Value creation with acquisitions of start-ups : a study of acquisition programmes in the high-technology industry

(1)

Value creation with acquisitions of start-ups

A study of acquisition programmes in the high-technology

industry

Abstract

This paper studies the acquisitions of private start-ups by established public firms in the high-technology sector. The event study methodology of Peterson (1989) and Mackinlay (1997) is used to analyse whether the announcement of acquiring a start-up creates value for the acquiring firm. The period from 1/1/2014-30/5/2016 is analysed and three event studies are performed: one for all the start-up acquisitions in this time period, one for the subset of only venture-backed start-up acquisitions and the last one for the subset of start-up without venture capital acquisitions. The event studies have shown that there is no statistical evidence that acquisitions of start-ups in general create value for the acquirer. However, the event studies do show that acquiring venture-backed start-ups creates value for the acquiring firm.

Keywords: event study, M&A announcement, high-technology JEL Classification: G34

Bachelor thesis Economics and Finance Faculty Economics and Business Author: Sjoerd van Dorp

Supervisor: dhr. dr. J. Lemmen

University of Amsterdam

(2)

Table of Figures

Table 1: First Hypothesis Standard t-test ... 19

Table 2: First Hypothesis Robustness Check Standard t-test ... 19

Table 3: Second Hypothesis Standard t-test ... 20

Table 4: Second Hypothesis Robustness Check Standard t-test ... 21

Table 5: Third Hypothesis Standard t-test ... 22

Table 6: Third Hypothesis Robustness Check Standard t-test ... 22

Table 7: Fourth Hypothesis two-sample t-test with unequal variances ... 23

(4)

1

This document is written by Student Sjoerd van Dorp who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

1 Introduction

The Netherlands hosted the Startup Fest Europe a couple of weeks ago. This European festival helps start-ups grow faster by attracting business leaders, founders and investors to the festival (“A full week of”, 2016). On the one hand, there is the presence of Tim Cook (Apple) and Eric Schmidt (Alphabet) who are dominating the high technology industry with their respective businesses. On the other hand, there is an abundance of promising and talented start-ups founders present at this festival. It is no coincidence that both parties are attending the Startup Fest Europe. Start-ups and established technology firms need each other to survive. Start-ups bring the innovation that is needed to survive in the technology industry, and established technology firms help start-ups by bringing in expertise and desperately needed funding.

An illustrative example is the acquisition of Dryft by Apple in 2015. Dryft is a start-up that focusses on keyboard applications. For example, an application was that the keyboard only appeared when the customer touches the screen. Apple acquired Dryft and therefore brought in new technology to the firm by an external acquisition (Lynley, 2015). Nowadays the technology of Dryft is implemented in the devices of Apple. This acquisition made Apple more competitive, because Google already had similar keyboard applications installed in their devices.

Bruno and Cooper noticed in 1980 that the high-technology industry is characterized by a remarkably high number of private start-up acquisitions. They examined 250 Silicon Valley start-ups that started on the San Francisco peninsula between January 1, 1960 and July 1, 1969. As of 1980, 32.4% of the firms had been merged or acquired.

Nowadays external acquisitions of new technology is a growing trend in the high-technology industry; firms try to complement their internal knowledge with acquisition

(5)

2

programs (Ransbotham & Mitra, 2010). Maturing strategic buyers, such as Microsoft and Cisco, acquire start-ups in order to access new growth markets and gamble on new

technologies (Ding & Eliashberg, 2002). Ransbotham and Mitra (2010) notice that it is not well understood if the acquisition of a private start-up creates value for the acquirer. The lack of knowledge on the profitability of these acquisitions is astonishing, because it is such a vital part of the high-technology industry. The primary objective of this thesis is to understand if acquiring start-ups creates value for the acquirer. It is of interest to study the value creation of buying private start-ups, because of the lack of knowledge about the profitability of acquiring private start-ups.

There are several ways to measure Merger & Acquisition (M&A hereafter)

profitability. This thesis will use the method of event studies to analyse if high-technology firms create value with buying start-ups. The methodology set out by Peterson (1989) and Mackinlay (1997) will be employed to analyse whether the announcement of acquiring a private start-up creates abnormal returns. This leads to the following research questions: “Does the announcement of acquiring a private start-up create shareholder value for a high-technology firm in the United States?”.

This thesis is divided as follows. The next chapter is devoted to describing a theoretical context for the reader. The first part helps explaining why acquiring start-ups is such a widespread phenomenon in the high-technology sector. The second part gives an overview of recent literature that analysed similar research questions. Chapter 3 outlines the research method of Peterson (1989) and Mackinlay (1997). The second part of chapter 3 gives detailed information about the dataset that is collected. The subsequent chapter presents the results and the analyses of the event study. Chapter 5 concludes; there will be determined if acquiring a private start-up creates value for the acquiring shareholders.

(6)

3

2 Theoretical framework

The first part of this chapter describes theories and ideas that concern the key concepts in the research question. Firstly, a closer look will be given to the start-ups; what defines a start-up and why are they important? The subsequent sectionexplains why acquisitions of start-ups are happening on such a frequent basis in the high-technology sector. Furthermore, all the

prominent ways of measuring M&A profitability are summarized. The chapter concludes with an extensive literature review on my research topic and also a widely studied subject in

finance: does an acquisition create value for the acquirer?

2.1 Start-ups

This section is devoted to the literature about start-ups. For this thesis, it is of importance to make a distinction between start-ups that are venture-funded and start-ups that do not have venture capital. To make this distinction, literature about start-ups in general will be described first. Afterwards, the role of venture capital for start-ups will be set out.

2.1.1 Start-ups in general

There are multiple definitions for a start-up. Kollman, Stöckmann, Linstaedt and Kensbock (2015) argue in the “European Startup Monitor” that a start-up has to be younger than 10 years and to focus on highly innovative technologies or business models.

Startup Fest Europe emphasizes the important role that start-ups play in our society, but is there any scientific justification for the importance of start-ups? Audretsch (2002) analysed the roll of start-ups in the United States. He finds that start-ups are among the

biggest stimulators of employment growth and innovation. For example, start-ups come ahead when comparing the patenting rate on a per-employee basis with larger firms. Another finding of Audretsch is that the net employment gain for start-ups is higher than for larger firms.

Zider (1998) argues that an important feature of start-ups is that a high level of uncertainty surrounds them. Once the start-up is trying to commercialize its innovation, it is very hard to distinguish the winners from the losers, because the financial statements and forecasts are all the same (Zider, 1998). The next section shows how venture capital reduces this uncertainty among start-ups.

