Research on the Relationship between the Profitability and R&D Investment of Enterprises of Different Scales in China’s Pharmaceutical Industry

Kudireti·shataer,Chen Yuwen

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

Abstract Objective To analyze the relationship between the profitability of pharmaceutical enterprises of different scales and their R&D investment.Methods The data of China’s pharmaceutical manufacturing industry from 1995 to 2016 were selected to establish a vector auto-regression mode,and then an impulse response function and variance decomposition were used to analyze the data.Results and Conclusion For large-scale pharmaceutical enterprises,1% increase in R&D input will lead to 0.792% increase of their main business profit.For medium-sized enterprises,if their main business profit increases by 1%,their R&D input will increase by 0.965%.Therefore,if their R&D input increases by 1%,their main business profit will increase by 1.036%.The improvement of the profitability of large-scale pharmaceutical enterprises cannot promote the increase of R&D capital investment.But the increase of profitability from medium-sized enterprises can promote the increase of R&D capital investment.As a result,the increase of R&D capital investment of large and medium-sized enterprises can improve corporate profitability.

Keywords:pharmaceutical enterprises;profitability;R&D investment;vector auto-regression (VAR) mode

Pharmaceutical manufacturing industry is one of China’s high-tech industries,and it is also a key development project for China manufacturing 2025 and strategic emerging industries.At present,innovation is the driving force in China’s relevant development planning of pharmaceutical manufacturing industry.While continuously strengthening the construction of innovation capability of pharmaceutical industry,we should improve the innovation system to promote the upgrading of pharmaceutical manufacturing industry.The integration of information technology and pharmaceutical manufacturing industry will lead the new development direction of pharmaceutical manufacturing industry[1].Apparently,China has attached great importance to the construction of an innovation system and the improvement of innovation capability of pharmaceutical manufacturing industry.At present,the problems of China’s pharmaceutical manufacturing industry are their weak innovation ability,backward basic research and transformation ability,and lack of high-quality innovation results.

One of the important factors restricting the innovation capability of China’s pharmaceutical manufacturing industry is the low investment in R&D funds.According to the statistics of national science and technology investment,the investment of science and technology expenditures in China’s pharmaceutical manufacturing industry in 2019 was only 2.55 %,while pharmaceutical industry in the developed countries used 10%–20% of sales on scientific research and development activities.For pharmaceutical enterprises,in addition to government subsidies,their own profits are also an important source of funds for R&D activities.Accordingly,the results of R&D activities will also have an important impact on corporate profits.Therefore,this paper selects China’s pharmaceutical manufacturing industry as the research object,empirically analyzes the relationship between the profitability of pharmaceutical enterprises of different scales and their R&D capital investment,which can provide theoretical and practical references for this industry to improve its profitability and innovation capability.

