Does working from home boost productivity

Abstract

This study investigates the impact of remote work on employee productivity, exploring the roles of autonomy, training, and individual characteristics in dictating employee performance. Using cross-sectional data from the Chinese General Social Survey (CGSS), I employ various methodological approaches to address selectivity bias and endogeneity concerns inherent in the analysis. Initially, ordinary least squares (OLS) regression is utilized to estimate associations between remote work, autonomy, training, and productivity. Subsequently, a multinomial logit model is employed to identify which individuals are more likely to do remote work. To mitigate potential bias, propensity score matching estimator is used, considering observed heterogeneity between remote and non-remote workers. Furthermore, instrumental variable (IV)-free Gaussian Copula and Lewbel methods are also utilized to assess the causal effect of remote work on productivity. The findings reveal that remote work positively impacts employee performance, however, long hours or frequent remote work negatively affect productivity. Notably, a significant causal relationship is identified. Additionally, greater autonomy and training are associated with enhanced productivity. Moreover, demographic factors such as gender, marital status, and parental responsibilities influence remote work tendencies, with educated individuals exhibiting lower likelihoods of remote work, while those receiving training and receiving greater independence tend to opt for remote work arrangements.