This article was downloaded by: [Memorial University of Newfoundland] On: 26 January 2015, At: 12:43 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Total Quality Management Publication details, including instructions for authors and subscription information: i nformation: http://www.tandfonline.com/loi/ctqm19
ISO 9000 certification: The financial performance implications Victor B. Wayhan , Elias T. Kirche & Basheer M. Khumawala Published online: 25 Aug 2010.
To cite this article: Victor article: Victor B. Wayhan , Elias T. Kirche & Basheer M. Khumawala (2002) ISO 9000 certification: The financial financial performance implications, implications, Total Total Quality Qua lity Management, 13:2, 217-231, DOI: 10.1080/09544120120102450 To link to this article: http://dx.doi.org/10.1080/09544120120102450
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor Taylor & Francis. The accuracy of the Content should not be relied upon and upon and should be independently verified with verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, supply, or distribution in any form to anyone is
expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
TOTAL QUALITY MANAGEMENT, VOL. 13, NO. 2, 2002, 217 - 231
ISO 9000 certi®cation: The ®nancial performance implications
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
V ictor B. W ayhan, E lias T . K irche & Basheer M . K humawala Department of Decision and Information Sciences, College of Business Administration, University of Houston, Houston, TX 77204-6282, USA
ISO 9000 certi®cation has become an increasingly popular option for US ®r ms seeking to improve their internal operations and competitive positions. The current study explores the relationship between ISO 9000 certi®cation and ®nancial performance. Most research in this area has been anecdotal in nature. The empirical studies that have addressed this relationship have not only been limited, but also have been largely contradictory. In an attempt to reconcile these con¯icting results, a multivariate, repeated measures research design was utilized. Our results indicate that ISO 9000 certi®cation has a very limited impact on ®nancial performance, as measured by return on assets, however this eþect dissipates quickly over time. abstract
Introduction
In an increasingly competitive global economy, long-term success is often contingent on a ®rm’s ability to reorganize aggressively and to improve their operations to match better new environmental contingencies (Hitt et al ., 1991). Often, in their attempts to remain competitive, companies grasp at one management program after another, in the hopes that one will prove eþective. When the latest program tried does not prove to be a panacea, it is soon discarded and eventually branded a management fad. But, hope springs eternal. One of the latest programs to oþer renewed hope to US ®rms facing intense global competition is ISO 9000 (Elmuti, 1996; Joubert, 1998; Rao et al ., 1997). However, much like its predecessors, ISO 9000 has already begun to engender debate over whether it is destined to slip quietly into oblivion, as just another fad, or whether it will become a business institution, having stood the test of time (Lamprecht, 1999; Parr, 1999). As is well known now, ISO 9000 is a set of ®ve international standards that establish procedures and requirements for the management of quality (Elmuti & Kathawala, 1997). This series of documents on quality assurance was written by the International Standards Organization in 1987, primarily to harmonize the plethora of national and international standards that were in eþect at the time (Zhu & Scheuermann, 1999). ISO 9000 certi®cation does not guarantee a quality product, it merely ensures that a company has a set of procedures Correspondence: V. B. Wayhan, Department of Decision and Information Sciences, College of Business Administration, University of Houston, Houston, TX 77204-6282, USA. Tel: (713) 743 4817; Fax: (713) 743 4940; E-mail: vwayhan@ uh.edu
ISSN 0954-4127 print/ISSN 1360-0613 online/02/020217-15 DOI: 10.1080/09544120120102450
© 2002
Taylor & Francis Ltd
218
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
V. B. WAYHAN ET AL.
by which quality is managed (Hill, 1996; Zhu & Scheuermann, 1999). ISO 9000 was ®rst embraced by Europe, which continues to lead the world in new certi®cations. Between 1987 and 1995, ISO 9000 was adopted by 101 countries as their national quality standard and more than 125 000 certi®cates were issued worldwide (Anderson et al ., 1999). In the early years of the program, US companies approached ISO 9000 certi®cation much more cautiously than the international business community, with certi®cations numbering well under 1000 ( Joubert, 1998). However, from 1992 onward, certi®cations in the USA increased dramatically, from approximately 600 in 1992 to almost 22 000 by 1998 (Elmuti, 1996). The prominence that ISO 9000 has recently achieved in the USA is aptly demonstrated by the fact that it now serves as the foundation for several industry-speci®c standards, such as QS-9000 (automotive), AS9000 (aerospace) and TL9000 (telecommunications) (Gupta & Pongetti, 1998; Zuckerman, 1998). This recent meteoric rise of ISO 9000 to prominence in the USA raises an important concern. Is the phenomenal growth of ISO 9000 driven by bene®ts accrued by the participating companies, or is it simply a case of mimetic behaviorЮrms rushing to implement for fear that the program might work for their competitors, placing them at a competitive disadvantage should they hesitate? As with many other popular business programs (i.e. total quality management (TQM) and business process reengineering), empirical research in this area has seriously lagged practitioner interest. To a certain extent, the dearth of empirical work in this area is understandable, given that ISO 9000 is still a new, albeit rapidly rising program in the USA, thus severely limiting the data necessary for conducting meaningful research. Unfortunately, this has left many important research questions concerning ISO 9000 certi®cation largely unexploredÐparamount of which is whether ISO 9000 certi®cation provides a ®rm with a competitive advantage over non-certi®ed companies, as evidenced by superior ®nancial performance (Elmuti, 1996; Hill, 1996; Skrabec et al ., 1997). The current study seeks to ®ll this gap in the ISO 9000 literature by determining whether ISO certi®cation actually results in superior ®nancial performance. This research is carried out by utilizing a doubly (multivariate) repeated measures design (Bergh, 1995). The balance of this paper is organized as follows. The next section provides a review of the relevant literature. The third section provides the hypotheses tested and the associated dependent variables selected. The fourth section outlines the research methodology employed. The ®fth section summarizes the results, while the sixth section discussion the implications of our ®ndings, before the conclusion.
Literature review
Prior ISO 9000 research has been primarily anecdotal in nature. Early survey work focused primarily on two aspects of ISO 9000: (1) the many reasons why ®rms decided to become ISO 9000 certi®ed; and (2) the perceived bene®ts that these ®rms received from being ISO 9000 certi®ed. The results of one such survey indicate that 77% of the respondents cited improved quality as the number one reason for implementing the program, while market advantage was cited by 73.4% can customer expectations by 68.3% (Litsikas, 1997). In general, other survey results echoed the results obtained above (Gupta & Pongetti, 1998; Skrabec et al ., 1997). These ®ndings are not surprising because the lack of ISO certi®cation has increasingly become a major hindrance to US companies attempting to do business in Europe. Therefore, major US multinational companies now view ISO 9000 certi®cation as not only a means to improve their quality levels, but also as a threshold that must be crossed to do business in Europe.
