The effects of pay and job satisfaction on the labour supply of hospital consultants

There is little evidence about the responsiveness of doctors' labour supply to changes in pay. Given substantial increases in NHS expenditure, new national contracts for hospital doctors and general practitioners that involve increases in pay, and the gradual imposition of a ceiling on hours worked through the European Working Time Directive, knowledge of the size of labour supply elasticities is crucial in examining the effects of these major changes. This paper estimates a modified labour supply model for hospital consultants, using data from a survey of consultants in Scotland. Rigidities in wage setting within the NHS mean that the usual specification of the labour supply model is extended by the inclusion of job quality (job satisfaction) in the equation explaining the optimal number of hours worked. Generalised Method of Moments estimation is used to account for the endogeneity of both earnings and job quality. Our results confirm the importance of pay and non‐pay factors on the supply of labour by consultants. The results are sensitive to the exclusion of job quality and show a slight underestimation of the uncompensated earnings elasticity (of 0.09) without controlling for the effect of job quality, and 0.12 when we controlled for job quality. Pay increases in the new contract for consultants will only result in small increases in hours worked. Small and non‐significant elasticity estimates at higher quantiles in the distribution of hours suggest that any increases in hours worked are more likely for consultants who work part time. Those currently working above the median number of hours are much less responsive to changes in earnings. Copyright © 2007 John Wiley & Sons, Ltd.