Cornell University ILR School

DigitalCommons@ILR
CAHRS Working Paper Series Center for Advanced Human Resource Studies (CAHRS)

5-1-1995

Employee Compensation: Theory, Practice, and Evidence
Barry A. Gerhart
Cornell University

Harvey B. Minkoff
TRW Corporation

Ray N. Olsen
TRW Corporation

Follow this and additional works at: http://digitalcommons.ilr.cornell.edu/cahrswp Part of the Human Resources Management Commons
This Article is brought to you for free and open access by the Center for Advanced Human Resource Studies (CAHRS) at DigitalCommons@ILR. It has been accepted for inclusion in CAHRS Working Paper Series by an authorized administrator of DigitalCommons@ILR. For more information, please contact jdd10@cornell.edu.

Employee Compensation: Theory, Practice, and Evidence
Abstract

[Excerpt] As organizations continue to face mounting competitive pressures, they seek to do more with less and do it with better quality. As goals for sales volume, profits, innovation, and quality are raised, employment growth is often tightly controlled and in many cases, substantial cuts in employment have been made. To accomplish more with fewer employees calls for effective management of human resources. Typically, the employee compensation system, the focus of this chapter, plays a major role in efforts to manage human resources better.
Keywords

employee, compensation, organization, profit, human, resource, manage, pay, market
Disciplines

Human Resources Management
Comments

Suggested Citation Gerhart, B., Minkoff, H. B. & Olsen, R. N. (1995). Employee compensation: Theory, practice, and evidence (CAHRS Working Paper #95-04). Ithaca, NY: Cornell University, School of Industrial and Labor Relations, Center for Advanced Human Resource Studies. http://digitalcommons.ilr.cornell.edu/cahrswp/194

This article is available at DigitalCommons@ILR: http://digitalcommons.ilr.cornell.edu/cahrswp/194

CAHRS / Cornell University 187 Ives Hall Ithaca, NY 14853-3901 USA Tel. 607 255-9358 www.ilr.cornell.edu/CAHRS/

WORKING PAPER SERIES
Employee Compensation: Theory, Practice, and Evidence
Barry Gerhart Harvey Minkoff Ray Olsen
Working Paper 9 5 – 0 4

Advancing the World of Work

Employee Compensation

WP 95-04

Employee Compensation: Theory, Practice, and Evidence

Barry Gerhart Center for Advanced Human Resource Studies Cornell University

Harvey B. Minkoff and Ray N. Olsen TRW Corporation

May 1994

Working Paper #95-04

www.ilr.cornell.edu/cahrs

An updated version of this working paper has been published in Handbook of Human Resource Management, Ferris, Rosen and Barnum (Ed.), Chapter (27), 1995.

This paper has not undergone formal review or approval of the faculty of the ILR School. It is intended to make results of research, conferences, and projects available to others interested in human resource management in preliminary form to encourage discussion and suggestions.

Page 1

Employee Compensation

WP 95-04

INTRODUCTION
As organizations continue to face mounting competitive pressures, they seek to do more with less and do it with better quality. As goals for sales volume, profits, innovation, and quality are raised, employment growth is often tightly controlled and in many cases, substantial cuts in employment have been made. To accomplish more with fewer employees calls for effective management of human resources. Typically, the employee compensation system, the focus of this chapter, plays a major role in efforts to manage human resources better. Employee compensation plays such a key role because it is at the heart of the employment relationship, being of critical importance to both employees and employers. Employees typically depend on wages, salaries, and so forth to provide a large share of their income and on benefits to provide income and health security. For employers, compensation decisions influence their cost of doing business and thus, their ability to sell at a competitive price in the product market. In addition, compensation decisions influence the employer's ability to compete for employees in the labor market (attract and retain), as well as their attitudes and behaviors while with the employer. Employee compensation practices differ across employment units (e.g., organizations, business units, and facilities) on several dimensions (Gerhart & Milkovich, 1990, 1992; Gerhart, Milkovich, & Murray, 1992). The focus of the employee compensation literature has been on defining these dimensions, understanding why organizations differ on them (determinants), and assessing whether such differences have consequences for employee attitudes and behaviors, and for organizational effectiveness. In the following discussion, we briefly describe the basic dimensions of compensation and summarize some of the key theories used to explain the consequences of different compensation decisions. A discussion of pay determinants can be found in Gerhart and Milkovich (1990, 1992). STRATEGIC PAY DIMENSIONS Pay practices vary significantly across employing units and to some degree, across jobs. We discuss the form, level, structure, mix, and administration of payment systems (Gerhart & Milkovich, 1992; Heneman & Schwab, 1979; Milkovich & Newman, 1993). First, pay can be in the form of cash or benefits (e.g., health care, retirement, paid vacation). On average, about 70 percent of payments to U.S. employees are in the form of cash, leaving 30 percent in the form of noncash and deferred cash benefits (Noe, Hollenbeck, Gerhart, & Wright, 1994). Health care has been the fastest growing benefit, and most employers describe the challenge of controlling this cost while providing quality coverage as one of their top human resource management challenges.

Page 2

Employee Compensation

WP 95-04

Second, both benefits and cash compensation can be described in terms of their level (how much). Most organizations use one or more market pay surveys to help determine what other organizations pay specific jobs in making their own pay level decisions. More broadly, total labor costs are a function of both compensation cost per employee and total employee headcount. Therefore, to assess competitiveness in the product market, organizations should not focus only on pay levels. They should compare total labor costs, and better yet, they should compare with other organizations the sort of return (or productivity) they receive in terms of profits, sales, and so forth for each dollar spent on labor costs. The now common announcements of major reductions in force attest to the importance of controlling labor costs. Such decisions are also sometimes driven by comparisons of revenue or profits per employee, or the ratio of sales or profits to labor costs. Labor costs and productivity are also key factors in decisions about where to locate production. Germany's high labor costs have led to what Business Week described as the "Exodus of German Industry." German companies are moving production to lower labor cost countries, such as those in Eastern Europe and the United States.1 BMW recently announced it would be building vehicles in South Carolina and Mercedes-Benz will produce vehicles in Alabama. Agreements such as the North American Free Trade Agreement.(NAFTA) and the General Agreement on Trade and Tariffs will only reinforce the globalization of production. However, contrary to what was heard in the debate over NAFTA, labor costs will not be the determining factor in most cases, except perhaps for labor-intensive production. Labor costs as a percentage of total costs is shrinking in many cases, and other factors such as access to markets and labor force quality will often be more important. The decision by BMW and Mercedes-Benz to build in the United States, not Mexico, is evidence of this. Third, the structure refers to the nature of pay differentials within an employing unit. How many steps or grades are in the structure? How big are the pay differentials between different levels in the structure? Large organizations often have over 20 such levels, although many organizations have recently reduced the number of steps ("delayered"). Are employees at the same hierarchical level in different parts of the organization (e.g., different product sectors or different occupational groups) paid the same? Yet another aspect of structure is the timing of payment over employees' careers. Some organizations may bring entry level people in at a relatively high rate of pay, but then provide relatively slow pay growth, while another

Labor costs in the United States have been lower- than those of Germany in recent years. However, this difference changes as currency exchange rates differ. The United States still has the highest purchasing power per capita of any country.
1

