University of Groningen – Faculty of economics and business
Master thesis
M&A in firm innovation activities
Contents
1. Introduction ... 3
2. Literature review... 5
2.1 Firm innovation activity in M&A... 5
2.2 M&A in post-M&A firm innovation activity... 8
3. Research methodology... 10
3.1 Data and sample ... 10
3.2 Measures ... 12
3.3 Statistical methods... 15
4. Results... 21
5. Discussion and Concluding remarks... 24
6. Reference... 26
7. Appendix... 29
Appendix 1: Descriptive statistics of variables regression model 1 ... 29
Appendix 2: Descriptive statistics of variables regression model 2 ... 30
Appendix 3: Selected literature summary... 31
Abstract
Theory suggests firm innovation activities become an important motive for firms to engage in M&A activities. This paper investigates the role of M&A activities in firm innovation activities in electrical and optical equipment industry during the sixth M&A wave from 2003 to 2008. We create a timeline to investigate the relationship between past firm innovation activities and M&A, and the relationship between level of M&A activities and post-M&A firm innovative activities. We find a positive and significant relationship between firm innovation activities and the propensity of firms engage in M&A activities. However, we do not find evidence to support the relationship between M&A activities and firms’ post-M&A innovation activities.
1. Introduction
Merger and acquisition implies that consolidation of firms either to form a new company or one firm purchases another firm without creating new subsidiaries, in order to achieve economies of scale and scope, higher growth rate in revenue, market share or better human resources etc. In addition, firms could achieve corporate “synergy” through M&A (Harrison et al., 1991), and M&As between firms are aimed at maximizing firm values (Salter and Weinhold, 1979). As the rapid development of business world, M&A has become one of the most popular modes of organizational collaboration in recent decades (Lamont and Anderson, 1985; Porter 1987). One of the most important purposes for M&A is seeking for the development and growth of organizations (Roll, 1986; Varian, 1988). Moreover, M&A allows firms to expand production scale, reduce costs, to enhance market power, to increase efficiency and to acquire better technologies, management skills and human resource (Haleblian et al., 2009).
One of the most important benefits that firms receive through M&A is to upgrade technologies. M&A allows firms to fully absorb and integrate technologies and innovative capabilities from other organizations. Meanwhile, it also allows firms to combine separate corporate identities into one new corporate identity (De Man and Duysters, 2003). Recent contributions show the growing importance of technology development and innovation in M&A activities, and firms achieve potential gains through technology integration (Cefis and Sabidussi, 2011). Therefore, future technology development and firm innovation activities are considered as important motivations for M&A activities beside financial and managerial motivations.
Hollanders and Arundals (2005) investigate the most innovative sector in European countries, and they find that electrical and optical equipment becomes one of the leading sectors in innovation activities in most EU countries. As Eurostat (Chapter 9,
states, the manufacturing of electrical and optical equipment is an important concept in European manufacturing industry, and most of these products are considered as high-technology goods, such as computers, transmission goods, semi-conductor etc. In addition, the demand of consumers and investment decisions depend on the economic environment and business cycle, the production of electrical and optical equipment should adapt the changes more frequently than other manufacturing industry (Eurostat). Therefore, firms in such industry are forced to upgrade and improve their technology more frequently, which require them to engage in M&A activities more frequently (Cooldt et al., 2006). According to NACE Rev.2, we take manufacture of computer, electronic and optical production industry in our research.
Firms engage in M&A since they could achieve better performance in technology, innovation efficiency and market power etc (Haleblian et al., 2009). Therefore, firms are more likely to acquire other firms if they are able to achieve synergy after M&A. Many empirical studies indicate that historical performance would lead to different post-acquisition performance (Heron and Lie, 2002; Servaes, 1991). For instance, Heron and Lie (2002) find post-M&A financial performance are strongly correlated to pre-M&A market-to-book ratio. Evidence also suggests that the pre-M&A financial performance of target firms affect post-M&A performance of firms (Bruner, 1988). However, many empirical studies focus on the impact of M&A activities on pre- and post-M&A firm performance from financial and managerial perspectives. As technology upgrading and firm innovation are also considered as important motivations of M&A activities, more attention should be paid on firm innovation perspective for highly innovative firms. Therefore, it is necessary to investigate whether the past firm innovative activities affect the propensity of the firms to engage in M&A, and how M&A activities affect post-M&A firm innovative activities. In this paper, we illustrate two research questions. In the first part, we investigate the effect of past firm innovation activities on M&A engagements.
In the second part, we focus on firms that engaged in M&A activities and investigate the effect of M&A activities on post-M&A firm innovative activities. The second research question is illustrated as follows.
Research question 2: For firms engaged in M&A activities, do M&A activities increase post-M&A innovative activities?
