Journal of Economics and Business 72 (2014) 30– 43

Contents lists available at ScienceDirect

Journal of Economics and Business

Antitakeover provisions, managerialentrenchment and firm innovation

Atreya Chakrabortya,∗, Zaur Rzakhanova, Shahbaz Sheikhb

a Department of Accounting and Finance, University of Massachusetts Boston, 100 Morrissey Boulevard,Boston, MA 02125, United Statesb DAN Management and Organizational Studies, The University of Western Ontario London, OntarioN6A5C2, Canada

a r t i c l e i n f o

Article history:Received 17 February 2012Received in revised form 5 June 2013Accepted 9 October 2013

JEL classification:O16O31O32G31G34

Keywords:PatentsCitationsR&DEntrenchmentCorporate governance

a b s t r a c t

We explore the relation between antitakeover provisions (i.e.managerial entrenchment) and firm performance in innovation.Empirical results indicate that an increase in antitakeover pro-visions is negatively related to number of patents and numberof citations to patents. Thus managers who are protected fromtakeover market perform worse on innovation. However, the nega-tive relation between antitakeover provisions and firm innovationholds only for low-tech firms. For high-tech firms, this relation isnot statistically significant. One possible explanation is that high-tech firms have to innovate continuously to survive in the longrun. The competitive pressure to innovate or perish dissipates thenegative effect of managerial entrenchment on firm innovation.Overall, our results support the agency based explanation of therelation between antitakeover provisions and firm performance ininnovation.

© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Firm innovation plays a critical role in creating, sustaining and adding firm value. However,investment in innovative activities is a higher risk investment compared to investment in capitalexpenditures due to higher probability of failure (Bhagat & Welch, 1995; Holmstrom, 1989). Since

∗ Corresponding author. Tel.: +1 617 287 7673; fax: +1 617 287 7877.E-mail address: atreya.chakraborty@umb.edu (A. Chakraborty).

0148-6195/$ – see front matter © 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.jeconbus.2013.10.001

A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43 31

R&D expenditures cannot be capitalized, investment in innovation depresses short term accountingearnings and reduces accounting based bonuses for managers. Moreover, investment in innovation ischaracterized by long gestation periods as cash flows could stretch beyond the tenures of managers(Dechow & Sloan, 1991; Gibbons & Murphy, 1992). Consequently, risk-averse managers may not haveincentives to invest in innovation. However, firms that do not innovate lose market value and becomedesirable takeover targets. Thus the presence of an active external control market disciplines managersand forces them to invest in risky but value-maximizing projects. Managers do not like the takeoverpressures and have incentives to favor antitakeover provisions that protect them from such pressures.

Although antitakeover provisions have been studied in literature, there is no theoretical consensuson their effect on firm innovation. The agency cost based “managerial entrenchment hypothesis” arguesthat protecting managers from market for corporate control makes management more entrenchedand further misaligns the interests of managers from those of shareholders (DeAngelo & Rice, 1983;Jensen & Meckling, 1976; Jensen & Ruback, 1983). The managerial entrenchment view thus predictsthat antitakeover provisions increase entrenchment and negatively affect managerial incentives toinnovate.

On the other hand, the “managerial myopia hypothesis” asserts that the threat of a hostile takeoverand the resulting career concerns about possible job loss make managers myopic and force themto make short-sighted investment decisions (Scherer, 1982; Stein, 1988). Stein (1988) argues thatshareholders cannot value investments in innovation due to information asymmetry and drive downthe market value of companies that make such investments, exposing these companies to takeoverthreats. Consequently, managers invest only in short term projects that the market can easily evaluateand avoid making investments in long term innovative projects. Protecting managers from takeoverthreats encourages them to take long term view and invest in innovation. Thus the managerial myopiaview predicts a positive relation between antitakeover provisions and firm innovation.

Given the absence of theoretical consensus, the relation between antitakeover provisions and firminnovation is an empirical question. This paper estimates the relation between antitakeover provisionsand firm innovation. Following the current literature on antitakeover provisions, we call the presenceof antitakeover provisions “managerial entrenchment”. Bebchuk, Cohen, and Ferrell (2009) constructan entrenchment index (E-index) from the corporate charter provisions of the Investor ResponsibilityResearch Centre (IRRC). We use their index to measure managerial entrenchment.

We use number of patents and number of citations to patents held by a firm to measure firmperformance in innovation. Our results indicate that managerial entrenchment is negatively relatedto firm performance in innovation. An increase in entrenchment (i.e. increase in E-index) leads to poorperformance on both measures of firm innovation. Prior research has shown that a subset of charterprovisions that increases managerial entrenchment is associated with lower firm value (Bebchuk et al.,2009).1 Our results provide an explanation of the link between entrenchment and firm value – thesame provisions that reduce firm value are also associated with lower performance in innovation. Thus,increased managerial entrenchment could lower firm value via its negative impact on firm innovation.

We also examine if the relation between entrenchment and innovation is sensitive to industrycharacteristics. Here our results indicate that the negative relation between entrenchment and inno-vation is sensitive to firm being a low-tech or high-tech. The negative coefficient on E-index is notstatistically significant at any conventional level for high-tech firms. It is statistically significant onlyfor low-tech firms. One possible explanation is that high-tech firms by definition are innovative. Theirsurvival and long-run profitability depends on constant innovation. Moreover, intense product mar-ket competition and rapid product obsolescence force such firms to innovate. The negative impact ofentrenchment for these firms is dissipated by positive competitive pressures.

Rest of the study is organized as follows. Section 2 provides motivation and hypotheses. Section 3discusses measures of managerial entrenchment and innovation. Section 4 describes data and sampleconstruction. Section 5 discusses empirical methodology, Section 6 presents results and Section 7concludes.

