_94003.html. JOURNAL OF APPLIED ECONOMICS 373 found that in a market with high degree of competition, ongoing subsidies would reduce the sensitivity of enterprises to competition pressure and cause the risk of so-called production only for obtaining subsidies as well as overcapacity. In consequence, whether subsidy ultimately strengthens or weakens market power depends on the offset between these two considerations. Additionally, debate on the heterogeneity of market power between China’s SOEs (stateowned enterprises) and non-SOEs (non-state-owned enterprises) has drawn a lot of attention. China’s SOEs have been criticized intensively for their low-cost efficiency and privileges on bank loans, investment, financing and so on (Ariff and Can, 2008; Fu & Heffernan, 2007; Wei & Wang, 2000; Zhao, Zhong, & Jiang, 2001; Zheng, Liu, & Bigsten, 2003). In addition, SOEs are more likely to be subsidized since the government has the socalled paternalism on SOEs (An et al., 2009; Wu & Shen, 2013). Therefore, the effect from subsidy on market power may include a part of indirect effect from ownership, and we will take it as a control variable to isolate this effect. In addition, it is helpful for us to test whether there is a significant difference in market power between SOEs and non-SOEs. To the best of our knowledge, previous studies have paid relatively little attention on examining the relationship between subsidy and market power, especially for China. This article sheds light on testing this relationship using rice processing industry as a case study. First, rice is one of the most important and heavily consumed food in China. The planting area of rice accounts for 18.16% of the total sown area of farm crops, producing nearly a third of China’s grain in 2015. 2 Second, the rice processing industry plays an important role in China’s food industry which links farmers and consumers directly. Third, the rice processing industry is heavily subsidized in China given its low profit, high labor-intensive character, and the important role on food security. The subsidies obtained by rice processing enterprises may include various items, e.g., Agricultural Industrialization subsidy, employment subsidy and Interest subsidy. Since we have no detailed information about the variety and amount of subsidies for our sample, the subsidy discussed in this article is a comprehensive variable. No matter what kind of subsidy a firm obtains, it does mean that firm acquires benefit or attention from the government. In this article, we estimate the market power for each individual rice processing enterprise with a stochastic frontier model and then investigate the effect of subsidy on market power empirically. The rest content of the article is arranged as follows. Section 2 introduces the methodology and data. Section 3 describes the empirical results. Section 4 explains the robustness check, and Section 5 makes the conclusion. 2. Methodology and data We construct our methodology framework starting at assuming that firms pursue maximum profit, and government subsidy is increasing with output, 3 then firms’ profit can be represented as follows: 2 Data source: China Statistical Yearbook 2016. 3 This assumption is consistent with the reality of China’s agribusinesses, since the main indicator according to which the government decides whether or how much to provide subsidy to a firm is its production size. Scaled firms are more possible to be subsidized since a large amount of production means big contribution to food security and employment, which are the most important policy targets in China. 374 J. DAI AND X. LI where pit and sit are firm i’s price and unit subsidy, respectively, Qit is output, C1 it and C2 it represent costs of production and subsidy, respectively, Sit ¼sit �Qit is total subsidy. The FOC of firm i is given as: Then, we have the expression of Lerner Index imbedding with subsidy: where εit ¼  dpit dQit Qit pit represents firm i’s demand elasticity. The marginal cost of subsidy MC2 it ¼dC2 it Sit ð Þ dSit is expected to be positive as is well known that obtaining subsidy always induces additional cost for firms subsidized in China, e.g., rent and expenditure for building relationship with the officials of the subsidy-related authorities. Market power has nothing to do with subsidy if MC2 it ¼1, which means the increasing subsidy can only compensate the corresponding cost. In that case, firms have no incentive to acquire subsidy from the government. If 