Age of Firms: Irrelevance Proposition

Age of firms is used with an alarming regularity in various studies in the fields of organizational behavioral, accounting and corporate finance/law/governance, industrial economics and the like. The variable “firm age” is typically called a control variable. Why it is necessary or what it controls is infrequently ex-plained. The variable’s statistical significance or insignificance is duly noted; explanation thereof is hardly ever provided. This paper demonstrates that this variable does not provide any sensible information its users would have us believe. The paper provides numerous examples demonstrating that managerial discretion changes the nature and value of the firm, thereby negating the value of the variable. We recommend that a conceptual foundation be se-cured before using the variable and it not be wantonly used as a control or fixed-effect variable. Researchers would do well to report the effect size instead of merely statistical significance at some level.


Introduction
"Age of firm" (also "firm age"; both phrases are used interchangeably) is used with an alarming regular frequency in various studies in the fields of organizational behavioral, accounting and corporate finance, law and law and economics, corporate governance, industrial economics and the like. Rarely is the variable used as an explanatory one; the variable is typically called a control variable. Why it is necessary or what it controls is infrequently explained. The variable's statistical significance or insignificance is duly noted; explanation thereof is hardly ever provided. 1 In this paper we take on the challenge of demonstrating the irrelevance of the use of the age of firm variable. We demonstrate that this variable does not provide any sensible information its users would have us believe. Scholars who use this variable in studies in management and strategic management often appeal to Cyert and March [2] [3] 2 or Penrose [4]. As the works of Penrose have become more popular, the use of the variable is attributed to a resource-based view of the theory of the firm. While management scholars have embraced Penrose and created an enormous amount of literature on the resource-based view and its extensions, nowhere does Penrose or her followers imply that firm age is deterrence to the production of knowledge or growth of the firm. The other two dominant views of the firm, viz., transaction-cost economics or agency theory do not make any prediction about the age and its effect on transaction or agency costs for organization. The vast literature on corporate governance is also nearly silent on the issue of firm age. 3 A review of the literature across disciplines lays bare the absence of conceptual, theoretical and empirical studies using firm age as an explanatory variable (or dependent variable!) in organizational behavior studies or strategic management studies, despite its often perfunctory inclusion as a control variable. The areas of finance and accounting have seldom used the variable and typically the use is not well-integrated. The literature does offer a few instances in industrial economics where the variable is justifiably used. A majority of studies claim to use firm age as a control variable. It will not be an exaggeration to say that these studies would not lose any of their potential if the variable of firm age were left out. At the minimum one would not be able to allege that they are using it superfluously.
We first review the literature where this variable is used. We then explore a more recent debate about the interpretation of variable effect sizes which might be used to argue that age of firm variable actually does explain something meaningful. However, we find this argument to be without merit. We provide numerous examples showing that managerial discretion changes the nature of the firm, thereby negating the value of the firm age variable. We conclude that these examples demonstrate the lack of relationship between the age of a firm and its ability to adapt to environmental changes. Despite this, researchers have continued to allude to ossification or perception thereof. 4 After Cyert and March an appeal may be made to Hannan and Freeman [8] through the concept of structural inertia. In their work, however, Hannan and Freeman [8] [9] lend support to the adaptability of firms to volatile environment through the lens of ecological-evolutionary process. They examine inertial pressures on organizations, labelling this tendency "structural inertia", and scrutinize how their strength varies with age, size and complexity. The reader, however, should recognize that the claim of strong inertial pressures does not equate to the claim that organizations never change, but that organizations respond dynamically and deliberately to threats and opportunities in their environments.
They state (p. 151) "structures of organizations have high inertia when the speed of reorganization is much lower than the rate at which environmental conditions change." They develop some testable propositions, viz., 1) Structural inertia increases monotonically with age, 2) Organizational death rates decrease with age (the "liability of newness hypothesis" due to Stinchcombe [10]), 3) Attempts at reorganizations increase death rates (depending on size, defined by the number of employees), 4) Complexity (of an organization which is non-hierarchical with multiple interconnected units) increases the risk of death due to reorganization.
Brüderl and Schüssler [11] study West German business organizations to introduce the concept of a "liability of adolescence" and provide empirical evidence. 4 Merton [13] actually anticipates March and his associates extremely well in his study of bureaucracy. He credits Mannheim [14], (revised and expanded in [15]) for the development of the concept of "rational organization". Hannan [12] provides three logical formalizations for empirical findings concerning the relation between organization age and the hazard of mortality. These formalizations depend on five concepts of endowment, imprinting, inertia, capability and position. Thus it is structural inertia which may hinder organizational choice in decision making, not firm age per se.
