The effect of R&D on the market value of companies in Europe

(1)

30-06-2020

R&D and Market Value

The effect of R&D on the market value of companies in Europe

Marijn Tielken 11331151

(2)

1

Abstract

This thesis investigates the effect of R&D expenditures on the market values of companies in Europe over a 9-year period, covering 670 firms. There has been made a distinction between the R&D investments- and firm performance linkage of manufacturing and service industries. This thesis also focusses on this difference between Western and Eastern Europe. After controlling for size, leverage, earnings per share, ROA, the conclusion is that R&D does not have a positive effect on the market value in the same year, but it does over a couple of years, because of the uncertainty of the benefits. Moreover, service firms seem to have a better response of R&D on market value than manufacturing firms. Finally, Eastern European firms have a worse response of R&D on market value, comparing to Western European companies.

(4)

3

1. Introduction

Firms have to spend money to earn money. R&D is like the blood of a company and they cannot survive without it. However, when it comes to a Research and

Development department, most of the time there is not an immediate payback. An R&D department focuses on investigating the already existing products and

developing new products (Stolowy & Ding, p. 299). R&D is part of the intangible assets of a company.

According to the accounting rules, R&D expenditures must be recognized and expensed when they incurred. This might have a bad influence on the market value of a company, since R&D expenses are usually related to future periods when the benefits for R&D will accrue. However, the future economic benefits that R&D potentially contributes to the firm value are not sufficiently objectively defined or certain to flow to the enterprise (Stolowy & Ding, p. 300).

Firm value is the value that an investment community is willing to give to a company. It is calculated by multiplying the outstanding shares with the value per share. For publicly traded companies, the owners of shares can trade the stocks independently, this results in the market price of a stock. These market prices of the shares are based on expectations of the

firm’s returns the potential and current shareholders see in the share. Each of these investors estimates the future growth of the net return, based on the past

decisions of the management, and their taking into account strategic and external parameters about technology, markets and competition. Since the share value, and thus market value, depends on the expectations of individual investors and future parameters, it is difficult to establish a firm’s market value (Stolowy & Ding, p. 25).

Regarding the effect of R&D on market value, several studies showed that R&D intensity contributes to a significant growth in sales and market value of a company (Stolowy & Ding, p. 307). Contrarily, other studies show that the management decisions regarding R&D are more based on rules-of-thumb than on actual objective evidence. According to them, budgets for R&D are discretionary variables and no guarantee for success. For example, pharmaceutical companies require a very high R&D investment to develop a new molecule and spending that much might not always result in the expected results. Stolowy and Ding (2017, p. 307) also state that many academic studies have shown the relationship between a company’s market value and

(5)

4 their R&D investment in the US and UK.

Therefore, it is important for the

management to inform their shareholders of the R&D expenses of the company.

Over the last couple of years, many companies in the Netherlands have spent more on R&D, especially in education, but also in the trade and industry (nu.nl, 2017). However, in Europe, the European

Commission is concerned about the R&D investments. Especially private companies stay behind with their investments,

compared to the United States and China. Where Europe is good for around 20% of the world-wide private R&D investments, the US and China score 28% and 24% higher. Contrarily, with the public investments, Europe scores higher with their 23% globally. The European

commission also pointed out that Europe is behind on the dissemination of knowledge.

Moreover, there exist less studies of R&D in Europe, comparing to the US and the UK. This might be caused by the fact that in Europe, the accounting and fiscal regulations do not require data on

R&D expenditures, so there is also less data available. For patents, a different kind of intangible assets, there is no centralized patent system, which increases the costs (Hall et al, 2007, p. 3).

Based on these previous studies, this thesis composed the following research question: Do public limited liability companies in Europe with a high R&D have higher market values? This thesis will answer this question with three hypotheses. The first is more general and says something about the effect of R&D on firm value. The next one focuses on the difference of this effectiveness between manufacturing and service industry. The final hypothesis is about R&D in Eastern Europe and if those expenditures are more profitable than in Western Europe. It contributes to the already extant literature, because there has been less research done on the effectiveness of R&D expenditures in Europe. Also, the results on the

difference between manufacturing and non-manufacturing industries are inconclusive.

(6)

5

2. Literature

Market value is the value an investment community is willing to give to a company. It is often used for the market capitalization of a publicly traded firm (Stolowy & Ding, 2017, p. 25). In short, this is called the market cap. It is

commonly calculated by multiplying the current share price with the number of outstanding shares. The share price is the net present value of the expected stream of cash flows. Because it is about the

expected stream, the value of the share is based on the future parameters as

estimated by individual shareholders. So, the value of a share is different for every shareholder, because no two investors ever agree exactly on the value of those future parameters. The value of a share, and therefore the market value, is based on future elements, such as sales, while net worth is based on elements from the past (Stolowy & Ding, 2017, p. 25).

