JOURNAL OF - jbm.johogo.comjbm.johogo.com/pdf/volume/0401/JBM-0401.doc · Web viewIt is essential to prevent submitting a boring, lackluster proposal to the government. Well-designed

FINANCE/ACCOUNTING

MANAGEMENT/MARKETING

QUANTITATIVE METHODS

Summer 1997 Vol. 4, No. 1

Delvin D. Hawley The Valuation of Exchangeable Securities: A

O. Felix Ayadi Pedagogic Approach

Joseph C. Montgomery Developing Customer-Centered Performance

David J. Lemak Measures Richard Reed

Robert S. Frey Effective Small Business Response Strategies to Federal Government Competitive Procurements

Golnaz Sadri Asian-Americans and Caucasians In the Workplace: Some Attitudinal Comparisons

Fred Glover An Integer Programming Approach to the Minimum

Ching-Chung Kuo Diversity Problem Krishna S. Dhir

JOURNAL OF BUSINESS AND MANAGEMENT

Published jointly by the Western Decision Sciences Institute and the School of Management, California State University, Dominguez Hills

JOURNAL OF BUSINESS AND MANAGEMENTTHE OFFICIAL PUBLICATION OF THE

WESTERN DECISION SCIENCES INSTITUTE (WDSI)

The Decision Sciences Institute is a professional society dedicated to the development and application of quantitative and behavioral methods to administrative problems. Most functional areas of business are represented among the membership. Through its journals, national and regional meetings, and other activities, the Decision Sciences Institute serves as a vehicle to advance and disseminate the theory, application, pedagogy, and curriculum development of the decision sciences.

Western Regional Officers 1996-97President, Thomas E. Callarman, Arizona State UniversityPresident-Elect, Richard L. Jensen, Utah State UniversityProgram Chair/Vice President for Programs/Proceedings Editor, Karen L. Fowler, University of Northern ColoradoVice President for Programs-Elect, Marc F. Massoud, Claremont McKenna CollegeVice President for Member Services, Paul Mallette, Colorado State UniversityVice President for Special Arrangements, Eldon Y. Li, California Polytechnic State University, San Luis ObispoSecretary/Treasurer, George A. Marcoulides, California State University, Fullerton

JOURNAL OF BUSINESS

AND MANAGEMENT

Vol. 4 , No. 1 Summer 1997

EDITORS

EDITORIAL ASSISTANT

Franklin StrierBurhan F. YavasPatty Ramirez

Editorial Offices:

JOURNAL OF BUSINESS AND MANAGEMENTSchool of ManagementCalifornia State University, Dominguez Hills1000 East Victoria StreetCarson, California 90747Phone: (310) 243-3472, (310) 243-3501Fax: (310) 516-3664, (310) 217-6964

Published jointly by Western Decision Sciences Institute (WDSI) and the School of Management, California State University, Dominguez Hills. The purpose of the JOURNAL OF BUSINESS AND MANAGEMENT is to provide a forum for the dissemination of contributions in all fields of business, management and related public policy of relevance to academics and practitioners. Original research, reports and opinion pieces are welcome. The style should emphasize exposition and clarity, and avoid technical detail and jargon.

The views expressed in articles published are those of the authors and not necessarily those of the Editors, Executive Board, Editorial Board, WDSI or California State University, Dominguez Hills. All submissions will be reviewed initially by the editors and, if judged appropriate, will be sent to knowledgeable referees for review. The authors assume responsibility for the accuracy of facts published in the articles.

Copyright 1997 WDSI and by the School of Management, California State University, Dominguez Hills. Subscriptions are $16/year. Manuscripts should be double-spaced and submitted in triplicate. Manuscripts and comments should be directed to the editors.


Executive Board Thomas E. Callarman, President, WDSI Richard L. Jensen, President-Elect, WDSI Yoram Neumann, Vice President, Academic Affairs, California State University, Dominguez Hills Franklin Strier, Editor Burhan F. Yavas, Editor

Editorial Board Dr. Joseph R. Biggs California Polytechnic State University, San Luis Obispo Dr. Henry Brehm University of Maryland Dr. Terry E. Dielman Texas Christian University Dr. Moshe Hagigi Boston University Dr. Ronald H. Heck University of Hawaii at Manoa Dr. Richard C. Hoffman Salisbury State University, Maryland Dr. Marc T. Jones University of Otago, Dunedin, New Zealand Dr. Erdener Kaynak Pennsylvania State University Dr. Thomas Kelly State University of New York, Binghamton Dr. George R. LaNoue University of Maryland Dr. George A. Marcoulides California State University, Fullerton Dr. John Preble University of Delaware Dr. Arie Reichel Ben-Gurion University of the Negev, Israel Dr. Elizabeth L. Rose University of Auckland, New Zealand Dr. Anne S. Tsui The Hong Kong University of Science and Technology, Hong Kong Dr. Michael Useem University of Pennsylvania

Reviewer Acknowledgments

The editors of the Journal of Business and Management wish to express their appreciation to the following individuals who have reviewed manuscripts submitted for consideration in this issue of the Journal of Business and Management.

Dr. Harvey ArbelaezDr. Catherine AtwongDr. Nizamettin AydinDr. Laurence BartonDr. Gabriel BassiryDr. Hamdi BiliciDr. Edward ChuDr. Sadik CokelezDr. Gregory A. DanekeDr. Nejdet DelenerDr. Prakash DheeriyaDr. Bob DowlingDr. Tammy DreznerDr. Zvi DreznerDr. Stephen P. FerrisDr. Dorothy FisherDr. Stephen FisherDr. Chic FojtikDr. Karen FowlerDr. Glenn FreedDr. Cristina GibsonDr. Zvi GoldsteinDr. James HeathDr. Ronald H. HeckDr. Veronica HortonDr. Swinder JandaDr. Stephen JennerDr. John KarayanDr. W. Fred KiesnerDr. Bob KoesterDr. Ching-Chung KuoDr. Craig C. LundberghDr. Yadong Luo

Dr. Richard MalamudDr. George A. MarcoulidesDr. James MartinoffDr. Bryant MillsDr. Isaac D. MontoyaDr. Khosrow MoshirvaziriDr. Edith NeumannDr. Ali M. ParhizgariDr. Cynthia PavettDr. Elizabeth RoseDr. Golnaz SadriDr. Donna ScottDr. Mark G. SimkinDr. Vernon StaubleDr. Elizabeth TrybusDr. Barbara WithersDr. Kosaku Yoshida


TABLE OF CONTENTS

From the Editor's Desk............................................................................7

The Valuation of Exchangeable Securities: A Pedagogic ApproachDelvin D. Hawley, O. Felix Ayadi...................................................... 9

Developing Customer-Centered Performance MeasuresJoseph C. Montgomery, David J. Lemak, Richard Reed....................24

Effective Small Business Response Strategies to Federal Government Competitive Procurements

Robert S. Frey...................................................................................40

Asian-Americans and Caucasians In the Workplace: Some Attitudinal Comparisons

Golnaz Sadri.....................................................................................75

An Integer Programming Approach to the Minimum Diversity ProblemFred Glover, Ching-Chung Kuo, Krishna S. Dhir.............................93

FROM THE EDITOR'S DESK

DELVIN D. HAWLEY and O. FELIX AYADI investigate the valuation of exchangeable securities by applying a variant of the Black-Scholes option pricing model. The results indicate that the market prices of these securities differ from their theoretical values, particularly at the time of issue. However, the deviation tends to shrink with time.

The measurement of organizational performance is a key issue in organizational effectiveness. JOSEPH C. MONTGOMERY, DAVID J. LEMAK and RICHARD REED discuss the recent shift away from the traditional manager-centered model to a customer-centered approach to measuring performance. The authors develop a model of customer-centered orientation and show how such a model would help improve organizational efficiency and effectiveness.

What are the critical success factors for small companies in the federal government competitive procurement process? ROBERT S. FREY emphasizes the importance of practical information and proven processes needed to plan, organize and prepare effective proposals.

Todays work force is more culturally diverse than ever before. The reality of working in multi-cultural environments underscores the need to understand how such diversity affects organizational goals. GOLNAZ SADRI studies two cultural groups, Asian-Americans and Caucasians, in an effort to assess differences in two important cognitions: locus of control and job satisfaction. Asian-Americans showed a more external locus of control and a lower level of job satisfaction than the Caucasian group.

In diversifying their portfolios to minimize overall risk, investors try to select securities that have low covariances. The minimum diversity problem, on the other hand, is one in which the emphasis is on selecting a set of elements exhibiting the smallest variation. FRED GLOVER, CHING-CHUNG KUO and KRISHNA S. DHIR formulate a number of

8

integer programs to model the minimum diversity problem. The complexity of the problem provides a motivation for the development of special procedures. The authors develop four heuristic approaches as a beginning and submit them for empirical analysis.

FRANKLIN STRIER

BURHAN F. YAVAS

9

THE VALUATION OF EXCHANGEABLE SECURITIES:

A PEDAGOGIC APPROACH

Delvin D. Hawley *

O. Felix Ayadi **

This study applies a variant of the Black-Scholes option pricing model to the valuation of exchangeable securities. An exchangeable security is a new hybrid bond which grants its holder the right but not the obligation to exchange the bond for the common stock of a firm other than the issuer. The results indicate that market prices of these securities deviate from their respective theoretical values but the deviation declines with time. This evidence supports the underpricing of initial public offering of securities already documented in the financial economics literature.

he securities market has in recent time been witnessing many developments in terms of the different packages of securities that are available to investors. In the bonds market for example, the familiar convertible bonds of the early 1960s have now been garnished to make them more appealing to investors. One variety of these instruments is the liquid yield option note (LYON) which Merrill Lynch introduced in 1985. A LYON is essentially a zero-coupon, convertible, callable, redeemable bond, whose callable and redeemable prices vary through time. Also, the bond contains a call protection for the investor because the issuer cannot call the bond for a prespecified period after issuance unless the issuer's stock price rises above a predetermined level.1

The authors appreciate the helpful comments of Michael Moghalu, two anonymous reviewers, editors, and participants at the 1996 Academy of Economics and Finance Conference.

* Delvin D. Hawley is an Associate Professor in the Department of Economics and Finance at the University of Mississippi, Mississippi.

** O. Felix Ayadi is an Associate Professor in the Department of Economics and Finance at Fayetteville State University, Fayetteville, North Carolina.

Manuscript received, November, 1995, revised, August, 1996.

T

Another recent innovation in the securities market is the introduction of exchangeable securities. An exchangeable bond for instance, grants its holder the right but not the obligation to exchange the bond for the common stock (or other securities) of a firm (called the convert firm) other than the issuer of the bond. For example, on May 15, 1986, National Distillers and Chemicals Corporation issued $49 million worth of exchangeable bonds due in 2011. Each exchangeable bond was convertible into the shares of Cetus Corporation (convert firm) at a conversion price of $49 per share. This new security is different from the traditional convertible bond that entitles the holder to convert the bond into shares of common stock of the issuing corporation according to certain terms and during a certain period.

There are several reasons why corporations issue exchangeable securities. Like convertible bonds, exchangeable bonds are issued as a substitute for common stock especially when the stock market is witnessing a downturn. A corporation will prefer to issue an exchangeable security than to issue common stock at low prices. Even when stock prices are not depressed, it has been argued that through some signaling mechanism, a new issue of common stock always results in depressing the price of outstanding common stock. More importantly, the "pecking order" theory of capital structure proposed by Myers (1984) and Myers and Majluf (1984) supports the view that corporations prefer to issue debt than common stock. Sometimes, corporations issue exchangeable bonds to avoid the prohibitive interest cost associated with the issuance of straight debt. Straight bonds bear a higher rate of interest relative to exchangeable bonds.

This financial instrument also offers unique tax benefits in the sense that it allows the issuing firm to defer the realization of capital gain on the convert's stock until the conversion is exercised. On the other hand, the issuing firm takes advantage of the tax deductibility of interest payment on exchangeables while at the same time receiving dividends from the convert firm. These dividends are subject to tax exclusion.2

The use of exchangeable securities is particularly important in mergers, acquisitions and divestitures. When a bigger firm takes over a somewhat weak one, the relative weakness can be overcome if the weak corporation can borrow under the cover of the parent firm. A corporation also can sell off its stake in another corporation without any unnecessary pressure on the stock price by issuing an exchangeable security. A perfect example is IBM's 6.375 1996 subordinated debenture (issued in 1986) that is exchangeable into capital stock of Intel Corp. Barber (1993) notes that firms issue exchangeable debt after a decision to divest of an intercorporate holding. He found this to be the dominant reason why firms issue exchangeables.

The objective of this paper is two-fold. First, the authors offer a valuation model which is an extension of the Black and Scholes (1973) contingent claims model. Leonard and Solt (1990) have argued extensively for the appropriateness of the Black-Scholes option pricing model to the valuation of warrants. In the case of exchangeable bonds, this model is more appropriate because no new shares are issued; the stocks into which the bonds are convertible have already been issued by the convert firm and held in escrow by the issuing firm.3 Second, an attempt is made to apply the new model to

real life data and observe the relationship between market prices of these bonds and their underlying theoretical values. In line with the above, this paper is organized as follows: Section 2 discusses the theoretical issues, and the pricing model is presented in the third section. The next part is the empirical application of the new model. The final section is the conclusion.

EXCHANGEABLE BONDS VERSUS CONVERTIBLE BONDS

The valuation of an exchangeable security is significantly different from the valuation of a regular convertible. An exchangeable bond grants the holder the right but not the obligation to exchange the bond for the common stock of a firm (hereafter referred to as 'convert' firm) other than the issuer of the bond. On the other hand a conventional convertible bond grants its holder the right to exchange the bond for the common stock of the issuing firm. In both cases the bond provides the investor with a fixed return. Also, the investor receives an implied warrant to convert the security into common stock and thereby participates in the possibility of capital gains associated with being a residual owner of a corporation. Convertible and exchangeable bonds are different to both holder and the issuer.

In terms of valuation, the value of an exchangeable bond to an investor is two-fold: its value as a straight bond, and its potential value as common stock. Like a convertible bond, investors obtain a hedge when they purchase an exchangeable bond. If the market price of the convert firm's stock rises, the value of the exchangeable security is determined largely by its conversion value.4 But, if the stock of the convert firm turns down (or if the convert firm goes bankrupt) the investor still holds a bond whose value provides a floor below which the price of an exchangeable is not likely to fall.

The application of contingent claims analysis to valuation of conventional convertible bonds has been well documented in Ingersoll (1977), King (1984), Brennan and Schwartz (1977,1980), McDaniel (1983), Stover (1983), and Marr and Thompson (1984) among others. Brennan and Schwartz (1980) observe that an increase in firm market value volatility would decrease the value of the debt component of a convertible bond because this pushes the firm closer to bankruptcy. On the other hand, the Black-Scholes (1973) option pricing model (OPM) predicts a positive relationship between volatility and the option value. This argument about volatility also applies to the risk-free rate variable. In view of this, McDaniel (1983) remarks that "unavoidable interdependencies" exist between the Black-Scholes option pricing model variables and the variables that determine the straight debt value of a convertible bond. Thus McDaniel argues that the option pricing model may not be appropriate for valuing regular convertible bonds. But this argument is not valid for exchangeable bonds because the observed interdependencies do not exist. For example, the unavoidable "interdependencies" reported by McDaniel in the valuation of regular convertibles do not exist for exchangeables. This is because the underlying stock is not the common stock of the corporation that issues the exchangeables. Consequently, variables such as volatility and conversion value that determine the warrant value of the exchangeable relate to the convert corporation.

Another issue is the dilution-adjusted OPM that is used for valuing regular convertibles. Unlike convertible bonds, the conversion of exchangeable bonds does not result in new common stocks being issued that alter the capital structure of the firm. The exercise of warrants issued in the case of an exchangeable does not directly affect the capital structure of the target corporation. The underlying common stocks are already outstanding. Thus, the Galai and Schneller (1978) and Asquith and Mullins (1986) dilution-adjusted OPM which is used to value regular convertibles is not applicable to exchangeable bonds. These issues will be explored further in the next section.

THE MODEL

The objective of this section is to decompose the valuation of an exchangeable bond into both the straight debt value and a warrant which captures the potential common stock value of the security. In other words, one can evaluate this security as the equivalent of a straight bond with an attached implied warrant issued by the firm to exchange the straight bond for the shares of the target firm. Thus

where,Vex = value of an exchangeable bondVb = straight debt or investment valueVw = value of implied warrant

The option to exchange (exchange option) is like a nondetachable warrant, with the conversion price replacing the exercise price. The existence of the exchange option is the rationale for the relevance of the option theory to the valuation of an exchangeable security.

Investment Value of An Exchangeable Bond

As already discussed above, this type of bond has two basic characteristics: a claim on a predetermined stream of cash inflows from the issuing firm, and an implied warrant to exchange the security for the common stock of a convert firm. Therefore, the value of this type of security reflects these two components. The value of the straight bond characteristics is usually called the bond's investment value. This investment value represents the current value of the bond without any conversion option, inclusive of accrued interest. In other words, it is the present value of all future cash inflows from holding a straight bond, using an appropriate discount factor. That is:

Install Equation Editor and double-click here to view equation.

where,Ct = coupon payment on the bond in period tM = maturity value of the bondr = discount rateN = maturity period of the bond

The investment value of a bond depends on the coupon rate and maturity as well as the risk of default that in turn reflects both the underlying asset risk of the issuing firm and the security provisions of the bond indenture. Brigham (1966) assumes that the bond value is a linear function of the years to final maturity. Thus, his valuation model does not allow for changes in the bond value because of changes in discount rates and default risk.

The implementation of the model represented by Equation 2 depends on the use of an appropriate discount rate. To get this we adopt the model in Billingsley, Lamy and Thompson [hereafter BLT] (1986). Their approach is based on market information and an appropriate discount rate that would exist for the issue if it were sold as straight debt. They define an appropriate discount rate as a linear function of bond rating, interest rate volatility, size of the issue, and the callable features of the bond issue. Their estimated regression equation is of the form:

where,

i = relative yield spread calculated as yield to maturity on issue I minus the yield on the 20-year constant maturity U.S. Treasury index on the date of issue divided by the yield on the 20-year constant maturity U.S. Treasury bond index;

RATING = zero-one variables for Moody's Investor Service rating (A, Baa, Ba, and B). .............................B issues serve as the reference group in the regression;

= volatility in interest rates computed as the previous ten days' (from issue date) mean absolute variation in the 20-year constant maturity U.S. Treasury bond index;

= the natural log. of the dollar size of the issue;CALL = years to first call or refunding divided by the years to maturity.

The BLT model has been shown to possess a better predictive power compared to the Brennan and Schwartz (1980) model. Schadler and Dudley (1993) and Dudley and Schadler (1994) also confirmed the superiority of the BLT model over the Brennan and Schwartz model. Since the relative performance of the BLT model in predicting an appropriate risk-adjusted discount rate is acceptable, we adopt their model in this paper.



Value of Implied Warrant

The insurance value of an exchangeable bond is closer to a warrant than to a call option. Listed call options have expirations that do not exceed nine months, but warrants typically have expirations which far exceed nine months. However, one main distinguishing feature of warrants is that they result in dilution when exercised. In the special case of exchangeable bonds, dilution constitutes no problem because bonds are not convertible into the stocks of the issuing corporation.

In order to apply the Black-Scholes option pricing model, it is necessary to define the relevant boundary conditions as in Brennan and Schwartz (1977, 1980); Ingersoll (1977); and McConnell and Schwartz (1986). From the investor's standpoint, an exchangeable bond can be redeemed at a prespecified call price or converted into common stock. Rationality demands that the investor selects the more valuable option. On the other hand, the firm will find it optimal to adopt a call strategy that minimizes the value of the exchangeable bond.

If we assume away bankruptcy, the value of an exchangeable bond at maturity is the higher of the conversion value and the face value of the bond plus the terminal coupon. However, at any time during the life of the security, its value must be equal to or greater than its conversion value. It is irrational investment behavior to convert the bond when its market value exceeds the conversion value. Furthermore, the call condition is such that at any time the value of the exchangeable bond must be equal to or less than the greater of the call price and the conversion value. These are the call and conversion conditions that serve as boundaries for the pricing of all convertible and exchangeable bonds.

Given that the underlying stock price obeys a geometric wiener process with the usual no-arbitrage assumption, the value of an exchangeable bond must satisfy the

following fundamental differential equation.5

Equation 3, together with the boundary conditions specified above, fully describe the warrant value of an exchangeable bond. The final equation is of the form:

where,



=the proportion of the convert firm's stock price that is paid out in dividend.Rf = the risk-free rateS0 = the market price of the shares of the convert firm that can be exchanged for

a warrantX0 = exercise price of the exchangeable bond = standard deviation of the stock price of the convert firmT = time to expiration/callZ(.) = cumulative probability under the normal curve

Equation 4 assumes that dividends are a constant proportion of the stock price and are paid out continuously over the life of a warrant. Insofar as dividends have effect on the firm's market value, the stock price as well as the price of a warrant will be correspondingly affected.6 In fact, when the underlying stock pays a dividend, an early conversion is highly likely.

Another significant feature of the Black-Scholes OPM implies simultaneous exercise of all options outstanding. According to Emanuel (1983), the differences that exist between warrants and options indicate that the simultaneous exercise may not be optimal if all the warrants are held by one investor. The reason is that the exercise of one warrant reduces the quantity of warrants in circulation and investors can use this to their advantage. So, it is argued that the differences between regular options and warrants render the Black-Scholes model inapplicable.7 Spatt and Sterbenz (1986) and Constantinides (1984) disagree with this inference. Constantinides notes that when warrants are held by competing investors, the OPM is still applicable. Spatt and Sterbenz examine the interaction between optimal warrant exercise strategies, on the one hand, and the firm's capital structure, dividend and reinvestment policies on the other. They argue that the firm can follow policies that will eliminate any advantage to sequential exercise strategies (cf. Cox and Huang, 1989). The authors concur with the conclusion of Constantinides.

DATA AND METHODOLOGY

The original exchangeable bond sample used in this study includes seventeen bonds. For each bond, the issue date, conversion and call features, rating, maturity date and other features were obtained from Moody's Bond Record, Moody's Bond Survey, Moody's Manuals, and the Wall Street Journal. The sample size was made up of exchangeable bonds issued between January 1984 and June 1987. The minimum criteria for the selection of the final sample are as follows:

- the underlying common shares are publicly traded- the exchangeable issues are also publicly quoted- the value of the issue is at least $45 million- data on the issue are available in Moody's and/or


Standard and Poor's manuals- the exchangeable bond is rated by one of the rating agencies- the bond is not a Eurobond.

The rationale for imposition of the screening criteria are as follows: We were interested in getting data on these bonds both at time of issue and beyond. Moreover, we wanted to avoid running into the confounding impact of other influences such as foreign exchange rate implications of Eurobonds. An issue size of $45 million was used because of low level of market activity associated with these small issues. Moreover, most of the issues were prematurely retired. Thus, the above screening process produced a final sample of seven exchangeable securities.8

The U.S. Treasury 20-year Constant Maturity Bond Index and the risk-free rate were obtained from Federal Reserve Statistical Release: Selected Interest Rates for various years. The historical record of the convert firm stock price is obtained from Standard and Poor's Stock Record.

The investment value of each bond was calculated using Equation 2. The BLT model whose validity was also confirmed by Dudley and Schadler (1994) is used to determine the appropriate discount rate. The warrant value was obtained from the application of the option pricing model represented by Equation 4. The estimate of the underlying stock price volatility is based on instantaneous rate of return on the convert firm's weekly data twenty-four weeks prior to the valuation period. The adjustment for dividend payments is on the usual assumption of a constant dividend yield. The yield is the ratio of average dividend for two years before the exchangeables are issued to the current price. The investment and warrant values were summed together to obtain the value of each exchangeable.

RESULTS

The attempt in this section is to compare the prices generated from the model discussed above with the market prices of exchangeables. We also intend to proffer possible reasons for the divergence between market prices and their respective theoretical values.

