essays on corporate strategy- evolution of corporate capabilities and the role of intangible assets

Therefore, it is natural to ask, `What type of assets are worth buying?' This paper investigates the long-run performance effects of acquiring intangible versus tangible targets.. Result

ESSAYS ON CORPORATE STRATEGY: EVOLUTION

OF CORPORATE CAPABILITIES AND

THE ROLE OF INTANGIBLE ASSETS

DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School

of The Ohio State University

By Asli Musaoglu Arikan, MBA

Professor Jay Barney, Adviser

Professor Karen Wruck

Professor Anita McGahan Adviser

Professor Konstantina Kiousis Business Administration Graduate Program Professor Oded Shenkar

Copyright by

Asli Musaoglu Arikan

2004

ABSTRACT

This dissertation is comprised of in depth analysis on the broader topic of

corporate strategy with emphasis of the role of intangible assets The first chapter looks at the performance implications of acquiring firms that have highly intangible assets

structures The second essay looks at dynamic characteristics as well as outcomes of developing intangible yet valuable corporate level capabilities in relation to managing alliances and acquisitions The final section looks at the role of intangible assets in

contracting between and within firms by utilizing property rights theory and the resource based view

A consistent finding regarding mergers and acquisitions (M&A) is that: on

average shareholders of target firms earn significant economic gains whereas

shareholders of acquiring firms break-even (Jensen and Ruback, 1983; Jarrell et al., 1988) Despite this general finding M&A activity has persisted, increasing in number and transaction value because, managers often perceive M&A activity as a mechanism for growth (e.g Penrose, 1959) Therefore, it is natural to ask, `What type of assets are worth buying?'

This paper investigates the long-run performance effects of acquiring intangible versus tangible targets Intangibility of target is proxied by multiple measures based on

R&D, advertisement and human capital stocks, and the Tobin's q 1-year prior to the corporate event Using a sample of M&A transactions spanning a 4 year period (1988-1991), long-run-buy-and-hold expected returns are calculated by constructing portfolios

of cohort firms that pursue M&A activity and tracked for 5 years Each firm's abnormal performance is calculated as the excess return to the benchmark portfolio Results show that on average, acquirers of intangible targets earn negative abnormal returns, whereas acquirers of tangible targets break-even However, for the whole sample, there is no evidence of long-run abnormal returns

long-run-Existence of asymmetric misvaluation between M&As of intangible versus

tangible targets is tested by regressing short-run returns on the buy-and-hold long-run returns Results provide evidence for market overreaction to the announcements that involve highly intangible targets Overall, findings suggest that on average, ownership claim to the target's intangible assets via M&A does not transfer the associated economic value

In the second section I investigate how long it takes for publicly traded firms within the United States to develop corporate capabilities for conducting alliances and acquisitions effectively The development of corporate capabilities has been difficult to study directly because little information has been available on the accumulation at the corporate level of performance-enhancing knowledge The research reported here relies

on a dataset that tracks the behavior of the 3,595 firms that went through an initial public

offering (IPO) between 1988 and 1999 to show how quickly corporate capabilities developed from the earliest years of firm formation

In particular, we conduct an event-study analysis to investigate how the abnormal returns to alliance and acquisition announcements changed as the firms accumulated experience in conducting deals of each type The results suggest that firms accumulated capabilities for executing and managing both alliances and acquisitions, and that

investors came to expect that firms would continue to exploit their specialized

capabilities into the future

Finally I provide discussion of the theoretical implications of the empirical findings and contribute to the literature on corporate strategy and resources based view

by incorporating insights from the property rights literature which can be considered as the recent development extension of transaction cost economics

Dedicated to my grandmother, Guzide Egilmez

ACKNOWLEDGMENTS

I wish to thank my adviser Jay Barney, and my committee members David

Hirshleifer, Konstantina Kiousis, Anita McGahan, Oded Shenkar, and Karen Wruck for their intellectual support, encouragement, and enthusiasm which made this thesis

possible

I am grateful to Ilgaz Arikan for his continued support and stimulating discussions

on all aspects of my research interests

I wish to also thank to Laurence Capron, Russ Coff, Ken Hatten, Anne-Marie Knott, Harbir Singh, Ralph Walkling, Julie Wulf, and Bernard Yeung for their helpful comments The author also benefited from discussions with Josh Lerner, Dan Levinthal, Jan Rivkin, Anju Seth, Jamal Shamsie, Scott Shane and Sid Winter All the errors remain mine

