تنظیمات نرم افزار-اجتماعی و خودانتخابی در مشاغل: شواهدی از پرستاران آفریقای جنوبی

Abstract Although a growing body of economic work has looked at the role of pro-social motives to explain self-selection into public or not-for-profit sector jobs, in particular in the delivery of social services, no attention has been given to the role of pro-social preferences in the decision to take up posts in rural and isolated areas. Yet there are reasons to believe that such choices involve a degree of self-sacrifice, in particular in developing countries where rural regions typically combine geographic hostile environment, high levels of poverty, low educational opportunities, limited access to basic services and worse career opportunities. On the other hand, as shortage of qualified staff is higher in rural areas, the returns on the presence of a health worker, and the benefits to the populations, are likely to be higher. Using data from a longitudinal study of nurses in South Africa this paper tests this hypothesis by linking experimental measure of pro-social preferences and revealed preferences outcomes. Three measures of pro-social preferences are constructed based on donations made by study participants in a dictator game played at baseline. Job choices are observed three years later for more than 97% of the initial sample. We show that the more dedicated the nurses – measured by their generosity towards patients in the dictator game – the more likely they are to have chosen a rural job. This result is robust to the inclusion of various demographic controls and to different econometric specifications. This finding contributes to the literature on role of pro-social values as an intrinsic motivation factor in labour supply decisions and it has policy implications for the provision of social services in difficult settings.

. Introduction Moving away from the traditional emphasis on purely selfish motives, a growing body of work has investigated the role played by pro-social preferences on individual decisions (see Meier, 2006 for a review). In relation to labour supply decisions, some attention has been paid to the role of pro-social preferences to explain the self-selection of individuals into the public sector or into not-for-profit organisations, in particular in the delivery of social services (Besley and Ghatak, 2005, Delfgaauw, 2007, Francois and Vlassopoulos, 2008, Kolstad and Lindkvist, 2012 and Serra et al., 2010). However, no attention has been given to the role of pro-social preferences in the decision to take up particular positions associated with low material welfare for the worker and high benefits for others, benefiting from the worker's presence. Yet if workers care about the benefits they bring to social service beneficiaries (e.g. patients, pupils), they might be willing to trade off some of their own welfare against the additional welfare recipients derive from having access to public services. Rural jobs of social sector in low- and middle-income countries typically fit this definition. Rural communities typically face challenging social and economic environment often exacerbated by isolation, extreme weather conditions, lack of public transport, difficulty of access to and limited choice of goods and services. Empirical evidence suggests that public goods, such as roads or health facilities, can generate welfare gains in rural areas (Jalan and Ravallion, 2002). Even when infrastructure such as hospitals, roads or schools are available, access to public services remains poor in rural areas as governments typically struggle to recruit and retain qualified staff in such posts (Dussault and Franceschini, 2007 and Mulkeen and Chen, 2008). There are many reasons why teachers, nurses or doctors are reluctant to be deployed in rural areas. Living in a rural isolated area compared to an urban one is associated with reductions in objective as well as subjective welfare (Fafchamps and Shilpi, 2009). Working conditions can also be harder in rural areas due to professional isolation, and professional advancement more limited as access to training and networking opportunities is more restricted (Hedges, 2002). Governments have used various strategies to encourage public servants to take up rural jobs. In health for example, many countries have imposed mandatory services in rural settings in exchange for subsidising nursing or medical studies (Frehywot et al., 2010). More frequently, financial incentives in the form of bonus payments, subsidised housing or hardship allowances have been used (Mulkeen and Chen, 2008 and Sempowski, 2004). Recognising instead that there might be some heterogeneity in the opportunity cost of working in rural areas (Hammer and Jack, 2002), some governments have sought to train as professionals individuals who are more likely to accept rural jobs at a later stage, for example by offering scholarships to increase the recruitment of graduates originating from rural areas (Grobler et al., 2009 and Ross and Couper, 2004). This paper tests whether pro-social preferences predict the choice of more altruistic positions. Using data from a panel of South African nurses, we are able to test whether generous behaviours observed in framed dictator games (DG) played at baseline, are associated with the choice of rural jobs, where health workers are able to help more needy people at the expense of their own personal welfare. Using decisions made by nurses played during the baseline survey, we construct three distinct measures of pro-social preferences: a generic measure of altruism (donation to a peer); a measure of pro-poor attitude (donation to a poor person) and a measure of nurses’ professional dedication (donation to a patient). We show that a higher dedication to patients is associated with an increased probability of taking up a job in a rural remote area. This result is robust to the inclusion of key socio-demographic variables generally associated with preferences for rural jobs, to different constructions of the dependent variable and dedication measure, and to various econometric specifications. On the other hand, we find weak evidence that a pro-poor attitude in the DG is associated with rural job choices, and no evidence of the effect of generosity towards one's peer. We find these results consistent with the idea that dedication measures social service providers’ willingness to sacrifice their well-being to increase the marginal benefits of service beneficiaries. This paper is organised as follows. Section 2 presents the related literature, and Section 3 describes the South African context and Section 4 the data used in the paper and the empirical strategy used. Section 5 reports the results and Section 6 presents different robustness checks. Section 7 briefly discusses the results and their implications.