More commonly, it is the "resource-based view" (RBV) (also known as "resource-based theory" (RBT)) attributed to Penrose [4] that is often cited for the inclusion of the variable "firm age" in various empirical studies in the fields of organizational studies and strategic management where the Penrosean viewpoint of the firm as a resource-based or knowledge-based or dynamic capabilities-based entity is one of the most widely accepted perspectives. Penrose's ideas are not easily summarized. Penrose and Pitelis's ( [16], pp. 11-12, Pitelis [17]) attempt, given below, is admirably clear: 1) Firms are bundles of resources, under internal direction, for use of goods and services, sold in markets for a profit. Their boundaries are defined by the area of coordination and "authoritative communication".
2) Firms differ from markets; transactions in the latter do not take place within "administrative coordination".
3) Resources render (multiple) services. The heterogeneity of services from resources gives each firm its unique character. Effective use of resources takes place when resources are combined with other resources. 4) Human, and in particular managerial, resources are of essence, because expansion requires planning and managerial resources that enable the firm to plan are firm-specific, they cannot be acquired in the market.
5) The cohesive shell of the firm helps to create knowledge. This can be "objective" (transmittable) or experience (hard to transmit). Experience renders managerial services firm-specific. 6) Unused resources always exist; they are released after the completion of an expansion and they are created through experience and new knowledge. They are an internal stimulus to growth and innovation, and determine in part the direction of expansion. 7) Firms are not defined in terms of products, but resources and (so) "diversification" is the normal state of affairs in firm expansion. 8) There are economies of growth, quite apart from any economies of size. 9) There are limits to growth, but not to size, and they are determined by the rate at which experienced managerial staff can plan and implement plans. The services of "inherited" managerial resources control the amount of new managerial resources that can be absorbed, and thus limit the rate of growth of firms.
10) The external environment is an "image" in the mind of the entrepreneur.
Firms' activities are governed by their "productive opportunity", i.e., all the productive possibilities that its entrepreneurs can see and take advantage of. 11) Entrepreneurs are in search of profits; firms desire to increase total long-term profits "for the sake of the firm itself and in order to make more profit through expansion" (1959, p. 29). In the long run, growth and profits are equiv- A search in the authoritative third edition of Penrose [29] and the extensive work and commentaries of Pitelis [17] [30]- [35] and Pitelis and Pseiridis [36] has not changed the result. the economic organisation we describe as "firm" defines both the nature and the limitations of such a body. A firm's growth is the result of certain mechanisms that govern expansion, and foremost among these are the managerial functions. They are the basis of the coordination and communication that hold together the resources the firm has at its disposal. There is an endogenous mechanism-the active search for better ways of using internal resources-which dynamically leads to the growth of the resources the firm has at its disposal, and thus further fuels growth. Expansion, therefore, is a recurrent and unbalanced phenomenon. ings. Nowhere, however, do Penrose or her followers imply that firm age is deterrence to the production of knowledge or growth of the firm.
Evans [38] [39] appears to be the first researcher to explore the relationship between firm growth, firm size and firm age. His concern is growth dynamics at the micro-level. His key finding is that the probability of firm failure, firm growth and the variability of firm growth decrease as firm ages. Evans claims that the finding of the inverse growth-age relationship is consistent with Jovanovic's [40] theory of firm-learning. Jovanovic models a theory of selection with incomplete information within an industry to conclude that "Firms learn about their efficiency as they operate in the industry. The efficient grow and survive; the inefficient decline and fail." We consider the implication of relationship between age on one hand and survival and growth on the other nearly tautological.