VITA

December 1, 1972…………Born – Izmir, Turkey

1994……….BS Istanbul technical University, Istanbul Turkey

1997 ………MBA University of North Carolina

2003-Current ……… Instructor, Boston University

PUBLICATIONS

Barney J.B., Arikan A.M 2002 The resource-based view Origins and implications In

Hitt M.A., Freeman R.E., Harrison J.S (eds.), Handbook of Strategic

Management Blackwell Publishers: Oxford, UK; 124-188

Arikan, A.M 2002 Does it pay to capture intangible assets through mergers and

acquisitions? Academy of Management Meetings Best Paper Proceedings

Arikan, A.M 2003 Does it pay to capture intangible assets through mergers and

acquisitions? In Strategic Management Society Book Series on M&A Summit

Arikan, A.M 2003 Cross-border mergers and acquisitions: What have we learned? In

B.J Punnett, and O Shenkar (Eds.) 2 nd Edition of Handbook of International

Management Research, University of Michigan Press

FIELDS OF STUDY

Major Field: Business Administration

Minor Field: Financial Economics

TABLE OF CONTENTS

Page

Abstract ii

Dedication vi

Acknowledgments vi

Vita……… vii

List of Tables x

List of Figures xii

Chapters: 1 Introduction 1

2 What Type Of Assets Is Worth Buying Through Mergers & Acquisition? 5

2.1 Resource-Based View And Competitive Advantage 8

2.2 Controls For Other Factors 11

2.2.1 Agency Motives 12

2.2.2 Information Asymmetry And Financing Of Intangible Assets 14

2.2.3 Market Over- Or Under-Valuation Of Growth Opportunities 17

2.3 Methodology And Data 20

2.3.1 Valuation Of Intangible Assets 21

2.3.2 Which Measure Of Performance? 24

2.3.3 Why Not Traditional Event Methodology? 26

2.3.4 Long Run Buy & Hold Abnormal Returns 29

2.3.5 Calculating Reference Portfolios 31

2.4 Data 32

2.5 Results 36

2.6 Discussion And Conclusion 44

3 How Long Does It Take To Build Corporate Capabilities For Conducting Alliances And Acquisitions? 50

3.1 Antecedents 51

3.2 Theory And Hypothesis 53

3.2.1 Industry And Time Effects 58

3.3 Data 59

3.4 Descriptive Statistics 61

3.5 Methods 64

3.6 Results 66

3.7 Conclusion 72

4 Why Do We Observe Heterogeneous Governance Choices For Similar Transactions? Theoretical Issues In Corporate Strategy 73

4.1 Overview 74

4.2 Theoretical Background 78

4.2.1 Transaction Cost Economics 78

4.2.2 Property Rights Theory 80

4.2.3 Capabilities View Of The Firm And Agency Costs 82

4.2.4 Hybrid Forms As Real Options 83

4.3 Model Setup And Intuition 84

4.4 A Real World Example 92

4.4.1 Who Should Own What? 94

4.4.2 Case 1: Agent B1j (Manufacturing Division Of Pfizer) Owns T Target B2 'S (Arqule's) Assets 95

4.4.3 Case 2: Target B2 (Arqule) Continues To Own Its Assets 96

4.5 Discussion 97

List Of References 99

LIST OF TABLES

1 Measures of Intangibility – Part 1 116

2 Measures of Intangibility –Part 2 117

3 Variable Descriptions 118

4 Descriptive Statistics 119

5 Correlation Matrix for Target Variables 120

6 Correlation Matrix 121

7 Test of Median Equality for the 60-Month Average Buy-and-Hold Abnormal Returns 122

8 Descriptive Statistics for the Average Monthly Buy-and-Hold Abnormal Returns [BHAR jt] 123

9 Sample Descriptio 130

10 Number of Alliance for each IPO year 131

11 Number of M&A for each IPO year 132

12 Mortality Rates of Firms and Deal Frequency 133

13 Average number of M&A deals per firm for each year following the IPO event 134

14 Descriptive Statistics for CARs per deal over -5,…,+5 days around the deal announcements 136

15 Descriptive Statistics for CARs per M&A deal over -5,…,+5 days around the deal announcements 137

16 Descriptive Statistics for CARs per Alliance deal over -5,…,+5 days around the deal announcements 138

17 Descriptive Statistics for CARs-5,…,+5 following the IPO year 139

18 Logit Analysis of Deal Type Choice (Alliance=1, M&A=0) and Past experience in the same-type deals 140

19 Logit Analysis of Deal Type Choice (M&A=1, Alliance =0) and Past experience in the same-type deals 141

20 Logit Analysis of Deal Type Choice and Market Reaction to the same-type deals 142

LIST OF FIGURES

1 Cumulative Abnormal Returns Around the Announcement Day, t=0 124

2 Average Monthly Abnormal Returns to the Acquirers with announcement year of 1988 125

3 Average Monthly Abnormal Returns to the Acquirers with announcement year of 1989 126

4 Average Monthly Abnormal Returns to the Acquirers with announcement year of 1990 127

5 Average Monthly Abnormal Returns to the Acquirers with announcement year of 1991 128

6 Average Monthly Abnormal Returns to the Pooled Acquirers with announcement years in 1988-1991 129

7 Model Payoffs 143

8 Corporate Capability as a System of Governance Mechanisms 144

However, firms intending to grow are more likely to create deviations from this ideal fit to accumulate intangible assets by, for example, following an overextension strategy First, firms that overextend know that they will not be able to do the new business effectively when they enter the new market; second, they know that they will eventually have to get into this new area; and third, those firms make sure that the intangible assets accumulated will be applicable beyond the segment that they were initially accumulated M&A activity serves as an investment mechanism to achieve

growth while possibly accumulating intangible assets1

A consistent finding regarding mergers and acquisitions (M&A) is that: on

average shareholders of target firms earn significant economic gains whereas

shareholders of acquiring firms break-even (Jensen and Ruback, 1983; Jarrell et al., 1988) Despite this general finding M&A activity has persisted, increasing in number and transaction value because, managers often perceive M&A activity as a mechanism for growth (e.g Penrose, 1959) It is natural ask, `What type of assets are worth buying?'

