THE INFLUENCE OF ACQUISITION EXPERIENCE AND …

THE INFLUENCE OF ACQUISITION EXPERIENCE ANDPERFORMANCE ON ACQUISITION BEHAVIOR: EVIDENCE

FROM THE U.S. COMMERCIAL BANKING INDUSTRY

JERAYR (JOHN) HALEBLIANUniversity of California, Riverside

JI-YUB (JAY) KIMNANDINI RAJAGOPALAN

University of Southern California

We draw upon theories of organizational learning to examine acquisition likelihood ina sample of banking industry acquisitions from 1988 through 2001. Although researchon organizational learning suggests that routines arising both from experience andfrom performance feedback guide organizational learning, the combined effect of thesetwo sources of learning has rarely been examined. Findings are consistent with ourtheoretical predictions: (1) prior acquisition experience, (2) recent acquisition perfor-mance, and (3) the interaction between acquisition experience and recent acquisitionperformance are all positively related to the likelihood of subsequent acquisition.

Research on organizational learning suggests thatorganizational behavior is guided both by routinesthat stem from experience (Nelson & Winter, 1982)and by performance feedback (Greve, 2003). Rou-tines arise from the experience dedicated to a par-ticular task. Greater experience with a specific rou-tine provides opportunities to refine the routineand increases the probability of the routine beingused. Underlying this perspective is the assump-tion that the learning process is largely indepen-dent of the performance outcomes of prior experi-ences (Levitt & March, 1988; Nelson & Winter,1982). On the other hand, the view that learningstems from performance feedback, which isgrounded in the Carnegie school of thought (e.g.,Cyert & March, 1963), emphasizes the role of theperformance outcomes of prior actions and sug-gests that these outcomes determine a firm’s futurebehavior (Greve, 2003).

Recent empirical research on organizationallearning has significantly improved scholars’ un-derstanding of the distinct effects of these twosources of organizational learning (e.g., Baum, Li, &Usher, 2000; Greve, 1998; Haunschild & Beckman,1998; Ingram & Baum, 1997). However, although

some theoretical work in the learning tradition hasimplied that these two sources of learning worktogether to influence firm behavior (Levitt & March,1988), their combined effect has rarely been exam-ined in an empirical setting. Further, no theoreticalarguments have been proposed to guide research onhow experience and performance feedback interactto impact learning outcomes. Our study addressesthese important gaps in the organizational learningliterature by investigating how experience, perfor-mance feedback, and their interaction affect firmbehavior.

We examine these two forms of learning withinthe context of corporate acquisitions. A few acqui-sition studies have documented the effects of rou-tines and have shown that prior acquisition expe-rience affects subsequent acquisition behavior(Amburgey & Miner, 1992; Baum et al., 2000). How-ever, the influence of prior acquisition performanceon a firm’s future acquisition behavior is yet to beexplored, and so understanding of whether firmsmodify their acquisition behaviors on the basis ofperformance feedback is limited. In addition, re-searchers do not know whether these two sourcesof influence—routines and feedback—interact toexplain firm acquisition behaviors. To addressthese gaps in the acquisition literature, we distin-guish here between the effects of acquisition expe-rience and performance feedback and explicate thetheoretical bases for their interaction in an attemptto understand the variance in acquisition behavior.We tested our research questions using longitudi-nal data (from the period 1988–2001) in a sample of

The authors contributed equally to this work. Wewould like to thank Christine Beckman, Jonathan Jaffee,Livia Markoczy, Anne Miner, David Mayers, Gerry Mc-Namara, and Kathleen Montgomery for their useful com-ments on earlier versions. We would also like to ac-knowledge Associate Editor Donald Bergh and threeanonymous AMJ reviewers for their valuable suggestions.

� Academy of Management Journal2006, Vol. 49, No. 2, 357–370.

357

acquisitions from the U.S. commercial bankingindustry.

THEORY AND HYPOTHESES

The Effect of Acquisition Experience onAcquisition Behavior

A central concept in behavioral learning theory isthe organizational routine that arises from organi-zational experience (Levitt & March, 1988). Rou-tines are programs of action that reflect the priorexperience of an organization with a particular task(Nelson & Winter, 1982). Once routines are estab-lished through experience, they are subject to iner-tial pressures (e.g., Szulanski, 1996). The organiza-tional literature offers ample evidence that themore experienced an organization’s members be-come with a particular strategic action or direction,the more likely they are to repeat it (e.g., Amburgey,Kelly, & Barnett, 1993; Gulati, 1995; Shaver, Mitch-ell, & Yeung, 1997). Routines can become a sourceof competitive advantage and often play a crucialrole in the formulation of a firm’s strategic choicesby supplementing, or even substituting for, calcu-lative, formal strategic decision-making rules(March, 1999). As a firm accumulates experience ina certain organizational routine, it may gain com-petence and expertise in that routine. However,even as these acquired capabilities make the firmproficient in performing a particular task, they mayalso lead it to overestimate its chances of futuresuccess in that domain and to focus on exploitingexisting capabilities instead of exploring new ones(March, 1991).

Persistence appears to be a pervasive force inorganizations. Miller and Friesen (1980, 1982) ar-gued that past practices and strategies tend to keepevolving in the same direction, perhaps eventuallyreaching dysfunctional extremes, a phenomenonthey described as strategic momentum. Kelley andAmburgey (1991), in their study of the airline in-dustry, found that firms that experienced strategicchanges of a particular type were more likely tocontinue making changes of the same type. Theyargued that although prior models had focused onthe effects of organizational size, age, and complex-ity on organizational inertia (Hannan & Freeman,1984), a more complete understanding of persis-tence would require an examination of an organi-zation’s history of experience. In the context ofacquisitions, Amburgey and Miner (1992) reportedresults consistent with the strategic momentum ar-gument. Specifically, they reported that acquirersthat made acquisitions of a particular type weremore likely to subsequently make the same type of

acquisition. Similarly, Baum and his coauthors(2000) found that Canadian nursing home chainstended to acquire targets similar to those of theirprior acquisitions, providing evidence that priorexperience is an important predictor of future ac-quisition behaviors.

Research in organizational learning implies thatonce a firm has accumulated experience in a stra-tegic action, not only favorable outcomes, but alsounfavorable consequences of the action will in-crease the likelihood that the action will be re-peated (Amburgey & Miner, 1992; March, 1981).Decision makers may not interpret poor perfor-mance as evidence that their strategies were inap-propriate or fundamentally flawed, but rather mayattribute it to poor execution of otherwise healthystrategies or to factors external to the strategies. Insummary, acquisition experience will lead to thedevelopment of routines associated with makingacquisitions, such as templates for selecting andevaluating targets or guidelines for postacquisitionintegration. That is, an acquisition routine, once inplace, may gain a life of its own and persist regard-less of the performance outcomes of prior acquisi-tions (Levitt & March, 1988), consequently increas-ing the likelihood of the acquirer making anotheracquisition. In view of these arguments, we pro-pose the following hypothesis:

Hypothesis 1. The greater an acquirer’s acqui-sition experience, the higher the likelihood ofthe acquirer making a subsequent acquisition.

