Printer Friendly

Trade-offs in objective and subjective performance evaluation: a case study examining the validity of agency theory predictions.

AN EXAMINATION OF THE CONTROL SYSTEM USED BY WEGMANS FOOD MARKETS, INC., TO MEASURE AND EVALUATE THE PERFORMANCE OF ITS MANAGERS LOOKS AT WHETHER THE PRINCIPLES OF AGENCY THEORY CARRY THROUGH IN PRACTICE. THE AUTHORS INTERVIEWED SEVERAL TOP-LEVEL EXECUTIVES TO ASCERTAIN THE USEFULNESS AND FUNCTIONALITY OF THE SUBJECTIVE CONTROL SYSTEM.

The control function in organizations includes systems and procedures for establishing performance goals and ensuring that all organization members work toward those goals. (1) An effective control system must (a) establish decision rights (authority) for managers, (b) set performance expectations, and (c) measure outcomes in comparison with expectations. Agency Theory is an economics-based control system design that has been used extensively to model the control system choices available to firms. (2) This article summarizes predictions that arise from the agency model and presents the results of a case study designed to assess the validity of agency model projections. Given that many business school professors continue to propound the relevance and validity of Agency Theory principles, practicing managers need to know if these principles apply in the real world, provide useful guidance, and, thus, have merit. This article addresses these issues head on.

CONCEPTUAL FRAMEWORK

Our assessment focuses on the use of control systems for specifying and measuring the performance of managers, an area where accounting information plays a critical role. Performance goals for managers are difficult to stipulate and evaluate. In fact, options available to managers and choices made by them might not be clear to superiors, so control of managers' behavior is often based on financial and operational proxies that may be linked to incentives. (3) Therefore, it typically falls to accountants, as the information experts in organizations, to provide managerial performance measures. (4)

Decision rights, performance expectations, and performance measures for responsibility area managers are summarized in Table 1. Note that none of the performance expectations can be measured objectively or reliably. Superiors cannot explicitly determine, for example, whether cost center managers have minimized the use of resources to meet output requirements and consequently can never be sure if optimal choices were made. Thus, superiors have to use indirect and more subjective proxies for measuring managerial performance, including financial performance measures such as standard costs and segment earnings and nonfinancial measures such as performance evaluations and job ratings.

Given these difficulties, one might ask whether control systems are needed at the managerial level at all. Is it not sufficient to admonish managers to "do the right thing"? Is it not sufficient to hire "hard-working" managers? (5) There are two reasons for applying control systems to managers in organizations. First, there is the goal congruence problem. Lack of goal congruence arises among managers, their superiors, and other stakeholders because managers value goals of self-interest, such as pay, promotion, further career opportunities, and leisure time, while owners of for-profit organizations value increases in their financial capital, characterized by rising firm value. (6) Stakeholders in not-for-profit organizations may have more altruistic goals, but a lack of goal congruence with managers still may persist.

A second reason for a managerial control system is the private information problem. Managers may have access to private information about their area of responsibility that is unavailable to their superiors. For example, they are often far more familiar with day-to-day operations and thus are better able to detect deviations from the norm, both good and bad. In fact, obtaining and implementing private, decision-relevant information is precisely the role of a subordinate manager. If the superior's decision-relevant information were as complete as the subordinate's, there would be little need to delegate authority to a cadre of middle managers. A superior could make all decisions and simply delegate execution to nonmanagement assistants and line workers. Thus, developing and acting on private information defines the function of management-level subordinates. It is this responsibility that distinguishes them from clerical/hourly employees. (7) The resulting private information gap, however, can put superiors in an uncomfortable position when evaluating managers. Because they often have less information about the local environment, superiors may be unable to identify options available to managers and be uncertain about how decisions are made. (8)

The Agency Theory model provides a framework for discussing the control system trade-offs in an organizational environment where managers lack goal congruence with their superiors (and other stakeholders) and possess private, decision-relevant information that is unavailable to their superiors. Above all, our assessment focuses on the fundamental choice in control system design: whether to use an objective or a subjective performance evaluation system. An objective control system is characterized by the phrase "pay-for-results," i.e., the main evaluation criteria are the results of managers' actions, not the quality of those actions. These results are typically measured by financial metrics, such as earnings. Rewards and continued employment follow from the achievement of expected results without regard for the level of effort. In contrast, a subjective control system concentrates on the quality of managers' actions: Effort and skill are rewarded, regardless of outcome. Admittedly, few firms use strictly objective or subjective control systems: Most implement a system somewhere between the extremes. We address the polar cases to facilitate analysis of the pros and cons inherent in each approach.