Several studies show that R&D intensity contributes to a significant grow in the sales and thus market value of a company (Stolowy & Ding, 2017, p. 307). However, the study of Hunter, Wyatt and Webster (2012) highlight that corporate spending decisions regarding R&D are more likely to be based on guidelines, instead of objective evidence. They argue that R&D budgets are discretionary

variables and are no guarantee for success. For example, pharmaceutical companies require an average spending of $1 billion on R&D to develop a new molecule, spending that much does not alway yield the expected results (Stolowy & Ding, 2017, p. 307).

Furthermore, according to Chen and Ibhagui (2019), the Tobin’Q provides a more accurate measure to get insight into a company’s financial situation. The Tobin’s Q does not suffer from scale biases that the method of market value added may have, because there is still a gap between the current capital market and a perfectly efficient market, so there are potential flaws with adapting market benchmarks (2019, p. 4).

The Tobin’s Q is calculated by the market value of a company divided by the replacement cost of its assets. This ratio expresses the relationship between the market valuation of a company and its intrinsic value. It is a mean to evaluate if a company is overvalued, undervalued or well-priced (Lang & Stulz, 1993, p. 2). Chen and Ibhagui (2019) studied the relationship between R&D and firm performance. As a measure of firm performance, they used the Tobin’s Q. They found there is a nonlinear

(7)

6 performance. Also, they found that firms

with high R&D intensity do not necessarily outperform firms with low R&D intensity (2019, p. 14).

Another study that used the

Tobin’s Q as a proxy for firm performance was done by Efstathios and Varsakelis (2010). They investigated the

consequences of R&D investment on firm value in emerging markets and used evidence from manufacturing and

computer firms listed on the Athens Stock Exchange. Their empirical results show that there is no significant difference between Greek’s R&D investment effect and that effect of US or European

companies. Also, they found the impact of R&D investment on the market value of a firm to be higher for small companies (2010, p. 359). This is in line with the theory of Syed Zulfiqar Ali Shah (2008), but in contrast with the theory of Chen and Ibhagui (2019). Chen and Ibhagui argued that larger firms might have an advantage over smaller firms, because they have larger financial and economic resources and market size (2019, p. 4).

Ehie and Olibe also investigated the effect of R&D expenditure on firm value (2010). They studied this association among US firms over a timespan of 18 years covering 26,500 firms. They examined the linkage between R&D investment and firm performance in the

manufacturing and service industries to distinguish their relative contributions to firm value. The study finds that there is a positive linear relationship between R&D investment on the market value of a firm (Ehie and Olibe, 2010, p. 134). Regarding the difference between manufacturing and service firms, they found that for both sectors about 14% of the variations in firm value can be attributed to R&D intensity, firm leverage, firm size and industry concentration (2010, p. 131). They also investigated the impact of a major economic disruption on the R&D

investment relative to firm’s performance. In 2008 there was a global

financial crisis and Lome et al. (2016) studied the effect of high R&D intensity on performance during the financial crisis in Norway, because little research has been done about the importance of R&D during a recession. They found that companies with high R&D intensity performed significantly better than other firms during a financial crisis. This connection was even stronger than the effect they found during times with normal growth, thus the importance of R&D is highlighted during a financial crisis. This might be caused by the fact that during a crisis, the dynamic capabilities of a company are accentuated, because it is a time with high external change. Dynamic capabilities are referred to as the firm’s ability to integrate, build,

(8)

7 and reconfigure external and internal

competences to address fastly changing markets. Since R&D contributes to the building of firm’s dynamic capabilities, R&D investments may make the firm more suited to handle a financial crisis (Lome et al., 2016, p. 66). Lome et al. also addressed the time lag between R&D investment and effect on revenue. On this subject they concluded that this average time lap is two years (Lome et al., 2016, p. 75).

In contrast with the study of Ehie and Olibe (2010), the study of Wang, Zhao and Voss (2016) found that for companies who want to improve their innovation capabilities, they have to invest in their technological capabilities or collaborate with their suppliers. It appeared that service firms were better at supplier collaboration. However, there was little difference between manufacturing and service firms for the technological capabilities. In short, they concluded that both manufacturing and service firms need to invest in their technological capabilities and supplier collaboration. However, for manufacturing firms, technological capabilities seem more important and for service firms, both are equally important (Wang, et al., 2016, p. 227).

Regarding the positive effect of R&D on firm growth, García-Manjón and Romero-Merino studied the effect for

European top R&D spending firms (2012, p. 1084). They found the same results as Ehie and Olibe (2010), so there is a positive effect of R&D on growth of companies. However, they found different results regarding the sectors. Namely, their study provides empirical evidence that R&D is essential only for some types of sectors. They observed that only the high-technology sector obtains clear growth from their R&D investment (2012, p. 1090).