Table 1 shows the theoretical prices of sampled exchangeable bonds at the time of issue. All the bonds are valued above their face values. The values range from $1079.40 to $1223.75. In all sampled bonds, the warrant portion of theoretical value is relatively small. The warrant value as a proportion of total value ranges from 9 percent for Panhandle Eastern Corp. to a high of 26 percent for Signal Companies.

Table 1

Theoretical Values of Exchangeable Bonds at Time of Issue

BOND WARRANT VALUE ($)

INVESTMENT VALUE ($)

TOTAL VALUE ($)

General Dynamics 136.31 960.04 1096.35

Petrie Stores 174.43 904.97 1079.40

Signal Cos. 323.23 900.52 1223.75

IBM 162.74 968.38 1131.12

National Dist. & Chem. 241.32 916.39 1157.71

General Host Corp. 246.02 1031.05 12277.07

Panhandle East. Corp. 102.59 1037.32 1139.91

In Table 2, we compared both the theoretical values and market prices of the underlying exchangeable bonds. At the time of issue, all the exchangeable bonds in the sample are significantly underpriced in the market relative to theoretical values. The ratio of theoretical value to market price (column 2) ranges from 1.079 to 1.277. On average, an exchangeable bond is sold at about 86 percent of its theoretical value. This result corroborates the following observation by McGuire (1991 p.39): "As a general rule, the theoretical value of an equity warrant often substantially exceeds its market value. Sometimes investors pay as little as one-half to two-thirds of what the warrant in theory is worth."

In column 3 of Table 2, we show the relationship between theoretical values and market prices at the time first public listing of the bonds. Results show that the market still underprices exchangeables, because an average bond sells for about 90 percent of its theoretical price. However, by the third month in the market, both theoretical and market prices of exchangeable bonds begin to converge. At this time, about 75 percent of the bonds sampled are priced within 10 percent of their respective theoretical values. On average, the market price is 92 percent of the theoretical price. The t-statistic suggests convergence in both prices at the 5 percent level. A higher degree of convergence is also achieved after the securities had traded for twelve months (column 5). There is still a tendency for the average market price to rise above its respective theoretical value.

Two important trends are found in this paper. First, the market price of exchangeable bonds are significantly lower than their respective theoretical prices at the time of issue. Second, within a period of three months after these bonds become listed on an exchange, we tend to achieve convergence between market and theoretical prices. The obvious conclusion is that there is a sort of "learning" process in the market for these securities. The more time they remain in the market, the more investors know about them and the more accurate one is able to value them. Market illiquidity has also been identified by McGuire (1991) as a possible factor explaining why market prices deviate from their underlying theoretical values. McGuire noted that the convertible market is illiquid because there are few market makers, reduced investor interest,

smaller market capitalization and less transparent methods of trading. Finally, the assumptions underlying the model discussed in this paper do not properly hold in practice. For example, dividend yield is not constant. Thus, the same types of bias found in the Black-Scholes OPM can be expected to occur in other applications of the OPM such as the one discussed here.

The pattern observed in this study is such that the trade in exchangeables bonds is continuous for an average of twelve months after they are listed. After this initial period, some do not record any trade for more than two years and thus result in an illiquid market. For example, exchangeable bonds issued by IBM in March 1986 was listed in the market in November 1986. It traded until (although not continuously) October 1987 and then recorded no further trade. Thus, the results obtained here confirm the view of McGuire (1991, p.48).9 Kuhn (1990) also provides a good description of the relative under performance of convertibles, especially in the later part of 1985, This incidentally coincides with our sample period.

Table 2

Relative Performance of the Valuation Model

BOND AT TIME ISSUED ($)

FIRST PUBLIC TRADE ($)

MONTH 3 OF PUBLIC TRADE ($)

MONTH 12 OF PUBLIC TRADE

General Dynamics

1096.35/ 996.00(1.101)

1187.27/ 1085.00(1.094)

1077.59/ 1145.00(0.941)

1179.42/ 1145.00(1.03)

General Host 1277.07/ 1000.00(1.277)

1086.06/ 1020.00 (1.065)

1045.09/ 945.00 (1.106)

N.A

IBM 1131.12/ 1000.00 (1.131)

1322.29/ 1221.25 (1.083)

1369.56/ 1205.00 (1.137)

N.A

National Distillers & Chemicals

1157.71/ 1000.00 (1.157)

995.81/ 935.00 (1.065)

955.55/ 885.00 (1.080)

1001.95/ 1180.00 (0.85)

Panhandle Eastern Corp.

1139.91/ 1000.00 (1.140)

1155.12/ 1055.00 (1.095)

1225.01/ 1120.00 (1.094)

1440.56/ 1340.00 (1.08)

Petrie Stores 1079.40/ 1000.00 (1.079)

1291.31/ 1140.00 (1.133)

1226.00/ 1180.00 (1.039)

1199.94/ 1420.00 (0.085)

Signal Cos. 1223.75/ 1000.00 (1.224)

1085.55/ 875.00 (1.241)

1036.67/ 830.00 (1.249)

1151.54/ 1010.00 (1.14)

AVERAGE 1157.90/ 999.43 (1.159)

1160.49/ 1047.32 (1.108)

1133.64/ 1044.29 (1.086)**

1194.68/ 1219.00 (0.98)**

* In each cell, the numerator is estimated value of an exchangeable. The denominator is the quoted price of the security. The values in brackets represent the ratio of model values to actual market prices of the exchangeables.

** Both theoretical and market prices are not significantly different at five percent level N.A. means no trade in the bond for the month.

CONCLUSION

The nature of exchangeable securities makes the application of the Black-Scholes option pricing model most relevant. An exchangeable bond can be viewed as a combination of a straight debt and an implied warrant. Therefore, the Black-Scholes OPM can be applied to value the warrant portion of the package without running into the "interdependencies" observed by McDaniel (1983) when the Black-Scholes model is applied to the traditional convertible bonds.

Although the results in this study indicate that market prices of these securities deviate from their respective theoretical values especially at the time of issue, the deviation tends to decline with time. This suggests that it takes some time before investors get to understand these complex securities. Market illiquidity has been identified by McGuire as one of the factors why market prices deviate from underlying theoretical values. The pattern observed in this study is such that the trade in exchangeables is continuous for an average of twelve months after they are listed. After this initial period, some do not record any trade for more than two years. For example, exchangeable issued by IBM in March 1986 was listed in the market in November 1986. It traded until (although not continuously) October 1987 and then recorded no further trade.

Finally, the results in this study support the practical usefulness of the OPM and the BLT model in the valuation of exchangeable bonds. Both models are used to determine a reasonable estimate of equity and debt components of an exchangeable security at any point in time.

The pricing formula presented in this paper, like the Black-Scholes OPM could be sensitive to both the estimates of the discount rate obtained from the application of BLT model and the volatility of the underlying stock price. If these estimates are biased, one expects the theoretical values to be biased too. Some of the convert firms' stocks in our sample do not have options listed on them. Therefore, the only approach for estimating stock price volatility is the approach adopted in this study. However. it should be noted that the BLT model has been validated by Dudley and Schadler (1994). Based on the foregoing, the valuation model presented in this paper has promise.

NOTES

1. For a more detailed analysis of this type of security, refer to McConnell and Schwartz (1986), Noddings (1982), McGuire (1991) and more recently, Barber (1993).

2. Barber (1993) examines the different reasons why firms issue exchangeable debt and the relative validity of each reason.

3. For a more comprehensive survey of the features of this security, refer to Barber (1993).

4. The convert firm in this case is the firm whose common stock an exchangeable bond would be converted.

5. All the variables here are as originally defined in Black and Scholes (1973). However, the stock price, S (conversion value of an exchangeable) is adjusted for dividend. H is the hedged portfolio. For this derivation refer to Brennan and Schwartz (1977, p. 1704).

6. Refer to Beenstock (1982, p. 36) for a discussion of the effect of dividends on the value of an option.

7. More recently, Emanuel (1983) argues that the Black-Scholes option pricing model exhibits some biases when used to value American call options. These biases were found to be related to the exercise price, time to expiration and the stocks volatility. In the original OPM, out-of-the-money options are systematically underpriced while in-the-money options are overpriced. It is also reported that the Black-Scholes model underprices call options on low variance stocks and overprices options on high variance stocks.

8. There are only a few exchangeable bonds listed in the market. Even with the small size of the market for this security, many of them are not regularly traded. For a survey of the market for exchangeable, refer to Barber (1993).

9. Mcguire (1991) observed that: Convertibles are complex securities whose price behavior requires more time and effort to understand than straight bonds and stocks, In addition, convertible portfolios need to be more actively monitored and traded than stock and bond portfolios. Furthermore, an unexpected rise in interest rates, or a takeover bid, or a decline in secondary market liquidity can lead to under performance of convertibles relative to bonds or stocks.

REFERENCES

Asquith, P. and Mullins, D.W. (1986). "Equity Issues and Stock Price Dilution." Journal of ...............................Financial Economics, 15, January/February, 61-89.

Barber, Brad M. (1993). "Exchangeable Debt." Financial Management, 22(2), 48-60.Beenstock, M. (1982). "The Robustness of the Black-Scholes Option Pricing Model."

Investment Analyst, October, 30-40.Billingsley, R.S, Lamy, R.E. and Thompson, G.R. (1986). "Valuation of Primary Issue

Convertible Bonds." Journal of Financial Research, 9(3), Fall, 251-259.Black, F. and Scholes, M. (1973). "The Pricing of Options and Corporate Liabilities."

Journal of Political Economy, 81(3), 637-659.Brennan, M.J. and Schwartz, E.S. (1977). "Convertible Bonds: Valuation and Optimal

Strategies for Call and Conversion." Journal of Finance, 32(5), 1699-1715.Brennan, M.J. and Schwartz, E.S. (1980). "Analyzing Convertible Bonds." Journal of

Financial and Quantitative Analysis, 15(4), 907-929.Brigham, E.F. (1966). "An Analysis of Convertible Debentures: Theory and Some

Empirical .............................Evidence." Journal of Finance, March, 21(1), 35-54.Dudley, L.W. and Schadler, F.P. (1994). "Reporting the Relative Equity Portion of

Convertible .....Debt Issues." Journal of Accounting, Auditing and Finance, 9(4), 561-577.

Emanuel, D. (1983). "Warrant Valuation and Exercise Strategy." Journal of Financial Economics, 12(2), 211-235.

Galai, D. and Schneller, M. (1978). "Pricing of Warrants and the Value of the Firm." Journal of Finance, 33, December, 1333-1342.

Gepts, Stefaan J. (1987). Valuation and Selection of Convertible Bonds. New York: Praeger.Ingersoll, J.E. (1977a). "A Contingent-Claims Valuation of Convertible Securities."

Journal of Financial Economics, 4, May, 289-321.Ingersoll, J.E. (1977b). "An Examination of Corporate Call Policies on Convertible

Securities." ................................................Journal of Finance, 32, May, 463-478.King, R. (1986). "Convertible Bond Valuation: An Empirical Test." The Journal of

Financial .................................................................Research, 9(1), Spring, 53-69.King, R.D. (1984). "The Effect of Convertible Bond Equity Values on Dilution and

Leverage." ..........................................The Accounting Review, 59, July, 419-431.Kuhn, Robert, L.(ed.) (1990). Corporate and Municipal Securities. Homewood, IL:

Dow Jones-Irwin.Leonard, D. C. and Solt, M.E. (1990). "On Using the Black-Scholes Model to Value

Warrants." ..................The Journal of Financial Research, 13(2), Summer, 81-92.McConnell, J.J and Schwartz, E.S. (1986). "Lyon Taming." Journal of Finance, 41,

July, .........................................................................................................561-577.

McDaniel, W.R. (1983). "Convertible Bonds in Perfect and Imperfect Markets." Journal of ....................................................Financial Research, 6, Spring, 51-65.

McGuire, Simon R. (1991). The Handbook of Convertibles. New York: Institute of Finance.

Myers, S. (1984). "The Search for Optimal Capital Structure." Midland Corporate Finance Journal, 1, Spring, 6-16.

Myers, S. and Majluf, N. (1985). "Corporate Financing and Investment Decisions When Firms ........Have Information Investors Do Not Have." Journal of Financial Economics, 13(3), 467-480.

Noddings, T.C. (1982). The Investor's Guide to Convertible Bonds. London: Thomas C. Noddings.

Schadler, F.P. and Dudley, L.W. (1993). "A Market-Based Model for Determining Common Stock Equivalence Improves EPS Reporting." Working Paper FINA 93-13, East Carolina University, Greenville, NC.

Schwartz, E. (1977). "The Valuation of Warrants: Implementing a New Approach." Journal of Financial Economics, 4(1), 79-93.

Spatt, C. and Sterbenz, F. (1989). "Warrant Exercise, Dividends and Reinvestment Policy." GSIA Working Paper No 33-84-85, Carnegie-Mellon University.

Stover, R.D. (1983). "The Interaction Between Pricing and Under-Writing Spread in the New Issue Convertible Debt Market." Journal of Financial Research, 6, Winter, 323-332.

DEVELOPING CUSTOMER-CENTERED

PERFORMANCE MEASURES

Joseph C. Montgomery *

David J. Lemak **

Richard Reed ***

This paper discusses the role that performance measures have in evaluating organizational effectiveness/efficiency, and the recent shift away from the traditional "manager-centered," cost accounting approach to a "customer-centered," performance- based perspective. The dysfunctional aspects of the former approach are discussed and contrasted with the advantages of the latter, more recent one. Then, a conceptual model for developing customer-based performance measures is developed. The paper concludes with a brief summary of the arguments presented, some limitations and caveats, and directions for future research.

he measurement of organizational performance is a key component of organizational effectiveness (Hall, Johnson, Thomas and Turney, 1991). Without

performance measurement information, the organization has no way of knowing whether it is achieving its mission, whether its strategies are appropriate to the situation, or whether implementation of the strategies is being successfully accomplished. The focus of this paper is on profit oriented, private sector organizations

This work was supported by the U.S. Department of Energy (DOE) under Contract DEAC06-76RLO 1830. Pacific Northwest National Laboratories are operated for DOE by Battelle Memorial Institute.

* Joseph C. Montgomery is affiliated with Battelle Pacific Northwest National Laboratories in Richland, Washington.

** David J. Lemak is a Professor of Management and Systems at Washington State University Tri-Cities, Richland, Washington.

***Richard Reed is a Professor of Management and Systems at Washington State University, Pullman, Washington.

Manuscript received, July, 1995, revised, February, 1996.

T

involved in some form of manufacturing and/or service activities, and, as such, the ultimate external measure of success is typically defined in terms of financial performance. However, it is argued here that performance measures internal to the firm can take on a variety of contexts and forms. The role of performance measures, in the context of organizational mission, strategies, and actions, is indicated in Figure 1 (adapted from Dixon, Nanni and Vollman, 1990).

The mission of the organization -- the definition of the organization's purpose, business areas, and "value-added" -- drives the development of strategies and plans. The strategies and plans, in turn, guide and determine the actions to be taken by organizational members. Actions include establishing and maintaining effective work processes, designing the organizational structure, emphasizing quality. These actions drive the development of performance measures. Performance measures provide a framework for the systematic collection and analysis of data summarizing performance within the organization. Once collected, these data provide feedback regarding achievement of the mission, successful implementation of strategies and plans, and success of the organization's actions. Effective performance measures provide feedback on both internal processes and on the level of satisfaction of customers and stakeholders. The result, when the mission, strategies, actions, and performance measures are operating, is an increasingly effective organization.

A NEW MODEL OF PERFORMANCE MEASUREMENT

Performance measures have traditionally been developed from a cost-accounting orientation, which has been labeled a "manager-centered approach" (Hall et al., 1991). From a historical perspective, cost accounting was developed in the late 1800s to help managers evaluate the total costs of operating textile mills, railroads, steel mills, retail stores, and similar businesses (Johnson and Kaplan, 1987). Attention was focused on variable costs (such as labor and materials) versus fixed costs (maintenance, facilities, and support functions). At that time, variable costs formed a considerable portion of total costs, typically 80% or more. Consequently, breaking the production process into components and assigning overhead charges to each component on some roughly rational basis was a sensible way of tracking costs. Standard financial indicators resulting from the cost-accounting approach include machinery utilization, labor utilization, net profit, return on investment, unit cost, and inventory turnover.

Production methods have changed dramatically however, particularly in the past two decades, and the assumptions upon which cost accounting were built are no longer applicable for most organizations. For example, variable costs have dropped from 80% to only 10-15% of total costs (Kaplan, 1984). As a result, not only do cost-accounting-based measures fail to provide useful feedback, but (as will be demonstrated shortly) they motivate patterns of counterproductive behavior (Drucker, 1990). The difference in orientation and actions of

Figure 1

Role Performance Measures in Organizational Effectiveness

MISSION

STRATEGY

ACTIONS


organizations that are manager-centered versus customer-centered is so great that they have been labeled as separate operational paradigms (Hall et al., 1991). A variety of familiar programs fall into the customer-centered paradigm, including Total Quality Management (TQM), Just-in-Time (JIT) manufacturing, Total Productive Maintenance (TPM), continuous improvement, total employee involvement, and quality function deployment.

Performance measures used by world-class organizations tend to be customer-centered rather than management-centered (Hall et al., 1991). Customer-centered performance measures are linked with product quality, dependability of service, waste reduction, timeliness, flexibility, innovation, and other indicators that are closely linked with actual work processes. Development and implementation of these measures has often resulted in dramatic improvements in internal work efficiency and effectiveness, thus driving the performance of products and services in the marketplace (Young, 1992). Manager-centered performance measures, on the other hand, tend to be based on financial measures and used for control purposes.

Key differences in assumptions between the manager-centered and customer-centered paradigms are summarized in Table 1 below (adapted from Hall et al., 1991). As can be seen, the manager-centered paradigm describes the traditional American organization.

Table 1

Differences Between Manager-and Customer-Centered Organizations

MANAGER-CENTERED ORGANIZATIONS CUSTOMER-CENTERED ORGANIZATIONS

A company is its assets; the company is a possession A company is its people; assets are people and people are assets

Economies of scale dominate; bigger is always better Economies of scope dominate; faster response is better

Managers manage; workers work; management initiates all improvement

Workers are thinkers; everyone in the organization works for improvement

Vertical organization; functional silos separate groups Horizontal organization, multi-directional communication; internal customer chain; focus on external customer needs

Profit is always first; costs are thought of in terms of tradeoffs

Quality is first; there is no compromise with quality and customer service

Company-centered operations; transaction driven management

Manufacturing-centered operations; improvement driven, problem solving/teamwork orientation

Performance measurement for control purposes; traditional cost accounting measures dominate

Performance measurement for work improvement; internal and external customer satisfaction and other non-cost

28

SUMMER 1997

operating measures dominate; measures of work processes for continuous improvement lead to high financial performance

Numerous writers (e.g., Schonberger, 1990; Suzaki, 1987) have presented detailed arguments as to why customer-centered organizations outperform manager-centered organizations. Manager-centered organizations simply cannot compete with customer-oriented organizations in such areas as innovation, quality, timeliness, cost, and customer satisfaction. As will be seen in the next section, cost-accounting-based performance measures make it very difficult for members of manager-oriented organizations to keep the overall mission in mind and to avoid succumbing to strong pressures, thus behaving in ways that hurt overall organizational performance.

Finally, as Cooper (1995) emphasizes, the shift in manufacturing from mass to lean production techniques must be accompanied by corresponding changes in assumptions about how to compete in the global market. Cooper has developed a conceptual tool labeled the "Survival Triplet," based on three critical dimensions of product effectiveness:

1) cost-price, 2) quality, and3) functionality.

Each of these components has a minimum and a maximum value for any given product to be competitive. The result can be thought of as a three-dimensional solid figure which defines a "survival zone" for each product. Thus the product will become non-competitive if it exceeds a certain price, drops below a certain level of quality, or fails to achieve a minimum level of functionality. The product will also become non-competitive, in a positive sense, if the producer can simultaneously drop below a certain price and exceed a certain level of quality and functionality-- other products simply will not compete with such a product. The traditional manager-centered model focuses attention on the area of cost-price, while the customer-centered model drives the manufacturer to optimize performance on all three dimensions. As a result, products coming from a manager-centered producer tend to drop out of the "survival zone" and cannot compete successfully with those coming from the customer-centered model.

DYSFUNCTIONAL IMPACTS OF COST-ACCOUNTING PERFORMANCE MEASURES

Once developed and implemented, performance measures tend to take on a life of their own, motivating a pattern of specific behaviors within the organization (Dixon,

29


Nanni and Vollmann, 1990). As the saying goes, "you get what you measure." Cost-accounting measures begin to drive both organizational actions and strategies because they 1) provide information to organization members as to what is important, and 2) indicate how their performance will be evaluated. Consequently, these performance measures become extremely powerful motivators of behavior throughout the organization (Johnson and Kaplan, 1987).

For example, once cost accounting measures are in place, management is highly motivated to show "good" performance on these measures. Consequently, they will work hard at maximizing output, minimizing unit cost figures, and maximizing performance on other measures such as machine- and labor-utilization. A side effect of these efforts, unfortunately, is excessive amounts of work-in-process (WIP) and inventory. Further, the goal of maximizing machinery and direct-labor utilization leads to tactics such as long production runs and sustained use of the workers. Unintended consequences of these tactics include additional WIP and inventory problems as well as inadequate attention to equipment maintenance and training of personnel. Thus, the drive to perform well on the performance indicators also contributes to breakdowns and operator errors.

Additional production problems will emerge as reductions in the training budgets (which reduce overhead expenditures) contribute to workers falling behind technically. Quality will also become an increasing problem because of the expanded production volume and the corresponding inability to adequately check for and correct defects. Consequently, resources devoted to rework increase and scrap rates will rise. Storage, transportation within the production area, and work scheduling become increasingly difficult. The overall result is a production process characterized by waste, inefficiency and low levels of effectiveness.

We want to stress that at department manager level, the actions and tactics used to perform well on the performance indicators appear very sensible. As a whole, however, they devastate the production system. The resulting production strategies are exactly the opposite of those recommended by world-class manufacturing experts (e.g., Black, 1991; Schonberger, 1990; Suzaki, 1987).

ENHANCEMENT OF ORGANIZATIONAL EFFICIENCY AND EFFECTIVENESS WITH CUSTOMER-CENTERED MEASURES

In sharp contrast, when customer-oriented performance measures are used, Figure 1 continues to hold true. This is because customer-oriented measures build upon the concepts of both external customers outside the organization and a "chain of customers" within the organization (Schonberger, 1990). The identification of external customers

30

SUMMER 1997

should be a relatively simple, straightforward process. However, our experience confirms what others have noted (e.g., Tenner and DeToro, 1992), namely, that most managers do not think in very specific terms of who their customer really is. For example, when dealing with a large company which has thousands of employees, the important task is to identify which of those people are really the customers. That is, the customers (buyers) of our product may not be, and usually are not, the ultimate users. Thus, the problem of identifying specific customer requirements can become very complex since the supplier has to identify who is the customer (i.e., the purchaser, the end user, or someone else), and what are the expectations of the customer(s) regarding the attributes of the product.

There is an equally complex identification problem within the organization for "internal customer chain." This concept takes the notion of the supplier-customer chain and extends it to input/output relationships within the firm. For example, if a production process involves a series of steps by different work groups then, in effect, a series of supplier-customer relationships results and, ideally, the initial "supplier" will provide the materials needed by the "customer" only in the quantity needed and only at the time they are needed. Customer identification in this process is aided by notions of output. "Outputs" are defined by Tenner and DeToro (1992, p. 54) as "the specific products or services that you produce, as part of your work process, and that you pass to others, who, in turn, use them in their work process." They go on to state the obvious but important point that if one can identify who receives your output(s), one can simply ask that person about such things as requirements and expectations which can then be translated into performance criteria (e.g., timeliness, quality and, defect rate).