Page 3

Employee Compensation

WP 95-04

organization may bring employees in relatively low but offer greater opportunities for promotion and pay growth over time. Fourth, payment systems differ in their mix (how and when cash compensation is disbursed). Some organizations pay virtually all employees a base salary that is adjusted approximately once per year through a traditional merit increase program. Merit increases become part of base salary and are supposed to depend on merit (performance), although there is a widespread belief that most employees get about the same percentage increase, regardless of their performance. As described below, an increasing number of organizations are using so-called variable pay or pay at risk, which means that some portion of employees' pay is uncertain and depends on some combination of future business unit or organization performance (e.g., profits, stock performance, productivity), group performance, and individual performance. Specific pay programs that influence pay mix are merit pay, incentive pay, gainsharing, profit sharing, and stock plans (e.g., stock options). Fifth, pay is administered differently in different organizations. The design of pay policies differs, for example, in terms of who is involved in the process. The roles of human resource departments, line managers, and rank and file employees differ across situations. In some organizations, line managers may design plans, often with assistance from the human resources department. Alternatively, human resources takes the lead in other cases. Employees to be covered by a payment system are sometimes involved, and in some cases, may actually design plans for themselves. Communication is another aspect of administration. The most technically sophisticated payment plan can generate desired employee reactions or exactly the opposite. The actual effect depends on whether the rationale for the payment plan is understood and accepted and whether employees' perceptions of the facts upon which the rationale is built (e.g., the company's financial health, the pay of employees in other jobs or organizations) are the same as the perceptions of those charged with seeing that the payment plan has the intended effects. We focus in this chapter on cash compensation issues. Benefits warrants a chapter of its own and discussions are available elsewhere (Beam & McFadden, 1992; Gerhart & Milkovich, 1992; Noe, Hollenbeck, Gerhart, & Wright, 1994). Further, our discussion of cash compensation is mostly limited to pay mix issues, an area that has been of great interest to organizations as they move (or consider a move) to "new" programs such as stock plans for non-executives, gainsharing, and profit sharing. In the remainder of this chapter, we provide a survey of theories that have been used to study the effects of pay decisions, describe specific pay programs and their expected consequences, and review recent empirical evidence on that question.

Page 4

Employee Compensation

WP 95-04

CONSEQUENCES OF PAY DECISIONS: THEORIES To understand what types of pay systems are most likely to be effective and how their effectiveness differs according to contingency factors such as business strategy, national culture, competitive environment, and employee characteristics, we need to have a good conceptual framework, or theory. In truth, there is as of yet no grand theory of compensation that takes these contingency factors into account, although recent work by Gomez-Mejia and Balkin (1992) is promising. In examining consequences, we need to recognize that effectiveness is a multi-faceted concept that could include at a minimum, cost, productivity, innovation, quality, financial, and attitudinal dimensions. Further, the relative importance of these dimensions will vary across organizations and business units. At the individual level of analysis, theories have been used to show how pay plans can be used to energize, direct, and control employee behavior. We briefly describe three such theories used in research on pay. Reinforcement and Expectancy Theories Reinforcement theory states that a response followed by a reward is more likely to recur in the future (Thorndike's Law of Effect). The implication for compensation management is that high employee performance followed by a monetary reward will make future high performance more likely. By the same token, high performance not followed by a reward will make it less likely in the future. The theory emphasizes the importance of a person actually experiencing the reward. Like reinforcement theory, expectancy theory (Vroom, 1964) focuses on the link between rewards and behaviors (instrumentality perceptions), although it emphasizes expected (rather than experienced) rewards (i.e., incentives). Motivation is also a function of two other factors: expectancy, the perceived link between effort and performance, and valence, the expected value of outcomes (e.g., rewards). Compensation systems differ according to their impact on these motivational components. Generally speaking, pay systems differ most in their impact on instrumentality: the perceived link between behaviors and pay, also referred to in the pay literature as "line of sight." Valence of pay outcomes should remain the same under different pay systems. Expectancy perceptions often have more to do with job design and training than pay systems.

Page 5

Employee Compensation

WP 95-04

Equity Theory Equity theory suggests that employee perceptions of what they contribute to the organization, what they get in return, and how their return-contribution ratio compares to others inside and outside the organization,' determine how fair they perceive their employment relationship to be (Adams, 1963). Perceptions of inequity are expected to cause employees to take actions to restore equity. Unfortunately, some such actions (e.g., quitting or lack of cooperation) may not be helpful to the organization. Two recent empirical studies provide good examples of the types of counterproductive behaviors that can occur as a result of perceived inequity. In the first study, Greenberg (1990) examined how an organization2 communicated pay cuts to its employees and the effects on theft rates and perceived equity. Two organization units received 15% across-the-board pay cuts. A third unit received no pay cut and served as a control group. The reasons for the pay cuts were communicated in different ways to the two pay-cut groups. In the "adequate explanation" pay-cut group, management provided a significant degree of information to explain its reasons for the pay cut, and also expressed significant remorse. In contrast, the "inadequate explanation" group received much less information and no indication of remorse. The control group received no pay cut (and thus no explanation). The control group and the two pay-cut groups began with the same theft rates and equity perceptions. After the pay cut, the theft rate was 54% higher in the adequate explanation group than in the control group. However, in the "inadequate explanation" condition, the theft rate was 141% than in the control group. In this case, communication had a large, independent effect on employees' attitudes and behaviors. Cowherd and Levine (1992) used a sample 102 business units in 41 corporations to examine whether the size of the pay differential between lower-level employees and top management had any impact on product quality. Cowherd and Levine suggest that individuals often compare their pay to that of people higher in the organization structure. If lower-level employees feel inequitably treated, they may seek to reduce their effort to achieve equity. Quality, in their study, was defined as customer perceptions of the quality of goods and services. They hypothesized that extrarole, or citizenship behaviors, such as freely offering to help others, following the spirit rather than letter of rules, and correcting errors that would ordinarily escape notice, would be less likely when pay differentials between hourly and top managerial employees were large. Their results supported this hypothesis, suggesting that
2

Employees may use other comparisons standards also, such as their previous or expected future jobs or cost of

living.

Page 6

Employee Compensation

WP 95-04

organizations need to take care that they not forget the potential adverse motivational consequences of executive pay for the motivation of other employees. Agency Theory Agency theory, until recently best known in the economics, finance, and law literatures, focuses on the divergent interests and goals of the organization's stakeholders, and the ways that employee compensation can be used to align these interests and goals (Eisenhardt, 1989; Fama & Jensen, 1983). ownership and management (or control) are typically separate in the modern corporation, unlike the days when the owner and manager were often the same person. With most stockholders far removed from day-to-day operations, so-called agency costs (i.e., costs that arise from the interests of the principals/owners and their agents/managers not converging are created. What is best for the agent/manager, may not be best for the owner. Examples of agency costs include management spending money on perquisites (e.g., "superfluous" corporate jets) or "empire building" (acquisitions that do not add value to the company but may enhance the manager's prestige or pay) rather than seeking to maximize shareholder wealth (Lambert & Larcker, 1989). In addition, the fact that managers and shareholders may differ in their attitudes toward risk gives rise to agency costs. Shareholders can diversify their investments (and thus their risks) more easily than managers can diversify risk in their pay. As a consequence, managers may prefer relatively little risk in their pay (e.g., high emphasis on base salary,-low emphasis on uncertain bonuses or incentives). Indeed, research shows that managerial compensation in manager-controlled firths is more often designed in this manner (Tosi & Gomez-Mejia, 1989). Agency costs also stem from differences in decision-making horizons. Especially where managers expect to spend little time in the job or with the organization, they may be more inclined to maximize short-run performance (and pay), perhaps at the expense of long-term success. Agency theory is also of value in the analysis and design of non-managers' compensation. In this case, the divergence of interests may exist between managers (now in the role of principals) and their employees (who take on the role of agents). In designing either managerial or non-managerial compensation, the key question is, "How can such agency costs be minimized?" Agency theory says that the principal must choose a contracting scheme that helps align the interests of the agent with the principal's own interests (i.e., reduces agency costs). These contracts can be classified as either behavior oriented (e.g., merit pay) or outcome oriented (e.g., stock options, profit sharing, commissions). At first blush, outcome-oriented contracts seem to be the obvious solution. If profits are high, compensation goes up. If profits go down, compensation goes down. The interests of "the