The Effect of Recent Acquisition Performance onAcquisition Behavior

The behavioral theory of the firm provides a use-ful framework for understanding the effects of prioracquisition performance on future acquisition be-havior. Surprisingly, this perspective, according towhich firms exhibit adaptive behavior over time(Cyert & March, 1963: 123), has not yet been ap-plied in the acquisition context. More specifically,firms are thought to demonstrate differing re-sponses to good and bad performance outcomes(March, 1981): strong performance increases afirm’s likelihood of persisting in prior strategic ac-tions (Miller & Chen, 1994), while poor perfor-mance increases the likelihood of strategic changeand promotes exploration of new strategies(Boeker, 1989). From this viewpoint, it is insuffi-cient to examine only the effects of the existence ofprior acquisition experience, because the perfor-mance feedback from prior acquisitions should alsohave an impact on future acquisition behaviors(Levitt & March, 1988).

358 AprilAcademy of Management Journal

Prior work on corporate acquisition has generallytreated acquisition performance as an outcomevariable (King, Dalton, Daily, & Covin, 2003) andhas examined the effects of firm, deal, and industrycharacteristics on acquisition outcomes (e.g.,Bergh, 1997; Bergh & Lawless, 1998; Capron &Pistre, 2002; Chatterjee, Harrison, & Bergh, 2003;Ramaswamy, 1997). However, in this study, weconceptualized acquisition performance as an an-tecedent rather than an outcome variable andstudied the influence of prior acquisition perfor-mance on future acquisition behavior.

In contrast to a learning perspective that empha-sizes the development and execution of routines, aperformance feedback approach to learning focusesmore on the outcomes of prior behaviors and theirinfluence on future behaviors. Specifically, poorperformance can drive managers toward strategicchange. Managers who detect poor performancewill try to reevaluate their current strategies toidentify what went wrong, and they will engage in“problemistic” search in an attempt to find an al-ternative strategy that might improve performance.In contrast, given initial success with an activity,an organization is likely to repeat the activity be-cause it has learned skills and capabilities associ-ated with it and finds it less risky and more reward-ing to repeat it than to try alternatives with whichorganizational experience is limited (Levitt &March, 1988). The successful execution of a strate-gic action is a source of self-assurance that makesfirms become more confident that they have thecapabilities and knowledge required to be success-ful in a specific strategic domain. Thus, strong per-formance decreases the intensity of search and ex-perimentation and promotes the persistentexploitation of strategies that have proven success-ful (Greve, 2003).

In keeping with the behavioral theory of the firm,empirical studies in the strategic change literaturehave shown that firms do react to past performancefeedback and that they are more likely to respondwith strategic change when firm performance ispoor (Rajagopalan & Spreitzer, 1997). In contrast,strong performance leads to organizational persis-tence by inducing firms to believe they have gottenit right—that is, it makes organizations resistant tochange in their strategies and routines (Miller &Friesen, 1984; Tushman & Romanelli, 1985). Insupport of these behavioral theory assertions, vari-ous firm contexts have demonstrated that strongperformance is associated with organizational per-sistence (e.g., Audia, Locke, & Smith, 2000; Greve,1998; Miller & Chen, 1994), but poor performanceis related to many forms of organizational change,including CEO turnover (e.g., Denis & Denis, 1995;

Puffer & Weintrop, 1991), strategic change (e.g.,Boeker, 1989), and divestitures (e.g., Duhaime &Grant, 1984; Hamilton & Chow, 1993; Kaplan &Wiesbach, 1992; Ravenscraft & Scherer, 1987; Tay-lor, 1988).

We propose that the theoretical arguments andevidence on performance feedback developed inthe broader strategic change literature are also ap-plicable to the acquisition context. Firms that havedone well with recent acquisitions interpret this aspositive feedback and an endorsement of the appro-priateness of their strategic choice. This reinforce-ment, in turn, increases the likelihood that theywill persist and repeat the same behavior—that is,make more acquisitions. In contrast, firms that re-ceived negative feedback, as manifested in poorperformance following prior acquisitions, are lesslikely to choose acquisition as a future strategicchoice.

Learning from performance feedback requires ameaningful evaluation of the performance of prioracquisitions. Such an evaluation is a difficult taskbecause varying performance levels of prior acqui-sitions may provide mixed signals to managers ofan acquiring firm. Acquisitions are complex eventsin which the causal relationship between perfor-mance and the factors contributing to that perfor-mance is often ambiguous. Hence, assessing thepattern of performance of multiple prior acquisi-tions presents a cognitive challenge to managers.Behavioral researchers have suggested that undersuch conditions decision makers depend heavilyupon the most recent information to reduce theircognitive burdens and simplify information pro-cessing (Hogarth & Einhorn, 1992; Steiner & Rain,1989). The reliance on the most recent performancefeedback may increase when the changes in thebusiness environment diminish the validity of in-ferences drawn from past performance feedback(Argote, 1999; March, 1999).

Taking the above arguments together, we proposethat performance feedback from a firm’s most re-cent acquisition (referred to as the focal acquisi-tion) will influence the likelihood of the firm’smaking a subsequent acquisition:

Hypothesis 2. The stronger the performance ofan acquirer’s focal acquisition, the higher thelikelihood of the acquirer making a subsequentacquisition.

The Effect of the Interaction between AcquisitionExperience and Performance on AcquisitionBehavior

In our first two research hypotheses we identi-fied two distinct mechanisms—routines and per-

2006 359Haleblian, Kim, and Rajagopalan

formance feedback—that may independently af-fect the future acquisition behavior of firms. Inthis section, we propose that these two distinctsources of learning may interact to produce ef-fects that explain further variance in acquisitionbehavior.

Although greater acquisition experience pro-vides opportunities to refine existing routines asso-ciated with making acquisitions, these routinesmay be challenged when performance feedback sig-nals decision makers that it may be undesirable topersist with current routines (Baum et al., 2000).Poor acquisition performance may eventually leadto the decay of an acquisition routine, suggestingthat performance feedback can either reinforce orweaken the effects of routines in driving subse-quent acquisition behaviors. Hence, we argue thatthe positive effect of acquisition experience on thelikelihood of making future acquisitions will beeven more pronounced when it is accompanied bystronger acquisition performance.

The combined effect of routines and performancefeedback is also consistent with the theoretical ideaof a competency multiplier. Specifically, successwith a particular routine reinforces the experientiallessons learned from using the routine, which in-creases the likelihood of further adoption of theroutine (March, 1981). In the acquisition context, afirm may develop an acquisition routine as it accu-mulates experience. Greater acquisition experiencemay make the firm a better acquirer by providingopportunities to refine its existing routines and de-velop more competencies in acquisition, encourag-ing it to further exploit those routines in the future.For several reasons, the effects of routines will bereinforced when acquirers receive strong perfor-mance feedback from their recent acquisitions.First, the positive performance feedback obtainedfrom recent acquisitions signifies the effectivenessof their existing routines. Second, it validates theirassumption that they have developed competen-cies required to successfully execute routines asso-ciated with acquisitions. Finally, it elevates theconfidence of decision makers in making furtheracquisitions.

By contrast, poor acquisition performance maychallenge the appropriateness of an existing acqui-sition-related routine and signal managers thattheir current routine may require major modifica-tions to be effective, which can lead them to reviewtheir acquisition programs. It may also erode thelegitimacy of the acquisition routine as a validstrategic choice and discount the perceived effec-tiveness of experiential lessons embedded in it.Thus, performance feedback will moderate theeffects of routines in such a way that strong per-

formance reinforces the effects of routine-basedpersistence and poor performance dampens theeffects of routine-based persistence. Taken to-gether, these arguments lead to our (final) inter-action hypothesis:

Hypothesis 3. The positive effect of acquisitionexperience on the likelihood of an acquirer’smaking a subsequent acquisition will be stron-ger at higher levels of focal acquisitionperformance.