As the internal customer becomes, in turn, the internal supplier for the next link in the chain, exactly the same process of performance measurement and feedback of data applies (see Figure 2). The end result of the chain of customers is essentially a "horizontal organization" (Ostroff and Smith, 1992). In the horizontal organization the products flow smoothly through the organization, unimpeded by departmental boundaries, quite unlike the deep departmental silos typically found in manager-centered organizations. While traditional organizations focus on vertical information flows, it is the lateral or horizontal flows of information created by performance measures that are critical to effective performance (Drucker, 1990).

In addition to the information flows, it is also important that the different roles that individuals or work groups perform within such organizations is understood. The internal chain of customers focuses on the notion that any one individual in the internal customer chain is simultaneously a customer (receiving output from another supplier) and a supplier (providing input to another customer farther down the chain). But, we can deduce from the literature that there is a third role, that of producer. That is, each individual or work group transforms and adds value to what eventually emerges as the

31


final product. In fact, much of what has been written recently focuses on the value added function, eliminating those activities which contribute to costs but fail to add value (Porter, 1985), and focusing only on immediate results (e.g., Myers, and Ashkenas, 1993; Schaffer and Thomson, 1992). While this is certainly important, the other two roles, those of customer and supplier, seem to have been overlooked, or at least down played. Granted, much has been written about the importance of customer orientation as it relates to external customers, but very little attention has been given to the concept of internal customers (Mohr-Jackson, 1991: Schonberger, 1992). If the requirements of the external customer are the ultimate driving force in this process, firms must also recognize that that the internal customer chain is composed of producers who are also, simultaneously, customers and suppliers.

One final point needs to be made with regard to supplier-customer relations, both internal and external. As the customer provides two types of information to the supplier (the product/service requirements and the customer's satisfaction with them), the supplier provides two other types of information to the customer (capabilities as they relate to what the supplier can reasonably deliver and expectations about the product/service quality). This set of exchanges is also depicted in Figure 2, where the interchange of information is continuous with all parties making continual and progressive adjustments. In any firm, resources are limited, and it is not reasonable to expect that a supplier can completely meet every customer requirement within some cost constraints. This holds true for the relationship between the external customer and the supplier firm, and within the internal customer chain as well. As the customer specifies a variety of requirements and the supplier responds with a realistic estimate of what can be delivered given a variety of constraints, the overall process becomes one of negotiation and optimization. That is, the customer has an idea of basic requirements, but is constrained by how much can be paid for a product or service. Likewise, the supplier evaluates what level of product quality/service can be supplied given cost constraints. The point is, customer requirements and supplier constraints are in a constant state of flux, and what constitutes "customer satisfaction" changes over time, as does the customer's expectations of what the supplier can deliver.

A CONCEPTUAL MODEL FOR DEVELOPING CUSTOMER-BASEDPERFORMANCE MEASURES

The literature provides little guidance regarding the process of actually developing customer based performance measures. The model shown in Figure 3 provides a framework for performance-measure development. The first four steps are accomplished by the process owner leading a cross-functional team composed of members who represent all the departments or groups that are involved in the production process. If there is no process owner (as is often true in functionally

32

SUMMER 1997

structure organizations), then someone -- perhaps a member of senior management -- will need to be named to this position. The performance measure development process must be led by staff who are able to rise above functional stovepipes and boundaries and work with the entire production process.

33

Figure 2

Customer - Supplier Information Flows

Note: This figure depicts only the final work steps of a work process flow that could involve a large number of upstream work steps.


The first steps in the model are the identification, by the process owner and the cross-functional team, of specific external customers and the initial determination of their needs as related to the products and/or services provided by the supplier. As noted above, the outcome of this stage should be a completed analysis of who will be receiving the final outputs and how they will be using them. It is crucial that these ultimate customers be asked to identify their requirements and standards for what is the minimum acceptable and what would be desirable. Customer requirements go beyond traditional notions such as product attributes, features, and warranties. In this age of global competition, customer requirements now include other things such as product quality, reliability, timely delivery, lot size, and after purchase services. It must be noted here that once the customer and supplier go through the process depicted in Figure 2, the supplier must then deliver what was agreed upon in this cycle if the customer is to be minimally satisfied.

The next step in the process is to mapping out the existing process flows. It is vital to thoroughly understand the entire existing process before attempting any improvements or creating any performance measures. It is likely the certain aspects of the existing plant layout, production flows and the organization of staff make it impossible to meet customer requirements. The completed process map should portray in a simple fashion each step in the production process, including what is accomplished and by whom and depicting how the materials and/or information flows through the process. The map may, for example, be laid out on a chart of the production workspace. Consideration of process flows overcomes the restrictions of functional boundaries, portrays the overall transformation process from inputs to outputs, and helps pinpoint critical links, bottlenecks and potential trouble spots.

Once the steps in the process have been identified, it is a simple step to identify the internal chain of customers. The process owner and his/her team simply identify all of the handoff points in the process. Those handing off materials or information are suppliers; those receiving materials or information are customers. As mentioned earlier, the identification of an internal chain of customers must be grounded in the concept of outputs. This step builds on Schonberger's (1992, p. 83) notion of organizing "...resources into chains of customers, each chain mostly self-contained and focused on a product or customer 'family.' "

What is not such a simple step is to educate the individuals or groups at each handoff point as to their roles as suppliers or customers. As Gunn (1992) points out, organization members must understand that their "customer" may well be in the next office. Firms must identify an internal "chain of customers" based on one person's output becoming another's input, be that on a product assembly line or in a billing process in the accounting department. Organization members need to understand that

36

SUMMER 1997

customer satisfaction and concern for quality and timeliness are just as important to those within the firm as to those who receive the final product outside the firm.

37

Figure 3

A Model for Developing Customer-Based Performance Measures

v

v

v

IDENTIFY EXTERNAL CUSTOMERS AND CUSTOMER REQUIREMENTS

MAP PROCESS FLOWS

IDENTIFY INTERNAL CUSTOMER CHAIN

>

IMPROVE/RE-ENGINEERWORK PROCESS

DEVELOP CUSTOMER-BASED PERFORMANCEMEASURES

FINANCIAL PERFORMANCE


In the next step in the model, labeled "Improve/Reengineer Work Processes," the process owner and the supporting team use the information from the previous steps to redesign the work processes so that external customer requirements can be met as effectively and efficiently as possible. Blaxill and Hout (1991, p. 94) contend, "Everything flows from robust processes: higher quality, better cycle time and much lower overhead." It may be that only minor adjustments to the work process are necessary, or it may be that major improvements and radical redesign of how the work is accomplished is required. At any rate, it makes no sense to perform work which does not contribute to meeting customer requirements or to develop and implement performance measures for flawed processes. While it is beyond the scope of this paper to summarize reengineering techniques, there are a variety of tools that can be used to implement this stage of the model (e.g., Hammer and Champy, 1993; Montgomery and Levine, 1996). Specific improvements may include reducing transport distance, flow time and space along the customer chain, cutting setup and changeover times, and using the customer's rate of usage for production scheduling as ways to streamline operations. Regardless of the techniques employed, it is critical that the improvements be initiated and driven by the organizational members and work teams (i.e., internal suppliers and customers) most closely involved with the work processes.

A final step in the reengineering process is for each supplier-customer pair (which may have changed from the original analysis) to negotiate exactly how the internal customer requirements are to be met. This discussion is truly a negotiation process, with the goal being a win-win agreement on how the supplier and customer are to work together. If resources are limited, it may be that not all demands of internal customers can reasonably be met. For example, it would be unreasonable for one internal customer to demand three day delivery on a component from an internal supplier simply to build in slack time for this stage of processing, when a two day delivery would suffice. The internal negotiation process must be driven by external customer requirements which cannot be compromised. In fact, each time there is a comprise on external customer requirements internally, the sum of all those compromises will be manifested many times over in that final product (Tenner and De Toro, 1992).

The next two steps in this process should be conducted simultaneously, by different teams working in direct communications with each other. First, a team composed of financial experts uses activity-based cost accounting (ABC) techniques to allocate costs across the process flows. While ABC is not a universal panacea, when used in concert with a process analysis and improvement efforts, it can be a very effective tool in helping to obtain an accurate estimate of the costs associated with each supplier-customer step in the production process. The reason we advocate implementing ABC at this stage in the model is best summarized by Johnson (1992, p. 154): "If your goal is competitive operations, don't waste time gathering data and compiling information in

40

SUMMER 1997

order to cost work you shouldn't be doing anyway." Thus, ABC information is not created until the production process has been thoroughly improved in the previous step.

While ABC is being used to allocate costs, the parallel step in the model is for the process owner and the team to the facilitate the development of customer-centered performance measures for each supplier-customer linkage. That is, performance measures are developed for each supplier-customer link by the supplier staff working in conjunction with their customers. The exact nature of the measures will, of course, depend on customer needs, as discussed earlier. Measures might include quality, dependability, waste, timeliness, and customer satisfaction. Finally, when the ABC financial analysis is completed, the cost attributed to each step of the process will simply be included as a performance measure for the corresponding step in the process. Consequently, not only will such measures as cycle time and queue time be tracked, but the cost of performing the step will be a measure of performance as well.

When the measures for each internal supplier have been completed, the process owner/team then can create a set of "rollup measures", such as total cycle time, total waste, total amount of rework, etc.. As a result, a small set of measures that reflect overall process performance will be available.

Finally, the model shows that the net result of the entire performance measure development process will be reflected in the firm's financial performance. If the firm truly adopts a customer orientation and if the customer requirements drive revising or reengineering the internal processes so that the final product meets or exceeds customer expectations, then bottom-line financial performance must necessarily be affected. We believe that this financial performance is optimized by following the process we have outlined.

CONCLUSION

The customer-centered orientation provides an approach to the measurement of organizational performance that, unlike the manager-centered one, should improve organizational efficiency and effectiveness. The cost-accounting measures used in the manager-centered approach drive both workers and managers to short-term efforts to maximize these measures, resulting in overall inefficiency, low effectiveness and possibly some dysfunction. Customer-centered measures provide motivation to organizational members to improve quality, reduce cycle time, develop innovations, and thus improve the value-added. When performance measures are developed for each link in the internal customer chain, the organization effectively becomes horizontal rather than vertical. The deep vertical silos typically found in traditional organizations are circumvented.

41


Customer-centered measures provide the information that top management needs to make operational decisions. Experts in the area of production and productivity strongly advise managers not to make operational decisions based solely on cost accounting data (Dixon, et al., 1990; Hall et al., 1991; Johnson and Kaplan, 1987; Peters, 1988). This is not to say that there is no role for cost accounting measures. Cost accounting measures may well be required for external reporting purposes, perhaps to regulatory agencies, or to fulfill legal requirements. Activity-based accounting provides all the data (and more) of traditional cost accounting measures except that it reflects reality better because of more accurate and appropriate overhead allocations. We argue strongly against using only accounting measures to guide the actual operation of an organization. The cost in terms of dysfunctional behaviors and loss of mission orientation is simply too great. Rather, these tools should be used as part of a larger process guided by external customer requirements and evaluated using customer-centered performance measures. Hopefully, such measures can be developed using the conceptual model presented here.

42

REFERENCES

Black, J. T. (1991). The Design of the Factory with a Future. New York: McGraw-Hill.Blaxill, M. F. & Hout, T. M. (1991). "The Fallacy of the Overhead Quick Fix."

Harvard Business Review, 69(4), 93-101.Dixon, J. R., Nanni, A. J., & Vollmann, T. E. (1990). The New Performance

Challenge: Measuring Operations for World-Class Competition. Homewood, IL: Richard D. Irwin

Drucker, P. F. (1990). "The Emerging Theory of Manufacturing." Harvard Business Review, 68(3), 94-102.

Gunn, T. G. (1992). Twenty-First Century Manufacturing. New York: Harper Business.Hall, R. W., Johnson, H. T., Thomas, P., & Turney, B. B. (1991). Measuring Up:

Charting Pathways to Manufacturing Excellence. Homewood, IL: Richard D. Irwin.

Hammer, M. & Champy, J. (1993). Reengineering the Corporation. New York: Harper- Collins.

Johnson, H. T. & Kaplan, R. S. (1987). Relevance Lost: The Rise and Fall of Cost Accounting. Boston, MA: Harvard Business School Press.

Johnson, H. T. (1992). Relevance Regained. New York: The Free Press.Kaplan, R. S. (1984). "Yesterday's Accounting Undermines Production." Harvard

Business Review, 4(4), 95-101.Mohr-Jackson, I. (1991). "Broadening the Market Orientation: An Added Focus on

Internal Customers." Human Resource Management, 30(4), 455-467.Montgomery, J.C. & Levine, L.O. The Transition to Agile Manufacturing,

Milwaukee, WI: Quality Press.Myers, K., & Ashkenas, R. (1993). "Results-Driven Quality..Now!" Management

Review, March, 40-44.Ostroff, F. & Smith, D. (1992). "The Horizontal Organization." The McKinsey

Quarterly, 1, 148-168.Peters, T. (1988). Thriving on Chaos. New York: Alfred Knopf.Porter, M. E. (1985). Competitive Advantage: Creating and Sustaining Superior

Performance. New York: The Free Press.Schaffer, R. H. & Thomson, H. A. (1992). "Successful Change Programs Begin with

Results." Harvard Business Review, 70(1), 80-89.Schonberger, R. J. (1990). Building a Chain of Customers. New York: The Free Press.Schonberger, R. J. (1992). "Is Strategy Strategic? Impact of Total Quality Management

on Strategy." Academy of Management Executive, 6(3), 80-87.Suzaki, K. (1987). The New Manufacturing Challenge: Techniques for Continuous

Improvement. New York: The Free Press.Tenner, A. R. & DeToro, I. J. (1992). Total Quality Management: Three Steps to

Continuous Improvement. Reading, MA: Addison-Wesley Publishing Co.

Young, S. M. (1992). "A Framework for Successful Adoption and Performance of Japanese Manufacturing Practices in the United States." Academy of Management Review, 17(4), 677-700.

EFFECTIVE SMALL BUSINESS RESPONSE

STRATEGIES TO FEDERAL GOVERNMENT

COMPETITIVE PROCUREMENTS

Robert S. Frey *

Practical information and proven-under-fire processes needed to plan, organize, and prepare effective proposals to Federal Government clients are introduced. Emphasis is given to producing highly focused, requirements-driven proposals that respond fully to client success criteria. Proposals are presented as marketing vehicles and as dedicated efforts in information management. Superior proposal writing and presentation techniques are provided, along with successful information management and proposal production scenarios. Winning approaches to strategic planning and long-term marketing relationships are offered. Human and organizational dynamics are shown to drive successful marketing and proposal processes within all companies.

inning. The Federal Government competitive procurement process1 is absolutely binarycontractors either win or lose with their proposals. With rare exception, there are no rewards for coming in second. To allocate your companys Bid and Proposal (B & P), marketing, and internal research and development (IR & D) funds to pursue procurements for which there is only a marginal probability of winning is, at best, questionable business planning. Federal agencies often have a variety of domestic as well as overseas2 contractor/vendor firms from which to select a specific supplier of goods or services. At a minimum, you have to know your potential client and his or her requirements, and in turn, your client must be made aware of your companys particular

* Robert S. Frey is the Director of Marketing Communications at EA Engineering, Science, and Technology, Inc., Hunt Valley, Maryland.

Manuscript received, April, 1996, revised, March, 1997.

W

technical capabilities, relevant contractual experience, available human talent, and financial stability in the context of an ongoing marketing relationship. One or two briefings from your company to top-level government agency administrators will most likely be insufficient to secure new business in the competitive federal marketplace.

Many small contracting firms which provide goods and services to the Federal Government are primarily or even solely dependent upon federal contracts for their survival and growth. Consequently, proposal planning, development, management, design, preparation, and follow up are the most important business activities that your company performs. Proposal development constitutes not just a full-time job; it can be a 12 to 16 hour-a-day, six or seven day-a-week effort just to keep from falling behind hopelessly.3 Your company should not start developing a proposal unless it intends to win. An exception to this guideline is if your company wants to propose on a particular procurement in order to gain experience in assembling proposals or to gain recognition from the government as a potential supplier.4 The American Graduate University suggests that as many as three-quarters of the proposals received by government procuring agencies are deemed to be non-responsive or inadequate. 5 If your company competes heavily in the federal marketplace, then proposals are your most important product. It does not matter how large your company is. For example, let s assume that yours is a company with $12 million posted in revenue during the last fiscal year. To simply maintain revenues at that level in the next fiscal year, you will burn $1 million each month in contract backlog. That means that you must win $1 million each month in new or recompete business just to keep the revenue pipeline full. Yet winning $1 million per month in new or recompete business will not allow your company to grow revenue-wise at all!

Without a plan, the proposal process will be chaotic and the product, at best, will be inferior.6 Gone is the time of last-minute, haphazard proposal preparation by a few individuals working in isolation from in-house review and other corporate or divisional guidance. Your company simply cannot compete effectively with the many U.S.-based and overseas contracting firms if every proposal you submit is not your very finest effort. Your company will, of course, not win every procurement25 to 40% is a reasonable win ratio but you must strive to have each and every proposal be in the competitive range7 from a technical, management, and cost standpoints.

It is important to note that a technically sound and competitively priced proposal is not enough. With content and cost must come readability, appearance, and format. And these elements require dedicated time to accomplish. Cover design, page formatting, editing, generating graphics, wordprocessing/publishing, proofreading, photoreproducing, collating, and assembling are all vital steps in the overall proposal preparation cycle. Put yourself in the role of a government evaluator. That person, along with his or her colleagues, has to look at many proposals for each procurement. Would you enjoy struggling through a poorly written, amateurishly prepared document on the evening or on the weekend? Indeed, there are an increasing number of small and large businesses chasing fewer and fewer federal dollars. Even relatively minor

procurements are resulting in 50 or more proposals submitted. Debriefings across a wide variety of agencies suggest that evaluators are spending 15 to 30 minutes on each companys proposal. There simply is no more time available to them. As a result, it is more imperative than ever that your companys proposal stand out in a positive way.

A fundamental reorientation of your companys collective thinking and attitude will most likely be required to begin the challenging shift from 8(a)-style8 procurement to competitive federal business acquisition. Changing attitudes can be a difficult and lengthy process. The process of change must begin and be fostered by senior management. Precisely the thinking that proved so successful and comfortable during the 8(a) days of your companys history is often the very thing that thwarts your companys potential for growth in the competitive arena. Entrepreneurial companies are often characterized by informal business organizations and cultures that are functional for small companies only. However, if companies are successful and grow in terms of revenue and human resources, they will reach a point at which an informal culture and organization is inadequate. This is particularly apparent in the areas of planning, management structure, internal communications, and support infrastructure. Successful, growing companies should reorganize, bring in new senior operations management as appropriate, and develop a strategic planning process. 9 And management responsibilities and authority should be delegated downward so that a small companys organizational structure is not so sharply hierarchical.10

Dedicated effort in accordance with a well-defined plan, broad-based and in-depth knowledge of your clients, and a formalized company organization and communication network all contribute to successful proposals.

SMALL BUSINESS CONSTRAINTS

In terms directly relevant to proposal development, design, and preparation, many small businesses must contend effectively with very limited B & P funds, lack of depth in human resources, a small business base, contract backlog deficit, low level of contractual experience, lack of name recognition in the federal marketplace, and line of credit challenges. A small business base, for example, can lead to higher indirect costs, which in turn can place a company at a competitive disadvantage during procurement efforts. And insufficient staff can translate to few or no people dedicated to the tasks of advanced and strategic planning, marketing to particular federal agencies, proposal operations, proposal reviews, proposal editing and proofreading, and proposal publication. Staffing challenges are seen in full-time project managers working 40 billable hours each week for the government client and then additional time to serve as proposal managers of proposal reviewers. (In predominately service-oriented contracting firms, the companys overall profitability is affected by the degree to which its personnel are fully billable. Transfer ratios, that is, billable time versus total time worked, must remain very high.) And thin contractual experience can lead to low scores received for Past Performance or Relevant Experience sections of the

Request for Proposal (RFP), areas that are given increasing weight in federal procurements in the late 1990s.

NEW TRAJECTORIES IN FEDERAL GOVERNMENT PROCUREMENT

The Federal Government spends approximately $200 billion each year in procuring goods and services. Managing and modifying the complex set of processes known as the Federal Acquisition Regulations (FAR) by which the government procures goods and services has proven to be daunting. David Osborne, a senior advisor to the Clinton Administration and co-author with Ted Gaebler of the best-selling Reinventing Government: How the Entrepreneurial Spirit Is Transforming the Public Sector (1992), foresees a profound movement away from large, centralized, command-and-control bureaucracies toward decentralized, entrepreneurial organizations. These new governmental organizations will be driven by competition and accountability to their customers for the results they deliver. Mr. Osborne supported Vice President Al Gores National Performance Review (NPR), which released a report in September 1993 that recommended sweeping changes in federal policies and procedures. NPR and the concept of service to the citizen involve putting people first. NPR played a significant role in the generation of the Federal Acquisition Streamlining Act.

Signed by President Clinton in October 1994, the Federal Acquisition Streamlining Act (FASA, Public Law 103-355) of 1994 repealed or substantially modified more than 225 provisions of law to reduce paperwork burdens, facilitate the acquisition of commercial products, enhance the use of simplified procedures for small purchases, transform the acquisition process to electronic commerce, and improve the efficiency of the laws governing the procurement of goods and services. FASA took effect on 1 October 1995. It emphasizes the acquisition of commercial-off-the-shelf (COTS) items, streamlines acquisition procedures under an elevated small purchase threshold, implements a government-wide electronic commerce system, establishes uniformity in the procurement system, improves protest and oversight processes, and authorizes specific pilot programs.

The Federal Acquisition Streamlining Act also established the Federal Acquisition Computer Network (FACNET), which requires the government to evolve its acquisition process from one driven by paperwork into an expedited Electronic Commerce (EC) process based upon Electronic Data Interchange (EDI). Federal agencies have been authorized to use EC/EDI for contracts ranging from $2,500 to $100,000. One hundred percent of government procurement will be accomplished via EC/EDI by the year 2003, in accordance with FASA. Electronic Commerce/Electronic Data Interchange should not be confused with the Internet. Currently, the Internet is not as secure as it needs to be for contractual transactions.

In addition to FASA, further modifications to government procurement occurred with enactment of the Federal Acquisition Reform Act (FARA, included in Defense

Authorization Act) of 1996. The most significant change introduced by the 1996 procurement rules is that federal officials have been given more latitude to award contracts based upon a contractors performance or expertise, rather than price alone.

The Federal Acquisition Computer Network (FACNET) is the Federal Governments EC/EDI for the acquisition of supplies and services. This pivotal network will facilitate electronic data interchange of acquisition information between the government and the private sector. It employs nationally and internationally recognized data formats and provides universal user access. FACNET is a universal electronic capability that will permit potential contractors to, at a minimum, obtain information on proposed procurements, submit responses, query the system, and receive awards on a government-wide basis. The system, which should be fully functional in five years, is being designed to inform the public about federal contracting opportunities, outline the details of government solicitations, permit electronic submission of bids and proposals, facilitate responses to questions about solicitations, enhance the quality of data available about the acquisition process, and be accessible to anyone with a personal computer and a modem.

EDI is intended to increase business opportunities through wider diffusion of procurement information. There are to be fewer errors in data, reduced processing times, less reliance on human interpretation of data, and reduced unproductive time. Greater competition and reduced prices to the government are the ultimate goals. Savings are to be realized through reductions in inventories, mailroom sorting/distribution time, elimination of lost documents, and reduction in postage and other mailing costs. EDI will also facilitate the flow of better and more up-to-date information for enhanced management decision making.

Given the trajectory of FASA (1994) and the Federal Acquisition Reform Act of 1996, the trend in federal acquisition is clearly toward electronic commerce, electronic data and information interchange, procuring commercial items, and streamlining the complex procedural framework of the FAR and associated regulations.