ISO 9000 CERTIFICATION
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
219
A few other surveys have attempted to ascertain the perceived bene®ts of implementing ISO 9000. In a review of several of these surveys, Elmuti (1996) notes that a British government survey indicates that 89% of respondent companies reported greater operational eýciency, 48% reported increased pro®tability, 76% reported improvements in marketing and 26% reported increased sales. The British Standards Institution (a British ISO 9000 Registrar) estimates that registered companies reduce operating costs by an average of 10%. In a Grant Thorton survey, also reviewed by Elmuti, US manufacturers cited increased international competitiveness, improved quality and a strategic advantage over non-certi®ed competitors as the primary bene®ts of ISO 9000. A survey conducted by Quality Systems Update and Dun & Bradstreet indicates that approximately 95% of the responding companies reported internal bene®ts from implementing ISO 9000, while approximately 85% reported external bene®ts. The top external bene®ts were: higher perceived quality (83%); competitive advantage (70%); reduced customer quality audits (56%); and improved customer demand (29%). The top internal bene®ts were: better documentation (88%); greater quality awareness by employees (83%); enhanced internal communication (53%); and increased operational eýciency (40%) (Sissell, 1996). In general, the surveys done in this area have almost uniformly found that ISO 9000 certi®cation has a positive impact on operational performance across a broad array of measures. In addition, the respondents believe emphatically that certi®cation will eventually lead to a competitive advantage over non-certi®ed companies (Skrabec et al ., 1997). Recent empirical work on why ®rms seek ISO 9000 certi®cation found that managers obtain certi®cation primarily to provide a credible signal of quality assurance to external parties. For the most part, complying with customer or regulatory requirements was found to be a secondary consideration (Anderson et al ., 1999). We are aware of only two published empirical studies to date which speci®cally address the ®nancial performance implications of ISO 9000 certi®cation and only one of these utilizes US ®rms. Terziovski et al . (1997) studied the relationship between ISO 9000 certi®cation and company performance in the presence or absence of a TQM program. The authors assessed diþerences between a large panel (n 5 858) of certi®ed and non-certi®ed companies from Australia and New Zealand across 13 ®nancial and operational measures. Terziovski et al . found no signi®cant relationship between ISO certi®cation and these performance measures, except for an improvement in cash ¯ow. However, the research was limited by the cross-sectional nature of the analysis and the fact that the data were self-reported by the companies instead of being obtained from archival sources (Podsakoþ & Organ, 1986). As part of their recommendations for future research, the authors suggested that a longitudinal research design would be better suited for determining whether a causal link exists between ISO 9000 certi®cation and ®nancial performance. Puderbach and Brown (1998) is the only study that focused on how ISO 9000 certi®cation impacts the subsequent ®nancial performance of US manufacturers. Unfortunately, this study had a very small sample size ( n 5 11), perhaps compromising the generalizability of the results. The authors tested whether ISO 9000 certi®cation has any impact on net income and revenue growth rates. They found that mean revenue and mean net income improved at a faster rate after certi®cation than before certi®cation. The researchers did employ a longitudinal research design (pre-test/post-test), utilizing Spearman’s rho to decide whether the pre-certi®cation and post-certi®cation trends were signi®cantly diþerent. However, the research was limited by the fact that they did not test their hypotheses against a control sample (as was done in the Terziovski et al . study). Since the sample ®rms were certi®ed during an economic recession (1991-92), the failure to use a control sample makes it diýcult to determine whether the trend diþerences found in the study were due to ISO 9000 certi®cation or macroeconomic forces (i.e. the US economy emerging from a recession).
220
V. B. WAYHAN ET AL.
Formulation of hypotheses
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
The ®nancial performance of an organization is most often viewed as a multidimensional construct (Law et al . , 1998), comprised of several distinct, but interrelated dimensions (Venkatraman & Ramanujam, 1986). Although the construct space has not been de®nitely speci®ed, two of its dimensions appear to have the requisite support of prior theory and research to warrant inclusion in the construct space at this time. These two performance dimensions are ®nancial growth and pro®tability (Campbell et al ., 1974; Murphy et al ., 1996; Steers, 1975). Each of these performance dimensions is not only viewed as a fundamental economic goal of any pro®t-seeking business organization (Mott, 1972), but has also served as a criterion in empirical studies (Aaker & Jacobson, 1987; Eisenhardt & Schoonhoven, 1990).