METHODS

Research Setting

We investigated our research questions in thecontext of the U.S. commercial banking industryduring the period between January 1, 1988, andDecember 31, 2001. We collected data on all of theacquisitions made by all publicly traded banks andbank holding companies that reported to the Fed-eral Reserve Board (FRB) during this period, but welimited our sample to the whole-bank acquisitionsthat did not receive government assistance.1 Wholebank acquisitions were defined as those in whichthe acquiring bank assumed complete ownership ofthe target bank. We excluded purchases of only thepartial assets of other banks, such as individualbranches or loans, because such transactions do notresult in a change in the ownership of the targetbanks, and they generally represent a routine man-agement practice rather than an important strategicevent that may have a significant impact upon fu-ture firm behavior and market valuation. Amongthe whole bank acquisitions that were announcedduring the study period, 232 deals were terminatedprematurely without being completed and thuswere dropped from the sample. The final sampleused in the study tracked 2,523 completed, wholebank acquisitions made by 579 publicly tradedbanks and bank holding companies.2

1 Government-assisted acquisitions refer to acquisi-tions in which a failing bank is merged with another bankwith financial assistance from a regulatory agency suchas the Federal Deposit Insurance Corporation (FDIC).Such acquisitions were excluded from the sample be-cause they do not represent voluntary strategic actions.We also limited our sample to stand-alone independentbanks; no sell-offs were included in our sample.

2 Each of the 579 banks entered the sample at a differ-ent time, depending upon when it made its first acquisi-tion after the beginning of the study period (January 1,1988). Each bank could also have exited from the samplebefore the end of the study period (December 31, 2001) ifit failed or was acquired by another bank. Hence, the total


Data on acquisitions were collected from theM&A module of the SNL Financial database. Dataon acquisition performance were measured withabnormal returns around the announcement dateand were calculated from data drawn from the Cen-ter for Research in Securities Pricing (CRSP). Fi-nally, demographic and financial data on banks inthe sample were obtained from the regulatory bank-ing data maintained by the FRB.

The empirical models built on these large-scalearchival data could provide systematic evidence onthe hypothesized causal relationships as well asfindings with greater generalizability, but we dis-cerned that they might not be able to fully capturethe fine-grained, intermediate processes that con-tribute to the relationships and other context-spe-cific boundary conditions. We therefore conductedcontinuous, iterative exploratory qualitative datacollection to inform and systematize our theoriesand empirical models (Eisenhardt, 1989; Strauss &Corbin, 1998). We conducted a series of exploratorysemistructured interviews with bank managers andexecutives at eight major commercial banking in-stitutions that had made at least one acquisitionduring our study period and six industry expertswho worked for major government agencies. Be-cause our study was deductive, these qualitativeinsights were not used to develop theories. Rather,we used these insights to (1) close potential gapsbetween our theories and empirical models, (2)check the validity of assumptions embedded in ourempirical models, (3) incorporate industry-specificboundary conditions or shared assumptions intoour study, and (4) help interpret our findings.

Our empirical setting was particularly appropri-ate for exploring our research questions for severalreasons. First, because our sample was drawn froma single industry, we were able to control for manyindustry-specific, exogenous factors (e.g., socioeco-nomic and technological conditions). Second, theseries of regulatory changes implemented duringthe 1980s altered the competitive landscape in theU.S. commercial banking industry. In particular,the Garns–St. Germain Deposit Institutions Act of1982 (GDIA) and the Competitive Equality BankingAct of 1987 (CEBA) reshaped the acquisition land-scape by relaxing the regulations that limited inter-state bank acquisitions. These regulatory changesnot only opened the door for interstate acquisitionsand promoted significant changes in acquisition

practices but also resulted in an explosion of acqui-sitions in the banking industry. This naturallyoccurring experimental setting was particularlyvaluable to our study design because such dis-continuity can alleviate potential biases thatcould be introduced by studying only a partialhistory of an industry.

Independent Variables

Acquisition experience. The benefits of prior ex-perience may not increase monotonically with theamount of experience that an organization accumu-lates because old experience becomes less usefulover time (Argote, 1999; Barnett, Greve, & Park,1994; Hayward, 2002). The regulatory changes ofthe 1980s changed how acquisitions were under-stood and performed in the banking industry. Ourqualitative data suggested that the regulatorychanges forced banks to review their existing ac-quisition strategies, to modify their templates formaking acquisitions, and to search for new prac-tices. Because those regulatory changes marked sig-nificant shifts in the industry, a bank’s acquisitionexperience before the changes might not representviable resources for future learning. Such experi-ence might even have led banks to adopt outdatedroutines and harm their performance by encourag-ing them to replicate strategies that worked wellunder different circumstances (Barney & Hesterley,1996). Following prior studies that have examinedacquisition experience (Haunschild, 1993; Hay-ward, 2002), we measured acquisition experienceas the total number of prior acquisitions that aninstitution in our sample made between 1988, theyear following the last major regulatory change(CEBA) that had significant impact upon bank ac-quisition strategies, and the year of a focalacquisition.

Focal acquisition performance. The short-termimportance of an event like an acquisition can beassessed by the price change in the acquirer’s secu-rity during a period surrounding the event; thisprice change is calculated as the difference be-tween the observed and the predicted, or normal,return for the same security. Hence, the short-termimpact of an event is measured by the part of thereturn that is unanticipated by an economicmodel of anticipated, normal returns. We mea-sured focal acquisition performance using abnor-mal returns.

To do this, we used an improved measure ofabnormal returns developed specifically for firmsin the banking industry. Rather than comparing thereturn for a security of a firm in our sample with the

number of spells in the data set (6,714) is smaller than thenumber of spells that would have been obtained if allbanks had entered and exited the sample at the sametime.


return on a general market index, we observedthe return for a given bank’s security and pre-dicted the normal return for the same security byassessing the performance of only other firms inbanking.3 This refinement allowed for an indus-try-specific abnormal return, which eliminatedextraneous noise that might have been brought infrom firms outside the industry. Specifically, weassessed returns of a security against the return ofa banking industry portfolio, using the followingformula:

E � Rit � (�i � �iRmt),

where E is the event, Rit is the return on stock i forday t; Rmt is the return on the banking industryportfolio for day t; �i is a constant; and �i is the betaof stock i.4

To determine the influence of an event on a firm,we averaged abnormal returns over an event win-dow beginning 5 trading days before and ending 15trading days after the announcement of an acquisi-tion event (–5, �15); a window of this duration iscommonly used for measuring the performance ofbank acquisitions (Hannan & Wolken, 1989;Rhoades, 1994). The mean cumulative abnormalreturn (CAR) was –0.01, and of the 2,523 acquisi-tions in the sample, 1,410 (56%) had negativeCARs, and 1,113 (44%) had positive CARs. Ourqualitative study also suggested that a medium-term event window (� 10 days) was preferable to ashorter event window in the banking industry be-cause the market reaction to an acquisition an-nouncement might not be initiated until regulatoryconcerns associated with the announcement werecleared.5

Ex ante measures of acquirer abnormal returnshave been found to be correlated with ex post mea-sures of acquisition performance, demonstratingthat event study methodology has predictive valid-ity. In the finance literature, Healy, Palepu, andRuback (1992) found a strong, positive relationship

between abnormal stock returns at merger an-nouncements and postmerger increases in operat-ing cash flows. Kaplan and Weisbach (1992) alsofound that unsuccessful divestitures, as comparedwith successful ones, were associated with signifi-cantly lower acquirer returns at an acquisition an-nouncement, suggesting that markets may reason-ably forecast subsequent acquisition performance.In the strategy literature, Sirower (1997) found thatacquirer returns at the time of an acquisition an-nouncement were representative of long-term per-formance. Thus, extant evidence on event studymethodology’s predictive validity is consistentwith our assumption that abnormal returns arevalid indicators of acquisition performance.

