Time Allocation and the Weather

TL;DR: In this article, a theoretical model is developed to distinguish malfeasant from legitimate forms of employee sickness absenteeism, and a weather quality index is constructed to capture the influence of the weather on workers' preferences for outdoor leisure activity.

Abstract: The overriding theme of my dissertation is the use of short-term weather fluctuations to study how people allocate their time across activities. In Chapter 1, a theoretical model is developed to distinguish malfeasant from legitimate forms of employee sickness absenteeism. In this model, individuals’ marginal utility of indoor leisure is increasing in their sickness levels, while their marginal utility of outdoor leisure is an increasing function of the interaction of their health and the quality of outdoor weather. In equilibrium, sickness absenteeism occurs at both ends of the sickness distribution – among the relatively sick and among the most healthy facing the best weather. The positive relation between marginal changes in weather quality and levels of sickness absenteeism in the workplace reflects the substitution of the inframarginal employees who are the least sick away from work activities towards outdoor leisure activities. The model in Chapter 1 suggests an empirical strategy to identify a shirking component in overall reported sickness absenteeism. Not only does this approach avoid attributing entirely legitimate forms of absenteeism to shirking, but unlike previous studies using employee dismissal rates, it is able to distinguish shirking activity whether or not that activity is detected by employers. In order to exploit exogenous weather fluctuations to identify shirking activity, we need a one-dimensional measure of weather “quality”. The primary objective of Chapter 2 is to construct a weather quality index that captures the influence of the weather on workers’ preferences for outdoor leisure activity. The weather quality index takes into account the multifaceted nature of weather conditions, and measures how various weather elements – temperature, humidity, precipitation, wind speed, and cloud cover – come together to affect the propensity of employees to engage in high-utility outdoor recreational activities. The resulting index provides a ranking of different weather conditions in terms of their outdoor recreational values, which can then be used to capture the incentives of employees to shirk contractual work hours in response to purely exogenous weather changes. Chapter 3 empirically tests the existence of weather-induced substitution between work and outdoor leisure activities and examines how this type of behaviour varies across workers facing different shirking incentives. Linking 12 years of employee data from Canada’s monthly Labour Force Survey (LFS), which queries reasons for employees’ absences, to weather quality measured using the index constructed in Chapter 2, a clear positive relationship is found between the quality of outside weather conditions and short-term reported sickness absenteeism. Moreover, consistent with a key proposition of the theoretical model in Chapter 1, the empirical relation between weather and sickness absenteeism tends to be larger when existing shirking incentives are low, such as when sick pay is less generous and when probability of getting fired if caught shirking is high. There is, however, little evidence that firms are able to adjust shirking incentives through the payment of efficiency wages.