SMALL BUSINESS STRENGTHS

Small businessescorporations in the manufacturing field that employ less than 500 people, or corporations in the services arena that generate less than $5 million in gross annual receiptsconstitute 99 percent of U.S. enterprises. These companies have certain competitive advantages: they are lean in terms of administration, they can position themselves in a market niche that large corporations cannot fill, and they can offer superior service to customers.11 In addition, small businesses have the potential to respond rapidly to emerging business opportunities because of fewer layers of management approval. Company policies can be modified quickly to meet client requests and requirements.12 Small businesses can carefully control their growth in terms of acquiring technical talent and penetrating new market sectors. The opportunity for excellent in-house communications up and down the ladder of

authority exists with small businesses. And because of the staffing deficit, people tend to become cross-trained and proficient in a wide variety of tasks. More people are given the chance to understand the big picture of the proposal life cycle and of specific business targets. Conversely, in large, multi-divisional corporations, very few staff fully understand the multi-dimensional complexities of massive procurement targets.

ORGANIZING YOUR COMPANY TO OBTAIN NEW BUSINESS

To support your companys efforts to obtain new and follow-on federal business, it would be well to consider forming a Business Development or Advanced Planning group.13 In many smaller firms, marketing and proposal efforts are handled exclusively through each division or line organization. One division may or may not be aware of such duplication with other divisions as marketing efforts, related contractual experience, and human talent in another division, and so forth. The formation of a centralized corporate Business Development Group (BDG) should not preclude a given divisions involvement in its own business planning and proposal development. Rather, the BDG can serve to focus, channel, and support divisional business-related activities in accordance with your companys formalized Mission Statement14 and Strategic Plan. Because of its corporate vantage, the BDG can help identify and make available the appropriate human talent, material resources, and information resident throughout your entire company in order to pursue a business opportunity. The functional charter of the BDG can also extend to include the following closely related activities:

Strategic, business, and marketing planning Business opportunity/Commerce Business Daily tracking and reporting Intelligence gathering: Marketing support/Client contacts Formalizing the process of establishing business objectives, gathering data,

analyzing data, prioritizing, and action planning Acquisition Team formation and guidance Coordinating the bid/no bid decision making process Coordinating teaming agreements Cost strategizing for proposals Company information management, distribution, and archiving (including

Proposal Data Center) Proposal and documentation standards development and dissemination Proposal coordination and production Company image development and public relations, as well as reputation

management15

Corporate communications (newsletters, etc.) Marketing and proposal management training

To ensure adequate connection with and visibility from senior management, a full-scale BDG should be under the leadership of a Vice President (VP) for Business Development. The most appropriate candidate for this pivotal position is an individual

with an advanced technical degree coupled with at least five years of demonstrated competitive business development experience in the Federal Government arena. Contacts alone are a necessary but insufficient gauge of a business developers successful performance.

Even under the constraints faced by very small companies (less than $5million in annual revenues), this VP functions most effectively when he or she is not obligated to be an on-the-road company marketeer as well as a business-development planner, organizer, and administrator. (In addition, an administrative assistant seems absolutely essential to enhance the functionality of the VP position.) Under this VPs guidance would be two primary functional groups: External and Internal Sales Support. The External Sales Support element might include full-time a corporate marketeer(s) as well as a key Division Manager(s).

The Internal Sales Support element might be subdivided further into Proposal Development/Production and Information Management, as depicted in Figure 1-1. Proposal Development/Production should logically include proposal publication. Proposal development and design becomes extremely challenging if it does not include oversight of the priorities and resources of the publications group. And finally, Information Management might include corporate communications, public relations, and corporate library activities. The function and focus of both the External as well as the Internal Sales Support elements are to project a professional, client-oriented corporate image. Understanding your clients business and demonstrating that understanding in every proposal you prepare is absolutely critical to your success in the federal marketplace in the mid-1990s and beyond (see Figure 1-2).

It has been the authors observation that small companies tend to undergo oscillations, and even convulsions, between centralized business development and control, and de-centralized divisional business activities. Instead of utilizing the BDG as a vital corporate support structure, some companies prefer to dissolve or de-scope the BDG, even after the one to two year extraordinary level of effort generally required to establish the BDG in the first place. Maintaining a full-time corporate business development staff can be significantly more effective than employees assigning to BDG support on a part-time basis only.

Figure 1-1

Suggested Business Development Group (BDG) Organization

Vice President for Business Development

Administrative Assistant

Internal Sales Support

Proposal Development/

Production

Information Management

Figure 1-2

Projecting An Appropriate Corporate Image

SUMMER 1997

STRATEGIC AND MISSION PLANNING

To ensure a planned pattern of growth for your company, strategic and mission planning are crucial considerations. According to Thompson and Strickland, a Strategic Plan is a comprehensive statement about the organizations mission and future direction, near-term and long-term performance targets, and how management intends to produce the desired results and fulfill the mission, given the organizations overall situation.16 Formulating, articulating, and implementing a meaningful Strategic Plan is critical for obtaining lines of credit through banking institutions, and for demonstrating to government auditing agencies that planned business growth over a period of time justifies an expansion in business base and therefore a long-term reduction in company overhead costs. Important questions to keep in mind when formulating a Strategic Plan for your company are listed below:17

What are your companys core competencies? Who are your companys principal clients? What are your principal services and/or products? What services or products

do your clients perceive you to be selling; these may be different than your internal perceptions?

Where is the intersection of what your clients value, and what your company does extremely well? What are the key commonalities shared by what your clients value?18

Where does your company want to be business-wise in five years? Develop a vision of the future state of your business.

What are your companys strengths and weaknesses (business, technical, human resources, public relations, fiscal, etc.)?

What business/economic opportunities and challenges face your company now? (Challenges include the entry of new competitors into the marketplace, adverse government policies, adverse demographic changes, etc.)19

Who are your primary competitors? Understanding how your competition thinks and acts is critical to your success. (Competitor analysis).20

EFFECTIVE MARKETING

Effective marketing involves far more than just selling. It is a long-term commitment and associated process that includes learning your clients requirements, constraints, and concerns, and understanding your clients respective business cultures. In the case of federal procurements, your company must determine that the particular target program is real and will be funded, that the RFP will be issued, and that political conditions within the specific government agency will allow for genuine competition in the case of a re-competition. Learn which civil servants will be on the evaluation board

55


for your procurement. Does the client expect to get the strongest proposal for the projected budget, or the most cost-effective solution? In effect, is budget or price the real driver of the acquisition? Over time and in a variety of ways, inform the client about your companys human talents, technical capabilities, past and present contractual experience, and financial stability in ways that demonstrate how you can help with his or her requirements. If your potential client does not know your firm, he or she is less likely to buy from you. Interact with your client at a variety of levelsfrom marketing and senior management staff to mid-level technical personnel and project managers.

One of the best ways to get information is to give information. Talk with your client; he or she is another human being who tends to respond to communication. Give your prospective client a chance to participate in your presentations. Get your technical people in front of the client to have meaningful discussions on issues prior to the release of the RFP. Introduce and reinforce themes or sales messages that will appear later in your proposal as authenticated assertions that set your company apart from your competitors. Indeed, a case could be made that the sale is made pre-RFP and that the proposal only closes the deal. A reasonably good proposal will not make up for inferior technical homework, poor management plans, and/or high costs.21 To which should be added inferior marketing intelligence about your client, teaming partners, and competitors. To be useful at all, the marketing information that your company collects must be analyzed, distributed, and archived so that it can be retrieved and updated easily. All information gathered that is relevant to a given procurement should be copied to the appropriate Proposal Manager. Appropriate marketing information must shape, and be fitted directly into, your proposal.

One potentially effective technique of intelligence gathering and analysis is for your companys external sales support staff and division managers to visit the contracting offices of your primary federal client agencies in order to obtain specific information on existing contracts. Learn the contract title and numbers of existing contracts, the incumbent contractor(s), the projected contract renewal date, and the nature of the work, as well as the staffing size, contract dollar amount, contract duration, and contract type. Ask for the names and telephone numbers of the Contracting Officers Technical Representative (COTR) and Contracting Officer (CO) for each contract. When you obtain this marketing intelligence, record it in a table, such as the one presented in Table 1-1. Try to gather information on both short- and long-term opportunities. One to two years is not too long a lead time for business planning.

56

Table 1-1

Sample Client Target Tracking Table

US Army Corps of Engineers

# Project Title

Acquisition Manager

Incumbent(s) Contract #

Anticipated RFP/CBD Synopsis Release Date

Duration of

Contract

Prime/Sub

Estimated Level of Effort (LOE) per Year

1

2

3

Once Client Target Tracking Tables are compiled for a given federal agency, preliminary bid/no bid decisions can be made in a planned and rational manner. Capture plans and associated call plans (direct contact plans) can then be generated to pursue the qualified list of targets. The types of information your company wants to collect in its capture plans include:

1. Project Name (according to government sources, i.e., use the clients name for this project)

2. Client Name (to the level of the line organization) and Organizational Chart 3. Client Mission Statement and Recent Developments in the Client

Environment 4. Incumbent Contractor (if applicable) 5. Potential Competitors 6. Specific Services or Products Being Procured 7. Key Evaluation Factors for Award/Client Success Criteria 8. Anticipated RFP Release Date 9. Anticipated Contract Award Date10. Source Evaluation Board (SEB) Membership11. Relevant Contractual Experience (for your company)12. Proposed Project Manager (for your company)13. Other Key Staff (from your company)

Call Plans

A call plan should be designed to clearly indicate who within your company will be visiting the client, as well as when, why, and at what level of the client organization. Information gathered from direct client contact should include the following:

Government organization and key personnel What is the client currently doing technically What is planned for the future Relevant documents/articles government and incumbent contractor personnel

have published Strengths and weaknesses of the incumbent. Whether your company can hire incumbent personnel Whether government personnel will sit on the SEB Who the competition is. What your company knows about it/them Whether your company is qualified to bid alone, or must team as a prime or

sub What your companys strengths and weaknesses are on the particular

opportunity. Whether you can correct your shortcomings in time

A very practical and relatively easy marketing technique applicable to contracts on which your company is the incumbent is for members of your companys executive and division-level management to visit the government staff as well as your on-site personnel on a regular basis. Build communication networks at both professional and appropriate social levels. Let your client know that his contract is important to your company by showing a real physical presence at his job site.

In addition, know your teaming partners, their technical expertise, human talent, cost strategies, contractual performance, financial stability, liabilities. Ask your client what they think of your prospective teaming partner. Consider the benefits and disadvantages of exclusive versus non-exclusive teaming arrangements.

Ethics in Marketing and Business Development

Ethical conduct in all business activities, whether domestic or international, should be your companys corporate standard at all times.22 The development and implementation of a formal set of ethics rules that prohibit any type of procurement fraud by employees, agents, or subcontractors would be a prudent step. Indeed, DoD FAR Supplement (DFARS) 203.7000, entitled Contractor Responsibility to Avoid Improper Business Practices, notes that a contractors management system should include a written code of business ethics and conduct, as well as an ethics training program for all employees. In their important work entitled Formation of Government Contracts, George Washington University law professors John Cibinic, Jr. and Ralph Nash, Jr. discuss standards of conduct in two broad categories: those dealing with

improper influence on Government decisions and those requiring honesty and disclosure of relevant facts in dealing with the Government.23

Part 3 of the Federal Acquisition Regulation (FAR) presents guidance regarding Improper Business Practices and Personal Conflicts of Interest. One example, the Anti-Kickback Act of 1986 (codified at 41 U.S.C. 51-58 [1988]) discussed in FAR 3.502-2, is legislation designed to deter subcontractors from making payments and contractors from accepting payments for the purpose of improperly obtaining or rewarding favorable treatment24 in connection with a prime contract or a subcontract relating to a prime contract. The term kickback means any money, fee, commission, credit, gift, gratuity, thing of value (e.g., promise of employment), or compensation of any kind. Unlike its precursor in 1946, the 1986 Act includes all types of government contracts. Finally, the 1986 Act places new reporting requirements on prime contractors and subcontractors. Possible violations of kickback laws must be reported in writing to the Inspector General (IG) of the appropriate federal agency, the head of the contracting agency if the agency has no IG, or the United States Department of Justice.

The Procurement Integrity Act, about which there is considerable confusion on the part of both civil servants and contractors, provides that the government cannot impart information to one bidder or proposer without making that information public to all potential bidders or proposers. And the False Claims Act Amendment of 1986 facilitated prosecution of complex defense procurement fraud cases.

Let us now take a closer look at what proposals really are; how they fit into a small business total marketing life cycle; and the planning, decisions, organization, and reviews required to prepare a successful proposal.

WHAT IS A PROPOSAL IN THE COMPETITIVE FEDERAL MARKETPLACE?

Technically, a proposal is an offer prepared by a contractor (such as your company) to perform a service, supply a product, or a combination of both in response to an RFP document issued by a procuring agency of the U.S. Federal Government. Functionally, a proposal is the final exam of your companys business acquisition effort. A proposal is designed to sell both technical and managerial capabilities of a company to accomplish all required activities on time and at a reasonable cost. Your companys proposal document(s) is scored, literally, by government evaluators against formalized, specific standards. A proposal is, first and foremost, a sales document. To be sure, it includes a host of technical, programmatic, institutional, pricing, and certification information, but it should remain sales oriented.

A proposal involves marketing your company to the federal government. It is a closing sales presentation, concretely supported by traceable and auditable credentials

and tailored to persuade your client to select your company for the award because you are best-qualified to achieve the results your client wants to achieve. 25 A proposal is, hopefully, continued dialogue with your customer. Proposals are the primary vehicle for winning new federal government business in the competitive arena.

To reiterate, a proposal (as well as a bid, i.e., a response to an Invitation for Bids [IFB]) is an offer or response.26 A proposal is not necessarily a contractual document. But under specific circumstances, an offer can be made into a contract by the government. The client uses your proposal and those of your competitors as the primary source of information upon which to base the selection of a winning contractor. Government evaluators are in no way obligated or encouraged to review publicly available material about your company that is not included within your proposal. That means that you have to ensure that all relevant and salient materials are included along with your proposal within the parameters of any page limitations.

A proposal is a package of carefully orchestrated arguments. Each section of a proposal should present arguments and meaningful evidence to convince your client that you should be awarded the contract because of the superiority of what you are proposing. To do this, you have to support the following messages in writing:

You understand your clients project requirements, critical issues, and success criteria.

Your approach satisfies all requirements (be careful not to imply that your approach exceeds requirements; this can result in the client thinking that you are proposing and charging for more than they are asking for).

Your approach offers tangible benefits to the client. Your approach minimizes schedule and cost risk. You are better (i.e., more reliable, more experienced, less expensive, etc.)

than your competitors.

Why us? and why not our competitors? These are the simple yet profound questions that internationally known and respected proposal consultant Hyman Silver 27

says proposals must answer effectively. Through the vehicle of your proposal, your company must finish the process of convincing your client to select you over your competition.

Proposals are important deliverable products. One carelessly written and poorly presented proposal can damage your companys reputation with your client or potential client. Experience suggests that clients tend to remember contractors which have submitted inferior proposal responses.

Figure 1-3

Overview of the Marketing and Proposal Lifecyle


The proposal life cycle (in particular, for procurement efforts on which your company is the prime contractor) begins well before the actual writing and production of a response to an RFP. Figure 1-3 presents an overview of the entire marketing and proposal life cycle.

THE RFP: AN OVERVIEW

The RFP is the culmination of a lengthy planning, budgeting, and approval process on the part of the Federal Government. It is a solicitation document issued by the government to obtain offers from contractors that propose to provide products or services under a contract to be awarded using the process of negotiation. The RFP is a complex document, often prepared under proposal-like conditions. That is to say, it is written and reviewed by a variety of civil servants under tight schedule constraints and is subject to delay caused by late inputs, protracted legal reviews, program modifications, changes in contractor support of the Statement of Work (SOW) development,28 etc. RFPs often contain conflicting or ambiguous requirements, particularly in Section L (Instructions to Offerers). This may result in part because RFP documents are often assembled using government boilerplate materials from previous or similar RFPs.

Contractors should not assume that the RFP reveals the full intent of the client s preferences, sentiments, or requirements. By the same token, care must be exercised to respond precisely to the RFP requirements and not build in additional levels of quality or uncalled-for services that will inflate your companys costs when compared with your competitors. Dont propose a Mercedes when the client wants a Buick! And a general rule is to take no deviations or exceptions to the stated RFP requirements and do not submit alternate proposals, even though they may be allowed.

RFPs can range from a few to thousands of pages and attachments for major aerospace and defense procurements. Certain RFPs are now being distributed via electronic media (diskettes, Internet, Federal Acquisition Computer Network [FACNET], and agency electronic bulletin boards [BBSs]) instead of in hard copy form. Some contractors elect to scan the RFP into electronic files using some type of optical character recognition (OCR) or intelligent character recognition (ICR) technology. Scanning can facilitate electronic searches for RFP requirements, which in turn can be transferred to electronically stored storyboard templates. A word of caution regarding scanningtime must be allotted to spell check electronically, review, and correct the scanned file. Scanning accuracy levels vary widely depending upon the scanner technology itself and the physical quality of the hard copy document.

62

SUMMER 1997

In accordance with the Uniform Contract Format (UCF) established at FAR 15.406-1, government Contracting Officers (COs) must prepare and assemble RFPs in a specific manner as enumerated below.

PART I THE SCHEDULE

Section A: Solicitation/Contract Form (Standard Form 33)Section B: Supplies or Services and Prices/CostsSection C: Description/Specifications/Work Statement(The SOW may include system specifications, contractor tasks and services, products, contract end items, data requirements, schedules, etc. It is an essential part of the RFP.)Section D: Packaging and MarkingSection E: Inspection and AcceptanceSection F: Deliveries or PerformanceSection G: Contract Administration DataSection H: Special Contract Requirements

PART II CONTRACT CLAUSES

Section I: Contract Clauses

PART III LIST OF DOCUMENTS, EXHIBITS AND OTHER ATTACHMENTS

Section J: List of Attachments

PART IV REPRESENTATIONS AND INSTRUCTIONS

Section K: Representations, Certifications and Other Statements of Offerors (Reps and Certs)

Section L: Instructions, Conditions, and Notices to OfferorsSection M: Evaluation Factors for Award(Used to determine proposal page allocations, writing emphases, and thematic structure.)

Presented a slightly different way, the RFP

Describes the requirement in Sections B, C, D, E, F, and (J) States the government agencys terms in Sections A, B, G, H, I, K, and (J) Describes the evaluation criteria in Section M Prescribes the proposal format and content in Section L Provides process information in Sections A and L

63


The contractors response to the governments RFP is called the proposal. Managing the proposal response process is a complicated and demanding task. Companies generally appoint a proposal manager to accomplish this task.

THE IMPORTANCE OF SECTION L

Section L of the RFP generally provides the specific instructions for preparing and structuring the proposal document. Margin requirements, font family and size, number of foldout (11" x 17") pages permitted, page count, and double-sided photocopying are among the publication parameters covered in this important section. In Section L you may also find specific guidance as to how your proposal should be structured in terms of the outline and numbering conventions. For example, you may see that your technical volume should consist of six major sections, each numbered according to the convention I.A, I.B, I.C, and so forth. Outlining of the proposal volumes should take into account guidance from Section L as well as Sections M and C, and other parts of the RFP as appropriate. Remember to use the verbiage from the RFP itself, particularly Section C, for building your proposal outline. Evaluators and their support staff will be looking for those same words.

SECTION M: THE EVALUATION CRITERIA

The general criteria by which the government will evaluate your proposal are presented in Section M of each RFP. In some cases (such as NASA RFPs), points are clearly allocated for each scored portion of the proposal. However, many times the evaluation criteria must be derived from somewhat vague narrative, as in the following example from an actual RFP.

CRITERIA

1. Offerors understanding of the problem and proposed technical approach.

2. Offerors applications-related experience.3. Experience and training of individuals who will work

under the contract.4. Offerors general experience in developing software

of comparable size, complexity and content to this project.

5. Offerors general experience in maintaining and operating (M&O) a computer-based system of

WEIGHT

Most important

Less importantSame importance as 2 above

Same importance as 2 above

Less important than 2 above

64

SUMMER 1997

comparable size, complexity, and content to this project.

6. Offerors proposed management plan.

Your companys strategy for responding to an RFP should definitely take into account the evaluation criteria. For example, if key and other résuméd personnel will count 50 out of 100 total points, then résumés and biographies should receive significant emphasis. Evaluation criteria should also serve as a guide for page allocations. Heavily weighted items should have an appropriately high number of pages allocated to them.

PRESENTING PROPOSALS IN EVALUATOR-FRIENDLY FORM

Publication of a set of high-quality proposal volumes is a professional-level, time-consuming, dedicated effort. It is not a lower-level clerical function that can be accomplished adequately and consistently by marshaling the secretarial and administrative support available within your company. Senior management is well served to recognize and support the publication professionals they have on staff, even if they number only two or three people. Support is most meaningful in the form of senior managements proactive involvement in answering that the proposal response schedule is met at every milestone. Otherwise, publication staff are faced with compensating for schedule slippages along the way. And when schedules are missed on a regular basis, and significant levels of night and weekend publication time become routine, publication staff morale is at risk. Turnover among publication staff can be highly detrimental to your companys ability to prepare outstanding proposal documents every time.

Proposal efforts are not complete when Technical and Management Sections are written and reviewed and the costing is completed. There remains:

Editing Proofreading Electronic and hardcopy configuration management Word processing/Desktop publishing Graphics generation/Preparation of prints for photocopying29

Outside printing coordination Photoreproduction Collation Assembly/Binding (3-ring vubinders, Velobinding, GBC, plasticoil, etc.) Quality checking Delivery

It is recommended that proposal publication be a centralized corporate function. Continuity of publication staff promotes uniformity of the image of the proposal documents from proposal to proposal. A core group of staff should become intimately familiar with and cross-trained in the document preparation process. Policies and procedures are much more easily implemented in a centralized working environment.

65


Document Configuration Management

Once sections of a proposal are written, it is critical to control the internal release of and changes to those sections throughout the proposal life cycle. Your centralized Publications Department should consider maintaining a master proposal book for each volume in three-ring binders. Such binders allow for quick page replacements. One set of master proposal books should be made available to the proposal manager, and another should be under the direct control of the Publications Department. The pages from a third and final master copy should be hung on the walls of your secure war room to facilitate comment and review. Then, as new proposal sections are created and existing ones modified with the authorization of the proposal manager, the Publications Department should be responsible for generating the change pages, inserting them in the master proposal books and hanging them on the walls of the war room. Establishing and maintaining a living master proposal document that iterates during the course of the response cycle will be of valuable assistance to the proposal contributors.

Your companys Publications Department should maintain electronic and hardcopies of the proposal documents at each review stage. For example, when the Publications Department produces the Red Team30 draft, all electronic files should be copied to a storage medium such as high-density diskette, compact disk (CD), or floptical (optical diskette). One hardcopy should be retained in the exact form and format of the Red Team draft. Then, if a computer hard disk fails, the network crashes, or electronic files are corrupted in some way, the last major draft version will be available, unmodified, in both electronic and hardcopy forms. It is advisable to back up all electronic proposal files to storage media at least once each work day.

A key aspect of proposal document configuration management is the use of time, date, file name on the hard drive, and diskette/CD/magneto-optical disk number headers on all draft proposal pages. Headers can be in a small font, and of course must be removed prior to submittal to the government. An example of a useful header for a page in the Technical Volume of a proposal to the Federal Aviation Administration (FAA) is provided below:

09:23 Tuesday, March 11, 1997 c:\proposal\A-1-3.FAA CD: Tech-A.1

Such headers allow easy identification of when a particular page was last modified, and on which storage medium the backup file for A-1-3.FAA resides. Another beneficial configuration technique is the use of colored paper for various proposal drafts pale blue paper for the Blue Team31, pale red for the Red Team, and so forth. Finally, do not discard draft proposal sections until after the proposal has been submitted. You can never be sure that you will not want to refer to an earlier version of the proposal. For security, keep all such materials in locked storage, and shred them when the proposal is submitted.