0 <MC2 it <1, it implies that the cost of an additional unit of subsidy is less than the increased revenue from the subsidy, and market power will be less than the case without subsidy. The higher the subsidy level (larger sit), the weaker the market power. In other words, firms in this situation do not need to struggle for strong market power to achieve maximum profit. While if MC2 it >1, i.e., adding one unit of subsidy requires paying more than one unit of cost; then, enterprises will increase market power. Under this condition, the higher the subsidy level, the more the company needs to increase prices or reduce marginal costs to enhance market power and therefore to maintain profit maximization. In general, firms’ market power increases with the marginal cost of subsidy MC2 it, every rational firm would not operate under the condition of MC2 it >1. The stochastic frontier cost function (SFCF) proposed by Schmidt and Lovell (1979) and improved by Berger and Hannan (1998) has been widely employed to estimate the market power for each individual enterprise (Guevara, Maudos, & Pérez, 2005; Maudos & Guevara, 2007; Solís & Maudos, 2008; etc.). Considering the special condition of China’s rice processing industry, we construct the SFCF as follows: where Cit ¼C1 it þC2 it and Qit represent total cost and output, respectively.ωit and T are input prices (including raw material, laborand capital) and time tendency representing technical progress, respectively. uit represents the cost inefficiency and is assumed to be independently half-normally distributed, i.e., Nþ0;σ2 u  �, and υit is a white noise with independent identical distribution, i.e., υit~ i:i:d:N0;σ2 υ  �. Taking derivative on both sides of Equation (4) with respective to Qit, we can get the expression for marginal cost: JOURNAL OF APPLIED ECONOMICS 375 and the Lerner Index can be calculated through the following equation: To estimate the effect of subsidy on market power, our model is as below: where Lit and SUBit are the Lerner Index and subsidy for firm i, respectively. Dit is set as an ownership dummy variable, where Dit= 1 if the firm is SOE, and otherwise Dit= 0. EDUit, ADVit, RDit, TAXit and XKTAit are control variables representing employee training expenditure, advertising expense, R&D expenditure, tax, and the ratio of capital to total asset, respectively. We apply an unbalanced panel data at the firm level from 1999 to 2011. The data of all the variables except prices of rice and paddy are taken from China’s Industrial Enterprise Database, which is also called Annual Survey of Industrial Firms (ASIF) in some other literature. This is the best and widely used micro dataset at the firm level in China. It covers all the state-owned industrial enterprises and non-state-owned ones with prime operating revenue above five million CNY. The database contains almost all the important financial variables but has no direct information on output and material prices. In consequence, we have to calculate the output of rice for each firm through dividing the output value by the rice prices at the province level. We obtained the price of rice and paddy from Statistical Yearbook of China and matched them to the firm-level database according to the location code for each firm. In that case, it has an underlying implication that firms in the same province face an identical output and material price, which is not very precise but acceptable as a second best choice. Table 1 illustrates the number of observations, mean, standard deviation, minimum and maximum for the main variables such as rice production, rice price and capital price. For example, the average rice price is 2.695 CNY per kg, with minimum 1.388 CNY per kg and maximum 5.05 CNY per kg. Table 1. The summary statistics for main variables (1999–2011). Variables Obs Mean Std. Dev. Min Max Rice production 3033 13,000,000 24,300,000 66,229 518,000,000 Rice price 3033 2.695 0.676 1.388 5.050 Capital price 3033 0.062 0.005 0.055 0.071 Wage 3033 18,063 6806 6195 37,441 Paddy price 3033 1.625 0.414 0.842 3.040 Subsidy 2727 37,977 399,434 −136,000 12,100,000 Advertisement 2035 13,479 125,573 0 3,769,000 R&D 1752 73,224 1,356,947 0 50,100,000 Tax 3033 912,358 4,496,134 1,000 174,000,000 Capital/total asset 2870 0.435 0.243 0.005 4.531 376 J. DAI AND X. LI 3. Empirical results Table 2 shows the estimation results of SFCF. Most of the coefficients are statistically significant at the 1% or 5% level. The estimate of γ is 0.917 and