ISO 9000 and ®nancial growth
From a normative perspective, determining the direction that a company should take is a primary function of the strategic management process. The most popular directional strategies are those designed to achieve corporate ®nancial growth, whether measured by variables such as revenue, market capitalization and stockholders’ equity. Two variablesÐgrowth in revenue ( REV ) and stockholder equity (STEQUITY )Ðwere chosen to tap into the ®nancial growth domain of the ®nancial performance construct. Growth in revenue was chosen as a measure because: (1) it was used in both Terziovski et al . (1997) and Puderbach and Brown (1998); (2) it is a prevalent measure in the ®nancial performance literature (Venkatraman & Ramanujam, 1986); and (3) it is a fundamental objective in competitive strategic management (Porter, 1980). ISO 9000 certi®cation is posited to improve revenue growth in a number of ways (Lorge, 1998): (a) First, it appears that many companies are beginning to insist that their suppliers be ISO 9000 certi®ed as part of their ongoing supplier certi®cation programs (Curkovic & Hand®eld, 1996; Elmuti & Kathawala, 1997; Zucherman, 1998). Thus, ®rms that obtain ISO 9000 certi®cation are expected to have a competitive advantage over their non-certi®ed rivals, particularly as ®rms reduce the number of vendors in their supply chains (Elmuti, 1996). (b) Second, ISO 9000 certi®cation is viewed as an independent aýrmation that a particular company has an established quality management program in place. Although this certi®cation does not guarantee a high-quality product (Zhu & Scheuermann, 1999), better quality management procedures are expected to lead to higher quality on average, all things being equal. This higher quality then is expected to lead to enhanced customer satisfaction and loyalty ( Joubert, 1998). The net result should be increased revenue growth from both existing and new business sources (Anderson et al ., 1999; Elmuti, 1996). (c) Third, ISO 9000 certi®cation may allow US ®rms to penetrate certain markets and sectors more successfully, particularly where certi®cation is viewed as a minimum entrance requirement (Barnes, 1998; Goodman, 1998). Growth in stockholder equity was chosen as a measure because it can indicate whether the strategies of top managers are building the value of the ®rm (Brealey et al ., 1995). As agents for stockholders, top managers of for-pro®t corporations should seek to implement strategies that increase the value of the ®rm and the wealth of these important stakeholders. Although the interests of top managers and their shareholders sometimes con¯ict (Fama, 1980; Jensen
ISO 9000 CERTIFICATION
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
221
& Meckling, 1976), this agency problem (Eisenhardt, 1989) may have been partially addressed by the recent proliferation of stock options and incentive-based contracts for top managers. By aligning the ®nancial interests of top management with those of shareholders in this manner, top managers should now have a vested interest in maximizing shareholder wealth. ISO 9000 certi®cation is expected to enhance shareholder wealth through pro®table revenue growth that increases the ®rm’s retained earnings. The ability of a ®rm to generate additional ®nancial resources internally can then be used to ®nance continued expansion. This can reduce the ®rm’s reliance on external debt (liability) to fund capital expenditures (assets), which should all eventually lead to an increase in shareholder equity. In addition, improved ®nancial performance, in general should improve the value of a ®rm’s common stock, thus increasing stockholder equity as well. The relationship between ISO 9000 certi®cation and ®nancial growth, as measured by both growth in revenue and stockholder equity, is tested formally through the following hypotheses: H1a: ISO 9000 certi®cation will have a positive impact on ®nancial g rowth, as measured by revenue growth. H1b: ISO 9000 certi®cation will have a positive impact on ®nancial growth, as measured by growth in stockholder equity. ISO 9000 and pro®tability
The theory of the ®rmÐthe most basic model of businessÐposits that ®rms have pro®t maximization as their primary objective. Although this model is overly simplistic and has been modi®ed or replaced in the ®nancial and economic literature by other theories, the fact remains that sustained pro®tability is an important indicator of a healthy and well-run ®rm. Two variablesÐgross pro®t (GRPROFIT ) and return on assets ( ROA)Ðwere chosen to tap into the pro®tability domain of the ®nancial performance construct. Gross pro®t was chosen as a dependent variable because it directly measures the cash generated by operations before selling, administrative and tax-related expenses are deducted (cf. Hendricks & Singhal, 1997). Thus, this measure should re¯ect the ability of the ®rm to reduce the direct costs associated with producing their products (cost of goods sold). ROA was chosen as a dependent variable because it should partially re¯ect (through net income) the interaction between revenue growth and cost reductions across the entire expense category. In addition, this variable should indicate how eýciently a ®rm uses its assets to generate pro®ts. Finally, ROA is one of the most widely used measures in the literature for assessing the ®nancial impact of various corporate strategies (Bergh, 1995; Hoskisson et al ., 1994; Robbins & Wiersema, 1995). ISO 9000 certi®cation is expected to impact ®rm pro®tability in several ways. An eþectively run quality management program, with its anticipated process improvements, is expected eventually to lead to reduced defects, rework and waste (Anderson et al ., 1999). Second, if ISO 9000 certi®cation enhances sales growth as anticipated, this will allow the company to take advantage of economies of scale and experience curve eþects, which can lead to a reduction in per-unit costs of products sold. These, in turn, can lead directly to increased pro®tability (Barney, 1997). Finally, as was noted earlier, ISO 9000 certi®cation is expected to lead to internal bene®ts (e.g. enhanced communication between employees) and external bene®ts (e.g. reduced number of customer audits) that can reduce the ®rm’s sales and administrative expenses, thus improving the ®rm’s pro®tability. The relationship between ISO 9000 certi®cation and ®rm pro®tability, as measured by both gross pro®t and ROA, is tested formally through the following hypotheses:
222
V. B. WAYHAN ET AL.
H2a: ISO 9000 certi®cation will have a positive impact on pro®tability, as measured by gross pro®t. H2b: ISO 9000 cer ti®cation will have a positive impact o n pro®tability, as measured by return on assets.
Research methodology ISO 9000 panel selection
The test sample of ISO 9000-certi®ed companies was chosen with extreme care in order to isolate, as much as possible, the impact of ISO 9000 certi®cation on ®nancial performance. 5 Information on ISO 9000 certi®cations was obtained from the 1996 edition of the ISO 9000 1 0 2 Reg istered Company DirectoryÐNor th A mer ica. The initial phase of the selection process began y r by identifying all ®rms/facilities that had received ISO 9000 certi®cation in the years a u 1993 - 94. Because the study relies upon archival ®nancial data, the sample was further n a J reduced, in phase two of the selection process, to those companies that were publicly traded 6 2 and had complete data for the entire time period covered by the study (i.e. from 1990 to 3 4 1998). Given that ISO 9000 certi®cations are awarded for individual production sites and : 2 1 not for the entire corporation (Anderson et al ., 1999), Fortune 500-type companies were t a excluded from the test sample. A second reason for excluding these large ®rms is that they ] d have numerous facilities that can become certi®ed at diþerent times (Anderson et al ., 1999). n a l For these types of ®rms, it would be practically impossible to isolate a de®nitive time when d n u ISO 9000 certi®cation actually took placeÐa prerequisite for longitudinal data analysis. o f w Finally, because these ®rms have such diversi®ed holdings, an attempt to link the ISO 9000 e N certi®cation of a plant with the ®nancial performance of the overall corporation would be f tenuous at best. Thus, the sample set included only small to medium size ®rmsЮrms where o y t an individual ISO 9000 certi®cation would cover a substantial amount of the ®rm’s production i s r capacity and revenue stream. Interestingly, after an initial rush of ISO 9000 certi®cations e v i among large, multinational conglomerates, the second wave of certi®cations have occurred n U among companies classi®ed as small to medium in size, with assets less than US$500 million l a i (Rao et al ., 1997). In fact, studies show that more than 50% of all US companies with ISO r o m 9000 registration have fewer than 500 employees, and 25% have less than 150 (Goodman, e 1998). Our ®nal sample set consisted of 48 companies. This size compares favorably with M [ other matched sample studies (Bergh, 1995). y b d e d a Control sample o l n w This sample set was then matched with a control sample of ®rms. The initial list of potential o D matching ®rms came from the four-digit SIC code of the corresponding panel ®rms, as provided by the FIS-Online database. The ®rms in the panel set represent 29 four-digit SIC codes, 20 three-digit codes and nine two-digit codes. The vast majority of ®rms (43 out of 48) were from the 3200-3800 SIC codes. The lists of potential matching ®rms were then further reduced by excluding ®rms that were not of comparable size as measured by 1993 total assets. Total assets had to be within a factor of three for a ®rm to be considered of comparable size (Hendricks & Singhal, 1997). To be included in the ®nal list, the potential ®rms also had to have ®nancial data covering the entire period of the study. The ®nal criterion for inclusion in the matched sample was that the ®rms, if possible, not be ISO 9000 certi®ed by 31 December 1996 (not in the ISO 9000 directory). The control sample ®rms were then selected randomly from the ®nal list of potential matching ®rms.