1 This result is consistent with the broader literature that suggests that weak shareholder rights are associated with greateragency problems and lead to poor firm performance (Becht et al., 2003).

32 A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43

2. Motivation and hypotheses

Protecting managers from the market for corporate control via antitakeover provisions has impli-cations for investment in firm innovation and for firm market value. The managerial entrenchmenttheory predicts that increase in entrenchment via antitakeover provisions weakens the discipliningeffect of external takeover threats and results in less investment in innovation. The myopia hypoth-esis argues that increased protection from takeover threats encourages managers to take more riskand invest in long-term risky but value-increasing projects. However, the empirical research gener-ally finds that antitakeover provisions discourage investment in innovation. Mahoney, Sundramurthy,and Mahoney (1997) find that the introduction of antitakeover provisions on subsequent long-terminvestments is negative. Lhuillery (2006) finds that firms with governance practices, that are shapedto defend shareholders’ rights, are more R&D intensive. Atanassov (2013) studies the relation betweenBusiness Combination Laws and innovation. He finds that firms that incorporate in states that enactantitakeover laws, innovate less than those that incorporate in states without such laws. Our studyis different from Atanassov (2013) in a number of ways. First, our measure of managerial entrench-ment is an index that captures the degree of entrenchment while his measure is a binary variable(business combination laws) that takes on only two values. Second, our study uses panel data for thetime period 1996–2006 while his study is for the period 1976–2000. Third, we control directly forCEO age, option based compensation, founding family relationship and board characteristics as othergovernance mechanisms. Moreover, our study examines whether industry type has any effect on thisrelation.

Prior empirical work has also shown that increased managerial entrenchment negatively affectsshareholder value (Bebchuk et al., 2009; Gompers, Ishii, & Metrick, 2003). Masulis, Wang, and Xie(2007) find that managers at firms protected from takeover market are more likely to engage inempire-building acquisitions that destroy shareholder value. Moreover, Wang and Xie (2009) findthat acquisition of poorly governed firms (with weaker shareholder rights) by acquirers with strongshareholder rights creates positive synergistic gains for shareholders. Thus protecting managers fromtakeover threat could actually destroy firm value. Based on prior empirical literature, we hypothesizethat managerial entrenchment is negatively related to firm performance in innovation.

Hypothesis 1. Managerial entrenchment is negatively related to firm performance in innovation.

The relation between innovation and managerial entrenchment may also depend on industry char-acteristics. Competitive success and long-run survival of high-tech firms depend on the effectiveintroduction of new and superior products. Firms in high-tech industries must innovate to stay inbusiness. Intense market competition and rapid product obsolescence force these firms to innovate.Kam, Kiese, Singh, and Wong (2003) find that high-tech firms are more innovative than traditionalfirms. We argue that the disciplining effect of competitive pressures on innovation in high-tech firmsmay dissipate the negative effect of managerial entrenchment, which leads to our next hypothesis.

Hypothesis 2. The negative effect of managerial entrenchment on firm performance in innovationis less significant for high-tech firms than for low-tech firms.

3. Measurement

3.1. Measuring entrenchment

Bebchuk et al. (2009) construct an entrenchment index (E-index) based on six provisions out of thetwenty-four provisions that the Investor Responsibility Research Centre (IRRC) monitors for institu-tional investors. Each of the six provisions carries equal weight in the E-index. Four of these provisions(staggered boards, limits to shareholder amendments of the bylaws, supermajority requirements formergers, and supermajority requirements for charter amendments) set statutory limits on shareholdervoting powers. The other two provisions (poison pills and golden parachutes) make hostile takeoverattempts more expensive. Bebchuk et al. (2009) argue that all six provisions of the E-index are designedto prevent change in control and discuss how each of the six provisions insulates management from

A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43 33

takeover pressures. Firms may add and drop provisions every year. The adoption of each provisionadds one point to the index while the removal of any provision subtracts one point from the index.Changes in the number of these six provisions in any given year measure changes in the level of theindex and entrenchment.

Since we are interested in entrenchment that protects managers from threat of takeover or makestakeover more expensive, we use the E-index of Bebchuk et al. (2009) as a measure of managerialentrenchment. The data on E-index comes directly from Bebchuk et al. (2009).2

3.2. Measuring innovation

Measuring firm innovation is difficult as innovation is a complex process. It involves various formsincluding the product, process and business methods innovation. Much of the previous literature usesR&D expenditures to measure innovation. This is a very useful measure as R&D expenditure is animportant input into the innovation process and is directly under managerial discretion. However,since we are interested in innovation output, we use patent data to measure innovation. Patents havelong been used to measure firm innovation output and technical change (Hall, Jaffe, & Trajtenberg,2005). Schmookler (1966), Scherer (1982) and Griliches (1984) are among the first studies that usepatents as a measure of innovation output. Simple patent counts have their own limitations (Griliches,1992) as patent counts do not convey any information about patent’s economic value. Moreover, insome industries firms prefer to keep trade secrets for various reasons (Kleinknecht, Van Montfort, &Brouwer, 2002; Mansfield, 1985). In such a case industries where firms primarily rely on trade secretswill appear to be less innovative than industries where firms choose to file a patent. Despite theselimitations, patent counts provide significant improvement over R&D expenditures as a measure ofinnovation output.

To overcome the limitations associated with patents, researchers use citations received by a patent.Citations not only measure innovation output, but also signal the economic “importance” or “quality”of this output (Jaffe, Trajtenberg, & Fogarty, 2000). If patent X cites patent Y, it shows that patentY could have valuable legal rights that could be used in generating future cash flows. Studies likeTrajtenberg (1990) and Hall et al. (2005) suggest that citations are correlated with market value ofa firm. Lanjouw and Schankerman (2004), using an index of patents that includes citations, find thatthere is a strong association between the stock market valuation of an organization and the averagevalue of patents it holds.