Note further that neither Evans nor Jovanovic mentions Penrose! Leonard-Barton [41] hints that culture of the firm produces rigidity harming innovation and new product development. Incidentally note that Leonard-Barton makes no mention of age! Therefore, unless firm age is shown to cause culture that causes rigidity that causes various ills, we must remain alert to other causes of rigidities and not firm age per se. Christensen  Economists, sociologists, political scientists, and organization scientists have developed life-cycles of products and industries. The temptation to port this idea to business firms is immense. Mueller [69] develops the life-cycle of a one-product profit-maximizing firm to conclude that managers pursue a growth-maximization policy. He further draws two inferences, viz., 1) "If large mature firms are investing too much, then someone in the economy must be spending too little." and 2) "The pursuit of growth by managers [will] increase overall concentration [in the control of economic activity]." One might deduce from Mueller's phrase "mature firm" implications for the word "age" or the phrase "firm age". This in turn leads to the conjecture that firm age must be important. 10  (except utilities and financial firms), they find support for their hypotheses. Similarly, a superb integrated use of firm age is provided by Ouimet and Zarutskie [75] 11 . They posit and empirically demonstrate that 1) young firms disproportionately employ and hire young workers, 2) young employees disproportionately join young firms with potential for greater innovation and growth, and 3) a causal link exists between young workers and new firm creation.
The discussion hitherto lays bare the absence of conceptual, theoretical and 10 Shumway [70] appears to be the first and unique in that he provides a justification for firm age in his conceptual framework and then empirical work and then uses the variable as an explanatory variable. the variable and typically the use is not integrated. A majority of studies claim to use firm age as a control variable. It will not be an exaggeration to say that these studies would not lose any of their potential if the variable of firm age were left out. At the minimum one would not be able to allege that they are using it superfluously. 12 In the next two sections we demonstrate that firm age is not exactly a meaningful variable regardless of the exact conceptual or numerical definition or statistical significance testing for vaguely-stated hypotheses.

Statistical Inference vs. "New Statistics" of Effect Size and Confidence Interval
The earlier section provided numerous studies from various business disciplines where the use of the variable firm age, however defined, was neither conceptually sound nor statistically meaningful. In addition the section highlighted a few studies where the use was conceptually appropriate for inclusion. In this section we discuss statistical inference as it relates to understanding of the importance of the age of firm variable. Specifically we allay doubts that there could be more importance to the variable if we examine statistics like effect size or confidence intervals in regressions using the age of firm variable. All of the studies in the earlier section deploy traditional statistical inference methods, i.e., drawing conclusions about a population parameter from the da- which the null is tested. The procedure is almost the same. In both cases, researchers often provide confidence intervals for population parameters. Typically 95% and 90% confidence intervals are given.
One of the earliest studies to use statistical methods (Arbuthnott [76]) shows that the excess of male births is statistically significant and the excess is due to Divine Providence. NHST is actually an accidental hybrid of the tests proposed by Fisher [77] [78] and Neyman and Pearson [79]. As noted elsewhere in the literature, criticisms of NHST are quite old (Boring [80], Berkson [81]). The modern criticisms of NHST can be found in Rozeboom [82], Carver [83], Cohen [84] [85], Schmidt [86], Anderson, Burnham and Thompson [87], Thompson [88]- [93] who also provide alternatives. These criticisms have been published in numerous disciplines, such as socio-economics, education and its sub-disciplines, psychology and its sub-disciplines, ecology and wildlife science, present the history of significance testing within psychology and education. One of the best summaries of NHST problems is Kline [99] who advocates for effect sizes and confidence intervals 14 . A few disciplines in the social sciences have made concerted effort to move away from the traditional methods of NHST in favor of "new statistics" of effect size and confidence interval (Cumming [102]).
The American Psychological Association (APA) has been a pioneer in this effort since 1994 when its fourth edition [103] of the Publication Manual advocated statistics that characterize magnitude of effect (e.g., Cohen's d, R 2 , η 2 , ω 2 ). Effect size quantifies by how much sample results diverge from the null hypothesis.
APA published the fifth edition [104] of the Publication Manual in 2001. It required the following: "For the reader to fully understand the importance of your findings, it is almost always necessary to include some index of effect size or 13 We wish that they had titled their paper "Men Are Highly Significant(ly) Idiots (p < 0.0001)", based on the title of the paper by Hubbard [95]. 14 Tyler [100] provides a "tutorial" in statistical significance. Arrow [101] discusses the choice of the level of significance for hypotheses in decision theory. strength of relationship in your results section…" (pp. [25][26]. The Manual admonished "… failure to report effect sizes …" is a "… defect in the design and reporting of research …" (p. 5). The APA published the sixth edition [105] of the Publication Manual in 2010. It is clear about the necessity of reporting effect size: For the reader to appreciate the magnitude or importance of a study's findings, it is almost always necessary to include some measure of effect size… Effect sizes may be expressed in the original units (e.g., the mean number of questions answered correctly; kg/month for a regression slope) and are often most easily understood when reported in original units. It can often be valuable to report an effect size not only in original units but also in some standardized or units-free unit (e.g., as Cohen's d value) or a standardized regression weights. (p. 34) "… mention all relevant results, including those that run counter to expectation; be sure to include small effect sizes (or statistically non-significant findings)" … (p. 32) But for purposes of the current paper, we need not summarize the history of NHST or the criticisms of NHST or counter-arguments or alternatives. 15 Cumming [102] has already provided an invaluable volume covering these topics. Cummig ( [102], p. 39), Huberty [107] and Kirk [108] discuss numerous effect size measures. Rather, the concern for us is whether or not these arguments against the use of NHST have relevance in our argument against the inclusion of firm age as a variable in organizational studies. Simply, we are trying to counter-argue against a mimetic isomorphism rampant in the management literature.
