Wang Yuzhuo,Wang Su,Chen Yuwen

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

Abstract Objective To explore the relationship between salary incentives for R&D personnel and R&D investment in China’s pharmaceutical industry through an empirical analysis so as to propose suggestions for increasing more R&D investment from the perspective of stimulating R&D personnel.Methods Based on the relevant data of the China’s pharmaceutical industry in the “China High-tech Industry Statistical Yearbook” and other data from the National Bureau of Statistics from 1995 to 2018,co-integration tests was used to construct error correction models and Granger causality tests to explore the relationship between R&D personnel salary incentives and R&D investment.Results and Conclusion There is a long-term balanced relationship between the salary level of R&D personnel in China’s pharmaceutical industry and the investment of R&D.For the per capita salary of R&D personnel increases by 1 unit,the internal expenditure of R&D investment will increase by 2.245 1 units.In the short term,the per capita salary of R&D personnel has a slight negative impact on the internal expenditure of R&D investment.In addition,the salary incentives for R&D personnel have the most significant role in promoting R&D investment after five years of implementation.

Keywords:pharmaceutical industry;R&D investment;salary incentives for R&D personnel;cointegration test

The pharmaceutical industry is a high-tech industry and it is most closely related to the health of people.Its degree of development can directly affect the quality of life and public health.The global outbreak of COVID-19 in early 2020 also confirms the importance of developing pharmaceutical industry.In recent years,with the support of national policies,the pharmaceutical industry in China has developed rapidly.According to the “China High-tech Industry Statistical Yearbook”,the total profit of China’s pharmaceutical industry was 5.148 billion yuan in 1995 and 318.7 billion yuan in 2018,with a compound annual growth rate of 19.65%[1].However,as the world’s second largest pharmaceutical market,China’s pharmaceutical industry is still dominated by generic drugs and active pharmaceutical ingredients(APIs) production,its innovation capability is insufficient.In addition to the weak foundation and late start,the lack of R&D investment is also an important factor restricting the improvement of innovation ability of China’s pharmaceutical industry.According to the annual report of Roche pharmaceutical in 2018,its R&D investment in 2018 was 10.98 billion US dollars,while that of the whole pharmaceutical industry in China was only 10.23 billion US dollars in the same year.The R&D investment level in pharmaceutical industry in China is low,which is far behind the foreign countries.Therefore,it is of great significance to study the influencing factors of R&D investment in China’s pharmaceutical industry to enhance its innovation ability.Pharmaceutical R&D personnel are the main body of scientific research innovation,and their work determines the degrees of innovation.The existing research shows that stimulating R&D personnel can promote the output of scientific research achievements[2].The increase of profit brought by the scientific research output can effectively promote the investment of R&D[3,4].Therefore,in theory,stimulating R&D personnel can promote R&D investment.But due to the intermediary effect and the uncertainty of objective environment,it is not known whether the incentives of R&D personnel can promote R&D investment.Therefore,this paper takes China’s pharmaceutical industry as the research object to explore the relationship between R&D personnel salary incentives and R&D investment.If the empirical results show that salary incentives for R&D personnel have a positive impact on R&D investment,then pharmaceutical enterprises can be proposed to increase their R&D investment from the perspective of salary incentives for R&D personnel.On the one hand,salary incentives can stimulate the enthusiasm of R&D personnel,retaining and attracting more talents to ensure the stable development of pharmaceutical enterprises.On the other hand,the role of salary incentives on R&D investment can be used to ensure the supply of innovation funds and improve the innovation capabilities of pharmaceutical enterprises.

1 Literature review

At present,the research on the influencing factors of R&D investment is mainly from the macro aspects of industry scale,competition degree and tax preference.Meanwhile,the number of researches on the relationship between R&D personnel salary incentives and R&D investment are still small.Therefore,this paper studies the relationship between R&D personnel salary incentives and R&D investment to enrich the existing research and put forward some suggestions for the innovation and development of pharmaceutical enterprises.

1.1 The impact of R&D personnel salary incentives on R&D investment

Cao Qinrun (2010) made an empirical study on the relationship between the changes of R&D personnel and R&D expenditure based on the research samples of listed enterprises in China from 2004 to 2007,pointing out that the proportion of R&D personnel’s salary had an impact on the R&D expenditure[5].Su Taoyong (2019),taking A-share listed enterprises as the research object,used the method of 3SLS regression to confirm that there was a significant correlation between the per capita salary and welfare and enterprise R&D investment[6].