Although other strategic management studieshave used accounting-based measures of acquisi-tion performance such as return on assets (e.g.,Ramaswamy, 1997), there were several reasons tobelieve that CAR was a more appropriate perfor-mance measure for our sample of aggressive acquir-ers in the banking industry during the study period(1988–2001). First, the effects of an acquisition arenot immediately reflected in the financial state-ments of an acquirer because it usually takes sixmonths to three years before the acquirer realizesthe effects (Rhoades, 1994). During this period,many confounding factors, such as changes inproduct mix, investment strategy, and manage-ment, as well as execution of additional acquisi-tions, may affect firm performance. Hence, al-though event study methodology allowed us todistinguish among the performance effects of indi-vidual acquisitions made in close temporal prox-imity, accounting data would not have done so.Second, firms often acquire other firms for nonfi-nancial reasons. For example, banks increasinglyengage in interstate acquisitions to expand theirgeographic scope. Such acquisitions often under-mine acquirers’ financial performance by increas-ing operating costs, and thus may be consideredfailures from an accounting standpoint, but theyare not failures if the acquirers achieve their stra-tegic objectives. Finally, performance measuresbased on accounting measures could have beenmisleading because accounting procedures are notuniform across firms (Bentson, 1985). In keepingwith these arguments, our systematic review of theempirical acquisition literature indicated thatevent study methodology (and, hence, cumulativeabnormal returns) has been the most frequentlyused analytical approach for measuring acquisitionperformance (e.g., Capron & Pistre, 2002; Finkel-stein & Haleblian, 2002; King et al., 2003; Rhoades,1994).

3 We also tested our models using a traditional marketindex abnormal return measure (e.g., Fama & French,1992) to examine sensitivity, and we obtained consistentresults.

4 It is assumed that � and � are stable and are calcu-lated during an arbitrary estimation period beginning 270trading days before an announcement date and ending 20trading days before the announcement date.

5 We also tested a shorter event window around theacquisition date (0, �2), a commonly used period in thefinance literature (e.g., Fama & French, 1992), to examinethe sensitivity of our findings, and we obtained consis-tent results.


Control Variables

Because we studied horizontal acquisitionswithin a single industry, our sampling strategy al-lowed us to control for both relatedness and indus-try effects. We also incorporated a series of controlvariables to rule out confounding factors related toan acquirer’s propensity to make future acquisi-tions. All the control variables except for deal char-acteristics were time-varying covariates that corre-sponded to each time period in which acquirerswere at risk of making an acquisition. Because dealcharacteristics reflected the characteristics of eachfocal acquisition, they were measured at the time ofthe focal acquisition.

Acquirer characteristics. We included four con-trol variables designed to capture acquirer charac-teristics that might influence an acquirer’s propen-sity to make acquisitions. Firm size tends toinfluence strategic choices in such a way that largerfirms are subject to stronger inertial forces and thusare more likely to repeat their prior actions (Han-nan & Freeman, 1989). Thus, we controlled foracquirer size, which was measured as the logarithmof an acquirer’s total assets. Prior research has sug-gested that a firm’s slack resources promote risktaking within the firm by encouraging managers tomake potentially risky strategic moves such as ac-quisitions (Lang, Stulz, & Walkling, 1991), whichimplies that organizational slack may be positivelyrelated to the propensity to make acquisitions. Ac-cordingly, we controlled for acquirer slack re-sources, which was measured as the ratio of anacquirer’s cash on hand that was not being used foroperations to the total assets of the acquirer. Wealso controlled for acquirer firm performance, mea-suring it as the acquirer’s return on assets, becausestrong financial performance could encourage man-agers to pursue an aggressive acquisition strategy.Finally, we controlled for acquirer financing capa-bility, measured as total interest expenses dividedby average interest-bearing liabilities,6 becausebanks with a better financing capability are morelikely to make acquisitions than those with a lowerfinancing capability.

Deal characteristics. We used a set of three vari-ables to control for the characteristics of each focalacquisition that might affect future acquisition like-lihood. Prior studies have provided ample evi-dence that the relative sizes of a target and anacquirer affect acquisition outcomes (e.g., Asquith,

Bruner, & Mullins, 1983). This ratio, called therelative size of a focal acquisition, may also influ-ence the acquirer’s future acquisition likelihood byaltering its perception of the importance of the ac-quisition and its interpretation of the performanceand the experience associated with the focal acqui-sition. For example, because a larger acquisitionmay be perceived as a more important event than asmaller acquisition, successful completion of alarge acquisition may raise the acquirer’s confi-dence more than a similar outcome obtained from asmaller acquisition—consequently, further promot-ing future acquisitions. Accordingly, we controlledfor relative acquisition size, which was measuredas the ratio of the target’s total assets to the acquir-er’s total assets in a focal acquisition. The type ofconsideration offered in an acquisition has beenfound to influence acquisition outcomes (Datta,Narayanan, & Pinches, 1992). For example, postac-quisition returns to an acquirer may decrease withthe fraction of the premium paid in the acquirer’sstock because stock offers may send a signal that anacquiring firm’s management feels its stock is over-valued. Thus, the type of consideration implicitlyindicates the acquirer’s performance in the finan-cial market, which may influence its propensity formaking future acquisitions. We controlled for stockconsideration, which was measured as the percent-age of the acquirer’s common stock paid for theequity of the target in a focal acquisition. The pres-ence of a lock-up agreement, a legally binding con-tract that prohibits insiders from a firm from sellingany share of stock for a specified period of time,may influence firms to make more acquisitions be-cause the insiders may seek an alternative way touse the locked-up stocks. Lock-up agreement wascoded 1 for a focal acquisition that was completedwith a lock-up agreement, and 0 otherwise.

Industry conditions. The industry conditions atthe time of a focal acquisition might affect acquisi-tion behaviors and outcomes (Bergh, 1997), in thatthey could determine the attractiveness of acquisi-tions as strategic choices. Hence, as have the au-thors of prior work (e.g., Greve & Taylor, 2000;Haveman, 1992), we included three industry fac-tors that could affect our dependent variable. In-dustry revenue, defined as the total amount of rev-enue of all banks during each year, was included tocontrol for the effects of overall industry revenuechanges over time. The intensity of acquisition ac-tivities in an industry is likely to affect the acqui-sition behavior of individual firms in the industrybecause firms frequently attempt to replicate pop-ular strategic actions (Haunschild & Miner, 1997).Industry-level acquisition activities also have sig-nificant implications for the availability of viable

6 Interest-bearing liabilities were calculated as the sumof deposits, federal funds purchased and repossessed,commercial paper, mortgage debt, subdebt, and otherborrowed money.


targets within the market in which a focal bankoperates. We controlled for the level of industry-level acquisition activities with industry acquisi-tion density and industry total acquisition value,which were respectively measured as the totalnumber of acquisitions and the sum of the totaldeal value of all the acquisitions in the market areain which a focal acquisition was made.7

Prior acquisition experience. Banks’ acquisitionexperience prior to our study period might havesome impact on the knowledge of institutions inour sample, although the series of regulatorychanges that occurred during the 1980s could havemade a substantial portion of such knowledge ob-solete. Beginning in 1978, some state legislaturesbegan to enact laws that allowed out-of-state bankholding companies to control banks within theirstates and thus opened the door for interstate bankacquisitions (Kane, 1996). We collected prior ac-quisition experience data from this point forwardbecause the erosion of geographic restriction inbank acquisition had substantially changed themeaning of acquisition experience in the industry.Hence, we controlled for prior acquisition experi-ence, 1978–87, measured as the total number ofacquisitions that each institution in the samplemade from 1978 to 1987.