Hall, Thoma and Torrisi (2007) studied the effect of patents and R&D on the market value. As García-Manjón and Romero-Marino, they also focussed mainly on Europe and compared the results with the United States. The focus of their study relies mainly on the patents in Europe, as Europe does not have a centralized patent system and this increased the costs for applications of patents, compared to the US (2007, p. 3). Regarding R&D, they found that most studies on R&D are focussed on US and UK firms and less on European firms. Moreover, most European fiscal and accounting regulations do not require data on R&D expenditures, so there is less data available for Europe, comparing to the US and UK (2007, p. 5).

The study of Keller et al. shows that R&D in Eastern Europe might be lucrative, because of lower salaries and

(9)

8 enhanced access to large and lucrative

market, compared to Western Europe (1995, p. 316). However, problems with bureaucratic structures within companies and risk-averse management might also arise more in Eastern Europe (1995, p. 316).

A more recent study about the efficiency of innovation systems in Eastern Europe was done by Kravstova and

Radosevic (2011). They studied the extent to which systems in Eastern Europe could be considered (in)efficient (2011, p. 109). They formed three conclusions on their research regarding R&D. First, Eastern European countries lost advantages regarding the size of R&D they inherited from the socialist period. Second, the R&D employment in combination with the production capacity explain these

differences in production. And third, they found that Eastern European countries have lower levels of productivity than would be expected given their production capabilities and R&D capacities (2011, p. 122).

Gupta et al. (2017) did an international study about the effect of R&D intensity on the value-creation of a firm. They used a database with 75 countries spanning 2004-2013. There focus relied mainly on country-specific characteristics and the change firm value, depending on these variations. Moreover,

they calculated the marginal effect of R&D intensity on firm value for high income and low-income countries separately, based on different degrees of competition. Their findings suggested that, for developing countries, the positive effect of R&D exists mostly for firms in a less competitive market. For developing countries, the positive effect was the same for competitive and less competitive industries (2017, p. 402).

In contrast with the studies of Keller, García-Manjón and Ehie, Wang et al. studied the effects of R&D policy choice on the market value of companies in China (2016, p. 545). They tested for the period between 2007 and 2014 and found that the choice of R&D policy affects a firm’s accounting performance and market value differently. Companies that choose to capitalize their R&D can increase market value but decrease their accounting performance. When firms choose to capitalize their R&D they focus on long-term projects and sustainable developments. Contrarily, firms that choose to expense their investments in R&D improve their accounting

performance and focus on short-term performance (Wang et al., 2016, p. 553). Also, they found that investing in R&D gives a positive image to the shareholders and other investors about the accounting

(10)

9 performance of the company (2016, p.

554).

Contrarily, the study of Syed Zulfiqar Ali Shah et al. shows that even though R&D is expensed, the market realizes that such investments are positive for the future and responds positively to those investments in R&D for firms of all sizes in the UK (2008, p. 87). They investigated effects of firm size and R&D expenses on the market value of

companies in the UK. They also found no clear advantage for large firms, compared to smaller sized firms. Finally, they found that the valuation of R&D is significantly positive for manufacturing, as well as non-manufacturing firms (2008, p.90).

Shevlin investigated whether investors in the capital market look at if R&D limited partnerships increase both the assets and liabilities of R&D firms, when assessing the market value of a company (1991, p. 1). He does so by interpreting the limited partnerships in an option pricing structure (1991, p. 7). He concludes that investors do not only look at the assets and liabilities included on the balance sheet of a company, but also at the footnotes that include R&D. This is in line with other literature. Also, he agreed with Syed Zulfiqar Ali Shah that investors in general capitalize R&D investments, even if companies expense these costs (1991, p. 18).

The study of Fantino investigated the interaction between R&D and

economic system (2010, p. 3). First, he studies the interaction between R&D and market value in a horizontally integrated market. In a horizontally integrated market, firms invest to differentiate their products. He concluded that R&D activity declines over time and that the production of non-differentiated products decline. This increasing specialization improves the utility of the consumers (2010, p. 59). Moreover, he found that companies in general underinvest in R&D. Second, he investigated whether the incentives of the Italian Ministry for Economic

Development increase the incentives to development of innovations. For this case, he did not find substantial evidence. The firms, however, did use the subsidies to finance current expenditures, increasing the assets (2010, p. 150). Third, he examined the relationship between R&D and market values of the firms. He investigated this by looking at

manufacturing firms in the US between 1975 and 1995 and found very small or negative effects of current R&D on the market value. So, he developed a model with uncertain R&D, dividing it in two compartments: work-in-progress R&D and already concretized assets. Risk aversion of investors might cause different

(11)

10 the investors are risk-averse and the

management team maximizes the long-term market value, the risk associated with the work-in-progress R&D reduces short-term market value, even though the long-term value increases (2010, p. 205). This is in line with the theory of Stolowy and Ding (2017).