Wang Xiaoli (2011) pointed out that there was a positive correlation between human capital and research investment,and suggested that incentive theory and performance appraisal should be utilized to reasonably allocate the salary of scientific research personnel[7].Research by Luo Laijun (2012) showed that low salaries of R&D personnel could lead to insufficient R&D investment and innovation[8].Fang Aiping (2012) calculated the research and development input efficiency of different provinces from 2008 to 2011 through the annual data envelopment analysis (DEA),pointing out that the per capita salary of personnel in scientific research generally has a positive impact on the research and development input efficiency[9].Yao Cheng(2018) selected 1 230 enterprises in the intensive manufacturing industry as the research sample to make an empirical study,which showed that the research and development intensity could promote the per capita salary of employees.In addition,a better proportion of employees with master’s degree or above and per capita salary could improve the R&D capability of enterprises[10].

Yan Qiang (2014) used the censored quantile regression to make an empirical study on the selected micro statistical data of large and medium-sized industrial enterprises in Shanghai.It showed that the regression coefficient of per capita salary at the middle and high quantile was negative,and the t-statistic was significantly improved.This means that with the increase of per capita salary,the innovation expenditure will decrease[11].

1.2 The impact of R&D investment on R&D personnel salary incentives

Deng Jun (2011) selected the data of 16 manufacturing industries in China from 1995 to 2007,and made an analysis by using the dynamic panel data model.The conclusion was that in order to obtain competitive advantages,enterprises increase research and development activities,which led to the demand for highly skilled labor.In turn,it resulted in the increase if salaries for high skilled labor[12].Gao Yanhui (2013) used the data of China’s high-tech industry from 1998 to 2008 to conduct an empirical study.The results showed that as one of the sources of enterprise R&D funds,government research and development subsidies had a more significant positive impact on the salary of scientific and technological personnel in state-owned enterprises[13].

Huang Xinfei (2005) found that the R&D investment of multinational enterprises was inversely proportional to the salary of R&D personnel[14].

Through the literature review,it is found that most of the research is based on the overall manufacturing industry,and the conclusions of scholars on the relationship between R&D personnel salary incentives and R&D investment are quite different.Therefore,this paper takes the pharmaceutical industry as the research object to empirically analyze the relationship between R&D personnel salary incentives and R&D investment,and put forward some suggestions for improving the level of R&D investment and innovation ability of pharmaceutical enterprises.

2 Research design

2.1 Data source and index selection

2.1.1 Data source

This paper takes the pharmaceutical industry in China as the research object,and studies the relationship between R&D personnel salary incentives and R&D investment.The research time span is from 1995 to 2018,a total of 24 years.The data related to this article are from the “China High-tech Industry Statistical Yearbook”[1]and the National Bureau of Statistics of the corresponding year.Excel and Eviews10 were used to make the analysis of data.

2.1.2 Index selection and description

(1) The explained variable includes R&D capital investment.It is measured by internal expenditure of R&D funds asRD(10 000 yuan).The internal expenditure of R&D investment is not only an important part of enterprises’ innovation activities,but also an important guarantee to improve the innovation ability of the enterprises.(2) The explanatory variable includes R&D personnel salary incentives,which refers to the calculation method of relevant scholars[5,10,13].It is measured by the per capita salary of R&D personnel,and the calculation method is personnel labor cost/full-time equivalent of R&D personnel (10 000 yuan/person year),which is recorded asPay.The salary level of R&D personnel reflects the importance that enterprises attach to R&D personnel,and the incentive mechanism and policies of enterprises.According to the efficiency salary theory,a high salary level can make workers cherish their job opportunities,thus improving their working enthusiasm and output efficiency.Therefore,it may reduce the cost of job conversion and the brain drain within the enterprise.Besides,it will attract excellent job seekers[15].High salary level is of great significance to improve the output efficiency of personnel and the innovation ability of enterprises.

In order to avoid heteroscedasticity of data,the original data was processed by logarithmic in this paper.Therefore,the correlation between variables does not change after processing.The processed data are recorded as lnRDand lnPayrespectively.

2.2 Research methods

2.2.1 Cointegration test

At present,there are two main methods for cointegration test of data:Engle-Granger twostep method and Johansen cointegration test.The Johansen cointegration test is mainly used to test the cointegration of multiple variables.Meanwhile,the Engle-Granger two-step method is for two variables.So the Engle-Granger two-step method is used for cointegration test in this paper.The main analysis process is shown in the following formula.