In summary, we illustrate a timeline graph to illustrate our research idea, and the graph is represented as follows.
The rest of the paper is organized as follows. In section 2, we provide relevant literature, and hypotheses are also illustrated in this section. In section 3, we provide the research methodology, which mainly consist of data description and collection methods, the measurement of variables and the illustration of our baseline regression models. In section 4, we present the regression results and discuss the implications. In section 5, we make concluding remarks.
2. Literature review
2.1 Firm innovation activity in M&A
According to corporate control theory, M&A is a business strategy to correct inefficiencies and capital market imperfections (Jensen and Ruback, 1983). Therefore, firms can generate higher amount of financial investment in developing technologies and enhance innovation on productions. Meanwhile, firms gain better technological resources from other firms to improve innovation efficiencies. Thus, firm innovation is an important motive for firms engage in M&A activities (Harrison et al., 1991; Gerpott,
1995). In addition, firms must be capable to generate innovative products to survive in the market, or otherwise exit the market. Firm innovation enables firms to obtain core competencies and be able to survive in competitive environment, and firms update technologies and acquire new knowledge to create competitive advantage. Moreover, innovation driven motives enhance firms to seek and utilize technologies both internally and externally. As a result, M&A plays an important role in high-tech industries.
high-tech industry. The vice president announced that innovation internally has higher risks, higher costs and more time consuming than acquire technologies from existing firms (Electronic business, Jan.7, 1991, page 31). As a software developer, Symantec Corporation seeks and acquires talented people and better products and technologies from other firms. Therefore, Symantec Corporation acquired 18 firms within 12 years operation. These examples indicate that M&A strategy becomes either the short-run or long-run strategy for the growth purpose of a firm, and becomes a trend for most high-tech firms. Although M&A strategy is a popular corporate strategy in high-tech industries, we do not know the relationship between firm innovation and M&A activities.
Hall (1990) and Blonigen and Taylor (2000) investigate the relationship between R&D intensity and M&A behaviour of firms in electrical equipment industries, and the results indicate that firms with relatively low level of ex ante innovation input are more likely to engage in M&A. However, Lehto and Lehtoranta (2003) find that more R&D expenses raise the probability to engage in M&A. Gantumur and Stephan (2011) find that acquirers with larger stock of accumulated knowledge and higher R&D intensity are more likely to engage in M&A activities. Moreover, larger acquirers with stronger R&D financing ability are more likely to become an acquirer, and acquirers with relatively low level of R&D financing ability pursue to develop knowledge and technologies internally and with relatively less M&A activities.
In summary, many empirical studies focus on a cross-industry data. However, the relationship between pre-M&A innovation activities and M&A activities is stronger in technology intensive sectors or industries. We focus the study on high-tech industries rather than a wide range of sectors, and expect the relationship between pre-M&A firm innovation activities and M&A to be either positive or negative. Hence,
Hypothesis 1b: Higher level of firm innovation activity decreases the propensity of firms to engage in M&A.
2.2 M&A in post-M&A firm innovation activity
industries (Bertrand and Zuniga, 2006). Milliou (2004) finds that M&A firms have more incentives to raise total R&D expenditure than non-M&A firms. The impact of M&A on innovative activities is separated across relatedness across partner firms. Cassiman et al (2005) find that R&D efficiency increases when two merging firms are technologically complementary, and R&D inputs decrease when they technologically substitute. Furthermore, Cefis and Sabidussi (2011) conduct panel data to test the effects of M&A activities on firm innovation, and find M&A activities promote innovation inputs, outputs and efficiencies in the short run. Furthermore, M&A activities positively affect post-M&A firm innovation activities. However, it depends on firms’ abilities to absorb and integrate knowledge (Cooldt et al., 2006). Therefore, we expect M&A activities have a positive impact on post-M&A firm innovation activities. Hence,
scale, it gives integrated firms less incentive to invest in innovation process. The argument on net effects of M&A activities on innovation process is ambiguous. The consolidation of two firms offer firms new opportunities to explore economies of scale and scope, enhancing research and development efficiency and stimulating innovation capabilities (Cohen and Levin, 1989). However, merger and acquisition may reduce the competition in upstream knowledge market and downstream production market, and reduce firms’ innovative activities. In summary, we expect that M&A activities negatively affect firm innovation activities based on empirical studies. Hence,
Hypothesis 2b: M&A activities negatively affect post-M&A firm innovation activities.
