Policies based on physicians' geographic preferences, such as quotas and university expansion, are more cost-effective than financial incentives
The delivery of basic services, such as healthcare, hinges on one key resource: human capital. The lack of qualified health professionals in rural and underdeveloped areas, therefore, forms a barrier to improving health outcomes for those living in such places (e.g. Banerjee et al. 2004, Bjorkman and Svensson 2009, Okeke and Abubakar 2020).
In Brazil, as in most countries, while some regions have an abundant supply of physicians, others lack qualified health professionals that provide basic healthcare to the population. Figure 1 shows the number of physicians per thousand inhabitants in each state's capital and countryside in 2014. While the number of physicians per thousand people ranges from 1.42 to 11.9 across state capitals, the supply of doctors outside the capitals is substantially lower, ranging from 0.1 to 2 doctors per thousand people. The poorest states, mostly located in the northern regions, have notably fewer doctors than the richest states.
Figure 1 Number of physicians per capita in capitals (blue) and countryside (grey) by state (per 1,000 people)
Imbalances in the geographic distribution of physicians
The imbalances in the geographic distribution of physicians – with most concentrated in metropolitan areas – have been a matter of concern in both developing and developed countries. Many governments have resorted to the use of financial and non-financial incentives to recruit specialised professionals to disadvantaged regions in need (e.g. Bolduc et al. 1996, Holmes 2005, Kruk et al. 2010, Rao et al. 2012, Kulka and McWeeny 2018, Falcettoni 2018, Carrillo and Feres 2019). However, as the thriving literature on recruiting of health and frontline providers indicates, attracting qualified personnel to poorer locations has been proven challenging (Dal Bo et al. 2013, Ashraf et al. 2016, Finan et al. 2017, Deserranno 2019).
The Brazilian government has implemented a number of programmes to mitigate the undersupply of physicians in disadvantaged areas. The most relevant one – Mais Médicos, or the More Physicians programme – was created in 2013, using three strategies: (i) expansion and building of new primary healthcare units in needy areas; (ii) increasing the number of medical schools and medical residency programmes in areas suffering from undersupply; and (iii) opening of primary healthcare jobs with good wages in underserved areas (Carrillo and Feres 2019). While the original formulation of the programme considered investments in health infrastructure and medical schools, the main focus was on the opening of healthcare jobs in needy regions. However, despite its efforts, the government was unable to fill all of the newly created physician vacancies with Brazilian doctors. Figure 2 shows the number of vacancies created by the Mais Médicos programme and the number of physicians who graduated in Brazil that filled the vacancies (per 1,000 people) between July 2013 to July 2014. As is evident, most of the vacancies remained unfilled, especially those in the countryside and poorer states.
Figure 2 Job vacancies created by the Mais Médicos programme (in blue) and the number of Brazilian physicians that filled the open positions (in grey) per 1,000 people.
Note: The top panel shows the metropolitan regions across states; the bottom panel shows the countryside regions across states.
These figures suggest that policies based mainly on financial incentives were not sufficient to reduce regional imbalances. Instead, the main hurdle to overcome this imbalance is not the lack of positions with good wages, but some other aspect behind physicians' locational preferences.
Empirically measuring physicians’ locational preferences
In our study (Costa, Nunes, and Sanches 2019), we estimate physicians’ location preferences by exploiting practice location choices of all 60,563 generalist physicians that received a medical degree in Brazil between 2001 and 2013. We focus on generalists because they are directly responsible for the supply of basic healthcare and are often the focus of policies designed to reduce regional imbalances in the provision of health services.
We combine a series of datasets to track generalist physicians from birth, through medical school, and the first years of their professional lives. We observe that more than 50% of the physicians in our sample choose to work in the same region as they were born or completed medical school in. Metropolitan areas in the richest regions of the country are the main destinations of physicians that decided to migrate to a different region. Curiously, real wages in these metropolitan areas are relatively lower than in other regions. Yet, these areas have better amenities and health infrastructure.
Where do doctors choose to practice?
To better understand how physicians’ preferences depend on their own characteristics (age, gender, place of birth, college where they studied) and the attributes of the places that physicians chose to work right after finishing medical school (real average salary, quality measures of health infrastructure and local amenities), we estimate a discrete choice model with random coefficients (Berry et al. 2004).
Our estimates show that physicians' supply function is inelastic, with mean and median wage elasticity ranging around 0.4 and 0.7 in metropolitan areas and the countryside, respectively. While this implies that physicians are less sensitive to locations based on changes in their wages, our results also suggest that health infrastructure and amenities impact positively physicians' utility. Importantly, we find that physicians derive great utility for working close to their place of birth and for staying in the same region from where they graduated.