Kuzikaite studied whether R&D and patents are important variables for the firm’s market values in Sweden (2013, p. 1). She stated that it is a difficult task to obtain data on the R&D activities of a firm, because not all companies publish their R&D expenses and not all firms even engage in R&D. She first tested whether patents and R&D can be associated in the context of Swedish firms. The results showed a positive and significant

association, but there was a lot of noise, so this result is questionable. Second, she investigated the effect of R&D and patents on the firm’s market value. This resulted in that R&D is inversely related to the market value and that patents did not show a significant relation to the value. She then applied the same study to the health

industry, because some industries are more dependent on innovation than others and in the health industry, a lot of innovation is needed. For this test, she found that both R&D and patents are positively related to the variations in the market value of companies in Sweden (2013, p. 37).

Finally, Hsu and Schwartz developed a real options model for R&D valuation, taking into account the time and cost of completion, the market demand for the R&D output and the uncertainty of the quality of the output (2003, p. 1). They applied this model to study the under-investment of R&D for vaccines in developing countries.

So, there has been a lot of research done on this subject with different results. However, another important subject within companies is the corporate social

responsibility. Lin (2017) studied the effect of R&D investment on firm performance under corporate social responsibility. Corporate social responsibility is a business model that helps a firm to be socially accountable, for itself, its stakeholders and the public. Engagement in Corporate Social

Responsibility could enhance production quality and restore society’s confidence (Lin, 2017, p. 217). The study of Lin (2017) investigates whether engagement in CSR strengthens the effect of a firm’s R&D expenditure on company’s performance for firms in Taiwan. She argued that the long-term profitability of firms was created by in-house R&D activities. In line with the previous literature, she also states that R&D investment has an impact on a firm’s operations and performance. She used

(12)

11 Tobin’s Q as a measurement for firms’

performance and she found that R&D investment has a positive effect on the firm’s performance, but only when a firm engages in CSR (Lin, 2017, p. 222).

Based on this previous extant literature this thesis formed the three hypotheses. First, this thesis focussed on the more general topic whether R&D has a positive effect on the market value of companies in Europe. Market value is difficult to establish, because it depends on the value of the shares and that value is mostly determined by the shareholders, who have different opinions about the future value of the company. Therefore, this thesis will use the Tobin’s Q as a proxy for market value. Europe is interesting, because less literature is written about Europe and also less data is available, as R&D expenses are not

required by European accounting and fiscal regulations.

This thesis second hypothesis is about whether there is a difference in the added value of R&D expenditure on the firm’s value in the manufacturing firms and service firms. Previous literature shows that service firms generally have less R&D expenses than manufacturing firms, because it is a less innovative industry.

Third, this thesis tested if the effect of R&D expenditure is higher in Eastern Europe. Eastern Europe used to be communist and the wages of the researchers are therefore lower, which contributes to lower R&D costs. Also, there is easier access to large and lucrative markets, such as Asia. However, there might be problems with the more bureaucratic structure within companies and more risk-averse management.

(13)

12

3. Methodology

To answer the research question “Do public limited liability companies in Europe with a high R&D have higher market values?” This thesis composed three hypotheses. There has been chosen for public limited liability companies, because more data is available for those firms. Also, this thesis excluded data of companies that went bankrupt or where not all relevant information is available.

3.1. The effect of R&D on firm value

The market value of a company is hard to estimate, because it depends on the estimations of the shareholders and future parameters (Stolowy & Ding, 2017, p. 25). The net worth is based on values of the past. Several studies show that R&D has a significant effect on the sales growth and thus the market value of firms. However, other studies showed that the decisions on R&D are more based on guidelines, instead of hard evidence. This suggests that there is a positive effect of R&D on firm value, but that there are also controversial thoughts. Therefore, this thesis formed the first hypothesis:

H1. Investment in R&D has a positive

effect on the market value of the firm

3.2. The effect of R&D in manufacturing and service firms

Ehie and Olibe studied the effect of R&D on the firm value in the US over a

timespan of 18 years and covering 26,500 firms (2010, p. 129). They investigated the difference of the effectiveness in R&D investment in manufacturing and service firms, because little attention has been given to this difference in previous

research. Manufacturing and service firms choose different strategies for their R&D investments, because manufacturing companies produce tangible products that are more distinguishable than the services service firms provide. They tested whether manufacturing firms adopt a more R&D intensive strategy for increasing the firm’s value, in comparison to the service firms. Also, García-Manjón and Romero-Merino provided with empirical evidence that the effectiveness of R&D is different per industry. Accordingly, this thesis tested the following hypothesis:

H2. The market value effect of R&D

investment is significantly more positive in manufacturing firms than in service firms.

(14)

13

3.3. the effect of R&D in Eastern and Western Europe.

Hall et al. (2007, p. 5) found that most studies on R&D are focussed on the UK and the US firms and less on European firms. Because, in Europe, fiscal and accounting regulations do not require any data on R&D expenditures, so there is less data available for Europe, comparing to the UK and US. Moreover, Keller et al. (1996, p. 316) shows in his study that R&D in Eastern Europe might be lucrative, because of relatively lower salaries, compared to Western Europe, and enhanced access to large and lucrative markets. However, the bureaucratic structures within those companies might cause problems and risk-averse

management might also arise in Eastern Europe. Based on these findings, this thesis stated the final hypothesis:

H3. The market value effect of R&D

investment is significantly more positive in eastern Europe than in western Europe

3.4 Regression

To test my hypotheses, this thesis uses a linear regression and compute this linear regression using OLS. Following, this thesis tests the significance of the

coefficients of the different variables using a t-test in Stata.

