Gao Yongbo,Cao Linlin,Xiao Yao,Sun Lihua

(School of Business Administration,Shenyang Pharmaceutical University,Shenyang 110016,China)

Abstract Objective To explore the relationship of market structure,efficiency and performance of Chinese traditional patent medicine (CTPM)industry,and to provide suggestions for the government and related enterprises to take the corresponding strategies.Methods Econometric model were empirically analyzed about 44 ongoing sample enterprises from 2007 to 2017,based on market power hypothesis and efficiency structure hypothesis of the industrial organization theory.Results and Conclusion The CTPM industry and the unlisted enterprises only conform to the relative market power hypothesis,and there is a significant positive correlation between performance and market share.The listed enterprises do not conform to the market power hypothesis and efficiency structure hypothesis:There is no significant correlation between performance and efficiency,market structure.There is no significant correlation between market share and efficiencies,while the industrial concentration has a significant correlation with efficiencies.The CTPM industry is relatively immature,so it's necessary for the government to establish the elimination mechanism of CTPM industry in line with market rules,to enhance the industrial concentration and resource utilization efficiency by encouraging enterprises to improve management and scale production level.The government should improve the innovation policy of CTPM industry to encourage the enterprises to pursue product innovation.The CTPM enterprises,especially the unlisted enterprises should increase the market shares and enterprise performance through production differentiation.

Keywords: CTPM industry;market structure;efficiency;performance

Chinese traditional patent medicine (CTPM)industry is one of the important high-tech industries,which has good economic benefits.The report of the Ministry of Industry and Information Technology of the People's Republic of China shows that the total assets,sales revenue and profits of CTPM industry are only lower than the chemical pharmaceutical preparation industry,but higher than the other six pharmaceutical sub-industries in 2017,such as the biological pharmaceutical industry and the medical device industry.The profit growth rate of CTPM industry has been higher than the growth rate of China's GDP.And the main business profit margin has long been on the top three in pharmaceutical industry in the recent years[1].At the same time,the CTPM industry has good social benefits,which is closely related to public health and modernization of traditional Chinese medicine.Therefore,it is of great significance to promote the healthy and orderly development of CTPM industry.Performance,which is the core of industrial development is concerned by the stakeholders of CTPM industry.According to the related hypothesis of industrial organization theory,performance comes from market structure or efficiency,while market structure will change with the change of efficiency.Because there are differences in market structure,efficiency and performance among different industries,it is necessary to conduct targeted analysis for the specific industry.This paper analyzes the relationship between market structure,efficiency and performance in CTPM industry to explore the source of performance and the key points of optimizing market structure.Some suggestions and references are given to the government and CTPM enterprise to formulate corresponding strategies to promote the stable development of CTPM industry and enterprises.

1 The relevant theory and literature review

1.1 The relevant theory

The industrial organization theory mainly researches on the contradiction between effective competition and scale economies effect in industry,which provides theoretical basis for maintaining reasonable and orderly market competition and economic operation effect.During the development of different schools of industrial organization theory,market power (MP)hypothesis,efficient structure(ES)hypothesis and the corresponding test models have been formed.The hypotheses and models mainly study the relationship between performance,market structure and efficiency.The market structure variables mainly include industrial concentration and market share,while efficiency variables mainly cover X-efficiency,scale efficiency and so on.

1.1.1 Market power hypothesis

The market power (MP)hypothesis holds that the market structure has an impact on the performance.It can be divided into structure-conduct-performance(SCP)hypothesis and relative market power (RMP)hypothesis.SCP hypothesis holds that the profit rate of enterprises depends on the scale of industrial concentration.Large enterprises will monopolize the market through collusion and other acts,such as raising the market entry barriers for other enterprises,setting higher prices,and reducing effective market competition to obtain excess profits[2].

RMP hypothesis holds that the market power should be reflected by the enterprise's market share[3].Even in the industries with lower concentration,enterprises with the higher market share can achieve the higher performance through the advanced technology,product differentiation and other advantages rather than monopoly.Therefore,Weiss[4],Smirlock[5]and other scholars introduced the market share into the model,such as formula (1):

Whereπrepresents the performance of enterprise,CRrefers to the industrial concentration,MSshows the market share,andXrepresents a group of control variables.The model can only test the MP hypothesis.Whenβ1is greater than 0 and significant,the SCP hypothesis holds in the market;Whenβ2＞ 0 and significant,the RMP hypothesis holds.