Page 7

Employee Compensation

WP 95-04

firm" and employees are aligned. An important drawback, however, is that such contracts increase the amount of risk borne by the agent. Furthermore, because agents are averse to risk, they may require higher pay (a compensating wage differential) to make up for it. Behavior-based contracts, on the other hand, do not transfer risk to the agent, and thus do not require a compensating wage differential. However, the principal must be able to monitor with little cost what the agent has done. Otherwise, the principal must either invest in monitoring/information or structure the contract so that pay is linked at least partly to outcomes. Which type of contract should an organization use? It depends partly on the following factors (Eisenhardt, 1989): ♦ Risk aversion. Risk aversion among agents makes outcome-oriented contracts more costly. ♦ Outcome uncertainty. Profit is an example of an outcome. Linking pay to profits (outcome-based contract) is more costly to the extent that profits vary and so there is a risk of low profits. ♦ Job programmability. As jobs become less programmable (i.e., less routine and less structured), and more difficult to monitor, outcome-oriented contracts become more likely. The increasing complexity of organizations and technology makes monitoring more difficult, and may help explain the growing use of variable pay programs (discussed below), which are examples of outcome-based contracts. Consistent with this idea, outcome-oriented contracts (e.g., profit sharing and stock plans) are more prevalent in research and development organizations, where monitoring is especially difficult (Milkovich, Gerhart, & Hannon, 1991). Pay levels are also higher, consistent with the idea that employees must be compensated for sharing more risk. ♦ Measurable lob outcomes. When outcomes are more measurable, outcome-oriented contracts are more likely. ♦ Ability to pay. outcome-oriented contracts contribute to higher compensation costs because of the risk premium. ♦ Tradition. A tradition or custom of using (or not using) outcome-oriented contracts will make such contracts more (or less) likely. Influences on Labor Force Composition Traditionally, the theories described above have been used to understand how using pay to recognize individual contributions can influence the behaviors and attitudes of current employees, whereas pay level and benefits have been seen as a way to influence so-called membership behaviors: decisions about whether to join or remain with the organization.

Page 8

Employee Compensation

WP 95-04

However, there is increasing recognition that individual pay programs may also have an effect on the nature and composition of an organization's work force (Milkovich & Wigdor, 1991; Gerhart & Milkovich, 1992). For example, it is possible that an organization that links pay to performance may attract more high performers than an organization that does not link the two. There may be a similar effect with respect to job retention. Breaking things down further, perhaps organizations that link pay to individual performance are more likely to attract individualistic types of employees, while organizations relying more heavily on team rewards are more likely to attract more team-oriented employees. Although there is no concrete evidence of this yet, it has been found that different pay systems attract different people depending on their personality traits and values (Bretz, Ash, & Dreher, 1989; Judge & Bretz, 1992). The implication is that the design of compensation programs needs to be carefully coordinated with the business and human resource strategy. Strategy Moving from the individual level of analysis to the business unit and corporate level, there are theories of what corporate and pay strategies fit best together. Stage in the product life cycle (Ellig, 1981) and the degree and process of diversification (Kerr, 1985) have been raised as contingency factors in the design of pay strategies (Milkovich, 1988). Briefly, organizations (or probably more precisely, business units) may go through growth, maintenance, and decline stages, each of which calls for a different compensation strategy. For example, in the growth stage, it was recommended that there be substantial pay at risk to provide high upside earnings potential (e.g., using stock plans) to spur innovation, growth, and risk-taking, combined with low fixed costs (base salary and benefits) to preserve scarce capital for investment. In the maintenance and decline stages, there would be less emphasis on pay at risk (except perhaps for more short term focused plans), and more dollars allocated to base salary and benefits. The literature on diversification and pay strategy suggests that single product firths and unrelated product firms (e.g., conglomerates) have more pay at risk than related product firms, and pay is more decentralized and tied to business unit rather than corporate performance in the unrelated product firms. This flexibility makes sense where each business unit has independent goals, and there is little need for coordination and thus for consistency in pay practices. From an agency theory point of view, it may be more necessary to rely on outcome-oriented contracts in unrelated products firms because the market-specific expertise is concentrated in the business units, making it difficult for corporate headquarters to make evaluations using behavior-oriented contracts. Finally, an unrelated firm that is a result of mergers and acquisitions is more likely to have the flexibility and pay linked to unit performance

Page 9

Employee Compensation

WP 95-04

than an unrelated firm that is the result of internal growth because there is often more interdependence and interaction in the latter case (Kerr, 1985). Gomez-Mejia and Balkin (Gomez-Mejia & Balkin, 1992; Gomez-Mejia, 1992) have summarized much of the research on these questions, and provided some of the first tests of whether firms that choose pay strategies consistent with the above frameworks actually perform better. The answer seems to be "yes." They have provided propositions about which types of pay practices are likely to be most effective based on various strategy frameworks. For example, the Miles and Snow (1978) model classifies business units as defenders (stable markets, focus on efficiency), prospectors (focus on new markets and technologies), and analyzers, which have elements of both defenders and prospectors. According to Gomez-Mejia and Balkin, variable pay, for example, should be higher in the prospector business units than in the defender business units. Table 1 shows other proposed differences.

Table 1. Matching Organization Strategy and Pay Strategy Business Unit Strategy Defenders Prospectors Low Short-term Above market Below market Above market Centralized Job High Long-term Below market Above Market Below market Decentralized Skills

Pay Strategy Dimensions Risk Sharing (Variable Pay) Time Orientation Pay Level (short run) Pay Level (long run potential) Benefits Level Centralization of Pay Decisions Pay Unit of Analysis

Source: Adapted from Gomez-Mejia, L.R. & Balkin, D.B. (1992). Compensation, organizational strategy, and firm performance, Appendix 4b.

Page 10

Employee Compensation

WP 95-04

Summary Reinforcement, expectancy, and agency theories all focus on the fact that behavior-reward contingencies can shape behaviors. However, agency theory is of particular value in studying variable pay because of its emphasis on the risk-reward trade-off, an issue that needs close attention when considering variable pay plans, which can carry significant risk. Equity theory is also very relevant because it can be applied to just about any pay decision, because fairness is always a key concern. Moving away from individual-level theories, life cycle and diversification-based contingency theories suggest that pay strategies should fit with corporate strategies. The evolving empirical literature provides tentative support for many of the specific propositions. PAY PROGRAMS Table 2 summarizes the key features of some of the most widely used pay programs. Key dimensions include the payment method (whether increases roll into base salary or are paid as bonuses or equity), the frequency of payouts, the nature of the performance measure, and who is typically covered under the different plans.