3.4.1. Dependent variable

As a dependent variable this thesis chose the Tobin’s Q as a proxy for the market value. This provides an accurate and simple way to gain insight in the company’s financial situation (Chen & Ibhagui, 2019, p. 4). It is always difficult to estimate the market value of a company, because it depends on the estimates of shareholders on future values. The Tobin’s Q suffers less from scale biases, since there is still a gap between the perfect market equilibrium and the current capital market (Chen & Ibhagui, 2019, p. 4). The Tobin’s Q ratio equals the market value of a firm divided by the replacement costs of its assets. The ratio expresses the relationship between a company’s intrinsic value and its market valuation. It is a mean of estimating whether a firm is overvalued, undervalued or well-priced. In

equilibrium, the market value is equal to the replacement costs, so Tobin’s Q is one.

(1) 𝑇𝑜𝑏𝑖𝑛&_{𝑠 𝑄 =}

+,-./ 0.123- 4./53 ,6 -73 6810 +,-./ .993- 4./53 ,6 -73 6810

3.4.2. Independent variable

As mentioned before, this thesis uses R&D expenditures as the main explanatory variable. For this, this thesis divides R&D

(15)

14 by the net sales of a company to check for

the proportion of R&D relative to net sales of a company. This is in line with previous research. Ehie and Olibe (2010, p. 130) used the R&D expenditures instead of the capitalized R&D, because of the inability to have an economic driven amortization rate. Furthermore, Chen and Ibhagui also used R&D expenditures over total sales. R&D intensity is favoured, because an absolute R&D measure only relates to firm size and may confound the relationship between firm performance and R&D investment (2019, p. 3). So, based on extant literature this thesis used R&D over Net sales as the independent variable.

(2) 𝑅𝐷𝑒𝑥𝑝 =𝑇𝑜𝑡𝑎𝑙 𝑅&𝐷 𝑒𝑥𝑝𝑒𝑛𝑑𝑖𝑡𝑢𝑟𝑒𝑠

𝑁𝑒𝑡 𝑠𝑎𝑙𝑒𝑠

3.4.3. Control variables

Based on the previous research, this thesis came to the following control variables to get a more accurate estimate of the market value the firms. Also, this thesis chose these variables to minimize the correlation between them and still give a valuable picture of the market value. The control variables are the following: Size, EPS (Hsu & Schwartz, 2016, p. 247), country, industry, ROA, total intangible assets per share and leverage. The size is measured by the natural logarithm of total assets. A

logarithm is not only easier to interpret, because they represent relative changes, but it also makes the distribution of the data closer to a normal distribution (Ehie & Olibe, 2010, p. 130). Size is included to control for economic scale, since larger companies are superior to small

companies, regarding financial resources, market size and economic resources (Chen & Ibhagui, 2019, p. 4).

Industry is included as a dummy variable and is important for my analysis, because according to García-Manjón and Romero-Merino (2012, p. 1084) stated that the R&D investment and firm’s growth is dependent on industry. For example, they argued that there is a positive response of R&D investment on firm value for high-tech industries, but a negative relationship for low-tech industries (2012, p.1085). In line with the studies of Chen and Ibhagui (2019) and Ehie and Olibe (2010), this thesis included the return on assets as another control variable as a proxy for firm performance. They showed there is a negative correlation between R&D intensity and ROA, which suggests that R&D investments negatively affects accounting performance. Finally, based on the studies of Hsu and Chen (2017), Ehie and Olibe (2010) and Chen and Ibhagui (2019), this thesis used leverage as another control variable. These studies have shown that leverage affects R&D intensity. This

(16)

15 variable is also used to control for

variances in firm valuation due to differences in capital structure. Bigger companies with more financial resources that have an appropriate debt ratio can reduce the company’s operating costs, which increases the short-term firm’s

performance (Chen & Ibhagui, 2019, p. 4). Leverage is calculated by the

debt-to-equity ratio to control for this effect. Mainly, this thesis used the variables country and industry to test my second and third hypotheses.

This results in the following final formula: (4) Tobin’s Q= α0+β1RDexp+

β2EPS+β3ROA+β5Country+β6Industry+β7 Leverage+β8Size +β9TIAPS+ε

Where EPS means Earnings per share, country and industry are dummy variables, and TIAPS stands for Total Intangible Assets per share.