1.1.2 Efficient structure hypothesis

The efficient structure (ES)hypothesis holds that efficiency is the main factor that determines performance and market structure.Enterprises will increase their market shares and industrial concentration by improving the efficiency level.As a result,they can also obtain higher profit margins.ES hypothesis can be divided into X-efficiency structure(ESX)hypothesis and scale efficiency structure(ESS)hypothesis.X-efficiency,which generally refers to the pure technical efficiency,is the sum of efficiency except for the impact of scale.It reflects the management level of the human resources,technology and other assets,measuring the management ability of enterprises to control costs and maximize output[6].The scale efficiency is the ratio of comprehensive technical efficiency to pure technical efficiency,which reflects the ability of enterprises to achieve the best production scale under the current conditions.The ESX hypothesis holds that the X-efficiency of an enterprise has a positive correlation with its performance and market structure,which means that an enterprise can achieve a higher performance and expand its market share by improving its management level[7-8].The ESS hypothesis holds that the scale efficiency of an enterprise has a positive correlation with its performance and market structure[9],which means that the enterprise can achieve its technology by expanding its scale.The improvement of technical level can improve the utilization rate of resources,reduce the average cost and achieve higher enterprise performance.With the maturity of efficiency measurement methods,many scholars such as Berger[10-11]use the panel data based on the microenterprise level to analyze the industry.They also bring efficiency into the model (2)(3)and (4).The new model (3)and (4)test the relationship between market share,industrial concentration and efficiency to supplement and verify the ES hypothesis.The model is following:

Where πitrepresents the performance of enterprise,CONCitrefers to the industrial concentration,MS shows the market share of enterprise,XEFitand SEFitrespectively refers to the X-efficiency and scale efficiency of enterprise,and X represents control variables.Whenβ1is greater than 0 and significant,the SCP hypothesis holds in the market.Whenβ2is greater than 0 and significant,the RMP hypothesis holds.Whenβ3,β5andβ7are all greater than 0 and significant,the ESX hypothesis holds.Whenβ4,β6andβ8are all greater than 0 and significant,the ESS hypothesis holds as well.

1.2 Literature review

Many foreign scholars have made descriptive analysis on the market structure and performance of pharmaceutical industry based on the structureconduct-performance paradigm[12-13].Some studies only have made descriptive analysis on some indexes,such as industrial concentration,market performance and market share,whereas there is no corresponding empirical analysis[14-15].Isola[16]researched on the relationship of performance,market structure and efficiency of six Nigerian pharmaceutical enterprises from 1996 to 2010.The conclusion shows that the performance is significantly affected by the market share and industrial concentration.It is suggested that the government should encourage domestic pharmaceutical enterprises to produce drugs that meet the quality standards,prohibit the import of inferior drugs from abroad,and improve the industrial concentration.

Many domestic scholars have also made descriptive analysis on the concentration and performance of Chinese pharmaceutical industry and its sub-industries based on the structure-conductperformance paradigm.It shows that the industrial concentration and performance of the industry are low[17-18].With regard to the relationship of market structure,efficiency and performance of pharmaceutical industry,some scholars only analyzed the relationship of the Chinese pharmaceuticalrelated industries' concentration and performance empirically.Some studies proved that there was a positive correlation between profit margin and concentration (CR4)of the production sales revenue[19],but the others found that there was a positive correlation between the concentration of pharmaceutical industry and the profit margin lagging behind the first stage[20].Chu Shuzhen[21]made an empirical analysis on the data of 16 listed medical devices companies from 2007 to 2016.The results showed that Chinese medical device industry conformed to the SCP hypothesis and the expansion of the assets scale of enterprises was also beneficial to improve the performance of enterprises,but it did not conform to the ESS hypothesis.

In summary,there are few studies on the relationship of the market structure,efficiency and performance of pharmaceutical industry,and the majority of scholars only conduct descriptive analysis or empirical research on industrial concentration and performance.There is no comprehensive study on the relationship of market structure,efficiency and performance of CTPM industry at home and abroad.Based on this,this paper uses the data of MENET database and China Statistical Yearbook.Drawing on the experience of listed companies' data to ref lect the whole industry[22-24],and taking into account factors such as the region and enterprise scales,44 Chinese enterprises operated from 2007 to 2017 are selected as the research objects.(See Table 1 for details).Combined with the relevant hypothesis of industrial organization theory and the model,this paper makes an empirical analysis of the relationship of performance,market structure and efficiency to provide suggestions for promoting the stable development of CTPM industry and enterprises.

Table 1 Distribution of 44 sample enterprises

2 Variable selections

2.1 Market structural variables

The MP hypothesis assumes that market structure is an important factor that affects performance.Therefore,market share and industrial concentration are chosen as the explanatory variables.

Market share is the proportion of product sales revenue,asset and the number of employee in the whole CTPM industry,which reflects the market control power of a CTPM enterprise.The general product sales revenue is selected as the calculation index[25]to reflect the market control power of the CTPM industry.