In the first step,ordinary least squares are used to estimate and calculate the unbalanced error:

It is called co-integrating or static regression[16].

The second step is to check the single integrity ofet.Ifetis a stable sequence I (0),then the variablesYtandXtare (1,1) order cointegration.Otherwise,the variablesYtandXtare considered to have no cointegration relationship.ADF test is adopted as the test method,and the intercept and time trend items are usually no longer needed in the test model,namely model (2):

In the test,if the null hypothesisH0:θ=0 is rejected,the residual termetwill be a stationary series,which indicates thatXandYare cointegrated[16].

2.2.2 Error correction model (ECM)

ECM is an econometric model with a specific form.Its main form is proposed by Davidson,Hendry,Srba and Yeo in 1978,which is also called the DHSY model[17].It can avoid false regression problems,eliminating possible multicollinearity problems in the model.In addition,it can be estimated by classic regression methods.The establishment of ECM is based on Granger’s formulation theorem:

Among them,ECMt-1=y t-1-b0-b1x t-1represents the unbalanced error in periodt-1,whileλ·ECM t-1represents the error correction term.λis called the correction coefficient,which shows the adjustment speed of the error correction term for △Yt.According to model 3,the value of △Ytdepends on the value of △Xand the previous error correction term.

2.2.3 Granger causality test

In the case of time series,the Granger causality between two economic variablesXandYis defined as the following.If the past information of the variablesXandYis included,the prediction effect of the comparative variableYis better than that of the onlyYfrom the past information,that is,the variableXhelps explain the future changes of the variableY.And then,the variableXis considered as the Granger cause of the variableY[16].

3 An empirical research

3.1 Descriptive statistical analysis

Table 1 shows the relevant data of internal expenditure of R&D funds and per capita salary of R&D personnel in China’s pharmaceutical industry from 1995 to 2018.Although the internal expenditure of R&D funds and the per capita salary of R&D personnel fluctuated slightly,both of them increased year by year.

Table 1 Related data of internal expenditure of R&D funds (RD) and per capita salary of R&D personnel(Pay) in China’s pharmaceutical industry from 1995 to 2018

3.2 Stationarity test

In this paper,ADF unit root test is used toexamine the stationarity of lnRDand lnPay,and the results of unit root test are shown in Table 2.

Table 2 ADF unit root test results

It can be seen from Table 2 that both the original sequence of lnRDand lnPayaccept the null hypothesis of the existence of unit roots at the significance level of 5%,indicating that the original sequence of variables is not stable.The first-order difference sequence △lnRDand △lnPayreject the null hypothesis that there is a unit root at the significance level of 5%.lnRDand lnPayare first-order singleintegration sequences,and they have the conditions for cointegration test.

3.3 Cointegration test

3.3.1 Ordinary least squares regression

An ordinary least square estimation (OLS) is carried out on lnRDand lnPay,and the estimation results are shown in Table 3.

Table 3 OLS estimation results

The regression equation of the two is obtained from Table 3:

The determination coefficient and the adjustment determination coefficient of equation (4) are both greater than 0.7,and theF-statistic is 58.10,indicating that the regression equation has good fitting degree and significance.

3.3.2 Unit root test of residual sequence

The residual sequence generated by equation(4) is namedECM,and its stationarity test is carried out to determine whether the long-term cointegration relationship between lnRDand lnPayis stable.The test results are shown in Table 4.

Table 4 ADF unit root test results of residual sequence

It can be seen from Table 4 that the residual sequence rejects the null hypothesis that there is a unit root at the significance level of 5%,indicating that there is a long-term stable co-integration relationship between lnRDand lnPay.From equation 4,the longterm elasticity coefficient of lnPayto lnRDis 2.245 1.From a long-term perspective,if the per capita salary of R&D personnel in China’s pharmaceutical industry increases by 1 unit,internal expenditure of R&D funds will increase by 2.245 1 units.The increase in per capita salary of R&D personnel can significantly promote the increase in internal expenditure of R&D funds.

3.4 Establishing an ECM

The cointegration test shows that there is a longterm stable cointegration relationship between the per capita salary of R&D personnel and the internal expenditure of R&D funds in China’s pharmaceutical industry.However,the short-term fluctuation between them needs to be further verified and analyzed.Therefore,an error correction model is established.The results of the model are shown in Table 5.

