Journal Of Accounting Research.docx

Virtual Issue on Empirical Management Accounting Research Margaret A. Abernethy Department of Accounting The University of Melbourne Dennis Campbell Harvard Business School Harvard University 1. Introduction Management accounting research typically studies how to direct and motivate people ‘to do the right things’ through incentives, information, or both. Since its very first issues, JAR has drawn our attention to the difficulty of doing so as Devine (1964) highlights in his critique of A Behavioral Theory of the Firm.The empirical management accounting research published over the past decade in the journal notably picks up on this theme with a substantial focus on the role of management accounting and control systems in organizational contexts where tasks are complex, ambiguous and difficult to measure. We focus on a set of 13 empirical papers published in JAR over the past 10 years that points to the idea that management accounting and control practices in complex contracting environments are not easily understood solely through the lens of traditional agency theory. These papers tend to draw on a variety of theoretical frameworks ranging from newer economic models of organizational culture and relational contracts to related theories from sociology and psychology. They collectively give us new and important insights into the roles of incentives, information, and employee selection in shaping norms, decision-making, and performance in organizations. 2. Review of Selected Empirical Management Accounting Research Published in JAR During the Past 10 Years At least three empirical papers over the past decade explicitly explore management control issues in settings where output is complex and difficult to measure. Kachelmeier, Reichert and Williamson (2008) draw on both economics and psychology to study the effects of incentives on output in the context of creativity, an objective that is important for many firms but can be ambiguous and difficult to measure. Their well-designed experiment draws our attention to the complexity of creating incentives that encourage both productivity and creativity. Despite the possibility that firms can implement multidimensional performance measurement systems (e.g. “balanced scorecards”) that include measures of both productivity and creativity, their results indicate that people ‘tend to simplify multidimensional objectives by prioritizing on one objective over others’ and that ‘adding creativity to multidimensional performance measures can do more harm than good’. While not the direct focus of the paper, their results point to the importance of other forms of management control, such as employee selection, in contexts where creative or other output can be difficult to measure.

Campbell (2012) draws on economic theories of corporate culture to examine this latter issue directly by studying the role of selection as a means of influencing difficult to measure performance outcomes in the context of consumer lending. Traditional incentive solutions to control in his setting are difficult because employees are required to use judgment in their current lending decisions, but the outcomes of these decisions can occur far into the future and be affected by a variety of external factors. The results provide empirical evidence for the idea that “input” controls – namely, the selection of employees whose preferences are better aligned with organizational objectives – can be efficient and effective in complex contracting environments where output measurement is difficult or costly. While Campbell (2012) relies at least partially on the theory that employee selection and incentive contracts can be substitutes under some conditions and complements under others, Abernethy, Dekker and Schulz (2015) study this issue directly. They use survey data across firms to assess if there is a trade-off between employee selection and the use of incentive contracts. They limit their main analysis to firms which are expected ex ante to operate in complex contracting environments, namely those for which learning is a key organizational objective. The results are intriguing in identifying conditions under which employee selection can act as either a substitute or as a complement to incentive contracts. Another set of papers draws our attention to the importance of more subtle forms of control. These studies collectively suggest an important link between management control system design and organization culture through the development and reinforcement of social norms. Tayler and Bloomfield (2011) examine how personal and descriptive norms can function as informal controls, and how these norms can be activated or muted by formal control systems. They find in an experimental setting that by activating individuals’ self-interested norms, formal controls can mute desired personal norms (e.g., altruism) of individuals as well as their tendency to conform to the desired behaviors of others around them (i.e., the descriptive norms). Importantly, their evidence suggests these effects are long lasting and persist even after controls are changed. Their results demonstrate the potentially influential effect that formal controls can have on norms and behavior within organizations. Chen and Sandino (2012) draw our attention to the role of compensation in facilitating the development of social norms among rank and file employees. They collect archival data from two retail stores and draw on efficiency wage theories to study the mitigation of employee theft – a problem that is hard to control and costly to firms. Their work goes beyond the traditional honesty-inducing control mechanisms in the accounting literature to examine how paying workers more than their counterparts in other organizations – so called “efficiency wages” – can activate norms of reciprocity . The results are consistent with the idea that efficiency wages facilitate norms of reciprocity whereby employees are less likely to steal and collude with each other. Maas and Van Rinsum (2013) use an experiment to show how the firm’s incentive system can interact with its information policy to influence managers’ social preferences and their misreporting behavior. They find that the design of an incentive contract can determine whether managers’ social preferences will lead to more, or less, performance misreporting. They also show that managers’ misreporting behavior can be mitigated by their tendency to be honest (following personal norm) and their desire to appear honest in front of others. The openness of the information policy also increases honesty. This paper is notable in documenting that incentive contracts are not only useful in rewarding and motivating effort, but can also be used to shape the norms in the organization. Cardinaels and Yin (2015) draw our attention to how ‘tone at the top’ can influence social norms within a firm. Using a two-stage experiment they demonstrate that it is not just the design of the incentive contract that matters but the ‘soft’ information that flows from the choices a principal makes about providing incentives in the first place. When the principal actively uses an incentive contract rather than a fixed salary to reward employees, it conveys to a new agent that the descriptive norm in the firm is self-interest. The agent responds to this soft information by conforming to what they perceive are the descriptive norms of their peers. This has a negative consequence as the agent is now more likely to engage in undesirable behavior (e.g., misreporting).