In this study, we use number of patents and number of citations to patents to measure the quantityand quality of innovation output. We also use R&D expenditures as a traditional measure of innovationto see if our results are sensitive to a change in the definition of innovation. Since firms with long historyof patenting and R&D expenditures are more likely to file a patent and receive citations, we includethe stock of previously held patents and R&D intensity in all regressions.

4. Data and sample construction

We construct our sample by combining datasets from a number of different sources. We useNational Bureau of Economic Research (NBER) Patent Database to measure firm level innovation. TheNBER data is available for the period 1976–2006. We recognize the limitations of the data and deviseour empirical strategy accordingly. Hall, Jaffe, and Trajtenberg (2002) suggest that application date isthe “relevant time placer for patents” as inventors “have a strong incentive to apply for a patent assoon as possible following the completion of the innovation”. Following Hall et al. (2002) we measurepatent counts at the application date as the data reports all patents that were ultimately granted. Ourresults are similar if we use the grant date.

The patent data however suffers from truncation problems. This is because there is an average lag oftwo years between the date when the application is filed and the date when the patent is granted. Thusas we come closer to the last years of data we find numerous missing values. Hall et al. (2002, 2005)

2 http://www.law.harvard.edu/faculty/bebchuk/data.shtml.

34 A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43

show an easy and straightforward way to correct the truncation bias in patent data. It involves usingthe application-grant empirical distribution to compute weight factors and multiplying the numberof patents by relevant weight factors to correct the truncation bias. We use the Hall et al. (2002, 2005)methodology to correct for truncation bias in patent counts. The details are given in Appendix A.However, there are still significantly fewer observations in the last two years of the data.

Our other measure of innovation output is the number of citations received by a patent. Like patentcounts, citation counts also suffer from truncation problem as patents continue receiving citations overlong periods of time but we only observe citations up to the last year of the available data. Since thecitation data ends in 2006, a patent granted in 1992 will show more citations in the data comparedto a patent received in 2002. Fortunately we do not need to correct the truncation in citations dataas the updated NBER data provides truncation corrected citations data using the Hall et al. (2002)methodology.

Hall et al. (2002) note “. . . standing by itself the fact that a given patent has received 10 or 100citations does not tell you whether that patent is “highly” cited. Intrinsically, information on patentcitations is meaningful only when used comparatively.” It is therefore important to benchmark meas-ures of innovation output to make them comparable across industries. We therefore normalize patent(citation) counts for each firm in a given year by dividing them by the mean number of patent (citation)counts for all firms in the same year and technology class. To control for the effect of patenting expe-rience we include the previous stock of patents measured by the cumulative number of patents in allregressions. We use Hall et al. (2002) methodology to construct patent stock using a 15% depreciationrate.

We use the E-index of Bebchuk et al. (2009) to measure entrenchment. This index is constructedfrom the IRRC data on corporate governance provisions. Although it spans a period from 1990 to 2006,the IRRC does not collect and publish such data every year. Bebchuk et al. (2009) replace missing yearobservations with the last available data.

Firm-specific accounting, financial and R&D data come from Compustat manufacturing files. Dataon CEO compensation and age are from the S&P ExecuComp database and board details come fromIRRC directors’ data set. The ExecuComp data start from 1992 and is available until 2006. The directors’data is only available from 1996. Following prior studies, we replace missing values of R&D expendi-tures with zero. We however include a binary variable called “missing R&D” to control for the missingR&D observations. The NBER PDP data project provides detailed help in matching the patent data toCompustat data.3 After merging the patent data with the Compustat firms, we merge the resultingdataset with ExecuComp and IRRC data on corporate charter provisions. Our sample for the period1996–2006 that includes all exogenous variables consists of 5469 firm-year observations. However,following Holthausen, Larcker, and Sloan (1995) we measure firm innovation output at t + 2 and controlvariables at period t. This two-year gap helps in mitigating concerns about simultaneity of entrench-ment and innovation output. With this lag between entrenchment and innovation, our final samplefor benchmark regressions consists of 4508 firm year observations.

5. Empirical methodology

Since we are interested in estimating the effect of managerial entrenchment on innovation, weestimate the following equation:

Innovationt+2 = f (entrenchmentt , firm and CEO characteristicst) (1)

Our dependent variables are:

patentsmean patents (year, technology class)

andcitations

mean citations (year, technology class)

3 The file NBER PDP Project User Documentation: Matching patent data to Compustat firms is available athttp://elsa.berkeley.edu/∼bhhall/bhdata/html along with updated NBER patent data.

A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43 35

We estimate Eq. (1) using OLS method with industry and year fixed effects. Atanassov (2013) findsthat most of the effect of antitakeover laws occurs two or more years after they are enacted. Holthausenet al. (1995) also use two-year lag to measure the impact of CEO compensation on innovation. Followingthese studies we use two-year lag between antitakeover provisions and firm innovation. We also useone year lag to see if our results are sensitive to this definition.

5.1. Control variables

It is important to control for other variables that affect the relation between managerial entrench-ment and innovation. Since all innovation results from investment in R&D, we use the ratio of R&Dexpenditures to sales in all regressions. There are a number of firms that have missing R&D observa-tions in the data. Following prior studies, we replace missing R&D with 0. However, to control for thiseffect, we include a binary variable “missing R&D” in all specifications. Missing R&D variable equals 1if the firm has missing R&D and 0 otherwise. Greater stock of previous patents indicates greater expe-rience in innovation. Firms with long and successful history of innovation are more likely to innovate.We use the cumulative number of patents that a company holds in each year to control for this effectusing a 15% depreciation rate.