SUBJECTIVE VS. OBJECTIVE CONTROL

If it could be implemented without cost, a subjective control system would be the most effective way to overcome the goal congruence problem. (9) Because subjective control is based on the quality of a manager's decision making, all options and actions are monitored. (10) Overseeing managerial decisions ensures that the superior's objectives remain paramount because the mutual knowledge that all decisions and actions are monitored deters subordinates from engaging in incongruent behavior. In any case, the superior can avoid incongruent decisions simply by overruling the subordinate before a decision is made.

But a thoroughly subjective system would be costly. Closely monitoring subordinates' options and actions is time-consuming and intrusive. It also requires a great deal of information and, thus, an expensive information system. (11) Consequently, subjective evaluation system costs arise in two ways. First, superiors must familiarize themselves with their subordinates' areas of responsibility. Given both cognitive and physical constraints, this typically requires a flatter organization structure with superiors responsible for fewer subordinates. Second, organizations must invest in advanced information technology to generate and distribute more information about managerial options. (12) Transaction and environmental monitoring has to be more sophisticated to ensure that superiors can access real-time information necessary for subordinate evaluations and early warnings, if necessary.

In addition to structural and information system expenses, subjective control systems can have substantial personal and political costs. Because the "quality" of subordinate managers' actions must be evaluated, superiors are required to identify and explain the flaws in a subordinate's performance. (13) This can be a very difficult task for many. It is troublesome to confront a subordinate with the judgment that, "You are not innovative enough" or "You missed opportunities to excel." Yet in a subjective evaluation system, superiors have no choice but to make such assessments because the perception of managerial quality is more important than measurable results. Moreover, political costs may arise in subjective systems because it can be difficult to defend qualitative performance evaluations to inside and outside observers. Internally, some managers may object to perceived favoritism shown their peers. Moreover, confronting subordinates may seem contradictory and counterproductive in an organization that prioritizes empowerment and autonomy. More importantly, however, subjective evaluation may be difficult to justify to external stakeholders, particularly when unanticipated changes to the environment (e.g., new competitors, severe economic turbulence, etc.) can confound the best efforts of hard-working, talented managers.

While significant structural, technological, personal, and political costs are inherent to subjective systems, objective control systems avoid many of them. (14) In contrast, little information is needed beyond a metric of actual vs. expected results. If earnings are used to measure results, either the expected earnings are achieved and incentives are provided, or expected earnings are not achieved and incentives are withheld. Objective systems can function with even a rudimentary information system infrastructure, and, because superiors are not required to evaluate the quality of subordinates' actions, a steeper reporting hierarchy is feasible. In comparison to a subjective system, substantial savings can result from a flatter corporate hierarchy and a less complex IT infrastructure. Moreover, objective evaluation systems may entail fewer personal and political implications. Superiors can avoid personally confronting subordinates with their shortcomings. In fact, objective rewards based on bonus formulas require little superior-subordinate interaction. Political costs are also lower because objective "pay for results" schemes are easier to clarify and justify to internal and external observers.

Yet all is not perfect with objective performance evaluations either. Such schemes can ignore the effect of uncontrollable environmental factors on managers' performance. (15) This so-called moral hazard problem can make it difficult for a superior to distinguish between a skilled, hard-working subordinate who overcomes a difficult environment and a less-skilled subordinate who meets objective goals through sheer luck or more favorable environmental factors. To be sure, objective results achieved by managers over time can establish reputations, and this certainly mitigates the moral hazard problem. Yet job rotation and job changes can exacerbate it in turn.

A more important problem with objective performance measures is that results-based criteria may affect managers' behavior in ways that are not always beneficial to the superior or to the overall organization. For example, using earnings as an objective outcome measure may cause managers to make decisions that increase earnings but not cash flow. Consider the production manager faced with an equipment replacement decision. To avoid an accounting loss on existing equipment that would decrease his or her earnings-based incentive, the manager may decide to reject an opportunity to acquire more-efficient equipment, even if the replacement would increase cash flow and, thus, firm value. (16)