. Discussion and conclusion This study presents the first empirical test of the role of pro-social preferences measured in the lab in the choice of jobs in deprived rural areas. Applied to the context of young nurses in South Africa, it explores the extent to which pro-social motives observed in the lab are predictors of career choices. The findings show that health workers who display higher levels of dedication towards patients in the lab are more likely to choose to work in rural areas, where their personal and professional lives are objectively more difficult. This result was particularly strong and consistent when looking at the relationship between dedication towards patients in the lab and rural jobs in health centres. We also found consistent evidence that generosity towards poor people in the lab was associated with choosing to work in a rural health centre. Two reasons probably explain this stronger association. First, the benefits derived by patients from the presence of a nurse are likely to be even greater than in hospitals. The latter are indeed typically staffed with several doctors and nurses, while it is not rare for nurses to be the only health care professional in health centres. Second, rural health centres are also located in more remote areas, where geographic barriers to access to alternative health care providers are greater. The presence of a professional nurse therefore makes a significant difference to the life of potential patients in the catchment area, more so than for rural hospitals. These findings have potential policy implications in South Africa. To ensure that a higher proportion of nurses are willing to choose rural posts, the recruitment processes of nursing students could try and ascertain the extent to which potential candidates show signs of dedication towards patients, or a commitment to improving patients’ wellbeing. Alternatively, assuming that dedication to patients can be nurtured, nurses’ curricula and experience during their studies could be adapted to cultivate positive attitudes towards patients. Finally, we found no evidence that pro-social preferences not specific to the particular decision-making context of nurses were relevant to their labour supply decisions. Specifically, generic altruistic decisions observed in a standard DG where the dictator sends money to fellow students, was never a predictor of job choices outside the lab. This finding suggests that lightly framed experiments might more relevant to explore the impact of social preferences on decisions made outside of the lab. This is particularly true for the investigation of the role of other-regarding preferences in a particular social setting, which can only be introduced in experimental economics by abandoning some degree of abstraction. This study contributes to the limited evidence on the relationships between decisions observed in the lab and choices made in real life, although the strength of the relationship observed here might be due to the particular characteristics of the population studied (nurses). With experimental (Jacobsen et al., 2011) and anecdotal (Miers et al., 2007 and Shaw and Degazon, 2008) evidence suggesting that nurses are likely to be more altruistic than the general population, does it make studying the role of social preferences in career decisions in that population less relevant or interesting? We believe not. In fact, if anything, detecting a relationship between pro-social motivation in the lab and career choices might be more difficult (and therefore more meaningful when detected) in a more altruistic population where one might observe more less variation in altruistic behaviour in the lab. What is more specific to nurses however, is the fact that part of their motivation to work is directly linked to social preferences, i.e. helping and serving others, more specifically patients. It is precisely the reason why they were a particularly relevant and interesting population in which to study relationships between social preferences in the lab and career decisions outside the lab. It would be interesting for further research to study the extent to which one can observe the same link between social preferences in the lab and career decisions made by individuals who are likely to choose their profession partly to benefit others (medical doctors, teachers).