A Structural Model of a Multitasking Salesforce: Multidimensional Incentives and Plan Design

The paper broadens the focus of empirical research on salesforce management to include multitasking settings with multidimensional incentives, where salespeople have private information about customers. This allows us ask novel substantive questions around incentive design and job while managing costs benefits information. To this end, introduces first structural model a in response incentives. also accommodates (i) dynamic intertemporal tradeoffs effort choice across tasks (ii) salespersoni?½s We apply our rich setting microfinance illustrate how address various identification estimation challenges. extend two-step methods used for unidimensional compensation plans by embedding flexible machine learning (random forest) first-stage policy function within an iterative procedure that accounts salesperson heterogeneity Estimates reveal two latent segments salespeople- i?½hunteri?½ segment is more efficient loan acquisition i?½farmeri?½ collection. Counterfactuals heterogeneous effects: huntersi?½ hurts firm as they engage adverse selection; farmersi?½ helps use it better collect loans. payoff complementarity induced multiplicative aggregation softens specialization hunters relative additive aggregation, but performance among farmers. Overall, task (acquisition) farmers (collection) selection harm overwhelms efficiency gain.