Lang and Stulz (1994) suggest that firms with valuable future growth

opportunities have highly intangible assets Intangible assets, such as managerial talent, corporate culture, R&D expertise, and brand capital have also been identified as sources

of competitive advantage (e.g Veblen, 1908; Grabowski and Mueller, 1978; Prahalad and Bettis, 1986; Barney, 1991) Thus, target firms with such assets appear very

attractive to buyers Can a buyer extract economic value associated with its target's intangible assets?

This chapter investigates the long-run performance effects of acquiring intangible targets versus tangible targets Using a sample of M&A transactions spanning a 4 year period (1988-1991), long-run-buy-and-hold expected returns are calculated by

constructing portfolios of cohort firms that pursue M&A activity in the 5-year post-event period Each firm's long-run-abnormal performance is calculated as the excess return to

the benchmark portfolio Intangibility of the target's assets is proxied by Tobin's q

1 Strategic alliances can be another external method to accumulate intangible assets and create growth opportunities However, the economic value associated with such growth opportunities is endogenous The firm's commitment to the alliance-related activities affects the value created In the case of M&As,

ownership and control rights of the buyer are more closely aligned

year prior to the corporate event This classification is robust to other measures of

intangibility, such as R&D and advertising stock, and human capital intensity Results show that on average, acquirers of intangible targets earn negative abnormal returns, whereas acquirers of tangible targets break-even The average long-run abnormal

performance of a buyer in the sample confirms the stylized fact that acquirers break-even

at best

The evidence on overconfidence is such that the individuals tend to be more confident in decision making situations where the feedback is delayed or inconclusive (Einhorn, 1980) The performance implications of M&As involving highly intangible targets are more likely to have delayed feedback or be inconclusive Also the expected returns to such corporate events are harder to forecast Thus, M&As of highly intangible targets are more likely to create situations where overconfidence can play a role in

forming expectations Moreover the behavioral model of Daniel, et al (1998) asserts that investors are more confident about their private signals and overreact to such

information In the same spirit with this model and above explanations, one would expect the M&A activity involving targets with intangible assets to trigger misvaluation due to overreaction

Existence of asymmetric misvaluation between M&As of intangible versus

tangible targets is tested by regressing short-run returns on the buy-and-hold long-run returns Results provide evidence for market overreaction to the announcements that involve highly intangible targets The market overreacts to the announcements regarding intangible targets and corrects its initial response over time On the other hand, there is no evidence of a misvaluation regarding the M&As of highly tangible targets However this

is not sufficient evidence to say that the overall market is inefficient Market efficiency suggest that there are classes of events that might be priced based on overreactions or underreactions to information, but on average these effects are cancel each other out (Fama, 1998) On the other hand, it is fair to say that the long-run underperformance of buyers of intangible targets imply that firms that develop an expertise to manage M&As

of such targets can create competitive advantage

The second chapter is organized as follows In the first section, the theoretical background and the hypotheses are presented In the second section methodology used and the data are discussed In the third section, the main results are presented Fourth section includes the theoretical discussion of and empirical tests for other confounding effects In the fifth section, the relevant robustness tests and their results are presented In the final section theoretical and managerial implications of the findings are discussed

possesses, which act as barriers to entry and generate abnormal returns Intangible assets are more likely to fall into the third category; such assets would have different economic value for different owner-firms, thus creating resource heterogeneity and

nonredeployability Intangible assets are information-based resources such as technology, know-how, innovativeness, patents, brand equity, employee motivation and commitment, customer service, corporate culture, and management skills Tangible assets, such as the plant, equipment, raw materials, and financial capital, have to be present for the business operations to take place whereas intangible assets are necessary for competitive success

2 Such resources are valued at their cost-reducing abilities (e.g a river whose water acts as a natural

coolant)

(e.g Prahalad and Bettis, 1986) It is also the case that well-managed firms with highly intangible assets have unrealized growth potential for future Well-managed bidders (high Tobin's q ) benefit positively from tender offers especially if the targets

were poorly managed (low Tobin's q ) (Lang et al., 1989) However, well-managed

targets benefit less than the poorly managed targets from a tender offer Two possible

explanations for this finding are offered (Lang et al., 1989) First, already well-managed

targets cannot be improved further through takeovers Second, the fact that the bidder

succeeds in acquiring such a high q target may mean that the target is not as valuable as

the bidder initially thought In both explanations, there is the underlying assumption that the motivation for the takeover is to improve the quality of the management of the target firm Based on this assumption, the ``surprise'' factor of announcing a takeover would be

less pronounced since the market also would most likely predict that the target is not

well-managed and who the potential buyers would be

Another possible reason why firms would want to buy targets with high

intangibles is to internalize the target firms' growth potential Bidder firms can grow through buying highly intangible targets ( q′>1 ) by funding, otherwise not funded, positive net present value (NPV) projects3 Acquirers that buy targets with less

3 If high q′ measures the growth opportunities stemming from intangible assets, above and beyond the

tangible assets, why would the target firm be willing to sell the firm? For target firms with high intangibles,

as the degree of nonredeployability increases, it will be inefficient for debt holders to finance new

investments because the increasing risk of default, coupled with high uncertainty regarding the flow of project cash-flows, would lead expected value of the debt holders' claims to decline In such cases target firms would have to forego some of the positive NPV projects because of financing Where projects face market breakdowns it is efficient to finance it through equity Therefore equity financing is an endogenous response to governance needs of suppliers of finance (in this context the bidder firms) who invest in nonredeployable projects These suppliers are the residual claimants who are awarded `control' over the board of directors

deployable assets foresee some positive NPV projects that only the merged company could undertake In this case, the information is most likely to be private to the buyer firm Therefore the ``surprise'' factor of an announcement of a takeover of a well-

managed target (high Tobin's q ) would be greater since the market becomes aware of

new information

Hypothesis 1a: Abnormal returns to highly tangible targets in the pre-announcement period would be higher than the returns to highly intangible targets