Analysis

Our dependent variable, the likelihood of makinga subsequent acquisition, was estimated as the haz-ard of a bank in the sample making a subsequentacquisition after a focal acquisition. A crucial issuein modeling the hazard rate of organizationalevents is selecting an appropriate functional spec-ification for the duration dependence of an event’soccurrence. Our theoretical arguments requiredmodels that allowed for the effects of acquisitionexperience and focal acquisition performance tovary with the characteristics of firms that were atrisk of making subsequent acquisitions, because thetime-varying nature of firm-specific characteristicsmay affect the interpretation of acquisition experi-ence and performance. Hence, we used the piece-wise exponential model, which splits the time axisinto predefined time segments and permits the haz-ard rate to vary in an unconstrained way acrosstime periods (Blossfeld & Rohwer, 1995). This ap-proach provides a flexible estimation of duration

dependence and offers a clear understanding ofhow duration affects acquisition patterns.

Our qualitative interviews with bank executivesindicated that serial acquirers often make manyacquisitions within a three-to-four-year time framethat is followed by an extended period with noacquisition activity. Hence, we divided durationinto four-year time periods. Given k time periods(Il � �t��l � t � �l � 1�, l � 1. . . k), the hazard rate ofthe piecewise exponential model we estimated canbe represented as:

r(t) � exp(�l � ��) if t � Il,

where �l is a constant coefficient associated withthe lth time period, � is a row vector of covariates,and � is an associated vector of coefficients that donot vary across time periods.

The piecewise exponential model allowed fordirect comparability to the Cox proportional haz-ards model, since a substantial difference in theresults obtained from these two methods could rep-resent evidence of an incorrectly parameterized un-derlying baseline model (Yamaguchi, 1991). Whenwe compared our results from the piecewise modelto estimates obtained from other functional speci-fications (including the exponential, Gompertz,and Weibull distributions), we obtained consistentresults, and the piecewise exponential model pro-vided the best fit, which supported the validity ofour parametric assumptions.8

The survival analysis we used has several advan-tages over alternative statistical methods, such as arandom-effects logistic regression. Because it usedthe information provided by “right-censored” cases(Allison, 1984; Tuma & Hannan, 1984), it did notrequire an ad hoc time window to determine whetheran acquirer made a subsequent acquisition after afocal acquisition. In addition, this modeling strategyprovided the most complete picture of acquisitionlikelihood by taking into account the information ontime elapsed since a prior acquisition.9

7 Our definition of market area was based on the geo-graphic market area classification of the Federal ReserveBank.

8 Because of space limitations, these comparisons arenot included in this article but are available from theauthors.

9 In addition to our event history analysis, we alsoestimated random-effects logistic regression models us-ing a dependent variable that measured the likelihood ofa subsequent acquisition within a two-year window aftera focal acquisition for both a full-sample analysis andsubsamples with only large acquisitions (10%, 20%, and30% of target/acquirer equity). Results remained consis-tent across these analyses.


RESULTS

Table 1 provides the descriptive statistics andcorrelations for the variables used in the study.

Table 2 reports maximum-likelihood estimatesfrom the piecewise exponential models for theanalysis of subsequent acquisition likelihood.Model 1 included only the control variables. Thethree predictor variables (acquisition experience,focal acquisition performance, and their interac-tion) were added hierarchically in models 2, 3, and4, respectively. To create the interaction term, wemean-centered the predictor variables to addressthe potential problem of multicollinearity (Aiken &West, 1991). The log-likelihood ratio tests showedthat the addition of each predictor variable signifi-cantly improved the overall model fit.

The coefficients for acquisition experience werepositive and significant (p �.01) in all models, sup-porting Hypothesis 1, which states that acquisitionexperience is positively related to future acquisi-tion likelihood. As Hypothesis 2 states, the coeffi-cients for focal acquisition performance were pos-itive and significant (p �.01) in models 3 and 4.These estimates indicated that focal acquisitionperformance was positively related to future acqui-sition likelihood. Finally, Hypothesis 3 states thatthe interaction between acquisition experience andfocal acquisition performance is significantly re-lated to future acquisition likelihood. As shown inmodel 4, the coefficient for the interaction term waspositive and significant (p � .01). This result sug-gested that the positive effect of experience on thelikelihood of making future acquisitions wasstrengthened at higher levels of a most recent acqui-sition’s performance, thus supporting Hypothesis 3.

To visually depict the interactive effect of thepredictors on subsequent acquisition rates, we cre-ated a three-dimensional graph that illustrates themultiplier effects of acquisition experience relativeto the levels of acquisition performance. Figure 1presents this graph.10 The vertical axis of the figureis the multiplier of the rate, which is the combinedinfluence of acquisition experience and acquisitionperformance on the hazard rate of an acquirer’smaking subsequent acquisitions, as determined byother covariates. The horizontal axes represent ac-quisition experience and acquisition performance,respectively, within the data range one standarddeviation below (low acquisition experience di-vided by performance) and above (high acquisitionexperience divided by performance) the mean ofeach predictor variable (Aiken & West, 1991; Cohen& Cohen, 1983). Supporting Hypothesis 3 and ourfindings from model 4, the plots showed that thepositive effect of acquisition experience on subse-quent acquisition rates was higher at higher levelsof focal acquisition performance.

10 The multiplier of the rate represents the risk ofexperiencing events relative to the baseline hazard. Theexponential hazard rate is represented as r(t) �h0(t)exp(�x), where h0(t) is the baseline hazard rate and�x is the regression coefficient. Because h0(t) is identicalfor all variables in the model, the relative effects betweenpredictors can be assessed by comparing the multiplier ofthe rate, exp(�x), after setting the multiplier to 1 for thebaseline hazard rate.

TABLE 1Descriptive Statistics and Correlations for Key Study Variablesa

Variableb Mean s.d. 1 2 3 4 5 6 7 8 9 10 11 12

1. Acquirer size 14.73 1.702. Acquirer slack resources 0.23 0.76 �.283. Acquirer firm performance 1.09 0.48 .13 .004. Acquirer financing capability 4.66 0.98 .08 .01 �.205. Relative acquisition size 0.17 0.25 �.33 .20 �.06 .046. Stock consideration 0.65 0.45 .16 �.06 .12 �.01 .027. Lock-up agreement 0.25 0.43 .13 �.09 .04 �.04 .20 .238. Industry revenue 2.84 6.28 �.13 .05 .12 �.05 .11 .05 .239. Industry acquisition density 88.44 47.18 .08 �.05 .14 �.10 �.14 .09 �.14 �.12

10. Industry total acquisition value 14.04 20.43 �.02 �.01 .09 .01 .06 .04 .13 .44 .0311. Prior acquisition experience,

1978–870.72 1.60 .21 �.06 .06 �.00 �.16 .07 �.04 �.11 .17 �.06

12. Acquisition experience, since1988

5.15 9.96 .49 �.10 .18 �.00 �.20 .16 .03 .11 .22 .07 .12

13. Focal acquisition performance �0.01 0.07 �.11 .01 �.01 .03 �.06 �.11 �.14 �.02 .01 �.01 �.03 �.03

a n � 6,714. Correlations greater than .03 are significant at p � .05, and correlations greater than .05 are significant at p � .01.b Industry revenue is expressed in trillions of U.S. dollars, and acquisition value is in billions.