3. Research methodology
3.1 Data and sample
Our dataset is a panel data from manufacture of computer, electronic and optical products industry. As Eurostat states that the manufacturing electrical and optical equipment becomes one of the most important concepts in European manufacturing industry, most of the products from this industry should adapt to the changing environment more frequently than other manufacturing goods. In addition, manufacturing of electrical and optical equipment is a highly innovative industry, most of the products in this industry are considered as high-tech productions. Therefore, we choose manufacturing of computer, electronic and optical equipment industry according to NACE Rev.2, Code 26. Moreover, since the sixth wave of M&A is from 2003 to 2008 (Lipton, 2006), we take period from 2003 to 2008 in our analysis.
M&A activities in current year and relationship between current year M&A activities and R&D intensity, efficiency in next year (see equation 1 and 2). Therefore, we collect data of R&D expenditures, total assets, sales, total liabilities, return on assets, cash flow and year of incorporation from 2002 to 2009. Moreover, because of studying the lagged effects in this paper, thus we need at least one year data before and after the M&A wave. In addition, we study the sixth wave of M&A (from 2003 to 2008). In order to avoid the missing value bias in our analysis, we eliminate firms with missing value of dependent variable in equation (2) (R&D intensity). Any firms with a value Not Applicable (N/A) in R&D intensity are removed from the dataset, and this keeps 1214 firms from computer, electronic and optical equipment manufacturing sector in our sample. The first regression model investigates the relationship between firm innovation activities and propensity of M&A transaction completion for M&A and non-M&A firms, thus we run the regression with 1214 firms consist of both M&A and non-M&A firms in 6 years, which gives a total sample size of 7284 firms in our first dataset. Moreover, the second regression model only focuses on firms engaged in M&A activities. Therefore, we eliminate firms do not complete any M&A transactions during 2003 and 2008, and only keep firms do at least one M&A activity during six years in our dataset. This keeps 3186 firms in our second dataset (531 firms in 6 years).
3.2 Measures
Regression model 1
Dependent variable
The dependent variable for equation (1) is M&A, and we use two different measurements for it. The measurement for probit model is dummy, we denote 1 as firms which complete M&A transaction(s) in that specific year and 0 as firms do not complete any M&A transaction(s) in that year. Moreover, the second measurement for our dependent variable in count model is frequency of M&As, and we measure it by counting the number of complete M&A transaction(s) for firms in that year. Moreover, by using the second measurement, we could treat firms that engage in M&A only once differ from those that engage in more than once in that specific year.
Independent variable
The independent variable in our first regression model is firm innovation activities, and we measure firm innovation activities by R&D intensity. In addition, we obtain R&D intensity by using R&D expenditure divided by total assets of a firm in a specific year.
Control variables
is usually calculated as after tax income plus non-cash charges (depreciation). Moreover, the debt situation of a firm may affect firms’ M&A engagement. Higher debt of a firm may reflect capital constraint, and higher debt ratio of a firm may reduce the possibility that a firm engage in M&A. The measurement of a firm’s debt situation is debt ratio, calculated by total debt divided by total assets of a firm. Furthermore, we add profitability to the model, and it has been argued that the profitability of a firm would contribute to M&A. According to empirical studies, we measure profitability of a firm by ratio of return on assets. The measurement of return on sales is net income divided by sales. Finally, we control the cross time effects and add year as the final control variable. In summary, we have 5 control variables in our model, which are firm size, cash flows, debt ratio, profitability and year. In addition, we take natural logarithm for variables firm size (total assets and number of employees), since the data for firm size (total assets and number of employees) may contain wide-ranging numbers, and natural logarithm could reduce wide-ranging numbers to smaller range (Spiegel and Stephens, 1998). Moreover, it is worthwhile to mention that we re-scale cash flow data by dividing all cash flow information by 10000 rather than natural logarithm for wide-ranging quantity variable cash flow in our first regression model. Because the data of cash flow variable contains negative number, natural logarithm would cause missing value in our dataset, thus we prefer to use re-scale for cash flow data.
Regression model 2
Dependent variable
Independent variable
The independent variable in our second regression model is M&A activities. We investigate the relationship between M&A activities and post-M&A firm innovation for both acquirers and target firms. In the second regression model, we are interested in identifying whether previously formed M&A activities affect post-M&A firm innovation. We use the number of M&A transactions which firms completed as the measurement of M&A activities. Moreover, we eliminate firms without M&A activities during the sixth M&A wave due to we only focus on acquirers and target firms in this case. Therefore, we keep firms do at least 1 M&A during 2003 and 2008 (firms with M&A activities). In addition, we rescale the data by dividing number of firms’ complete transactions by 100 since the number of M&A transaction completion has large scale.
Control variables
year of a firm’s incorporation. Mairesse and Hall (1996) and Blundell and Bond (1998) argue that higher remuneration on labour could reduce resources that available for intangible investments, which substitute R&D investments. However, higher labour cost also implies higher quality of employees, which can facilitate R&D investments or R&D performance. Therefore, we consider the average costs of employee as a control variable, and we measure labour costs by using total wage and remuneration divided the number of employees. Finally, we add year as a control variable to control the cross time effect. To sum up, we have 5 control variables in our second regression model. We take natural logarithm for firm size (total assets and number of employees), firm age and labour costs to reduce the scale of our data.