Hypothesis 1b: Announcement-day abnormal returns to highly intangible targets would

be higher than the returns to highly tangible targets

Intangible assets of a firm, such as R&D projects, patent stocks, and human capital are more likely to be undervalued by the market when they are bought by another company Such assets would generally have high target-firm specificity and therefore lower second-best use, which in turn leads to the undervaluation This expected

undervaluation is common to both the market and the potential buyers Given this adverse setup, if the market observes a bid for a highly intangible target, in theory it should signal the buyer's expectations to redeploy the target's intangible assets and create new growth opportunities Potentially, buying a firm with high intangibles is a more noisy way to obtain a particular subset of intangible assets Even though successful post-event

integration of targets with highly intangible assets as opposed to targets with highly tangible targets is more problematic, this is also expected by the market participants as

well as the buyers Therefore, this difference should have less bearing on the post-event

long-run abnormal performance of buyers 4

Overall, intangible sources of firm value are of a differential character, in that the advantage of those firms who own them may lead to competitive disadvantage of those who do not (Veblen, 1908) Conversely, tangible resources would not lead to competitive advantage over firms that lack such resources5 The main reason is that the price charged

by the owners of those tangible resources in factor markets would be equal to the income that would be generated by the buyers of those resources in product markets Also if the assets of the target have high redeployability (high ratio of tangibles), then acquiring such targets would be, on average, equivalent to internally developing the same resources because the costs associated with both methods would be approximately the same

2.1 Resource-Based View and Competitive Advantage

There has been a systematic effort to distinguish the types of assets (tangible or intangible) and their effect on the firm's competitiveness (e.g Coff 1999a, 1999b; Delios

and Beamish, 2001; Finkelstein and Haleblian, 2002; Hall 1992, 1993; Hitt et al., 1990,

4 What could make a difference is if managers' and the market's expectations of post-event integration of highly intangible targets differ significantly Managers may either have favorable private information that justifies the acquisition of highly intangible target, or act in self-interest as a result of agency conflicts (Jensen, 1986) specific to the context of buying highly intangible targets These two factors affect long-run abnormal firm performance in the opposite directions However, the related theories are less explicit about the aggregate direction

5 What about the highly synergistic acquisitions even though the target's assets are highly intangible? There

is no theoretical reason to believe that the probability and the magnitude of post-event synergies would systematically differ in cases where the target is highly tangible versus intangible The only assumption

required to follow through with this logic is the following: the potential for synergies is equally likely to

exist for both the acquirers of highly intangible targets as well as highly tangible targets

1991a, 1991b; Mowery et al., 1998) For example, Prahalad and Bettis (1986)

emphasized a ``dominant logic'' as an intangible asset that could be shared between firms through diversification to create economic value Firms that develop their core

competency, defined as ``the collective learning in the organization, especially how to coordinate diverse production skills and integrate multiple streams of technologies'', are more likely to have a strategic advantage over their competitors (Prahalad and Hamel, 1990:82)

According to the resource-based logic, resources that are rare, valuable, and inimitable are the real sources of competitive advantage (Barney, 1991a, 1991b; Barney, 1986; Conner, 1991; Rumelt, 1984; Wernerfelt, 1984) Of these firm-specific resources, intangible assets are more likely to be the source of sustainable competitive advantage6

because they are harder and more time-consuming to accumulate, provide simultaneous uses, and are both inputs and outputs of business activities Another characteristic of these intangible assets is that they are likely to be causally ambiguous (Dierickx and Cool, 1989) making them less likely to be imitated by competitors (Barney, 1991a) Therefore firms that seek to internalize intangible assets through acquiring highly

intangible targets are, on the one hand, trying to internalize new growth opportunities, but

on the other hand more likely to suffer from potential pricing, integration and

maintenance problems of the targets due to causal ambiguity, complexity and tacitness of the very same intangible assets

6Villalonga (1999) tested this assertion by using the predicted value from a hedonic regression of Tobin's q

as a measure of resource intangibility and found supporting results

Knowledge, one of the most important firm-specific intangible assets, has been developed as a reason for a firm's existence (Liebeskind, 1996; Spender, 1996)7 Highly firm-specific knowledge would be harder to transmit because fewer parties other than the innovator can benefit from the application of that knowledge (Henderson and Cockburn, 1996; McEvily and Chakravarthy, 2002) If the firm is an accumulation of idiosyncractic knowledge that is valuable, what are the methods of developing that firm-specific

knowledge base? One of the direct methods is to pursue M&A activity and try to

internalize knowledge intensive targets Such target firms are necessarily the ones with highly intangible asset stocks Buying a firm with high intangibles is a more noisy way to obtain a particular subset of intangible assets8

Although resource-based view and other related approaches to defining sources of competitive advantage favor the accumulation and utilization of intangible assets, one

cannot extend these arguments to suggest any systematic differences between the two

