Flexibility, Specialization and Individual Productivity: Evidence from Call Center Data

Bundling tasks into jobs is an essential element of organizational design. We here examine whether workers should be assigned broader or narrower sets of tasks, i.e. should job design be “flexible” or “specialized”. We investigate the individual productivity effects of flexibility using a unique data set from a call center in Germany with daily performance data of 477 agents over a period of 19 months. At different times in their career, agents acquire new skills enabling them to carry out additional tasks that are related to different consumer types, different types of services, or activities (for instance, sales versus provision of information). Individual productivity is measured in the duration of calls (shorter calls being better) and the sales conversion rate. We show that flexibility (i) decreases individual productivity, both in a statistical and economic sense, but (ii) increases the capacity utilization rates of agents. To our knowledge, the paper is the first to show the trade-off associated with flexibility, which goes back to Adam Smith’s reflections on the gains of specialization, and how the extent of the market constrains the scope for it. We also investigate likely channels and show that switching costs are the most likely cause for the drop in individual productivity.