significant at the 1% level, indicating that applying the stochastic frontier model is sensible. We calculate the marginal cost for each enterprise through substituting the relevant coefficients estimated into Equation (5) and then compute the Lerner Index through Equation (6). Figure 1 illustrates the median value of estimated Lerner Index annually from 1999 to 2011. It shows that market power of China’s rice processing enterprises is relatively weak (i.e., less than 0.06). One possible explanation could be that China’s rice processing industry has very low entry barrier and the product is relatively homogenous. Thus, it is quite competitive rather than monopolistic. In addition, an Inverted-U shape variation is easy to be found from 2001 to 2007, following with an increasing trend after 2007. Table 3 presents the effect of subsidy on market power. It includes five regressions in which column (1) does not control any fixed effect and other variables which may affect market power, column (2) adds four control variables. Year and province fixed effects are gradually introduced into columns (3) and (4), while column (5) controls the year by province fixed effects to capture the fixed effects varying with year and province simultaneously. Robust standard errors are reported in parentheses. Empirical results show that subsidy has a negative and statistically significant effect on Lerner Index, suggesting that subsidy on rice processing enterprises tends to weaken their market power. After controlling the year and (by) province fixed effects, the coefficients of subsidy turn to be Table 2. The estimation results of SFCF. lncit Coefficient Std. Dev. γk9.283** 4.609 γl0.499 0.999 γm−98. 300*** 16.00 γq17.53*** 2.868 γkk 3.712*** 1.203 γlk −0.240 0.362 γmk −1.340*** 0.440 γml 1.001*** 0.179 γll −0.213*** 0.067 γmm −1.868*** 0.131 γqq 0.023*** 0.001 γkq 0.151*** 0.027 γlq 0.055*** 0.012 γmq 0.043*** 0.010 ρt0.145*** 0.023 ρq−0.008*** 0.001 ρk– – ρl– – ρm0.046*** 0.008 CONS −274.00*** 46.00 σ20.504*** 0.033 γ0.971*** 0.002 σ2 u0.490*** 0.033 σ2 υ0.015*** 0.000 *** and ** represent significance at the 1% and 5% level, respectively. The coefficients of the cross terms of T and labor, as well as T and raw material, are omitted due to the collinearity. Robust standard errors are reported in the parentheses. JOURNAL OF APPLIED ECONOMICS 377 slightly smaller (from −0.021 to −0.019) but still significant, validating the negative relationship between subsidy and market power. This is probably due to the increased marginal cost induced by acquiring subsidy from the government. Given the rice price in a competitive market with low entry barrier and highlevel ofhomogeneity, the increase of marginal cost for firmiwould weaken its market power. We will further discuss and test this intuition at the end of this section. What is more, the results also demonstrate that SOEs have lower market power. The main reason is that SOEs are generally found to have Figure 1. The median value of Lerner index in each year. Table 3. The estimation results for the effects of subsidy on market power. Variables (1) Without controlling any effect (2) Controlling related variables (3) Controlling year fixed effect (4) Controlling year & province fixed effects (5) Controlling province by year fixed effects Subsidy −0.021** −0.022** −0.022*** −0.017** −0.019** (0.009) (0.009) (0.008) (0.008) (0.008) SOEs −0.204** −0.219** −0.150** −0.317*** −0.345*** (0.087) (0.087) (0.066) (0.065) (0.069) Training expenditure −0.023*** −0.029*** −0.025*** −0.025*** (0.007) (0.006) (0.006) (0.006) Advertising expense 0.016** 0.003 0.001 0.001 (0.007) (0.007) (0.006) (0.007) R&D expenditure −0.002 0.016 0.016 0.144 (0.012) (0.011) (0.011) (0.011) Tax −0.020* 0.038*** 0.030*** 0.031*** (0.012) (0.010) (0.011) (0.011) Capital/total asset 0.081** 0.076** 0.062** 0.041 0.039 (0.032) (0.032) (0.028) (0.027) (0.028) Constant −3.171*** −2.915*** −3.856*** −3.710*** −3.809*** (0.041) (0.148) (0.196) (0.197) (0.349) Year FE NO NO YES YES NO Province FE NO NO NO YES NO Year*Province FE NO NO NO NO YES R 2 0.009 0.021 0.052 0.139 0.205 Observations 2870 2870 2870 2870 2870 ***, ** and * represent significance at the 1%, 5% and 10% level, respectively. All the variables are taken the natural logarithm. Robust standard errors are reported in the parentheses. In column (5), we add the cross term of year and province to capture the fixed effects varying with year and province simultaneously. 378 J. DAI AND X. LI