3.3 Statistical methods
which the measurement of M&A engagement is a dummy. Therefore, we estimate the probit regression model with M&A engagement as the dependent variable to the first test for the impact of the independent variable on the likelihood that a firm’s decision on “go” or “no go” to M&A. On the other hand, we also use a count model to treat firms do 1 M&A and firms do more than 1 M&As separately, because it is able to solve the specification bias problem in probit model. In general, two testing models can be expressed as follows.
MA i, t= β0+ β1 FI i, t - 1+ β2 log(FS i, t - 1) + β3SCF i, t – 1+ β4 DEBT i, t – 1+ β5Πi, t - 1+ β6 Yeari, t - 1+ ei, t – 1 (1)
Where:
MAi. t= Propensity of M&A transaction(s) of firm i in year t FIi, t-1= Firm innovative activities of firm i in year t-1 FSi, t-1= Firm size of firm i in year t-1
CFi, t-1= Scaled free cash flow of firm i in year t-1 DEBTi, t-1= Debt ratio of firm i in year t-1
Πi, t-1= Profitability of firm i in year t-1 Yeari, t – 1 = Time year of firm i in year t-1 ei, t-1 = error term of firm i in year t-1
Alternatively, we take the number of M&A transactions that firms complete into consideration in the second test, and use count data mode. In addition, the dependent variable is the number of M&As that firms completed. Count model treat M&A firms differently, and the results of count data model suggests firm innovation may influence the number of complete M&A transactions which firms take, and gives robustness results.
The second regression model intends to investigate the relationship between M&A activities and post-M&A firm innovation, and we focus on firms engaged in M&A activities. In this model, our dependent variable is post-M&A firm innovation activities and the independent variable is M&A activities, and the second regression model can be illustrated as follows.
FI i, t+1= α0+ α1 MAAi, t+ α2log(FS i, t) + α3Π i, t+α4FAi, t+ α5LCi, t+ α6Yeari, t+ e i, t (2)
Where:
FIi, t+1= Firm innovative activities of firm i in year t+1 MAA i, t = M&A activities of firm i in year t
FSi, t= Firm size of firm i in year t Πi, t= Profitability of firm i in year t FAi, t= Firm age of firm i in year t
LCi, t = Average labour costs of firm i in year t Year i, t= Time year of firm i in year t
e i, t= Error term of firm i in year t
engaged in at least one M&A between 2003 and 2008, and this keeps 531 firms. Therefore, we run the regression for 531 firms between 2003 and 2008.
for the second regression model suffers from autocorrelation problem. Since data for second regression model also suffers from heteroskedasticity problem, we use heteroskedasticity and autocorrelation consistent (HAC) standard errors to correct heteroskedasticity and autocorrelation problems.
In order to generate the relationship mentioned in our first regression model, we first generate a panel probit regression model to investigate the relationship between previous innovative activities and M&A propensity (measured by dummy). Moreover, we generate either a panel Poisson regression model or Negative Binomial regression model for our count model mentioned in equation (1). The strong assumption of using Poisson model is the conditional variance equals to the conditional mean. Based on the assumption, we present the means and variances of each variable in the regression model, and investigate that the variances do not equal to the means (see Appendix 1). Therefore, we use panel Negative Binomial regression model for the analysis, which can be used for over dispersed count data.
Hausman test, which we confirm that the differences in coefficients are not consistent. Therefore, the estimators of fixed effects are closer to the true value. In the section of empirical results, we will perform the results of fixed effect regression model.
We should also check multicollinearity problem. If the explanatory variables are correlated to each other, then it might be difficult to distinguish the individual effect on dependent variables (Hill et al., 2011). Therefore, we present a multicollinearity check of our explanatory variables in both regression equations, and the result tables are as follows.
Table 1: Partial correlation regression model 1
MA NOMA RDI LTA DEBT LNE SCF ROA
MA -NOMA 0.225* -RDI 0.112** 0.030* -LTA 0.269** 0.193** -0.019 -DEBT -0.113** -0.011 -0.133** -0.173** -LNE 0.243** 0.183** -0.034** 0.935* -0.092** -SCF 0.256** 0.149 0.015 0.186* 0.006 0.178* -ROA 0.002 0.036** -0.187** 0.062** -0.264** 0.083** 0.094** -*p<0.10, **p<0.05
Table 2: Partial correlation regression model 2
*p<0.10, **p<0.05
Table 1 and Table 2 present the partial correlation between dependent variable and each explanatory variable, and most of the correlation coefficients are significant at 5% level. In both table 1 and 2, number of employee is strongly correlated with total assets of a firm. However, this is not surprising, since both variables measure firm size. In addition, we do not find strong partial correlation between explanatory variables in both tables.