Industrial concentration reflects the market monopoly and competition of CTPM industry.Because the industrial concentration rateis easier to calculate than the Herf indahl-Hirschmann index (HHI),Gini coeffi cient and other indicators,its data are also easy to obtain and it's more used widely.CR4andCR8of product sales revenue and total assets of CTPM enterprises are calculated respectively (See Figure 1).

Fig.1 Concentration ratio of CTPM industry from 2007 to 2017

On the whole,the total assets and sales revenue ofCR4andCR8have the same trend.Whether measured from the perspective of total assets or product sales revenue,the concentration of CTPM industry achieved a rapid growth in 2010,followed by a slow growth as a whole.TheCR4andCR8of the total assets and sales revenue of CTPM industry are far less than 30% and 40% respectively from 2007 to 2017.This is in line with Bain's competitive market characteristics,which means the market has low industrial concentration and fierce intra-industry competition[26].

2.2 Efficiency variables

According to the ES hypothesis,efficiency is an important factor affecting performance.Drawing experience from the previous scholars[27],data envelopment analysis (DEA)is used to measure the X-efficiency (pure technical efficiency)and scale efficiency of 44 samples of CTPM enterprises.

DEA indicators should be selected.The relevant input indicators include total assets,annual average of all employees,product sales expenses,fixed assets,current assets,etc.Output indicators mainly include product sales revenue,total profits,gross industrial output value,prime operating revenue[28-30].Based on the representativeness,availability and the relations between the indicators,combined with the life cycle of drugs and DEA,the number of indicators is required.The input indicators selected in this study refer to the total assets,including the intangible assets(such as patent rights),fixed assets and current assets,which reflect the R&D investment and production of enterprises.Product sales expenses represent the input of sales in enterprises.Administrative expenses,which include the total cost of production management,organizational activities and employee wages,represent the management input of enterprises.The selected output indicators include product sales revenue,substituting for the missing operation income which reflects the overall outcome of the enterprise.The total profit reflects the profitability level of the enterprise.

2.3 Performance variables

The indicators that measure the performance include return on assets (ROA),main business profitability (MBP),earnings before interest and taxes (EBIT),return on sales,ratio of profits to cost and so on.ROA reflects the profitability and sustainable operation of CTPM enterprises,which is a common index to measure the performance of enterprises.Most ROAs in CTPM enterprises are relatively stable.A few ROAs fluctuate greatly,such as Hengjin Pharmaceutical Co.Ltd.(AE)and Meihe Pharmaceutical Co.Ltd.of Wantong Pharmaceutical Group (AH).Therefore,individual differences among CTPM enterprises may affect the ROAs of enterprises.

3 Empirical research

To research on the relationship between market structure,efficiency and performance of CTPM industry is based on model (2)(3)and (4)that were constructed by Berger et al.The performance is represented by ROA,which is the explanatory variable.In terms of market structure,CR4represents the concentration (CONC),which is the sum of market share of product sales revenue in the top four major CTPM enterprises,and MS represents the market share.Efficiency variables are divided into X-efficiency (XEF)and scale efficiency (SEF).X represents the control variables,including the growth rate of China's gross domestic product (GDP),enterprise asset scale (InAS),enterprise asset liability ratio (OR).

Stata 14.0 is used to analyze the short panel data which include 44 full-sample CTPM enterprises (25 listed enterprises and 19 unlisted enterprises)from 2007 to 2017.The descriptive analysis of all sample variables is as follows (See Table 2).

Fig.2 Time trend map of ROAs in 44 CTPM enterprises

Firstly,model (2),(3)and (4)are tested by F test,LM test and Hausman test in turn to determine whether fixed effect,random effect or mixed regression is used.For the whole sample model (2),Ftest shows that the fixed effect is better than the mixed regression.LM test shows that the random effect is better than the mixed regression.Hausman test shows that the fixed effect is better than the random effect.Then the year dummy variable is defined and the time effect of model (2)is analyzed.The results show that there is an individual effect but the time effect doesn't exit,so the fixed effect should be chosen for model (2).Similarly,the random effect and the mixed regression should be chosen for the full sample models (3)and(4)respectively.The results are shown in Table 3.

Table 2 Descriptive statistics of variables

Table 3 Model formal test for master sample

The whole sample regression results are shown in Table 3.In model (2),there is no significant positive correlation between performance and concentration,X-efficiency and scale efficiency.But there is a significant positive correlation between performance and market share at the confidence level of 0.1%.Therefore,the CTPM industry does not conform to the SCP hypothesis and the ES hypothesis.It only conforms to the RMP hypothesis.It shows in the competitive CTPM industry there are no many CTPM enterprises with monopoly position which obtain excess profit margin by conspiring to damage the effective competition of the market.But there are market dominant forces in the industry,and enterprises with larger market share can obtain higher profit margin through product differentiation such as improving product quality.At the same time,enterprises with higher X-efficiency and higher scale efficiency cannot achieve higher profit margins because of the fierce market competition.There is a significant negative correlation between performance and asset size at the confidence level of 0.1%,which indicates that enterprises' profitability does not increase with the expansion of asset size.There is a significant negative correlation between performance and GDP growth in China,which implies that the profitability of CTPM enterprises didn't decline though China's economy showed a downward trend from 2007 to 2017.