Another set of related papers focuses on the prevalence of implicit incentives in organizations – that is incentives that do not reside in any explicit performance contract but depend on the expected future relationships between employees, managers, and firms. These studies examine different forms and consequences of implicit incentive contracts. Hales and Williamson (2010) examine implicit incentive contracts in a multi-period experimental setting. In their work, implicit incentives arise based on expectations about how the manager will allocate post-production resources between owners and employees. Their findings indicate that reputation concerns in this setting encourage higher firm productivity and employee payoffs, but only under conditions in which manager pay is relatively insensitive to the owner's ex post allocation. The findings are important in providing a potential explanation for why executive compensation is relatively insensitive to owner returns. Campbell (2008) explores implicit incentives by examining the role of nonfinancial performance metrics in organizational promotion decisions. Promotions provide incentives when they reward past performance with increased pay and rank in the organization. Promotions serve a matching function when they sort employees into the jobs for which their skills and abilities are best suited. The results show that non-financial performance measures facilitate this latter matching role by providing important information about the capabilities and expected future performance of lowerlevel employees. This use of nonfinancial measures for matching creates powerful multi-period implicit incentives in his research setting: employees subject to the strongest promotion incentives exert more effort and, more intriguingly, more learning on nonfinancial dimensions of performance. Grabner and Moers (2013) explore the incentive versus sorting role of performance measures even more directly. Using unique panel data from a retail bank, they show that as the difference between the current task environment and that at the next level of promotion increases, promotion decisions of superiors become less sensitive to objective measures of performance in the current job and more sensitive to subjective assessments of individual ability. In using novel data and clever empirical design to directly measure these dynamic shifts in the weighting of different types of metrics, their work provides important new insights on how performance measures are chosen and incorporated in promotion-based implicit incentive contracts. While the above papers focus largely on various types of formal and informal incentive mechanisms, a more recent stream of empirical research uses novel fieldexperiments to examine important questions on the potential impact of information on decisionmaking, productivity, and performance. Li and Sandino (2018) conduct a field experiment at a retail organization in which they were able to exogenously increase access to information about the creative work of peers through a randomized implementation of a new information sharing system. While they find no average effect of this system on the quality of creative work, job engagement, or financial performance, they do find important heterogeneous treatment effects. Positive effects of information access materialize only under three conditions in their study: (1) when users engage more actively with the system, (2) when users have greater potential to learn from new information given their lack of creativity, lack of exposure to others’ creative work (due to distance from other business units), or both; and (3) their need to customize their creative work to divergent local market conditions. Going beyond the dichotomy of access versus no access to information, Casas-Arce, Lorenco, and Martinez-Jerez (2017) conduct a field-experiment in which they exogenously vary both the frequency and level of detail of feedback on customer satisfaction and other nonfinancial performance measures. Their results are important and surprising in showing that more detailed and frequent feedback does not necessarily lead to better performance. In their field-setting, better performance is achieved when feedback is detailed but infrequent. Their results are consistent with a dynamic by which previous information is disregarded as new information becomes available. This information “salience” effect has potentially important theoretical implications for the optimal frequency of performance feedback. Eyring and Narayanan (2018) explore information effects in the context of peer comparison. Using the novel setting of online education, they conduct a field-experiment to measure the performance effects of setting high reference points (top quartile versus median). Their empirical analysis yields a performance effect that is concave in initial performance, with the effect being