In terms of the reported statistics for the firm age variable, these counter-arguments are two. 1) Whether NHST is a problem, regardless of the extent, depends on the context of the discipline. 2) Whether "new statistics" of effect size, confidence intervals and meta-analysis can shed any light on the proposition that firm age is an irrelevant variable in most of the literature of management sub-disciplines (which is not the same as business disciplines).
Ziliak and McCloskey ( [109], p. 527) (ZM) in a follow-up study to the contribution of McCloskey and Ziliak [110] state that in economics Significance testing as used has no theoretical justification. … significance testing is getting worse. A super majority (81%) believed that looking at the sign of a coefficient sufficed for science, ignoring size. … The confusion between fit and importance is causing false hypotheses to be accepted and true hypotheses to be rejected. We propose a publication standard for the future: "Tell me the oomph of your coefficient; and do not confuse it with merely statistical significance." Zellner [111] first documented the unsatisfactory use of NHST in a sample of 1978 issues of several leading economics and econometric journals and then ad- 15 For a recent review and survey, see Mbengué ([106], especially page 119) with its incendiary title. … pushing an economically large effect that is statistically insignificant is usually a stretch. Suppose I sample 10 high-school graduates from a population, and estimate a return to each year of college of 30%, but with a t statistic of one. Should I then make policy based on the 30% figure? If another researcher uses 5000 observations and estimates the return to be 12% with t = 20, should I prefer the 30% estimate because it is economically more important? I see a growing problem of instances where researchers ignore statistical significance, and statistical issues more generally, and focus too much on magnitude in order to push a particular story or policy. As an example, it has become a cottage industry in empirical microeconomics to find a clever instrumental variable for an endogenous explanatory variable. Naturally, ordinary least squares estimation is set up as the straw man, and then instrumental variables estimation (usually two stage least squares) is used to solve the endogeneity problem. Often-especially if the paper is published in a good journal-the 2SLS estimate is bigger in magnitude than the OLS estimate. Even in cases where a Hausman test fails to reject that OLS is consistent, researchers often appeal to the 2SLS estimate for policy discussions. Fisher is the best known of the inadvertent "fathers" of the null [NHST] ritual. His influence has divided psychologists deeply, and interestingly, the 16 See also Lecoutre [112] and Lecoutre, Lecoutre and Poitevineau [113] for advocacy of Bayesian methods. For the endemic problems and pitfalls of NHST and poor statistical reporting practices in scholarly research both Gigerenzer ([119], pp. 592-294) and Altman [120] blame the culture of publishing and journal editors.
While the substantive differences in conceptual framework, theoretic modeling and empirical methods have diminished the ardor for new statistics in economics and finance, the similarity of framework, instruments and methods has brought the crusade of new statistics of effect sizes and confidence intervals over into business (functional) disciplines such as marketing (Sawyer and Peter [121]) and (support) disciplines such as forecasting, 18 human resources (HR) and international business (IB) management.
Since the seminal articles by Huselid [124] and MacDuffie [125] on the impact of HR practices on firm performance, a large number of papers has examined various aspects of these practices. Methodological issues are examined by numerous papers beginning with Huselid [126] and continuing with Gerhart [127], Gerhart, Wright, McMahan and Snell [128], Gerhard, Wright and McMahan [129], Huselid and Becker [130] and Wright et al. [131]. Combs, Liu, Hall and Ketchen [132] provide a meta-analysis of high-performance work practices and organizational performance using effect size. 19  χ is significant, we assume residual variance is heterogeneous. Otherwise, homogeneity is assumed." In the text (p. 513) the authors explain, "When 2 1 K χ − was not significant, the effect was considered homogeneous (i.e., one population effect with no remaining moderators). All variance is assumed to be due to sampling error, and the standard error of sampling error variance was used to create confidence intervals for the homogeneous case. When significant variance remained unexplained, a wider confidence interval was used based on the standard error of the total effect size variance." It is hard for an uninitiated reader to appreciate whether 2 1 K χ − is Cochran's Q and whether the individual studies meet the criteria for combinability. We will not go into the issues of field studies and their conceptual foundations in terms of alternative explanations, omitted variables, measurement errors, simultaneity and structural causation.