Perhaps the most pervasive problem with objective control, however, is caused by the risk that is inherently transferred to managers. Because objective control focuses strictly on results, the risk of an uncertain environment is imposed on managers because outcomes are jointly determined by their actions and the environment. An adverse environment can overcome the best efforts of even a capable manager. Unlike owners who can diversify financial risk across their investment portfolio, (17) managers are unable to diversify performance risk to their human capital. (18) There are two fundamental problems with imposing performance risk on managers. First, risk-averse managers can become overly cautious in their decision making if their performance is evaluated strictly according to results. To mitigate risk to their human capital, they may avoid opportunities with potentially large and longer-term payoffs even if the opportunities seem desirable to their superiors or the owners. Thus, while objective schemes that "pay for results" may seem less costly than subjective control systems, the organization may miss out on high-expected-value opportunities that managers perceive as too risky. In addition, at the executive level, managers are able to mitigate the risk imposed on their human capital by diversifying firm holdings. (19) Research has shown, however, that executives undertake costly subsidiary acquisitions and dispositions in an apparent attempt to diversify firm-specific risk. For example, Philip Berger and Eli Ofek demonstrated that nonsynergistic intrafirm diversification holds little value for owners because they can independently diversify their investment portfolio. (20) Yet such diversification provides managers with the opportunity to diversify human capital risk. (21)

Finally, even in an objective control system, managers may find it necessary to explain the uncontrollable nature of a detrimental environmental effect. The argument may be: "It's not my fault I couldn't control the local economy and the actions of my competitors." Within the objective results-based system, however, the superior might not be able to evaluate the merits of the subordinate's case because of a lack of detailed information about the options or actions of the manager. Moreover, if explanations or excuses are accepted from one subordinate, they must be accepted from all; otherwise, superiors may be perceived as biased and inequitable. Therefore, even in a mainly objective evaluation system, circumstances can give rise to subjective issues requiring information system capabilities necessary for judging the quality of managers' actions and for taking into consideration mitigating circumstances that explain performance shortfalls. If ex-post explanations/excuses mitigate penalties for negative outcomes, an objective control system can transform into a subjective control system.

CASE STUDY: WEGMANS FOOD MARKETS, INC.

In order to assess these fundamental concepts of Agency Theory about control system alternatives, we conducted a case study of Wegmans Food Markets, Inc., an innovative and highly regarded food retailer headquartered in Rochester, N.Y. (22) Founded in 1916, Wegmans is a privately held, family-owned company that currently operates 71 stores, 48 of which are located in New York and the rest in other eastern states. Danny Wegman is CEO, and Colleen Wegman, his daughter, is president. Danny's father, Robert Wegman, was chairman of the company until his death at the age of 87 in April 2006. Wegmans' motto is "Every day you get our best," which it lists on its website along with four guiding principles under the heading "What We Believe":

(1) "At Wegmans, we believe that good people, working toward a common goal, can accomplish anything they set out to do."

(2) "In this spirit, we set our goal to be the very best at serving the needs of our customers. Every action we take should be made with this in mind."

(3) "We also believe that we can achieve our goal only if we fulfill the needs of our own people."

(4) "To our customers and our people we pledge continuous improvement, and we make the commitment: 'Every Day You Get Our Best.'"

Wegmans appears on Forbes' annual list of "Largest Private Companies in the U.S." In 2008 the company had annual sales of $4.8 billion and more than 37,000 employees. It ranked 34th on the 2007 Supermarket News list of the Top 75 by Sales Volume. Wegmans ranked third among the Fortune "100 Best Companies to Work For"[R] list in 2008. Data suggests that companies on the "100 Best Companies to Work For" list provide superior returns to investors, and the company has been on this list every year since it began in 1998. In fact, it was ranked first in 2005 and is ranked fifth in the current 2009 list. According to the Russell Investment Group, the Fortune "100 Best Reset Annually," which is a hypothetical portfolio containing stock from the companies on the "100 Best Companies to Work For" list, had an overall rate of return of 14.16% vs. a 5.97% return for the S&P 500 during the 1998-2006 time frame. (23) In other words, practices adopted by the "100 Best Companies" appear to lead to significant superior financial returns and are thus worthy of consideration (if not emulation) by other firms.

Given Wegmans' continued success in achieving a top-10 ranking among the "100 Best Companies," its operations can serve as an excellent case study for assessing control system practices with reference to Agency Theory. In other words, Wegmans' decisions about its internal control systems should be instructive to companies that aspire to the results the firm has achieved on a consistent basis. To analyze the control practices at Wegmans, we conducted structured interviews with four executive-level managers in finance, human resources, and store operations. (24) As summarized in the Appendix, we asked a number of specific and open-ended questions concerning three general themes: compensation, monitoring/information systems, and performance evaluation. The interviews lasted between 30 minutes to an hour and were, with one exception, conducted at the company's corporate offices in Rochester, N.Y. Transcripts of the interviews were verified by the subjects for accuracy and completeness.