(7)

4

2.1.2 Start-ups and venture capital

The world recognizes the venture capital industry in the United States as an engine of economic growth (Zider, 1998). At first, it is remarkable that venture capital is the engine of growth and not the start-ups. Why is this? As noted in the previous section, start-ups are surrounded with a lot of uncertainty. This uncertainty restrains the start-up from growing, since it has a hard time attracting financing. On the other hand, the founder is most of the times not capable of growing the firm (Zider, 1998). Venture capital helps the start-up growing by supplying the necessary funding and playing an active role in the company. Therefore, venture capital plays an important role in the professionalization of start-ups (Hellmann and Puri, 2002).

First of all, venture capitalists invest in a start-up’s balance sheet and infrastructure (Zider, 1998). On the one hand, investments in fixed assets and working capital increase the size and credibility of the start-ups. This is needed to attract investments from public equity, such as institutional investors. On the other hand, investments in infrastructure, such as manufacturing and sales, help the firm grow (Zider, 1998).

Secondly, venture capitalists play a more active role than traditional financial intermediaries. Hellmann and Puri (2002) find that there is a significant relation between venture capital and organizational milestones, such as the formulation of human resource policies and hiring of a manager responsible for marketing. Aside from the organizational milestones, venture capital plays another vital role: start-ups gain attention and recognition when they obtain their first venture capital sponsor (Hellmann & Pure, 2002). This

recognition makes it easier to find new funding and also opens the eyes of potential acquirers. The period of investing and playing an active role in the start-up usually takes five years. The venture capitalist sells the start-up – most of the time – to a corporation after this timeframe (Zider, 1998). The idea is that the start-up has reached a sufficient size and credibility; therefore, it is less of a gamble for the corporation to acquire the start-up. This brings us to the high-technology sector; a sector that is famous for its high amount of start-up acquisitions.

2.2 High-technology industry & start-ups

The first step in understanding why so many acquisitions of start-ups occur, is looking into the innovative activity of the high-tech sector. Acs and Audretsch (1988) use the number of

(8)

5

patents as an indicator of innovative activity. Their findings imply that the more an industry is composed of larger firms, the higher the level of innovative activity will be, but this increase in innovative activity comes from the smaller firms. Applying these findings on the high-technology sector implies that start-ups are the true innovators of the industry and they are so, because this is the only way to survive in the high-tech industry.

2.2.1 Reasons for acquiring start-ups

It is now established that the innovation in the high-tech industry comes from start-ups, but this does not explain why most start-ups are being acquired. The reason for this high number of acquisitions has to do with the fact that a firm can acquire knowledge in two ways: they acquire the knowledge their selves by investing in the firm, or they acquire outside knowledge by acquiring another firm (Ahuja & Katila, 2001). It is of importance that high-technology firms increase their knowledge base, because a higher knowledge base increases the

innovation output of the firm (Hall, Griliches & Hausman, 1986). A higher innovation output implies that more ideas are able to reach the market; therefore, high-technology firms will be more competitive. Hence acquiring the innovating start-ups makes sense, because it is the easiest way to increase the knowledge base and therefore the competitiveness.

Cloodt, Hagedoorn and Van Kranenburg (2006) analysed mergers and acquisitions in four high-tech sectors: aerospace and defence, computers and office machinery,

pharmaceuticals, and electronics and communications. They confirmed that an important rationale for an acquisition was increasing the knowledge base, because these industries are primarily knowledge driven. However, they stress that not all acquisitions are with the

intention to learn. Other rationales may be: entering new markets and market-structure related decisions, or the desire to move the firm’s businesses overseas (Cloodt, Hagedoorn & Van Kranenburg, 2006). Thus not all acquisitions in the high-tech sector increase the innovation output and thereby the competitiveness of the acquirer.

Cloodt, Hagedoorn and Van Kranenburg (2006) also explain why acquisitions of start-ups are common in the high-tech industry. This has to do with the fact that acquiring

companies also have to integrate the acquired knowledge base in order to improve innovation output. The integration of a new knowledge base can disrupt existing innovation activities; therefore, acquiring firms prefer to acquire start-ups because of the relative small knowledge base (Cloodt, Hagedoorn & Van Kranenburg, 2006). This entails that the integration goes fast and no existing practices will be disrupted.

(9)

6

Finally, Cloodt, Hagedoorn and Van Kranenburg provide an explanation for the high number of acquisitions in the high-tech sector. This is because the rate at which the value of knowledge decreases is steep in the high-tech sector; therefore, established technology firms need to keep acquiring start-ups in order to compensate for the depreciating knowledge of earlier acquisitions.

There is more behind the aggressive acquisition programmes of tech giants than the explanation of compensating knowledge depreciation that Cloodt, Hagedoorn and Van Kranenburg give. Carew and Mandel (2011) state that the acquisition of a start-up is very risky, but the benefits can potentially be very large. The acquiring firm does not know what acquisition will give the largest economic benefit; therefore, established high-tech firms make a long-term commitment to many acquisitions. They know that most acquisitions will not create value, but the ones that do make up for it (Carew & Mandel, 2011).

The above-mentioned arguments all explain why high-tech firms want to acquire start-ups, but why not solely rely on the internal processes to increase the innovative output? The answer comes from Prabhu, Chandy and Ellis (2005) who argue that industry-defining ideas can arise outside the firm. A strategy that focusses solely on internally acquired knowledge is likely to inhibit access to knowledge that competitors already use. For example, Digital Equipment Corporation was a firm who solely relied on its own internal processes and is now acquired by a rival (Prabhu, Chandy & Ellis, 2005).

2.2.2 Theory that does not support acquiring start-ups

The previous section described the reasons that managers of high-tech firms can use to explain their acquisition programmes. However, there is also literature that does not support the aggressive acquisition programmes of start-ups.

One of the main arguments for acquiring a start-up is that it increases the innovation output. However, Ernst and Vitt (2000) find that acquisitions hurt innovation. The arguments they give are that many acquisitions distract managers from their own internal processes and key employees of the start-up may leave the firm after the acquisition. Another interesting argument comes from Hitt, Hoskisson, Ireland and Harrison (1991) who argue that the

acquisitions are financed with a high amount of debt and the associated interest payments and repayments supress needed funds for innovation.

Furthermore, Carew and Mandel (2011) acknowledge that high-technology firms may also just be increasing their market power and this could harm competition. However, Carew

(10)

7

and Mandel are not convinced that this is the reason why high-tech firms acquire start-ups, because they need to keep innovating. Otherwise, they will lose substantial market share to competitors who keep innovating.

2.3 Ways of measuring M&A performance

It is now established that there are reasons that confirm value creation of acquiring start-ups and there are also researchers who argue that acquisitions of start-ups destroy value. This controversy helps explaining why measuring M&A performance is an important topic in the finance literature.

It is common practice to use a specific benchmark for measuring performance of the acquisition. The benchmark is the return that investors could have earned on another

investment with similar risk (Bruner, 2002). Against this benchmark, Bruner (2002) defines three possibilities. The value is conserved if the investment is the same as the benchmark. Secondly, value is created if the investment return exceeds the benchmark and, lastly, value is destroyed if the investment return is lower than the benchmark.

