By Keith A. Bender and Ioannis Theodossiou
The 19th Century African-American folk hero, John Henry, was said to be the best of all the steel-drivin’ men. According to some musical accounts of his life, he was bet that he could not beat a new steam powered hammer in a tunnelling contest. Henry won, but it cost him his life, as shown in a later line in the song, ‘And he laid down his hammer and he died’. The linkage between working ‘too hard’ and its consequences on health are clearly shown in the death of this legendary folk figure.
What might generate such a relationship? Two linkages immediately come to mind. First, performance pay may give incentives to workers to take risks to produce as much as possible, thereby increasing chances of on-the-job injury. Second, people may work more hours in order to produce more, substituting away from time that could be spent on safeguarding health, e.g. working rather than going for exercise, eating take-away rather than cooking a healthy meal, etc.
Although Smith made this observation over 200 years ago, there is little economic research on this linkage and most has focused on the performance pay-injury link. For example, Keith Bender, Colin Green and John Heywood examine European data to find a strong link between piece rate payments and injury.
Our paper investigates the second link – the relationship between health and performance related pay. The study uses the British Household Panel Survey (BHPS) to identify workers who were in good health in 1998 and then follow their health over time to see if they report a worsening in their health, looking at differences in whether they are employed in jobs which have a performance pay element or not. We look at four different measures of health: a subjective evaluation of overall health and three objective measures – heart problems, stomach problems and anxiety or depression problems.
First, we examine graphs of how long it takes a worker to go from good to bad health (in the statistical literature, this is called a ‘survival plot’). The graph below shows how the percentage of workers who started off in good subjective health continue to be in good health each year after 1998. Unsurprisingly, that percentage falls over time. However, the interesting finding is that the percentage of the healthy falls more quickly for those who have spent more than 50% of their time in performance pay (the red, dashed line). The story is no different for the three objective measures of health – the heart, stomach and anxiety or depression measures all have lower probabilities the longer a worker is in performance pay.
Of course, there could be other factors that drive these results – differences in gender or industry or previous health, etc. So next, we statistically control for such factors and find that the odds to drop into poor health increase dramatically as the percentage of time spent in performance pay increases. For subjective health, the odds are nearly double for a worker with 50% of his/her time in performance pay compared to a worker with no performance pay. Furthermore, the odds for developing a heart problem is 3.5 times higher for workers with 50% of time spent in performance pay. The health measures for stomach problems and anxiety/depression show an increase in the odds by more than 2.5 times.
What might explain these differences? Evidence from the BHPS is limited, but we find that performance pay is correlated with a higher prevalence of eating out and drinking heavily, behaviours linked to health problems in the medical literature. A more direct linkage is through increased stress. For a series of stress measures (for example, problems sleeping, making decisions, feeling under strain, etc.) the longer one is in a job with performance pay, the more likely one is to experience increased stress.
Thus, while performance pay may be linked to desirable economic outcomes like increased productivity, firms should be careful in implementing such payment systems and workers should be aware of the risks in taking jobs with these payment systems, as there may unintended health consequences of such jobs. Adam Smith was right after all!
Keith A. Bender and Ioannis Theodossiou are authors of ‘The Unintended Consequences of the Rat Race: The Detrimental Effects of Performance Pay on Health’, published in Oxford Economic Papers. Keith A. Bender is the SIRE Professor of Economics at the University of Aberdeen Department of Economics and Centre for European Labour Market Research (CELMR). His research examines labour market and heath interactions, as well as the labour market effects of educational mismatches. Ioannis Theodossiou is Professor of Economics in the University of Aberdeen Department of Economics and CELMR). His research focuses on issues related to the effect of socioeconomic conditions and unemployment on health and wellbeing outcomes, on the analysis of the unemployment problem and on issues related to pay determination.
Oxford Economic Papers is a general economics journal, publishing refereed papers in economic theory, applied economics, econometrics, economic development, economic history, and the history of economic thought.
Subscribe to the OUPblog via email or RSS.
Subscribe to only business and economics articles on the OUPblog via email or RSS.
Image credits: (1) John Henry Lies Dead After Beating the Steam Drill, painting by Palmer Hayden, Smithsonian American Art Museum, via Wikimedia Commons; (2) and (3) Both graphs courtesy of the authors. Do not reproduce without permission.