M&A strategies: buying highly tangible versus intangible targets If one argues that

7 Leibeskind conceptualized firms as structures to keep knowledge proprietary (1996) This assumes, in essence, that there is a fully efficient market for knowledge, and that without the firm the knowledge would have been diffused which in essence is similar to Porter's idea of entry barriers (1980) Conner and

Prahalad (1995) developed a resource-based theory of the firm based on knowledge as a valuable asset The main argument is that, absent opportunism, firm organization would provide a better mechanism to allow

an owner to provide his/her knowledge as input in the team production setting with higher value than in a market setting Information and knowledge are factors of production that could be sources of competitive advantage However, these factors of productions are also very hard to price; moreover their value is context- and owner-specific Given this, how would a strategic factor market for knowledge, and more generally intangible assets, work? I argue that the firm is an internal market for knowledge that decreases the inefficiencies of the external market for knowledge Once an individual offers his/her knowledge to the team production, the internal processes would translate it into a firm-specific knowledge base (Kogut and Zander, 1992)

8 An alternative and more precise way would be to develop intangible assets internally through firm-specific processes such as employee training and R&D In this case, because the direct method of internal

development would be more precise and less risky, the expected rate of return would more likely be lower when compared to the expected rate of return to the acquirers of highly intangible targets

buying highly intangible targets would more likely be a source of competitive advantage, then as a corporate strategy it constitutes a ``rule for riches'' and generates no sustainable competitive advantage (abnormal returns) This is analogous to the performance

implications of related versus unrelated acquisitions (Seth, 1990a; Singh and

Montgomery, 1987; Clark and Ofek, 1994) Empirical evidence supports the theoretical argument that both related and unrelated acquisition strategies can create significant synergies

As mentioned above, targets, on average, appropriate most of the economic value associated with the acquisition synergies, while buyers on average breakeven (e.g Jensen

and Ruback, 1983; Lubatkin, 1983, 1987; Agrawal et al., 1992) Buyers can create

sustainable competitive advantages only if there are unique, valuable and inimitable synergies with the targets (neither the target nor other bidders have this information) at the time of the acquisitions (Barney, 1988) However, this condition can equally apply for acquisition strategies of both highly tangible and intangible targets

Hypothesis 2: On average, there is no systematic above-normal performance differences between buying intangible versus tangible targets

Hypothesis 3: On average, corporate strategies of buying intangible or tangible targets cannot be a source of systematic competitive advantage

2.2 Controls for Other Factors

There are three alternative explanations9 that could affect the performance of M&A strategies: agency motives, financing and tax treatments, and behavioral

explanations regarding market reactions Controlling for these alternative explanations also serves as robustness checks for the tests of the above hypotheses

favorably, have an incentive to delay or advance the project resolution This type of

manipulation of information arrival can be achieved by greater investment in execution projects (which tend to resolve early) than exploratory projects (which tend to resolve late) (Hirshleifer et al., 2001) High ability managers are more likely to choose execution

projects and low ability managers are more likely to choose exploratory projects

M&A activity of highly tangible targets would be similar to the execution projects

in the sense that the project resolution (realization of synergies) is less likely to be

delayed However, M&A of highly intangible targets would be akin to exploratory

projects where the resolution of outcomes arrive later Therefore, if the agency motives play a differential role in target selection, then low ability managers are more likely to

9 I would like to thank Anita McGahan and Karen Wruck for providing useful insights

pursue highly intangible targets to take advantage of the longer horizon of project

resolution Also cash reserves increases the likelihood of engaging in M&A activity and cash-rich bidders destroy 7 cents in value for every excess dollar of cash reserves held (Harford, 1999)

Based on the above discussion, the buyers that pursue targets with highly

intangible targets are expected to be firms with low ability managers and higher free cash flows or cash reserves prior to the acquisition when compared to the buyers of highly tangible targets As a result, buyers of highly intangible targets are expected to

underperform in the long-run However, this is only a necessary but not sufficient

condition to cause a systematic abnormal performance between the two types of

acquisition strategies For such inefficient capital allocations to persist over time, there have to be other systematic factors that impede corrections to irrational expectations If there are such impediments, than it is plausible to entertain the existence of irrational expectations of firm performance when M&A of intangible targets are concerned

For the purposes of testing for agency motives in this context, the following hypotheses are developed:

Hypothesis 4a: If buyers of target firms with highly intangible assets have significantly higher levels of pre-event free-cash-flows or cash reserves then such targets tend to attract firms with costly agency problems

Hypothesis 4b: If the market does not expect agency motives to systematically drive the acquisition of targets with highly intangible versus tangible assets then there should

be no significant differences between the market reactions to the two types of M&A announcements

2.2.2 Information Asymmetry and Financing of Intangible Assets

There are two main concerns that might affect the long-run performance of buyers adversely The first one is related to the increased debt burden of the buyers to finance large transactions, such as M&A of highly intangible targets that also have large market capitalizations The second one is related to the adverse effects of information

asymmetries If the highly intangible targets are already overvalued then the buyers of such targets end up paying an excessive amount of premium