Bruner (2002) finds that there are four ways to measure M&A profitability. The event study is the most dominate way of measuring M&A profitability. It measures the abnormal returns to shareholders surrounding the M&A announcement. Secondly, accounting studies use financial reports to investigate if financial performance has changed. Furthermore, surveys of executives try to measure M&A profitability by asking managers if it is likely that the acquisition will create value. Finally, clinical studies make use of case studies that provide a detailed background of a deal in order to find new insights on M&A profitability.

Schoenberg (2006) finds the same methods as Bruner, but argues that there is no comparability between the M&A performance generated by the alternative methods. Hence, acquisition literature should consider multiple methods in order to have a holistic view of outcome. Furthermore, Schoenberg states that ex-ante measures are captured by CARs (event studies) in the sense that they capture the immediate effect on shareholder value. However, if the researcher is interested in the long-term performance of the acquisition, CARs should be complemented with other methods.

(11)

8

2.4 Literature Review

The literature review starts with a short summary of important results – in all the research disciplines that study M&A performance. Subsequently, the literature review will become more specific by reviewing the results about acquirer returns. Subsections 2.4.3 and 2.4.4 conclude with describing theories that justify the empirical evidence for positive and negative acquirer returns.

2.4.1 Multiple disciplines that study mergers and acquisitions

The complex phenomenon of mergers and acquisitions (M&As) has received wide-spread attention by academia. Cartwright and Schoenberg (2006) argue that most research draws the same conclusion: target-firm shareholders earn positive short-term returns, shareholders of the acquiring firm experience negative returns, and shareholders that hold both firms as a

portfolio earn – insignificant – overall wealth gains. Multiple disciplines have tried to explain this conclusion.

The disciplinary of finance tackles M&A research by studying if the event (an M&A) creates or destroys shareholder value (Cartwright & Schoenberg, 2006). Most studies agree that acquisitions create value for shareholders for the target firm, i.e. the short-term abnormal returns of the target firm are positive. On the other hand, there is doubt among academia if M&As create value for the acquiring firm. Agrawal and Jaffe (2002) give an overview of finance studies about M&A. They find that cumulative abnormal returns in the years

following acquisitions are – in the best case – not statistically different from zero. However, it has to be noted that a high level of variance surrounds the abnormal returns (Conn et al., 2001). Other disciplines try to explain this variance.

The research within the discipline of strategic management can be divided in ‘strategic fit’ literature and ‘process’ literature (Cartwright & Schoenberg, 2006). ‘Strategic fit’

literature focusses on the link between acquisition performance and the extent to which the business of the acquirer is the same as the target’s. While this line of research explains some of the variation, the M&A underperformance cannot solely be explained by ‘strategic fit’. On the other hand, ‘process’ literature emphasizes that wrong decision-making, negotiation and integration processes explain part of the M&A underperformance (Cartwright & Schoenberg, 2006).

(12)

9

A new field that tries to explain M&A performance finds its origins in psychology. This field pays attention to the cultural dynamics of M&As (Cartwright & Schoenberg, 2006). The relationship between culture and performance seems to account for some M&A

underperformance. An important result of Van Dick, Ullrich and Tissington (2006) is that communication can be used to increase work effort. For example, communication should be used to make employees aware of the positive consequences of the merger and to create a sense of belonging for all the employees.

The last field that studies M&As are the longitudinal studies. Unfortunately, these type of studies are hard to find. One of the few longitudinal studies is from Kavanagh and

Ashkanasy (2006). They analysed three Australian university mergers over a seven-year period and concluded that leadership and a gradual pace of change are important to make an M&A successful.

2.4.2 A closer look at acquirer returns of acquisitions

In estimating acquirer returns to start-ups acquisitions, my study builds on the work of finance studies that tries to explain M&A performance. The literature that studies returns of the

acquirer is divided.

On the one hand, there are several studies that find negative returns for the acquiring firm. For example, Mitchell and Stafford (2000) use the Fama and French 3-Factor model to construct a benchmark return and find that abnormal returns are negative. On the other hand, Kohers and Kohers (2002) find that value is created when they look at a sample of 1634 firms in the period 1987-1996.

However, there seems to be a discrepancy between acquiring a private firm and public firm. Benson and Ziedonis (2005) show that the evidence is growing that shareholders of acquiring firms earn abnormal returns from acquiring private companies. For example, Chang (1998) looks at a sample of 281 private firms acquisitions and concludes that on average a 2.6% abnormal return is earned when stock is used to finance the takeover. Furthermore, Fuller, Nettor, and Stegemoller (2002) find significant positive abnormal returns in

acquisitions of private firms and negative returns when the acquirers acquire public firms. Andrade, Mitchell and Stafford (2001) give an overview of the acquisition literature and conclude that it is still a wide open field whether or not acquisitions create or destroy value for the acquirer. This study hopes to add to this field by analysing a segment within the private firms; namely, private start-ups in the high-tech industry.

(13)

10

2.4.3 Theory about positive acquirer returns

Empirical evidence of positive bidder returns, if the targets are private, is provided in the previous section. It is equally as important to look at literature that explains this phenomenon. Chang (1998) analysed bidder returns at the announcement of a takeover proposal when the target firm is privately held. He reports that bidders experience a positive abnormal return when the acquisition is financed with stock offers. On the other hand, Chang reports that bidders gain no abnormal returns when the deal is financed with cash. He provides the following reasons for positive bidder returns when acquiring private firms.

First of all, Chang (1998) argues that if the takeover market is competitive, the acquisition will be a zero net present value project. However, Chang states that the takeover market of private firms can be less competitive; therefore, bidding firms are likely to

experience abnormal returns, because the probability of underpaying is higher.

Secondly, there is a monitoring argument. That is, if the acquisition is financed with a stock exchange, it is likely that block holders will be created. The assumption made here is that private firms are owned by a small group of shareholders. These block holders serve as effective monitors of managerial performance; therefore, financing a private firm acquisition with a stock exchange creates value.

Finally, Chang uses the argument that the willingness of a party to hold a large block of shares results in a positive stock price reaction. This means that acquirers, who finance the acquisition by offering stocks to a small number of target shareholders, experience abnormal returns. The assumption is that the target shareholders did their due diligence and concluded that it is profitable to hold the stock.

2.4.4 Theory that contradicts evidence of positive acquirer returns

On the contrary, several reasons may explain why acquiring firms end up with negative returns when acquiring private firms.