Table 2. Comparison of Different Pay Programs
Individual Incentives Payment Method Payout Frequency Performance Measurement Bonus Merit Pay Merit Bonus Gainshairing Profit Sharing Ownership Skill Based Pay Changes in base pay When skill required Skill acquisition

Changes in base pay Annually

Bonus

Bonus

Bonus

Equity Changes When stock sold Stock value

Weekly

Annually

Monthly or quarterly Production or controllable costs Production or service unit

Semi-annually or annually Profit

Output, productivity, sales Direct labor

Performance rating

Performance rating

Coverage

All employees

All employees

Total organization

Total organization

All employees

Source: Adapted and extended from Lawler, E.E. III. (1989). Pay for performance: A strategic analysis. In L.R. Gomez-Mejia (Ed.), Compensation and benefits. Washington, D.C.: Bureau of National Affairs.

Page 11

Employee Compensation

WP 95-04

In compensating employees, an organization does not have to choose one program over another. Instead, a combination of programs is often the best solution. For example, one program may foster teamwork and cooperation but not enough individual initiative. Another may do the opposite. Used in conjunction, a balance may be attained. We now turn to a discussion of some recent trends in pay and an evaluation of .where such trends are likely to lead us. Recent Developments The Shift to Variable Pay According to a survey of over 2,000 U.S. companies by Hewitt Associates (Tully, 1993), the percentage of companies having a variable pay policy covering all salaried employees increased from 47% in 1988 to 68 in 1993. Moreover, whereas the standard merit increase (which rolls into base salary) was larger (5% versus 3.9%) in 1988 than the merit bonus (a lump sum payment that does become part of base salary), by 1993 the situation was reversed with the merit bonus being larger on average than the standard merit increase (5.9% versus 4.3%). Those in the human resource management field expect the movement toward variable pay to continue. In the Workplace 2000 study conducted by Dyer and Blancero (1993), 57 human resource executives, consultants, academics, and others were asked to describe how the workplace was likely to change by the year 2000. Dyer and Blancero provided study participants the characteristics of a hypothetical service organization in 1991 and asked how it would look in the year 2000. One expectation of participants was that pay would become more variable. As Table 3 indicates, variable pay as a percentage of total direct compensation was expected to increase significantly for each of the four occupational groups studied. Table 4 provides some examples of how variable pay programs operate.

Table 3. Variable Pay as a Percentage of Total Direct Compensation 1991 Scenario and Year 2000 Projection Occupational Group Executives Managers Professional/Technical Support 1991 20 % 10 % 10 % 10 % 2000 33 % 23 % 18 % 14 % Percentage Change 65 % 130 % 80 % 40 %

Source: Dyer, L. & Blancero, D. (1993). Workplace 2000: Adelphi study. Center for Advanced Human Resource Studies, Cornell University.

Page 12

Employee Compensation

WP 95-04

Table 4. Examples of Variable Pay Programs for Managers Company Plan Participants Base Pay Policy Bonus Policy

Nucor Steel

14 plant managers

$80,000 to $150,000 (25 % below market)

5 % of every dollar earned beyond 10 % return on equity goes into bonus pool. Last year, average plant manager bonus equaled base salary

General Mills

Managers (Marketing manager in this example)

$75,000 (versus $90,000 market midpoint)

$10,000 if profits growth and return on capital are at market average; up to $40,500 if profits growth and return on capital are in top 10 % of market

AT&T

80,000 middle managers& 30,000 scientists, researchers, technical employees

Between 1986 and 1989, pay raises less than one-half of competitors--move from pay leader to below midpoint

1. Individual or team bonus-pool depends in part on corporate net profitability (5 to 15 % of base depending on individual/team performance) 2. Business unit net profitability (about 2 % of base or less this year) 3. Corporate net profitability (7 to 11 % of base)

Source:

Tully, S. (1993, November 1). Your paycheck gets exciting. Fortune, p. 83+.

Page 13

Employee Compensation

WP 95-04

Group and Organization-Based Variable Pay Dyer and Blancero (1993) also found a strong belief that, in the future, variable pay would be based to a lesser degree on individual performance and to a much greater degree on firth, business unit, and work group performance (see Table 3). The examples in Table 4 are consistent with this expectation. It should be noted, however, that despite these significant changes, Dyer and Blancero found that individual performance is expected to remain as the single most important determinant of variable pay for all occupational groups. Why are organizations making greater use of variable pay, and why are they moving away from an individual focus to more of a group and organization focus? Variable pay is seen as a way of both controlling costs (especially in the case of organization-wide plans) and re-directing employee behavior. Better cost control is expected to be gained by replacing standard merit increases with merit bonuses that are linked to firm or business unit performance. Thus, when profits or stock returns are good, they can be shared with employees. However, when profits or stock returns are poor or nonexistent, the organization is not saddled to the same degree with high fixed labor costs. In theory, the use of variable pay plans to control labor costs is fine and it even works in practice under the right conditions; namely, if employees see a compelling business need to stay competitive in this manner. However, as in the widely discussed case of the DuPont Fibers division variable pay plan (Santora, 1991),3 employee opposition to downside variability in pay when profit targets are not met can lead to such plans being discontinued as soon as the labor cost control aspect is supposed to kick in, and employees forego bonuses and receive only their (below market) base salary. This result is consistent with agency theory's prediction that outcome-oriented contracts are less successful when there is high outcome uncertainty. Some organizations seek to avoid this "problem" by setting base pay at a higher level, and then sharing profits or stock with employees on top of their base salary during good years. These "gravy" plans do not control labor costs and, in fact, raise them. Yet, unless there is a compelling reason to believe that such pay plans significantly raise employee or for organization

Under the DuPont plan, base salary was about 4 percent lower than for similar employees in other divisions, unless 100 percent of the profit goal (a 4 percent increase over the previous year's profits) was reached. However, if the profit goal was exceeded, employees would earn more than similar employees in other divisions. For example, if the division reached 150 percent of the profit goal (i.e., 6 percent growth in profits), employees would receive 12 percent more than comparable employees in other divisions. In 1989, when the profit goal was exceeded, the plan seemed to work fine. However, in 1990, profits were down 26 percent from 1989, the profit goal was not met, and employees received no profit-sharing bonus. Instead, they earned 4 percent less than comparable employees in other divisions. Employees were not happy and DuPont eliminated the plan and returned to a system of fixed base salaries with no variable component.
3