ISO 9000 CERTIFICATION
223
Unfortunately, in nine out of the 48 cases (18%), a suitable match could not be found that satis®ed all of the criteria. In seven cases, the four-digit SIC code was not broad enough to generate a suitable match. In ®ve cases, this parameter was loosened to the three-digit SIC code level, and in two cases it was loosened to the two-digit level (Werner & Tosi, 1995). In two additional cases, the potential matches were all certi®ed before 31 December 1996. In these two cases, the ®rm that was ISO 9000 certi®ed last (thus providing the largest competitive diþerential) was selected as the match. In these two cases, the minimum acceptable ISO 9000 certi®cation diþerential was 24 months. This ®gure matches the minimum `known’ diþerential that would exist between a panel ®rm certi®ed in December 1994 (end of the focal period) and a ®rm not found to be certi®ed in the ISO 9000 directory (December 1996). The maximum `known’ diþerential was 48 months ( January 1993- December 1996). In all cases, the matching ®rms were of comparable size at the start of the focal period. In summary, the priorities in this three-step matching process were as follows: (1) an ISO diþerential of at least 24 months; (2) comparable size (within a factor of three), as measured by 1993 total assets; and (3) a matching or related SIC code.
5 1 0 2 y r a u n a J 6 2 Analytical model 3 4 : The analytical model tests for diþerences in the four dependent measures over the time period 2 1 t 1990 - 98, utilizing multivariate analysis of variance (MANOVA) with repeated measures. This a ] multivariate model tests all four of the dependent measures simultaneously (Bergh, 1995), d n while conducting an omnibus test of main and interaction eþects. In addition, the interaction a l d eþects are deconstructed into separate univariate tests, so that the individual eþects of each n u dependent variable, over time, can be speci®ed. The time period of analysis allows the model o f w to consider both pre-certi®cation (1990- 92) and post-certi®cation ®nancial performance e N (1995- 98), when determining longitudinal eþects. A between-groups design was utilized, f o with companies in the test sample (coded 1) paired with the matched sample (coded 0). This y t i between-groups variable is labeled (ISO93 ). The within-groups factor is PERFORM , with s r e the four repeated measures (dependent variables) nested under this factor for each of nine v i time frames (i.e. the dependent variables are tested simultaneously). n U Since the analytical model tests for diþerences in the dependent variables over time l a i r (repeated measures), certain data analysis concerns must be addressed. Although repeated o m measures research designs have grown increasingly popular over the last decade (Bergh, e 1993), many longitudinal studies in the management literature are potentially ¯awed due to M [ this failure to satisfy certain analytical assumptions (Bergh, 1995). In a longitudinal research y b design, the repeated measures are rarely independentÐa critical assumption in regular d e d analysis of variance (ANOVA). This lack of independence precludes proper analysis until a o strict analytical assumptions are met and speci®c techniques are employed (O’Brien & l n w Kaiser, 1985). These analytical assumptions are referred to as the symmetry conditions or o D multisample sphericity (Girden, 1992; Huynh, 1978). When data are found to violate the symmetry conditions, s peci®c analytical techniques must be employed that adjust for this lack of independence. Without the proper adjustment (orthonormalization), in¯ated F -statistics may be produced for both omnibus tests of main eþects and interactions involving the withinsubjects factors (Boik, 1981). Since MANOVA for repeated measures was the principle analytical technique used in this study, the necessary adjustments to account for any lack of multisample sphericity were made automatically. Statistical analysis and results
The mean 1993 total assets for the panel set was US$128 246 and US$111 057 for the control sample, mean diþerences that were not signi®cant (t 5 2 0.46, n.s.). In addition, the