Barney (1988) argues that managers bid the target’s price up to an amount equal or beyond its value, because the synergies that can be acquired are rarely unique to the acquiring firm. This means that more firms will bid up the price, because they all can extract the same synergies. Therefore, the acquirer will pay a premium that is equal to or more than the value of the synergies. Hence the acquiring firm will not earn abnormal returns (if the acquiring overpaid, that is the premium is more than the value of the synergies).

(14)

11

The following two arguments are important for acquisitions in the high-technology sector. In 2.2.1 it was shown that the ability to innovate is an important determinant of competitiveness for high-tech firms. Franko (1989) reports the significant positive relationship between R&D expenditures and innovation output. Hitt et al. (1991) use this argument by stating that an acquisition carries opportunity costs with respect to R&D: if funds are used to finance an acquisition, they cannot be used to finance R&D. These costs can be so high that the firm is better off investing in R&D and not acquiring another firm.

Another argument that Hitt et al. (1991) give is that acquisitions intervene in the internal R&D processes. This means that managers are less stimulated to create ideas on their own, because there is always the possibility to acquire external technology. Therefore,

acquisitions in the high-tech sector harm shareholder value, because the R&D output goes down (Hitt et al., 1991).

This chapter provided a theoretical framework that forms the basis of my study. The first section about start-ups helps with forming the hypothesises. Furthermore, the reasons for acquiring start-ups in the high-tech sector will place the results of my thesis in the right context. Also, the literature review was necessary to conclude if the results that I find are contradicting or supporting recent literature. Finally, sections 2.4.3 and 2.4.4 will prove helpful when quantifying the right hypothesis.

The next chapter describes the research method that will be used to test if acquirer returns are positive when acquiring start-ups in the high-tech sector.

3 Research Method & Data

This thesis will use the method of event studies to analyse if high-technology firms create value with buying private start-ups. There are many variations in the applications of event studies. Luckily, Peterson (1989) and Mackinlay (1997) describe methodologies that can be used to analyse whether an M&A announcement creates abnormal shareholder value. Every event study starts with a timeline that includes the M&A announcement. The next step is to come up with a random variable that can be used for testing. Finally, the null hypothesis will be rejected, or not, on basis of the t-test that uses the constructed random variable and its standard deviation.

The next section is devoted to the first step of an event study: setting up a timeline. Next, I will summarize how the random variable is constructed that is used to test the

(15)

12

hypothesis. Subsequently, the hypothesises will be stated in 3.3. Section 3.4 provides a thorough description of the collected data. Finally, the chapter concludes with limitations of the event study that I used.

3.1 Timeline

The first step in the event study is to set up a timeline for the event study. This timeline can be represented as follows:

Figure 1 Time-line event study

where

tb = The first day of the estimation period;

tpre = The last day of the estimation period and first day of the event period

te = The day of the event

tpost = The last day of the event period

The first interval on the timeline is the estimation period. The estimation period is used to calculate the benchmark return. When choosing the length of the estimation period, you have to balance the benefits of a longer period (statistical power) versus the costs (parameter bias). A typical length of the estimation period for M&A announcements is 120 days (Mackinlay, 1997).

The second interval is the event period. In the case of an acquisition, Peterson advices to analyse the value creation over an interval of time. This is because uncertainty may remain at the time of announcement; for example, most of the time deals still need shareholder approval. Also, some information may already leak before the M&A announcement. There is no standard width for the event period; therefore, I decided to analyse multiple event periods in this thesis. The event period is defined as follows: (tpre;tpost) where tpre is the number of days

(16)

13

periods: only the announcement day (0;0), the day before and the announcement day itself (-1;0), (-1;1), (-2;2), (-5;5), (-7;7), (-10;10). This means that I will analyse 7 different event periods for each M&A announcement.

3.2 Construct a testable random variable

The first step in constructing a testable random variable is calculating the abnormal returns for the event period. Peterson (1989) defines the abnormal returns as follows:

𝐴𝑅𝑖,𝑡 = 𝑅𝑖,𝑡− 𝑅𝑖,𝑡∗ , 𝑡 = 𝑡𝑝𝑟𝑒, … , 𝑡𝑝𝑜𝑠𝑡, 𝑖 = 1, … , 𝑛 (1)

where

ARi,t = Abnormal return stock for stock i period t;

Ri,t = Return on stock i in period t;

Ri,t* = Expected return on stock i in period t.

I use the market model, suggested by Peterson (1989), to calculate Ri,t*:

R∗_i,t = 𝛼̂ + 𝛽̂ ∗ 𝑅_𝑚,𝑡 , 𝑡 = 𝑡_𝑝𝑟𝑒, … , 𝑡_{𝑝𝑜𝑠𝑡}, 𝑖 = 1, … , 𝑛 (2)

where 𝑅_𝑚,𝑡 is the return of the S&P 500 or NASDAQ (robustness check) at time t and the parameters are calculated from the time-series regression in the estimation period:

𝑅_𝑖,𝑡 = 𝛼_𝑖+ 𝛽_𝑖∗ 𝑅_𝑚,𝑡 + 𝜖_𝑖,𝑡 , 𝑡 = 𝑡_𝑏, … , 𝑡_𝑝𝑟𝑒, 𝑖 = 1, … , 𝑛 (3)

Furthermore, 𝑅_𝑖,𝑡 is calculated as follows:

𝑅_𝑖,𝑡 = 𝑃𝑖,𝑡− 𝑃𝑖,𝑡−1

𝑃_{𝑖,𝑡−1} (4)

where P is the adjusted stock price, i is the stock index and t refers to the time (day).

(17)

14

𝐶𝐴𝑅𝑖 = ∑ 𝐴𝑅𝑖,𝑡

𝑡=𝑡𝑝𝑜𝑠𝑡

𝑡=𝑡𝑝𝑟𝑒

, 𝑖 = 1, … , 𝑛 (5)

CARi is the sum of the abnormal returns over the entire event window. The cumulative

average abnormal return (CAAR) can now be constructed by taking the average of all the acquisitions their CARs:

𝐶𝐴𝐴𝑅 =1 𝑛∑ 𝐶𝐴𝑅𝑖 , 𝑖 = 1, … , 𝑛 𝑛 𝑖=1 (6) 3.3 Hypothesises

The method set out in the previous sections will be used to test the hypothesises in this section. I will start with a short explanation for the two-tailed hypothesis that I use.

The conclusion from 2.4.2 is that positive acquirer returns for public firm acquisitions are questionable; however, there is considerate empirical evidence that acquiring private firms creates value for the acquirer.