(17)

16

Data

4.1. Data preparation

For the regression of this thesis a sample is used of 9,253 companies in Europe with information available about the R&D expenses over the timespan 2010-2018, because the data of 2019 was not yet available. So, the data of 2010-2018 is the most recent data and doesn’t include the financial crisis. This thesis has removed the companies where not all data was available, so there is a clean sample. After removing all the companies where not all data was available, there were 2,698 companies left. This thesis also removed all the companies for which R&D was zero, there were 721 companies left after that. The OLS method is used to estimate the regression. This thesis tested the significance of the coefficients by a standard t-test in STATA.

This thesis has selected some industries that were excluded. Industries that are identified with the SIC codes 4000-4999 and 6000-6999 are excluded from the sample of the European

companies. The industries with these SIC codes are heavily regulated by the

government or are financial industries. The firms in these industries may not be able to determine their own strategy or choices; this could influence the financial data of

the different countries from Europe in this sample (Ehie & Olibe, 2010). By

excluding these industries, the sample of this thesis is further reduced, the sample of deals is reduced to 670 firms.

4.2. Testing data for normal distribution

To test for a normal distribution, the Shapiro-Wilk test is used. This test tests whether a sample is normally distributed in statistics. The null hypothesis is that the sample is normally distributed and when it is rejected, it means there is not a normal distribution.

Significant for: *** p<0.01, ** p<0.05, * p<0.1

Table 1.1 shows that the p-values of the variables are all significant, which means that the variables are not normally

distributed. This is as expected for financial research, since finance data is known for its heteroskedasticity (Li & Hong, 2010). Because the variables are

(18)

17 not normally distributed, robust standard

errors in Stata are used. In appendix 2.1 these tests are shown for the years 2010-2017, which show the same results.

4.3. Testing data for correlation

This thesis also checks for correlation between the variables, using a correlation matrix. The correlation matrix is a table that shows the correlation coefficients between variables. Each cell of the table shows the correlation between two variables. When the different coefficients highly correlate with each other, when the value is close to one, it suggests that linear regression estimates will be unreliable. When the correlations are zero, it means there is no correlation at all.

Table 1.2. shows that there are no high correlations between the different estimates, which means the coefficients of the linear regression are reliable. Also, a lot of correlations are negative, which means that one variable increases when the other one decreases, and vice versa. In this thesis leverage, size, intangible assets per share and Eastern Europe have a negative

relation to the Tobin’s Q. So, when these values increase, the Tobin’s Q, or market value, decreases. For leverage this is logical, namely when a company has more debt, the market value goes down. The correlation between size and Tobin’s Q is also negative, which is in line with previous research (Efstathios &

Varsakelis, 2010). The matrix also shows that when there are higher total intangible assets per share, the Tobin’s Q goes down, which means that shareholders expect lower future market returns when there are more intangible assets. This might be caused because of the fact that intangible assets have uncertain returns. The

correlation between Eastern Europe and Tobin’s Q is also negative, which means that companies in Eastern Europe

generally have a lower market value. The correlations for earnings per share, research and development, return over assets and manufacturing firms are positive, which implies that when these values go up, the Tobin’s Q will also increase. This is in line with the expectations. Appendix 2.2. shows the

(19)

18 correlation matrices for the years of

2010-2017, which has the same results.

4.4. The effect of R&D on market value

Table 2.1. shows the linear regression of the effect of research and development on Tobin’s Q. As mentioned before, this thesis uses robust standard errors to correct for the fat tails. The different robust

standard errors are small, which means there is low variance and the regression is more reliable. The coefficients are so small that they seem zero in the table, but they have positive and negative values. These small coefficients are less meaningful, but they show a small but significant effect. The earnings per share, leverage, size, intangible assets per share and Eastern Europe are significant, which means they

all have a significant effect on the Tobin’s Q and thus market value. For the earnings per share the coefficient is positive and significant, which means that for

companies in Europe, the market value of a company is higher when the EPS are higher. Leverage has a significant and negative effect. This implies that when leverage goes up, the market value of a company decreases. Size also has a negative and significant effect, which means that the effect of R&D on market value is larger for smaller firms than for larger firms. This is in line with the theory of Efstathios and Varsakelis (2010), who found the impact of R&D investment to be higher for small companies, but in contrast with the theory of Chen and Ibhagui (2019). The effect of total intangible assets per share on market value is also negative and significant. This means that companies

(20)

19 with more intangible assets generally have

lower market values. This can be

explained by the fact that intangible assets have uncertain benefits and these benefits are usually gained over a couple years and not directly visible. The coefficients of return over assets and R&D expenditures are both not significant. The return over assets has a positive coefficient, and thus a positive effect, on the Tobin’s Q, which means that companies with higher returns have higher market values. The coefficient of R&D expenditures is negative, which can also be explained by the uncertainty of the gains of R&D, so shareholders

estimate the future cashflows lower. This is confirmed by Fantino (2010), who found that the risk associated with the work-in-progress R&D reduces short-term market value. Appendix 3.1 shows these similar regressions for the years 2010-2017 and have the same results.

