Many service organizations rely on information sharing systems to boost employee creativity to meet customer needs. We conducted a field experiment in a retail chain, based on a registered report accepted by JAR, to test whether an information sharing system recording employees’ creative work affected the quality of creative work, job engagement, and financial performance. We found that, on average, this system did not have a significant effect on any of these outcomes. However, it significantly improved the quality of creative work in stores that had accessed the system more frequently and in stores with fewer same‐company nearby stores. It also improved creative work and job engagement in stores in divergent markets, where customers needed more customization. We found weak evidence of better financial results where salespeople had lower creative talent before the system was introduced. Our findings shed light on those conditions in which information sharing systems affect employees’ creative work.

We examine how changing the allocation of hiring decision rights in a multiunit organization affects employee-firm match quality, contingent on a unit's circumstances. Our research site, a U.S. retail chain, switched from a decentralized hiring model (hiring by business unit managers—in our case, store managers) to centralized hiring (in this study, by the head office). While centralized hiring can ensure that enough resources are invested in hiring people aligned with company values, it can also neglect the unit managers' local knowledge. Using difference-in-differences analyses, we find that the switch is associated with relatively higher employee departure rates and, thus, poorer matches if the business unit manager has a local advantage; that is, if the store serves repeat customers, serves a demographically atypical market, or poses higher information-gathering costs for headquarters. In these cases, the unit manager may be more informed than headquarters about which candidates best match local conditions. Data Availability: The analyses presented in this study are based on data shared by a U.S. retail company. The data are confidential, according to a nondisclosure agreement between the company and the authors.

We provide fresh evidence regarding the relation between compensation consultants and CEO pay. First, firms that employ consultants have higher-paid CEOs—this result is robust to firm fixed effects and matching on economic and governance variables. Second, while this relation is partly due to consultant conflicts of interest, it is largely explained by the impact consultants have on the composition and complexity of CEO pay plans; notably, this impact fully mediates the consultant-CEO pay relation. Third, firms with higher-paid CEOs and more complex pay plans are more likely to hire a consultant. Last, Say-on-Pay voting patterns suggest shareholders view positively the advice consultants provide, but only when consultants provide no other services. We also find suggestive evidence of boards “layering” new equity incentive plans over existing ones, thereby increasing the impact of composition and complexity on CEO pay beyond the premium the CEO would demand for bearing additional compensation risk. JEL Classifications: J33; M12; M52; M48. Data Availability: Data are available from the public sources cited in the text.

Many companies operate units that are dispersed across different types of markets, serving significantly divergent customer bases. Such dispersion is likely to compromise headquarters' ability to control local managers' behavior and satisfy the needs of different customer types. In this study we find that market-type dispersion is an important determinant of the delegation of decision rights and the provision of incentives. Using a sample of convenience store chains, we show that market-type dispersion is positively associated with the degree of franchising at the chain level as well as the probability of franchising a given store within a chain. Our results are robust to alternative definitions of market-type dispersion and to other determinants of franchising such as the stores' geographic dispersion. Additional analyses suggest that chains that do not franchise cope with market-type dispersion by decentralizing operations from headquarters to their stores and providing their store managers higher variable pay.

This study explores decisions related to formal empirical tests of business models and interpretations and uses of those tests. Business models describe managers' rationales as to how their organizations will achieve success. This study documents a test of one company's business model under seemingly favorable conditions for such a test—a successful single-product firm following a consistent strategy over a long period of time with stable management and publicly traded stock. Although the findings provide only weak support for the hypothesized business model, the confidence of the company's top managers in their business model remained high. Further analyses reveal that the managers' response to the test results is consistent with that expected of Bayesian-rational agents. Our analyses provide the basis for development of a framework for understanding the expected value of testing business models in various circumstances. This framework might explain apparent contradictions between previous studies containing normative statements regarding the value of testing business models. Data Availability: The data used in this study are derived from a proprietary dataset and public sources.