224
V. B. WAYHAN ET AL.
panel set enjoyed, on average, a known competitive advantage of 34 months over the control sample, with respect to ISO 9000 certi®cation. Before the data were analyzed and interpreted, the analytical assumptions associated with longitudinal data research were tested. The Box M homogeneity test for between-subjects (W 53160.97563, v 2 51934.48436 , df 5666, p< 0.001) showed that the variance homogeneity assumption had indeed been violated. MANOVA for repeated measures was employed to orthonormalize the data to allow proper analysis. Table 1 reports mean diþerences between the two groups, for all four dependent variables, across the nine time periods. Omnibus test and interaction eþects
5 Table 2 reports the results of the omnibus test. The omnibus test indicates that the main 1 0 eþect of ISO certi®cation (ISO93 ) was insigni®cant over the entire nine time periods 2 y (F 5 0.62865, p 50.643). In addition, the eþect size (multivariate) of ( ISO93 ) was extremely r a u small (0.027). However, the power of the model was also very small (0.20). PERFORM , the n a J second main eþect, was signi®cant over time, although this is expected of most ®nancial 6 2 performance data and is of little theoretical importance (F 52.48737, p< 0.001). The 3 interaction eþect (PERFORM 3 ISO93 ) was not signi®cant ( F 50.74545, p 5 0.816), indicat 4 : 2 ing that contrasts between cell means did not diþer across any of the four dependent variables 1 t a (Girden, 1992). ] d n a l Post hoc power analysis d n u o Whenever non-signi®cant results are obtained in a study, it is appropriate to question the f w ability (power) of the model to detect eþects (Hoskisson et al ., 1994). Although the model e N has a reasonable sample size (n 596), it is readily apparent that the model does not have f o suýcient power to detect an eþect size this small. Table 3 provides a summary of the post y t i hoc power analysis, indicating the results that would be obtained if the original sample size s r e were increased to enhance the power of the model. When the original sample size is doubled v i n to 192, the interaction eþect becomes signi®cant (F 5 1.881, p < 0.05). A review of the U l univariate tests indicates that ROA is driving much of the post hoc results, being the only a i r statistically signi®cant dependent measure (F 52.00976, p< 0.05) in the study. When the o m original sample size is tripled to 288, the only change in the results is that the ROA variables e becomes statistically signi®cant at a higher level (F 5 3.02521, p< 0.01). Finally, when the M [ y original sample size is increased by a factor of four ( n 5384), the main eþect of (ISO93 ) b d becomes signi®cant (F 52.618, p< 0.01). However, a review of the univariate tests indicates e d that the ROA measure is still the only dependent variable that is signi®cant ( F 5 4.04067, a o l n p < 0.01). Thus, the three dependent measuresÐgrowth in revenue, stockholder equity and w gross pro®tÐappear to have no practical signi®cance, as well as no statistical signi®cance, o D whereas ROA is both practically signi®cant and statistically signi®cant at a reasonable sample size of (n 5 192) and power level of (0.40). Given that ISO 9000 certi®cation is statistically linked to ROA , given a modest increase in sample size and power level, a second `custom’ model (Norusis/SPSS, 1994, p. 140) was employed, utilizing a simple eþects analysis, to specify the form and stability of the empirical relationship over time. This analysis provides a test for diþerences in ROA between the two groups, for each of the nine time periods, in a series of one-degree-of-freedom tests (O’Brien & Kaiser, 1985). Table 1 indicates that the mean diþerences between the two groups approached signi®cance (F 52.61, p 5 0.109) in 1991, and became signi®cant in 1992 (F 5 3.26, p 5 0.074) and 1993 (F 54.84, p 50.03). However, these statistically signi®cant diþerences dissipated in 1994 and were not signi®cant in subsequent years.
ISO 9000 CERTIFICATION
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
s e c n e r e þ i d n a e m : e c n a m r o f r e p l a i c n a n ® d n a n o i t a c ® i t r e c 0 0 0 9 O S I . 1 e l b a T
n o i t a c ® i t r e c t s o P
n o i t a c ® i t r e c e r P
8 9 9 1
6 8 2 2 6 4 0 . . 3 . 3 6 6 0 6 8 2 3 3 2
8 6 2 4 3 1 2 . . 5 . 7 3 9 3 3 1 1 1 1
3 9 4 7 7 9 9 . 1 . 7 . 3 8 5 9 8
7 9 9 1
5 1 4 5 6 9 8 . 8 . 9 . 6 4 1 3 2 1 3 3
7 4 3 6 6 0 0 . 5 . 5 . 5 3 1 3 0 3 1 1
3 5 8 5 0 4 7 . 5 . 2 . 0 8 2 9 7 1
6 9 9 1
3 8 5 0 2 7 7 . 7 . 9 . 4 2 1 0 8 2 3 2
7 8 9 4 9 4 0 . 5 . 4 . 1 7 3 2 8 3 1
2 9 3 1 3 7 5 . 8 . 6 . 5 0 4 8 7 1
5 9 9 1
4 1 3 2 1 1 7 . . 0 . 7 2 7 5 6 4 1 2 2
2 4 8 0 1 8 6 . 4 . . 1 6 0 6 9 7 2
0 2 8 7 6 0 2 . 0 . . 2 4 3 1 7 6 1
4 9 9 1
1 4 7 4 9 4 6 . 4 . 1 . 4 2 2 1 9 2 2 1
6 8 8 5 7 7 9 . 5 . 3 . 1 9 2 7 4 2
0 7 3 1 5 5 9 . 4 . 4 . 1 0 1 6 5 1
3 9 9 1
3 9 4 6 5 0 3 . 9 . 4 . 4 2 1 8 6 2 1 1
2 2 2 8 6 5 5 . 3 . 2 . 5 3 2 5 3 2
1 1 0 4 1 3 8 . 3 . 5 . 2 3 9 5 4
2 9 9 1
7 7 0 7 1 6 1 . 4 . 7 . 3 3 9 5 3 1 1 1
2 2 0 7 6 1 4 . 3 . 1 . 2 1 1 4 3 1
2 7 5 8 1 6 1 . 3 . 8 . 5 8 6 4 3
1 9 9 1
3 4 9 5 3 1 0 . . 8 . 2 7 3 3 3 2 1 1 1
8 2 6 5 8 7 4 . 1 . . 2 1 0 1 4 2 2
4 6 8 7 4 2 6 . 8 . 8 . 2 3 8 4 3
0 9 9 1
5 9 6 2 6 5 1 . 0 . 0 . 9 9 0 1 1 1 1
3 4 9 7 6 0 2 . 2 . 0 . 3 7 6 3 1 1
2 5 7 6 1 4 6 . 4 . 2 . 6 4 2 3 3
e l b a i r a V
3 9 O S I
7 5 2 9 4 4 6 8 5 4 . 4 . 9 . 2 1 3
1 3 4 . 0
6 7 9 5 8 6 7 2 4 0 . 9 . 1 . 2 0 1
2 5 8 . 0
2 3 9 0 8 1 9 8 0 7 . 3 . 4 . 5 1 4
4 7 3 . 0
2 7 5 6 7 8 9 5 3 8 . 9 . 9 . 5 3 1
6 4 5 . 0
5 6 9 3 6 6 3 9 3 5 . 5 . 9 . 6 3 2
1 4 3 . 0
7 8 5 3 1 5 4 3 7 6 . 1 . 7 . 8 1 9
0 3 0 . 0
5 0 5 4 2 2 2 0 2 2 . 3 . 9 . 7 2 4
4 7 0 . 0
5 1 4 3 3 0 9 5 4 4 . 9 . 5 . 5 0 4
9 0 1 . 0
4 9 3 1 3 5 0 7 7 0 . 4 . 4 . 0 4 4
5 5 4 . 0
2
2
2 2 2
2
2
2 2
T Y Y ) ) ) I T I s s T s T n V V n I I n F F ) A A o o U U o O O i i i l l l E E O % R R l R R : i l Q Q : i l ( O i R R : P P : E E s s s s : m m T e ( m R R e e e ( c ( T c c c ) p S S G G n n n n ( e e e e r r r r e e e e e c þ þ þ þ i i i i n . d d d d a c 6 ® n n n n i 9 a a a a n e e e e i g 5 1 0 M 1 0 M 1 0 M 1 0 M S N
225
226
V. B. WAYHAN ET AL.
Table 2. ISO 9000 certi®cation (ISO93) and ®nancial performance multivariate analysis of variance: omnibus test
Source of variation
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
Wilks value
df
Between subjects: ISO93
0.97311
4, 91
Within subjects: PERFORM PERFORM 3 ISO93
0.44181 0.72535
32, 63 32, 63
Univariate results: REV STEQUITY GRPROFIT ROA
F
0.643
2.48737** 0.816
0.05263 0.31870 0.23065 0.99430
N 5 96; eþect size: 0.027; power: 0.20; ** p < 0.01.