Among firm specific control variables, firm size is very important. Previous studies have identi-fied decreasing returns to scale in innovation (Cockburn & Henderson, 1996), suggesting that sizecould be an important determinant of innovation output. We measure size by the natural log of salesand expect this variable to be positively related to innovation. Firm operating performance may alsoaffect innovation. It is argued that firms with better operating performance are better at innovation(Eberhart, Maxwell, & Siddique, 2004). We use return on assets (ROA) as a measure of firm operat-ing performance. Internal financing is an important source of funds for innovation-driven firms. Wecalculate coverage ratio as the ratio of operating income to the sum of interest, taxes and preferreddividends.

Innovation is inherently risky and depends on managerial risk taking incentives. There are studiesthat show that managerial compensation incentives affect firm investment policy (Coles, Daniel, &Naveen, 2006) and R&D activities (Lerner and Wulf, 2007; Xue, 2007). We include CEO age and shareof options in total compensation to control for managerial risk taking incentives. It is also importantto control for other forms of monitoring and governance. The proportion of outside directors on theboard measures board independence and indicates better governance. We include the percentage ofoutside (independent) directors on the board in our specifications and expect a positive sign on thisvariable. We also include total number of directors (board size) on the board. Similarly, many high-tech firms are founded by researchers who retain large equity holdings after the firm goes public. Insuch case the agency explanation of the relationship between innovation and entrenchment may nothold. We include a founding family binary variable that equals 1 if the CEO is a founder or belongs toa founding family and 0 otherwise.

Prior work shows that market structure is an important determinant of firm innovation (Blundell,Griffith, & Van Reenen, 1999; Koeller, 1995). We include 2-digit Herfindahl index of industry concen-tration in all specifications. Lower values of this index indicate more competition and more equallydistributed market share. We do not have a prior sign expectation on this variable. The propensityto patent and cite previous patents could vary by industry (Hall et al., 2002; Mairesse & Hall, 1996).It is necessary to control for the systematic differences in patent application and citation behavioracross industries. We include industry indicator variables as defined in Mairesse and Hall (1996) inall specifications. We also include year indicator variables to control for any systematic differences innumber of patents and citations across years.

A detailed description of variables is given in Appendix B. Table 1 presents the summary statis-tics of sample variables and Table 2 gives the correlation matrix of these variables along withp-values.

A cursory look at Table 1 shows that an average firm has 31 patents and 6 citations. However,high-tech firms hold an average of 68 patents and 10 citations while low-tech firms hold an averageof 15 patents and 4 citations.

36 A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43

Table 1Summary statistics for full sample and the sub-samples of high-tech vs. low-tech firms.

Variables Full sample High-tech firms Low-tech firms

Mean Std. dev. Mean Std. dev. Mean Std. dev.

Patents 30.99 166.32 68.36 266.95 15.21 91.83Citations 5.79 16.39 10.13 23.40 3.95 11.81Patent stock 64.67 324.97 142.17 519.65 32.14 181.74E-index 2.43 1.30 2.07 1.17 2.53 1.32Sales ($ millions) 7.14 1.63 6.14 1.85 7.40 1.46R&D intensity (%) 4.56 12.52 20.01 21.17 0.50 1.07R&D missing 0.47 0.50 0.00 0.00 0.60 0.49Coverage (times) 1.62 9.13 0.96 4.02 1.81 10.12ROA (%) 12.07 15.61 7.84 27.21 13.21 10.21Herfindahl index 0.11 0.10 0.07 0.05 0.12 0.11Family 0.05 0.22 0.08 0.27 0.04 0.20CEO age (years) 55.57 7.59 53.69 7.61 56.07 7.51Option compensation (%) 38.27 24.04 50.25 26.46 34.74 22.08Total directors 9.78 5.66 8.55 3.71 10.18 6.12Outside directors (%) 64.43 18.47 66.18 16.48 63.86 19.04

The data is for the period 1996–2006. Patents is the number of patent applications filed by a firm that and Citations is thenumber of citations received by the patents of a firm. E-index is the entrenchment index of Bebchuk et al. (2009). Patent stock isthe cumulative number of patents filed by a firm. R&D intensity is R&D expenditure to sales. Coverage is interest coverage ratio-proxy for leverage. ROA is return on assets. Herfindahl index is based on sales in 2-digits SIC industries. Outside directors is theratio of outside directors to total directors. Total directors are the total number of directors on the board. Option compensationis the ratio of Black and Scholes value of CEO stock options to total compensation.

6. Results

6.1. Managerial entrenchment and firm innovation

Table 3 presents benchmark results on the relation between managerial entrenchment and firminnovation. The coefficients on the E-index are negative for both patents and citations. An increasein the E-index is associated with a decrease in both the number of patents and number of citationsto patents. Thus our results support hypothesis 1. It seems that protecting managers from corporatecontrol market via antitakeover provisions distorts managerial incentives to innovate and adverselyaffects firm performance in innovation. This result is quite important as it provides a possible explana-tion of why antitakeover provisions reduce firm market value as documented by previous studies. Ourresults show that the same provisions that reduce firm market value also reduce firm performance ininnovation. Thus increased managerial entrenchment could lower firm value via its negative impacton innovation.

Other control variables are also statistically significant in this specification. The stock of pre-vious patents and R&D intensity both have positive impact on patents and citations. Firms thathave prior experience in successful patents and firms that invest more in R&D as a percentage oftheir sales outperform firms with less prior innovation experience and lower R&D intensity. Thecoefficients on missing R&D are negative and statistically significant indicating that firms with miss-ing R&D values have lower innovation output. Firm size and profitability have positive coefficientsand show that large and more profitable firms also perform better on innovation. All the CEOcharacteristics (i.e. CEO age, relation to the founding family and option based compensation) havestatistically insignificant effect on firm innovation. Among the board characteristics, the coefficientson outside directors are positive and statistically significant. It seems that firms with indepen-dent boards perform better in terms of innovation. However, the size of the board in terms oftotal number of directors does not seem to have any significant effect on firm performance ininnovation.

