Firm Size, Compensation and Firm Deaths in SMEs: A Evidence from America

BRITISH JOURNAL OF INTERDISCIPLINARY RESEARCH © 2018 OPEN ARCHIVES INITIATIVE | Volume 8| Issue 1 | ISSN: 2308-3218 Firm Size, Compensation and Firm Deaths in SMEs: Evidence from America 1 Daniel Quacoe 1, Xuezhou Wen 1, Dinah Quacoe 1, Ann Dodor1, Isaac Asare Bediako 1 School of Management, Jiangsu University, 301 Xuefu Road, Zhenjiang, Jiangsu, P.R. China Corresponding Author: Daniel Quacoe ABSTRACT This paper mainly investigates how SME compensation affects their collapse and renders them unsustainable. We utilize annual time series data provided by America’s Small Business Administration from 1988 to 2014. We decompose SMEs into three size categories (micro, small and medium) and analyse the impact of compensation on the collapse of each of these sub-divisions to capture the extent this might help inform policy. Our ARDL Bounds Test shows that compensation and firm death have long-run relationship. Through VECM Granger causality analysis, we find a unidirectional causality running from compensation to firm death indicating that the compensation system to workers of SMEs significantly contributes to their sustainability. Keywords: Firm size, Compensation, Firm Deaths, Sustainability, America . INTRODUCTION Some key factors are broadly known to greatly influence the survival and sustainability of SMEs. One of such broad influence factors is firm-specific factors, regularly mentioned are the age and size of firms (Box 2008). Most SMEs are young firms and they are known to suffer liability of newness (Freeman, Carroll et al. 1983). SMEs are also known to be small firms so they suffer the liability of smallness hence face higher risk of death (Carroll and Hannan 2000). One reason behind the risk of liabilities of newness and smallness is that they negatively influence access to finance (Carreira and Silva 2010). One firmspecific factor necessary to abate the risks of the liabilities of newness and smallness of SMEs is the quality of human resource that can turn the fortunes of the firm around and be able to withstand all risks of firm collapse. According to (Gomes and Kuehn 2017), a more educated work force raises firm size and there exist correlation between a small-business’ ability to hire and ability to get financing (Peek 2013). In addition, (Bailey 1993) argues that human resources are frequently "underutilized" because employees often perform below their maximum potential and that organizational efforts to elicit discretionary effort from employees are likely to provide returns in excess of any relevant costs. One critical firm-specific influence factor on the quality and effectiveness of human capital is Compensation (Gupta and Shaw 2014). Compensation influences the quality of the people who are employed, the motivation and performance level of the employees, and the quality of those workers who stay with the company (Shaw and Gupta 2007, Dineen and Williamson 2012). It is evident that compensation has powerful incentive and sorting effects (Gerhart and Rynes 2003) despite contrary arguments by (Pfeffer 1998). In every part of organizational execution, compensation can shape employee behavior and their effectiveness. In the view of Gupta and Shaw, compensation systems not only influences employee motivation, but also can be used to improve safety, quality, creativity, innovation and many other outcomes vital in a successful workplace (Gupta and Shaw 2014). Despite the importance of employees compensation to the sustainability of firms especially SMEs, literature is only rife with research on CEO compensation (Duong and Evans 2015, Olaniyi, Obembe et al. 2017), executive compensation (Lahlou and Navatte 2017, Yarram and Rice 2017), BJIR105-384-3 |Received: 08 April 2018 | Accepted: 05 May 2018 | January-December-2018 [(8)1: 010-019] Board compensation (Liu and Stark 2009, Müller 2014, Dah and Frye 2017) but completely silent on employee compensation (Gupta and Shaw 2014). According to a meta-analysis of 40 years of research on financial incentives and performance (Jenkins Jr, Mitra et al. 1998) yielded about one per year employee compensation study and the situation has not improved in recent years (Gupta and Shaw 2014). This study takes a step towards improving the situation. The main contribution of this paper is to investigate the extent to which SME’s inadequate employee compensation contributes to their collapse or sustainability. This paper seeks to inform government, corporate and other institutional policies on how best compensation package could be