Illustrations
With a select few examples, we highlight the incontrovertible fact that firm age is unlikely to capture much information because many things in both internal and The clear-cut implication of the foregoing is that the age of the firm becomes less informative as an explanatory variable for the behavior of the firm and redundant as a control variable. Given this conclusion we posit that the best way to conceptualize a firm is to consider it a dynamic system (contrasting from a static system) which evolves in response to internal and external stimuli. The factors influencing a firm's behavior are different at different periods of time and there is unlikely to be a single dominant factor constant through the firm's evolution.
The following examples will demonstrate this thesis emphatically.

Nokia
Today Nokia is known as a Finnish communications and information technology company with headquarters in Espoo, Uusimaa, Finland. 20 The company provides Internet services as well as applications, games, music, media and messaging services, etc., and telecommunications network equipment and services. In Can we consider Nokia as a static firm? 22 No. Would we expect firm age to 21 See http://en.wikipedia.org/wiki/Nokia. 22 A static firm can be defined by contrast with a dynamic firm. A dynamic firm is trying to sustain its competitive advantage by constantly adapting to changing circumstances in various markets in which it participates. This adaptation would help the firm to be a continuing firm, yet the adaptation would change the inherent nature of the firm, including its business model. If a firm wishes to be a long-lived firm, then it must continually evolve; it cannot remain the same. Even the fundamental accounting principle of "going concern" inform us that a company will be able to continue operating for a period of time that is sufficient to carry out its objectives, obligations, promises without jeopardizing the invested capital and this is possible only if the company adapts.
provide any explanatory power with respect to asset rigidity, managerial discretion, etc.? No.

CBS
CBS Corporation is one of the most interesting evolutions of a company. Additionally we will need to distinguish between CBS as a brand and CBS as a company, recognizing the phenomenon of brand names as assets that can be licensed or sold. Additionally this example can be labelled as "CBS/Westinghouse/Viacom" because of the complex history, even though we will retain the focus on CBS.
Today's CBS Corp. is a well-known commercial broadcast television network which also owns radio network, numerous TV-and radio-stations in large-and mid-size markets, publishing groups and some cable networks. The name reflects a convoluted history. The company was incorporated in 1927 as United Independent Broadcasters, Inc. The name was changed to Columbia Broadcasting System in 1928 by William S. Paley as the president who later in the same year became the majority owner. In 1974 the company changed the name to CBS Inc. In 1995 the Westinghouse Electric Corp. acquired CBS Inc. and in 1997 changed its own name to CBS Corp. (Note that Westinghouse Electric Corp. itself was founded in January 1886 as Westinghouse Electric Co. and later renamed by the founder George Westinghouse himself. The company was engaged in long-distance high-voltage AC power transmission. Later it designed and developed nuclear reactors for the US military's submarine and aircraft carrier. 23 ) CBS continued to prosper as a radio network concentrating on music, comedy and variety shows, a large number of which were directed at women. In 1934 it launched, much to the consternation of newspapers and wire services, its independent news division hiring a year later Edward R. Murrow. Murrow and Murrow's Boys (William L. Shirer, Charles Collingwood and Eric Sevareid) developed the News Round-Up format which is the mainstay of radio and TV news today. The decade of 1950s saw CBS struggle against the RCA-system of color TV preferred by both NBC and ABC as well as the rivalry between its preferred UHF frequency band versus more capacious VHF band. TV proved to be a disruptive invention such that most of the successful programs moved from the radio medium to the TV medium; unsurprisingly, money and profits followed suit. During these years, CBS continued with buying and selling radio stations and TV stations throughout the country. The decade of 1970s saw CBS do better than other networks in terms of programming and profits.