Graphics Are an Integral Part of the Proposal

Graphics, photographs, and appropriate images of all kinds will increase the government evaluators interest in and positive response to your companys proposal. It is essential to prevent submitting a boring, lackluster proposal to the government. Well-designed graphics can convey complex information in an easily understood format. And in page- or word-limited

66

SUMMER 1997

proposals, graphics can present significant quantities information in very limited space. Most proposals benefit from and many in fact require graphical presentations of Personnel Skills, Contractual Experience, Client-Contractor Interfaces, Company Organization, Project Organization, and Project Milestone Schedules. Photographs might be added to personalize your companys resumes, particularly if your client knows your key proposed staff. Appropriate photographs of company facilities, specialized computer equipment, engineering and manufacturing centers, off-the-shelf (OTS) products, and design prototypes can also greatly enhance your proposals sales value. Photodocumentation brings projects and products to life.

INCORPORATING TECHNICAL BRILLIANCE UP TO THE LAST MINUTE

One of the most challenging aspects of responding to a federal RFP is incorporating the very best materials into the final document within a very limited time frame. Technical and programmatic input, tailored boilerplate plans, marketing intelligence, nuances of corporate image, résumés, project descriptions of past performance, cost and pricing data, and legal opinionall of these elements must be brought together quickly and effectively. However, precisely because the proposal preparation cycle must be a controlled, choreographed process of planning, analysis, writing, multiple review and publication, potentially brilliant technical or programmatic ideas might not be incorporated into the final document. There simply may not be enough hours to integrate the change(s) into the text from a publication standpoint. Technical, management, and cost volumes of a proposal are ecological in the sense that a change in a system design concept affects cost, and an alteration in the Work Breakdown Structure (WBS) and Bid Task List affects the Program Plan and perhaps the cost as well. These wide-ranging changes take time to identify and make consistent across the proposal volumes in both the text and graphics. Unfortunately, the publication staff invariably are caught in the crossfire of last-minute changes. Their task is to generate a document, which in final form always takes more time to produce than in draft form.

The balancing act for the proposal manager is to allow the entire proposal process to remain fluid enough to accommodate evolutionary change, while simultaneously maintaining firm commitment to milestone schedules and completion of action items. Human and organizational dynamics come into play in graphic relief during the proposal process. Management commitment of sufficient human and material resources, bonus and incentive programs, effective cross-training, and the infusion of a winning attitude all can be brought to bear on the often arduous schedule of responding to federal procurements.

PIVOTAL ROLE OF YOUR PUBLICATIONS DEPARTMENT

A small companys Publications Department is often tasked with multiple responsibilitiesproposals, contract deliverables, presentations, marketing brochures, and so forth. The core staff should be cross-trained in a variety of operations, software applications, communications protocols, and hardware platforms. And it is critical that the Publications Department be able to accept, incorporate, and manage outside assistance. The Departments policies and procedures, software, and hardware should not be tailored so that temporary

67


employees or other in-house staff cannot support the Department effectively and efficiently during crunch (crisis) documentation periods. The supervisor or manager of your companys Publications Department might consider maintaining an active list of local freelance, part-time graphic artists, wordprocessors, desktop publishers, editors, and proofreaders for on-call requirements. And keep updated lists of local photography/visual imaging vendors and photoreprographics houses.

The continuity of Publications Department core personnel is very important for smoothly operating proposal efforts. Proposal managers benefit from seasoned, competent documentation staffers. Having established successful proposal production departments from the ground up for four federal contracting firms, I can testify to the benefits of human continuity within the publication group. Continuity, cross-training, and a positive attitude are salient elements in proposal publication success. Loss of skilled and cross-trained staff can cut deeply into productivity and inflate overall proposal publication costs.

In order to facilitate a smooth proposal operation, your company will need to determine the document throughput capacity of the Publications Department. How many new proposals of average size (for your firm) per month can be handled adequately with the core staff? If your bid/no bid process causes additional proposals to enter the publication pipeline each month, will additional staff, computer equipment, and floor space be needed? Keep in mind that an informed, aggressive bid/no bid process is the checkpoint that controls proposal document flowthrough. Your company should avoid overloading your Publications Department and your business development infrastructure with proposal efforts that have low-win probabilities. That practice wastes B & P money, and can be very detrimental to morale. Preparing and submitting proposals on a law of averages basisthe more we submit the better the chance of winningis a devastating and debilitating practice.

MODIFYING YOUR THINKING TO WIN

Your companys entry into the arena of competitive federal procurements brought with it the requirement for a fundamental shift in thinking and business-related behavior throughout the ranks of management as well as the professional and support staff. Senior management, for example, must appreciate the importance of federal proposals and proactively support the development and enhancement of the proposal response infrastructure. And each division of your company must make full use of the collective human and contractual expertise resident throughout the entire firm. Divisions cannot operate effectively in isolation from each other or from corporate direction. The dynamics of teamwork take on accentuated meaning in the competitive marketplace. People at all levels within the company must work together to collect and assess marketing intelligence, write proposals, review proposals, and publish proposals.

Proposals do not follow a fully democratic process. Companies must recognize that effective proposal management follows from authoritative, informed decision making. A proposal effort can become mired if people are not instructed to heed the guidance of the proposal manager. And senior management must clearly and repeatedly reinforce the role and authority of the proposal manager and the importance of meeting the schedule milestones.

68

SUMMER 1997

BUILDING A CLIENT-CENTERED WORK ETHIC

It is likely that your companys very existence and future growth depends upon winning 25-40% of the federal proposals you submit. There is no more important corporate activity than proposal development. Every resource in your company must be made available to the successful completion of any and all proposals. There are many details that must be attended to in the course of preparing a proposal which do not require the direct, hands-on attention of the proposal manager. Such activities include electronic searches of the résumé and project description subdirectories or data bases to determine appropriate staff to meet position descriptions and to ascertain relevant company contractual experience. Technical editing and tailoring of résumés, project descriptions, and the actual proposal volumes would be of invaluable assistance. In addition, administrative support in the form of photocopying, meal arrangements for evening and weekend work, and text-entry word processing are each vital to the overall success of the proposal preparation process.

No one person can address all of the details of technical content quality, writing consistency and compliance with the RFP. Therefore, your company would probably benefit from more people being trained and available on an as-needed basis to assist in the preparation of the proposal in addition to the technical authors, proposal manager, and Publications Department. The burden of preparing winning proposals must be shared among all professional and support staff. Everyones future depends upon winning!

PROPOSALS ARE IMPORTANT SALES DOCUMENTS

The essence of effective proposal writing is responding to the RFP requirements while convincing the client that your team, contractual experience, and technical and management approach are the very best. Remember that proposals are, first and foremost, sales documents. They are not technical monographs or users manuals.

Proposal managers must articulate clearly to their technical writing team exactly what their expectations and acceptance criteria are. The proposal manager must foster ongoing communication and feedback not only to himself, but also among all of the writers. This will help to ensure a consistent approach and mitigate rewriting.

Most proposals for a given procurement look and read essentially the same. The challenge is to incorporate well-substantiated information in your proposal that only your company can say. Identify precisely what will separate your company from the competition. If your company is the incumbent contractor for a particular project, use the names of your incumbent staff people throughout your proposal. Write to a level of technical detail that exceeds what non-incumbents could gain from your project monthly reports (obtained through the Freedom of Information Act [FOIA]),32 published articles, and conversations with government staff. Demonstrate that your company understands the technical risks as well as the success criteria.

Strive to have your proposals not be boring. Contributors should keep in mind the ABCs of proposal writing:

69


Accuracy Brevity Clarity

Twenty-five pages of thematically integrated, well-constructed, and compliant prose are much preferred to 50 pages that contain most every technical detail the writers happened to know. Technical data dumps glued together produce a very uneven and most likely non-compliant proposal. Keep your ideas focused on the clients requirements as stated in the RFP, and keep your sentences short. This author has seen sentences in proposal narrative that have exceeded 80 words.

Active Voice Adds Strength and Saves Space

Proposal writers should make every effort to employ the active voice. We accomplished XYZ, not XYZ was accomplished by us. The active voice adds strength to the proposal presentation, and can save as much as 14% in terms of space as compared with the passive voice.

Contributors should also attempt to vary sentence and paragraph structure. If you examine a proposal and most of the paragraphs on a page begin with The or your companys name, the narrative requires editing to infuse variety.

Writers should also attempt to employ strong and descriptive verbs, adjectives, and adverbs, such as in the following examples:

Our incumbent personnel all exceed the stipulated requirements . . . Over the course of 11 years of successful service to the Air Force Space Command,

our senior programmers developed and implemented . . . Our company offers field-tested prototyping experience . . . Fifteen analysts embody a knowledge base and legacy of experience that is

unmatched . . . We are a people-oriented firm with a very low turnover rate as measured against the

industry standard in the Washington metropolitan area . . . We will maintain technical continuity on this important Department of State

program . . . Our technical understanding is greatly enhanced by our past performance on two

critical contracts for the Office of Personnel Management . . . With an extensive inventory of ADP equipment that includes both Silicon Graphics

and Sun workstations to support off-site data processing, our company . . . Eight years of progressive support of the VA in Texas validates our companys

commitment to . . .

Guide the Government Evaluators Through Your Proposal

Formulate ideas precisely, concretely, and simply. To include most every technical detail each writer knows about a certain topic will probably not result in a successful proposal. This is

70

SUMMER 1997

where bullet drafts come into playto guide your writers in preparing an even, thematically consistent response. Proposals must appear as if written by a single, well-organized individual. Your proposal narrative must contribute to fast and easy comprehension by a variety of government evaluators. Use frequent subheadings to break up text and to facilitate the evaluation of your proposal.

Contributors should write from the general to the specific, from the easy-to-understand to the more difficult to comprehend. Proceed from an overview of each major topic to the finer-grained technical and programmatic details. Writers should identify and discuss tangible benefits to the client of your companys technical and management approach. There is a natural tendency to dwell on familiar ground. Often, this is of the least significant as far as the RFP requirements are concerned. Demonstrate how the features of your approach translate into benefits. Use factssupport all statements with concrete examples. Include:

What your company will do for the client. How your staff will do it. Why you will do it that way. What you did in the past, i.e., previous or current contractual experience. Highlight

relevant lesson learned from similar contractual experience.

Summarize the content of each proposal section in the first paragraph of discussion. Write for a variety of readers, including the skim or executive-level reader. To be a winner, a proposal must contain concise, understandable, and closely related thoughts. Identify the critical technical areas. In discussing them, use a level of detail that exceeds what a non-incumbent could use, but do not drown the reader in jargon or equations. Your proposal should be built upon solution-oriented writing. Writeups should be risk-aware and solution-oriented, and demonstrate an understanding of the evolutionary changes likely to occur over the life of the contract.

Make your companys responsiveness apparent to the evaluator over and over again. Use RFP terminology exactly (or in shortened form) for your proposal headings and subheadings. Employ the RFP terminology as a point of departure for further original writing. Do not simply recite the RFP verbiage in the actual narrative of your proposal. A DoD COTR once told me that he felt personally insulted when contractors, both large and small, merely replaced the words the contractor shall that appear in the SOW, with our company will. Such an approach demonstrates no technical understanding whatsoever.

Define acronyms and abbreviations the first time they are used in each major proposal section. Not all evaluators see the entire technical and management proposal volumes, so re-definition is certainly acceptable. Avoid sectional or page references in the text because references must be changed each time the section or page numbers change. References such as See Section 5.1.1 on page 5-34" should not be used. Let the compliance matrix33 and table of contents assist the evaluators in locating specific pages or sections. In addition, avoid fifth-order headings (e.g, Section 4.3.1.3.1).

71


Proposal writers should attempt to think graphically as well as in words. Every proposal section should have a figure or table associated with it that is referenced and discussed clearly in the text. Figure captions should appear centered beneath the appropriate figure. Table legends should appear centered above the appropriate table.

WELL-PLACED MANAGEMENT SUPPORT ENHANCES PROPOSAL SUCCESS

Small firms are in the enviable position of being able to respond rapidly to changing business environments. You should maximize the opportunity to establish as soon possible in your corporate history appropriate mechanisms and patterns of business thinking and behavior that will facilitate winning proposals in the federal marketplace. Genuine teamwork is a critical element of success in the federal contracting arena. Proposalmanship is a capability that needs to be cultivated throughout the levels of your organization in order to fully harness the talent and energy located there. Many companies struggle because only a few of their staff are trained and experienced in proposal management, design, and development.

Ongoing, positive communication throughout your company is a second measure of your successful business culture. That bi-directional communication extends between the Acquisition Team and the Proposal Team, Business Development staff and technical staff, Business Development staff and your client, proposal manager and senior management, proposal manager and proposal writers, and proposal manager and publications staff.

Management support must necessarily assume many different forms. But it must be manifested clearly in order to be fully effective. That support should be present in ensuring that efforts to build appropriate résumé and project summary files on your companys staff and contractual history are met with complete and timely support throughout the technical and programmatic ranks. Management support should take the shape of assisting the proposal manager in enforcing the proposal milestone schedule, and of underwriting the proposal managers authority for a given proposal effort. And management involvement and support also lies in committing the resourceshuman, financial, equipment, floor space, and so forthto make every proposal your company elects to pursue a superior sales document.

As entrepreneur the senior business leader, you are able to infuse an ethos of rational and formal planning into your companys business development and proposal development infrastructure. Expend the time and effort early in your firms history to develop a mission statement and strategic plan. Generate and have all of your management team follow written, albeit revisable, business development protocol. This applies particularly to bid/no bid decision making. Too many times, small and large businesses dilute their collective resources in pursuing marketing opportunities that do not support their lines of business. Planning also extends to developing the proposal Kickoff Package prior to conducting the formal Kickoff Meeting. Upfront planning, analysis, and decision making yield significant dividends downstream in the proposal process.

And finally, recognition.34 People need to know in a variety of ways that they are meeting your companys business and proposal development expectations. They need to be recognized

72

SUMMER 1997

when they contribute in any one of several ways to a proposal victory. Senior management should develop and implement a definitive recognition and incentive plan that is communicated clearly to everyone in your company. Success should be noted in a big way. Those people in your firm who are thinking to win should be recognized and should receive tangible benefits. We must remember that people are the most important ingredient in any companys success in the federal marketplace.

73

NOTES

1. See Federal Acquisition Regulation (FAR) Part 6. Established for the codification and publication of uniform policies and procedures for acquisition by all executive agencies, the Federal Acquisition Regulations System consists of the FAR, which is the primary document, and agency acquisition regulations that implement or supplement the FAR. The FAR System is articulated in Title 48 of the Code of Federal Regulations (CFR). The FAR is organized into Chapters, Subchapters, Parts (of which there are 53), Subparts, Sections, and Subsections. Example25.108-2: Part 25, Subpart 1, Section 08, and Subsection 2. The FAR is issued and maintained by the Department of Defense (DoD), the General Services Administration (GSA), and NASA under several statutory authorities of those agencies. Any critical understanding of the FAR must include the protest decisions of the Comptroller General and the General Services Board of Contract Appeals (GSBCA).

2. Competition is growing from Japanese, Western European, and emerging Eastern Bloc nations for U.S. Government contracts.

3. Do-It-Yourself Proposal Plan, TRW Space & Defense Proposal Operations, Rev. 3, February 1989, p. iii.

4. Rodney D. Stewart and Ann L. Stewart, Proposal Preparation (New York: John Wiley & Sons, 1984).

5. Proposal Preparation Handbook (Covina, Calif.: Procurement Associates, Inc., 1989), p. 1-2 (Vol. II).

6. Do-It-Yourself Proposal Plan, p. ii. 7. See FAR 15.609. Competitive range consists of those proposals that have a reasonable

chance of being selected for contract award. Win percentages can be presented in two very distinct ways: in terms of numbers of proposals won versus total number submitted, and also in terms of dollars awarded versus total potential dollars for all procurements on which you company proposed.

8. 8(a) refers to the U.S. Small Business Administrations (SBA) program to assist qualified small, woman-owned, and minority-owned businesses during the early years of their operation. 8(a) refers to Section 8(a) of the Small Business Act (15 U.S.C. 637(a)), which established a program that authorizes the SBA to enter into all types of contracts with other agencies and let subcontracts for performing those contracts to firms eligible for program participation based upon the criteria established in 13 CFR 124.101-113. The SBAs subcontractors are referred to as 8(a) contractors. (See FAR Subpart 19.8.) Eligibility in the program focuses on those groups that have been historically denied access to capital, educational resources, and markets. In 1995, the SBA listed 5,500 8(a) firms throughout the United States.

9. H. S. Cranston and Eric G. Flamholtz, The Problems of Success, Management Decision 26 (September 1988): 17.

10. Curtis Hartman and Steven Pearlstein, The Joy of Working, INC., November 1987, p. 62.11. Harold W. Fox, Strategic Superiorities of Small Size. Advanced Management Journal 51

(Winter 1986): 14.12. Hartman and Pearlstein, The Joy of Working. p. 67.13. The name of these business and planning groups varies from company to company. Some

are called Business Development Groups (BDGs), Advanced Planning, Strategic Planning and Business Development Groups (SP & BDGs), Special Programs groups, Marketing Departments, etc.

14. A Mission Statement is a concise, written expression of a companys long-term business, technical, and programmatic goals. It is a narrative outline of who we are, what we do, and where we are headed. (Arthur A. Thompson and A. J. Strickland, Strategy

Formulation and Implementation: Tasks of the General Manager , 4th ed. (Homewood, Ill. and Boston: BPI/IRWIN, 1989), p. 23.) Once formulated, this Statement should be communicated publicly to employees, clients, vendors, etc. One effective mechanism is to add a pull quote from your companys Mission Statement to your marketing brochures, annual reports, company letterhead, or newsletter. Some companies have poster-sized copies of their Mission Statement displayed prominently in their facilities to clearly convey their identity.

15. See Shari Caudron, Forget Image: Its Your Reputation That Matters, Industry Week, 3 February 1997, pp. 13-14, 16.

16. Arthur A. Thompson and A.J. Strickland, Strategy Formulation and Implementation, p. 19.17. Certain elements in this listing were adopted and modified from Thomas S. Piper (1989). A

Corporate Strategic Plan for General Sciences Corporation. Paper presented at the University of Maryland, College Park, Maryland.

18. See W. Chan Kim and Renée Mauborgne, Value Innovation: The Strategic Logic of High Growth. Harvard Business Review 75(1)(January-February 1997): pp. 103-104.

19. For additional business challenges, as well as opportunities, strengths, and weaknesses, see Thompson and Strickland, Strategy Formulation and Implementation, pp. 109-11.

20. Benchmarking is a rigorous process for linking competitive analysis to your company s strategy development. Benchmarking is a method measuring the performance of your best-in-class competitors relative to your industrys key success factors. It also is a mechanism for determining how the best-in-class achieve those performance levels.

21. Selected Viewgraphs from Judson LaFlash Seminars on Government Marketing and Proposals, Government Marketing Consultants, July 1980.

22. For a variety of articles that all focus on business and professional ethics (including international marketing ethics), see BRIDGES: An Interdisciplinary Journal of Theology, Philosophy, History, and Science 2(3/4) (FALL/WINTER 1990), entire issue.

23. John Cibinic and Ralph C. Nash, Formation of Government Contracts, 2d ed. (Washington, D.C.: The George Washington University, 1986), p. 107.

24. According to Ropes & Gray, examples of favorable treatment that are illegal include receiving confidential information on competitors bids, obtaining placement on a bidders list without meeting requisite qualifications, obtaining unwarranted waivers of deadlines, obtaining unwarranted price increases, and recovering improper expenses. Contractors should also be aware of the Fraud Awareness Letter issued in September 1987 by the DoD Council on Integrity and Management Improvements, which identified indicators of potential subcontractor kickbacks. See Complying With The Anti-Kickback Act: Guidelines and Procedures, Developments in Government Contract Law No. 10 (September 1990).

25. Herman Holtz and Terry Schmidt, The Winning Proposal: How to Write It (New York: McGraw-Hill, 1981).

26. The IFB, or sealed bid procedure, should result in a firm fixed price contract. This procedure is especially well-suited for government purchases of commercial off-the-shelf (COTS) items. Bids are opened publicly and the award is made immediately to the lowest priced, responsible, responsive bidder. No discussions are allowed, and no exceptions may be taken by the contractor.

The concept of responsiveness was developed in sealed bidding, wherein a Contracting Officer is prohibited from considering any bid that deviates from the IFB. This prohibition does not apply to negotiated procurements. (See Cibinic and Nash, Formation of Government Contracts, pp. 523-24).

An RFP, which results in competitive proposals for competitive acquisitions under Public Law 98-369 (Competition in Contracting Act of 1984), is a more flexible approach used in sole source or competitive situations. Any contract type from fixed price to cost reimbursement can be used. The government opens and reviews proposals privately. The number and identity of offerors is not revealed (immediately), and award may be based upon considerations other than price. Considerable discussion between the government and the offeror may or may not take place. The offeror may take exception to the solicitation terms or may submit alternate proposals. And modifications to proposals by the offerors are permitted.

27. This phrase was coined in 1975 by Hyman Silver. Mr. Silver, President of H. Silver and Associates (HSA) based in Los Angeles, California, held senior engineering and marketing positions with McDonnell Douglas and Rockwell International before founding HSA.

28. Many Statements of Work (SOWs) and Specifications are researched, written, and prepared by contractor personnel under separate contract to the government procuring agency. Companies which have contracts to prepare SOWs cannot legally compete on particular procurements for which they have prepared the Statement of Work.

29. For the best reproduction quality without scanning technology, black-and-white as well as color prints should be statted and screened prior to being pasted down and photocopied. Most printing companies can prepare your photos for photocopying for a nominal charge. And certain photocopiers come equipped with a specialized capacity to reproduce prints with reasonable quality.With a flatbed scanner, both black-and-white and color prints can be scanned directly into such applications as WordPerfect and Adobe PhotoShop. In a two-step process, these same prints can be scanned and imported into CorelDRAW!, Aldus Persuasion, and Aldus PageMaker.

30. The Red Team functions as an internal Source Evaluation Board, critiquing the proposal for compliance with RFP instructions and evaluation criteria (as found in Sections L and M, respectively, of the RFP). The Red Team looks for consistency and continuity among sections and volumes (inter-volume compatibility), inclusion of sales messages and win strategies, presence and substantiation of themes, clarity of text and artwork, and overall believability and persuasiveness. The Red Team serves as a recommending group. Red Team reviewers should include a number of people from within your company with appropriate technical competence and persons with a high degree of marketing and management competence. No reviewer should have participated in the proposal writing effort.

31. Blue Team evaluations are generally early course correction reviews. The fundamental technical architecture and programmatic direction of your companys proposal should be evaluated at this stage, deficiencies identified and articulated, and corrective action offered.

32. Freedom of Information Act (5 U.S.C. 552, as amended) provides that select information is to be made available to the public either by publication in the Federal Register, providing an opportunity to read or copy records, or providing copies of records. FOIA requests should be made to the appropriate agencies in written form. Letters should contain only one specific request. The reason for this practice is that an agency could decline to process one item of a multi-item request, and then the entire request would be returned to the contractor and the process would have to begin again. It is suggested that you begin your request with, Pursuant to the Freedom of Information Act, 5 U.S.C., Section 552 as amended, Company XYZ hereby requests a copy . . . . Keep an electronic log of each FOIA request, and place follow up calls to the agencies after 2 weeks has elapsed.

33. Compliance, cross reference, or traceability matrices are all terms for tables that clearly map the RFP requirements (from Sections C, L, M, and H as well as the Contract Data

Requirements List (CDRL), Data Item Descriptions (DIDS), and RFP attachments as appropriate) to the specific sections within your proposal in which your company responded to a given RFP requirement. These matrices generally appear in the front matter of your proposal volumes, but may be placed at the very end of each proposal volume as an 11" x 17" foldout page that will allow the government evaluators to be able to refer constantly and easily to the matrix.

