Labour Supply Decisions of Stockholm Taxicab Drivers
Sammanfattning: In this paper, panel data on hours worked and wage rates of taxicab drivers in the city of Stockholm are used to analyse whether reference-dependent preferences play a significant role governing labour supply decisions. The results suggest that only a small fraction of wage variation (about 1/5) is unanticipated implying that reference-dependence (which is relevant only in response to unanticipated variation in the wage rate) plays a limited role determining labour supply decisions. The result is confirmed by applying the discrete-choice stopping model (Farber, 2005) and the findings imply that the probability of ending a shift is positively related to accumulated hours (conditional on accumulated income) and seemingly unrelated to accumulated income (conditional on accumulated hours). This is inconsistent with the model of preferences dependent on a reference income level. The paper finds no evidence of a dual target suggested by Crawford and Meng (2011). In addition, heterogeneous behaviour across drivers' labour supply decisions is found. One possible factor explaining the differences is that driving a taxicab may be a "learning by doing"-process (LBD). Heterogeneous behaviour across drivers' labour supply decisions also highlights the importance of data quality and it could be argued that some of the spurious results found by Camerer et al. (1997) and Chou (2002) can be explained by a non-random sampling selection. The findings of this paper are generally consistent with Farber's (2015) results. Conclusively, extending the time horizon, some evidence for a longer planning horizon is found and precise wage elasticity estimates are positive and ranging from 0.20-0.35.
HÄR KAN DU HÄMTA UPPSATSEN I FULLTEXT. (följ länken till nästa sida)