TABLE 2Maximum-Likelihood Estimates of Subsequent Acquisitionsa

Variables Model 1 Model 2 Model 3 Model 4

ControlsAcquirer size 0.21** (0.02) 0.09** (0.02) 0.10** (0.02) 0.10** (0.02)Acquirer slack resources �0.37** (0.09) �0.35** (0.08) 0.35** (0.08) �0.35** (0.08)Acquirer firm performance 0.62** (0.06) 0.49** (0.06) 0.48** (0.06) 0.48** (0.06)Acquirer financing capability 0.01 (0.02) 0.01 (0.02) 0.00 (0.02) �0.01 (0.02)Relative acquisition size �1.07** (0.14) �0.93** (0.14) �0.89** (0.14) �0.88** (0.14)Stock consideration 0.45** (0.06) 0.37** (0.06) 0.38** (0.06) 0.38** (0.06)Lock-up agreement 0.12* (0.06) 0.26** (0.06) 0.27** (0.06) 0.27** (0.06)Industry revenue �9.14** (0.62) �1.11** (0.63) �1.11** (0.63) �1.08** (0.63)Industry acquisition density 0.05** (0.00) 0.03** (0.00) 0.03** (0.00) 0.03** (0.00)Industry total acquisition value �3.88 (1.74) �2.93 (1.61) �3.11 (1.61) �4.79** (1.70)Prior acquisition experience, 1978–87 �0.01 (0.01) 0.01 (0.01) 0.01 (0.01) 0.01 (0.01)

Duration since previous acquisition0–4 years �8.54** (0.34) �6.12** (0.36) �6.23** (0.36) �6.26** (0.36)5–8 years �8.29** (0.36) �5.91** (0.38) �6.01** (0.38) �6.05** (0.38)9–12 years �8.22** (0.40) �6.10** (0.41) �6.20** (0.41) �6.25** (0.41)� 13 years �8.42** (0.49) �6.48** (0.50) �6.56** (0.51) �6.47** (0.51)

PredictorsAcquisition experience since 1988 0.04** (0.00) 0.04** (0.00) 0.04** (0.00)Focal acquisition performance 1.44** (0.38) 1.15** (0.39)Experience � performance 0.06** (0.02)Log-likelihood �14,024.72 �13,812.11 �13,804.97 �13,798.82Incremental likelihood-ratio chi-square 425.22** 14.28** 12.29**df 15 16 17 18df 1 1 1

a n � 6,714. Unstandardized coefficients are reported, with standard errors in parentheses.* p � .01, two-tailed tests.

FIGURE 1Interaction between Experience and Performancea

a The actual lower-bound value of acquisition experience was �4.81 (5.15 � 9.96). Because acquisition experience could not benegative, we replaced the lower-bound value with zero.


DISCUSSION

We explored the effects of routines stemmingfrom experience, performance feedback, and theirinteraction in the context of acquisitions made inthe U.S. commercial banking industry. We testedthe theoretical idea that, although routines thatarise from experience with a particular task impactorganizations, they also use performance feedbackto update routines (Baum et al., 2000). We foundsupport for our hypothesized main effects in thatboth acquisition experience and focal acquisitionperformance positively influenced the likelihoodof a firm’s making a subsequent acquisition. Wealso found an interaction effect between these twopredictors of acquisition behavior—the positiveeffect of acquisition experience on subsequentacquisition likelihood was even more pro-nounced when it was accompanied by strongeracquisition performance.

From a learning perspective, our results are con-sistent with the idea that organizational routines,once established, are subject to inertial pressures(Levinthal & March, 1993; March, 1981). We foundthat acquirers were more likely to make subsequentacquisitions as they gained acquisition experience.In addition, in keeping with learning theories thatemphasize performance feedback, we found thatfirms adjust their behaviors in view of prior out-comes, because the likelihood of a future acquisi-tion was positively related to recent acquisitionperformance. These findings suggest that managersrespond to performance feedback by repeating re-warded behaviors but are less likely to persist inpunished behaviors.

To the best of our knowledge, our study is thefirst to directly examine the interaction betweenexperience and performance in predicting futurebehaviors. We found that the joint effect betweenthese two variables was more pronounced whenboth acquisition experience and performance werehigh (i.e., positive). This finding implies that a highlevel of experience accompanied by strong recentacquisition performance sends a clear signal to afirm that it has developed the routines required tomake successful acquisitions from its prior acqui-sition experience, which in turn increases its con-fidence in making subsequent acquisitions. How-ever, poor performance of a focal acquisitionweakens the effect of acquisition experience, pre-sumably because it calls into question the decisionmakers’ belief that the acquisition routines thathave arisen from their prior acquisition experienceare appropriate strategic choices. Thus, our find-ings contribute to a more refined understanding ofthe influence of routines on organizational behav-

ior because they show that while firms do persist intheir routines, these routines appear to be sensitiveto performance feedback, especially at highly pos-itive levels of acquisition experience.

Our interaction results also reveal an interestingbut potentially counterintuitive insight, that acqui-sition performance has a greater effect at a higherlevel of acquisition experience than at a lower levelof acquisition experience. When recent acquisitionperformance is strong (x̄ � 1 s.d.), the multiplierof the hazard rate of making a subsequent acqui-sition is 1.07 (e.04�0�1.15�.06�.06�0�.06) at a low levelof acquisition experience (x̄ � 1 s.d.). However,the multiplier (for strong acquisition perfor-mance) is 2.07 (e.04�15.11�1.15�.06�.06�15.11�.06) at ahigh level of acquisition experience. A similarpattern is observed when acquisition perfor-mance is poor. Some theorists would expect theopposite to be true—that is, they would expectperformance to have a greater effect when a firmhas little acquisition experience because it hasless data to use in evaluating the effect of acqui-sition, and/or the organization is less inert. Incontrast, we found that the effect is actually sym-metric for both decreases and increases in acqui-sition performance, so that the acquisition rateresponds more readily to acquisition perfor-mance at stronger levels of acquisition experi-ence.11 This inference would seem to be contra-dictory to organizational inertia theory, whichimplies slower adjustment to performance as ex-perience increases.

Our interviews also provided qualitative evi-dence to bolster the face validity of our empiricalconclusions. The following illustrative quote froman executive at a bank holding company located inthe Midwest supports our inference that perfor-mance feedback moderates the effects of routineson organizational learning:

Our bank is quite acquisitive. During the past twoyears, our unit alone made five acquisitions, andthat figure does not even include branch acquisi-tions. All those acquisitions have made us better atmaking acquisitions over the years. We have built acomprehensive acquisition template from our expe-rience that guides us through when making an ac-quisition. It makes the whole process much moreefficient, and acquisition more attractive. I believeacquisition is here to stay as our primary growthstrategy. The only thing that will slow us down inour acquisition program will be questionable resultsof an acquisition we make. If numbers go down[after an acquisition], we will be forced to review

11 We thank an anonymous reviewer for bringing thispoint to our attention.


our strategy and make any necessary adjustment toour template because it tells us our template may beincorrect. But if numbers look good, we will be onanother negotiation table shortly after.