4.5. The effect of R&D on market value over time

Table 5.1. shows the effect of R&D expenditures of the previous years on the Tobin’s Q of the most recent year. The years of 2010, 2014 and 2015 are not

(21)

20 significant. Again, the problem is that the

coefficients are very small, so probably not very meaningful. The coefficients of 2017, 2013 and 2011 are significant and

negative, which means that those years the R&D expenses do not have a positive effect on the market value of 2018. Contrarily, the coefficients of 2018, 2016 and 2012 are significant and positive, which implies those have a positive effect. The negative effects of 2017, 2013 and 2011 are a contradiction of the study of Syed Zulfiqar Ali Shah et al. (2008). They stated that even though R&D is expensed, the market will realize that such

investments are positive for the future and thus should increase the market value of the firms. The R&D coefficients of the other years are in line with this theory.

4.6. The difference between manufacturing and service firms regarding the effect of R&D on market value

To look at the difference between

manufacturing and service firms regarding the effect of R&D on market value, this thesis removed the industries with SIC codes 01-19 and 50-59, because these are agricultural, mining, construction and wholesale trade industries. The dummy of manufacturing is included, which has a value of 0 for manufacturing firms and a

value of 1 for service firms, so there are 616 firms left See table below.

The table shows that there are 522 companies within in the manufacturing industry and 94 companies within the service industries. The companies with SIC codes 20-49 are within manufacturing industries and the companies with SIC codes 70-99 are within the service industries. As shown in table 2.1. the coefficient of the manufacturing dummy is positive and significant, which implies that service firms have a more positive effect on the effect of R&D on market value, than manufacturing firms. This is in line with the theory of Wang, Zhao and Voss (2016), who stated that companies that want to improve their innovation capabilities, they have to invest in their technological capabilities or collaborate with their suppliers. In their study it appeared that service firms were better at supplier collaboration. Appendix 3.1 shows these same results for the different years.

(22)

21

4.7. The difference between West and East European firms regarding the effect of R&D on market value

See Appendix 4.1. for the definition of the countries. As can be seen in table 4.1., after removing the companies with not all information available or with values R&D=0, there are only seven companies with R&D from Eastern Europe left in the

regression. Because there are few

companies left, this is not a representative study anymore. However, for those few companies, table 2.1. still shows a negative and significant coefficient, which could imply that companies in Eastern Europe have less effect of R&D on market value, then companies in Western Europe, but this is not very meaningful for this small sample. This is in line with the theory (Kravstova & Radosevic, 2011). Appendix 3.1. shows the same results for the years 2010-2017.

(23)

22

7. Conclusion

This thesis is set out with an overview of how innovation has an impact on the market value of companies in Europe. Over the last couple of years, many companies have spent more on innovation and it is believed that it is an important phase in company growth and survival. However, the European commission is concerned about the R&D investments. Especially private companies stay behind on these investments, comparing to the United States and China. Very few studies have focused on the effects of R&D expenditures on market value of companies in Europe. This thesis deals with this issue using databases and computing those results in STATA.

The results show that on the short term, the R&D expenditures do not have a significant effect on the Tobin’s Q.

However, when the R&D expenditures of previous years were included, the results did show a significant effect for most years. Some of them were positive and some of them were negative.

For the difference between Eastern and Western Europe, the results showed a

significant and negative effect, which could imply that companies in Europe have less effect on the market value then companies in Western Europe. However, after removing all data that was not available, there were very few companies in Eastern Europe left, so this result is not very representative. For the results

between manufacturing and service firms, the results showed a positive and

significant effect for service firms, which means that service firms have a more positive effect on the effect of R&D on market value, than manufacturing firms.

Concluding, to answer the research question if R&D has a positive effect on market value, the results were

inconclusive. However, in previous literature, there is shown a positive effect of R&D expenditures on market value (Ehie & Olibe, 2010, and García-Manjón & Romero-Merino, 2012). The different results can be caused by the fact that for Europe, less studies have been done before and less data is available, because it is not obligatory for European companies to disclose that data (Hall et al., 2007).

(24)

23

6. Discussion

Not much research has been documented on how innovation affects European firm’s market values, so this is an open field for more studies. This study used public limited liability companies for its sample. However, because there were very few companies left in Eastern Europe after removing companies for which not all data was available, the results were less

meaningful. An idea to fix this problem is to also include privately owned companies to create a larger sample, this might also increase the small coefficients. Another

suggestion for future studies is to include more industries, this thesis only focused on the difference between manufacturing and service firms, but other industries like utilities could also be included to cover all industry classifications, to see which industry has a greater effect on R&D. Another thing that can be included in next studies is capitalized R&D, since studies have shown that R&D usually only has a positive effect on market value over a longer time period. However, it is key to find the right amortization rate in that case.