CASE STUDY FINDINGS

Our field interviews indicated that while the company maintains an extensive database of quantitative information, in essence, Wegmans uses a subjective control system. According to Agency Theory, a subjective control system determines rewards on the quality of managers' decisions rather than on the results of these decisions. This seems to be the case at Wegmans. For example, one respondent stated that, at the company, "There is no bonus program.... We build growth and learning into every performance evaluation, and pay for knowledge." Agency Theory also predicts that in a subjective control system, performance evaluations will be highly dependent on subjectively determined abilities and perceived decision-making skills. Our interviews also found this to be the case at Wegmans. One respondent commented, "We look at career advancement as a journey to pick up different skills and intentionally move high-potential people into ... different functional areas." Another said, "Once an individual is identified as having high potential, they are given stretch assignments."

At Wegmans, the identification and rewarding of highly skilled and high-potential managers involve carefully planned, but clearly subjective and heuristic processes. When asked to list the criteria for performance evaluation, one respondent replied, "the ability to work with others and to work on teams; creativity; [and] ideas." Another described Wegmans' evaluation process for managers as follows: "We review them in two-way conversations, identifying ... things they have done well and ... things [they need] to improve." Therefore, personal one-on-one assessments are a key component of Wegmans' control system--clearly marking it as subjective, according to Agency Theory.

Furthermore, our interviews confirmed that Wegmans' subjective performance evaluation process is time-consuming and costly, requiring extensive effort on the part of the organization. This finding is likewise consistent with Agency Theory predictions that the success of a subjective control system is dependent on the organization's ability to identify and reward "high-potential" and "hard-working" managers. Agency Theory also suggests that an effective monitoring system is key to this effort. As described above, there are two dimensions to effective monitoring. First, the organization must invest in a sophisticated information system infrastructure. Second, a "flat" organizational hierarchy is required to permit superiors to evaluate the quality of subordinate managers' options and decisions. Figure 1 is an internal human resources document that Wegmans uses to delineate its managerial performance evaluation process. According to this chart, the company's subjective control system focuses on two main factors: (1) building trust and (2) documenting performance.

As Agency Theory would anticipate, our interviews revealed that in order to assess managers according to these criteria, Wegmans has to undertake extensive monitoring. All respondents consistently and repeatedly emphasized the capabilities of the company's information system infrastructure and the important role it has in their control methods. For example, when asked how familiar superiors are with subordinates' decisions, one respondent said, "We are very familiar because we have such strong monitoring systems.... We know the daily shrink on almost every one of our 212,000 items." Another commented, "We get daily information on product shrinkage and look at the performance of each store at least weekly. We do it every month as well, and that is done more formally. Labor is reported and worked on daily, and knowledgeable labor is rewarded." Another respondent said, "[We] are very attuned to what data [is] available and how subordinates should use the data. We supply a lot of data." Finally, another comment was, "Our systems are superior--nobody can hide from the information. It is widely and highly disseminated."

As indicated earlier, the second dimension of effective monitoring in a subjective control-oriented company, according to Agency Theory, is a "flat" organization structure. Flat structures afford superiors better understanding of their subordinates' responsibilities and thus help them to effectively identify and fast-track highly skilled subordinate managers. As would be expected, opportunities at Wegmans for both formal and informal contact between executives and managers are extensive and repeatedly emphasized by respondents as critical for effective control. One respondent commented, "We tend to be a flat organization, and superiors have a lot of direct interest and input. They are very familiar with [subordinates'] decisions." Another said, "An effective manager is wired to [his or her subordinates], and there is a lot of two-way communication." When asked about the extent of informal communication between managers and superiors, one respondent observed, "We believe in hallway management and do a lot over lunch." Superiors also engage in pre-action assessments of initiatives taken by subordinates. According to another respondent, "We try to understand the whole project before we approve specific requests." Finally, informal monitoring at Wegmans is extended to central management: "Our top managers visit every store between Thanksgiving and Christmas." It is also noteworthy that respondents agree that the ability to maintain informal contacts between corporate and store-level managers becomes increasingly difficult as the company expands geographically.