1 Literature review

1.1 Research on the relationship between profitability and R&D investment

Yu Xuebin[2]took China’s GEM listed companies from 2011 to 2013 as a sample to analyze the relationship between corporate innovation input and profitability.The research showed that corporate innovation investment in human resource,R&D funds and product differentiation was significantly positively correlated.Wang Yueming[3]took 131 listed manufacturing companies from 2013 to 2015 as a sample.Through an empirical analysis,he concluded that the R&D expenditure of China’s manufacturing enterprises had a positive correlation with profitability.Zhao Wen,et al[4]took pharmaceutical listed companies as the research object and used their released data to conduct an empirical analysis on the relationship between the R&D investment and their profitability.The results showed that the R&D investment of listed pharmaceutical companies and their profitability had a significant positive correlation,while R&D human investment and the company’s profitability didn’t have a significant correlation.Yuan Huili[5]took the Shenzhen Stock Exchange SME board listed companies as a sample from 2010 to 2014 to conduct an empirical study on the relationship between R&D investment and corporate profitability.The results showed that the overall R&D investment of SMEs was positively correlated with profitability.Zhang Zhanzhen[6]selected the data of 21 listed furniture companies from 2010 to 2014 as object to measure the impact of R&D investment on profitability,operating capacity,and solvency.The empirical results showed that there was a positive linear correlation between R&D investment of listed furniture companies and their profitability.Wu Jizhong and Miao Pengfei[7]selected 43 GME private equity investment funds to support listed companies as a sample.They used the panel data from 2007 to 2012 to obtain that manufacturing capabilities,human resources,profitability,and comprehensive capability of enterprises were significantly positively correlated with R&D investment.Han Xinchao and Wang Wenfei[8]used the panel data of 137 listed companies in the computer,communication and other electronic equipment manufacturing industries from 2012 to 2014 as the research object.They applied regression analysis to study the impact of R&D investment on the company’s profitability.The results showed that the R&D investment of these companies was positively related to their profitability.Zhai Ge and Bai Xiansheng[9]studied the relationship between corporate capital structure,profitability and technological innovation input of the listed companies in Shanxi and neighboring provinces.The results showed that the capital structure of listed companies was negatively correlated with the R&D investment,but profitability was positively correlated with R&D funds and technical personnel input.Wang Xiaoting[10]selected 91 small and medium board listed companies from 2010 to 2012 as the research object to test the correlation between R&D investment and corporate performance.The study found that the R&D investment of small and medium board listed companies had a significant positive correlation with corporate profitability.Wang Chen and Wang Xinhong[11]took China’s high-tech listed companies as the research object,and they used empirical research methods to analyze the correlation between the R&D investment and profitability of these companies.They found that companies profitability had a significant positive impact on R&D investment,which showed that they were mutually reinforcing.Zhang Yajing[12]took the data of pharmaceutical manufacturing enterprises from 1995 to 2015 as a sample,and empirically analyzed the relationship between profitability and R&D investment.The results showed that the improvement of the profitability of pharmaceutical manufacturing industry could increase R&D investment.But it took a long time to increase the profitability of these enterprises by increasing their R&D investment.

Some scholars have reached different conclusions.For example,Deng Yuzhou[13]conducted an empirical study on the data of listed companies,and the results showed that the financialization of asset had a positive saving effect on enterprises,but there was no inevitable positive or negative relationship between profitability and R&D investment.Luo Shaode et al.[14]analyzed the financial data of 161 manufacturing companies in the Shanghai and Shenzhen stock markets in 2006,mainly exploring the profitability of these companies,and confirmed that the profitability of these companies didn’t have a positive correlation with their R&D investment.Zhang Zichao[15]selected the panel data of China’s high-tech enterprises from 2013 to 2017 as the research object.The results showed that the R&D investment of these enterprises was negatively correlated with profitability,which would continue into the next two periods.Qin Zhimin and Huang Sheng[16]used the comparative data of Chinese and American IT companies from 2012 to 2014 to conduct an empirical research on the relationship between their R&D investment and profitability.The results showed that R&D investment of Chinese IT companies had no significant correlation with profitability.R&D investment of US IT companies was negatively correlated with their profitability.

1.2 Research on the impact of R&D investment on profitability

Chen Hongming and Xu Si[17]selected the data of listed companies on the small and medium board from 2009 to 2011 to empirically analyze the impact of R&D investment on profitability.The results showed that the R&D investment had a significant positive impact on profitability.Although this impact had a certain lag,the impact on the current period was still the most significant.Liu Guiping and Yin Zongcheng[18]took A-share listed companies in Shanghai and Shenzhen as the research object and collected relevant data for 4 consecutive years from 2011 to 2014.The results showed that the R&D investment of listed companies in Anhui Province had a significant impact on the company’s profitability.Feng Chen Feng[19]took China’s listed companies from 2004 to 2009 as the research object.Then he used panel data models to empirically analyze the impact of R&D investment on profitability.The results showed that the R&D investment of listed companies had a positive impact on the profitability for the current period.Both the lag phase one and three had a significant positive force,and the contribution of R&D investment to profit growth tended to gradually decrease with the expansion of the lag period.

1.3 Research on the impact of profitability on R&D investment

Bai Xuyun[20]took the small and medium board and GEM listed companies before the end of 2009 as the research object.He empirically analyzed the impact of the profitability of small and medium enterprises on their R&D investment,and the research proved that the effect of profitability of enterprises on the R&D funds in lag phase one and two was positive.Zhao Zhihui[21]used multiple linear regression analysis methods to conduct related research on the relationship between R&D investment and capital in the telecommunication industry.The results showed that profitability had a negative effect on R&D investment,but it was not significant.