However, we find that preferences are heterogeneous according to the quality and prestige of the medical school from which the physicians graduated. Those who graduated from better medical schools value more local amenities, are more inelastic to wages, derive lower value for returning to their region of birth, and are the most inclined for staying in their locale of graduation.
Combined, these findings may explain why financial incentives in Brazil have not been effective to attract physicians to underserved areas.
Attracting physicians to underserved areas
We use our model estimates to study counterfactual policies that aim at reducing imbalances in the geographic distribution of physicians. Our benchmark geographic distribution of generalist physicians is such that the physician to population ratio is the same in all regions.
We find that policies exploiting physicians' geographic preferences are the most cost-effective:
- First, affirmative action policies in the form of quotas on student enrolment aimed at increasing the proportion of students born in underserved areas in medical schools appear to improve the geographic distribution of physicians by 63% at little cost.
- Second, the opening of new vacancies in medical schools in areas lacking generalists reduces this imbalance by 65%, but at higher costs compared to quotas.
- Third, an increase of 50% in wages paid by the public sector to doctors in needy areas is also effective, but at higher costs than the first two alternatives.
- Last, investing in health infrastructure is less effective and the costliest option.
Our study sheds light on new dimensions relevant to the locational choices of qualified professionals and provides cost-effectiveness analyses of different policies. We show that geographic preferences are decisive to explain the locational decisions of qualified health professionals. Importantly, our counterfactuals clarify that the cost-effectiveness of policies acting on physicians' preferences for working close to their home region (birthplace) and on preferences on working close to their graduation place are substantially different.
References
Ashraf, N, O Bandiera, E Davenport and S S Lee (2020), “Losing prosociality in the quest for talent? Sorting, selection, and productivity in the delivery of public services”, American Economic Review 110(5): 1355–94.
Banerjee, A, A Deaton and E Duflo (2004), “Wealth, health, and health services in rural Rajasthan”, AER Papers and Proceedings 94(2): 326–331.
Berry, S, J Levinsohn and A Pakes (2004), “Differentiated products demand systems from a combination of micro and macro data: The new car market”, Journal of Political Economy 112(1): 68–105.
Bjorkman, M and J Svensson (2009), “Power to the people: Evidence from a randomized field experiment on community-based monitoring in Uganda”, Quarterly Journal of Economics 124(May): 735–769.
Bolduc, D, B Fortin and M A Fournier (1996), “The effect of incentive policies on the practice location of doctors: A multinomial probit analysis”, Journal of Labor Economics 14(4): 703.
Carrillo, B and J Feres (2019), “Provider supply, utilization, and infant health: Evidence from a physician distribution policy”, American Economic Journal: Economic Policy 11(3): 156–96.
Costa, F, L Nunes and F Sanches (2019), “How to attract physicians to underserved areas? Policy recommendations from a structural model”, Working Paper, first draft 2015.
Dal Bo, E, F Finan and M Rossi (2013), “Strengthening state capabilities: The role of financial incentives in the call to public service”, Quarterly Journal of Economics 128(3): 1169–1218.
Deserranno, E (2019), “Financial incentives as signals: Experimental evidence from the recruitment of village promoters in Uganda”, American Economic Journal: Applied Economics 11(1): 277–317.
Falcettoni, E (2018), “The determinants of physicians’ location choice: understanding the rural shortage”, Working Paper.
Finan, F, B Olken and R Pande (2017), “The personnel economics of the developing state”, Handbook of Economic Field Experiments 2: 467–514.
Holmes, G M (2005), “Increasing physician supply in medically underserved areas”, Labour Economics 12(5): 697–725.
Kruk, M E, J C Johnson M Gyakobo, P Agyei-Baffour, K Asabir, S R Kotha, J Kwansah, E Nakua, R C Snow and M Dzodzomenyo (2010), “Rural practice preferences among medical students in Ghana: A discrete choice experiment”, Bulletin of the World Health Organization 88(5): 333–341.
Kulka, A and D McWeeny (2018), “Rural physician shortages and policy intervention”, Working Paper.
Okeke, E and I Abubakar (2020), “When a doctor falls from the sky: The impact of easing physician supply constraints on mortality”, Working Paper.
Rao, K D, S Zubin, R Sudha, N Khandpur, M Seema, H Indrajit, R Mandy, B Peter and V Marko (2012), “How to attract health workers to rural areas? Findings from a discrete choice experiment in India, volume 6”, The World Bank, Washington, D.C.