Start-ups form only a subset of private firms. Therefore, it would be too much of a generalization to apply the results of acquisition literature about private firms to start-ups. Nevertheless, section 2.4.3 provides scientific justification for positive acquirer returns when acquiring private start-ups. On the other hand, 2.4.4 gives several reasons why acquisitions of start-ups might not create value. Thus there is empirical and theoretical proof that acquiring start-ups creates values, but the contradicting theory in 2.4.4 withholds me from using an alternative hypothesis that assumes that acquisition of start-ups create value. Therefore, the first hypothesis is set up as follows:

H0: Announcement of acquiring a private start-up conserves value for the acquiring

high-technology firm

H1: Announcement of acquiring a private start-up creates or destroys value for the

high-technology firm (so this will result in two-sided tests)

If H0 is rejected, then there is statistical evidence that acquiring private start-ups creates or

(18)

15

operationalised. CAARs follow a distribution with a population mean and variance. I define the population mean for CAAR as CAAR* and this gives the following hypothesis:

H0: CAAR* = 0

H1: CAAR* ≠ 0 (not right-tailed, because of the above-mentioned part about theory

contradicting and supporting the value creation of acquiring start-ups)

This hypothesis can be tested with a t-statistic due to the central limit theorem:

𝑡 = 𝐶𝐴𝐴𝑅 − 0_𝑠 √𝑛

(7)

where s is the sample standard deviation and n is the sample size. The sample standard deviation is calculated as:

𝑠 = √ 1 𝑛 − 1∑(𝐶𝐴𝑅𝑖 − 𝐶𝐴𝐴𝑅) 2 𝑛 𝑖=1 (8)

The t-statistic and accompanying p-value can now be constructed to conclude if there is enough statistical evidence to reject the null hypothesis.

After testing the first hypothesis, I will analyse the sample on a deeper level. The theory in 2.1 seems to favour an acquisition of a start-up backed with venture capital over a start-up without venture capital. This is because venture capital reduces uncertainty and increases the professionalization of a start-up (see 2.1 for a more detailed explanation). Therefore, acquiring a venture-backed start-up is less risky and this should be preferred by risk-averse shareholders.

To find empirical evidence for this theory, I will divide the sample in acquisitions of start-ups backed with venture capital and acquisitions of start-ups that do not have venture capital.

First, I will do two more event studies for each subset. For each subset, I will test the same hypothesis as for the entire sample (stated on the previous page). This gives two more hypothesises. Hypothesis two is set up as follows:

(19)

16

H0: CAAR*venture-backed = 0 (the backed subscript stands for the population of

venture-backed start-ups)

H1: CAAR*venture-backed ≠ 0

Furthermore, hypothesis three:

H0: CAAR*without venture capital = 0 (the without venture capital subscript stands for the

population of start-ups without venture capital) H1: CAAR*without venture capital ≠ 0

Finally, I will use a two-sample t-test to determine if the cumulative abnormal returns for the acquisitions of venture-backed start-ups are higher than for the acquisitions of start-ups without venture capital. This gives the fourth, and final, hypothesis:

H0: CAAR*venture-backed = CAAR*without venture capital

H1: CAAR*venture-backed > CAAR*without venture capital

3.4 Data

I used the Thomson One database to find the sample of acquisitions that I will analyse. The “Mergers & Acquisitions” database of Thomson One allows one to filter for a number of categories. The first filter I used is that the public status of the acquirer is public – otherwise there is no stock data to analyse – and the public status of the target is private. Furthermore, both companies have to be in the United States and in the high-technology industry. Finally, I want to analyse recent data; therefore, I chose for the following announcement period:

01/01/2014 – 5/30/2016. After accounting for the above-mentioned filters, the database gives 414 hits. Furthermore, I imported additional stock prices to construct the estimation period for acquisitions in the beginning of 2014.

Once I imported the 414 hits, I scanned the deals one by one and made the following adjustments. I deleted the firms that are Over-The-Counter (OTC)-listed, because these stocks are less transparent and trustworthy than exchange stock prices. Furthermore, I deleted firms who went bankrupt or were acquired shortly after the M&A announcement. I also deleted firms that are older than 10 years at the acquisition date, because this study analyses start-ups.

(20)

17

A couple of M&A announcement were just rumours and did not go through – hence they were deleted from the dataset. Furthermore, I deleted all other deal forms than acquisitions, such as mergers, because I am analysing acquisitions. Finally, firms were deleted from the sample because of a lack of data; they went public just before the M&A announcement. The sample exists of 225 M&A announcements after correcting for all the above-mentioned firms

Finally, I used Datastream to come up with adjusted stock prices for the 225 firms and adjusted levels for the S&P 500 and NASDAQ Composite. I will analyse the adjusted prices, because this accounts for stock dividends and splits.

3.5 Methodology limitations

There are several possible limitations of the conducted research that need to be taken into account. The first limitation is specific to the high-technology industry. The second limitation is more general to the event study methodology.

First of all, due to the aggressive acquisition programmes it is possible that there is overlap between event periods. For example, Google acquired Softcard on 29/04/2015 and Red Hot Labs on 24/02/2015. This means that the acquisition of Red Hot Labs is in the estimation period of the acquisition of Softcard. Therefore, the parameters of the market model are biased due to the abnormal returns surrounding the Red Hot Labs event. The ideal solution would be to delete the event period of Red Hot Labs from the estimation period. However, no other researchers have done this and I expect that the results of the deleted and undeleted sample will be comparable (only a small part of the estimation period is affected by the overlap). Nevertheless, it is important to be aware of this limitation, because of the bias it brings into the market model and therefore into the calculation of abnormal returns.

The second limitation has to do with the properties of daily stock returns. Brown and Warner (1985) studied potential problems that may arise from using daily stock prices in event studies. Two important problems they investigated are non-normality of daily returns and issues of autocorrelation in daily returns. These problems can potentially lower the statistical power of parametric tests (Brown & Warner, 1985). They conclude that methodologies based on the OLS market model and using standard parametric tests are well specified. Hence, following Brown and Warner, this study does not have limitations caused by the

characteristics of daily stock returns.

However, the previous analysis of Brown and Warner (1985) only holds up when there is a sufficient amount of data. This thesis only analyses 2,5 years and therefore the number of

(21)

18

acquisitions is limited. An analysis of a longer time period, e.g. starting at the end of the financial crisis, gives a smaller bias and justifies using the central limit theorem for t-tests.

This chapter described the research method and the accompanying hypothesis. Furthermore, the data collection has been addressed and possible limitations of the research method are discussed. Next chapter presents the results and analyses of my inquiry.

4 Results & Analyses

This chapter is divided in four sections , each of the sections analyses a hypothesis mentioned in the previous chapter. The chapter begins with the event study of the entire sample. I also perform a robustness check for each event study. This means that I run another event study where the NASDAQ Composite represents the market proxy (instead of the S&P 500).

4.1 Event study of all start-ups

The first hypothesis concerns all start-ups; therefore, the entire sample set is equal to 225 observations (acquisitions). The first hypothesis can be written as:

H0: CAAR* = 0

H1: CAAR* ≠ 0 (not right-tailed, because of the above-mentioned part about theory

contradicting and supporting the value creation of acquiring start-ups)

The null-hypothesis is rejected for all event periods. This means that there is no statistical evidence that acquiring start-ups create abnormal shareholder value. On the same token, there is also no statistical evidence that acquisitions of start-ups destroy shareholder value.