According to the summary and analysis of the existing literature,it is found that most of the research on the relationship between profitability and R&D investment focuses on listed companies and high-tech industries,but the conclusions obtained are different.There are few studies on the relationship between the two in pharmaceutical enterprises,which is not pertinent.Based on this,this paper selects companies of different sizes in pharmaceutical manufacturing industry as the research object,and uses the vector auto-regression (VAR) mode to study the relationship between their profitability and R&D investment.

2 Research design

2.1 Research theory and model construction

2.1.1 VAR model

VAR mode is mainly used to predict the interconnected time series system and analyze the dynamic impact of random disturbances on the variable system.Based on this,it can explain the impact of economic shocks on economic variables.The general expression of a VAR mode withKtime series andPperiod lag is as follows:Yt=A1Yt-1+A2Yt-2+...+ApYt-p+εt,t=1,2,3 ...,t.In the expression,Ymainly represents theK-dimensional endogenous variable vector,andArepresents the corresponding coefficient matrix,wherePis the lag order of the endogenous variable[22].When the VAR mode is used for related research and analysis,it is necessary to apply Granger causality test,impulse response function(IRF),variance decomposition and other tools for simultaneous analysis.Among them,Granger causality can test whether all the lags of a variable will have a certain impact on the current value of another or multiple variables.And the IRF can be drawn to more comprehensively reflect the inter-variable dynamic impact.Besides,variance decomposition can decompose a variable in the VAR mode system to each perturbation item,so as to analyze the contribution of each impact to the change of endogenous variables in detail[23].

2.1.2 Constructing the VAR mode

This article uses the VAR mode in the analysis and research.Some methods such as co-integration theory,impulse response analysis,and variance decomposition are applied to study the relationship between the profitability of companies of different sizes in China’s pharmaceutical manufacturing industry and their R&D investment.In general,the form of the variables entering the VAR mode is required to be a stationary series,so the ADF unit root test is first performed on the selected time series[24].After the co-integration test is passed,because there are two variables in this article,the co-integration regression equation is established with the E-G twostep method.Through the unit root test of the residual,the equation is valid when there is no unit root.Because different lag orders will cause differences in the model estimation results,so it is necessary to determine the optimal lag order before establishing the VAR mode.When establishing the VAR mode,a stationarity test should be conducted to ensure the stability of the mode.Since VAR mode has many coefficients,it is impossible to analyze the mode by studying the estimated value of the mode coefficients.So Granger causality test,impulse response function and variance decomposition are used to study the relationship between the two variables.

2.2 Indicator selection and data sources

This article mainly uses the profitability of the company as an explanatory variable,and its principal operation profit rate (POPR) as the measurement index.The explained variable is the R&D capital investment of the company and R&D intensity (RDI)is used as the measurement index.Among them,POPR=total profit/main business income,and the RDI=R&D expenses internal expenditure/main business income.In the “China High-tech Industry Statistical Yearbook”,the missing values of the total profits and main business income of pharmaceutical manufacturing industry in 2017 and 2018 are obtained through data sorting.Therefore,this article sort out the total profit,main business income,and R&D expenses internal expenditure of large and mediumsized pharmaceutical manufacturing enterprises in China from 1995 to 2016.Finally,the time series of pharmaceutical manufacturing enterprises of different sizes are analyzed respectively.

In order to ensure that there is no heteroscedasticity in the relevant research data,all variables are processed by logarithmic,and the processed data will not change the relationship between the variables.In this paper,the logarithm of the variables is marked as:lnRDI1and lnPOPR1(large pharmaceutical manufacturing enterprises),lnRDI2and lnPOPR2(medium-sized pharmaceutical manufacturing enterprises).The relevant data are shown in Table 1 and Table 2.

Table 1 Relevant data of China’s large pharmaceutical manufacturers from 1995 to 2016

Table 2 Relevant data of China’s medium-sized pharmaceutical man ufacturersfrom 1995 to 2016

3 Empirical analysis

3.1 Unit root test

This article uses the ADF unit root test method to check the stationarity of lnRDI1,lnPOPR1,lnRDI2,and lnPOPR2in large and medium-sized pharmaceutical enterprises,and a 5% significance level is selected.The lag order is determined by the Akaike information criterion (AIC).The results of the inspection are shown in Table 3.