Page 14

Employee Compensation

WP 95-04

productivity, organizations following this approach run the risk of investing extra money in the form of labor costs without realizing any return on the investment. Therefore, consistent with agency theory, employees may demand a compensating pay premium to assume risk. So, pay risk costs the organization more money, but gains in effectiveness are not certain. Organizations that use variable pay, with or without downside risk, often believe that such plans do generate significant returns. In agency theory terms, profit-sharing, stock plans, and gainsharing are examples of outcome-based contracts that seek to align the interests of employees and management with those of owners. As such, they are expected to re-direct behavior away from parochial individual goals, and more toward what it takes in terms of cooperation, commitment, and innovation to make the group, business unit, or organization a success. A change to variable pay may be a way to send a message to employees that things are going to change in important ways and therefore, may be helpful in supporting other major human resource changes. For example, variable pay may support a move to a team-based organization. As another example, variable pay may help eliminate the "entitlement" mentality or culture that can result from so-called merit increase plans that (in fact) fail to differentiate between employees with different performance levels, roll the increase into base salary so the cost remains in future years, and ignore the performance of the business. With a merit bonus, the pay has to be re-earned each year. Past individual performance does not matter, and is not reflected in base salary. Therefore, employees cannot rest on their past laurels. Moreover, the bonus pool may be linked to organization or business unit profitability. Again, the idea is to align employee interests with those of the organization. In this case, the goal is to encourage continuous improvement and a forward-looking perspective. Agency theory suggests that group and organization incentives can also contribute to greater overall levels of performance monitoring by, in effect, making each employee a principal who monitors other employees (Levine & Tyson, 1990). So, if your pay and my pay depend on what we do as a team, we will be more likely to monitor each other's performance and give feedback to one another when performance needs improvement: Similarly, according to equity theory, if a person feels that his or her inputs (e.g., work effort) are greater than another member of the work group, but they receive the same reward, one way to restore equity would be to encourage (or pressure) the other person to put forth more effort. Group size, however, is a key contingency variable in discussions of the behavioral impact of group and organization variable pay plans. According to expectancy theory, the larger the number of employees covered by a pay plan, the weaker the link they see between their

Page 15

Employee Compensation

WP 95-04

own performance and pay (Schwab, 1973), and thus the weaker is their motivation. Similarly, a theme from the shirking, social loafing, and free rider literatures is that individual effort decreases as the size of the group increases (Kidwell & Bennett, 1993). The implication, therefore, is that the ability of group and organization plans to change employee behavior may be very limited in cases where large numbers of employees are covered. On this dimension, gainsharing plans, which typically cover smaller groups of employees, probably have an advantage over organization-wide plans like profit sharing and stock-based plans. Another advantage is that the performance measures in gainsharing plans (e.g., labor costs, quality) are often more controllable, again fostering greater employee motivation to change behavior. The trade-off, however, is that gainsharing plans can pay off big even when the company is losing money. Another difficult situation arises when management would like to bring more work into the plant, but cannot afford to because the plan payouts would become too costly. In these cases, one might say that gainsharing plans (consistent with the general history of incentive plans) sometimes "fail" because they are too "successful." The payouts of any incentive plan must walk the fine line between being too low to motivate employees and being too high for management to afford. Even when standards work well initially, changes in production level and technology often result in the plan being unacceptable to one party or the other. In some cases, management may choose to "buy out" employees by paying a lump sum settlement in exchange for being able to redesign the plan with different standards, especially in unionized settings. An implication is that any sort of variable pay program should have a "sunset" provision that requires evaluation of the plan after a specific number of years, to avoid having the pay program becoming irrelevant because the organization changed, but it did not. A final reason we discuss for the growth in variable pay plans is that the increased use of total quality management (TQM) often entails a movement toward a team-based organization and empowerment of employees to go beyond their traditional roles to make decisions in a broader range of areas that are likely to have an impact on organization performance. Individual-oriented systems may not be adequate for encouraging employees to pursue broad organization goals, and to engage in the cooperative team and group-based decision-making necessary. A survey conducted for the American Compensation Association (ACA) asked organizations that implemented TQM programs how their pay practices changed (Davis, 1993). As Table 5 indicates, major changes included less reliance on supervisors as the only source of

Page 16

Employee Compensation

WP 95-04

performance appraisals, more reliance on team and organization results in setting pay, greater use of variable pay, and fewer, broader pay grades.

Table 5. Changes in Pay to Support Total Quality Before Performance Appraisal (n = 91) Only supervisors as source Peer/team appraisals Add quality criteria/goals Have team goals Plan Increase Policies (n = 38) Increases tied to individual performance appraisals Increases tied to team/organization results Increases tied to quality results Increases tied to skill/knowledge levels Incentive Program Policies (n = 56) Incentives based on individual results Incentives based on individual/ team results Incentives based on team/organization results Salary Structure Policies Other (e.g., more pay at risk) Fewer grades, broader range widths --52 % 38 % 26 % 23 % 20 % 31 % 37 % 52 % 88 % 8% --60 % 60 % 49 % 33 % 59 % 2% --12 % 25 % 68 % 41 % After

Note: n refers to the number of organizations (out of 196 total) that made changes in each pay area. Source: Davis, J.H. (1993, Autumn). ACA Journal. "Quality Management and Compensation."

From an equity theory perspective, placing the entire employee population on such plans may also create a greater sense of fairness among non-executive employees who typically have not been covered by such plans in the past, but saw that executives were. Of course, this effect may be limited to plans where variable pay is used to provide additional upside earnings potential, as opposed to cases where it replaces a portion of base salary. Banding, De-Layering, and Paying the Person Rather than the Job In the traditional pay system, the worth of jobs is assessed on the basis of job evaluation data in combination with market survey data. Job evaluation focuses on measuring and valuing

Page 17

Employee Compensation

WP 95-04

the specific characteristics and requirements of the job. Critics, however, suggest that job-based systems tend to spawn too much bureaucracy, too much emphasis by employees on doing only what is in their job description, and a lack of focus on market comparisons, which are critical for competitiveness. In addition, job levels become status indicators, which can get in the way. For example, an employee may be reluctant to accept a temporary assignment, that would be good from a developmental point of view, unless it has at least as high of a job level. There have been at least two types of responses. First, organizations like General Electric have cut levels of management and the corresponding pay grades. The goals are to improve communication and speed decisions by reducing the levels of management, and to provide wider pay grades (or bands) in order to allow more flexibility to recognize individual contributions, and to make lateral movements simpler by reducing the likelihood of a job being in a different (in this case, lower) grade (and looking like a demotion). The participants in the Dyer and Blancero (1993) study were also asked how the number of pay levels in the hypothetical service organization would change by the year 2000. Across the four occupational groups, the 36 pay levels in 1991 were expected to decrease to 23 pay levels by 2000, a decrease of about one-fourth. Whether the hoped for advantages of delayering and banding will offset the potential drawbacks (e.g., less opportunity for promotion) remains to be seen. Aside from allowing more flexibility in moving employees, banding, by virtue of a greater spread between the minimum and maximum in each pay grade, is also intended to provide more opportunity to recognize individual differences in performance. So, within-level pay growth for high performers will increase, while promotion opportunities and related pay growth will decrease. It remains to be seen whether this will, on balance, be a good trade for motivational purposes. Further, banding carries the risk of becoming very expensive. Topping out of employees near the maximum would be very expensive under a banding system. Some organizations have implemented sub-bands or zones within bands to avoid this problem. However, one might then reasonably ask what the difference is between an old system with 30 grades and a new system with 10 bands, each with 3 sub-bands. Another trend is for some organizations to move away from linking pay to job content through job evaluation, and instead pay workers for the skills they possess. Skill-based pay links pay to the breadth or depth of employee skill. The goal is to encourage learning, which in turn facilitates flexibility in work assignments and encourages learning as a way of life to help with future organization change.