RDI MAA LTA ROA FA LC LNE
-Therefore, we confirm that our dataset for both regression model 1 and regression model 2 do not suffer from strong multicollinearity problem.
4．Results
Table 3 presents the results of our regression models. We conclude that three models have good explanatory power from the R-square values in our 3 models. The probit model in regression 1 has the highest R-square value of 0.2703, which means 27.03% of variation in M&A engagements is explained by all explanatory variables in our model. Moreover, the R-square value of Negative Binomial model in regression 1 is lower than probit model with the value of 0.1787, which means the explanatory power between explanatory variables and dependent variable is lower in Negative Binomial model than probit model. In addition, regression model 2 has the lowest R-square value of 0.1242, and regression model 2 with fixed effect has the lowest explanatory power.
cash flow and M&A activity confirms Jensen’s (1988) results, which larger amount of free cash flows allow firms to finance M&A costs.
Table 3: Regression results
Regression model 1 Regression model 2 Probit model
Dependent variable: MA
Negative binomial model Dependent variable: NOMA
Fixed effect model Dependent variable: RDI Intercept 17.638 (29.247) -75.397**(30.937) 0.175***(1.525) MAA 0.003 (0.009) RDI 3.890*** (0.513) 2.263***(0.825) LTA 0.469*** (0.043) 0.304***(0.068) -0.026***(0.005) LNE -0.036 (0.043) -0.180***(0.070) 0.028***(0.006) SCF 0.013** (0.007) 0.002(0.003) DEBT -0.599*** (0.156) -0.030 (0.234) ROA 0.0006 (0.002) 0.004(0.003) -0.0003(0.0002) FA -0.001 (0.001) LC 0.014*** (0.005) Log likelihood -2162.92 -2510.69 R-squared 0.2703 0.1787 0.1242 Sample size 7284 7284 3192
*p<0.10, **p<0.05, ***p<0.01 (standard errors in parentheses)
consistent with most empirical studies (Miyazaki, 2009; Lehto and Lehtoranta, 2003; Gantumur and Stephan, 2007).
5．Discussion and Concluding remarks
Firm innovation activities are one of the most important sources to generate competencies and growth. In this paper, we investigate the role of M&A in firm innovation activities during the sixth M&A wave from 2003 to 2008. More specifically, we create a timeline to investigate the role of M&A activities in firm innovation activities. On one hand, we find a positive and strongly significant relationship between firm innovation activities and the propensity of firms engage in M&A activities. On the other hand, we do not find evidence to support the relationship between M&A activities and firms’ post-M&A innovation activities. In addition, we collect panel data and use both probit model and negative binomial model to investigate the relationship between firm innovation activities and the propensity of firms engage in M&A activities, which provide robust evidence to prove the relationship. Such relationship implies that firms in high-tech industry with larger R&D investments and R&D resources are more likely to become either an acquirer or a target firm, and more likely to engage in M&A activities. For firms engaged in M&A activities, higher level of M&A activities do not affect post-M&A firm innovation activities. We do not consider the impact of macro-economical change on firms’ M&A behavior and post-M&A innovative activities. However, the insignificant relationship between M&A activities and post-M&A activities might be due to the technology bubble burst in 2000, and future studies could explore the relevance. Overall, our results indicate that innovative activities raise the propensity of firms involves in M&A activities. However, our results do not show the relationship between M&A activities and post-M&A innovative activities.
6．Reference
Bertrand, O., Zuniga, P., 2006, “R&D and M&A: Are cross-border M&A different? An investigation on OECD countries”, International Journal of industrial organization, No. 24, pp. 401-423.
Bertrand, O., 2009, “Effects of foreign acquisitions on R&D activity: Evidence from firm-level data for France”, Research Policy, 38(6), pp. 1021–1031.
Blonigen, B.A., Taylor, C.T., 2000, “R&D Intensity and Acquisitions in High-Technology Industries: Evidence from the US Electronic and Electrical Equipment Industries”, The journal of industrial economics, Vol. 48, issue 1, pp. 47-70. Blundell, R., Bond, S., 1998, “Initial conditions and moment restrictions in dynamic panel data models”, Journal of Econometrics, 87, 115–143.
Bresman, J., Birkinshaw, J., Nobel, R., 1999, “Knowledge transfer in international acquisitions”, Journal of International Business Studies, No. 30, pp. 439-462.
Bruner, R., 1988, “The Use of Excess Cash and Debt Capacity as a Motive for Merger”,
Journal of Financial and Quantitative Analysis,Vol. 23, issue 2, pp. 199-217.