Table 3 Unit root test results

According to the data analysis in Table 3 above,it can be seen that when the original sequences of lnPOPR1,lnRDI1,lnPOPR2and lnRDI2are at the 5% confidence level,the absolute values of the ADF statistics are all less than the critical value.Therefore,all the original sequences have unit roots,and it is a non-stationary time series.At the same time,the first-order difference series of ΔlnPOPR1,ΔlnRDI1,ΔlnPOPR2,and ΔlnRDI2are at the 5% confidence level.The absolute values of the ADF statistics are greater than the critical value.There is no unit root in the differential sequence,which is a stationary time series.Therefore,lnPOPR1,lnRDI1,lnPOPR2,and lnRDI2are all first-order single integer sequences,which can be tested for co-integration.

3.2 Co-integration test

This article mainly uses the E-G two-step method to test the co-integration relationship between variables.A regression mode was constructed for variables such as lnPOPR1and lnRDI1of the large enterprises,lnPOPR2and lnRDI2of the medium-sized pharmaceutical enterprises.Ordinary least squares(OLS) estimation was used to estimate,and then the regression residual series were tested for their relevant stationarity.According to the regression results,the co-integration equations are listed as follows:

According to formula (1),theR2of the regression equation for large enterprises is 0.648,the adjustedR2is 0.63,and thetvalue is in parentheses,which means the equation fits well.According to formula (2),R2of medium-sized enterprises is 0.809 5,adjustedR2is 0.799 9,and DW value is close to 2,which also means the equation fits well.Then get the residual sequence of the regression equation of large enterpriseset,et=lnRDI1– 1.262 0 lnPOPR1+1.195 6,and get the residual sequence of the regression equation of medium-sized enterprisesez,ez=lnRDI2– 0.964 89 lnPOPR2+2.075 3.Then,ADF test on the residual sequence is performed respectively.The results are shown in Table 4.

Table 4 ADF test results of residual items

From the data in Table 4 above,it can be seen that the ADF test values of the residual sequence of et andezare less than the critical value of 1%significance level.Therefore,the residual sequence is relatively stable.It is believed that there is a cointegration relationship between the two variables of large and medium-sized enterprises in China’s pharmaceutical manufacturing industry.According to the co-integration equation,the following assumptions can be made.

For large pharmaceutical manufacturing enterprises:

(1) Hypothesis 1:When the main business profit rate increases by 1%,the R&D investment intensity of the enterprise will increase by 1.262%.

(2) Hypothesis 2:When the R&D investment intensity increases by 1%,the main business profit margin of the enterprise will increase by 0.792%.

For medium-sized pharmaceutica l manufacturers:

(1) Hypothesis 3:When the main business profit rate increases by 1%,the R&D investment intensity of the enterprise will increase by 0.965%.

(2) Hypothesis 4:When the R&D investment intensity increases by 1%,the main business profit margin of the enterprise will increase by 1.036%.

3.3 Establishment of VAR mode

VAR modes of different scale enterprises in China’s pharmaceutical manufacturing industry are established.Then,EVIEWS 8.0 is applied to determine the model lag orderP.Finally,the method of joint determination of multiple criteria is adopted to judge LR,FPE,AIC,SC,HQ criteria.The results are shown in Table 5.

Table 5 Selection results of VAR lag order P

According to the results of the research data in the above table,the lag order of VAR of China’s large pharmaceutical manufacturers is mainly obtained by LR,FPE,AIC,SC,and HQ.When the lag order is 1,the VAR mode is optimal.Therefore,the optimal lag order selected by a large-scale pharmaceutical enterprise is 1,that is,k=1.Based on the same principle,the optimal lag order of a medium-sized pharmaceutical enterprise is 4.

3.4 Estimation of VAR mode

By comprehensively considering the stability of the mode and the information criteria for EVIEWS 8.0 is used to estimate the relevant parameter.The results are shown in Tables 6 and 7.

Table 6 VAR mode coefficient estimation analysis (large enterprises)

Table 7 VAR mode coefficient estimation (medium-sized enterprises)

Continued Table 7

According to the research data in Table 7 above,theR-square and adjustedR-square of the two equations of large pharmaceutical enterprises are both above 0.8.Therefore,the two equations are well fitted.According to the data in Table 8,the analysis results show that theR-squares of the two equations of the medium-sized pharmaceutical enterprises are both above 0.8,and the adjustedR-squares are 0.796 and 0.876 respectively.So the two equations fit well.