In addition, our study is the first to show thatrecent acquisition performance influences subse-quent acquisition behavior. Most prior acquisitionstudies in the organizational learning traditionhave focused primarily on the impact of prior ac-quisition experience on acquisition performance bytreating performance as an outcome. Yet the effectof prior acquisition performance on subsequent ac-quisitions has not been explored, despite its poten-tial theoretical and practical importance. By treat-ing recent performance as an independent variable,we investigated this unexplored question andfound empirical support for our theoretical conjec-ture that researchers should take performance feed-back from a prior acquisition into account to pro-vide a more complete understanding of a firm’ssubsequent acquisition behavior.

In conclusion, our study provides useful insightsinto the effects of learning from experience andperformance in the context of corporate acquisi-tions. However, there are several ways in which ourfindings could be extended both theoretically andempirically. Although our empirical models fo-cused on the organizational level of analysis andused archival data to measure organization-levelvariables, we did not directly examine the role ofdecision makers in the studied firms (e.g., Bergh,2001). The characteristics of top managementteams and boards of directors may have importantmoderating or mediating influences on the relation-ship between past acquisition performance andsubsequent acquisition decisions. Future researchthat incorporates measures of managerial character-istics may help explain more variance in organiza-tional decisions. In addition, although we exam-ined the effects of acquisition experience, prioracquisition performance, and their interaction, wedid not examine the processes that underlay oureffects. It may be that for acquisition success firmsneed to manage their acquisition experience andhave some corporate activity to facilitate its transferto the next acquisition. A useful direction for futureresearch would be to explore the ability of organi-zational skills or capabilities to explain additionalvariance in acquisition behaviors. Moreover, inkeeping with prior work on acquisitions and thecharacteristics of our empirical context, we mea-sured acquisition performance in terms of cumula-tive abnormal returns. However, other measures ofacquisition performance, such as return on invest-ment, Tobin’s Q, or acquisition survival, might be

more appropriate in other industry contexts. Fi-nally, we focused our study on a single industry ina particular historical period so that we could de-velop and test a more accurate empirical model.However, we recognize this focus may limit thegeneralizability of our findings to other industrycontexts and time periods. Future research that ex-amines our theoretical predictions in other empiricalsettings and time periods could help develop a moregeneralizable theory of organizational learning.

REFERENCES

Aiken, L. S., & West, S. G. 1991. Multiple regression:Testing and interpreting interactions. NewburyPark, CA: Sage.

Allison, P. D. 1984. Event history analysis: Regressionfor longitudinal event data. Newbury Park, CA:Sage.

Amburgey, T. L., Kelly, D., & Barnett, W. P. 1993. Reset-ting the clock: The dynamics of organizationalchange. Administrative Science Quarterly, 38: 51–73.

Amburgey, T. L., & Miner, A. S. 1992. Strategic momen-tum: The effects of repetitive positional, and contex-tual momentum on merger activity. Strategic Man-agement Journal, 13: 335–348.

Argote, L. 1999. Organizational learning: Creating, re-taining, and transferring knowledge. Hingham,MA: Kluwer Academic Publishing.

Asquith, P., Bruner, R. F., & Mullins, D. W. 1983. Thegains to bidding firms from merger. Journal of Fi-nancial Economics, 11: 121–139.

Audia, P. G., Locke, E. A., & Smith, K. G. 2000. Theparadox of success: An archival and a laboratorystudy of strategic persistence following a radical en-vironmental change. Academy of ManagementJournal, 43: 837–853.

Barnett, W. P., Greve, H. R., & Park, D. Y. 1994. Anevolutionary model of organizational performance.Strategic Management Journal, 15: 11–28.

Barney, J. B., & Hesterley, W. 1996. Organizational eco-nomics: Understanding the relationship between or-ganizations and economic analysis. In S. R. Clegg, C.Hardy, & W. Nord (Eds.), Handbook of organization-al studies: 115–147. London: Sage.

Baum, J. A. C., Li, S. X., & Usher, J. M. 2000. Making thenext move: How experiential and vicarious learningshape the locations of chains’ acquisitions. Admin-istrative Science Quarterly, 45: 766–801.

Bentson, J. G. 1985. The validity of profits-structure stud-ies with particular reference to the FTC’s line ofbusiness data. American Economic Review, 75: 37–67.

Bergh, D. D. 1997. Predicting divestiture of unrelated


acquisitions: An integrative model of ex ante condi-tions. Strategic Management Journal, 18: 715–731.

Bergh, D. D. 2001. Executive retention and acquisitionoutcomes: A test of opposing views on the influenceof organizational tenure. Journal of Management,27: 603–622.

Bergh, D. D., & Lawless, M. W. 1998. Portfolio restruc-turing and limits to hierarchical governance: Theeffects of environmental uncertainty and diversifica-tion strategy. Organization Science, 9: 87–102.

Blossfeld, H. P., & Rohwer, G. 1995. Techniques of eventhistory modeling: New approaches to causal anal-ysis. Hillsdale, NJ: Erlbaum.

Boeker, W. 1989. Strategic change: The effects of found-ing and history. Academy of Management Journal,32: 489–515.

Capron, L., & Pistre, N. 2002. When do acquirers earnabnormal returns? Strategic Management Journal,23: 781–794.

Chatterjee, S., Harrison, J. S., & Bergh, D. D. 2003. Failedtakeover attempts, corporate governance and refo-cusing. Strategic Management Journal, 24: 87–96.

Cohen, J., & Cohen, P. 1983. Applied multiple regres-sion/correlation analysis for the behavioral sci-ences (2nd ed.). Hillsdale, NJ: Erlbaum.

Cyert, R. M., & March, J. G. 1963. A behavioral theory ofthe firm. Englewood Cliff, NJ: Prentice-Hall.

Datta, D. K., Pinches, G. E., & Narayanan, V. K. 1992.Factors influencing wealth creation from mergersand acquisitions: A meta-analysis. Strategic Man-agement Journal, 13: 67–84.

Denis, D. J., & Denis, D. K. 1995. Performance changesfollowing top management dismissals. Journal ofFinance, 50: 1029–1057.

Duhaime, I. M., & Grant, J. H. 1984. Factors influencingdivestment decision-making: Evidence from a fieldstudy. Strategic Management Journal, 5: 301–318.

Eisenhardt, K. M. 1989. Building theories from casestudy research. Academy of Management Review,14: 532–551.

Fama, E., & French, K. 1992. The cross section of ex-pected stock returns. Journal of Finance, 47: 427–465.

Finkelstein, S., & Haleblian, J. 2002. Understanding ac-quisition performance: The role of transfer effects.Organization Science, 13: 36–47.

Greve, H. R. 1998. Performance, aspirations, and riskyorganizational change. Administrative ScienceQuarterly, 43: 58–86.

Greve, H. R. 2003. Organizational learning from per-formance feedback. New York: Cambridge Univer-sity Press.

Greve, H. R., & Taylor, A. 2000. Innovations as catalystsfor organizational change: Shifts in organizational

cognition and search. Administrative Science Quar-terly, 45: 54–83.

Gulati, R. 1995. Social structure and alliance formationpatterns: A longitudinal analysis. AdministrativeScience Quarterly, 40: 619–652.

Hamilton, R. T., & Chow, Y. K. 1993. Why managersdivest: Evidence from New Zealand’s largest compa-nies. Strategic Management Journal, 14: 479–484.