Page 18

Employee Compensation

WP 95-04

Empirical Evidence Where are these recent developments in pay likely to lead us? We know that money can be a powerful motivator. Indeed, a literature review of four motivational programs (individual monetary incentives, goal-setting, job redesign, and participation in decision making) found that monetary incentives were associated with the largest average increase in physical productivity (Locke, Feren, McCaleb, Shaw, & Denny 1980). Therefore, changes in pay practices have the potential to significantly change attitudes, behaviors, and organization functioning. The challenge, however, is to realize the potential of money as a motivator without running afoul of the many roadblocks that arise in terms of measuring performance, setting standards that are perceived as fair, and choosing the right mix of individual, group, and organization objectives to reward. As one recent example of a variable pay program gone wrong, consider the problems Sears encountered in some of its automotive repair shops in New Jersey and California. In a State of California undercover investigation, 38 visits to 27 Sears repair shops resulted in 34 cases of unnecessary service or repair recommendations. Edward A. Brennan, the chairman of Sears, stated that "the incentive compensation program and sales goals created an environment where mistakes occurred" (Fisher, 1992). In essence, repair shop employees had been rewarded for driving revenue (i.e., selling repairs to customers). Sears subsequently changed its pay system to one that focused on "quality." Although specific examples are useful to demonstrate specific points, what does the broader research literature tell us regarding the typical outcomes of variable pay and other pay for performance programs? At the organization level, evidence suggests that greater emphasis on short-term bonuses and long-term incentives (relative to base pay) is associated with higher subsequent profitability, at least among top and middle level managers (Gerhart & Milkovich, 1990). Specifically, an organization with a bonus/base ratio of 10%, and 28% of its managers eligible for long term incentives had an average return on assets of 5.2%. In contrast, an organization with a 20% bonus to base ratio, and 48% of its managers eligible for long term incentives, had an average return on assets of 7.1%. The fact that organization-based bonuses and incentives work for high-level managers does not necessarily mean they will work for other types of employees, most of whom have less influence over organization performance and thus, weaker instrumentality perceptions. Still, even if the motivational impact (in terms of sheer effort) of organization-based incentives is

Page 19

Employee Compensation

WP 95-04

weaker for such groups, cost control and a re-focusing of behavior toward broader organizational goals may still be possible with such programs. The empirical evidence on profit sharing plans, in fact, generally paints a positive picture, with organizations using profit sharing having higher productivity (usually defined as value added per employee) on average than organizations that do not use profit sharing (e.g., Weitzman & Kruse, 1990; Kruse, 1993a, 1993b). Still, there has yet to be a convincing demonstration that profit sharing actually causes better organization performance (Gerhart & Milkovich, 1992). It may be that organizations with higher profit levels are more likely to adopt profit sharing plans. In addition, if a profit sharing plan does not work out, it is likely to be discontinued. So, the only profit sharing plans that are studied are those that have proven successful, and we do not hear about the plans that failed or needed to be replaced after they served their purpose. An organization that is deciding whether to adopt a profit sharing plan must know how often such plans work and how often they fail or get discontinued, not just how well the successful plans work. The evidence on stock plans is very limited, aside from the Gerhart and Milkovich (1990) study of top and middle level managers. The evidence that is available pertains mostly to employee stock ownership plans (ESOPs). Like profit sharing, the evidence is generally favorable (Jones & Takao, 1993; Conte & Svejnar, 1990), but the same cautions regarding causality apply. In any case, research suggests any beneficial effects of ESOPs may be stronger where employees have greater participation in making decisions, perhaps because it gives the employee a stronger feeling of ownership (Pierce, Rubinfeld, & Morgan, 1991). The costs of stock plans, especially options, may not always be obvious, but purchasing stock or issuing new stock (and the resulting dilution of the value of other shares) are costly moves. Indeed, U.S. Senator Carl Levin of Michigan introduced a bill in January 1993 that would require companies to show the granting of stock options as an expense. The Financial Accounting Standards Board (FASB) also has proposed changes to its rules in this area. What is the evidence on gainsharing programs? Again, it is generally positive. Although the types of cautions cited above regarding causality apply, the fact that from a theoretical standpoint, gainsharing programs offer employees a better line of sight (or instrumentality) between their performance and rewards (Lawler, 1989; Schwab, 1973) suggests that the motivational impact of such programs may be stronger than is the case with organization-wide programs, like profit sharing and stock plans. Gainsharing payouts are typically based on measures like value added, sales value of production, or hours saved, which are more controllable by employees than profits or stock performance.

Page 20

Employee Compensation

WP 95-04

A time series study by Schuster (1983) of six gainsharing plans found substantial (around 30 %-) increases in productivity in four cases following the implementation of gainsharing. A fourth plant had an initial increase in productivity, but increases in the costs of raw materials subsequently decreased the value added per worker, leading to no bonuses. A fifth plant, although not showing an increase in productivity, had gainsharing in place for twenty years, suggesting that the plan was working, but the productivity increase had already occurred before the study. Other studies have also found significant productivity improvements from gainsharing programs (e.g., Kaufman, 1992). In addition to having a payout measure that is controllable, gainsharing plans often have the advantage of covering a smaller number of employees, which is also beneficial for motivation, because there is less likelihood of employees "free riding" (i.e., working less hard because others will work hard). Indeed, one study estimated that a doubling of employees covered by a gainsharing plan from around 200 to 400 would reduce the expected productivity gain by almost one-half (Kaufman, 1992). The implication is that the number of employees covered can have a substantial impact on the plan's success. The fact that gainsharing (or any pay program) has a positive impact on productivity is no guarantee that it will continue to be used. A study of a gainsharing plan at an electrical utility estimated a net savings of between $857,000 and $2 million, but the plan was discontinued because employees in other divisions (all represented by the same union) felt unfairly treated because they were not covered (Petty, Singleton, & Connell, 1992). The organization was then faced with two difficult options. First, it could include all employees under the same plan, but that would likely increase the free rider problem and reduce the motivational impact. Second, it could have a separate plan for each division, but this could easily result in unequal payoffs to employees in different divisions, raising the same problems originally encountered with employees and the union. There would also need to be a means of preventing between-division competition. A profit sharing or stock plan combined with gainsharing plans would be one option. Other evidence also indicates that plans which appear to save money do not necessarily survive very long. Kaufman (1992) found that discontinued plans had improved productivity nearly as much as continuing plans. A study sponsored by the American Compensation Association (McAdams & Hawk, 1992) may shed some light on this question. They found that gainsharing plans, on average, were associated with net gains per employee of between $1,300 and $3,700 per year. Nevertheless, when asked to rate the effectiveness of gainsharing plans in improving effectiveness in areas like business performance, fostering teamwork, strengthening

Page 21

Employee Compensation

WP 95-04

the pay-for-performance link, and so forth, the average effectiveness ratings all fell between 2.63 and 3.25 on a 1 (no effectiveness) to 5 (high effectiveness) scale. In other words, most respondents were pretty lukewarm about gainsharing. Many organizations are moving to group and organization variable pay plans because they are frustrated with what they see as the failure of more traditional merit pay plans. Commonly cited problems include a lack of adequate differentiation between good and poor performers, employee and supervisor resistance, and the fact that merit increases sometimes seem to have become viewed as an entitlement by employees that is costly, and does not vary with business performance. Although there is truth to many of these assertions, one sometimes wonders if perhaps merit pay has been pronounced dead too soon. So-called studies of merit pay have often had significant limitations (see Gerhart & Milkovich, 1992). In addition, the notion that there is no merit pay is open to question. It is common to conclude that there is no individual merit pay because raises received by good and poor performers differ by only a few percentage points. Two employees, each with a base salary of $40,000, one receiving a 5% increase, the other 6%, would receive raises differing by $400 per year before taxes, or about $8 per week. Framed this way, the difference does indeed seem small and unlikely to motivate performance. On the other hand, the example ignores the fact that high performers are more likely to be promoted and thus, will have greater earnings growth. This is part of pay-for-performance, but it may not always be communicated as well as it could. Further, even limiting one's attention to the annual increase process, it can be shown that small differences in pay raises accumulate into significant differences over time. As Table 6 shows, the present value (or "real" payoff) to raises higher by 1 percentage point adds up to about $76,000 over 20 years. Factoring in promotion based on performance and pay-linked benefits (e.g., retirement) would further increase the payoff to higher performers. Factoring in taxes would decrease the payoff. Would communicating the payoff to performance in this manner change the way employees react to merit pay? Our conversations with managers yields a wide array of opinions on the matter, suggesting a good area for future research.