used to attract the best talents to surmount the challenges bedevilling the SME sector. The paper also seeks to draw the attention and interest of the research community to the importance of SME employee compensation. To the best of our knowledge, no such research has been conducted. Effects of Firm Size on SMEs One of the most important factors influencing the risk of SMEs death is their size. The liability of smallness (Aldrich and Auster 1986) states that small firms are more likely to collapse than their larger counterparts. One important reason that accounts for this relationship is that small firms are more likely to face financial constraints (Carreira and Silva 2010). To raise their size and productivity, a more educated work force is necessary but this requires competitive wage (Gomes and Kuehn 2017). Furthermore, smaller firm size means smaller scale and therefore face cost disadvantages compared to larger firms (Audretsch and Mahmood 1994). Effects of Compensation on SMEs Because of surge in competitions, organizations seek to be more efficient and effective to get competitive edge. This means SMEs must accomplish more with fewer employees and this calls for effective management of human resources. Typically, employee compensation system plays a major role in order to better manage human resources. Employee compensation plays such a key role because it is at the heart of the employment relationship, being critically important to both employees and employers. In addition, compensation decisions influence the employer's ability to compete for employees in the labour market (attract and retain), as well as their attitudes and behaviours whilst with the employer (Gerhart and Rynes 2003). Sustainability in SMEs Individual SME’s contribution towards sustainable development is small but collectively, SMEs have a very large impact on the development of a specific geographic area. The more presence of SMEs in the economy of a particular area, the more important is the SMEs’ role for achieving sustainability (A. 1993). But for SMEs to help achieve sustainability, their business must not collapse but be sustainable into the future (Brundtland 1987). Brundtland report emphasizes economic development as a key component in sustainable development. Theoretical Background This study draws direction from the combined effects of Expectancy, Reinforcement, Equity and Agency theories. Expectancy theory focuses on the link between rewards and behaviours, although it emphasizes expected (rather than experienced) rewards. The implication of this theory in line with the well-established SME’s liability of smallness is that employees of high quality may elude most SMEs since they will be unable to compete favourable in the labour market because most SMEs will be unable to offer the competitive compensation package for the best talents in the market. Even when SMEs have managed to get the required talents, Reinforcement theory suggests that they may require higher compensation to be retained. Equity theory suggests that employee perceptions of what they contribute to the organization, what they get in return, and how their return-contribution ratio compares to others inside and outside the organization determines how fair they perceive their employment relationship Adams (Adams 1963). Where the comparison is not favourable as compared to larger firms, multinational corporations, world bodies and governmental institutions etc. employees may take actions such as quitting or lack of cooperation, thievery (Greenberg 1990) which will eventually contribute to the collapse of the firm. Similarly, according to Agency theory (Fama and Jensen 1983), managers and owners may not be able to align the interest of the SME to that of the employees and this will be detrimental to the sustainability of the firm. Data We make use of a rich data set provided by the U.S. Small Business Administration (SBA) 1 . SBA is a 1 Small Business Administration known as “SBA” was created in 1953 as an independent agency of the federal government to aid, counsel, assist and protect the interests of small businesses in the US. For more details: http://www.sba.gov/. Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index United States government agency that provides support to entrepreneurs and small businesses. According to the SBA, its mission is to maintain and strengthen the US economy by enabling the establishment and viability of small businesses and by assisting in the economic recovery of communities after disasters. The large data set spans from 1988 to 2014. The data describes employment size of firms that died from 1988 to 2014. It also includes a data set on the payroll of employment by size from 1988 to 2014. The data groups size of firm per the number of persons employed from 1-4, 5-9, 10-19, 20-49, 50-99, 100-249, 250-499 range of employment. We define a firm as ‘micro’ if it has less than 20 employees; ‘small’ if it has 20 to 99 employees; and ‘medium’ if it has 100 to 500 employees. We measure firm size by the number of employees in the firm and measure firm death by the number of employees those firms shed within the period of the study. We measure compensation by the yearly payroll of firm per the number of employees in the firm. METHODS We use econometrics model to analyse the data. Non-Stationary and Stationary Tests Our data covers 26 years and because different years have unique disturbances, the long-period time series data may experience random drift. For this reason, we first test non-stationary of our series by instituting autoregressive time series (Dickey and Fuller 1981), The Phillips-Perron (PP) unit root test (Phillips and Perron 1988) and we follow this with stationary test (Kwiatkowski, Phillips et al. 1992). Augmented Dickey-Fuller test is one of the best known and most widely used unit root tests. Formulation: …(1)   p p p yt     t   yt 1   i yt i 1   t where  = - 1   bi  and i  - b j i 1 i 1 j 1   We use the generalized ADF equation that allows for deterministic specifications such as intercept, time trend and a number of lags up to p. In conducting the ADF tests, correct specification is important. Intercept and trend should be included only when it is appropriate. We make this decision by running the regression and finding the intercept and or trend significant at 5% before we include in the model specification. We are also mindful that structural break could lead the test to indicate nonstationary when it’s actually stationary. We graph our series data and examine them before making decision to test for structural break. We use Eviews 9 software to test if the coefficient on the lagged variable yt 1 = 0. H 0 : yt = 0 The Phillips-Perron (PP) unit root test differs from the ADF test mainly in how they deal with serial correlation and heteroskedasticity in the errors. Formulation: y      yt 1  ut ………………. (2) H 0 :  0 One advantage of the PP tests over the ADF tests is that the PP tests are robust to general forms of heteroskedasticity in the error term ut . Another advantage is that the user does not have to specify a lag length for the test regression (Phillips and Perron 1988). The ADF and PP unit root tests are for the null hypothesis that a time series yt is I(1) or higher and therefore called non-stationary tests. Stationary tests, on the other hand, are for the null that yt is I(0). The most commonly used stationary test is the KPSS test, (Kwiatkowski, Phillips et al. 1992). Formulation: yt  0 Dt  t  u ………………………(3) Where Dt contains deterministic components and t is a random walk and u is white noise ARDL model If at least one of the series data is found to be integrated not in the same order of the other variables, then the traditional co-integration tests, usually produce false results since they require variables to be integrated in same order (Engle and Granger 1987). To avoid the possibility of chaining out spurious regression, this paper institutes ARDL model developed by Pesaran et al. (Pesaran, Shin et al. 2001) to test the long run and short run relationship of our time series data. ARDL model institutes a bound test for the co-integration. ARDL approach of co-integration now enjoys increasing use in various disciplines like macroeconomics, applied finance, education economics, tourism, etc. Like many other researchers, one of the reasons why we utilize ARDL "bounds" testing approach to cointegration in our study is that this model does not require all the series to be stationary in the same order even though it is inefficient if at least one variable is integrated in order two. The other reason is the ARDL Bounds Test has the power to detect long term relationship even in small samples such us ours. Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index We formulate our ARDL model as 1stModel: fdt   0  SME Compensation   t 1 ….