A.

Chakraborty et

al. /

Journal of

Economics

and Business

72 (2014) 30– 4337

Table 2Correlation matrix with p-values in parantheses.

Variable I II III IV V VI VII VIII IX X XI XII XIII XIV XV

I Patents 1II Citations 0.10 1.00

0.00III Patent stock 0.96 0.09 1.00

0.00 0.00IV E-index −0.13 −0.08 −0.14 1.00

0.00 0.00 0.00V Sales 0.23 −0.04 0.24 0.00 1.00

0.00 0.00 0.00 0.99VI R&D 0.04 0.10 0.05 −0.10 −0.41 1.00

0.00 0.00 0.00 0.00 0.00VII R&D missng −0.13 −0.14 −0.13 0.05 0.12 −0.34 1.00

0.00 0.00 0.00 0.00 0.00 0.00VIII Coverage −0.01 −0.02 −0.01 0.00 0.19 −0.11 0.05 1.00

0.50 0.04 0.47 0.66 0.00 0.00 0.00IX ROA 0.03 0.01 0.03 0.01 0.25 −0.42 −0.01 0.15 1.00

0.00 0.20 0.00 0.21 0.00 0.00 0.41 0.00X Herf index −0.02 −0.05 −0.03 −0.01 0.11 −0.14 0.07 −0.01 0.09 1.00

0.01 0.00 0.00 0.06 0.00 0.00 0.00 0.37 0.00XI Family 0.00 0.04 0.00 −0.08 −0.07 0.08 −0.06 0.00 0.03 0.02 1.00

0.79 0.00 0.98 0.00 0.00 0.00 0.00 0.72 0.00 0.01XII CEO age −0.02 −0.06 −0.02 0.00 0.13 −0.12 0.07 0.01 0.04 0.04 0.05 1.00

0.06 0.00 0.05 0.82 0.00 0.00 0.00 0.15 0.00 0.00 0.00XIII Opt comp 0.06 0.08 0.06 −0.13 −0.08 0.20 −0.17 0.01 0.00 −0.05 0.06 −0.13 1.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.50 0.74 0.00 0.00 0.00XIV Tot dir 0.04 0.00 0.04 0.08 0.21 −0.09 0.05 −0.01 0.02 0.04 −0.06 0.03 −0.08 1.00

0.00 0.90 0.00 0.00 0.00 0.00 0.00 0.16 0.01 0.00 0.00 0.00 0.00XV Out dir 0.06 −0.05 0.07 0.16 0.15 0.05 −0.14 0.00 0.00 −0.04 −0.11 −.02 −0.01 0.05 1.00

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.88 0.75 0.00 0.00 0.02 0.22 0.00

38 A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43

Table 3Antitakeover provisions, managerial entrenchment and firm innovation: benchmark results.

Variables Patentst+2 Citationst+2

E-index −0.0270*** −0.0271***(0.000) (0.000)

Size 0.1298*** 0.0477***(0.000) (0.000)

Patent stock 0.0009*** 0.0002***(0.000) (0.000)

R&D intensity 0.6324*** 0.4027***(0.000) (0.000)

R&D missing −0.1410*** −0.1269***(0.000) (0.000)

Coverage ratio −0.0126** −0.0052(0.017) (0.395)

ROA 0.3348*** 0.1557(0.000) (0.161)

Hefindahl index −0.3364*** −0.3644***(0.000) (0.000)

Family 0.0203 0.0019(0.566) (0.966)

CEO age 0.0008 −0.0004(0.419) (0.762)

Option compensation 0.0404 0.0287(0.225) (0.452)

Outside directors 0.0008** 0.0021***(0.040) (0.000)

Total directors 0.0005 −0.0008(0.580) (0.512)

Year fixed effects Yes YesIndustry fixed effects Yes YesR2 0.585 0.166Observations 4508 4508

Coefficients reported here are from OLS estimation of the impact of antitakeover provisions on firm innovation for the period1996–2006. Patents are the number of patents divided by mean patents in the same year and technology class. Citations arethe number of citations received/mean citations in the same year and technology class. E-Index is the entrenchment index ofBebchuk et al. (2009). Size is natural log of sales. Patent stock is the cumulative number of patents filed by a firm. R&D intensityis the ratio of R&D expenditure to sales. R&D missing is a binary variable that equals 1 if R&D expenditures are missing and 0otherwise. Coverage is interest coverage ratio- proxy for leverage. ROA is return on assets. Herfindahl index is based on sales in2-digits SIC industries. Outside directors is the ratio of outside directors to total directors. Total directors are the total numberof directors. Option compensation is the ratio of Black and Scholes value of CEO stock options to total compensation. Robustp-values are reported in parentheses. *, ** and *** indicate significance at 10%, 5% and 1% respectively.

6.2. Industry type, managerial entrenchment and firm innovation

Next, we attempt to see if industry characteristics and the technological environment affect therelation between managerial entrenchment and innovation. For this purpose, we divide our sampleinto high-tech and low-tech firms based on R&D intensity. Massachusetts High Technology Council(MHTC) classifies a firm as high-tech if its R&D intensity is 5 percent or more. We follow this definitionand classify a firm as a High-tech firm if its R&D expenditures in any given year are greater than 5percent of its sales.4 We then create two interaction variables high-tech × E-index and low-tech × E-index to see the effect of technology on innovation. The results are presented in Table 4.