[FIGURE 1 OMITTED]

Overall, evidence drawn from our interviews about the structure of Wegmans' control system is remarkably consistent with the analytical framework described above. As anticipated by Agency Theory, Wegmans' clearly subjective control system is accompanied by a strong monitoring system consisting of a sophisticated information system and a flat organization structure, which allow formal and informal monitoring of subordinate managers' options and decisions. In other words, the presence of an accurate and timely information system assures that unwanted consequences of subjective control (bluffing, flattery, etc.) will be uncovered quickly. While this finding was unanticipated and not specifically discussed in the Agency Theory literature, it is consistent with the theoretical framework.

We also asked respondents to assess the effectiveness of their control system. In response, Wegmans executives reported general satisfaction with the current subjective control system, highlighting three benefits in particular. First, the current system allows a high level of communication through the management hierarchy. As one put it, "There is a great deal of informal communication, so we know about an issue right away--long before it can be formally reported.... We don't have a lot of surprises." In this way, Wegmans' subjective control system promotes communication that can provide subtle performance cues to subordinates. The second benefit of subjective control noted by Wegmans executives is that it helps to instill managers with the "culture" of the organization. For example, one respondent suggested that, "Our team approach helps keep down [functional] silos. In many cases, one manager will help another." Another commented that, "[We have] a family-oriented nature." Along the same lines, a respondent said, "The biggest thing that gets people in trouble ... is when word filters up that a manager is not a team player and doesn't treat people with respect." The final benefit afforded by Wegmans' current system, according to its executives, pertains to employee empowerment. Interviewees often repeated the premise that Wegmans' managers are held responsible for the company's "values." While these comments fully support Wegmans' "What We Believe" list, they are not of themselves strong evidence of employee empowerment. That said, these sentiments are supported by employee actions frequently and consistently observed by one of the authors, who regularly shops in Wegmans and sees employees taking personal initiative and responding to customers in ways that benefit the company and its customers.

Although executives expressed general satisfaction with Wegmans' current control system, we also asked whether they or the company had ever considered a more objective method. Several respondents did acknowledge desire for greater formality and objectivity in the process, in part because of the tremendous growth the firm had recently experienced. In fact, there has been discussion at Wegmans about the need to incorporate more results-oriented, formal metrics rather than to continue to rely on a more subjective system. For example, one respondent noted that "there is an ongoing debate [about whether] we should pay on merit.... But it isn't a driving need for us right now." Another said, "We have decided to focus on the subjective part and then attack the objective piece. This frees us to have more conversations on the subjective part and sustain the 'it's never good enough' mind-set." Still, others highlighted the problems they perceived might arise with more-objective control. One noted that "if we pay on product turnover, [managers] could bring in junk," implying that managers could "game" an objective evaluation system. Another mentioned that "the benefit of subjective evaluation is that you don't get hung up on making a specific goal," perhaps suggesting that objective goals would limit innovation.

Overall, Wegmans executives perceived the present control system as appropriate for the firm. They made very few critical comments and reported no serious consideration of change. But some do recognize that the company's family-oriented approach to managerial control is being tested by rapid growth and geographic expansion, both of which bring new challenges. In fact, one respondent indicated that being a privately held company helps mitigate market pressure for continued double-digit growth rates in key performance metrics. While we did not ask the respondent to explain these comments in greater detail, we infer that Wegmans could not maintain the intensity of its subjective-oriented control system and still sustain recent growth rates if it were publicly held. Alternatively, if maintaining recent growth rates became a prerequisite, more-objective controls would be needed--if only to evaluate a growing cadre of store and middle managers. In point of fact, Wegmans' actual growth rate has slowed in the past two years, although that may be due to the financial recession as well as to a decision to limit growth due to control considerations. In either case, all of this suggests to us that subjective assessment has its limits and, ultimately, can be at loggerheads with rapid growth.

AGENCY THEORY IN ACTION

Our survey of Wegmans executives provides strong support for the majority of Agency Theory predictions about control system design. For example, the company deploys many features of a subjective control system while at the same time committing significant resources to the firm's monitoring system. It has invested in a sophisticated information system that reduces information asymmetry between executives and managers. It also has an organizational structure that allows ample opportunities for formal and informal monitoring of subordinate managers. Moreover, at least according to our interviewees and the results of the "100 Best Companies" ratings, Wegmans has been successful both in building trust and creating common goal congruence. All of these are key management control issues, according to the Agency Theory framework.