Table 5 ECM results

The error correction equation is:

It can be seen from equation (5) that the estimated value of the error correction term coefficient is–0.034 3.When the short-term fluctuations of lnRDand lnPaydeviate from the long-term equilibrium,the unbalanced errors of the previous year will be corrected reversely to the value of lnRDin this year with the strength of 0.034 3,bringing it back to equilibrium.The ECM model reflects the dynamic equilibrium mechanism between the per capita salary of R&D personnel and the growth of internal expenditure of R&D funds.In addition,the short-term elasticity coefficient of lnRDto lnPayis–0.218 4.In the short term,every increase of 1 unit in the per capita salary of R&D personnel will reduce its internal expenditure of R&D funds by 0.218 4 units.

3.5 Granger causality test

From the above analysis,we can see that there is a cointegration relationship between internal expenditure of R&D funds (lnRD) and per capita salary of R&D personnel (lnPay).But whether there is a causal relationship between the two and the extent of the impact is still unknown.Therefore,the Granger causality test is performed on lnRDand lnPay.Since the lag order has a great influence on the Granger causality test,this paper makes the Granger causality test for the lag order 1–6,respectively.The results are shown in Table 6.

Table 6 Granger causality test results

Table 6 shows that when the lag order is 1 to 4,the assumptions that “lnRDis not the Granger cause of lnPay” and “lnPayis not Granger cause of lnRD” are accepted at the confidence level of 5%.It means that there is no Granger causality between lnPayand lnRD.When the lag order is 5 and 6,the hypothesis that“lnPayis not Granger cause of lnRD” is rejected at the confidence level of 5%,but “lnRDis not Granger cause of lnPay” is accepted.Therefore,lnPayis the Granger cause of lnRD,but lnRDis not the Granger cause of lnPay.

4 Results and discussion

4.1 Results

In this paper,after conducting a co-integration test,constructing an error correction model and carrying out Granger causality test on the per capita salary of R&D personnel (lnPay) in China’s pharmaceutical industry and internal expenditure of R&D funds (lnRD),the following results are obtained.

(1) There is a long-term equilibrium relationship between the per capita salary of R&D personnel and the internal expenditure of R&D funds in China’s pharmaceutical industry,and the long-term elasticity coefficient is 2.245 1.In the long run,R&D per capita salary increases by one unit,internal expenditure of R&D funds will increase by 2.245 1 units.

(2) In the short term,the deviation of longterm equilibrium relationship between the per capita salary of R&D personnel and the internal expenditure of R&D funds will be corrected by 3.4% in the next period,and the short-term elasticity coefficient is–0.218 4.If R&D per capita salary of R&D personnel increases by one unit in China’s pharmaceutical industry,the internal expenditure of R&D funds will decrease by 0.218 4 units,indicating that the increase of R&D personnel’s per capita salary has a negative impact on promoting the internal expenditure of R&D funds in the short term.

(3) When the lag order is 1–4,there is no Granger relationship between the R&D personnel’s per capita salary and the internal expenditure of R&D funds.When the lag orders are 5 and 6,the per capita salary of R&D personnel is the one-way Granger cause of internal expenditure of R&D funds.The per capita salary of R&D personnel has a lagging effect on the promotion of internal expenditure of R&D funds.After five years of salary incentives for R&D personnel,its role in promoting R&D investment is most significant.Granger causality test results also confirm that the per capita salary of R&D personnel has a significant effect on the internal expenditure of R&D funds in the long term,but it has a weak inhibitory effect in the short term.

4.2 Discussion

The empirical analysis results show that the longterm and short-term effects of the per capita salary of R&D personnel in China’s pharmaceutical industry on internal expenditure of R&D funds are different,and there is a lag effect.This section will discuss the empirical results.

(1) In the short term,salary incentives for R&D personnel in the pharmaceutical industry have a weak inhibitory effect on R&D investment.As we all know,drug R&D has the characteristics of longterm,high investment and high risk.It usually takes an average of 10 to 15 years from the start of research and development to the successful launch of a new drug,and the average cost is more than 800 million US dollars[18].Even if the salary incentives for R&D personnel stimulate their enthusiasm and speed up their results,it is difficult to shorten the drug research and development cycle by more than half.Therefore,some short-sighted enterprises believe that their incentives have not produced the desired results and will weaken their salary incentives for R&D personnel and R&D investment.Secondly,in order to avoid the high risk of drug research and development,even if the salary incentives for R&D personnel accelerates the output of research and development results,some enterprises still reduce R&D investment to avoid risks when they do not see the economic benefits brought by research and development achievements.