Cassiman, B., Colombo, M.G., Garrone, P., Veugelers, R., 2005, “The impact of M&A on the R&D process: An empirical analysis of the role of technological- and market-relatedness”, Research Policy, No. 34, pp. 195-220.
Cefis, E., Marsili,O., 2006, “Survivor: The Role of Innovation in Firms’ Survival”,
Research Policy, 35 (5):626-641.
Cefis, E., Sabidussi, A., Schenk, H., 2007, “Do mergers of potentially dominant firms foster innovation? An empirical analysis for the manufacturing sector”, Discussion
paper series, 07-20.
Cefis, E., Sabidussi, A., 2011, “Innovation in potentially dominant firms: the role of M&A”, working paper series, Paper presented at the DIME Final Conference.
Coase, R., 1937, “The nature of the firm”, Economica 4, pp. 386-405.
Cohen, W., Levin, R., 1989, “Empirical Studies of Innovation and Market Structure”,
Handbook of Industrial Organisation, North-Holland, chapter 18, pp. 1060-1107.
Cooldt, M., Hagedoorn, J., Van Kranenburg, H., 2006, “Mergers and acquisitions: Their effect on the innovative performance of companies in high-tech industries”,
Research policy, No. 35, pp. 642-654.
Evangelista, R., Sandven, T., Sirilli, G., Smith, K., 1998, “Measuring innovation in European industry”, International Journal of economics and business, Vol. 5, No. 3, pp. 311-333.
Gantumur, T., Stephan, A., 2011, “Mergers & acquisitions and innovation performance in the telecommunications equipment industry”, Journal of industrial and corporate
change, Vol. 21, issue 2, pp. 277-314.
Gerpott, T.J., 1995, “Successful integration of R&D functions after acquisition: an exploratory empirical study”, R&D management, Vol. 25, issue 2, pp. 161-178.
Haleblian, J., Devers, C.E., McNamara, G., Carpenter, M.A., Davison, R.B., 2009, “Taking stock of what we know about mergers and acquisitions: A review and research agenda”, Journal of Management, Vol. 35, NO. 3, pp. 469-502.
Hall, B.H., 1988, “The effect of takeover activity on corporate research and development”, Corporate Takeovers: Causes and Consequences, Chicago: University ofChicago (NBER Conference Volume).
Hall, B.H., 1990, “The impact of corporate restructuring on industrial research and development”, Brookings Papers of Economic Activity: Microeconomics 1990, pp. 85-135.
Harrison, J., Hitt, M., Hoskisson, R., Ireland, R., 1991, “Synergies and post-acquisition performance: differences versus similarities in resource allocations”, Journal of
management,Vol. 17, NO. 1, pp. 173-190.
Heron, R., Lie, E., 2002, “Operating performance and the method of payment in takeovers”, Journal of Financial and Quantitative Analysis, Vol. 37, issue 1, pp. 137-155.
Hill, R.C., Griffiths, W.E., and Lim, G.C., 2011, Principles of econometrics. John Wiley & Sons, Inc,Hoboken, US.
Hitt, M.A., Hoskisson, R.E., Johnson, R.A., Moesel, D.D., 1996, “The market for corporate control and firm innovation”, The Academy of Management Journal, Vol. 39, No. 5, pp. 1084-1119.
Hollanders, H., Arundel, A., 2005, “European sector innovation scoreboard”,
European Commission, Discussion paper from innovation/SMEs programme.
Jensen, M.C., 1988, “Takeovers: Their Causes and Consequences”, Journal of
Economic Perspectives, Vol. 2, pp. 21-48.
Jensen, M.C., Ruback, R.S., 1983, “The market for corporate control: The scientific evidence”, Journal of financial economics, Vol. 11, pp. 5-50.
Lehto, E.L.O., Lehtoranta, M.O., 2003, “Becoming an acquirer and becoming acquired”, Technological forecasting and social change, Vol. 71, issue 6, pp. 635-650. Lipton, M., 2006, “Merger waves in the 19th, 20th, 21stcenturies”, The Davis Lecture,
York University.
Mairesse, J., Hall, B., 1996, “Estimating the productivity of research and development: an exploration of GMM methods using data on French and United Sates manufacturing firms”, NBER Working Paper 5501.
Milliou, C., 2004, “Vertical integration and R&D information flow: is there a need for ‘firewalls’?”, International Journal of industrial organization, No. 22, pp. 25-43. Miyazaki, H., 2009, “An analysis of the relation between R&D and M&A in high-tech industries”, Applied economic letters, No. 16, pp. 199-201.
Porter, M.E., 1987, “From competitive advantage to corporate strategy”, Harvard
Business Review, 56 (3): 28-46.