The responses of the Wegmans survey do diverge from the predictions of Agency Theory in one significant way. While the theory suggests that subjective control systems afford managers less decision-making autonomy than objective methods might, our respondents suggested otherwise. One executive stated, "If you empower employees, like we do at Wegmans, we don't really need to know the options." Another admitted, "There are a lot of decisions we [superiors] don't know about." And a third recognized that, even in a subjective control system, "You have to trust your people to do the right thing." While these comments appear to diverge from the Agency Theory framework, they are supported by an extensive database that is relied on and routinely consulted by top executives--a practice that is fully compatible with theory. In addition, as reported earlier, bluffing, flattery, and other undesirable consequences of subjective control are quickly discovered and cannot be sustained by managers who achieve career advancement and longevity at Wegmans.

Our findings also indicate that while the company's control system practices are generally consistent with theoretical predictions, there is no evidence that Wegmans executives explicitly considered Agency Theory or any other conceptual framework in designing their control system. Although rational economic and behavioral theories are common in the academic literature, these theories have had no impact on control system design at Wegmans. Respondents expressed no awareness of economics-based theories of organizational control or corporate behavior. Instead, their control system choices were based solely on intuition and heuristics. For example, none of them mentioned the costs associated with the monitoring component of subjective control, i.e., extensive information system infrastructure and high levels of management time and effort. Moreover, even when asked leading questions, respondents did not seem to recognize that a more objective control system could be administered with a leaner management hierarchy and lead to significant information system savings. Our respondents also seemed unaware of the personal and political costs of subjective control. According to Agency Theory, superiors must bear the personal cost of confronting poorly performing subordinates and defending subjective performance evaluations. But none of these issues came up in the interviews.

Agency Theory suggests that, in an objective system, base pay would be reduced and supplemented with bonuses based on simple profitability or ROI measures. Such an objective system would also be much easier to administer and would require far less intervention by superiors. While this approach may not seem an optimal choice for a family-owned and managed company such as Wegmans, it is nonetheless remarkable that its executives have hardly considered implementing a more objective control system that would reduce the company's need to invest so much in monitoring systems and time-consuming subjective control procedures. That said, these findings are not altogether unexpected given that three of the four executive managers we interviewed are senior executives who have spent their entire careers of 25-30 years with the company and are therefore committed to its system. Also, because managers self-select their employment, the organization is likely attracting managers whose personal characteristics make them comfortable with a subjective control system, and it may discourage managers who would prefer more-objective controls. As objective control is more risky for subordinate managers, the firm may also be discouraging managers with higher risk tolerance.

Similarly, given the close monitoring of subordinates, problem-solving approaches tend to self-perpetuate under a subjective control system like Wegmans'. For example, when asked about the familiarity of superiors with managers' options and decisions, one respondent said, "We are pretty familiar with that because we have all done the work and moved up through the organization." Such close personal monitoring and the opportunity to pass on problem-solving techniques is a necessary prerequisite for effective subjective control. Finally, the top managers we interviewed have known each other for many years, were personally acquainted with the company's founder, remain close with current family executives, have fully embraced the Wegmans culture, and have experienced firsthand the benefits of subjective control. The challenge these executive managers face, as several readily admitted, is to maintain the benefits of a family-oriented (i.e., subjective) control system as the company moves into new markets, faces stiffer competition, and employs a greater number of managers who will be less influenced by company tradition and thus more responsive to monitoring practices and results-based incentives associated with objective control systems.

Our findings based on the Wegmans survey suggest a number of areas regarding the application of Agency Theory in the design of control systems that deserve further research. To corroborate the conclusions we have drawn from interviews at this large, privately held company, additional case studies should be undertaken at one or more publicly held companies. Moreover, to build on the current study--and in light of the strong emphasis on building trust and empowering employees we discovered at Wegmans--further studies could specifically examine the impact of a culture of trust and empowerment on both objective and subjective control systems. Future studies could also test Agency Theory principles in other real-world settings, including professional service firms and not-for-profit organizations.

Practicing managers as well as academics need to know if Agency Theory principles are applicable in today's competitive environment, in which an increasing number of firms stress the importance of qualitative values such as trust and empowerment. Our findings suggest that the majority of Agency Theory principles hold true at Wegmans, one of the country's most successful privately held companies. Although none of the executives we interviewed is aware of Agency Theory, its principles still apply, especially when firms embark on significant growth and geographic expansion. In other words the trade-offs between objective and subjective controls have important consequences that must be considered in theory as well as practice.