While the businesses of radio and TV broadcasting were moving up, the management of the company was undergoing substantial changes. Earlier CBS's 23 Westinghouse's own interesting history can be provided as another illuminating illustration of companies undergoing unbelievable changes. The company itself is now defunct. In 1982 the company made another attempt, after abandoning its production unit in 1972, at film production in a joint venture, TriStar Pictures, with Columbia Pictures and HBO. The stake was sold in 1985 to Coca-Cola Company, the then-owner of Columbia Pictures. The company's romance with home video market started in 1978 in a joint venture with MGM and ended in 1982.
By mid-1990s, majority stakeholder Tisch was becoming less satisfied at the trajectory of profits. The company was sold outright to Westinghouse in 1995 for $5.4 billion. Westinghouse acquired and traded numerous radio stations, TV stations and cable channels. In 1997, Westinghouse changed its name to CBS Corp. while the headquarters was moved from Pittsburgh to New York. The new company put up all non-entertainment assets for sale. By the end of 1999, all pre-CBS elements of the industrial past of Westinghouse had disappeared.
By the end of the year 1999, however, a new suitor emerged. A division started by the original CBS in 1952 and spun off as Viacom in 1971 that had become a giant communication and media conglomerate under the majority ownership of Sumner Redstone announced a takeover of CBS for $37 billion. The takeover was completed in 2000.
By 2005, however, having found that the communication and media empire he had assembled was not performing at the level he had hoped, Redstone split Viacom into two under the umbrella of his holding company National Amusement. One part became CBS Corp. keeping the earlier CBS, Paramount Television's production operations, United Paramount Network (UPN), Viacom

G. Vora
Outdoor advertising, Showtime Network, Simon & Schuster publishing house (later sold), and Paramount Parks (later sold). The other part retained the name Viacom and kept the portfolio of assets such as MTV Network, Black Entertainment Television (BET) and some music and recording studios. Since this "reorganization", CBS has undertaken many other acquisitions to become the owner of one of the largest libraries of film and TV programs.
Can we consider CBS/Westinghouse/Viacom a static company? No. In addition, recognize the elements and pace of change because of extremely strongly-held views and management practices of the majority stakeholders such as Paley, Tisch and Redstone, showing considerable managerial discretion unrelated to firm age.

International Business Machines (IBM)
Today IBM is known as one of the largest, multinational computer technology, information technology (IT) and IT consulting companies in the world. It has a continuous history since 1880s. 24 The forerunner is the Computing- Tabulating , and others such as Dey Time Register Co., Stimson Computing Scale Co. and Moneyweight Scale Co. CTR was established as a holding company whose board sought an executive not previously affiliated with any of the portfolio companies to integrate them into one harmonious business. The board hired Thomas J. Watson, Sr., on May 4, 1914 as general manager. 25 In February 1924, the name was changed to International Business Machines Corp., a name which had been registered in New York since 1918 and first used in 1917 when International Business Machines Co. Ltd., was established in Canada. Therefore, IBM recognizes 1911 as the establishment year and celebrated its centennial in 2011.
Since 1911 IBM has undergone many changes of personnel and businesses and employees. Today the company has scientists, engineers, consultants, as well as marketing and sales professionals in over 170 countries. Impressively IBM employees have received five Nobel Prizes, 26 six Turing Awards, 27 nine National 24 See IBM FAQ at http://www-03.ibm.com/ibm/history/. 25 IBM has in the past traditionally recognized this date as its anniversary. 26 See http://ibmresearchnews.blogspot.com/2013/05/ibm-research-zurich-officially-turns- 50 Medals of Technology and Innovation (not counting the two corporate awards) 28 and five National Medals of Science 29 among many accolades 30 . The history of IBM's evolution is exceedingly well presented by IBM itself and others. 31,32 Our interest in this history starts from 1980s-1990s. These decades are marked by the twin computer tech revolutions, viz., the PC revolution that put computers directly on the desks of the users and the distributed network revolution that and R&D were deployed to upgrade its existing product lines. As the Internet, 28 See  Can we consider IBM a static company? No. In addition recognize the elements and pace of change because of rapidly fluctuating external environment which resonated with a new CEO who found the wherewithal to respond meaningfully and successfully. Again, IBM's history shows a lack of relationship between asset rigidity, managerial discretion, and firm age.