Roll, R., 1986, “The hubris hypothesis of corporate takeovers”, Journal of Business, Vol. 59, No.2,197-216.
Salter, M., Weinhold, W., 1979, “Diversification through acquisition: strategies for creating economic value”, New York: Free express.
Servaes, H., 1991, “Tobin’s Q and the gains from takeovers”, Journal of finance, Vol. 46, NO.1, pp. 409-419.
Sevilir, M., Tian, X., 2011, “Acquiring innovation”, working paper series, AFA 2012 Chicago Meetings Paper.
Spiegel, M.R., Stephens, L.J., 1998, “Schaum’s outline of theory and problems of statistics”, McGraw-Hill.
Tremblay, V.J., Tremblay, C.H., 1988, “The Determinants of Horizontal Acquisitions: Evidence from the U.S. Brewing Industry,” Journal of Industrial Economics, Vol. 37, pp. 21-45.
Varian, H.R., 1988, “Symposium on takeovers”, Journal of Economic Perspectives, 2: 3-5.
Warner, A.G., Fairbank, J.F., Steensma, H.K., 2006, “Managing Uncertainty in a Formal Standards-Based Industry: A Real Options Perspective on Acquisition Timing”,
Journal of Management, 32(2), pp. 279-298.
Williamson, O. E., 1975, “Markets and hierarchies: Analysis and antitrust implications”, New York: Free Press.
7．Appendix
Appendix 1: Descriptive statistics of variables regression model 1
Variable Description Mean Standard
Deviation
Median Variance
MA M&A propensity, measured by 0, 1 0.196 0.397 0 0.157 NOMA Number of M&A engagement of a firm 0.898 8.068 0 65.099 RDI R&D intensity, measured by R&D
expenditure divided by total assets
0.058 0.066 0.039 0.004 LTA Natural logarithm of total assets in US
dollars
10.884 2.424 10.715 5.874 LNE Natural logarithm of number of
employees of a firm
5.662 2.203 5.316 4.852 SCF Free cash flow divided by 10000 in US
dollars
1.067 6.538 0.016 42.745 DEBT Debt ratio, measured by total debt
divided by total assets of a firm
0.480 0.258 0.469 0.067 ROA Return on assets 1.606 14.885 3.66 221.558
Appendix 2: Descriptive statistics of variables regression model 2
Variable Description Mean Standard
Deviation
Median Variance
MAA M&A activities, measured by scaled
number of M&As of a firm 2.052 12.103 0 146.490 RDI R&D intensity, measured by R&D
expenditure divided by total assets
0.071 0.060 0.056 0.004 LTA Natural logarithm of total assets in US
dollars
12.703 2.044 12.534 4.179 LNE Natural logarithm of number of
employees of a firm
7.111 2.011 6.935 4.044 ROA Return on assets 1.105 14.867 3.77 221.030 FA Firm age, measured by days between
date of completed M&A and date of incorporation
29.745 24.424 23 596.554
LC Natural logarithm of average labour costs, measured by total wage and remuneration divided by number of employees
3.064 1.286 3.332 1.654
Appendix 3: Selected literature summary
Author Year Variables Conclusions/Propositions Bertrand and
Zuniga 2006 Dependent-M&A activity
Independent
-R&D expenditure
Control
-Taxation rate
-Average labour costs -Import penetration rate of the industry -Time dummy -Industry dummy -Country dummy
1) Domestic and cross border M&A have different impact on industrial innovative performance.
2) Last M&A wave did not have any significant impact on domestic R&D activities at the aggregated industrial level.
3) Domestic M&A stimulates R&D activity in low-technology intensive industries.
4) Domestic M&A diminished R&D investment across OECD nations in medium-technology intensive industries. Bertrand 2009 Dependent -R&D expenditure Independent -M&A Control -Market share -Export intensity -Firm size -Advertising intensity -R&D skill -Capital intensity -Debt -Profitability
1) The acquisitions of French firms by foreign companies strongly increase the level of R&D budget. They have a positive and significant impact on both the internal and external R&D expenditures of French acquired firms.
2) R&D is more contracted out to domestic providers, especially to local public laboratories and universities.
3) The increase in the R&D budget appears to be financed by a rise in internal resources, but also by foreign external partners, and parent
companies finance more affiliate R&D. Blonigen and
Taylor 2000 Dependent-M&A activity
Independent -R&D intensity Control -Firm size -Profitability -Debt ratio -Free cash flow
1) There is a negative relationship between R&D activity and M&A activity. Firms with relatively low R&D intensity are likely to acquire.
2) Acquisition may be used as short term or long term strategy.