APPENDIX

Summary of Interview Questions

COMPENSATION SYSTEM

Q1. To what extent is your career advancement and compensation based on objective, results-based, performance measures, regardless of managers' effort?

Q2. To what extent is career advancement and compensation based on factors beyond a manager's control?

MONITORING (INFORMATION SYSTEM)

Q1. In general, how familiar are superiors with the options available to subordinates to effectively perform their jobs?

Q2. How familiar are superiors with the decisions subordinates have made to effectively perform their jobs?

Q3. To what extent do the following items affect managerial decision making?

1. approval limits for capital expenditures

2. approval limits for headcount

3. formal meetings to review your decisions

4. informal contacts with your superiors

5. required explanations for variance from plan

6. pre-action reviews for specific projects

7. pre-action reviews for day-to-day activities

8. policies and procedures manuals

Q4. How much detail is included in control reports from subordinate to superior managers, such as reports showing budget and actual data?

Q5. What is the reporting frequency of control reports from subordinate to superior managers, such as reports showing budget and actual data?

OPEN-ENDED DISCUSSION QUESTIONS

A1. Can you talk about managers' understanding of the choices that are made in control system design?

A2. Do managers really understand the trade-offs explained in Agency Theory (i.e., objective vs. subjective controls)? If not, do they act as if they understand?

A3. The fundamental trade-off that we examine in our paper is between objective control, which incurs the cost of risk-transfer (i.e., causing overly cautious decision making), and subjective control, which incurs the cost of monitoring (i.e., information system). Are you aware of this trade-off?

A4. Agency Theory suggests that information system costs arise in two ways. First, subjective control requires a flatter organization structure, with superiors responsible for fewer subordinates. Second, a more sophisticated information technology infrastructure is needed to generate and distribute more decision-relevant information to superiors about subordinates' options and actions. Do you view information system cost trade-offs in this way?

A5. The information system characteristics described in

A4 should be more apparent at higher levels in a corporate hierarchy simply because higher-level managers have a wider range of decision options available and there are many more potential actions leading to success. Is this an accurate description of information system characteristics?

A6. In a subjective control system, at least according to the literature, superiors "must" have a mechanism to understand the options available to their subordinates along with an understanding of the actions the subordinates have taken. Does this theory apply?

ENDNOTES

(1) Kenneth A. Merchant, "Evaluating General Managers' Performances," Strategic Finance, May 2007, p. 12.

(2) Kathleen M. Eisenhardt, "Agency Theory: An Assessment and Review," Academy of Management Review, January 1989, pp. 5774; and Leslie Kren, "The Role of Accounting Information in Organizational Control: The State of the Art," in Vicky Arnold and Steve G. Sutton, eds., Behavioral Accounting Research: Foundations and Frontiers, American Accounting Association, Sarasota, Fla., 1997, pp. 1-48.

(3) David S. Crandon and Kenneth A. Merchant, "Principles to Guide the Development and Use of Effective Performance Measures," Performance Improvement, February 2006, pp. 17-48.

(4) Anil Arya, Jonathan C. Glover, Brian Mittendorf, and Lixin Ye, "On the Use of Customized vs. Standardized Performance Measures," Journal of Management Accounting Research, 2005, pp. 7-21. We focus here on control for managers where accounting information plays a critical role. For clerical/hourly employees, where required behavior can be more easily specified and control can be based on standard operating procedures and HR policies, accounting information plays little or no role.

(5) The latter suggestion has merit, although the characteristic of being a "hard worker" can be difficult to identify, particularly in new hires.

(6) Michael Jensen and William H. Meckling, "Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure," Journal of Financial Economics, July 1976, pp. 305-360.

(7) Stanley Baiman and John H. Evans, III, "Pre-decision Information and Participative Management Control Systems," Journal of Accounting Research, Autumn 1983, pp. 371-395.

(8) Bengt Holmstrom, "Moral Hazard and Observability," Bell Journal of Economics, Spring 1979, pp. 74-91.

(9) Stanley Baiman and Joel S. Demski, "Economically Optimal Performance Evaluation and Control Systems," Journal of Accounting Research, 1980 (Supplement), pp. 840-848.

(10) Robert H. Chenhall and Kim Langfield-Smith, "Performance Measurement and Reward Systems, Trust, and Strategic Change," Journal of Management Accounting Research, January 2003, pp. 117-143.

(11) Ibid.