3.5 Stationarity test of VAR mode

According to the above-mentioned VAR mode,the stationarity test is performed,and the test results are shown in Fig.1 and Fig.2.The spot in the figure represents the inverse of the root of the AR characteristic polynomial.

Fig.1 VAR mode stability test (large enterprises)

Fig.2 VAR mode stability test (medium-sized enterprises)

According to Fig.1 and Fig.2,the modulus of each feature and the reciprocal are in the unit circle,and there is no root outside the unit circle.Therefore,the VAR modes of large and medium-sized pharmaceutical enterprises in China have passed the stability test and can be further analyzed.

3.6 Granger causality test

The above mentioned co-integration test cannot explain whether there is a statistically causal relationship between the two variables,but it can show that there is a possibility of Granger causality between them.In order to further determine whether there is a causal relationship between the two,the Granger causality test is used to verify the causal relationship between lnPOPR1and lnRDI1,lnPOPR2and lnRDI2.The results are shown in Table 8.

Table 8 Granger causality test of large and medium-sized enterprises

For large pharmaceutical enterprises,because the original hypothesis that lnRDI is not the Granger cause of lnPOPR is rejected under the VAR mode,it can be concluded that lnRDI is the Granger cause of lnPOPR.After accepting the original hypothesis that lnPOPR is not the Granger cause of lnRDI,we can learn that lnPOPR is not the Granger reason of lnRDI.So hypothesis 1 in the above-mentioned co-integration test is not established,but hypothesis 2 is established.It means if the R&D investment intensity of China’s large pharmaceutical enterprises increases by 1%,the profit margin of main business will increase by 0.792%.

For medium-sized pharmaceutical enterprises,lnPOPR is the Granger cause of lnRDI under the VAR mode,and lnRDI is also the Granger cause of lnPOPR.Therefore,hypothesis 1 and hypothesis 2 in the above co-integration test are both valid.It means when the profit rate of the main business of China’s medium-sized pharmaceutical enterprises increases by 1%,their R&D investment intensity will increase by 0.965%.When the enterprise’s R&D investment intensity increases by 1%,their main business profit rate will increase by 1.036%.

3.7 Impulse response analysis

EVIEWS 8.0 is used to obtain the impulse response results of the main business profit rate and R&D investment intensity of large and medium-sized pharmaceutical enterprises.The number of impulse response tracking periods (unit:year) is selected to set a 10-year tracking period.The results are shown in Table 9.

Table 9 Variable impulse response analysis results

It can be seen from Table 9 that the impulse response analysis results of large pharmaceutical enterprises are as follows.

From the perspective of the response of the main business profit rate to a unit impact of R&D investment intensity,the response is 0 in the first period,and the second period (–0.000 636) begins to react negatively,and then the response value continues to decrease.The fourth period is the smallest(–0.000 858).Then it gradually increases,and the tenth period is the largest (–0.000 368).The overall trend is to first decrease and then increase.During the entire analysis period,the cumulative response value of lnPOPR1to lnRDI1is–0.005 758,which means the overall impact of lnPOPR1on lnRDI1is negative.The change of lnRDI1is not obvious,and the response value is low.This is consistent with the Granger causality that the change of lnPOPR1 cannot significantly affect the change of lnRDI1.It shows that the increase in profitability of large pharmaceutical enterprises cannot greatly affect their R&D investment.As to the response of R&D investment intensity to a unit impact of the main business profit rate,lnPOPR1immediately responds in the first period,with a response value of 0.008 530.It gradually increases,and the response value reaches the maximum in the third period(0.042 401).But the fourth period slowly drops to 0.042 071,and then gradually decreases until the tenth period gets to the smallest (0.017 632).Therefore,the overall trend is to increase first,then slowly decrease.During this period,the cumulative effect of lnRDI1on lnPOPR1is 0.292 234,which means the overall effect of lnRDI1 on lnPOPR1is positive,and the change in lnPOPR is obvious.This is consistent with the Granger causality that changes in lnRDI1can significantly affect lnPOPR1.It shows that the increase in R&D investment of large pharmaceutical enterprises can improve their profitability.