General Electric Co. (GE)
Though General Electric has gone through many transformations which we argue render firm age a meaningless measure with regard to the company, a more recent example drives this point home. On Sept. 8, 2014, GE, which commercialized the electric toaster and self-cleaning oven, announced that it would sell its appliance business to Sweden-based Electrolux AB for $3.3 billion. Electrolux will continue with GE brand-name. The sale will leave one of the original conglomerates in only a few business lines, namely, non-consumer finance and industrial equipment such as power-generation turbines, aircraft engines and medical devices like CT scanners. The shift is attributed to Jeffrey R. Immelt who in his 14th year as CEO was adjusting to competitive pressures and trying to boost a long-sluggish stock price. Under Immelt GE has sold off businesses as diverse as insurance, plastics, media, consumer finance and now appliances.
And, under Immelt GE has spent around $14 billion acquiring oil-and-gas service companies. In 2013 energy and energy-related activities contributed about one-third of the company's revenues and more than 40 percent of its operating profit. Its non-bank finance division, GE Capital, contributed about one-third of the company's revenues and around 50 percent of its operating profit.
Can we consider GE a static company? No. The ascension of Immelt after the long-tenure of John F. "Jack" Welch (became Vice Chairman in 1979, Chairman and CEO in 1981, resigned in 2001) heralded the New GE Way of innovation and transformation (Magee [138], Owles [139]). Immelt's moves again unders-

Unilever (London and Rotterdam)
The Anglo-Dutch firm, Unilver, is one of the most well-known international firms. It is a dual-listed firm comprising of Unilever PLC (London) and Unilever N.V. (Rotterdam) with two headquarters at London and Rotterdam. Its products span the space of food and beverages, cleaning items and personal care. Unilever's history since 1871 is a history of changes, expansion, and innovations. The web site of the firm at https://www.unilever.com/about/who-we-are/our-history/#timeline+2D+none+ closed provides a detailed history of its existence in one form or another since 1860. The firm gained its current form in 1929 after the merger of the businesses of Dutch Margarine Unie and British soap-maker Lever Brothers. The Wikipedia page on the firm at https://en.wikipedia.org/wiki/Unilever gives a nice summary of the changes in business lines and divisions through mergers and acquisitions and spin-offs. Unilever is considered so attractive that a much-smaller firm, Kraft Heinz, made a $143 billion unsolicited bid to acquire it. The bid was rejected and Unilever remains independent continuing its businesses across the world. As the history of Unilever makes clear the firm has hardly stayed the same though the years; it has constantly been changing. Currently it has organized itself in four multi-country divisions, viz., 1) Personal Care, 2) Foods, 3) Refreshments, and 4) Home Care. Unilever's main international rivals are Nestlé and Procter & Gamble. We devote the next subsection to a brief history of Nestlé.

Nestlé
Nestlé is an international Swiss firm, founded in 1905 by the merger of An- nutrition and health-related products, 5) pet-related products, 6) confections, and 7) bottled water. Many of these business lines were acquired and expanded as well as developed in its R&D labs. In many countries, Nestlé has run into numerous food-related controversies and legal suits on safety of its products. As Nestlé's web site reveals, the firm has constantly adapted to the market place.
Because it operates in most of the countries in the world, it, its products, and its management-corporate and local-are constantly evolving in the face of pressures in each country and market. As the evolution occurs, the firm changes in its essence.

Further Thoughts
We can pick almost any medium or large company, publicly-or privately-held, based in the U.S. or elsewhere. 33 For foreign companies we could have picked other well-known examples of 1) Bayer AG whose nature has been changing as it has acquired firms in animal health, crop science and pharmaceuticals. Bayer has not been to digest the products lines bought from Merck and has been saddled with large legal problems related to products of Monsanto; or 2) Fiat Chrysler Automobiles whose nature has been changing as operations of two brands are still not on an even keel; or 3) Renault-Nissan-Mitsubishi Alliance whose nature has been changing because of brands, markets, marketing and promotion, and inter-country rivalries. For US companies we could have picked other well-known examples of a) United Technologies Corp. whose nature has been changing as it acquired and is now trying to spin itself off into three companies; or b) Amazon.com, Inc. whose nature has been changing as it began its life as a book-seller and has now become a major force in e-commerce, music and video streaming, cloud computing, and many other lines of business; or c) Land O'Lakes, Inc., a co-operative established in 1921, with more than 1700 farmers as current owners, that started in butter business and expanded into many different lines of business and business models. If we dissect a firm's history properly we will see that change is constant. The company has changed its essential character in response to various internal and external factors. Even the most mature companies such as Procter & Gamble, Coca-Cola, and PepsiCo as well as specialized companies such as Steinway Pianos, DuPont, and Kodak go through dramatic changes. 34 This phenomenon is 33 We reluctantly refrain from providing more illustrations because of limitation of space. 34 We are not sure how one can discuss a firm such as AT&T because it represents an extreme case of changes through its history. not restricted to US companies. Almost all firms, regardless of domicile, undergo changes. 35 To ascribe any particular behavioral traits to firms because of their age is not sensible.