34. In a recent reward and motivational survey conducted by Jo W. Manson of the Andrulis Research Corporation in Arlington, Virginia, it was determined that during the actual proposal response life cycle, special meals were used as a motivator by 25% of the respondents. After a proposal had been submitted, expressions of appreciation, bonuses, and time off were used as motivational tools. The survey was distributed to 400 members of the Association of Proposal Management Professionals (APMP ) and resulted in a 25% response rate. Expressions of appreciation included letters, verbal praise, win parties, and publicity within the company.

REFERENCES

Caudron, S. (1997). Forget Image: Its Your Reputation That Matters. Industry Week, 246(3), 13-14, 16.

Cibinc, J. & Nash, R.C. (1986). Formation of Government Contracts. 2d ed. Washington, DC: TheGeorge Washington University.

Complying with the Anti-Kickback Act. (1990). Developments in Government Contract Law, No. 10. Ropes & Gray 1001 Pennsylvania Avenue, N.W. Suite 1200 South

Washington, DC 20004.Cranston, H.S. & Flamholtz, E.G. (1988). The Problems of Success. Management Decision,

26, 17.Do-It-Yourself Proposal Plan. (1989). TRW Space & Defense Proposal Operations, Rev.

3.Contractor Responsibility to Avoid Improper Business Practices, DFARS 203.7000Federal Acqusition Streamlining Act Enacted. (1995). Business Credit, 97, 6.Fox, H.W. (1986). Strategic Superiorities of Small Size. Advanced Management Journal,

51, 14.From Red Tape to Results: Creating a Government That Works Better and Costs Less,

(1993). (see www.npr.gov)Tests conducted by Robert S. Frey for EER Systems, Vienna, Virginia.Hartman, C. & Pearlstein, S. (1987). The Joy of Working. INC., November 1987, 61-71.Holtz, H. & Schmidt, T. (1981). The Winning Proposal: How to Write It . NY: McGraw-

Hill.Kim, W.C. & Mauborgne, R. (1997). Value Innovation: The Strategic Logic of High

Growth. Harvard Business Review, 75(1), 103-112.Osborne, David & Gaebler, Ted. (1992). Reinvesting Government: How the Entrepreneurial

Spirit Is Transforming the Public Sector. NY: Addison Wesley.Piper, T. S. (1989). A Corporate Strategic Plan for General Sciences Corporation.

(unpublished manuscript)Proposal Preparation Manual. (1989). 2 vols. Covina, CA.: Procurement Associates, Inc.

(revised).Selected Viewgraphs from Judson LaFlash Seminars on Government Marketing and

Proposals. (1980) Government Marketing Consultants.Stewart, R.D. & Stewart, A.L. (1984). Proposal Preparation. NY: John Wiley & Sons.Thompson, A.A. & Strickland, A.J. (1989). Strategy Formulation and Implementation:

Tasks of the General Manager. 4th ed. Homewood, IL and Boston: BPI/IRWIN.

ASIAN-AMERICANS AND CAUCASIANS IN THE

WORKPLACE: SOME ATTITUDINAL COMPARISONS

Golnaz Sadri *

The work force is now composed of a greater mix of culturally diverse employees than ever before, making an understanding of how such diversity impacts the work place of critical importance to organizational success. The present project examines differences in two important work-related cognitions, locus of control and job satisfaction, across two cultural groups, Asian-Americans and Caucasians. Studies show that social cognitions have an impact on a wide range of organizational behaviors including assessment and training; goal-setting; job performance; leadership effectiveness and employee job satisfaction. T-tests to examine differences across the two groups included in the present research showed that the Asian-American sample obtained a more external locus of control and a lower level of job satisfaction than the Caucasian sample. Implications of the results for the management of behavior in organizations are discussed.

number of trends within the United States make an understanding of differences between culturally diverse groups imperative for effective management in today's work place. Between now and the year 2010, the population in the United States is projected to grow by 42 million. Hispanics are projected to account for 47% of this growth; Blacks for 22%; Asians and other people of color for 18%, while Whites will account for only 13% of the increase (Randle, 1990). By the year 2000, women and immigrants will constitute 85% of the net growth in the United States labor force (Johnston and Packer, 1987). A national study of almost 3000 worker attitudes indicated that more than 20% of people of color reported being discriminated against, with a corresponding higher incidence of burnout, reduced willingness to take risks and increased intention to quit (Shellenbarger, 1993). Understanding differences between cultural groups can assist not only in the management of a diverse labor force but can also contribute to the

* Golnaz Sadri is an Associate Professor of Management in the School of Business Administration and Economics at the California State University, Fullerton, Fullerton, California.

Manuscript received, June, 1996, revised, March, 1997.

A

progressive notion of valuing diversity as a competitive resource (Cox, 1993; Loden and Rosener, 1991). One major contributor to the process of valuing diversity is the availability of research on how diversity affects organizational behavior, such research is evolving (Fukuyama and Greenfield, 1983; Hackett, Lent and Greenhaus, 1991; Weitzel and Waller, 1990; Zane, Sue, Hu and Kwon, 1991). The present project aims to contribute to this stream of research by examining differences in two important work-related cognitions, locus of control and job satisfaction, across two cultural groups, Asian-Americans and Caucasians.

LITERATURE REVIEW

The role of social cognitions in influencing organizational behavior has gained momentum over the past several decades. Studies looking at the influence of social cognitions on organizational behavior include such diverse topics as goal-setting (Latham and Yukl, 1975; Taylor, Locke, Lee and Gist, 1984), vocational behavior (Betz and Hackett, 1981; Betz and Hackett, 1983; Multon, Brown and Lent, 1991; Taylor, 1981; Taylor and Betz, 1983), assessment and training (Brockner, 1988; Frayne and Latham, 1987; Luthans and Kreitner, 1985; Manz and Sims, 1980; Snyder, 1987) and job performance (Barling and Beattie, 1983; Robertson and Sadri, 1993; Sadri and Robertson, 1993). House (1971; 1987) suggests that, among other variables, social cognitions affect leadership effectiveness. He suggests that leaders need to choose from four styles of leadership (directive, participative, achievement oriented and supportive) by examining two classes of contingency factors: environmental factors (task structure, formal authority systems and work group relationships) and subordinate characteristics (locus of control, experience and perceived ability). The correct match between leadership style and these contingency factors will affect job performance and employee job satisfaction. The present research aims to examine two constructs within this model (locus of control and job satisfaction) within a cross-cultural framework.

Locus of Control. Locus of control may be defined as a general expectancy regarding an individual's ability to control the major events in his/her life. Those who believe that they are masters of their fate are labelled internals while those who believe that their lives are reliant on luck, chance, fate or powerful others are classified as externals (Rotter, 1966). There is evidence that internals and externals interpret information with respect to their occupations in different manners (Sadri and Marcoulides, 1994; Spector, 1987). Internals perceive a greater degree of control over their work situation; they report having more autonomy and receiving more feedback than externals (Hammer and Vardi, 1981; Kimmons and Greenhaus, 1976). Internals are reported to view their jobs as more purposeful. They hold higher expectancies that effort will lead to enhanced performance, and exhibit greater personal career effectiveness when they feel that effort and performance will lead to valued rewards. Internals also show higher levels of job involvement and display a higher level of work motivation (Organ and Greene, 1974; Spector, 1982).

Externals tend to have higher absenteeism rates, are more anxious, and are more alienated from the work setting (Spector, 1982). Externals tend to perceive their jobs as more stressful than internals (Anderson, Hellriegel and Slocum, 1977; Gemmill and Heisler, 1972). Furthermore, externals report greater psychological strains resulting from job specificity; more somatic complaints as a result of role conflict; and are more likely to respond to normal organizational frustrations with aggression, sabotage, or withdrawal than are internals (Anderson et al, 1977; Fusilier, Ganster and Mayes, 1987; Gemmill and Heisler, 1972; Marino and White, 1985; Sadri and Marcoulides, 1994; Storms and Spector, 1987).

Job Satisfaction. Job satisfaction has been the subject of job-related research for over 50 years (Iaffaldano and Muchinsky, 1985). Job satisfaction refers to an affective or emotional response toward different facets of one's job, including financial compensation, colleagues, opportunities for promotion and challenge (Quinn and Shepard, 1974). It has been suggested that job satisfaction is linked to a number of work behaviors and attitudes such as job performance, absenteeism, turnover, organizational commitment and pro-union voting behavior. A brief review of studies examining job satisfaction is presented below.

The strongest empirical link between job satisfaction and work-related behavior relates to the negative relationship between job satisfaction and employee turnover. Mobley (1977) suggested that a primary result of job dissatisfaction is the stimulation of thoughts of leaving the organization which in turn, lead to the withdrawal decision and behavior. This general pattern of relationships has been supported in a number of studies (Griffeth and Hom, 1988; Lance, 1988; Lee and Mowday, 1987; Mobley, Horner and Hollingsworth, 1978). Job satisfaction has also been shown to be negatively related to pro-union voting behavior (Heneman and Sandver, 1983; Klandermans, 1989; Mellor, 1990) and positively related to organizational citizenship behavior (behaviors that exceed the call of duty; Fisher and Locke, 1992; Organ, 1990; Podsakoff, Mackenzie and Hui, 1993).

While job satisfaction has also been linked to higher levels of absenteeism and lower levels of job performance, such relationships have demonstrated a much weaker magnitude than that to employee turnover (Hackett and Guion, 1985; Scott and Taylor, 1985). However, job satisfaction does show a link with a number of very important work-related and life attitudes. There is a strong link between job satisfaction and an individual's level of organizational commitment (defined as the extent to which an individual identifies with an organization and is committed to its goals; Mathieu and Zajac, 1990). Furthermore, job satisfaction is negatively linked to levels of physical and mental ill-health and positively related to life satisfaction (Cooper and Sadri, 1991; Ivancevich and Matteson, 1980; Jenkins, 1971; Judge and Hulin, 1993).

Locus of Control and Job Satisfaction. A number of studies employing diverse samples (including police officers, nurses, managers and professionals) have examined the relationship between locus of control and job satisfaction. The general finding

which emerges from these studies is that internals tend to be more satisfied with their jobs than are externals (Gemmill and Heisler, 1972; Lester and Genz, 1978; Munoz, 1973; Organ and Greene, 1974). Moreover, there is a stronger relationship between job satisfaction and performance for internals than externals. Evidence indicates that internals obtain higher salaries and greater salary increases than externals (Andrisani and Nestel, 1976; Nystrom, 1983).

Self-Expectancies and Cultural Background. One variable which remains under-researched in the study of both locus of control and job satisfaction is culture. Hofstede (1984) defines culture as a form of collective mental programming of a group of people; and research shows that such programming operates as a powerful influence on a person's beliefs, values and actions (Hackett, Lent and Greenhaus, 1991; Lent and Hackett, 1987; Weitzel and Waller, 1990). Fukuyama and Greenfield (1983) found that Asian-Americans students obtained lower assertion scores than Caucasian students on a number of dimensions (such as expressing feelings or making difficult requests, making requests in public, expressing compliments, disagreements with parents and expressing annoyance to the opposite sex). Zane, Sue, Hu and Kwon (1991) also found differences between Asian-Americans and Caucasians in self-efficacy (self-expectations of performance) and self-reported assertive responding.

HYPOTHESES

In attempting to formulate some hypotheses concerning how the two groups included in this study (Asian-Americans and Caucasians) might differ, it is necessary to review previous research in this area. Hofstede (1984; 1991) found differences between the dominant American culture and many of the Asian cultures which he surveyed on three of his four dimensions of national culture (power distance, uncertainty avoidance and individualism versus collectivism). On the dimension of power distance (the pattern of interpersonal relationships when differences in power are perceived) America emerged as below average while all of the Asian countries included in Hofstede's study (the Philippines, Singapore, Hong Kong, Thailand, Taiwan and Japan) emerged as above average. On the dimension of uncertainty avoidance (the extent to which people are threatened by ambiguous situations or stimuli and have beliefs and institutions that help them to avoid this uncertainty) America was again below average, while most of the Asian countries (included here were Japan, Taiwan and Thailand) were above average. In terms of individualism versus collectivism (the extent to which individuals are concerned with the welfare of themselves and their immediate family as opposed to the welfare of the group) America emerged as a highly individualistic country whereas all of the Asian countries surveyed emerged as collectivistic.

Hall's construct of high-versus low-context has been used extensively in cross-cultural research. He suggests that in high-context cultures, people rely heavily on situational cues such as status or position for meaning when communicating with others, while in low-context cultures, written and spoken words are heavily relied upon

in important communication (Hall, 1973; 1976; 1983). America is a low-context culture, Asian cultures are predominantly high-context (Dulek, Fielden and Hill, 1991; Gudykunst, 1988, 1991).

Three of the dimensions of culture identified by Kluckhohn and Strodtbeck (1961) serve to further identify differences between the two groups identified here. The three dimensions are: relationship to the environment, time orientation, and focus of responsibility. Relationship to the environment addresses societal perceptions of whether individuals are subjugated to their environment, in harmony with it or able to dominate it. Individuals living in North America generally believe that they can control nature while most Asian cultures seek harmony with it. Time orientation addresses whether the culture focuses on the past, present, or future. The American culture has typically focussed on the present and the immediate future while the Asian cultures have generally taken a longer term perspective. Focus of responsibility addresses the issue of where responsibility lies for the welfare of others in society. As stated above, Americans are basically individualistic and believe that responsibility lies with the individual; Asian cultures are more group oriented and believe that responsibility lies with the larger group (Kluckhohn & Strodtbeck, 1961).

The dimensions of national culture identified above serve to show that there is some reason to believe that there are different perceptions of reality between the two cultural groups included in this study. The following are some tentative suggestions about how such cultural differences might impact perceptions of locus of control and job satisfaction. Cultures which place greater importance on the individual and his/her roles and responsibilities (as opposed to the group) are likely to lead towards a greater internal locus of control in the general population. Also, cultures which adopt an orientation of control toward the environment would be expected to produce a more internal locus of control orientation. Conversely, uncertainty avoidance as a cultural attribute communicates less tolerance of risk and ambiguity (with the implication that such risk and ambiguity is less controllable by individual members of society). Thus, cultures high on uncertainty avoidance would be likely to exhibit greater levels of external locus of control among the general population. The two dimensions of power distance and high-context are cultural attributes which place great importance on one's position and other contextual factors (such as age). This implies that an individual's position in life is, in large part, a determinant of the important events which occur in his/her life. Again, this would tend to promote a more external locus of control orientation.

Based on these tentative suggestions, it is hypothesized that the Caucasian group included in this study will obtain a more internal locus of control than the Asian-American group. Consistent with previous research in the field demonstrating that internals are more satisfied with their jobs than are externals (Gemmill and Heisler, 1972; Lester and Genz, 1978; Munoz, 1973; Organ and Greene, 1974), it is hypothesized that the Caucasian sample included here will also show a higher level of

job satisfaction than the Asian-American group. In summary, the present research aims to test the following two hypotheses:H1: The Caucasian group included in this study will obtain a more internal locus of control than the Asian-American group.H2: The Caucasian group included in this study will demonstrate a higher level of job satisfaction than the Asian-American group.

METHOD

Subjects

Data were collected from respondents at a major university in the Southern California area. Questionnaires were distributed on campus in the library and the student union building by volunteers. The initial number of questionnaires distributed was 270, of which 205 were returned (a response rate of 76%). In total, 85% of the sample respondentswere employed at the time of data collection: 64% of respondents worked full-time, 21% worked part-time and 15% were unemployed. The unemployed respondents were asked to think about their last job when responding to the job satisfaction questions described below. Across the sample, 47% were male and 53% female. The median age of the sample was between 25 and 34 years (62% fall into this category). Data collected on the ethnic-racial background of participants showed that 100 were Caucasian, 90 Asian-American, and 15 from a number of other ethnic-racial backgrounds. Due to the small sample size, the third group was excluded from the present study. The final sample size was 190 respondents, 53% of whom were Caucasian and 47% Asian-American. The two groups were comparable in age, gender and working status.

Measures

Locus of Control. Locus of control was assessed though an abbreviated version of Rotter's locus of control scale (Rotter, 1966). The measure used in the present study consists of seven items. Each item presents two alternatives, and respondents were required to choose the response which most closely matched their perception. Each response which indicated an internal locus of control orientation was given a positive score. Hence, the higher the score, the greater the internal locus of control orientation.

Job Satisfaction. Job satisfaction was measured using a 35-item questionnaire designed to measure six aspects of the job (Quinn and Shepard, 1974). These six components include satisfaction with the degree of comfort (7 items), financial rewards (3 items), resource adequacy (11 items), challenge (6 items), co-workers (4 items) and promotion (3 items). Responses were summed for each set of items to derive facet-specific subscale scores. In addition to the facet-specific questions, one facet-free question asked respondents how satisfied they were with their jobs on a general level.

Copies of the locus of control and job satisfaction questionnaires used in this study are provided in Appendix A.

RESULTS

All the quantitative analysis was conducted using the Statistical Package for the Social Sciences (SPSS; Norusis, 1990).

Table 1

Descriptive Statistics and Alpha Coefficients of Reliability forLocus of Control and Job Description

Variable Mean S.D. Alpha

Locus 4.28 1.84 .64

Comfort 20.01 3.54 .67

Challenge 17.8 4.04 .85

Rewards 7.99 2.30 .70

Co-workers 12.08 2.55 .78

Adequacy 32.04 6.71 .90

Promotion 7.38 2.48 .84

Facet-free 2.88 0.86

Table 1 shows summary statistics and Cronbach alpha coefficients of reliability for the locus of control scale and the six facets of the job satisfaction measure. As can be seen, the alpha coefficients for all the scales are at acceptable levels (.64 and above) which indicates reasonable levels of internal consistency for the locus of control scale and the six subscales of the job satisfaction measure. No alpha is presented for facet-free job satisfaction as this is a single item indicator.

Table 2 shows the correlation matrix for locus of control and the seven aspects of job satisfaction measured in this study. It can be seen that there is a high level of correlation between the seven aspects of job satisfaction. This indicates that individuals satisfied with one facet of their jobs were also satisfied with other aspects of their job and were also generally satisfied (facet-free). Furthermore, respondents with an internal locus of control showed a significantly higher level of satisfaction with the challenge offered in their jobs (p<.001) and the rewards received from their jobs (p<.01).

Table 2

Correlation Matrix for Locus of Control and Job Satisfaction

Variables 1. 2. 3. 4. 5. 6. 7.

1. Locus

2. Comfort .12

3. Chanllenge .25** .29**

4. Rewards .21* .39** .44**

5. Co-workers .18 .44** .47** .49**

6. Adequacy .14 .45** .50** .47** .62**

7. Promotion .14 .33** .50** .45** .46** .56**

8. Facet-free .14 .43** .54** .53** .64** .62** .62**

*p<.01; **p<.001

Table 3 shows t-tests for locus of control and job satisfaction across the two groups. It may be seen that the Asian-American group obtained a significantly more external locus of control (p<.000). Furthermore, the Asian-American respondents included in the present study were significantly less satisfied with a number of aspects of their jobs. The Asian-American group obtained significantly lower mean scores for levels of physical comfort at work, the amount of challenge their jobs offered, the type and amount of rewards, their co-workers, the adequacy of resources necessary to do their jobs and the promotional opportunities offered by their jobs. The Asian-American group also scored lower on the facet-free item, indicating a lower level of overall satisfaction with their jobs.

Table 3

T-tests for Differences in Mean Scores Across the Caucasian and Asian Groups on Locus of Control and Job Satisfaction

Variable Mean S.D. T-value

Locus of Control

Caucasian 4.81 1.75 3.72***

Asian 3.83 1.84

Comfort

Caucasian 21.12 3.33 4.50***

Asian 18.86 3.52

Challenge

Caucasian 18.82 4.14 3.25**

Asian 16.96 3.71

Rewards

Caucasian 8.55 2.26 3.10**

Asian 7.52 2.31

Co-workers

Caucasian 12.71 2.53 3.34**

Asian 11.49 2.48

Adequacy

Caucasian 33.87 6.18 4.24***

Asian 29.88 6.65

Promotion

Caucasian 7.99 2.26 3.14**

Asian 6.88 2.58

Facet-free

Caucasian 3.00 .82 2.20*

Asian 2.72 .91

*p<.01 **p<.001 ***p<.0001DISCUSSION AND CONCLUSIONS

As stated earlier, the reality of working in multi-cultural environments, in multinational companies and in a global marketplace has made an understanding of cultural differences imperative. The United States labor force is becoming more culturally diverse and valuing diversity is likely to become a key competitive advantage (Kiplinger and Kiplinger, 1989; Loden and Rosener, 1991). Valuing diversity assumes that differences between groups based on cultural background are understood and accepted. The present research aimed to take one step toward understanding such differences by examining differences in locus of control and job satisfaction across two cultural groups: Asian-Americans and Caucasians. Prior research demonstrates that the dominant American culture does differ from the dominant cultures observed in Asian countries. For example, Hofstede (1984; 1991) found major differences across these cultures on three of his four dimensions of culture (power distance, uncertainty avoidance and individualism versus collectivism). The dominant American culture is low-context while Asian cultures are predominantly high-context (Dulek, Fielden and Hill, 1991). Individuals living in North America generally believe that they can control nature while most Asian cultures seek harmony with nature. Americans and Asians also differ in terms of their time orientation, Americans typically take a short-term perspective whereas the Asian cultures generally take a longer term view (Kluckhohn & Strodtbeck, 1961). The present results highlighted some more micro-focussed differences between these two cultural groups: the Asian-American group included in this study showed a more external locus of control and a lower level of job satisfaction than the Caucasian group.

The constructs of locus of control and job satisfaction are important because they influence perceptions, experiences and behavior in organizations. Studies show that individuals with an internal locus of control are more satisfied with their work, suffer from less stress-related outcomes and have higher levels of organizational commitment and life satisfaction (Anderson et al, 1977; Fusilier et al 1987; Gemmill and Heisler, 1972; Sadri and Marcoulides, 1994; Storms and Spector, 1987). Research emanating from the Path-Goal theory of leadership (House, 1971; 1987) shows that employees with an external locus of control are more satisfied with a directive style of leadership while those with an internal locus of control are more satisfied with a participative style (Indik, 1986; Keller, 1989; Wofford and Liska, 1993).

Further research is needed to investigate whether Asian-American respondents consistently report a more external locus of control. Should such a pattern emerge, then it is plausible to assume that the experience of Asian-Americans in a workplace which overlooks this will be adversely affected. This was substantiated in the present research. The present results have significant implications for the management of a diverse work

force. Currently, work in most American organizations is structured in a manner which tends to motivate or satisfy individuals with an internal locus of control. The current interest in employee empowerment is likely to make the work context more so. Research shows that employees with an external locus of control prefer more structure, more direction and higher levels of supervision (Indik, 1986; Keller, 1989; Wofford and Liska, 1993). Individuals with an external locus of control and those from collectivistic cultures are likely to value group level reward systems and a more collaborative work environment which fosters mutual (rather than individual) dependence.

More research of a similar nature is needed to determine whether different environments are consistently perceived as more satisfying for individuals from different cultural backgrounds. If so, often relatively minor changes in the structure of the workplace or the job itself may accommodate such differences. Clearly, it is not always possible to adapt one's organization structure to suit different types of employees, and in such instances, managers (avoiding the assumption that employees are predominantly homogeneous) should explicitly identify expectations and explain procedures. Previous studies have shown that job redesign techniques such as increased participation in decision making and improvements in communication within the American organization have resulted in positive individual and organizational outcomes (Jackson, 1983; Pierce and Newstrom, 1983; Wall and Clegg, 1981). Future research needs to address the role of such techniques within a multi-cultural setting. It is important to note that the present results do not imply differences in productivity or effectiveness across the two groups included here, and are not intended to suggest that one way of observing reality is better than another. However, these results do suggest that to ignore differences may be detrimental to organizations operating in today's multi-cultural environment.