Target firms that have highly intangible targets have unrealized but valuable growth opportunities How can these firms finance their growth opportunities, such as R&D projects? First, retained earnings can be used Second, the firm can seek external financing from debt and/or equity financial markets Since governance is costly, the general rule is to reserve complicated forms of financing for complicated investments (Williamson, 1991, 1988) `Expressed in terms of asset specificity, fungible assets can be leased, semi-specific assets can be debt financed, and equity is the financial form of last

resort to be used for assets of a very nonredeployable kind' (Williamson, 1991: 84)10 Nonredeployability also suggests that the value of the assets in its first-best use is

significantly higher than the value of that same asset in its second-best use Therefore, we would expect nondeployable assets that are financed by equity to be intangible assets

10 Emphasis added

Overall, firms with highly intangible assets have a lower concentration of debt financing Titman and Wessel (1988) find a negative relationship between the measures of

uniqueness (e.g high R&D expenditure) and its debt ratio (Debt/Equity) Specifically, firms with low employee turnover and large R&D expenditures have relatively low debt ratios R&D intensive firms receive higher returns to firm shares following new debt issues (Alam and Walton, 1995; Zantout, 1997)

There are two methods of equity financing: a target firm can either issue new equity (seasoned equity offering-SEO) or be bought out by another company There are differences between the two methods of financing First, there is well-documented

negative stock market reaction to announcements of SEOs for the issuing firm The dominant explanation for this empirical regularity is based in information asymmetry between the firm's insiders and outsiders Myers and Majluf (1984) show that with

information asymmetry, insiders have an incentive to issue new equity when the firm is overvalued The stock market knows this, and therefore discounts the firm that issues SEO Second, especially for firms with highly intangible assets, financing through SEO is not preferred because of adverse effects of disclosing proprietary information about the firm's projects that were to be financed with the proceedings The only other method of equity financing is through M&As

Hypothesis 5: Buyers of highly intangible targets will decrease debt ratio when compared

to the buyers of highly tangible targets in the post-M&A period

According to Modigliani-Miller theorem (1958, 1961), firms should be indifferent

between internal and external sources of financing for the marginal R&D project since in both cases the cost of capital would be the same However, as widely researched, this theorem fails in practice due to several reasons As discussed in detail by Hall (2002), the divergence between the internal and the external cost of capital is due to the information asymmetries between the inventor and the investor, moral hazard on the part of the inventor due to the separation of ownership and control, and tax considerations

Asymmetric information creates a ``lemons'' market (Akerlof, 1970) for R&D project financing because the investors have a hard time distinguishing good projects from the bad ones when the projects are long-term R&D projects (Leland and Pyle, 1977) Therefore investors require a ``lemons'' premium (Hall, 2002) In the case of highly intangible targets, buyers would require a premium for the ``lemons'' problem associated with the information asymmetry between the target (inventor) and the acquirer (investor) On the other hand, a takeover would decrease the moral hazard problems that would have been present for the other potential investors if the target firm had issued equity or new debt instead of being acquired

Empirically there seem to be limits to leveraging strategy in R&D intensive industries such that the leveraged buy-outs in the 1980s that were financed by high levels

of debt were almost exclusively in industries where R&D intensity was insignificant (Hall, 2002, 1994, 1990; Opler and Titman, 1994) Tax treatment of R&D lowers the required rate of return, because the effective tax rate on R&D assets is lower than that on other types of tangible assets (Hall, 2002) On average, although these two effects act in opposite directions, the rate of return expected by the buyers of highly intangible targets would be higher than the expected rate of return expected by the buyers of highly

tangible targets

Moreover, buyer's of highly intangible targets expect to have post-event

integration problems (e.g Greenwood et al., 1994) This expectation increases the

riskiness of fully realizing the expected synergies As a result, buyer's expected rate of return increases, and the price (premium) to be paid decreases The effective tax rate on R&D assets is lower than tangible assets such as plant, property and equipment because R&D is expensed as it is incurred (Hall, 2002) This would mean that the rate of return for such investment would be lower

In comparing the acquisition of highly tangible targets versus highly intangible targets, while moral hazard, post-event integration, and the ``lemons'' problems are more severe for the case of intangible assets, tax considerations are more favorable As the riskiness of a project increases, the expected rate of returns increases As the expected rate of return increases the investor is willing to pay less

Hypothesis 6: Buyers of highly intangible targets will pay a lower premium in

comparison to buyers of highly tangible targets

2.2.3 Market Over- or Under-Valuation of Growth Opportunities

Firms are producers of tangible and intangible information about themselves Investors utilize firm-generated as well as other sources of information to decide on a

course of action in both financial and labor markets11 Tangible information is explicit performance measures such as book-to-market ratio of equity, earnings, and sales,

therefore any information generated by using financial statements will be tangible

(Daniel and Titman, 2001) However, this definition of tangible information may not correspond one-to-one to the tangibility of the asset base In essence, the validity of this tangible information is more in question when the firm's asset base is highly intangible A firm with highly tangible assets is more likely to provide all the relevant information about its nature as tangible information in the form of financial statements

On the other hand, a firm with highly intangible assets, has a harder time

reporting information about itself in the form of financial statements; rather, it is more likely to produce intangible information such as reputation or corporate culture (e.g

Louis et al., 2001) Moreover a common ratio, such as book-to-market, will be downward

biased due to the lack of a book value of intangible assets in the numerator This bias taints its validity as a measure of tangible information Investors are more likely to

overreact to intangible information, but rationally react to the tangible information