(12) Leslie Kren and Jeffrey L. Kerr, "The Effect of Behavior Monitoring and Uncertainty on the Use of Performance-Contingent Compensation," Accounting and Business Research, Spring 1993, pp. 159-167.

(13) Michael Gibbs, Kenneth A. Merchant, Wim A. Van der Stede, and Mark E. Vargus, "Determinants and Effects of Subjectivity in Incentives," The Accounting Review, April 2004, pp. 409-436.

(14) Jan Bouwens and Laurence van Lent, "Performance Measure Properties and the Effects of Incentive Contracts," Journal of Management Accounting Research, July 2006, pp. 55-75.

(15) Eisenhardt, January 1989.

(16) This example also illustrates the potential for a private information problem in an organization: The need for and availability of new equipment is likely to be the private information of managers, unavailable to superiors. Yet as the firm's information system capabilities are increased (as in a subjective control system), more private information becomes available to superiors.

(17) Shareholders are concerned only about market risk, not firm-level risk.

(18) The opportunity to diversify performance risk increases as one moves up the corporate hierarchy because superior managers can diversify performance risk among a number of different subordinate managers. Thus, superiors are less subject to performance risk than their subordinates, compounding the goal congruence problem. But managers are unable to completely diversify performance risk to their human capital.

(19) Anup Agrawal and Gershon N. Mandelker, "Large Shareholders and the Monitoring of Managers: The Case of Antitakeover Charter Amendments," Journal of Financial and Quantitative Analysis, June 1990, pp. 143-162.

(20) Philip G. Berger and Eli Ofek, "Diversification's Effect on Firm Value," Journal of Financial Economics, January 1995, pp. 39-65.

(21) Yakov Amihud and Baruch Lev, "Risk Reduction as a Managerial Motive for Conglomerate Mergers," Bell Journal of Economics, Autumn 1981, pp. 605-617.

(22) For example, Wegmans reports on their website that in 2007 they received 4,284 requests from people asking the company to open a store in their community, and another 4,100 customers wrote to say how much they enjoy shopping at Wegmans.

(23) Rates of return are computed by "Reset Annually" portfolio, which invests equal dollar amounts (at the beginning of 1998) in the stock of each of the 1998 "100 Best" publicly traded companies. The portfolio is liquidated at the end of 1998, and the proceeds are invested in the 1999 list by buying equal dollar amounts of each publicly traded firm on the 1999 list. This process of liquidating the portfolio at the end of the year and using the proceeds to invest in the new list of "100 Best" is repeated annually.

(24) The interviewees were Jim Leo, senior VP & CFO; Gerry Pierce, senior VP of Human Resources; Karen Shadders, VP of People; and Jack De Peters, EVP of Store Operations.

Leslie Kren, Ph.D., CPA, is an associate professor at the University of Wisconsin-Milwaukee. You can contact Leslie at (414) 229-6075 or lkren@uwm.edu.

Tom Tyson, Ph.D., CMA, is a professor of accounting at St. John Fisher College in Rochester, N.Y. You can contact Tom at (585) 3985-8431 or ttyson@sjfc.edu.
Table 1: Performance Expectations and Accounting-based
Performance Measures

 Performance Performance
 Decision Rights Expectations Measures

Cost Center Input mix Meet output Input standards,
 (labor, requirements flexible
 materials, and while minimizing budgets,
 technology) the use of activity-based
 inputs management
 (ABM),
 nonfinancial
 operational
 measures

Profit Center Input mix, Maximize present Measures above
 product mix, value of cash plus earnings,
 prices, with flow flexible budgets
 control over
 product-market
 constraints

Investment Input mix, Maximize present Measures above
Center product mix, value of cash plus return on
 prices, with flow investment,
 control over residual income,
 product-market Economic Value
 and resource Added (EVA[R]),
 constraints balanced
 scorecard
COPYRIGHT 2009 Institute of Management Accountants
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

 Reader Opinion

Title:

Comment:



 

Article Details
Printer friendly Cite/link Email Feedback
Author:Kren, Leslie; Tyson, Tom
Publication:Management Accounting Quarterly
Article Type:Case study
Geographic Code:1USA
Date:Jan 1, 2009
Words:6616
Previous Article:Measuring the costs of responding to business risks.
Next Article:The proposed financial reporting overhaul and business valuation: proper disclosure of financial information can be critical to a company's health....
Topics:

Terms of use | Copyright © 2015 Farlex, Inc. | Feedback | For webmasters