Page 22

Employee Compensation

WP 95-04

Table 6. Pay for Performance: Accumulation over Time Employee 1 Average Employee 2 1 point above above average 6%

Performance Rating Annual Pay Growth Year 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Total Difference *Using 5 % discount rate.

5%

Nominal $40,000 $42,000 $44,100 $46,305 $48,620 $51,051 $53,604 $56,284 $59,098 $62,053 $65,156 $68,414 $71,834 $75,426 $79,197 $83,157 $87,315 $91,681 $96,265 $101,078 $1,322,638

Real*

Nominal $40,000

Real'*

$40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $40,000 $760,000

$42,400 $44,944 $47,641 $50,499 $53,529 $56,741 $60,145 $63,754 $67,579 $71,634 $75,932 $80,488 $85,317 $90,436 $95,862 $101,614 $107,711 $114,174 $121,024 $1,471,424 $ 148,785

$40,381 $40,766 $41,154 $41,546 $41,941 $42,341 $42,744 $43,151 $43,562 $43,977 $44,396 $44,819 $45,245 $45,676 $46,111 $46,551 $46,994 $47,441 $47,893 $ 836,690 $ 76,690

Page 23

Employee Compensation

WP 95-04

The empirical evidence on the effects of banding is basically non-existent. Research on skill-based pay is just beginning to emerge. An ACA survey of organizations using skill-based pay illustrates some of the potential advantages and disadvantages. Most survey respondents felt that skill-based pay was successful in contributing to greater workforce flexibility and adaptability and in supporting work teams. However, relatively few saw any reduction in labor costs or layoffs. (Indeed, skill based pay is thought to permit a leaner headcount because of cross-training.) So, one must consider whether possible higher labor costs are justified by the advantages having to do with flexibility, adaptability, and the use of teams. Further, it must be recognized that if a plan is implemented, there are several factors that can contribute to its termination. The ACA survey found the following to be most important in terminating skill based pay plans: inadequate management commitment, unwillingness to endure short-term implementation problems, poor plan designs that increase labor costs without providing offsetting organizational benefits, conflicts between employees included and those excluded from the plan, inadequate training opportunities, and the failure of management to require meaningful skill certifications prior to pay increases. Only one skill based pay study to date (Murray & Gerhart, 1994) has used objective measures of productivity and quality, a control group, and a time series before and after implementation of the plan. In a comparison of two automobile parts plants, Murray and Gerhart found that a significant increase in productivity and product quality took place in the plant that implemented skill based pay. Globalization and Compensation The continued globalization of markets means that we will have to increasingly consider whether the effect of different pay strategies is likely to differ from country to country, or between cultures within a country. Hofstede's (1993) work on identifying culture differences on dimensions such as power distance (i.e., the degree of inequality considered normal), individualism, masculinity, uncertainty avoidance; and short- versus long-term time orientation has been used by Hodgetts and Luthans (1993) to begin studying this question. Certain hypotheses flow readily from the national differences depicted in Table 7.4 Variable pay (pay at risk) may face difficulties in countries that have a high need for uncertainty avoidance such as Japan, South Korea, and Taiwan. Individualistic programs such as merit pay could be a problem in cultures where collectivism is a stronger norm than individualism (e.g., the Pacific Rim countries). Still, average differences in culture are just that, averages, and should not
Long- versus short-term orientation is not shown. Japan, Hong Kong, and China have a long-term orientation, whereas the United States has a more short-term oriented culture.
4

Page 24

Employee Compensation

WP 95-04

necessarily be viewed as factors that must be taken as a given. Honda, in Japan, for example, just recently announced that it would be changing many of its managers over to a merit pay system. On the other hand, U.S. companies that have attempted to export pay practices overseas have often encountered difficulties. Lincoln Electric, famous for its history of success using variable pay, has thus far not been successful in implementing variable pay in its overseas acquisitions. Our own experience with gainsharing plans in Western Europe has not been successful (Chilton, 1993). The cultural differences described by Hofstede and related customs are often difficult to overcome. It is probably significantly easier to implement pay practices that are not typical of a country in a greenfield setting as opposed to an acquisition. Indeed, Japanese (e.g., Honda, Nissan) and German (e.g., BMW,Mercedes-Benz) automobile plants opened in the United States have often been in greenfield sites, where the company has maximum flexibility in screening and choosing employees who will fit well with their corporate culture, human resource management, and pay philosophies.

Table 7. National Culture Clusters Region or Country Power Distance Individualism Masculinity Uncertainty Avoidance

Pacific Rim

Hong Kong, Malaysia, Philippines, Singapore

High

Low

High

Low

Japan

High

Low

High

High

South Korea, Taiwan

High

Low

Low

High

United States, Great Britain

Low

High

High

Low

Sources:Hofstede, G. (1993, February). Cultural constraints in management theories. Academy of Management Executive, 7(1), 81-94; Hodgetts, R.M. & Luthans, F. (1993, March-April). U.S. multinationals' compensation strategies for local management: Cross-cultural implications. Compensation & Benefits Review, 25, 42-48. Review, 25, 42-48.

Page 25

Employee Compensation

WP 95-04

CONCLUSION Our goal in this chapter has been to describe the theory and practice of compensation, as well as provide an overview of recent empirical evidence on the consequences of different compensation practices. The theory section points to the many trade-offs in designing employee compensation policies. Examples of trade-offs include maximizing high individual effort versus teamwork and cooperation, controlling costs versus maximizing employee effort, and providing incentives for promotion versus producing feelings of inequity due to large pay differentials. Our message has been that the nature of such trade-offs should depend on the corporate and business strategies and that the trade-offs can be made less of a problem by combining pay programs in a way that helps balance competing objectives. As a final comment, we would like to emphasize that, although it is important to keep abreast of what other organizations are doing (benchmarking) in the area of employee compensation, it is crucial to remember that what works for one organization may not work at all for another. Therefore, surveys of "best practices" are useful to the extent that the surveys report a diversity of best practices and the reasons why different practices are best for different organizations. The ultimate choice of a best compensation strategy rests, of course, on its fit with other human resource activities and its fit with the business strategy.