negative for students below the median and positive for those between the median and top quartile. Their results point to reference points as a critical design choice in information display for purposes of peer comparison. Overall, these 13 papers collectively address the challenges of designing management accounting and control systems when firms face complex contracting environments. A key theme is that economic frameworks provide useful insights to study how informal controls and the design of the internal information environment can influence behavior, build desired social norms and perhaps even shape organizational culture. These frameworks recognize the importance of integrating social psychology into traditional economic models in order to solve complex control problems. This set of empirical papers also points to the opportunities of using unique databases and well-designed experiments to study accounting and control systems in practice, and they demonstrate the symbiotic relation that has developed between experimental and field-based research. We believe there remains great scope for management accounting researchers to get into the field and document empirically the nature, causes, and consequences of these often complex systems. Research in this area should also continue to be informed by theoretically grounded and well implemented experimental studies. We are optimistic that JAR has been, and seems poised to remain, at the forefront of publishing high quality research in this area. References 1. ABERNETHY, M. A., DEKKER, H. C. and SCHULZ, A. K. (2015), Are Employee Selection and Incentive Contracts Complements or Substitutes? Journal of Accounting Research, 53: 633-668. 2. CAMPBELL, D. (2008), Nonfinancial Performance Measures and Promotion‐Based Incentives. Journal of Accounting Research, 46: 297-332. 3. CAMPBELL, D. (2012), Employee Selection as a Control System. Journal of Accounting Research, 50: 931-966. 4. CARDINAELS, E. and YIN, H. (2015), Think Twice Before Going for Incentives: Social Norms and the Principal's Decision on Compensation Contracts. Journal of Accounting Research, 53: 985-1015. 5. CASAS‐ARCE, P. , LOURENÇO, S. M. and MARTÍNEZ‐JEREZ, F. A. (2017), The Performance Effect of Feedback Frequency and Detail: Evidence from a Field Experiment in Customer Satisfaction. Journal of Accounting Research, 55: 1051-1088. 6. CHEN, C. X. and SANDINO, T. (2012), Can Wages Buy Honesty? The Relationship Between Relative Wages and Employee Theft. Journal of Accounting Research, 50: 967-1000. 7. DEVINE, CARL THOMAS. “A Behavioral Theory of the Firm: A Review Article.” Journal of Accounting Research, vol. 2, no. 2, 1964, pp. 197–220. 8. EYRING, H. and NARAYANAN, V. G. (2018), Performance Effects of Setting a High Reference Point for Peer‐Performance Comparison. Journal of Accounting Research, 56: 581-615. 9. GRABNER, I. and MOERS, F. (2013), Managers' Choices of Performance Measures in Promotion Decisions: An Analysis of Alternative Job Assignments. Journal of Accounting Research, 51: 1187-1220. 10. HALES, J. and WILLIAMSON, M. G. (2010), Implicit Employment Contracts: The Limits of Management Reputation for Promoting Firm Productivity. Journal of Accounting Research, 48: 147-176. 11. KACHELMEIER, S. J., REICHERT, B. E. and WILLIAMSON, M. G. (2008), Measuring and Motivating Quantity, Creativity, or Both. Journal of Accounting Research, 46: 341-373. 12. LI, S. X. and SANDINO, T. (2018), Effects of an Information Sharing System on Employee Creativity, Engagement, and Performance. Journal of Accounting Research, 56: 713-747. 13. MAAS, V. S. and VAN RINSUM, M. (2013), How Control System Design Influences Performance Misreporting. Journal of Accounting Research, 51: 1159-1186.

14. TAYLER, W. B. and BLOOMFIELD, R. J. (2011), Norms, Conformity, and Controls. Journal of Accounting Research, 49: 753-790.