Hannan, M. T., & Freeman, J. 1984. Structural inertia andorganizational change. American Sociological Re-view, 4: 149–164.

Hannan, M. T., & Freeman, J. H. 1989. Organizationalecology. Cambridge, MA: Harvard University Press.

Hannan, T. H., & Wolken, J. D. 1989. Returns to biddersand targets in the acquisition process: Evidence fromthe banking industry. Journal of Financial ServicesResearch, 3: 5–16.

Haunschild, P. R. 1993. Interorganizational imitation:The impact of interlocks in corporate acquisitionactivity. Administrative Science Quarterly, 38:564–592.

Haunschild, P. R., & Beckman, C. M. 1998. When dointerlocks matter? Alternate sources of informationand interlock influence. Administrative ScienceQuarterly, 43: 815–844.

Haunschild, P. R., & Miner, A. S. 1997. Modes of interor-ganizational imitation: The effects of salience anduncertainty. Administrative Science Quarterly, 42:472–500.

Haveman, H. A. 1992. Between a rock and a hard place:Organizational change and performance under con-ditions of fundamental environmental transforma-tion. Administrative Science Quarterly, 37: 48–75.

Hayward, M. L. A. 2002. When do firms learn from theiracquisition experience? Evidence from 1990–1995.Strategic Management Journal, 23: 21–39.

Healy, P. M., Palepu, K. G., & Ruback, R. S. 1992. Doescorporate performance improve after mergers? Jour-nal of Financial Economics, 31: 135–175.

Hogarth, R. M., & Einhorn, H. J. 1992. Order effects inbelief updating: The belief adjustment model. Cog-nitive Psychology, 24: 1–55.

Ingram, P., & Baum, J. A. C. 1997. Opportunity and con-straint: Organizations’ learning from the operatingand competitive experience of industries. StrategicManagement Journal, 18: 75–98.

Kane, E. J. 1996. De jure interstate banking: Why onlynow? Journal of Money, Credit and Banking, 28:141–161.

Kaplan, S. N., & Wiesbach, M. S. 1992. The success ofacquisitions: Evidence from divestitures. Journal ofFinance, 47: 107–138.

Kelley, D., & Amburgey, T. L. 1991. Organizational iner-tia and momentum: A dynamic model of strategic


change. Academy of Management Journal, 34: 591–612.

King, D. R., Dalton, D. R., Daily, C. M., & Covin, J. G.2003. Meta-analyses of post-acquisition perfor-mance: Indications of unidentified moderators. Stra-tegic Management Journal, 25: 187–200.

Lang, L., Stulz, R. M., & Walkling, R. 1991. A test of thefree cash flow hypothesis. Journal of Financial Eco-nomics, 29: 315–335.

Levinthal, D. A., & March, J. G. 1993. The myopia oflearning. Strategic Management Journal, 14: 95–112.

Levitt, B., & March, J. G. 1988. Organizational learning. InW. R. Scott (Ed.), Annual review of sociology, vol.14: 319–340. Palo Alto, CA: Annual Reviews.

March, J. G. 1981. Footnotes to organizational change.Administrative Science Quarterly, 26: 563–577.

March, J. G. 1991. Exploration and exploitation in organ-izational learning. Organization Science, 2: 71–87.

March, J. G. 1999. The pursuit of organizational intelli-gence. Oxford, U.K.: Blackwell.

Miller, D., & Chen, M. J. 1994. Sources and consequencesof competitive inertia: A study of the U.S. airlineindustry. Administrative Science Quarterly, 39:1–23.

Miller, D., & Friesen, P. H. 1980. Momentum and revo-lution in organizational adaptation. Academy ofManagement Journal, 23: 591–614.

Miller, D., & Friesen, P. H. 1982. Innovation in conserva-tive and entrepreneurial firms: Two models of stra-tegic momentum. Strategic Management Journal,3: 1–25.

Miller, D., & Friesen, P. H. 1984. Organizations: A quan-tum view. Englewood Cliffs, NJ: Prentice-Hall.

Nelson, R. R., & Winter, S. G. 1982. An evolutionarytheory of economic change. Cambridge, MA: Har-vard University Press.

Puffer, S. M., & Weintrop, J. 1991. Corporate perfor-mance: The role of performance expectations. Ad-ministrative Science Quarterly, 36: 1–9.

Rajagopalan, N., & Spreitzer, G. M. 1997. Toward a theoryof strategic change: A multi-lens perspective andintegrative framework. Academy of ManagementReview, 22: 48–79.

Ramaswamy, K. 1997. The performance impact of strate-gic similarity in horizontal mergers: Evidence fromthe U.S. banking industry. Academy of Manage-ment Journal, 40: 697–715.

Ravenscraft, D. J., & Scherer, F. M. 1987. Mergers, sell-offs, and economic efficiency. Washington, DC:Brookings.

Rhoades, S. A. 1994. A summary of merger performancestudies in banking, 1980–1993, and an assessment ofthe “operation performance” and “event study”

methodologies. Federal Reserve Bulletin, July:1–37.

Shaver, J. M., Mitchell, W., & Yeung, B. 1997. The effectof own-firm and other-firm experience on foreigndirect investment survival in the United States,1987–1992. Strategic Management Journal, 18:811–824.

Sirower, M. 1997. The synergy trap: How companieslose the acquisition game. New York: Free Press.

Steiner, D. D., & Rain, J. S. 1989. Immediate and delayedprimacy and recency effects in performance. Journalof Applied Psychology, 74: 136–142.

Strauss, A., & Corbin, J. 1998. Basics of qualitative re-search: Techniques and procedures for developinggrounded theory. Thousand Oaks, CA: Sage.

Szulanski, G. 1996. Exploring internal stickiness: Imped-iments to the transfer of best practice within thefirm. Strategic Management Journal, 17: 27–42.

Taylor, M. L. 1988. Divesting business units: Makingthe decision and making it work. Lexington, MA:Lexington Books.

Tuma, N. B., & Hannan, M. T. 1984. Social dynamics:Models and methods. New York: Academic Press.

Tushman, M. L., & Romanelli, E. 1985. Organizationalevolution: A metamorphosis model of convergenceand reorientation. In L. L. Cummings & B. M. Staw(Eds.), Research in organizational behavior, vol. 7:171–222. Greenwich, CT: JAI Press.

Yamaguchi, K. 1991. Event history analysis. NewburyPark, CA: Sage.

Jerayr (John) Haleblian ([email protected]) is anassistant professor of strategic management at the Ander-son Graduate School of Management at the University ofCalifornia, Riverside. He received his Ph.D. from theUniversity of Southern California. His work focuses oncorporate governance, mergers and acquisitions, strategicdecision making, and top management teams.

Ji-Yub (Jay) Kim ([email protected]) is an assis-tant professor of management and organization in theMarshall School of Business at the University of South-ern California. He received his Ph.D. in business admin-istration from the University of Wisconsin–Madison. Hiscurrent research focuses organizational learning, mergersand acquisitions, and organizational evolution.

Nandini Rajagopalan ([email protected]) is aprofessor of management and organization in the MarshallSchool of Business at the University of Southern California.She received her Ph.D. in strategy from the University ofPittsburgh. Her research focuses on CEO succession, topmanagement compensation, strategic change, strategic de-cision processes, and global strategic alliances.


Documents

THE INFLUENCE OF ACQUISITION EXPERIENCE AND …