(25)

24

Appendix

(26)

25 Appendix 2.1. Shapiro-Wilk tests for 2010-2017

(27)

26

(28)

27

(29)

28

(30)

(31)

30 Appendix 3.1. Linear regressions for the years 2010-2017

(32)

(33)

(34)

(35)

34 Appendix 4.1.

Orbis defined Western Europe as the countries: Andorra, Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Gibraltar, Greece, Iceland, Ireland, Italy, Liechtenstein, Luxembourg, Malta, Monaco, Netherlands, Norway, Portugal, San Marino, Spain, Sweden Switzerland,

Turkey, United Kingdom and Vatican City. Eastern Europe consists of the countries: Albania, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Kosovo, Latvia, Lithuania, Republic of Moldova, Montenegro, North Macedonia, Poland, Romania, Russian Federation, Serbia, Slovakia, Slovenia and Ukraine.

(36)

35

References

Andzelina Kuzikaite, V. (2013). R&D, Patents and Firm Value: The Effects of Intellectual Property on Firms’ Market Values. Jönköping International Business School. Cardol, T. (21 february 2018). Private R&D investments in Europe are lagging.

https://www.scienceguide.nl/2018/02/private-investeringen-blijven-achter/. Consulted on May 22nd 2020.

Chen, Y. & Ibhagui, W. (January 2019). R&D-firm performance nexus: New evidence from NASDAQ listed firms. North American Journal of Economics and Finance. pp. 1-16 Efstathios, G. & Varsakelis, N. (July 2010). R&D and Tobin’s Q in an Emerging Financial

Market: The Case of the Athens Stock Exchange. Managerial and decision Economics, Vol. 31, No. 5. pp. 353-361.

Ehie, I., Olibe, K. (June 2010). The effect of R&D investment on firm value: An examination of US manufacturing and service industries. Int. J. Production Economics, Vol. 128. pp. 127-135.

Fantino, D. (July 2010). Innovation, activity, R&D incentives, competition and market value. The London School of Economics and Political Science.

García-Manjón, J., Romero-Merino, E. (March 2012). Research, development, and firm growth. Empirical evidence from European top R&D spending firms. Research Policy, Vol. 41. pp. 1084-1092.

Gupta, K., Banerjee, R. & Onur, I. (2017). The effects of R&D and competition on firm value: International evidence. International Review of Economics and Finance, Vol. 51. pp. 391-404.

Hall, B., Thoma, G. & Torrisi, S. (September 2007). The market value of patents & R&D: Evidence from European firms. National Bureau of Economic Research, Cambridge. Working paper 13426.

Hsu, J., Schwartz, E. (October 2003). A model of R&D Valuation and the design of research incentives. National Bureau of Economic Research, Cambridge. Working paper 10041.

Hunter, L., Webster, E. & Wyatt, A. (March 2012). Accounting for Expenditure of Intangibles. Abacus, Vol. 48, No. 1, pp. 104-145

Keller, R, Kedia, B. Julian, S. (September 1995). R&D Team productivity in Eastern and Western European countries. European Management Journal, Vol. 13, No. 3. pp. 316-321.

(37)

36 Kravstova, V. & Radosevic, S. (April 2011). Are systems of innovation in Eastern Europe

efficient? Economic Systems 36. pp. 109-126.

Lang, L. & Stulz, R. (1993). Tobin’s Q, Corporate Diversification and Firm Performance. National Bureau of Economic Research, Cambridge. Working paper 4376.

Li, H. & Hong, Y. (2010). Financial volatility and forecasting with range-based autoregressive volatility model. Finance Research Letters, Vol. 8. pp. 69-76

Lin, Y-C. (2017). Does R&D investment under corporate social responsibility increase firm performance? Investment Management and Financial Innovations, Vol. 14, Issue 1, pp. 217-226.

Lome, O., Heggeseth, A. & Moen, Ø. (2016). The effect of R&D on firm performance: Do R&D-intensive firms handle a financial crisis better? Journal of High Technology Management Research, Vol. 27. pp. 65-77.

Nu.nl (27 october 2017). Organizations spend record amount on Research and Development. https://www.nu.nl/economie/4980722/organisaties-geven-recordbedrag-onderzoek-en ontwikkeling-.html. Consulted on May 22nd, 2020.

Shevlin, T. (January 1991). The valuation of R&D firms with R&D Limited Partnerships. The Accounting Review, Vol. 66, No. 1, p. 1-21.

Stolowy, H. & Ding, Y. (2017). Financial Accounting and Reporting. Singapore: Seng Lee Press. pp. 25-307.

Syed Zulfiqar Ali Shah, Andrew, W., Stark & Saeed Akbar (2008). Firm size, sector and market valuation of R&D expenditures. Applied Financial Economic Letters, 4:2, p. 87-91.

Wang, Y., Du, R., Koong, S. & Fan, W. (2016). Effects of R&D policy choice on accounting performance and market value. R&D Management, Vol. 47, No. 4, p. 545-556 Wang, Q., Zhao, X., Voss, C. (2016). Customer orientation and innovation: A comparative

study of manufacturing and service firms. Elsevier ScienceDirect Journals. pp. 221-230.

The effect of R&amp;D on the market value of companies in Europe

R&D and Market Value