(2) In the long term,salary incentives for R&D personnel in the pharmaceutical industry will significantly promote R&D investment.Increasing the per capita salary of R&D personnel can effectively stimulate their work enthusiasm,improve their output efficiency,and shorten the drug research and development cycle.Although increasing the salary of R&D personnel can only shorten the time for drug research and development to a certain extent,the products of enterprises can seize market opportunities and obtain greater profits than those of enterprises without incentive policies.At the same time,shortening the drug research and development cycle can save enterprises a lot of R&D funds for follow-up research.After making huge profits,enterprises will be more willing to increase R&D funds to obtain more profits.Second,the per capita salary of R&D personnel reflects the labor cost of R&D personnel.The increase in salary of R&D personnel has prompted enterprises to strengthen R&D investment and technological innovation,which in turn can promote the optimization and upgrading of industrial structure[19].

(3) Salary incentives for R&D personnel have a lag effect on the promotion of R&D investment.Due to the long-term characteristics of drug research and development,salary incentives can stimulate the enthusiasm of research and development personnel and accelerate the speed of new drugs on the market,but it does not mean the new drugs can make profits in one or two years.Enterprises can continue to invest more in R&D funds only after gaining profits from research and development results.Therefore,salary incentives for R&D personnel have a lag effect on the promotion of R&D investment and the effect is more significant in the long term.

5 Conclusions and recommendations

5.1 Conclusions

In the short term,the salary incentives for R&D personnel in China’s pharmaceutical industry have a weak inhibitory effect on R&D investment.However,in the long-term,R&D personnel salary incentives have a significant positive effect on R&D investment.It has the strongest role in promoting R&D investment after 5 and 6 years of salary incentives for R&D personnel.

5.2 Recommendations

Pharmaceutical enterprises should make good use of the relationship between R&D personnel salary incentives and R&D investment to strengthen their salary incentives policy.On the one hand,it can enhance the research and development enthusiasm of R&D personnel,attract and retain high-level talents,and ensure the stable and good development of the enterprise.On the other hand,it can promote enterprises to increase investment in research and development to ensure the innovation capabilities of enterprises and the entire industry.

5.2.1 Establishing a diversified incentive system

Pharmaceutical enterprises’ incentives for R&D personnel should not be limited to salary and research funds,but performance salary and subsidies can also be used as means to motivate them to innovate.Then,enterprises can reap the dividends of research and development innovation,which in turn,will increase investment in research and innovation.

5.2.2 Establishing a dynamic adjustment mechanism for the salary payment of R&D personnel

The problem of uneven development in China is still serious.At the same time,due to inflation,the salary of R&D personnel should be dynamically adjusted.In different regions,pharmaceutical enterprises should pay R&D personnel based on regional consumption levels and the difficulty of research and development projects.

5.2.3 Improving the welfare of R&D personnel

Pharmaceutical enterprises can provide reasonable welfare to core R&D talents in terms of housing loans and children’s education expenses.This means they can strengthen the role of incentive for R&D personnel.In addition,the state and government can introduce corresponding policies to appropriately reduce personal income tax on R&D personnel in the pharmaceutical industry.The personal tax on patent income,labor income and copyright income should be reduced or exempted within a reasonable range[20].

5.2.4 Standardizing the salary incentive mechanism and improving the evaluation system

Pharmaceutical enterprises should improve the evaluation system to enable high-level R&D personnel to get more salaries,while low-level R&D personnel will gradually step out under such a system.On the other hand,it also avoids waste of resources.In addition,enterprises can also give excellent R&D personnel more opportunities for exchanges and advanced studies to stimulate their enthusiasm for research and development and innovation in the pharmaceutical industry.

5.2.5 Enhancing the importance and training of R&D personnel by pharmaceutical enterprises

(1) R&D personnel are the main body of pharmaceutical enterprises’ research and development activities and they are the guarantees of research and development innovation capabilities.Excellent R&D personnel contributes greatly to the output of high-quality research and development results,thereby improving the innovation capabilities of pharmaceutical enterprises.Therefore,pharmaceutical enterprises should consciously attach importance to R&D personnel with more materials and spiritual rewards.(2) The training of R&D personnel is a long term process.Therefore,pharmaceutical enterprises should insist on cultivating R&D personnel and reserve enough talents so that their innovation activities and innovation capabilities can be continuously improved.