Page 26

Employee Compensation

WP 95-04

REFERENCES Adams, J.S. (1963). Toward an understanding of inequity. Journal of Abnormal Psychology, 67, 422-436. Beam, B.T. Jr., & McFadden, J.J. (1992). Employee benefits. Chicago: Dearborn Financial Publishing. Bretz, R.D., Ash, R.A., & Dreher, G.F. (1989). Do people make the place? An examination of the attraction-selection-attrition hypothesis. Personnel Psychology, 42, 561-581. Chilton, K. (1993, November-December). Lincoln Electric's incentive system: Can it be transferred overseas? Compensation and Benefits Review, 21-30. Conte, M.A., & Svejnar, J. (1990). The performance effects of employee ownership plans. In A.S. Blinder (Ed.), Paying for productivity, pp. 245294. Washington, D.C.: The Brookings Institution. Cowherd, D.M., & Levine, D.I. (1992). Product quality and pay equity between lower-level employees and top management: An investigation of distributive justice theory. Administrative Science Quarterly, 37, 302-320. Davis, J.H. (1993, Autumn). ACA Journal. Quality management and compensation. Dyer, L., & Blancero, D. (1993). Workplace 2000: A delphi study. Center for Advanced Human Resource Studies, Cornell University. Eisenhardt, K.M. (1989). Agency theory: An assessment and review. Academy of Management Review, 14, 57-74. Ellig, B.R. (1981). Compensation elements: Market phase determines the mix. Compensation Review, (Third Quarter), 30-38. Fama, E.F., & Jensen, M.C. (1983). Separation of ownership and control. Journal of Law and Economics, 26, 301-325. Fisher, L.M. (1992, June 23). Sears auto centers to halt commissions. New York Times, D1. Gerhart, B., & Milkovich, G.T. (1990). Organizational differences in managerial compensation and financial performance. Academy of-Management Journal, 33, 663-691. Gerhart, B., & Milkovich, G.T. (1992). Employee compensation: Research and practice. In M.D. Dunnette , & L.M. Hough (Eds.), Handbook of industrial and organizational Psychology, 2nd Edition, pp. 481-569. Palo Alto, CA: Consulting Psychologists Press, Inc. Gerhart, B., Milkovich, G.T., & Murray, B. (1992). Pay, performance, and participation. In D. Lewin, O. Mitchell, & P. Sherer (Eds.), Research Frontiers in Industrial Relations, pp. 193-238. Madison, WI: Industrial Relations Research Association. Gomez-Mejia, L.R. (1992). Structure and process of diversification, compensation strategy, and firm performance. Strategic Management Journal, 13, 381-397. Gomez-Mejia, L.R., & Balkin, D.B. (1992). Compensation organizational strategy and firm performance. Cincinnati: South-Western.

Page 27

Employee Compensation

WP 95-04

Greenberg, J. (1990). Employee theft as a reaction to underpayment of inequity: The hidden cost of pay cuts. Journal of Applied Psychology, 75, 561-568. Heneman, H.G., III , & Schwab, D.P. (1979). Work and rewards theory. In D. Yoder , & H.G. Heneman, Jr. (Eds.), ASPA handbook of personnel and industrial relations. Washington, D.C.: Bureau of National Affairs. Hodgetts, R.M., & Luthans, F. (1993, March-April). U.S. multinationals' compensation strategies for local management: Cross-cultural implications. Compensation & Benefits Review, 25, 42-48. Hofstede, G. (1993, February). Cultural constraints in management theories. Academy of Management Executive, 7(1), 81-94. Jones, D.C., & Takao, K. (1993). The scope, nature, and effects of employee stock ownership plans in Japan. Industrial and Labor Relations Review, 46, 352-367. Judge, T.A., & Bretz, R.D. Jr. (1992). Effect of values on job choice decisions. Journal of Applied Psychology, 77, 261-271. Kaufman, R.T. (1992). The effects of Improshare on productivity.- Industrial and Labor Relations Review, 45, 311-322. Kerr, J.L. (1985). Diversification strategies and managerial rewards. Academy of Management Journal, 28, 155-179. Kidwell, R.E. Jr., & Bennett, N. (1993). Employee propensity to withhold effort: A conceptual model to intersect three avenues of research. Academy of Management Review, 18, 429-456. Kruse, D.L. (1993a). Profit sharing: Does it make a difference? Kalamazoo, MI: Upjohn Institute. Kruse, D.L. (1993b). Does profit sharing affect productivity? Tests using panel data on profit sharing features and personnel policies. Working Paper, Rutgers University. Lambert, R.A., & Larcker, D.F. (1989). Executive compensation, corporate decision-making, and shareholder wealth. In F. Foulkes (Ed.), Executive compensation (pp. 287-309). Boston: Harvard Business School Press. Lawler, E.E. III. (1989). Pay for performance: A strategic analysis. In L.R. Gomez-Mejia (Ed.), Compensation and benefits. Washington, D.C.: Bureau of National Affairs. Levine, David I., & Tyson, Laura D. (1990). Participation, productivity, and the firm's environment. In Alan S. Blinder (Ed.), Paving for productivity. Washington, D.C.: Brookings Institution. Locke, E.A., Feren, D.B., McCaleb, V.M., Shaw, K.N., & Denny, A.T. (1980). The relative effectiveness of four methods of motivating employee performance. In K.D. Duncan, M.M. Gruenberg, & D. Wallis (Eds.), Changes in working life (pp. 363-388). New York: Wiley. McAdams, J.L., & Hawk, E.J. (1992). Capitalizing on human assets through performance-based rewards. ACA Journal, 1(1), 60-73. Miles, R.E., & Snow, C.C. (1978). Organizational strategy, structure, and process. New York: McGraw-Hill. Milkovich, G. T. (1988). A strategic perspective on compensation management. Research in Personnel and Human Resources Management, 6, 263-288.

Page 28

Employee Compensation

WP 95-04

Milkovich, G.T., Gerhart, B., & Hannon, J. (1991). The effects of research and development intensity on managerial compensation in large organizations. Journal of High Technology Management Research, 2, 133-150. Milkovich, G.T., & Newman, J. (1993). Compensation (4th Ed.). BPI/Irwin: Homewood, IL. Milkovich, G.T., & Wigdor, A.K. (1991). Pay for performance. Washington, D.C.: National Academy Press. Murray, B., & Gerhart, B. (1993). Organizational outcomes from the introduction of a skill-based pay program. Working Paper. Center for Advanced Human Resource Studies, Cornell University. Noe, R.A., Hollenbeck, J.R., Gerhart, B., & Wright, P.M. (1994). Human resource management: Gaining a competitive advantage. Burr Ridge, IL: Austen Press/Irwin. Petty, M.M., Singleton, B., & Connell, D.W. (1992). An experimental evaluation of an organizational incentive plan in the electric utility industry. Journal of Applied Psychology, 77, 427-436. Pierce, J.L., Rubenfeld, S., & Morgan, S. (1991). Employee ownership: A conceptual model of process and effects. Academy of Management Review, 16, 121-144. Santora, J.E. (1991, February). DuPont returns to the drawing board. Personnel Journal, 34-36. Schuster, M. (1983). The impact of union-management cooperation on productivity and employment. Industrial and Labor Relations Review, 36, 415-430. Schwab, D.P. (1973). Impact of alternative compensation systems on pay valence and instrumentality perceptions. Journal of Applied Psychology, 58, 308-312. Tosi, H.L. Jr., & Gomez-Mejia, L.R. (1989). The decoupling of CEO pay and performance: An agency theory perspective. Administrative Science Quarterly, 34, 169-189. Tully, S. (1993, November 1). "Your Paycheck Gets Exciting." Fortune, p. 83+. Vroom, V.H. (1964). Work and motivation. New York: Wiley. Weitzman, M.L., & Kruse, D.L. (1990). Profit sharing and productivity. In A.S. Blinder (Ed.), Paying for productivity. Washington, D.C.: Brookings Institution.

Page 29