(4)  2nd model: fd = β + β Δfd + δ Δmicro + σ Δsmall + α Δmedium … t 0 i t-i i t-i i t-i i Empirical Results and Discussions Table 1: Descriptive analysis t-i θ1fd t-1 +θ 2 micro t-1 +θ 3small t-1 +θ 4 medium t-1 +ε t ………… (5) In line with US SBA’s categorization of SME using the number of employees, we decompose compensation into micro-firm compensation, small firm compensation and medium firm compensation where “fd” represents firm deaths (firms that could not be sustained), “micro” is compensation paid by micro firms, “small” is compensation paid by small firms and “medium” is compensation paid by medium scale firms. Prior to testing whether there is a long-run relationship for the ARDL equation, it is essential to determine the most appropriate lag length for the model. Extremely short-lag lengths may lead to incorrect specification but we risk losing too many degrees of freedom if lag length is too long. By way of art, we use the rule of the thumb that says that maximum lag for annual data is either 1 or 2. We opt for 2 and use AIC information criteria to select the optimum lag length. This and other information criteria are based on a high log-likelihood value, with a "penalty" for including more lags to achieve the maximum lag length. One of the key assumptions of ARDL Bounds testing is that the errors in equation (4) above must not be serially correlated. To test the serial independence of our series data, we conduct residual diagnostic tests using Ramsey RESET Test, Breusch-Godfrey Serial Correlation LM Test and Breusch-Pagan-Godfrey Heteroskedasticity Test as our residual diagnostic Tests (See table 3) after which we conduct Bounds test (see table 3). VECM Granger Causality Test After examining the long run relationship between the variables, we use the Granger causality test to determine the causality between the variables. If there is co-integration between the series, then the VECM can be developed as follows: Firm  death t   a1   b11i  Medium   a   b t  =  2  +  21   a3  b31i Small t       Microt   a4  b41i 1  L  =   b11i b +  21 b31i  b41i b12i b13i b22i b23i b32i b33i b42i b43i b12 i b13i b14i  Firm  death t 1  b22i b23i b24i   Medium t 1   +K b32i b33i b34i   Small t 1     Micro t b42i b43i b44i    b14i  Firm  death t 1     1t    b24i   Medium t 1      +   ECM T 1   2t   3t  b34i   Small t 1            Microt b44i       4 t  (6) All micro to medium figures are in hundred millions of US Dollars and SME figures are in billions of US Dollars except Min. figures that are also in hundred millions of US Dollars. Table 1 above presents descriptive analysis. It can be seen that, on the average, small enterprises pay higher compensation than both micro and medium scale enterprises. Medium scale enterprises which pay the least on average also have the largest standard deviation. Stationary/Non Stationary Test In co-integration analysis, testing the nonstationary of the variables is a necessary condition even though the ARDL method does not require pretesting the integration order of each variable, however, the non-stationary and stationary results can be used to confirm whether the ARDL model should be used or not since the approach requires that none of the variable under consideration must be integrated in order two (Pesaran, Shin et al. 2001). The result of non-stationary and stationary tests is presented in table 2. Table 2: Non-stationary and stationary tests From equations 1, 2 and 3 for ADF, PP and KPSS respectively, the following results are generated. According to the ADF and PP tests, the null hypothesis that the time series are integrated in order one could not be rejected at 5% significance level for each of our variables since all the t-statistic values are greater than their respective critical values. Similar result is suggested by the KPSS, with the exception of firm deaths, where the t-statistics at level is greater than the critical value (see table 2). The combined effect of this result provides sufficient ground to conclude that the series are not stationary at level. At first difference analysis, all the t-statistic values are less than their respective critical values. This leads us to conclude that our variables are not stationary at level but become stationary at first difference. However, KPSS makes firm death result inconclusive. Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index Table 3: Bounds Test for SME The ADF, PPP and KPSS regressions include an intercept and trend. Optimal lags are determined using the Akaike Information Criterion. ADF and PP tests represents non-stationary test whilst the KPSS represents the stationary test. Each critical value is at 5% significant level. The ADF, PPP and KPSS regressions include an intercept and trend. Optimal lags are determined using the Akaike Information Criterion. ADF and PP tests represents non-stationary test whilst the KPSS represents the stationary test. Each critical value is at 5% significant level. Co-integration Analysis According to the KPSS stationary test, at least one of our series data is integrated in l(0) and others in l(1). Under this inconclusive result, co-integration tests may produce false results since they require variables to be integrated in same order. So, this paper institutes ARDL model developed by Pesaran et al. to test the long run and short run relationship of our time series data (Pesaran, Shin et al. 2001). ARDL model institutes bounds test for the cointegration and using equation 4 and 5 the results are presented in table 3. In our model 1 equation, where we use firm death as dependent variable and compensation paid to SME employees as independent variable, our results show that, Fstatistic value of 9.26 is greater than the upper bound value of 8.27 at 2.5% significant level. However, at 1% significant level, our F-statistic fall in-between the lower and upper bounds (see table 3).This result rejects the hypothesis of no co-integration. It therefore means that, at 2.5% significant level, sustainability of SMEs is co-integrated with SMEs compensation to their employees. Similarly, when we run model 2 with firm death as dependent variable and compensation paid to Micro firm, Small firm and Medium firm employees as independent variables, the F-statistic figure of 7.48 is higher than the upper bound value of 5.61 at 1% significant level. This result is interpreted that, at 1% significance level, the null hypothesis of no co-integration is rejected and it is concluded that there is co-integration relationship between firm death and SME compensation payment. When the SME compensation is used as dependent variable, the results show no evidence of cointegration. We do not report these figures because they are not of interest in this paper. We use critical bounds from Pesaran et al. (2001) to make decision whether co-integration exists or not. Intercept and but no trend are used. AIC was used to select the optimum lag length for diagnostic tests To test the validity of our model with our target variable, firm death as dependent variable, we report the result in column 2 of table 3. We employ Breusch-Godfrey Serial Correlation LM Test and Breusch-Pagan-Godfrey Heteroskedasticity Test as our residual diagnostic Tests. According to the Breusch-Godfrey Serial Correlation LM Test with calculated F-statistic value p-values, the null hypothesis of no serial correlation is accepted. Similarly, according to the Breusch-Pagan-Godfrey Heteroskedasticity Test for residual diagnostic, our models are free from Heteroscedasticity problem and this result suggests good models for our analysis. Ramsey RESET Test for model stability also suggests our two models are dynamically stable. Table 4: Long and Short Run Analysis Once co-integration relationship has been established, we now examine the long and short run impact of SME compensation on SMEs firm death. We present this result in table 4 above. Our results show a negative relationship between firm death and compensation. It is statistically at 5% level of significant. With the given coefficient, it is interpreted that a 100% increase in compensation for SMEs, micro, small and medium-scale enterprise employees will result in a 28%, 24%, 33% and 23% respective drop in the number of SMEs that collapse in the long term. Our finding confirms the combined effects Reinforcement and Expectancy Theories, Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index Equity Theory as well as Agency Theory. Also, Matching Theory argues that effective matches between SMEs firms that demand talent and employees with desired abilities occur when there are complementarities between employees productivity and firm-specific productivity (Pissarides 2000). Our results suggest SME are placed at the disadvantage side in seeking talents that can foster sustainable growth rather than collapse. Even when SMEs have managed to get talents through recruitment or training, Relative Deprivation Theory argues that individuals experience deprivation when they find that they have received fewer rewards than they deserve compared to rewards received by their reference groups (Cowherd and Levine 1992). Equity theory suggests that SMEs employee perceptions of what they contribute to the organization, what they get in return, and how their return-contribution ratio compares to others outside the organization, will fuel perceptions of inequity and cause employees to take actions to restore equity. Unfortunately, some such actions (e.g., quitting or lack of cooperation) may contribute immensely