Conclusions
This paper argues against the unjustified use of the variable "firm age" in business disciplines, especially in management studies. A theoretic justification for the variable must be provided either as an explanatory variable or as a control variable. The casual, perfunctory use of the variable in empirical studies is decried because it diminishes the value and insights of the studies. Illuminating illustrations are provided to highlight changes in the essential nature of a company.
This importance of the irrelevance proposition is grounded in the importance of theoretical and empirical parsimony. The use of firm age as a control variable is ubiquitous in the management literature. We question its inclusion first on theoretical grounds by showing that the authors cited to support the use of firm age actually offer no such support. But, we also question its inclusion based on the proxies that are used to measure this variable. Simply, the idea that the number of years a firm has been in existence or listed on a major stock should explain anything related to rigidity in and constraints upon managerial decision making is hard to support given the rampant restructuring undertaken by large firms. We suggest that the statistical non-significance often seen for firm age is related to its measurement. The variable, as measured, does not actually capture what the authors argue is theoretically important, that being a measure of managerial and asset rigidity.
The irrelevance proposition is innovative in the way that anyone willing to say "The emperor has no clothes" is innovative. We are challenging conventional wisdom in this article. It is also innovative in this age of ever-more powerful computers which allow for the easy inclusion of explanatory variables without actually thinking about the parsimony of the models being tested. In our discussion, we surface and challenge the strongly held beliefs about the importance of the firm age variable, including the citations to the authors who supposedly argue for its importance, such as Penrose.
We hope that the scholars who study empirically issues in strategic management, management, corporate governance, theory of the firm, finance, accounting and entrepreneurship heed our advice. Some of these groups may not be regular readers of some of these journals but they are likely to use the same journals as a major source in literature searches for cross-disciplinary research.
Scholars in traditional business disciplines such as finance, management or accounting, provide some justification for the use of the variable firm age, but our sense of its use is that of mimetic isomorphism rather than a theory-driven ne- 35 For example, the history of Shell says more about the owners, managers and their vision than about the age of the company. cessity for its inclusion. Indeed, scholars in empirical research in various branches of management routinely use the said variable without providing much justification for the use of the variable. This paper is intended to speak to any researchers who include a measure of firm age simply because they learned in graduate school that one needed to include the years that a firm has been in existence or listed on a stock exchange because it has some mystical importance related to firm rigidity. Thus our real purpose is to have researchers ask a basic question: Why do we include firm age as a variable when our measure for it probably has little relationship to the theoretical idea we are trying to capture?
The scholarly writers of management, strategic management, and organizational behavior should ideally stop using the variable firm age unless they can provide a conceptual or theoretical foundation for it. Their use of the variable is all too often predicated on flimsy grounds that this researcher or that researcher has used the variable as a control variable. The clear-cut implication of the foregoing is that the age of the firm becomes less informative as an explanatory variable for the behavior of the firm and redundant as a control variable. Given this conclusion we posit that the best way to conceptualize a firm is to consider it a dynamic system (contrasting from a static system) which evolves in response to internal and external stimuli. The factors influencing a firm's behavior are different at different periods of time and there is unlikely to be a single dominant factor constant through the firm's evolution. The casual, perfunctory use of the variable in empirical studies is decried because it diminishes the value and insights of the studies. If parsimony is as valued as researchers suggest, the inclusion of the firm age variable, as measured, is a distraction from the theoretical and empirical insights that researchers are offering. As demonstrated earlier, no firm is able to continue to exist on a single, historically-established form. If the firm does not adapt, it will perish. Even the leadership changes bring about changes in the behavior of the firm.
The future direction of research in this area is easy to point, and yet difficult to travel. A promising start would be for each discipline to undertake a meta-analysis. A meta-analysis would serve at expounding on at least two functions: 1) Statistical tests of significance and effect size and 2) Nature and role of the variable. Such research would provide a conceptual foundation for further research.