Table 3. ISO 9 000 certi®cation (ISO93) and ®nancial performance post h oc power analysis
Sample size: Statistical power: Eþect size:
n 5 192 0.40 0.27
n 5 288 0.59 0.27
n 5 384 0.73 0.27
0.643
1.292
1.955
2.618*
Within subjects: PERFORM PERFORM 3 ISO93
2.487** 0.816
6.278** 1.881**
10.068** 3.017**
13.858** 4.153**
Univariate results: REV STEQUITY GRPROFIT ROA
0.05263 0.31870 0.23065 0.99430
0.10638 0.64418 0.46621 2.00976*
Source of variation Between subjects: ISO93
n 5 96 0.20 0.27
0.16013 0.96966 0.70177 3.02521**
0.21389 1.29514 0.93734 4.04067**
Note: F -statistics are reported; * p< 0.05; ** p< 0.01.
Discussion of results
This study was undertaken in an eþort to test empirically the following research question: Do ®rms that are ISO 9000 certi®ed have a competitive advantage over non-certi®ed companies, as evidenced by superior ®nancial performance? Two hypotheses were tested to determine whether ISO 9000 certi®cation has a positive impact on ®nancial growth (as measured by revenue and stockholder equity) and pro®tability (as measured by gross pro®t and ROA ). Both of the formalized hypotheses [H1a and H1b] are rejected, which implies that ISO 9000 certi®cation does not impact ®nancial growth. The results indicate that during the period of analysis, ISO 9000 certi®cation did not result in additional sales. Although sales did increase for ®rms in the test sample, for most of the period of analysis ®rms in the matched sample largely equaled this increase. This lack of impact in the area of sales improvement may be due to the fact that US companies have only recently begun to require their suppliers to be ISO 9000 certi®ed, which would tend to
ISO 9000 CERTIFICATION
227
reduce the eþect s ize of certi®cation. In addition, the panel set was comprised of small to medium size companies that may be less likely than large multinational ®rms to export to EuropeÐwhere supplier restrictions concerning ISO 9000 certi®cation are often the most stringent. Without the in¯uence of these restrictive European supplier requirements, new sales gained at the expense of non-certi®ed companies are expected to be minimal. Continued tightening of supplier requirements in the USA and increased export activity may eventually cause ISO 9000 certi®cation to be a source of competitive advantage with respect to revenue growth. The results also indicate that companies desiring to implement strategies that will materially enhance the value of the ®rm and the wealth of stockholders should search elsewhere. Given that ISO 9000 certi®cation did not enhance pro®table revenue growth, the 5 1 ®nding that stockholder equity was also not signi®cantly impacted is not unexpected. This 0 2 lack of impact on stockholder equity may also indicate that the stock-market places very little y r a value (as re¯ected in the common stock component of stockholder equity) on ISO 9000 u n certi®cation. a J The results indicate that ISO 9000 has a positive impact on pro®tability, however this 6 2 link is somewhat tenuous. ISO 9000 did not impact gross pro®t, but was positively linked to 3 4 : 2 ROA , after a modest increase in sample size and power level. Therefore, hypothesis two 1 receives only partial support (H2a is rejected and H2b is accepted). Apparently, the costs t a ] directly associated with a company’s operations were not impacted by ISO 9000 certi®cation, d n given that mean diþerences in pro®tability, at the gross pro®t level, were not statistically a l d signi®cant throughout the period of analysis. This may suggest that the improvements in n u quality management record keeping, due to ISO 9000 certi®cation, have not yet translated o f w into concrete improvements in actual operations that would result in a lower cost structure e N (Barnes, 1998). f o The signi®cant link between ISO 9000 and ROA suggests that certi®ed companies y t i improved pro®tability by reducing indirect costs, primarily in the sales and administrative s r e expense categories, which were re¯ected in the increased diþerences in net income. Interest v i n ingly, the subsequent test of simple contrasts indicates that ISO 9000 had a positive impact U on mean diþerences in ROA starting in approximately 1991. Since obtaining ISO 9000 l a i r certi®cation c an take anywhere from 9 to 28 months of preparation before cer ti®cation is o m achieved (Anderson et al ., 1999), these improvements in ROA mean diþerences appear to e have begun at approximately the time when the certi®ed companies began their initial M [ y preparation work. These positive bene®ts, however, appear to have dissipated quickly, at b approximately the time when ISO 9000 certi®cation was achieved (see Table 1). In fact, the d e d positive mean diþerences in ROA eventually turned negative by 1997, although these last a o l two diþerences were not statistically signi®cant (a disquieting trend none the less). n w One of the chief criticisms of ISO 9000 is that once certi®cation is achieved, the program o D provides little or no motivation for additional improvement (Anderson et al ., 1999). Unlike TQM and other quality management programs, ISO 9000 does not focus on continuous improvement, but instead focuses on achieving certi®cation ( Joubert, 1998; Zhu & Scheuermann, 1999). Thus, ISO 9000 certi®cation may be interpreted by a company as an indication that the ®rm has `arrived’, which may lull the company into a false sense of s ecurityÐa mindset that may keep the ®rm from zealously protecting its competitive advantage. Over time, it appears that this small competitive advantage gained during the preparation for certi®cation phase eroded quickly once certi®cation was achieved. Thus, the parabolic orientation of the mean diþerences in the ROA (Fig. 1) may indicate that the ISO 9000 ®rms were satis®ed merely to attain certi®cation, while the ®rms in the matched panel perhaps continuously improved through other means.
228
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
V. B. WAYHAN ET AL.
Figure 1. Mean diþerences (ROA).