This result is supported by researchers of earlier studies (Agrawal & Jaffe, 2001; Jensen, 1988). On the other hand, this result contradicts the work of (Chang, 1998; Fuller, Nettor & Stegemoller, 2002).

Table 1 shows the results of the event study. The p-value of the announcement day is almost significant at the 10%-level. It is in line with the efficient market hypothesis that the abnormal returns of the acquisition should be realised on the same day as the information is

(22)

19

available to the financial markets. Therefore, it makes sense that the p-value on the announcement day is the lowest of all event windows.

Finally, the cumulative abnormal returns of the most event periods are positive. This is hopeful for research that suggests that acquisitions of private firms create value. Nevertheless, the positive CAARs can be due to outliers, i.e. positive returns due to market perception of an acquisition that will be very successful.

Table 1: First Hypothesis Standard t-test

Event Window _CAAR _s _t _p _Significance

(0,0) 0.004 0.038 1.524 0.129 No (-1,0) 0.002 0.043 0.607 0.545 No (-1,1) 0.000 0.056 0.038 0.970 No (-2,2) -0.001 0.067 -0.171 0.864 No (-5,5) -0.003 0.080 -0.633 0.527 No (-7,7) 0.000 0.091 0.007 0.994 No (-10,10) 0.000 0.104 -0.015 0.988 No

*at 10%-level significance, ** at 5%-level significance

I repeated the same analysis with the NASDAQ Composite as a robustness check. The results are displayed in table 2 and the same analysis applies as for table 1. Therefore, the event study seems robust to different market proxies.

Table 2: First Hypothesis Robustness Check Standard t-test

Event Window CAAR s t-statistic p-value Significance

(0,0) 0.004 0.038 1.554 0.122 No (-1,0) 0.002 0.043 0.634 0.527 No (-1,1) 0.000 0.056 0.077 0.938 No (-2,2) 0.000 0.067 -0.031 0.975 No (-5,5) -0.002 0.080 -0.412 0.681 No (-7,7) 0.001 0.090 0.220 0.826 No (-10,10) 0.001 0.102 0.162 0.871 No

4.2 Event study of only venture-backed start-ups

The second hypothesis concerns only venture-backed start-ups; 145 acquisitions in the sample are acquisitions of venture-backed start-ups. The second hypothesis was stated as follows:

(23)

20

H0: CAAR*venture-backed = 0 (the backed subscript stands for the population of

venture-backed start-ups)

H1: CAAR*venture-backed ≠ 0

The null-hypothesis of the second hypothesis is rejected for all event periods, except for the (0,0) event period. This means that there is statistical evidence that acquisitions of start-ups create or destroy value. In this particular case, it is very likely that the acquisition creates value, because the CAAR is 0,7% with a p-value of 0,038%. Therefore the null hypothesis is rejected with 96,2% certainty.

This result is in line with earlier studies that find that acquisitions of private firms create value for the acquirer (Chang, 1998; Fuller, Nettor & Stegemoller, 2002). It adds to this literature by studying the acquisitions of private start-ups. It seems to be an important feature if start-ups are backed with venture capital. On the other hand, this study contradicts earlier studies that find that acquisitions destroy value for the acquirer (Mitchell & Stafford, 2000). This is not concerning, because this literature finds these results when the target firms are public. Therefore, the results are in line with the two split in the acquisition literature; acquisitions of public firm seem to destroy value for the acquirer and acquisitions of private firms are likely to create value for the acquirer.

Table 3 shows the results of the event study. The significance of the (0,0) event window has already been noted. Interestingly, the p-value of the (-1,0) event window has dropped with a great amount in comparison to tables 1 and 2. This is in line with Chang (1998) who analysed the acquirer return of private firms for the (-1,0) event window and showed statistical evidence that these acquisitions create value. Unfortunately, the evidence is not significant here, but the drop in p-value supports the view of Chang (1998) who argues that acquisitions of private firms create value for the acquirer.

Table 3: Second Hypothesis Standard t-test

(0,0) 0.007 0.042 2.097 0.038 Yes** (-1,0) 0.005 0.044 1.464 0.145 No (-1,1) 0.003 0.060 0.534 0.594 No (-2,2) -0.001 0.069 -0.121 0.904 No (-5,5) 0.000 0.085 -0.015 0.988 No (-7,7) -0.001 0.095 -0.081 0.936 No (-10,10) 0.001 0.106 0.152 0.880 No

(24)

21

Table 4 contains the results of the robustness check with the NASDAQ Composite (the robustness check has the same estimation period, but a different market proxy). The

differences in p-values are not that different, but it is remarkable that all the CAARs are now positive. This is in line with the view that acquisitions of venture-backed start-ups create value. Nevertheless, the p-values do not provide statistical evidence: the p-values – except for (0,0) – are all above 0.05. Therefore, not much value should be attached to all the positive CAARs. Finally, the event study seems robust to changes in the market proxy (the same main results are reported).

Table 4: Second Hypothesis Robustness Check Standard t-test

(0,0) 0.007 0.042 2.085 0.039 Yes** (-1,0) 0.005 0.044 1.467 0.145 No (-1,1) 0.003 0.061 0.573 0.568 No (-2,2) 0.000 0.069 0.033 0.974 No (-5,5) 0.001 0.086 0.146 0.884 No (-7,7) 0.001 0.095 0.084 0.933 No (-10,10) 0.003 0.105 0.288 0.774 No

4.3 Event study of only start-ups without venture capital

The final event study focusses on start-ups without venture capital – there are 80 acquisitions of start-ups without venture capital in the sample. Hypothesis three was stated as follows:

H0: CAAR*without venture capital = 0 (the without venture capital subscript stands for the

population of start-ups without venture capital) H1: CAAR*without venture capital ≠ 0

The third hypothesis is not rejected for all the event windows. This means that there is no statistical evidence that acquiring private start-ups without venture capital creates or destroys value. This result is the same as the result in 4.1 and therefore the same analysis applies. However, there are some dissimilarities between the entire sample of acquisitions (4.1) and the sample of start-ups without venture capital (this section). This means that there seem to be

(25)

22

differences – for the acquirer – between acquiring start-ups in general and start-ups without venture capital.

The first difference is that the CAARs for table 5 are all more negative than the CAARs in table 1, except for the (-7,7) event window. This indicates that the acquisition of start-ups without venture capital might destroy value. Another indicator of this, is that the p-values have dropped in comparison to table 1. This means that it became more likely that the null-hypothesis might not be right; yet again, there is no statistical evidence for this at any respectful significance level. The differences in tables 1 and 5 can be explained by the fact that the venture-backed start-ups are not included in the sample of 4.3. This means that a lot of positive CARs are deleted from the sample, because the previous section concluded that it is very likely that acquiring a venture-backed start-up creates value.