to firm deaths and make them unsustainable. This is because lack of adequate compensation from SMEs incapacitates owners to align workers’ interest to that of the company and this contributes to their collapse. The negative and statistically significant estimates for ECTt 1 of 0.396 lends support to long run relationship at 5%, significant level (see table 4 above). In the short run, our results show that there is a statistically significant negative relationship between compensation and SME collapse. Specifically, we find that a 100% increase in compensation for SME, Micro, Small and Medium-Scale Enterprise employees will result in a 33%, 21%, 29% and 33% respective drop in the number of firms that were otherwise unsustainable. Impulse Response Function From our result presented in fig 1 and 2, a one standard deviation positive shock to Compensation, receives both positive and negative reaction of firm death. In the case of small firms, firm death shows a negative impact as there is a decrease of close to seven thousand firm deaths in the first two years but these numbers inches down to above one thousand in the tenth year. A higher impact is seen in medium enterprises as there is a reduction of over thirteen thousand firm deaths in the firth year and a downward trend is seen after the tenth year. Micro enterprise rather shows a positive impact but overall, we conclude that higher compensations in SME will help reverse the increasing number of firm that collapse on yearly basis. Response of FIRM_DEATH to Cholesky One S.D. Innovations 25,000 20,000 15,000 10,000 5,000 0 -5,000 -10,000 -15,000 1 2 3 4 5 MICRO 6 7 SMALL 8 9 10 MEDIUM Fig. 1. Impulse-Response Function Accumulated Response of FIRM_DEATH to Cholesky One S.D. Innovations 200,000 150,000 100,000 50,000 0 -50,000 -100,000 1 2 3 4 MICRO 5 6 SMALL 7 8 9 10 MEDIUM Fig. 2. Accumulated Impulse-Response Function VECM Granger Causality To complete our analysis, we carry out Granger causality test to describe the direction of relationship between the variables. Since co-integration is confirmed in our study, there must be a unidirectional or bi-directional causality among the series. We examine this relation within the VECM framework. From equation 6 above, we run the model and report the VECM Granger causality result in Table 5 below. In our 1st model which has firm death as the dependent variable and SME compensation as an independent variable, we find that the speed of adjustment towards long run equilibrium has the expected negative sign and it is statistically significant at 5% significance level, therefore it retains its economic interpretation. Because it is negative, it ensures that if there is a deviation in one direction, the correction will reverse in the other direction so to ensure that equilibrium is retained. This indicates that SME compensation granger causes firm death and makes them unsustainable in the long run dynamics. In our 2nd Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index model which also has firm death as the dependent variable but decompose SME compensation into micro firm compensation, small firm compensation and medium firm compensation as a separate independent variables, we find long run causality running from compensation of Micro, Small and Medium scale enterprise to firm death at 5%. We also find short run causality running from Small and Medium scale enterprises to firm death. Table 5: The VECM Granger Causality Analysis 12 8 4 0 -4 -8 -12 01 02 03 04 05 06 07 08 CUSUM 09 10 11 12 13 14 5% Significance Fig. 5. Plot of CUSUM of Recursive Residuals with Micro as dependent variable 12 8 4 0 We conduct stability diagnostic on our data. We use cumulative sum (CUSUM) of Recursive Residuals and the cumulative sum of squares (CUSUMsq) of Recursive and present our result from fig. 3 to fig. 7. The graphs of CUSUM confirm stability of parameters (Brown 1975) but CUSUMsq test does not lie within the 5% critical bounds. However, on the whole, we find the graphs are within the critical bounds at 5% level of significance. This ensures the stability of long run and short run coefficients. -4 -8 -12 01 02 03 04 05 06 07 CUSUM 08 09 10 11 12 13 14 5% Significance Fig. 6. Plot of CUSUM of Recursive Residuals with Small as dependent variable 12 8 4 Stability Diagnostics for VECM Walt Test 12 0 8 -4 4 -8 0 -12 01 02 03 04 05 06 07 08 09 10 11 12 13 14 -4 CUSUM 5% Significance Fig. 7. Plot of CUSUM of Recursive Residuals with Medium as dependent variable -8 -12 01 02 03 04 05 06 07 CUSUM 08 09 10 11 12 13 14 5% Significance Fig. 3. Plot of CUSUM of Recursive