Link with previous research studies
As discussed in the literature review section, the only two empirical studies to investigate this speci®c research question arrived at contradictory conclusions, perhaps due to methodological diþerences between the two studies and the limitations inherent in both research designs. We utilized a doubly multivariate, repeated measures, research design in our research to address these limitations and to provide a robust, longitudinal test of the hypotheses. Our results do not provide support for the Puderbach and Brown (1998) study that concluded that ISO 9000 certi®cation improved revenue growth. This may suggest that the lack of a control group and the fortuitous choice of a focal period (the recession of 1991-92) contributed substantially to the positive link found between ISO 9000 certi®cation and revenue growth in the Puderbach and Brown (1998) work. The second variable in the Puderbach and Brown study (net income) was not tested directly, although it was the numerator in the ROA ratio employed in the current study. Given that we have found ISO 9000 certi®cation to be statistically linked to ROA , at least in the short run, this result may provide tentative support for the Puderbach and Brown study with respect to this particular variable. Our results appear to provide support for the study of Terziovski et al . (1997), which found that ISO 9000 did not provide certi®ed companies with a competitive advantage, as evidenced by superior ®nancial performance, across a broad array of ®nancial performance measures. The only exception was in the case of increased cash ¯owÐa ®nding that our study did not support. The common conclusion drawn from the Terziovski et al . study and our own work is that there is, at best, a very limited link between ISO 9000 certi®cation and subsequent ®nancial performance. This conclusion, however, raises the troublesome issue of why there is such a disconnect between surveys, which almost uniformly report that ISO 9000 certi®cation has a positive impact on ®nancial performance, and empirical research, which arrives at almost the exact opposite conclusion. This may indicate that the self-
ISO 9000 CERTIFICATION
229
reported data are somewhat biased, perhaps due to the fact that the respondents often have a vested interest in the eventual success of the ISO 9000 initiative (Podsakoþ & Organ, 1986). Future research
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
Given the prevalence of ISO 9000 and the limited amount of research done in the area, additional empirical work is a must. Research should continue to explore the possible link between ISO 9000 certi®cation and other domains of the ®nancial performance construct. Given that our work focused exclusively on accountingÐbased ®nancial performance measures, it would be advisable for a future study to explore the impact of ISO 9000 certi®cation on market-based measures of ®nancial performance such as growth in market capitalization (Venkatraman & Ramanujam, 1986). In addition, researchers should investigate whether ISO 9000 certi®cation impacts non-®nancial measures such as productivity, market share and product quality, particularly since surveys indicate that certi®ed ®rms anticipate bene®ts in these areas ( Joubert, 1998; Litsikas, 1997). Fortunately, the potential link between ISO 9000 and product quality is already drawing the attention of researchers (Rao et al ., 1997). However, more research in this area is needed since critics of the program insist that this link (direct or indirect) is non-existent (Hill, 1996; Joubert, 1998; Skrabec et al ., 1997). Surveys indicate that one of the major bene®ts of ISO 9000 certi®cation is that it can provide an entre e into foreign markets, particularly Europe (Goodman, 1998). Researchers should investigate whether ISO 9000 certi®cation has a positive impact on export sales. This research may have to be conducted at the industry level to reduce potential confounding. An interesting corollary would be to determine empirically whether the export orientation of a particular industry is an indicator of how many ®rms are certi®ed. In addition, researchers could determine whether this level of export orientation moderates the relationship between ISO 9000 certi®cation and ®nancial performance. Yet another potential productive research stream would be to explore the eþects of ISO 9000 on workers at the plant level. The work of Elmuti and Kathawala (1997) in the area of employee attitudes and job performance after implementation of ISO 9000 is an excellent start in this direction. Another welcome addition to the literature is the work by Anderson et al . (1999), which explores the decision-making process behind ISO 9000 implementation. Besides the potential export-orientation eþect noted above, researchers should extend this work by focusing on other potential antecedents of ISO 9000 certi®cation. These could include ®rm-speci®c attributes such as ®rm size and age, market share position and quality reputation (i.e. previous quality awards, if any). In addition, industry-speci®c criteria could be included, such as the level of competition (i.e. number and strength of competitors), the market share held by foreign companies (globalization of the industry) and the technological orientation of the industry. Conclusion
Survey results indicate that ISO 9000-certi®ed companies expected to gain a competitive advantage over their non-certi®ed rivals, as evidenced by superior ®nancial performance. Although the panel set enjoyed a known 34 months head-start on the ®rms in the matched sample, these ®rst mover advantages did not result in the competitive advantages that these companies apparently expected at the onset of implementation. The current study did establish a link between ISO 9000 certi®cation and ROA , but this ®nding must be tempered by the fact that the eþect size was extremely small and that the eþect dissipated quickly. Even