Bresman et
al. 1999 Dependent-Knowledge transfer
Independent
-Communications -Visits and meetings
1) Communications and visits and meetings were significant predictors of technological know-how transfers.
-Articulability of knowledge -Time elapsed
Control
-Firm size
3) Beyond the quantity of transfer, the quality and type of transfer changed as well. Knowledge transfer from acquiring to acquired unit in early stage, but knowledge transfer in both direction in the later stage.
4) Knowledge transfer process in acquisitions is distinctly different from the process under other modes of governance, because of the
rapidly-evolving relationship between the two parties. Cassiman et al. 2005 Dependent -5 principle component questions on R&D inputs -3 principle component questions on R&D outputs -2 principle component questions on R&D performance -3 principle component questions on R&D organization and management -2 principle component questions on R&D mission Independent -Number of M&As Control -M&A type -Industry dummy
1) The ex-ante-relatedness between merger partners matters and that market- and technology relatedness have important
separately identifiable consequences for the
impact of a M&A on the new entity’s R&D and innovation process.
2) The underlying drivers of the aggregate effects on R&D inputs, outputs and performance can be quite different depending on the
ex-ante-relatedness of partners.
3) When merged entities are technologically complementary, they increase their R&D efficiency, while merged entities which are technologically substitutive decrease their R&D inputs after the M&A.
Cefis and Marsili 2006 Dependent -Survival probability Independent -Firm innovation Control -Firm age -Firm size -Industry/sector
1) Innovation increases the probability of a firm survives in an industry, and the effect becomes more pronounced over longer time period. 2) The impacts of innovation on survive probability are more important for small and young firms.
3) The survival probability of small and young innovators is higher than those non-innovators. Cefis et al. 2007 Dependent
-R&D expenditure -Firm’s sales due to new products
1) M&A activities have positive and significant impact on R&D investments and innovation expenses.
-R&D efficiency Independent -M&A activities Control -Firm size -Firm age
-Interaction size and age
their markets, M&As do not reduce the incentive or the willingness to innovate.
3) Potentially dominant firms use M&A to build new competences and capabilities or to expand the existing ones in order to innovate.
Cooldt et al. 2006 Dependent -Post innovative performance Independent -M&A activities Control -Firm size -Industry -Nationality -Cultural distance -Time -Unobserved heterogeneity
1) Non-technological M&As have a negative impact on post-M&A firm innovative performance.
2) Technological M&As have a positive impact on post-M&A firm innovative performance, but the effects depend on firms’ ability to integrate the knowledge.
3) Firm size has positive relationship with post-M&A innovative performance.
4) Cultural distance has positive and significant relationship with post M&A innovative
performance, which indicates international M&A has positive impact on innovative performance.
Evangelista et al.
1998 Case study 1) The largest part of firms’ expenditure for innovation is linked to the adoption and diffusion of technologies through machinery and equipment, which absorbs 50% of the firm’s innovation costs.
2) R&D activities are an important component of firms’ technological activities which account 20% of total firm expenditure. Harrison et al. 1991 Dependent -Return on assets Independent -Capital intensity difference -Debt intensity difference -Interest intensity difference -R&D intensity difference
1) Firms seeking synergy should acquire target firms with the greatest similarities.
2) Differences in resource allocation patterns may provide unique and valuable synergy. 3) There is no support that acquiring and target firms with similar resource allocation could create valuable synergy.
Lehto and Lehtoranta
2003 Dependent
-Number of purchases per firm (M&A
activity) Independent -R&D intensity Control -Turnover at fixed price -R&D stock -Number of firms in the same industry -average R&D stock of other firms in the same industry -debt ratio -Profit ratio
2) Firms do not seem to specialize in their strategies, either invest in their own R&D or to buy innovations from the market.
3) R&D investments raise the firm’s attraction as a target of M&A. Miyazaki 2009 Dependent -M&A activity Independent -R&D intensity Control -Tangible assets -Cash flow -Debt ratio
-Time dummy variable
1) There is a positive relationship between R&D investments and M&A in Japanese high-tech industries.
2) Firms are expecting to obtain synergy or higher level of R&D intensity by announcing their willingness to obtain new knowledge and the capacity to obtain it and by confirming their bargaining power when the terms of M&A are negotiated. Sevilir and Tian 2011 Dependent -Post-M&A firm innovation performance (Patent) Independent -M&A volumes Control -R&D intensity -Capital expenditure intensity -Firm size -profitability -asset tangibility -Leverage -Growth opportunity (Tobin’s q)
1) There is a positive relationship between M&A volume and innovation output.
2) The relationship between M&A and innovation output is stronger when firms are more mature and older.
3) Acquiring innovation is a more efficient strategy for mature and old firms than investing in innovation internally.
4) M&A activity leads to superior long-term stock price performance if target firm is moving innovative.