Residuals with Firm death as dependent variable 1.6 1.2 0.8 0.4 0.0 -0.4 01 02 03 04 05 06 07 CUSUM of Squares 08 09 10 11 12 13 14 5% Significance Fig. 4. Plot of CUSUM of Squares Recursive Residuals with Firm death as dependent variable CONCLUSION AND POLICY IMPLICATIONS This paper mainly investigates how SME compensation affects their collapse. We utilize annual time series data provided by America’s SBA from 1988 to 2014. We decompose SMEs into three size categories (micro, small and medium) and analyse the impact on each of the sub-division to capture the extent this might help inform policy at corporate level and beyond. Our stationary and non-stationary analysis shows that compensation and firm death series are not stationary but have unit root. Our ARDL Bounds Test shows that compensation and firm deaths have long-run relationship. Through VECM Granger Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index causality analysis, we find a unidirectional causality running from compensation to firm death indicating that the inappropriate compensation payment to workers of SMEs contributes to collapsing that all important sector of various economies using the American economy as an example. Through the impact response analysis, we find that compensation package has more telling effect on medium scale enterprise followed by small scale and micro enterprises in that order. This separation suggests that compensation package may differ among the SME sub-categories. Our study is consistent with theories such as Expectancy theory Vroom (1964) Equity theory Agency theory Fama and Jensen (1983) that predict doom for inadequate compensation for best talents. The study is also consistent with literature that suggest that compensation could determine the success or otherwise of firms (Gomes and Kuehn 2017). This finding is important to policy makers. In the current efforts being made across the world to boost SMEs and save them from collapse, policies should be made to boost SME compensations and make it more competitive. Some of the reasons why increase in SME compensation will help them succeed is that their current salary levels are not competitive to that of the large corporations. This makes high talent employees prefer jobs offered by large corporation. Employees who end up in SMEs may not give up their best especially when they compare their compensation to their peers in large corporations and other institutions. We call for similar research in other countries. Some limitations have to be considered with this paper. The paper did not analyse the financial strength of SMEs to be able to increase the compensation of their employees. The results of this study should therefore be implemented taking SME’s financial strength and other factors into consideration. FUNDING Funding for this paper was granted by Project of the National Social Science Foundation of China (No. 13BGL038) and Project of Social Science Foundation of Jiangsu (No.15JD007) REFERENCES [1]. A., W. R. G. (1993). "Environmental Management and Business Strategy." Pitman. London. [2]. Abdullah, M. M. B. and J. J. Tarí (2012). "The influence of soft and hard quality management practices on performance." Asia Pacific Management Review 17(2): 177-193. [3]. Adams, J. S. (1963). "Towards an understanding of inequity." The Journal of Abnormal and Social Psychology 67(5): 422. [4]. Aldrich, H. and E. R. Auster (1986). "Even dwarfs started small: Liabilities ofage and size and their strategic implications." Research in organizational behavior 8(1986): 165-186. [5]. Audretsch, D. B. and T. Mahmood (1994). "The rate of hazard confronting new firms and plants in US manufacturing." Review of Industrial organization 9(1): 41-56. [6]. Bailey, T. (1993). Discretionary effort and the organization of work: Employee participation and work reform since Hawthorne, Teachers College and Conservation of Human Resources, Columbia University. [7]. Beck, T., A. Demirgüç‐Kunt and V. Maksimovic (2005). "Financial and legal constraints to growth: Does firm size matter?" The Journal of Finance 60(1): 137177. [8]. Box, M. (2008). "The death of firms: exploring the effects of environment and birth cohort on firm survival in Sweden." Small Business Economics 31(4): 379-393. [9]. Brown, R. L., J. Durbin and J.M. Evans (1975). "Techniques for testing the constancy of regression relationships over time." Journal of the Royal Statistical Society 37: 149–192. [10]. Brundtland, G. H. (1987). "Our common future—Call for action." Environmental Conservation 14(4): 291-294. [11]. Carreira, C. and F. Silva (2010). "No