There are a number of shortcomings with the present research. First, only two ethnic-racial groups were included which is not representative of the truly multi-cultural environment of the bulk of organizations operating within America and across the world. Future research of a similar nature might try to incorporate a more ethnically diverse mix of respondents. A second shortcoming with this research is that the respondents were working for a diverse range of companies in a diverse range of occupations and there are no statistical controls in the analysis for differences in occupation, job type and job tenure. Further research which controls for such disparities may shed more light on the exclusive influence of cultural background on work-related attitudes. Third, the present study examined work attitudes (which are of great importance to the practice of management) but lacked any measures of job performance. Again, future researchers would make a significant contribution to the field if performance measures are incorporated into their research along with measures of work-related perceptions and attitudes. This research presents one small step in the evolving stream of research addressing the effects of a culturally diverse work force on organizational behavior. Such research serves to show where culture is important and where it is not so that the role of culture is neither over- nor under-stated (Yeh and

Lawrence, 1995). Research of this nature is likely to have significant implications for the management of behavior in organizations in the 1990s and beyond.

Appendix A

Questionnaire Used to Measure Locus of Control

Locus of Control

1. A. Many of the unhappy things in people's lives are partly due to bad luck.B. People's misfortunes result from the mistakes they make.

2. A. Unfortunately an individual's worth often passes unrecognized no matter how hard he tries.

B. In the long run, people get the respect they deserve.3. A. Without the right breaks one can not be an effective leader.

B. Capable people who fail to become leaders have not taken advantage of their opportunities.

4. A. I have often found that what is going to happen will happen.B. Trusting to fate has never turned out as well for me as making a decision to

take a definite course of action.5. A. Most people don't realize the extent to which their lives are controlled by

accidental happenings.B. There really is no such thing as "luck."

6. A. In the long run, the bad things that happen to us are balanced by the good ones.

B. Most misfortunes are the result of lack of ability, ignorance, laziness or all three.7. A. Many times I feel I have little influence over the things that happen to me.

B. It is impossible for me to believe that chance or luck plays an important role in my life.

Questionnaire Used to Measure Job Satisfaction

Job Satisfaction

1. I have enough time to get the job done. 2. The hours are good. 3. Travel to and from work is convenient. 4. The physical surroundings are pleasant. 5. I can forget about my personal problems. 6. I am free from the conflicting demands that other people make of me. 7. I am not asked to do excessive amounts of work. 8. The work is interesting. 9. I have an opportunity to develop my own special abilities.10. I can see the results of my work.11. I am given a chance to do the things I do best.12. I am give a lot of freedom to decide how I do my own work.13. The problems I am expected to solve are hard enough.14. The pay is good.15. The job security is good.16. My fringe benefits are good.17. The people I work with are helpful.18. The people I work with are friendly.19. I am given a lot of chances to make friends.20. The people I work with take a personal interest in me.21. I have enough information to get the job done.22. I receive enough help and equipment to get the job done.23. I have enough authority to do my job.24. My supervisor is competent in doing his/her job.25. My responsibilities are clearly defined.26. The people I work with are competent in doing their jobs.27. My supervisor is very concerned about the welfare of those under him/her.28. My supervisor is successful in getting people to work together.29. My supervisor is helpful to me in getting my job done.30. The people I work with are helpful to me in getting my job done.31. My supervisor is friendly.32. Promotions are handled fairly.33. The chances for promotion are good.34. My employer is concerned about giving everyone a chance to get ahead.35. Now, taking everything into consideration, how do you feel about your job as a whole?

REFERENCES

Anderson, C.R., D. Hellriegel, and J.W. Slocum. (1977). "Managerial Response to Environmental Induced Stress." Academy of Management Journal, 6 (2),

260-272.Andrisani, P.J. and G. Nestel. (1976). "Internal-External Control as Contributor to and

Outcome of Work Experience." Journal of Applied Psychology, 61 (2), 156-165.Barling, J., and R. Beattie. (1983). "Self-Efficacy Beliefs and Sales Performance."

Journal of Organizational Behavior Management, 5 (1), 41-51.Betz, N., and G. Hackett. (1981). "The Relationship of Career-Related Self-Efficacy

Expectations to Perceived Career Options in College Women and Men." Journal of Counseling Psychology, 28 (4), 399-410.

Betz, N., and G. Hackett. (1983). "The Relationship of Mathematics Self-Efficacy Expectations to the Selection of Science-Based College Majors." Journal of Vocational Behavior, 23 (3), 329-345.

Brockner, J. (1988). Self-Esteem at Work. Lexington, MA: Lexington Books.Cooper, C.L. and G. Sadri. (1991). "The Impact of Stress Counselling at Work."

Journal of Social Behavior and Personality, 6 (7), 411-423.Cox, T. (1993). Cultural Diversity in Organizations. San Francisco: Berrett-Koehler

Publishers, Inc. Dulek, R.E., J.S. Fielden, and J.S. Hill. (1991). "International Communication: An

Executive Primer." Business Horizons, 15 (1), 20-26.Fisher, C.D. and E.A. Locke. (1992). "The New Look in Job Satisfaction Research

Theory." In C.J.Cranny, P.Cain Smith and E.F.Stone (eds.) Job Satisfaction: How People Feel About Their Jobs and How it Affects Their Performance . New York: Lexington Books.

Frayne, C.A. and G.P. Latham. (1987). "Application of Social Learning Theory to Employee Self-Management of Attendance." Journal of Applied Psychology, 72 (3), 387-392.

Fukuyama, M.A., and T.K. Greenfield. (1983). "Dimensions of Assertiveness in an Asian-American Student Population." Journal of Counseling Psychology, 30 (4), 429-432.

Fusilier, M.R., D.C. Ganster, and B.T. Mayes. (1987). "Effects of Social Support, Role Stress, and Locus of Control on Health." Journal of Management, 13 (1), 87-98.

Gemmill, G.R. and W.J. Heisler. (1972). "Fatalism as a Factor in Managerial Job Satisfaction, Job Strain, and Mobility." Personnel Psychology, 25 (2), 241-250.

Griffeth, R.W. and P.J. Hom. (1988). "A Comparison of Different Conceptualizations of Perceived Alternatives in Turnover Research." Journal of Organizational Behavior, 9 (1), 103-111.

Gudykunst, W.B. (1988). "Uncertainty and Anxiety," in Y.Y. Kim & W.B. Gudykunst (eds.) Theories in Intercultural Communication. Newbury Park, CA: Sage.

Gudykunst, W.B. (1991). Bridging Differences. Newbury Park, CA: Sage. Hackett, R.D. and R.M. Guion. (1985). "A Reevaluation of the Absenteeism-Job

Satisfaction Relationship." Organizational Behavior and Human Decision Processes, 35 (3), 340-381.

Hackett, G., R.W. Lent, and J.H. Greenhaus. (1991). "Advances in Vocational Theory and Research: A 20-Year Retrospective." Journal of Vocational Behavior, 38 (1), 3-38.

Hall, E.T. (1973). The Silent Language. Garden City, N.Y.: Anchor/Doubleday. Hall, E.T. (1976). Beyond Culture. Garden City, N.Y.: Anchor/Doubleday.Hall, E.T. (1983). The Dance of Life: The Other Dimension of Time . Garden City,

NY: Anchor/Doubleday.Hammer, T.H. and Y. Vardi. (1981). "Locus of Control and Career Self-Management

Among Nonsupervisory Employees in an Industrial Setting." Journal of Vocational Behavior, 18 (1), 13-29.

Heneman III, H.G. and M.H. Sandver. (1983). "Predicting the Outcome of Union Certification Elections: A Review of the Literature." Industrial and Labor Relations Review, 36 (4), 537-559.

Hofstede, G. (1983). "Dimensions of National Cultures in Fifty Countries and Three Regions." In J. Deregowski, S. Dziurawiec and R. Annis (eds.) Explications in Cross-Cultural Psychology. Lisse, The Netherlands: Swets and Zeitlinger.

Hofstede, G. (1984). Culture's Consequences: International Differences in Work-Related Values. Beverly Hills, CA: Sage Publications.

Hofstede, G. H. (1991). Cultures and Organizations: Software of the Mind. New York: MacGraw Hill.

House, R.J. (1971). "A Path-Goal Theory of Leader Effectiveness." Administrative Science Quarterly, 16 (1), 321-338.

House, R.J. (1987). "Retrospective Comment," in L.E. Boone and D.D. Bowen (eds.) The Great Writings in Management and Organizational Behavior, New York: Random House, 354-364

Iaffaldano, M.T. and P.M. Muchinsky. (1985)." Job Satisfaction and Job Performance: A Meta- Analysis". Psychological Bulletin, 97, (2), 251-273.

Indik, J. (1986). "Path-Goal Theory of Leadership: A Meta-Analysis," Paper presented at the National Academy of Management Conference, Chicago, Il.

Ivancevich, J.M. and M.T. Matteson. (1980). Stress and Work: A Managerial Perspective. Glenview, IL: Scott, Foresman.

Jackson, S.E. (1983). "Participation in Decision Making as a Strategy for Reducing Job-Related Strain." Journal of Applied Psychology, 68 (1), 3-19.

Jenkins, C.D. (1971). "Psychological and Social Precursors of Coronary Disease." The New England Journal of Medicine, 284 (3), 307-316.

Johnston, W.B. and A.H. Packer. (1987). Workforce 2000. Indianapolis: Hudson Institute.Judge, T.A. and C.L. Hulin. (1993). "Job Satisfaction as a Reflection of Disposition: A

Multiple Source Causal Analysis." Organizational Behavior and Human Decision Processes, 56 (3), 388-421.

Keller, R.T. (1989). "A Test of the Path-Goal Theory of Leadership with Need for Clarity as a Moderator in Research and Development Organizations." Journal of Applied Psychology, 74 (1), 208-212.

Kimmons, G. and J.H. Greenhaus. (1976). "Relationship Between Locus of Control and Reactions of Employees to Work Characteristics." Psychological Reports, 39

(1), 815-820.Kiplinger, A.A. and K.A. Kiplinger. (1989). America in the Global 90s. Washington

D.C.: Kiplinger Books.Klandermans, B. (1989). "Union Commitment: Replications and Tests in the Dutch

Context." Journal of Applied Psychology, 74 (6), 869-875.Kluckhohn, F. and Strodtbeck, F.L. (1961). Variations in Value Orientations.

Evanston, IL: Row Peterson.Lance, C.E. (1988). "Job Performance as a Moderator of the Satisfaction-Turnover

Intention Relation: An Empirical Contrast of Two Perspectives." Journal of Organizational Behavior, 9 (2), 271-280.

Latham, G.P. and G.A. Yukl, G.A. (1975). "A Review of Research on the Application of Goal Setting in Organizations." Academy of Management Journal, 18 (4), 824-845.

Lee, T.W. and R.T. Mowday. (1987). "Voluntarily Leaving an Organization: An Empirical Investigation of Steers and Mowday's Model of Turnover." Academy of Management Journal, 30 (4), 721-743.

Lent, R.W. and G. Hackett. (1987). "Career Self-Efficacy: Empirical Status and Future Directions." Journal of Vocational Behavior, 30 (3), 347-382.

Lester, D. and J.L. Genz. (1978). "Internal-External Locus of Control, Experience as a Police Officer, and Job Satisfaction in Municipal Police Officers." Journal of Police Science and Administration, 6 (3), 479-481.

Loden, M. and J.B. Rosener. (1991). Workforce America: Managing Employee Diversity as a Vital Resource. Homewood, Il: Richard D. Irwin.

Luthans, F. and R. Kreitner. (1985). Organizational Behavior Modification and Beyond. Glenview Il: Scott Foresman.

Manz, C.C. and H.P. Sims Jr. (1980). "Self-Management as a Substitute for Leadership: A Social Learning Theory Perspective." Academy of Management Review, 5 (2), 361-367.

Marino, K.E. and S.E. White. (1985). "Departmental Structure, Locus of Control and Job Stress: the Effect of a Moderator." Journal of Applied Psychology, 70 (4), 782-784.

Mathieu, J.E. and D.M. Zajac. (1990). "A Review and Meta-Analysis of the Antecedents, Correlates, and Consequences of Organizational Commitment." Psychological Bulletin, 108 (2), 171-194.

Mellor, S. (1990). "The Relationship Between Membership Decline and Union Commitment: A Field Study of Local Unions in Crisis." Journal of Applied Psychology, 75 (3), 258-267.

Mobley, W.H. (1977). "Intermediate Linkages in the Relationship Between Job Satisfaction and Employee Turnover." Journal of Applied Psychology, 62 (2), 237-240.

Mobley, W.H., S.O. Horner, and A.T. Hollingworth. (1978). "An Evaluation of Precursors of Hospital Employee Turnover." Journal of Applied Psychology, 63 (4), 408-414.

Multon, K.D., S.D. Brown, and R.W. Lent. (1991). "Relation of Self-Efficacy Beliefs to Academic Outcomes: A Meta-Analytic Investigation." Journal of Counseling

Psychology, 88 (1), 30-38.Munoz, M. (1973). "Job Satisfaction in Policemen and its Relations to Locus of

Control, Ego Strength, and Performance." Doctoral Dissertation. City University of New York. Dissertation Abstracts International, 34, 2946A-2947A.

Norusis, M.J. (1990). SPSS/PC+ Advanced Statistics 4.0. Chicago, Illinois: SPSS, Inc..Nystrom, P.C. (1983). "Managers' Salaries and Their Beliefs About Reinforcement

Control." Journal of Social Psychology, 120 (2), 291-292.Organ, D.W. (1990). The Motivational Basis of Organizational Citizenship Behavior. In

B.M. Staw and L.L. Cummings (eds.) Research in Organizational Behavior. Greenwich, CT: JAI Press.

Organ, D.W. and C.N. Greene. (1974). "Role Ambiguity, Locus of Control and Work Satisfaction." Journal of Applied Psychology, 59 (1), 101-102.

Pierce, J.L. and J.W. Newstrom. (1983). "The Design of Flexible Work Scedules and Employee Responses: Relationships and Processes." Journal of Occupational Behavior, 4 (1), 247-262.

Podsakoff, P.M., S.B. Mackenzie, and C. Hui. (1993). Organizational Citizenship Behaviors and Managerial Evaluations of Employee Performance: A Review and Suggestions for Future Research. In G. Ferris (ed.) Research in Personnel and Human Resources Management. Greenwich, CT: JAI Press.

Quinn, R.P. and L.J. Shepard. (1974). The 1972-1973 Quality of Employment Survey. Ann Arbor, Michigan: The Survey Research Center of the Institute for Social Research, 50-69.

Randle, W. (1990). "Some Firms Facing Facts on Workforce Diversity Dictates Major Changes." Chicago Tribune (January 2), 1.

Robertson, I.T. and G. Sadri. (1993). "Managerial Self-Efficacy and Managerial Performance." British Journal of Management, 4 (1), 37-45.

Rotter, J.B. (1966). "Generalized Expectancies for Internal Versus External Control of Reinforcement." Psychological Monographs: General and Applied, 80 (1), 1-

28.Sadri, G. and G.M. Marcoulides. (1994). "The Dynamics of Occupational Stress:

Proposing and Testing a Model." Research and Practice in Human Resource Management, 2, (1), 1-19.

Sadri, G., and I.T. Robertson. (1993). "Self-Efficacy and Work-Related Behavior: A Review and Meta-Analysis." Applied Psychology: An International Review, 42 (2), 139-152.

Scott, K.D. and G.S. Taylor. (1985). "An Examination of Conflicting Findings on the Relationship Between Job Satisfaction and Absenteeism: A Meta-Analysis."

Academy of Management Journal, 28 (3), 599-612.Shellenbarger, S. (September 3, 1993). "Workforce Study Finds Loyalty is Weak,

Division of Race and Gender are Deep." The Wall Street Journal: B1, B8.Snyder, M. (1987). Public Appearances/ Private Realities: The Psychology of Self-

Monitoring. New York: W.H. Freeman.

Spector, P.E. (1982). "Behavior in Organizations as a Function of Employees' Locus of Control." Psychological Bulletin, 91 (4), 482-497.

Spector, P. (1987). "Interactive Effects of Perceived Control and Job Stressors on Affective Reactions and Health Outcomes for Clerical Workers." Work and Stress, 1 (1), 155-162.

Storms, P.L. and P.E. Spector. (1987). "Relationships of Organizational Frustration with Reported Behavioral Reactions: The Moderating Effect of Locus of Control."

Journal of Occupational Psychology, 60 (2), 227-234.Taylor, K.M. (1981). "Applications of Self-Efficacy Theory to Career Decision Making

and Vocational Indecision in College Women and Men." Paper presented at the annual meeting of the American Psychological Association, Los Angeles, CA.

Taylor, K.M., and N.E. Betz. (1983). "Applications of Self-Efficacy Theory to the Understanding and Treatment of Career Indecision." Journal of Vocational

Behavior, 22 (1), 63-81.Taylor, M.S., E.A. Locke, C. Lee, and M.E. Gist. (1984). "Type A Behavior and

Faculty Research Productivity: What are the Mechanisms?" Organizational Behavior and Human Performance, 34 (3), 402-418.

Wall, T.D. and C.W. Clegg. (1981). "A Longitudinal Study of Group Work Redesign." Journal of Occupational Behavior, 2 (1), 31-49.

Weitzel, M.H. and P.R. Waller. (1990). "Predictive Factors for Health-Promotive Behaviors in White, Hispanic, and Black Blue-Collar Workers." Family and Community Health, 13 (1), 23-34.

Wofford, J.C. and Liska, L.Z. (1993). "Path-Goal Theories of Leadership: A Meta-Analysis." Journal of Management, 19 (4), 857-876.

Yeh, R. and Lawrence, J.J. (1995). "Individualism and Confucian Dynamism: A Note on Hofstede's Cultural Root to Economic Growth." Journal of International Business Studies, 26 (3), 655-669.

Zane, N.W.S., S. Sue, L. Hu, and J.H. Kwon. (1991). "Asian-American Assertion: A Social Learning Analysis of Cultural Differences." Journal of Counseling Psychology, 38 (1), 63-70.

INTEGER PROGRAMMING AND HEURISTIC

APPROACHES TO THE MINIMUM DIVERSITY

PROBLEM

Fred Glover *

Ching-Chung Kuo **

Krishna S. Dhir ***

The minimum diversity problem is concerned with selecting a set of elements from some larger collection such that those chosen exhibit the smallest variety of characteristics. In this paper, we propose a framework within which the concept of diversity can be defined and the diversity of a set of elements can be measured. We then formulate a number of integer programs to model the minimum diversity problem and its variation. We also show the theoretical complexity of the problem and provide specialized methods for efficiently obtaining approximate solutions, which are submitted to empirical analysis.

onsider a set of elements (e.g., items to be produced in a job shop) and some of their attributes (e.g., size, shape, and type of processing). For every element in the set, each of its attributes can be in one of several possible states (e.g., large, medium, or small; rectangular or triangular; grinding, cutting, machining, or painting). The objective of the minimum diversity problem is to select a predetermined number of elements from

* Fred Glover is the US West Chaired Professor of Systems Science at the University of Colorado, Boulder, Colorado.

** Ching-Chung Kuo is an Associate Professor of Operations Management at Pennsylvania State University at Harrisburg, Middletown , Pennsylvania.

***Krishna S. Dhir is the S.B.A. Professor of Business Administration at Pennsylvania State University at Harrisburg, Middletown, Pennsylvania.

Manuscript received, September, 1995, revised, March, 1996.

C

the set that are most similar in that they exhibit the lowest degree of difference in attribute states.

The concept of minimizing diversity is useful in many real-life applications. In cellular layout (Bedworth, Henderson, & Wolfe, 1991), for instance, component parts to be produced are grouped into families with similar processing requirements by using the technique of group technology (Burbidge, 1975). A set of machines are then dedicated to each of the part families to form a manufacturing cell to realize such benefits as reduced setup time, decreased material handling cost, faster throughput time, and lower work-in-process inventory level. In capital budgeting, portfolio selection from a set of interrelated investment proposals involves minimizing the overall risk, which is measured by the covariance of the security returns, at a given level of expected return (Laughhunn, 1970). In industry, workers are frequently placed in teams on the basis of compatibility in terms of similarity in sociometric choice. This has resulted in increased job satisfaction and lower turnover rate (Van Zelst, 1952). Finally, as another example, consider the scenario in which a college student majoring in comparative history is interested in identifying, say, three cultures in the world that are most similar to each other. It is obvious that this may be considered as a minimum diversity problem.

These examples, which provide only a sampling, clearly show the prevalence of the need to account for diversity minimization in practical settings. Nevertheless, the minimum diversity problem is largely ignored in management science and related literature. The main focus of this paper is on building a broad foundation which allows the wide variety of diversity minimization issues to be viewed within a common framework. We will present several integer programming formulations to model the minimum diversity problem, including a special "minimax" variation. Also included is an examination of the connections of the problem to the class of NP-complete problems. Finally, we propose heuristic algorithms for minimizing diversity and conduct computational tests that reveal the value of these special methods.

LITERATURE REVIEW

In recent years, diversity analysis has been utilized in empirical studies in major functional areas of business (Jones, 1987; Nayak & Gastworth, 1989; Varadarajan & Ramanujam, 1987). Typically, an entropy-based index is calculated and used to compare the degrees of diversification of several systems with respect to some variable of interest (e.g., income distribution), or to examine the change in the degree of diversification of an economic system over time. In most cases, the treatment is descriptive in nature; no mathematical model has been described that includes diversity minimization as a general goal.

In a sense, statistical clustering (Johnson & Wichern, 1988) offers a limited approach to minimizing diversity in the presence of representivity requirements since a

form of diversity is created by selecting elements from different clusters that have been identified. However, as pointed out by Kuo, Glover, & Dhir (1993b), this method suffers from several drawbacks. Dhir, Glover, & Kuo (1993) show how the minimum diversity concept can be applied to the selection of dependent engineering projects, where the risk of committing to two proposals is measured by the covariance between their returns. These, along with the study by Glover, Kuo, and Dhir (1994), account for the sparse research on diversity minimization in the existing literature.

In contrast, there has been a growing interest in the maximum diversity problem over the past few years. A succinct review of the latest developments in this area can be found in Kuo, Glover, & Dhir (1993a). More recently, Weitzman (1992) proposes a unified diversity theory using a function which satisfies a basic dynamic programming equation. Solow, Polasky, and Broadus (1993) formulate optimal policies for preserving biological diversity by taking economic factors into account. Glover, Kuo, and Dhir (1995) present an interesting application of the maximum diversity model for the conversation of crane species based on genetic distinctiveness.

In summary, while the subject of maximizing diversity has been researched from different perspectives, there have been few studies on its minimization counterpart. This observation provides ample motivation for the study of the minimum diversity problem.

MATHEMATICAL MODELING

We begin with a rudimentary formulation of the minimum diversity problem that will subsequently be refined and elaborated. Consider a set of elements s1, s2, ..., sn and some of their common attributes. Each of the attributes can be in one of several possible states. Our goal is to select m of the n elements so that the diversity of the subset is minimized.

In order to characterize the diversity of the elements in the subset to be chosen, we stipulate the existence of a value d ij for each pair of elements si and sj representing a distance measure over them. This measure can be the Euclidean distance, the Manhattan distance, or something in the nature of an affinity relationship (Francis, McGinnis, & White, 1992). For the purpose of our initial discussion, we will assume that these values are nonnegative though this assumption will be relaxed later.

According to these conventions, we introduce zero-one variables x i, where xi = 1 indicates that si is selected and x i = 0 if it is not, 1 I n. Our starting formulation of the minimum diversity problem is a quadratic zero-one integer program shown as follows.

Model (M1) n-1 n

Minimize Z = dijxixj I=1 j=I+1

n

subject to xi = m I=1

xi = 0 or 1, 1 I n

EQUIVALENT LINEAR MODEL AND VARIATION

The nonlinearity of the objective function in (M1) is inconvenient for most of the existing integer programming solution methods. Although a number of quadratic algorithms have been devised (see, for example, Laughhunn, 1970; Lawler & Bell, 1966; Taha, 1972a; 1972b), they are useful only for small problems and have not found widespread applications in the real world. A more efficient approach is to derive a linear equivalent of the nonlinear model and solve it. Techniques for converting polynomial zero-one programs into linear zero-one formulations are available in literature. Particularly, it can be readily established that (M1) can be transformed into (M2) below (Glover & Woolsey, 1974).