The coefficient on high-tech × E-index is negative but not statistically significant at any con-ventional level. For high-tech firms, entrenchment does not seem to have any significant effect oninnovation. This result supports our second hypothesis. Since the competitive success and long runprofitability of high-tech firms depends on the effective introduction of new and superior products,

4 We also define high-technology firms as firms with R&D to sales ratio above sample median, the low-technology firms arefirms with R&D to sales ratio below sample median and find similar results.

A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43 39

Table 4Antitakeover provisions and firm innovation: effect of industry characteristics.

Variables Patentst+2 Citationst+2

High-tech × E-index −0.0181 −0.0221(0.170) (0.172)

Low-tech × E-index −0.0271*** −0.0276***(0.000) (0.001)

High-tech 0.1270*** 0.0517(0.008) (0.314)

Size 0.1345*** 0.0497***(0.000) (0.000)

Patent stock 0.0008*** 0.0002***(0.000) (0.000)

R&D intensity 0.4151*** 0.3088***(0.000) (0.003)

R&D missing −0.1130*** −0.1147***(0.000) (0.000)

Coverage ratio −0.0179*** −0.0075(0.001) (0.222)

ROA 0.3259*** 0.1522(0.000) (0.171)

Hefindahl index −0.2901*** −0.3446***(0.002) (0.000)

Family 0.0195 0.0014(0.580) (0.974)

CEO age 0.0007 −0.0004(0.433) (0.755)

Option compensation 0.0181 0.019(0.588) (0.617)

Outside directors 0.0009** 0.0021***(0.027) (0.000)

Total directors 0.0005 −0.0008(0.593) (0.507)

Year fixed effects Yes YesIndustry fixed effects Yes YesR2 0.589 0.167Observations 4508 4508

Coefficients reported here are from OLS estimation of the impact of antitakeover provisions on firm innovation for the period1996–2006. Patents are the number of patents divided by mean patents in the same year and technology class. Citations arethe number of citations received divided by mean citations in the same year and technology class. E-Index is the entrenchmentindex of Bebchuk et al. (2009). Size is natural log of sales. High-tech is a binary variable that equals 1 if firm’s R&D intensityis greater than 5 percent. Low-tech is a binary variable that equals 1 if firm’s R&D intensity is less than or equal to 5 percent.Patent stock is the cumulative number of patents filed by a firm. R&D intensity is the ratio of R&D expenditure to sales. R&Dmissing is a binary variable that equals 1 if R&D expenditures are missing and 0 otherwise. Coverage is interest coverage ratio-proxy for leverage. ROA is return on assets. Herfindahl index is based on sales in 2-digits SIC industries. Outside directors is theratio of outside directors to total directors. Total directors are the total number of directors. Option compensation is the ratioof Black and Scholes value of CEO stock options to total compensation. Robust p-values are reported in parentheses. *, ** and*** indicate significance at 10%, 5% and 1% respectively.

such firms must innovate to stay in business. The disciplining effect of product market competitionon innovation in high-tech firms seems to dissipate the negative effect of managerial entrenchmenton firm performance in innovation. However, the coefficient on low-tech × E-index is negative andsignificant in all specifications. In low-tech firms entrenchment seems to be more likely to exacerbateagency problems and negatively affect the incentives to innovate.

6.3. Robustness

Our results may be sensitive to our definition of firm innovation. In order to check the robustnessof our empirical results to a change in the definition of innovation, we re-estimate our benchmarkregression using R&D intensity as a measure of firm innovation. We also report results from models

40 A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43

Table 5Antitakeover provisions and innovation: robustness.

Variables R&D intensityt+2 Non-missing R&D intensityt+2 Patentst+1 Citationst+1

E-index −0.0077*** −0.0107*** −0.0248*** −0.0106*(0.000) (0.000) (0.000) (0.095)

Size −0.0227*** −0.0272*** 0.1320*** 0.0456***(0.000) (0.000) (0.000) (0.000)

Patent stock 0.0000*** 0.0000*** 0.0009*** 0.0001***(0.000) (0.000) (0.000) (0.000)

R&D intensity 0.6673*** 0.4265***(0.000) (0.000)

R&D missing −0.1378*** −0.1266***(0.000) (0.000)

Coverage ratio −0.0028 −0.0022 −0.0091** −0.0064(0.320) (0.552) (0.050) (0.214)

ROA −0.3559*** −0.4082*** 0.2846*** 0.1524*(0.000) (0.000) (0.000) (0.097)

Hefindahl index −0.0516*** −0.0535** −0.3015*** −0.3739***(0.001) (0.028) (0.001) (0.000)

Family 0.0265*** 0.0320*** −0.004 −0.0357(0.010) (0.010) (0.900) (0.281)

CEO age −0.0008*** −0.0011*** 0.0009 −0.0012(0.001) (0.001) (0.296) (0.258)

Option Compensation 0.0679*** 0.0713*** 0.049 0.0127(0.000) (0.000) (0.107) (0.714)

Outside directors 0.0004*** 0.0003** 0.0007** 0.0015***(0.001) (0.025) (0.047) (0.000)

Total directors −0.0003 0.0001 0.0004 −0.0004(0.144) (0.994) (0.523) (0.622)

Year fixed effects Yes Yes Yes YesIndustry fixed effects Yes Yes Yes YesR2 0.349 0.362 0.580 0.151Observations 4514 3367 5183 5183