Summarizing, there seems to be a difference in the distribution of CAARs of venture-backed start-up acquisitions and start-up without venture capital acquisitions. Section 4.4 looks into this by constructing a two-sample t-test.

Table 5: Third Hypothesis Standard t-test

(0;0) -0,002 0,029 -0,747 0,457 No (-1;0) -0,005 0,042 -0,992 0,324 No (-1;1) -0,004 0,047 -0,842 0,402 No (-2;2) -0,001 0,064 -0,126 0,900 No (-5;5) -0,009 0,070 -1,192 0,237 No (-7;7) 0,001 0,084 0,137 0,891 No (-10;10) -0,003 0,101 -0,241 0,810 No

Table 6 presents the results of the robustness check with the NASDAQ Composite as the market proxy. There are no remarkable differences between table 5 and 6. This is in favour of the robustness of the event study.

Table 6: Third Hypothesis Robustness Check Standard t-test

(0;0) -0,002 0,029 -0,657 0,513 No (-1;0) -0,004 0,042 -0,956 0,342 No (-1;1) -0,004 0,047 -0,840 0,404 No (-2;2) -0,001 0,063 -0,103 0,918 No (-5;5) -0,008 0,068 -1,054 0,295 No (-7;7) 0,003 0,081 0,276 0,783 No (-10;10) -0,001 0,096 -0,137 0,892 No

(26)

23

4.4 Comparison of CAARs

The previous section already indicates that there might be a difference in the CAARs distribution of venture-backed start-up acquisitions and CAARs distribution of start-up

without venture capital acquisitions. An indicator for this statement is that the CAARs of table 5 are almost all negative and the CAARs of table 3 are mainly positive. This means that the CAARs of acquiring venture-backed start-ups may be higher than the CAARs of acquiring start-ups without venture capital.

I constructed a two-sample t-test to test hypothesis 4 only for the announcement day event window. Thus the difference with the previous sections is that only one event window is analysed in this section – the event window with significant results. Since there is no evidence suggesting an equal or unequal variance of the CAARs distributions, I constructed two tests. As a quick reminder, hypothesis 4 was stated as follows:

H0: CAAR*venture-backed = CAAR*without venture capital

H1: CAAR*venture-backed > CAAR*without venture capital

The first test used the assumption of unequal variances and the results are reported in table 7. There are 213 degrees of freedom and the null-hypothesis for both event studies (the one that uses the S&P 500 as market proxy and the one that uses the NASDAQ Composite as market proxy) is rejected. There is a high level of statistical evidence: the null-hypothesis is rejected for a 2.5% significance level. This means that CAARs distribution of the venture-backed start-ups has a higher population mean than the CAARs distribution of start-ups without venture capital (with more than 97,5% certainty). This result can be placed in the light of the results in previous sections. The higher population mean for CAARs of venture-backed start-up acquisitions helps to explain why there is statistical evidence that acquiring venture-backed start-ups creates value and acquiring start-ups without venture capital does not.

Table 7: Fourth Hypothesis two-sample t-test with unequal variances

Type of Event Study v t-statistic p-value Significance

S&P 500 213 2,0493 0,0208 Yes***

NASDAQ Composite 213 1,9790 0,0246 Yes***

(27)

24

The second two-sample t-test used the assumption of equal variances and rejects the null-hypothesis as well. The statistical evidence is less certain than the previous t-test, but there is still more than 95% certainty that the CAARs distribution of venture-backed start-ups has a higher population mean than the CAARs distribution of the start-ups without venture capital.

Table 8: Fourth Hypothesis two-sample t-test with equal variances

Type of Event Study v t-statistic p-value Significance

S&P 500 223 1,8432 0,0333 Yes**

NASDAQ Composite 223 1,7806 0,0382 Yes**

*at 10%-level significance, ** at 5%-level significance, *** at 2,5%-level significance

This chapter presented all the results and concluded if the four null-hypotheses in chapter 3 can be rejected. The next chapter answers the research question and summarizes all the important results.

5 Conclusion

This study focussed on the high-technology sector – a sector that is famous for its high amount of start-up acquisitions. I tried to explain these acquisitions by asking a simple

question: “Does the announcement of acquiring a private start-up create shareholder value for a high-technology firm in the United States?”. This is an important question for shareholders of maturing strategic buyers, such as Microsoft and Cisco, because these companies use a lot of funds to buy start-ups. If these acquisitions do not create value, then it is perhaps time to use the funds for other investments, such as R&D, that might create more value.

This thesis builds on the work of finance that tries to explain M&A performance. The acquisition literature is still a wide open field and it is not decided if acquisitions create or destroy value for the acquirer. Benson and Ziedonis (2005) showed that the evidence is growing that acquiring firms earn positive abnormal returns from acquiring private start-ups. It would be too much of a generalization to apply the evidence of Benson and Ziedonis (2005) to private start-ups, since private start-ups are only a subset of private firms. Therefore, the answer to my research question tests the robustness of the results of Benson and Ziedonis by looking at a subset of private firms (start-ups).

I analyzed 225 start-up acquisitions in the United States from 01/01/2014-5/30/2016. The event study for the entire dataset has shown that there is no statistical evidence to

(28)

25

conclude that acquiring private start-ups creates value for the acquirer. However, this result is different when the sample is divided in start-ups that received venture capital and start-ups that did not. An important result is that the announcement of acquiring a venture-backed start-up creates value for the acquirer. However, announcing the acquisition of a start-start-up without venture capital – in the best case – does not create value for the acquirer. Finally, a two-sample t-test compared the population mean of two distributions: the distribution of cumulative average abnormal returns for venture-backed start-up acquisitions and of

cumulative average abnormal returns for starup acquisitions without venture capital. The t-test has shown that the first distribution has a higher population mean. This result helps to explain why the event study has shown that the announcement of acquiring a venture-back start-up creates value and the announcement of acquiring a start-up without venture capital does not.

A limitation of this study is that there is overlap between estimation periods and event periods, because a firm acquired multiple firms in the same period. This overlap results in a bias when calculating abnormal returns. Another limitation is addressed by Bruner who argues that the event study should be complemented with other methods, such as accounting studies, in order to have a holistic view of outcome. Furthermore, the sample size is low and therefore estimated parameters may have a bias. Finally, only one possible t-test is used. The conclusions drawn from the results in this thesis would be stronger if they were supported by a battery of t-tests, such as the MacKinlay t-test and the Brown Warner t-test.