The results of impulse response analysis of medium-sized pharmaceutical enterprises are as follows.

As to the response of the main business profit rate to a unit impact of R&D investment intensity,the response is 0 in the first period,and it reaches the maximum value in the second period(0.038 565).Then it rapidly decreases in the third period to a negative value (–0.027 141).Then it gradually increases to positive (0.021 89) in the fourth period.But it gradually decreases in the fifth period(0.019 653),and stabilizes at about 0.001 in the ninth and tenth periods.The overall trend is first to increase and then decrease.During this period,the cumulative response value of lnPOPR2to lnRDI2is 0.050 601 7,which means the overall impact of lnPOPR2on lnRDI2is positive.It indicates that the increase in profitability of medium-sized pharmaceutical enterprises can promote the increase of R&D investment.As to the response of R&D investment intensity to a unit impact of the main business profit rate,lnPOPR2immediately responds in the first period with a response value of 0.006 658,and the response value reaches the maximum (0.044 766) in the second period.It first drops and then rises to the fourth period (0.042 654).It gradually decreases to the eighth period,which is the smallest (–0.002 844).Finally,it rises and stabilizes at about 0.02.The overall trend is to increase first and then decrease.During this period,the cumulative impact of lnRDI2on lnPOPR2is 0.150 647.It means the overall impact of lnRDI2on lnPOPR2is positive,indicating that the increase in R&D investment of medium-sized pharmaceutical enterprises can improve their profitability.

3.8 Variance decomposition

The full name of variance decomposition is prediction error variance decomposition.EVIEWS 8.0 is used to perform variance decomposition to predict the future trends of R&D investment intensity and the main business profit rates of large and medium-sized pharmaceutical enterprises in China.The specific results are shown in Tables 10 and 11.

Table 10 Variable variance decomposition results (large enterprises)

Table 11 Variable variance decomposition results (medium-sized enterprises)

It can be seen from Table 10 that the profit margin of the main business of large pharmaceutical enterprises in the first phase is affected by 98.91% of their volatility,but the impact of their disturbances continues to decline.The impact of R&D investment intensity on the disturbance of the profit margin of the main business exists in the first period.It continues to increase until the tenth period it reaches the peak.Therefore,about 52.84% of the variance of the main business profit rate can be explained by the change in the R&D input intensity.However,the disturbance of the first period main business profit rate to the R&D input intensity is 0.Afterwards,it continues to increase,but the impact is little and remains stable at around 0.012%.

It can be seen from Table 11 that the profit margin of the main business of medium-sized pharmaceutical enterprises in the first phase is affected by 98.6% of their fluctuations,but the impact of their disturbances gradually decreases.The impact of R&D investment intensity on the profit margin of the main business is significant from the second period,and it reaches the peak in the seventh period.Therefore,about 58.698% of the variance of the main business profit margin can be explained by the change in the R&D investment intensity.In the first period,R&D investment intensity is totally affected by the impact of its fluctuation,and then its disturbance.Gradually it decreases,but the impact of the main business profit rate on the intensity of R&D investment increases.It reaches a peak in the fifth period,and it is about 18.51%.Then it slowly decreases and stabilizes at about 18%.

4 Results and discussion

Based on the “China High-tech Industry Statistical Yearbook” during 1995–2016,the main business income,total profit,and internal expenditure of R&D funds of large and mediumsized pharmaceutical enterprises in China are systematically studied through co-integration test and VAR mode.The relationship between the profitability of enterprises of different sizes in the pharmaceutical manufacturing industry and the investment in R&D funds is as follows.

(1) There is a co-integration relationship between the main business profit rate of China’s large and medium-sized pharmaceutical enterprises and their R&D investment intensity,which is a long-term stable equilibrium relationship.Through Granger causality test,impulse response function analysis,and variance decomposition methods respectively,the assumptions of the co-integration analysis of large and mediumsized pharmaceutical enterprises are assessed.When the R&D investment intensity of large pharmaceutical enterprises increases by 1%,their main business profit rate will increase by 0.792%.However,the increase in the profit rate of the main business of large pharmaceutical enterprises cannot increase the R&D investment intensity.When the profit rate of the main business of medium-sized pharmaceutical enterprises increases by 1%,their R&D investment intensity will increase by 0.965%.Similarly,every 1% of increase in R&D investment intensity will increase the profit margin of its main business by 1.036%.