Coefficients reported here are from OLS estimation of the impact of antitakeover provisions on firm innovation for the period1990–2006. Patents are the number of patents divided by mean patents in the same year and technology class. Citations arethe number of citations received divided by mean citations in the same year and technology class. E-Index is the entrenchmentindex of Bebchuk et al. (2009). Size is natural log of sales. The dependent variables are patents and citations at period t + 1 andR&D Intensity at period t + 2. E-Index is the entrenchment index of Bebchuk et al. (2009). Patent stock is the cumulative numberof patents filed by a firm. R&D intensity is the ratio of R&D expenditure to sales. R&D missing is a binary variable that equals1 if R&D expenditures are missing and 0 otherwise. Coverage is interest coverage ratio- proxy for leverage. ROA is return onassets. Herfindahl index is based on sales in 2-digits SIC industries. Outside directors is the ratio of outside directors to totaldirectors. Total directors are the total number of directors. Option compensation is the ratio of Black and Scholes value of CEOstock options to total compensation. Robust p-values are reported in parentheses. *, ** and *** indicate significance at 10%, 5%and 1% respectively.

that exclude all firm-year observations that have missing R&D values. The results are given in thefirst two columns of Table 5. The coefficients on E-index are negative and significant indicating thatentrenchment results in lower investment in R&D. Thus measuring innovation in terms of investmentin R&D does not change the relation between entrenchment and innovation.

We also investigate whether our results change when we use a different lag period betweenentrenchment and firm innovation. We re-estimate our benchmark regressions with one year lagbetween entrenchment and innovation instead of a two-year lag. We find similar results (the lasttwo columns of Table 5). Entrenchment has negative and significant effect on firm performance ininnovation.

7. Conclusion

We explore the relation between managerial entrenchment and firm performance in innovation.Results indicate that managers that are protected from corporate takeover market via antitakeoverprovisions innovate less than managers that are not protected. Prior research has shown that

A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43 41

antitakeover provisions that increase managerial entrenchment are associated with lower firm value(Bebchuk et al., 2009). Our results provide an explanation of this relation – the same provisions thatreduce firm value are also associated with poor performance on innovation. Thus increased managerialentrenchment could lower firm value via its negative impact on innovation.

We also find that entrenchment has no significant effect on firm innovation in high-tech firms.One possible explanation is that high-tech firms need to continuously innovate for long run survivaland profitability. The disciplining effect of product market competition on innovation in high-techfirms seems to dissipate the negative managerial entrenchment effect. However, in low-tech firmsmanagerial entrenchment is more likely to exacerbate agency problems and negatively affect theincentives to innovate. Overall, our results support the agency based explanation of the relationbetween antitakeover provisions and managerial entrenchment.

Appendix A. Truncation bias in patent counts:

The truncation bias in patent counts exists because there is an average lag of two years between thetime an application is filed and the time the patent is granted. This is not a problem in the beginningof our sample (1990) as the average patent filed by the year 1990 was granted by the year 1992.However, as we approach toward the end of the sample, the data does not report the patents that willeventually be granted in the coming few years. We therefore observe large number of missing patentdata in the last years of the sample. Hall et al. (2002, 2005) show that correcting the truncation biasin patent counts is simple and straight forward. We follow Hall et al. (2005) to correct the truncationin patent counts. The method involves in calculating weight factors based on the application-grantdistribution of patents in the sample and then multiplying number of patent counts by the respectiveweight factors. The patent counts are corrected for the years 2001–2006. Based on Hall et al. (2005)the following formula is used to correct for the truncation in patent counts.

Patentt = patentt∑2005−t

k=0 weightk

2006 ≤ t≥2001

Truncation bias in citations:The truncation bias in citations exists because citations to a patent keep on coming long after the

patent was granted but we only observe citations in the data that are available by the last year of thedata. Hall et al. (2002, 2005) use the “quasi-structural” approach to correct the truncation bias in cita-tions. They first estimate the life time citation-lag distribution over the entire data period (1976–2006).Given this distribution they estimate the total citations of any patent for which only a portion of itscitations life is observed by dividing the observed citations by the fraction of the population distribu-tion for the time interval during which the citations are observed. A detailed description of the methodcan be found in Hall et al. (2002). We do not need to correct for the truncation bias in citations in oursample as the updated NBER data now provides truncation corrected data on citations using Hall et al.(2002).

Appendix B.

Variable Definition

Patents Number of patent divided by mean patents in the same year and technology classCitations Number of citations divided by mean citations in the same year and technology classE-index E-index is the sum of six provisions. The four provisions that limit shareholder voting

powers are: (1) staggered boards, (2) limits to shareholder amendments of the bylaws, (3)supermajority requirements for mergers, (4) supermajority requirement for charteramendments. The two anti-takeover provisions are poison pill and golden parachutearrangements. See Bebchuk et al. (2009).

Stock of patents Cumulative number of firm’s approved patents. See Hall et al. (2002) for methodology.R&D intensity Ratio of research and development expenditures to salesMissing R&D Binary variable equals 1 if R&D value is missingSize Natural log of sales

42 A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43

Variable Definition

Coverage Ratio of operating income to sum of taxes, interest and preferred dividendsROA Return on assets measured by the ratio of operating income to total assetsFamily Binary variable equals to 1 if the CEO belongs to a founding familyCEO age Age of the Chief Executive OfficerOption compensation Ratio of the value of CEO stock options to total compensationHerfindahl index Herfindahl index measures industry concentration, calculated as the sum of squared

market share based on 2-digit SIC code industries.Outside directors Ratio of outside directors to total directors on the board. Measures board independenceTotal directors Total number of directors on the board. Measures board size

References

Atanassov, J. (2013). Do hostile takeovers stifle innovation? Evidence from antitakeover legislation and corporate patenting.Journal of Finance, 68(3), 1097–1131.