(Daniel and Titman, 2001) Distinctions between public versus private information follow

a similar logic (Daniel et al., 1998) Private information, such as the growth opportunities

of a firm, would be more ambiguously defined and is heterogeneous among investors12

11 For example, public firms, by law, produce more tangible information which is coded in financial

statements On the other hand private firms do not produce as much tangible information and yet the investors might benefit from other types of intangible information like reputation to form their expectations about the firm's performance

12The behavioral model of Daniel et al asserts that investors are more confident about their private signals

and overreact to such information (1998)

If we allow for the possibility of investor irrationality in the form of

overconfidence, stock prices reflect both systematic risk and misperceptions of firm's

prospects (Daniel et al., 2001)13 Therefore, it is reasonable to expect mispricings due to overconfidence to be more severe for firms with highly intangible assets, such as R&D

firms with relatively long-run projects (Daniel et al., 2001; Chan et al., 1999; Leland and

Pyle, 1977)

The mispricing will be equivalent to the divergence between the market's initial reaction to the announcement and the post-event long-run stock market performance of the buyers14

Hypothesis 7: Market is more likely to correctly evaluate and price the buyer's synergies

with the highly tangible target

Hypothesis 8: Market is less likely to correctly evaluate and price the buyer's synergies

with the highly intangible targets

13 The evidence on overconfidence is such that the individuals tend to be more confident in decision making situations where the feedback is delayed or inconclusive (Einhorn, 1980)

14However this divergence could be due to the revelation of unexpected but negative news after the event,

which could not have been incorporated into the market's reaction at the time of the announcement This case also requires the assumption that the unexpected but negative reaction to bad news is much more severe than the unexpected but positive reaction to good news Conversely, holding the severity of the reaction equal for both cases, the likelihood of unexpected negative news should be significantly higher than the likelihood of unexpected positive news

Hypothesis 8a: Negative deviation between the market's initial response and the long-run

performance of the buyers represent market overreaction

Hypothesis 8b: Positive deviation between the market's initial response and the long-run

performance of the buyers represent market underreaction

2.3 Methodology and Data

Management and financial economics literature consist of many event studies that detect abnormal stock returns following major corporate events or decisions, such as earning announcements, acquisitions, stock splits, or seasoned equity offerings However, studies that are concerned with long-run abnormal returns in the context of M&A are fewer in number15 Modified Tobin's q is used as a proxy for the measure of intangible

assets in the target firms which will be discussed in detail The analysis is concerned with measuring abnormal economic performance The most important step in measuring abnormal performance is to define a theoretically sound benchmark to proxy the expected

performance First, a brief discussion of modified Tobin's q will be provided Second, the

traditional method of calculating long-run abnormal returns will be discussed Second,

15 The main concern in these event studies is to determine whether there are abnormal returns associated with the firm-specific events There is considerable variation among these studies regarding the calculation

of abnormal returns and the statistical tests carried out to detect the presence of abnormal returns Refer to McWilliams and Siegel (1997) for a detail discussion of event studies Some representative studies are

Seth, (1990a, 1990b), Lahey and Conn (1990), Conn et al (1991), Haleblian and Finkelstein (1999)

the shortcomings of this method will be presented Barber and Lyon (1997) provide evidence that the common techniques used to calculate short-run abnormal returns, when applied over a longer horizon, are conceptually flawed and/or lead to biased test statistics Finally, the data and the methodology used in this study will be discussed

2.3.1 Valuation of Intangible Assets

In this paper, we use 1-year pre-event Tobin's q (Tobin, 1969) as an indicator of the target's intangible assets (Daniel et al., 2001; Klock and Megna, 2000; Loughran and Vijh, 1997; Lang et al., 1989) True q ratio of i th firm is defined as the market value

of all financial claims on the firm, MV , relative to the firm's total assets calculated as i

the sum of the i th firm's replacement values of tangible assets T and intangible assets i

i

I

i i

i

MV q

equal to 1 (Megna and Klock, 1993) The MV i of the target firm is measured as the sum of all outstanding claims on the firm including book value of debt D i and preferred stock P i , and market value of common equity C i Book value of total assets is denoted

as A i

Debtterm-Longs

Inventorieof

Book value

AssetsCurrent s

LiabilitieCurrent

where,

++

−

≡

++

≡

i

i i i i

D

C P D MV

Since in reality we do not have the theorized homogeneity due to differences in tax provisions, depreciation schedules, heterogeneous production functions given firm-

specific resources and capabilities, etc., the true equilibrium value of i th firm's q

defined by q' is unobservable Therefore one observes

i

i i

Tompkins, 1999; Bharadwaj et al., 1999) Also, the advantage of this approximation is

that bidder firms and the investors are more likely to use this simpler formula that

requires publicly available financial and accounting data It is reasonable to assume that the observed q′ values will approximate the portion of market value of the firm

explained by the firm's tangible assets If q′ is greater than 1 then there are specific valuable intangible assets contributing above and beyond the firm's tangible assets

firm-A q′ that is less than 1 would suggest that the firm's tangible resources and capabilities are underutilized or that there are value destroying intangible resources (e.g bad management) Such intangible resources, in theory, would have negative replacement value If we can fully explain the market value of a firm based on its tangible resources then the firm's q′ is equal to 1 In the mean time, the equilibrium value of q′ for any firm will change from year to year as there are changes in the mix of the old capital, new capital, and intangible capital, as well as changes in the macroeconomic and regulatory environment (Megna and Klock, 1993) Another source of change in q′ is M&As that are most likely to alter the mix of a firm's asset base and its economic value