(2) Impulse response analysis finds that within the predicted 10 years,the R&D investment intensity of large and medium-sized pharmaceutical enterprises has a positive effect on their profitability of the main business,reaching the peak in the lag 3 and 2 phases.But then the R&D investment intensity has less driving effect on the profit rate of the main business.This may be related to the technological innovation mode of introducing advanced technology,imitating,digesting and improving by large and medium-sized pharmaceutical enterprises,resulting in a short time effect of R&D capital investment.The profit margin of medium-sized pharmaceutical enterprises has a positive effect on the R&D investment intensity,reaching a peak in the lag 2 phrase,and then this positive driving effect will gradually decrease.This may be because after promoting the profitability of most medium-sized pharmaceutical enterprises,their enthusiasm for research and development activities gradually decreases,thus reducing the driving effect.

(3) According to variance decomposition analysis,the variance of above 50% of the main business profit rate of large pharmaceutical enterprises can be attributed to their R&D input intensity.But the main business profit rate has a low impact and contribution to the R&D input intensity.The impact of large-scale pharmaceutical enterprises R&D investment on profitability is far greater than the impact of profitability on R&D investment.The R&D investment intensity of medium-sized pharmaceutical enterprises can attribute to more than 58% of the variance of the main business profit rate.But the main business profit rate can only attribute to about 18%of variance of R&D investment intensity.It indicates that the R&D investment and profitability of mediumsized pharmaceutical enterprises influence each other.In addition to profitability,there are other factors that affect the R&D investment of enterprises.

5 Conclusions and recommendations

5.1 Conclusions

(1) The increase in R&D capital investment of large-scale pharmaceutical enterprises can improve their profitability.Conversely,the increase in profitability cannot promote their R&D capital investment.Meanwhile,the increase in the profitability of medium-sized pharmaceutical enterprises can promote their R&D capital investment,and the increase in R&D capital investment also can improve their profitability.

(2) The R&D investment of China’s large pharmaceutical enterprises has a lag in the improvement of their profitability,and the lag time is generally about 3 years.However,the profitability of medium-sized pharmaceutical enterprises has a lagging effect on the positive promotion of their R&D capital investment,and the lag time is 2 years.Besides,the positive promotion of R&D capital investment on profitability has a lagging time,which is 2 years.Investment and profitability influence each other.In addition to profitability,there are other factors that affect R&D investment.

5.2 Suggestions

(1) Strengthening the awareness of technological innovation and increasing R&D investment.As a high-tech industry,pharmaceutical enterprises should be aware that the innovation and the accumulation of technological capital can improve their competitiveness.However,due to the coexistence of benefits and risks in R&D activities,pharmaceutical enterprises should view R&D investment from a longterm development perspective.If they want to remain invincible in the fierce competition,they must depend on their innovation,and increase R&D investment.This article proves that R&D investment of large and medium-sized enterprises has a significant positive effect on profitability in pharmaceutical manufacturing industry through an empirical research,which provides a reference for pharmaceutical enterprises to make investment decisions.

(2) Improving the technological innovation mechanism.From the results of co-integration test and impulse response analysis,the role of R&D capital investment for China’s large and mediumsized pharmaceutical enterprises can promote their profitability for 2 to 3 years,and then it begins to decrease.This may be because the technological innovation mode that enterprises introduce,imitate,assimilate and absorb the advance science and technology.Compared with the technological innovation in developed countries,such an innovation mode is not the true innovation.It is just another extension of advanced technology development.This causes a shorter output cycle of enterprises R&D capital investment.Besides,this model is not very conducive to the long-term development of pharmaceutical enterprises.Therefore,pharmaceutical enterprises should continuously improve and perfect their innovation mechanisms,enhance their innovation capabilities for a long-term development.

(3) Enhancing profitability and using the profits for R&D activities.Through an empirical study,it can be seen that the increase in profitability of mediumsized pharmaceutical enterprises has a positive effect on the investment of R&D funds.Therefore,while improving profitability,medium-sized pharmaceutical enterprises should invest a certain proportion of their profits into the R&D activities.Once the overall profitability of the enterprise is improved,the economic benefits will be generated[19].