Bebchuk, L., Cohen, A., & Ferrell, A. (2009). What matters in corporate governance? Review of Financial Studies, 22, 783–827.Becht, M., Bolton, P., & Röell, A. (2003). Corporate governance and control. In G. Constantinides, M. Harris, & R. Stultz (Eds.),

Handbook of Economics and Finance, Part I (pp. 1–109). Amsterdam: North Holland.Bhagat, S., & Welch, I. (1995). Corporate research and development investments: International comparisons. Journal of Accounting

and Economics, 19, 443–470.Blundell, R., Griffith, L., & Van Reenen, J. (1999). Market share, market value and innovation in a panel of British manufacturing

firms. Review of Economic Studies, 66, 529–554.Cockburn, I., & Henderson, R. (1996). Public–private interaction and the productivity of pharmaceutical research. Proceedings of

the National Academy of Sciences of the United States of America, 93(23), 12725–12730.Coles, J. L., Daniel, N. D., & Naveen, L. (2006). Managerial incentives and risk taking. Journal of Financial Economics, 79(2), 431–468.DeAngelo, H., & Rice, E. M. (1983). Antitakeover charter amendments and stockholder wealth. Journal of Financial Economics, 11,

329–359.Dechow, P., & Sloan, R. (1991). Executive incentives and the horizon problem: An empirical investigation. Journal of Accounting

& Economics, 14(1), 51–89.Eberhart, A., Maxwell, W., & Siddique, A. (2004). An examination of long-term abnormal stock returns and operating performance

following R&D increases. Journal of Finance, 59(2), 623–650.Gibbons, R., & Murphy, K. (1992). Optimal incentive contracts in the presence of career concerns: Theory and evidence. The

Journal of Political Economy, 100(June), 468–505.Gompers, P. A., Ishii, J. L., & Metrick, A. (2003). Corporate governance and equity prices. Quarterly Journal of Economics, 118(1),

107–155.Griliches, Z. (1992). The search for R&D spillovers. Scandinavian Journal of Economics, 94(0), S29–S47. SupplementGriliches, Z. (1984). R&D, patents, and productivity. In NBER conference proceedings Chicago: University of Chicago Press.Hall, B. H., Jaffe, A., & Trajtenberg, M. (2002). The NBER patent citation file: lessons, insights, and methodological tools. In A.

Jaffe, & M. Trajtenberg (Eds.), Patents, citations and innovations. Cambridge, MA: The MIT Press.Hall, B. H., Jaffe, A., & Trajtenberg, M. (2005). Market value and patent citations. Rand Journal of Economics, 36(1), 16–38.Holmstrom, B. (1989). Agency costs and innovation. Journal of Economic Behavior & Organization, 12(3), 305–327.Holthausen, R., Larcker, D., & Sloan, R. (1995). Business unit innovation and the structure of executive compensation. Journal of

Accounting and Economics, 19(March–May), 279–313.Jaffe, A. B., Trajtenberg, M., & Fogarty, M. S. (2000). The Meaning of Patent Citations: Report on the NBER/Case-Western Reserve

Survey of Patentees. NBER Working Papers 7631. National Bureau of Economic Research, Inc.Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal

of Financial Economics, 2, 305–360.Jensen, M. C., & Ruback, R. S. (1983). The market for corporate control: The scientific evidence. Journal of Financial Economics,

11, 5–50.Kam, W. P., Kiese, M., Singh, A., & Wong, F. (2003). The Pattern of Innovation in Singapore’s Manufacturing Sector. Singapore

Management Review, 25(1), 1–34.Kleinknecht, A., Van Montfort, K., & Brouwer, E. (2002). The non-trivial choice between innovation indicators. Economics of

Innovation and New Technology, 11(2), 109–121.Koeller, T. C. (1995). Innovation, market structure and firm size: A simultaneous equations model. Managerial and Decision

Economics, 16(3), 259–269.Lanjouw, J. O., & Schankerman, M. (2004). Patent quality and research productivity: Measuring innovation with multiple

indicators. Economic Journal, 114(495), 441–465.Lerner, J., & Wulf, J. (2007). Innovation and incentives: Evidence from corporate R&D. Review of Economics and Statistics, 89(4),

634–644.Lhuillery, S. (2006). The impact of corporate governance practices on R&D intensities of firms: An econometric study on French largest

companies. CDM Working Papers cemi-report-006.Mahoney, J. M., Sundramurthy, C., & Mahoney, J. T. (1997). The effects of corporate antitakeover provisions on long-term

investment: Empirical evidence. Managerial and Decision Economics, 18(5), 349–365.Mairesse, J., & Hall, B. H. (1996). Estimating the productivity of research and development: An exploration of GMM methods

using data on French & United States manufacturing firms. In van Ark, Bart, & Karin Wagner (Eds.), International productivitydifferences, measurement and explanations. Amsterdam: Elsevier Science.

Mansfield, E. (1985). How rapidly does industrial technology leak out? Journal of Industrial Economics, 34, 217–223.

A. Chakraborty et al. / Journal of Economics and Business 72 (2014) 30– 43 43

Masulis, R. W., Wang, C., & Xie, F. (2007). Corporate governance and acquirer returns. Journal of Finance, 62(4), 1851–1889.Scherer, F. M. (1982). Inter-industry technology flows and productivity growth. Review of Economics and Statistics, 64, 627–634.Schmookler, J. (1966). Invention and economic growth. Cambridge: Harvard University Press.Stein, J. C. (1988). Takeover threats and managerial myopia. Journal of Political Economy, 96(1), 61–80.Trajtenberg, M. (1990). A penny for your quotes: Patent citations and the value of innovations. Rand Journal of Economics, 21(1),

172–187.Wang, C., & Xie, F. (2009). Corporate governance transfer and synergistic gains from mergers and acquisitions. The Review of

Financial Studies, 22(2), 829–858.Xue, Y. (2007). Make or buy new technology: The role of CEO compensation contract in a firm’s route to innovation. Review of

Accounting Studies, 12(4), 659–690.