230
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
V. B. WAYHAN ET AL.
more importantly, a link with just one ®nancial performance variable does not establish a competitive advantage. This suggests that ISO 9000 certi®cation should only be implemented when a compelling reason for certi®cation is evident (i.e. doing business in Europe). When this is the case, ISO 9000 should only be viewed as a very limited component of a ®rm’s quality management program, not as a miracle panacea that will remedy all competitive disadvantages that a company must overcome in the market-place. It must be stressed that ISO 9000 certi®cation is not intended to be either a quality or a ®nancial performance improvement program. It simply certi®es the existence and adherence to a set of quality procedures. The test sample’s poor relative ®nancial performance after certi®cation may suggest that these ®rms unwittingly placed more expectations on ISO 9000 than the program was able to bear. References A aker, D. & J acobson, R . (1987) The role of risk in explaining diþerences in pro®tability, Academy of Management Journal , 30, pp. 277-296. A nderson, S.W ., Daly, J .D. & J ohnson, M .F. (1999) Why ®rms seek ISO 9000 certi®cation: Regulatory compliance or competitive advantage?, Production and Operations Management , 8, pp. 28-43. Barnes, F.C. (1998) ISO 9000 myth and reality: a reasonable approach to ISO 9000, SAM Advanced Management Journal , 63, pp. 23- 30. Barney, J . (1997) Gaining and sustaining competitive advantage (Reading, MA, Addison-Wesley). Bergh, D. (1993) Watch the time carefully: the use and misuse of time eþects in management research, Journal of Management , 19, pp. 683- 705. Bergh, D. (1995) Problems with repeated measures analysis: demonstration with a study of the diversi®cation and performance r elationship, Academy of Management Journal , 38, pp. 1692-1708. Boik, R . J . (1981) A priori tests in repeated measures designs: eþects of nonsphericity, Psychometrika , 46, pp. 241- 255. Brealey, R .A ., M yers, S.C. & M arcus, A . J . (1995) Fundamentals of Corporate Finance (New York, McGraw-Hill). Campbell, J ., Brownas, E ., Peterson, N . & Dunnette, M . (1974) The Measurement of Organizational Eþectiveness : A Review of Relevant Research and Opinion (San Diego, CA, Naval Personnel Research Center). Curkovic, S. & Handfield, R . (1996) Use of ISO 9000 and Baldridge Award criteria in supplier quality evaluation, International Journal of Purchasing and Materials Management , 32, pp. 2-11. E isenhardt, K . (1989) Agency theory: an assessment and review, Academy of Management Review, 14, pp. 57-74. E isenhardt, K . & Schoonhoven, C. (1990) Organizational growth: linking founding team, strategy, environment, and growth among US semiconductor ventures, Administrative Science Quarterly, 35, pp. 504- 529. E lmuti, D. (1996) World-class standards for global competitiveness: an overview of ISO 9000, Industrial Management , 38, pp. 5-9. E lmuti, D. & K athawala, Y . (1997) An investigation into the eþects of ISO 9000 on participants’ attitudes and job performance, Production and Inventory Management Journal , 38, pp. 52- 57. Fama, E .F. (1980) Agency problems and the theory of the ®rm, Journal of Law and Economics , 26, pp. 288- 307. G irden, E .R . (1992) ANOVA Repeated Measures (London, Sage University Publications). G oodman, D . (1998) Thinking export? Think ISO 9000, World Trade, 11, pp. 48-49. G upta, P. & Pongetti, D. (1998) Are ISO/QS-9000 c erti®cations worth the time and money?, Quality Progress, 31, pp. 19- 24. Hendricks, K .B. & Singhal, V .R . (1997) Does implementing an eþective TQM program actually improve operating performance? Empirical evidence from ®rms that have won quality awards, Management Science , 43, pp. 1258- 1274. Hill, S. (1996) ISO certi®cation provides long-term payoþ, Quality, 35, pp. 66- 68. Hitt, M .A ., H oskisson, R . & H arrison, J . (1991) Strategic competitiveness in the 1990s: challenges and opportunities for US executives, Academy of Management Executive , 5, pp. 7-22. Hoskisson, R .E ., J ohnson, R .A . & M oesel, D.D. (1994) Corporate divestiture intensity in restructuring ®rms: eþects of governance, strategy, and performance, Academy of Management Journal , 37, pp. 1207- 1251.
ISO 9000 CERTIFICATION
5 1 0 2 y r a u n a J 6 2 3 4 : 2 1 t a ] d n a l d n u o f w e N f o y t i s r e v i n U l a i r o m e M [ y b d e d a o l n w o D
231
Huynh, H. (1978) Some approximate tests for repeated measurement designs, Psychometrika , 43, pp. 161-175. J ensen, M . & M eckling, W . (1976) Theory of the ®rm: managerial behavior, agency costs, and ownership structure, Journal of Financial Economics, 3, pp. 305-360. J oubert, B. (1998) ISO 9000: international quality standards, Production and Inventory Management Journal , 39, pp. 60-65. L amprecht, J . (1999) Is ISO 9000:2000 the beginning of the end?, Quality Progress, 32, pp. 47- 52. L aw, K .S., W ong, C.S. & M obley, W .H. (1998) Toward a taxonomy of multidimensional constructs, Academy of Management Review, 23, pp. 741-755. L itsikas, M . (1997) Companies chose ISO certi®cation for internal bene®ts, Quality, 36, pp. 20- 26. L orge, S. (1998) Can ISO certi®cation boost sales?, Sales and Marketing Management , 150, p. 19. M ott, P. (1972) Characteristics of Organizations (New York, Harper & Row). M urphy, G ., T railer, J . & H ill, R . (1996) Measuring performance in entrepreneurship research, Journal of Business Research, 36, pp. 15- 23. O’Brien, R . & K aiser, M . (1985) MANOVA method for analyzing measures designs: an extensive primer, Psychological Bulletin, 97, pp. 316-333. Parr, G .L . (1999) ISO 9000 has staying power, Quality, 38, p. 6. Podsakoff, P.M . & Organ, D.W . (1986) Self-reports in organizational research: problems and prospects, Journal Management , 12, pp. 531-544. Porter, M .E . (1980) Competitive Strategy: Techniques for Analyzing Industries and Competitors (New York, The Free Press). Puderbach, L . & Brown, K . (1998) Financial eþects of I SO 9000 cer ti®cation in American corporations, Proceeding of the Decision Science Institute , Las Vegas. R ao, S.S., R agu-N athan, T .S. & Solis, L .E . (1997) Does ISO 9000 have an eþect on quality management practices? An international empirical study, Total Quality Management , 8, pp. 335- 346. R obbins, J . & W iersema, M .F. (1995) A resource-based approach to the multibusiness ®rm: empirical analysis of portfolio interrelationships and corporate ®nancial performance, Strategic Management Journal , 16, pp. 277- 299. Sissell, K . (1996) Survey rates ISO 9000 success, Chemical Week, 150, p. 33. Skrabec, Q .R ., R agu-N athan, T .S., R ao, S.S. & Bhatt, B.T . (1997) ISO 9000: Do the bene®ts outweigh the costs?, Industrial Management , 39, pp. 26-32. Steers, R . (1975) Problems in the measurement of organizational eþectiveness, Administrative Science Quarterly, 20, pp. 546-558. T erziovski, M ., Samson, D. & Dow, D. (1997) The business value of quality management systems certi®cation: evidence from Australia and New Zealand, Journal of Operations Management , 15, pp. 1-18. V enkatraman, N . & R a manujam, V . (1986) Measurement of business performance in strategy research: a comparison of approaches, Academy of Management Review , 11, pp. 801- 814. W erner, S. & T osi, H.L . (1995) Other people’s money: the eþects of ownership on compensation strategy and managerial pay, Academy of Management Journal , 38, pp. 1672- 1691. Zhu, Z. & Scheuermann, L . (1999) A comparison of quality programs: total quality management and ISO 9000, Total Quality Management , 10, pp. 291-297. Zuckerman, A . (1998) GM enforces QS-9000 mandate, Purchasing , 125, pp. 131-132.