Model (M2) n-1 n

Minimize Z = dijyij I=1 j=I+1

n


xi + xj - yij 1, 1 I < j n (1)-xi + yij 0, 1 I < j n (2)-xj + yij 0, 1 I < j n (3) yij 0, 1 I < j n (4) xi = 0 or 1, 1 I n (5)

A few observations about (M2) are in order. First, note that y ij 1, I = 1, 2, ..., n - 1; j = I + 1, I + 2, ..., n. Moreover, the y ij variables will automatically receive zero-one values whenever the xi variables are assigned such values. Constraints (1) through (4) also validly model yij as the product of x i and x j in the case where one of them is continuous, relaxing the integer requirements on these variables in (5). Finally, if all d ij

0 then (2) and (3) can be disregarded since they will be nonbinding at an optimal solution.

While it is possible to improve further on (M2) using the results of Glover (1975; 1984), the more efficient formulations will not be presented here since the present one

is sufficient for treating the minimum diversity problem. Instead, we give an additional model based on an objective of minimizing the maximum diversity among the elements to be selected. In the following "minimax" model, M represents an extremely large positive number.

Model (M3)

Minimize Z = w

n


(M + dij)yij - w M, 1 I < j n xi + xj - yij 1, 1 I < j n-xi + yij 0, 1 I < j n-xj + yij 0, 1 I < j n yij 0, 1 I < j n xi = 0 or 1, 1 I n

Although w is unrestricted in the above formulation, the constraint w 0 may be included to facilitate the solution of (M3) if all d ij are nonnegative since this is a minimization problem.

AN EXAMPLE

In this section we provide an illustrative example based on (M2), which is the simplest minimum diversity model. Several distance measures have been in use in facility layout and location and, depending on the area of application, one measure may be preferred to the others (Klastorian, 1985). The Euclidean distance is chosen here because it is one of the most commonly used measures (Francis, McGinnis, & White, 1992).

Consider a set of elements S = {si: I N} with the index set N = {1, 2, ..., n}. Each element si contains r attributes whose states are given by s ik, k R = {1, 2, ..., r}. Hence, si = (sik: k R) is the vector of attribute states of s i. By the Euclidean distance measure, the diversity of a subset containing two elements s i and sj is defined as

r

dij = [ (sik - sjk)2]1/2 (6) k=1

From this, we may define the diversity of a set of elements, Z, as the sum of the Euclidean distances between each distinct pair of elements, i.e.,

n-1 n n-1 n r

Z = dij = [ (sik - sjk)2]1/2

I=1 j=I+1 I=1 j=I+1 k=1

Now, suppose a manager wishes to select four engineers to work on a certain project as a group. He has rated each of the eight engineers in his department on a seven-point differential scale with respect to five personality attributes: considerate, open, pleasant, supportive, and trustworthy. The results are summarized in Table 1. The manager's goal is to identify the four engineers who are most compatible with each other in that the overall personality difference among them is as small as possible.

Table 1

Evaluation of Engineers with Respect to Relevant Attributes

No. Name Considerate

Open Pleasant Supportive Trustworthy

1 Alan 4 2 5 5 4

2 Beth 3 4 5 1 7

3 Carl 6 6 6 7 1

4 Dave 5 5 4 4 1

5 Eileen 7 6 5 5 6

6 Frank 7 3 3 3 1

7 Greg 3 5 1 3 3

8 Hal 4 5 5 3 4

In order to formulate a minimum diversity model for the problem, we need to compute the personality distance between each pair of the engineers using the formula described in (6). For example, the Euclidean distance between Beth (No. 2) and Frank (No. 6) is d26 = [(7 - 3)2 + (3 - 4)2 + (3 - 5)2 + (3 - 1)2 + (1 - 7)2]1/2 7.8102. The resulting symmetric personality distance matrix is shown in Table 2.

Table 2

Personality Distance Matrix

1 2 3 4 5 6 7 8

1 0.0000 5.4772 5.8310 4.5826 5.3852 5.1962 5.5678 3.6056

2 5.4772 0.0000 9.2736 7.1414 6.0828 7.8102 6.0828 3.8730

3 5.8310 9.2736 0.0000 3.8730 5.5678 5.9161 7.4162 5.5678

4 4.5826 7.1414 3.8730 0.0000 5.6569 3.1623 4.2426 3.4641

5 5.3852 6.0828 5.5678 5.6569 0.0000 6.4807 6.7823 4.2426

6 5.1962 7.8102 5.9161 3.1623 6.4807 0.0000 5.2915 5.0990

7 5.5678 6.0828 7.4162 4.2426 6.7823 5.2915 0.0000 4.2426

8 3.6056 3.8730 5.5678 3.4641 4.2426 5.0990 4.2426 0.0000

It follows that the problem of determining the four most compatible engineers can be modeled as the following mixed integer program based upon (M2).

Model (M4)

Minimize Z = 5.4772y12 + 5.8310y13 + ... + 4.2426y78

subject to x1 + x2 + ... + x8 = 4 x1 + x2 - y12 1 : x7 + x8 - y78 1-x1 + y12 0 :-x7 + y78 0-x2 + y12 0 :-x8 + y78 0y12, y13, ..., y78 0x1, x2, ..., x8 = 0 or 1

Using QSB+ (Chang & Sullivan, 1991) to solve (M4), we obtain an optimal solution given by (x1

*, x2*, x3

*, x4*, x5

*, x6*, x7

*, x8*) = (1, 0, 0, 1, 0, 1, 0, 1) with a total diversity

of Z* = 25.1098. In other words, to minimize the total personality difference of the team

of engineers, the manager should select Alan, Dave, Frank, and Hal. If it is desirable to include at least one female engineer in the group, the constraint x2 + x5 1 may be added to (M4). The solution to this new problem indicates that the following four will be chosen with a slightly higher overall diversity of Z* = 26.9370: Alan, Dave, Eileen, and Hal. If, alternatively, it is required that Carl be included in the group due to his previous experience with similar projects, the constraint x3 = 1 may be incorporated into (M4). The optimal solution shows that the team will then consist of Alan, Carl, Dave, and Hal and the total diversity becomes Z* = 26.9241. It is easy to see that other modeling considerations can be accommodated in a similar way.

HEURISTIC ALGORITHMS

Because the example problem in the previous section is quite small, there is a good chance of reaching a verified optimal solution. We will show, however, that the general problem of minimizing diversity is NP-hard (Garey and Johnson, 1979); namely, it is sufficiently difficult that no existing procedures can be expected to solve the problem within a reasonable amount of time as its size grows. A theoretical basis of such an argument is presented below (Kuo, Glover, & Dhir, 1993a).

Theorem 1. The maximization version of (M1) is NP-hard, both with and without restricting the dij coefficients to nonnegative values.

Since, given the same set of constraints, minimizing an objective function is equivalent to maximizing the negative of the objective function, the following result follows immediately:

Corollary 1. The problem of (M1) is NP-hard.

As the minimum diversity problem is intractable, attempts to solve it by general-purpose integer programming codes may encounter significant difficulties. Consequently, it is important to identify special-purpose heuristics capable of obtaining high quality solutions efficiently for handling these problems in practice. Toward that end, we will propose two constructive heuristics and two destructive heuristics for minimizing diversity.

As before, S = {si: I N} is a set of elements with the index set N = {1, 2, ..., n}. Each element si contains r attributes whose states are given by s ik, k R = {1, 2, ..., R}, and si = (sik: k R) is the vector of the attribute states of si. The problem is to select m of the n elements in S with the minimum possible diversity. Alternatively, we may express this as selecting an m-element subset M of N and the set of elements chosen may be represented by S_Selected = {si: I M}.

Each of the two constructive heuristics begins with S_Selected empty, and then progressively transfers elements from S to S_Selected. On the other hand, each of the

two destructive heuristics begins with S_Selected = S, and then progressively eliminates elements from S_Selected. In both cases, the procedure terminates when S_Selected contains exactly m elements. Relative to the index sets N and M, we note that transferring an element from S to S_Selected corresponds to transferring an index from N to M, while eliminating an element from S_Selected corresponds to deleting an index from M. The algorithms terminate when |M| = m. Moreover, the final solution assigns values to the zero-one variables in the vector x = (x1, x2, ..., xn) of the preceding models according to the following rule: x i = 1 if I M and x i = 0 otherwise.

Some additional notation is required to characterize the heuristics. Let X = {s i: I I} represent either S or S_Selected (depending on whether I = N or I = M) and define s_center(X) to be the center of gravity of X; that is, s_center(X) = ( si)/|X|. We denote the operation of deleting or adding a selected index I* to I (hence deleting or adding a selected element si

* to X) by I = I - I* or I = I + I*, respectively. We also allow the distance dij between elements si and sj to be represented by d(si, s j); hence the distance between s i and s_center = s_center(X) is expressed as d(si, s_center). Finally, we define the composite between si and X as D(si, X) = (d(si, sj), j I - {I}). Observe that this composite distance stipulates that I is removed from {I}, and it is the sum of the distances between si and all the remaining elements in X. Of course, if s i X then I = I - {I}.

The four heuristic algorithms for the minimum diversity problem follow:

Constructive Heuristic 1 (C1)

Step 0. Let S_Selected be empty and s_center = s_center(S).Step 1. If |M| = m, terminate. Otherwise, identify I* such that d(si

*, s_center) = Min(d(si, s_center): I N).

Step 2. Set N = N - I*, M = M + I*, and s_center = s_center(S_Selected). Then return to Step 1.

Destructive Heuristic 1 (D1)

Step 0. Let S_Selected = S and s_center = s_center(S_Selected).Step 1. If |M| = m, terminate. Otherwise, identify I* such that d(si

*, s_center) = Max(d(si, s_center): I M).

Step 2. Set M = M - I* and return to Step 1.

Constructive Heuristic 2 (C2)

Step 0. Let S_Selected and M be empty and X = S.Step 1. If |M| = m, terminate. Otherwise, identify I* such that D(si

*, X) = Min(D(si, X): I N).

Step 2. Set N = N - I* and M = M + I* and let X = S_Selected. Then return to Step 1.

Destructive Heuristic 2 (D2)

Step 0. Let S_Selected = S.Step 1. If |M| = m, terminate. Otherwise, identify I* such that D(si *, S_Selected)

= Max(D(si, S_Selected): I M).Step 2. Set M = M - I* and return to Step 1.

In order to help understand the heuristics proposed above, we apply C1 to solve the problem of selecting four engineers discussed previously. A summary of the results from implementing the algorithm is included in Table 3. The heuristic solution suggests that the project team be formed by Alan, Dave, Frank, and Hal with a total diversity of 25.1098, which happens to be the optimal choice. As a matter of fact, it can be demonstrated that each of D1, C2, and D2 will also provide the same solution to this particular minimum diversity problem.

Table 3

Summary of Results from Implementing C1

Iteration I* d(si*, s_center) M N s-center

0 Ø {1, 2, 3, 4, 5, 6, 7, 8} (4.88, 4.50, 4.25, 3.88, 3.38)

1 8 1.65 {8} {1, 2, 3, 4, 5, 6, 7} (4.00, 5.00, 5.00, 3.00, 4.00)

2 4 3.46 {4, 8} {1, 2, 3, 5, 6, 7} (4.50, 5.00, 4.50, 3.50, 2.50)

3 1 3.74 {1, 4, 8} {2, 3, 5, 6, 7} (4.33, 4.00, 4.67, 4.00, 3.00)

4 6 3.99 {1, 4, 6, 8} {2, 3, 5, 7}

A COMPUTATIONAL STUDY

The heuristics presented in the previous section are all very efficient to execute. In fact, it has been shown that the theoretical complexities of C1, D1, C2, and D2 are of orders O(mnr), O(n2r), O(nm2), and O(n3), respectively (Kuo, Glover, & Dhir, 1993a). In order to evaluate them more completely, each of the four procedures was empirically tested on a sample of 60 minimum diversity problems. We sought the optimal solutions as benchmarks by applying the branch-and-bound method to the test problems based on (M2).

The problems were randomly generated by computer, with the total number of elements considered (n), the percent selection rate (100m/n), and the number of relevant attributes possessed by each element (r) ranging, respectively, from 10 to 30, from 15 to

75, and from 2 to 21. For each n-m combination, five random examples were solved by each of the four algorithms. The deviation of the heuristic solution from the optimal solution was computed, and the average deviation (d) of the five individual deviations was then calculated. The results have been summarized in Table 4.

RESULTS AND DISCUSSION

We see from Table 4 that the first constructive heuristic (C1), though intuitive, turns out to be the least efficient of those tested. More specifically, C1 admits a largest individual deviation of 123.61%, but the majority (73%) of the heuristic solutions are within 95% of optimality. In other words, the occurrence of large individual deviation is very rare. The value d can be as large as 26.59% although the overall average deviation is 7.48%. Moreover, about 45% of the solutions obtained by C1 are optimal.

Using the first destructive heuristic (D1), the largest individual deviation is significantly smaller, at 23.00%. Thus the worst solution obtained over all runs is still 77% of optimality. The d value is never greater than 5.37%, and the overall average deviation from optimality is a mere 1.78%. In addition, D1 yields an optimal solution about 68% of the time.

The second constructive heuristic (C2) nearly rivals the performance of C1. The largest individual deviation is 123.61%, which is not as good as that with D1, and d reaches a maximum value of 26.59%. On the other hand, the overall average deviation is 7.13%, which indicates that in the large the average deviation from optimality is less than 8%. Furthermore, about 47% of the heuristic solutions obtained through C2 are in fact optimal.

Finally, the second destructive heuristic, D2, appears to perform somewhat similarly to D1. The largest individual deviation under D2 is 23.00%. The d value does not exceed 6.73%, and the overall average deviation is quite impressive for this procedure also, at 1.76%. Moreover, about 67% of the time the solution given by D2 is an optimal solution.

To gain fuller insight into the behaviors of the approximate methods, the average deviations are plotted against the percent selection rates for each of the four heuristics. An examination of Figures 1, 2, 3, and 4 reveals that, with a few exceptions, d generally decreases as 100m/n increases.

When similar graphs are plotted for sets of elements of different sizes, we see from Figures 5, 6, and 7 that C1 and C2 give the worst solutions in the majority of the example problems, whereas D1 and D2 appear to be comparable. These observations are consistent with the results presented in Table 4.

CONCLUSION

In this paper, we have argued that the diversity minimization problem has widespread applications as well as practical importance. Our concern has been to build a broad foundation that allows the rich variety of diversity minimization problems to be viewed within a common framework. As a basis for this, we have formulated a number of interrelated integer programming models. We have also presented an illustrative example showing how our formulations can be flexibly modified to handle additional constraints.

We have shown that the theoretical complexity of the minimum diversity problem places it in the realm of NP-hard problems. This observation motivates the development of special procedures in order to generate solutions of high quality efficiently, since standard integer programming packages cannot be expected to fulfill this need. We therefore have developed four heuristic approaches using simple constructive and destructive moves. Applied to a testbed of 60 problems of varying dimensions, two of the approximate algorithms gave solutions whose average deviation from optimality was approximately 7%, while the other two rendered solutions with an average deviation of less than 2%. From an empirical standpoint, this experimentation represents only a beginning, but it suggests that problems from the diversity minimization class can be handled effectively by appropriately designed heuristics.

In taking these first steps, we concede that a great deal remains to be done. The diversity minimization problem poses a challenge to researchers and practitioners to develop an extended repertoire of models and solution approaches to deal with important variants. The scope of this area and the need to address the problem directly call for further examination of its fundamental elements in future investigations.

Table 4

Summary of Empirical Results

d = (h - 0)/0 * 100%

n m

C1 D1 C2 D2

2 26.59 1.61 26.59 1.61

3 4.11 5.37 4.11 6.73

10 4 4.88 0.73 4.88 0.73

6 2.27 0.38 0.17 0.38

8 0.14 0.00 0.14 0.00

2 12.59 1.02 12.59 4.02

4 4.63 3.12 2.26 2.40

15 6 0.37 0.82 0.51 0.82

9 3.11 0.42 3.11 0.08

11 0.00 0.00 0.00 0.00

5 24.24 2.51 24.24 1.3330

8 6.89 2.42 6.97 3.02

Largest individual deviation 123.61 23.00 123.61 23.00

Overall average deviation 7.48 1.78 7.13 1.76

% of time solution is optimal 45 68 47 67

% of time solution is within 95% of

optimum 73 90 77 90

0 : optimal total diversityh: heuristic total diversityd: average percent deviation from optimality based on 5 random examples

Figure 1

Empirical Performance of C1

Figure 2

Empirical Performance of D1

Figure 3

Empirical Performance of C2

Figure 4

Empirical Performance of D2

Figure 5

Empirical Performance of Four Heuristics with n = 10

Figure 6


Figure 7


SUMMER 1997

REFERENCES

Bedworth, B.D., Henderson, M.R., & Wolfe, P.M. (1991). Computer-integrated Design and Manufacturing. New York, NY: McGraw-Hill.

Burbidge, J.L. (1975). The Introduction of Group Technology. New York, NY: John Wiley & Sons.

Chang, Y.-L. & Sullivan, R.S. (1991). QSB+: Quantitative Systems for Business Plus. Englewood Cliffs, NJ: Prentice-Hall.

Dhir, K.S., Glover, F., & Kuo, C.-C. (1993). "Optimizing Diversity for Engineering Management." In Proceedings of the Institute of Electrical and Electronic

Engineers (pp. 23-26). International Engineering Management Conference held in New Delhi, India (December 17-18).

Francis, R.L., McGinnis, L.F., & White, J.A. (1992). Facility Layout and Location: An Analytical Approach. Englewood Cliffs, NJ: Prentice-Hall.

Garey, M.R. & Johnson, D.S. (1979). Computers and Intractability: A Guide to the Theory of NP-completeness. New York, NY: Freeman.

Glover, F. (1975). "Improved Linear Integer Programming Formulations of Nonlinear Integer Problems." Management Science, 22(4), 455-460.

Glover, F. (1984). "An Improved MIP Formulation for Products of Discrete and Continuous Variables." Journal of Information & Optimization Sciences, 5(1), 69-71.

Glover, F., Kuo, C.-C., & Dhir, K.S. (1994). "An Integer Programming Approach to the Minimum Diversity Problem." Proceedings of the Western Decision Sciences

Institute (pp. 674-676). 23rd annual meeting held in Maui, HI (March 29 - April, 2).Glover, F., Kuo, C.-C., & Dhir, K.S. (1995). "A Discrete Optimization Model for

Preserving Biological Diversity." Applied Mathematical Modeling, 19(11), 696-701.

Glover, F. & Woolsey, R.E. (1974). "Converting the 0-1 Polynomial Programming Problem to a 0-1 Linear Program." Operations Research, 22(1), 180-182.

Johnson, R.A. & Wichern, D.W. (1988). Applied Multivariate Statistical Analysis. Englewood Cliffs, NJ: Prentice-Hall.

Jones, L.E. (1987). "Optimum Product Diversity and the Incentives for Entry in Natural Oligopolies." Quarterly Journal of Economics, 102(3), 595-613.

Klastorian, T.D. (1985). "The p-median Problem for Cluster Analysis: A Comparative Test Using the Mixture Model Approach." Management Science, 31(1), 84-95.

Kuo, C.-C., Dhir, K.S., & Glover, F. (1993a). "Maximizing Diversity: Concept, Formulations, and Applications." Proceedings of the Decision Sciences Institute (pp. 1092-1094). 24th annual meeting held in Washington, D.C. (November 21-23).

Kuo, C.-C., Glover, F., & Dhir, K.S. (1993b). "Analyzing and Modeling the Maximum Diversity Problem by Zero- one Programming." Decision Sciences, 24(6),

1171-1185.

115


Laughhunn, D.J. (1970). "Quadratic Binary Programming with Application to Capital-Budgeting Problems." Operations Research, 18(3), 454-461.

Lawler, E.L. & Bell, M.D. (1966). "A Method for Solving Discrete Optimization Problems." Operations Research, 14(6), 1098-1112.

Nayak, T.K. & Gastworth, J.L. (1989). "The Use of Diversity Analysis to Assess the Relative Influence of Factors Affecting the Income Distribution." Journal of Business & Economic Statistics, 7(4), 453-460.

Solow, A., Polasky, S., & Broadus, J. (1993). "On the Measurement of Biological Diversity." Journal of Environmental Economics and Management, 24, 60-68.

Taha, H.A. (1972a). "A Balasian-based Algorithm for Zero-one Polynomial Programming." Management Science, 18(6), B328-B343.

Taha, H.A. (1972b). "Further Improvements in the Polynomial Zero-one Algorithm." Management Science, 19, B226-B227.

Van Zelst, R.H. (1952). "Validation of a Sociometric Regrouping Procedure." Journal of Abnormal and Social Psychology, 47, 299-301.

Varadarajan, P. & Ramanujam, V. (1987). "Diversification and Performance: A Reexamination Using a New Two-dimensional Conceptualization of Diversity in Firms." Academy of Management Journal, 30(2), 380-393.

Weitzman, M.L. (1992). "On Diversity." Quarterly Journal of Economics, 107(2), 363-406.

116

CALL FOR PAPERS

JOURNAL OF BUSINESS AND MANAGEMENTThe JOURNAL OF BUSINESS AND MANAGEMENT is soliciting papers for its upcoming issue. Its purpose is to provide a forum for the dissemination of fresh ideas and research in all areas of business, management and public policy which would be of interest to business persons, public officials and academics. The JOURNAL invites both empirical and theoretical articles. Reports on research and opinion pieces as well as original manuscripts are welcome. Book reviewers are requested to submit articles on recent publications in related topics. Only articles not previously published or currently under review elsewhere can be considered. All submissions are blind refereed. It should not be assumed that readers are completely familiar with the concepts and terminology of the specific subject under study. Directness and clarity of presentation are desired.

Manuscripts must be typed and double-spaced. Three copies should be submitted and become the property of the JOURNAL. The length of the paper should not exceed 20 pages, excluding notes, references, tables, figures and appendices. Notes and references should appear at the end of the article. Each chart, figure or table should be camera ready, that is, an original on a separate page with instructions indicating the placement in the article. A separate cover sheet indicating the title of the manuscript, as well as the name, affiliation, position and address of the author should accompany each submission. To facilitate blind review, the body of the text should include the title, but not the author's name. Include also an abstract of not more than 100 words summarizing the article.

The views expressed in published articles are those of the authors and not necessarily those of the Editors or the Editorial Review Board. Responsibility for the correctness of quotations and citations rests with the author.

The JOURNAL OF BUSINESS AND MANAGEMENT is a bi-annual publication of the Western Decision Sciences Institute and the School of Management, California State University, Dominguez Hills.

Submit Papers to:Burhan F. Yavas, EditorJOURNAL OF BUSINESS AND MANAGEMENTSchool of ManagementCalifornia State University, Dominguez HillsCarson, CA 90747

Subscription to the


The JOURNAL OF BUSINESS AND MANAGEMENT is published bi-annually by the Western Decision Sciences Institute and the School of Management, California State University Dominguez Hills.

The annual subscription fee is $16.00. To receive the

JOURNAL OF BUSINESS AND MANAGEMENT, please complete the form below, and make your check or money order payable to the CSUDH Foundation.

Dean's Office

School of Management

California State University Dominguez Hills

Carson, CA 90747

Name

Organization

Address

Please make your check or money order payable in the amount of $16.00 payable to the CSUDH Foundation.

Documents

JOURNAL OF - jbm.johogo.comjbm.johogo.com/pdf/volume/0401/JBM-0401.doc · Web viewIt is essential to prevent submitting a boring, lackluster proposal to the government. Well-designed