Accounting based measures of intangibility are based on R&D and advertisement

expenditures (Lev and Sougiannis, 1996; Louis et al., 2001), and labor costs (Qian,

2001) Each firm's R&D (advertisement) capital is estimated from its pre-event history of R&D (advertisement) expenditures based on Lev and Sougiannis (1996) and Louis et al., (2001) as follows17:

17The financial information is taken from the COMPUSTAT/CRSP merged database provided by Wharton

Research Database Services R&D expenditure is annual data item 46; sales is annual data item 12; net

income is annual data item 172; dividends and book value of common equity are measured as annual data

4 3

2 1

4 3

2 1

2.04

.06

.08

.0

2.04

.06

.08

.0

∗+

∗+

∗+

=

∗+

∗+

∗+

∗+

=

it it

it it

it it

it it

it it

it it

ADV ADV

ADV ADV

ADV ADVC

RD RD

RD RD

RD RDC

where RDC it and ADVC it are the R&D and advertisement capital respectively

for firm i in year .t These estimates of R&D and advertisement capital measure the

proportion of past spending that is still productive in a given event-year t ={−5, , } based on current and past R&D and advertisement expenditures of RD it and ADV it This approximation assumes that the productivity of each dollar of spending declines linearly by 20% a year As a robustness check the approximations are recalculated by using a 15% capital amortization rate that is used by Hall et al (1988) for the database compiled on R&D activity The results are qualitatively unaffected The intangibility of the target firm is measured by the estimated R&D (advertisement) expense as a

percentage of either total sales, and cost of goods sold These ratios are recalculated using R&D (advertisement) capital Other measurers include Tobin's q′ , 4 -digit SIC

adjusted leverage ratio, total intangibles (R&D and advertisement capital) as a percent of cost of goods sold, and cash as a percent of sales As discussed earlier as the intangibility

of a firm's assets increases, its debt ratio decreases whereas its cash reserves increases to fund projects internally In Table-1 the descriptive statistics (in Panels A and B) and between-group equality tests are reported

2.3.2 Which measure of performance?

Strategic management is concerned with improving firm performance Thus any strategy such as M&A activity, is assumed to have an effect on firm performance, which

is the main concern of this paper Performance can be measured (accounting versus economic performance) in multiple ways over various lengths of time (short-run versus long-run) In M&A what is the relevant performance measure that would allow detection

of sustainable competitive advantage? This line of inquiry allows whether or not

acquiring firms internalize the economic value associated with the intangible assets of the target firm

Intangible assets of a firm are akin to latent assets (Brennan, 1990) in the sense that they cause a potential bias in the firm's market value mainly because they are

expensed in the accounting statements It would be the case that the accounting measures would understate the true return in the early years for investments in capacity, new product R&D, etc.18 On the other hand, firm value can be viewed as being generated by its tangible assets and intangible assets Accounting measures that use book values of the firm's assets would cause a downward bias in the performance measures as the

concentration of firm-specific productive intangible assets increases This downward bias would be most severe in the short-run because those development projects would not generate any income in the early years Overall, based on the extensive literature on the drawbacks of using accounting measures, economic measures of performance based on stock returns will be employed in this paper

18 This is especially critical for the valuation of growth opportunities of firms Although, by conventional accounting measures a highly intangible target may appear to be trading at a premium, for the buyer company the price paid can be justifiable

2.3.3 Why not traditional event methodology?

The traditional approach in corporate event studies is to calculate Cumulative

A bnormal Returns ( CAR ) A security's performance can only be considered `abnormal' relative to a particular benchmark (Brown and Warner, 1980) Therefore a model

generating normal returns (exante expected returns) has to be specified Almost

exclusively all event studies of this kind apply the Capital Asset Pricing Model (CAPM)

or market model19 to estimate the normal returns This method focuses on average market model residuals of the sample securities for a number of periods around the event date The null hypothesis is such that if there are no significant effects associated with the corporate event, CAR s will be a random-walk.20 The operationalization is as follows First, we define R it as the month t simple return on a sample firm i , E(R it) as the

month t expected return for the sample firm, and AR it as the abnormal return in

month t To calculate E(R it), we regress R it on the market portfolio R Mt over an estimated period of t=−1L−k days preceding the event as R it =a i +b i R Mt , where

i

a and b i are the ordinary least square parameter estimates.21 Then

19Refer to Chatterjee et al (1999) for a discussion of the CAPM and the strategic theory of risk premium

20 The average residual in the event time are independent and identically distributed, with a mean of zero

21 It is important to notice that the coefficient a i is in fact the systematic risk factor, β in the CAPM

Mt i i

Generally these studies identify τ =±5 or some other shorter window of

analysis around the event date of t=0 The main argument behind this identification is that the stock prices reflect the discounted economic value of all future expectations As discussed earlier, M&A activity is an event that is much more complex than any other corporate event such as an earnings announcement Also the nature of the event is

conducive to exacerbate any potential investor biases such as overconfidence Therefore studies as early as 1974 have started looking at longer post-event periods to gauge long-run performance effects of M&A activity.22 All of these studies used CAR and overall document negative CAR for mergers and positive CAR for tender offers Also most of these studies document a less than 3% CAR in magnitude for combined M&A activity with varying signs