Deep Pockets: Some Stylized Empirical Results On Firms’financial Constraints." Journal of Economic Surveys 24(4): 731-753. [12]. Carroll, G. R. and M. T. Hannan (2000). The demography of corporations and industries, Princeton University Press. [13]. Cowherd, D. M. and D. I. Levine (1992). "Product quality and pay equity between lower-level employees and top management: An investigation of distributive justice theory." Administrative Science Quarterly: 302-320. [14]. Dah, M. A. and M. B. Frye (2017). "Is board compensation excessive?" Journal of Corporate Finance 45(Supplement C): 566585. [15]. Dickey, D. A. and W. A. Fuller (1981). "Likelihood ratio statistics for autoregressive Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index [16]. [17]. [18]. [19]. [20]. [21]. [22]. [23]. [24]. [25]. [26]. [27]. time series with a unit root." Econometrica: Journal of the Econometric Society: 10571072. Dineen, B. R. and I. O. Williamson (2012). "Screening‐oriented recruitment messages: Antecedents and relationships with applicant pool quality." Human Resource Management 51(3): 343-360. Duong, L. and J. Evans (2015). "CFO compensation: Evidence from Australia." Pacific-Basin Finance Journal 35(Part A): 425-443. Engle, R. F. and C. W. Granger (1987). "Cointegration and error correction: representation, estimation, and testing." Econometrica: journal of the Econometric Society: 251-276. Fama, E. F. and M. C. Jensen (1983). "Separation of ownership and control." The journal of law and Economics 26(2): 301325. Freeman, J., G. R. Carroll and M. T. Hannan (1983). "The liability of newness: Age dependence in organizational death rates." American sociological review: 692-710. Gerhart, B. and S. Rynes (2003). Compensation: Theory, evidence, and strategic implications, SAGE publications. Gomes, P. and Z. Kuehn (2017). "Human capital and the size distribution of firms." Review of Economic Dynamics 26(Supplement C): 164-179. Greenberg, J. (1990). "" Employee theft as a reaction to underpayment inequity: The hidden cost of pay cuts": Correction." Gupta, N. and J. D. Shaw (2014). "Employee compensation: The neglected area of HRM research." Human Resource Management Review 24(1): 1-4. Jenkins Jr, G. D., A. Mitra, N. Gupta and J. D. Shaw (1998). Are financial incentives related to performance? A meta-analytic review of empirical research, American Psychological Association. Kwiatkowski, D., P. C. Phillips, P. Schmidt and Y. Shin (1992). "Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root?" Journal of econometrics 54(1-3): 159-178. Lahlou, I. and P. Navatte (2017). "Director compensation incentives and acquisition performance." International Review of Financial Analysis 53(Supplement C): 1-11. [28]. Liu, L. S. and A. W. Stark (2009). "Relative performance evaluation in board cash compensation: UK empirical evidence." The British Accounting Review 41(1): 21-30. [29]. Lucas Jr, R. E. (1978). "On the size distribution of business firms." The Bell Journal of Economics: 508-523. [30]. Mata, J. (1996). "Markets, entrepreneurs and the size of new firms." Economics Letters 52(1): 89-94. [31]. Müller, V.-O. (2014). "Do Corporate Board Compensation Characteristics Influence the Financial Performance of Listed Companies?" Procedia - Social and Behavioral Sciences 109(Supplement C): 983-988. [32]. Olaniyi, C. O., O. B. Obembe and E. O. Oni (2017). "Analysis of the Nexus between CEO Pay and Performance of Non‐Financial Listed Firms in Nigeria." African Development Review 29(3): 429-445. [33]. Peek, J. (2013). "The impact of credit availability on small business exporters." US Small Business Administration working paper. [34]. Pesaran, M. H., Y. Shin and R. J. Smith (2001). "Bounds testing approaches to the analysis of level relationships." Journal of applied econometrics 16(3): 289-326. [35]. Pfeffer, J. (1998). "Six dangerous myths about pay." Harvard business review 76(3): 109-119. [36]. Phillips, P. C. and P. Perron (1988). "Testing for a unit root in time series regression." Biometrika 75(2): 335-346. [37]. Pissarides, C. A. (2000). "Equilibrium Unemployment Theory, second edn." MIT Press, Cambridge. [38]. Shaw, J. D. and N. Gupta (2007). "Pay system characteristics and quit patterns of good, average, and poor performers." Personnel Psychology 60(4): 903-928. [39]. Yarram, S. R. and J. Rice (2017). "Executive compensation among Australian mining and non-mining firms: Risk taking, long and short-term incentives." Economic Modelling 64(Supplement C): 211-220. Volume 8 | Issue 1 | Januray-December-2018 [(8)1: 010-019] | http://onlinejournal.org.uk/index.php/BJIR/index