Model (3)supplements and verifies that the ES hypothesis is not valid,and market share has no significant correlation with X-efficiency and scale efficiency.However,China's GDP growth and the expansion of enterprises' assets scale have a positive role in promoting their market shares.Concentration is positively correlated with X-efficiency and scale efficiency,while performance is not significantly correlated with industrial concentration in model (4).Therefore,the CTPM industry still does not conform to the SCP hypothesis.

Table 4 Regression results for master sample

For the listed CTPM enterprises,it is possible to adopt the strategies such as reducing prices or expanding bottom-end markets to maintain the market share because of their abundant capital.Therefore,there is no significant positive correlation between enterprise performance and market share.At the same time,there is no significant correlation between the performance and industrial concentration,X-efficiency and scale efficiency.There is a significant negative correlation between the market share and assetliability ratio at the confidence level of 1%.This shows that increasing the asset-liability ratio of CPTM enterprises will make them bear heavy debts and lack sufficient funds to exploit sales market.Therefore,the listed CTPM enterprises do not conform to the MP hypothesis and the ES hypothesis.

Table 5 Regression results for the listed companies

For the numerous unlisted CTPM enterprises,their profit mainly comes from the growth of market share,so there is a significant positive correlation between their performance and market share at the confidence level of 1%.And there is no significant correlation between the performance and industrial concentration,X-efficiency,scale efficiency.But there is a significant positive correlation between the asset scale and market share at the confidence level of 0.1%,which indicates that enlarging the assets scale of unlisted enterprises reasonably is beneficial to the development of sales market.Similar to the listed companies,industrial concentration has a significant positive correlation with X-efficiency and scale efficiency.Thus the unlisted CTPM enterprises only conform to the RMP hypothesis.

Table 6 Regression results for the unlisted companies

4 Conclusions and suggestions

Based on the empirical analysis of the relationship between market structure,efficiency and performance of CTPM industry from 2007 to 2017,the results show that CTPM industry only conforms to the RMP hypothesis on the whole.And performance and market share have a significant positive correlation.There is no significant correlation between the performance and industrial concentration,X-efficiency and scale efficiency.In addition,there is also no significant correlation between market share and efficiency,but there is a significant positive correlation between the industrial concentration and efficiency.After grouping 44 firms,it is found that there is a significant positive correlation between the performance of the unlisted companies and market share,which is in line with the RMP hypothesis.However,the listed firms do not conform to the MP hypothesis and the ES hypothesis.In order to promote the healthy development of CTPM industry and enterprises,the specific suggestions are as follows.

First,at the industrial level,the elimination mechanism of the CTPM industry should be established.There is no significant correlation between performance and industrial concentration,but the low concentration of CTPM industry can lead to excessive competition among enterprises easily.Therefore,the government should raise the market access threshold reasonably to eliminate the enterprises with backward equipment and low management level.Besides,the preponderant enterprises should be encouraged to improve their management efficiency and scale efficiency,and enterprises should have annexations and reorganizations to improve resource utilization efficiency.

At the same time,the government should improve the innovation policy of CTPM industry.Because there is market power in the CTPM industry,product innovation is of great significance to enlarge the market share of enterprises.Thus the government should tilt to the enterprises with product innovation in tax revenue,public bidding,and expanding essential drug list,which will have a positive incentive for the innovation in the CTPM industry.At the same time,the government should improve the intellectual property protection system of CTPM and enhance the innovation transformation and protection ability in the CTPM industry.

Secondly,at the enterprise level,they should increase market share and pay attention to the management of their assets.Because the market share represents the comprehensive competitive position of enterprises,it also reflects the product quality and market recognition of CTPM enterprises.And the increase of market share of CTPM enterprises,especially the unlisted enterprises,can improve the enterprise performance.Therefore,it is necessary for enterprises to strengthen innovation and R&D investment under the relevant pharmaceutical laws and regulations,and combined with ancient prescriptions and modern technology.In order to improve the market share and performance,CTPM enterprises should enhance the quality,formulation and packaging of their products.Besides,they should strengthen the brand effect to form brand products with quality performance and service as the core which will meet the market demand and ultimately have product differentiation advantages.For CTPM enterprises,performance is negatively related with the assets size of enterprises.Therefore,enterprises need to optimize their assets to promote large-scale operation by improving the management level of enterprise assets.