For future studies, it could be interesting to see if the performed event study gives the same results for different countries. For example, the acquisition literature recently started to focus on the M&A activity in southeast Asia. It will be interesting to see if acquiring a venture-backed start-up in southeast Asia also creates value for the acquirer. Another suggestion is to keep the focus on the United States and analyze if the obtained results also hold up for a longer time period, e.g. the last decade. Finally, an event study can be used to analyze the buy and build strategy: acquiring firms within a specific field and subsequently acquiring firms that have a strategic fit with the acquired firms. It would be interesting to see if this strategy creates more value than acquiring firms that are not necessarily similar to already acquired firms.

(29)

26

References

A full week of match-making at Startup Fest Europe (2016, May 24). Retrieved from https://www.startupfesteurope.com/site/what-is-startupfest-europe/

Acs, Z. J., & Audretsch, D. B. (1988). Innovation in large and small firms: an empirical analysis. The American Economic Review, 78(4), 678-690.

Agarwal, A. and J. F. Jaffe (2000). ‘The post-merger performance puzzle’. In: C. Cooper and A. Gregory, Advances in Mergers and Acquisitions, 1, 7–14. Elsevier Science,

New York.

Ahuja, G., & Katila, R. (2001). Technological acquisitions and the innovation performance of acquiring firms: A longitudinal study. Strategic management journal, 22(3), 197-220.

Andrade, G., Mitchell, M., Stafford, E. (2001). New evidence and perspectives on mergers.

Journal of Economic Perspectives, 15(2), 103–120.

Audretsch, D. B. (2002). The dynamic role of small firms: Evidence from the US. Small

Business Economics, 18, 13-40.

Barney, J. 1988. Returns to bidding firms in mergers and acquisitions: Reconsidering the relatedness hypothesis. Strategic Management Journal, 9(special issue), 71. Benson, D. F., & Ziedonis, R. H. (2005, December). Corporate venture capital and the

returns to acquiring entrepreneurial firms. Paper presented at Harvard

Entrepreneurship Conference, Cambridge, Massachusetts.

Brown, S. J., & Warner, J. B. (1985). Using daily stock returns: The case of event studies. Journal of Financial Economics, 14(1), 3-31.

Bruner, R. F. (2002). Does M&A pay? A survey of evidence for the decision-maker. Journal

of Applied Finance, 12(1), 48-68.

Bruno, A. V., & Cooper, A. C. (1982). Patterns of development and acquisitions for Silicon

Valley startups. Technovation, 1(4), 275-290.

Cartwright, S., & Schoenberg, R. (2006). Thirty years of mergers and acquisitions research: Recent advances and future opportunities. British Journal of Management, 17(1), 1-5. Chang, S. (1998). Takeovers of privately held targets, methods of payment, and bidder returns. Journal of Finance, 53(2), 773-784.

Cloodt, M., Hagedoorn, J., & Van Kranenburg, H. (2006). Mergers and acquisitions: Their effect on the innovative performance of companies in high-tech industries. Research

(30)

27

Conn, C., & Michie, J. (2001). Long-run share performance of UK firms engaging in cross- border acquisitions. ESRC Centre for Business Research, University of Cambridge. Dick, R., Ullrich, J., & Tissington, P. A. (2006). Working under a black cloud: How to sustain

organizational identification after a merger. British Journal of Management, 17(1), 69-79.

Ding, M., & Eliashberg J. (2002). Structuring the new product development pipeline.

Management Science¸48(3), 343–363.

Ernst, H., & Vitt, J. (2000.) The influence of corporate acquisitions on the behaviour of key inventors. R&D Management, 30(2), 105-119.

Fama, E. F., & French, K. R. (2004). The capital asset pricing model: Theory and evidence. Journal of Economic Perspectives, 18(3), 25-46.

Franko, L. G. (1989). Global corporate competition: Who's winning, who's losing, and the R&D factor as one reason why. Strategic Management Journal, 10(5), 449-474 Fuller, K., Netter, J., Stegemoller, M. (2002). What do returns to acquiring firms tell us?

Evidence from firms that make many acquisitions. Journal of Finance, 57(4), 1763– 1793

Hall, B. H., Griliches, Z., & Hausman, J. A. (1986). Patents and R&D: Is there a lag?.

International Economic Review, 27(2), 265-283.

Hitt, M. A., Hoskisson, R. E., Ireland, R. D., & Harrison, J. S. (1991). Effects of acquisitions on R&D inputs and outputs. Academy of Management Journal, 34(3), 693-706.

Hellmann, T., & Puri, M. (2002). Venture capital and the professionalization of start‐up firms: Empirical evidence. The Journal of Finance, 57(1), 169-197.

Jensen, M. C. (1988). Takeovers: Their causes and consequences. The Journal of Economic

Perspectives, 2(1), 21-48.

Kavanagh, M. H., & Ashkanasy, N. M. (2006). The impact of leadership and change management strategy on organizational culture and individual acceptance of change during a merger.British Journal of Management,17(1), 81-103.

Kohers, N., & Kohers, T. (2000). The value creation potential of high-tech mergers. Financial

Analysts Journal, 56(3), 40-51.

Kollman, T., Stöckmann, C., Linstaedt, J., & Kensbock, J. (2015). European Startup Monitor.

Retrieved from: http://europeanstartupmonitor.com

Lynley, M. (2015, April 8). Apple Quietly Bought Dryft. Retrieved from

http://techcrunch.com/2015/04/08/apple-quietly-bought-dryft-a-keyboard-app/ MacKinlay, A. C. (1997). Event studies in economics and finance. Journal of economic

(31)

28

literature, 35(1), 13-39.

Mandel, M., & Carew, D. (2011). Innovation by Acquistion: New Dynamics of High-Tech Competition. Retrieved from

http://progressivepolicy.org/wp- content/uploads/2011/11/11.2011-Mandel_Carew-Innovation_by_Acquisition-New_Dynamics_of_Hightech_Competition.pdf

Mitchell, M. L., & Stafford, E. (2000). Managerial Decisions and Long‐Term Stock Price Performance. The Journal of Business, 73(3), 287-329.

Peterson, P. P. (1989). Event studies: A review of issues and methodology. Quarterly journal

of business and economics, 36-66.

Prabhu, J. C., Chandy, R. K., & Ellis, M. E. (2005). The impact of acquisitions on innovation: poison pill, placebo, or tonic?. Journal of Marketing, 69(1), 114-130.

Ransbotham, S., & Mitra, S. (2010). Target age and the acquisition of innovation in high- technology industries. Management Science, 56(11), 2076-2093.

Schoenberg, R. (2006). Measuring the Performance of Corporate Acquisitions: An Empirical Comparison of Alternative Metrics. British Journal of Management, 17(4), 361-370. Zider, B. (1998). How venture capital works. Harvard business review, 76(6), 131-139.

Value creation with acquisitions of start-ups : a study of acquisition programmes in the high-technology industry