Health workers are at the center of health systems, and the health workforce plays a key role in increasing access to health services for populations in developing countries. There are numerous challenges in this critical area of health policy in developing countries. At the global level, a 2006 World Health Organization analysis found that an additional 4.3 million health workers were needed to provide basic health-care services to populations in developing countries. A more recent analysis found that for 31 countries in sub-Saharan Africa, there will be a needs-based shortage of 800 000 health workers by 2015 and addressing this shortage would require more than 2.5 times the projected financial resources that will be set aside for health worker salaries in these countries. Various global, national, and regional analyses have demonstrated the link between having an adequate number of health workers relative to population and achieving key health service delivery and population health targets. The evidence suggests clearly that having an inadequate number of health workers is limiting the effectiveness of health service delivery in many developing countries.
However, the availability of health workers is not the only health workforce policy challenge in developing countries. In fact, growing empirical evidence would suggest that it is not even the main issue, at least in the short term, in many settings. Geographic maldistribution of health workers is one of the most persistent and widespread issues in developing countries. A recent analysis by the World Health Organization found that one-half of the world’s population lives in rural areas, but these areas are served by only 38% of the total nursing workforce and by less than a quarter of the total physicians workforce. Lack of health workers in rural areas is a major constraint to improving health service delivery. Low health worker productivity and quality limit the effectiveness of the existing health workforce. An analysis in five countries found an average health worker absenteeism rate of 35%. A recent review found that in India and Tanzania, doctors completed less than one quarter of the medically required tasks for patients presenting with tuberculosis (TB), diarrhea, or malaria. If these issues of geographic maldistribution, low productivity, and poor quality of care delivered by health workers is resolved, this could often have an immediate impact on health service delivery and population health outcomes in developing countries.
Why is it that countries with relatively similar epidemiological and disease profiles have vastly different numbers of doctors and nurses? Why are there unemployed nurses in countries that have far fewer nurses than needed to deliver basic care? Why do rural areas that often have the highest need for health services have the lowest staffing levels? Why are doctors absent in public facilities yet see patients in their private office? Why do health workers deliver care that is of lower quality than what they are trained to deliver?
As shown in this article, a labor economics perspective is extremely useful in understanding the underlying causes of these and other health workforce challenges developing countries are facing. Specifically, this article reviews the key factors that influence the demand for and supply of health workers and reviews the special features of the health labor market in developing countries. It also discusses how the labor economics perspective is extremely useful for policy makers when designing policy responses to the numerous challenges developing countries face.
A Labor Economics Perspective
A major focus of health workforce policy in developing countries historically has been to identify the ‘need’ for health workers of various skill sets, in various types of facilities and locations. Need can be defined as the number of health workers required to provide some mix of health services to the population. Need is a completely normative concept and takes into consideration only the epidemiological profile of a population, the preferences of policy makers over disease priorities, and technology considerations such as optimal skill mix, models of care delivery, and the expected productivity of health workers. Determining the need for health workers involves a great deal of priority setting among policy makers, but no economic factors such as prices or budgets enter into the needs discussion. There have been numerous studies that focus on identifying needs-based staffing levels. The World Health Organization estimates that worldwide greater than 4 million additional health workers are needed to deliver basic health services to the population. In Ethiopia an analysis estimated that 36% more physicians are required in order to expand antiretroviral treatment to target level. To scale up 42 priority health services, Tanzania is estimated to require greater than 100 000 full-time health workers by 2015, compared to a projected availability of less than 30 000.
Demand for health workers is defined according to the standard definition from labor economics: the total amount of labor, or in the simplest sense the total number of health workers, employers are willing to hire at current wages, holding constant other important variables such as health worker productivity, household income levels, political considerations, and government budget levels. The key distinction between need and demand is that many factors other than the health status of the population influence the demand for health workers. In other words, financial, economic, and political factors can be thought of as driving a wedge between the demand and the need for health workers. More importantly, the demand for health workers – and not the need – drives hiring behavior and, as a result, policy makers need to understand employer behavior in order to influence hiring decisions.
In the health sector, particularly in developing countries, there is a diverse set of employers of health workers. The main ones include global nonprofit employers (e.g., multilateral organizations that directly employ health workers), a country’s public sector (e.g., national, state, or local government directly or a government-owned hospital), a country’s for-profit sector (e.g., for-profit clinics), the nonprofit sector (e.g., mission clinics), and individuals (e.g., sick people who seek care from health workers and pay for their services out of pocket). The factors that influence the demand for health workers within each employer category are different. For the global nonprofit employers, a key factor is the level of resources various bilateral and multilateral agencies provide for health initiatives. For example, the large increases in donor resources for human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) in recent years led to a sharp increase in the demand for health workers who provide HIV care.
The public sector is a significant employer of health workers in both developed and developing countries. Within the public sector, hiring decisions are often influenced as much by political, macroeconomic, and social factors as by the needs of the population. In settings where health workers are employed as civil servants, the demand for health workers is influenced heavily by the total wage bill allocated to the health sector, which, in turn, is often a highly politicized process dependent on macroeconomic and fiscal policy priorities. In developing countries there tend to be constraints, for very sound reasons, on how fast the overall wage bill can expand. For various reasons, these overall wage bill constraints often, but not always, restrict the demand for health workers in the public sector. For example, in Kenya in the mid-2000s the overall wage bill policy of the government limited the Ministry of Health’s ability to hire health workers and expand service delivery, leading to high health worker unemployment rates. Even in decentralized settings, fiscal transfers to subnational governments (or even to facilities) are only very loosely based on the health-care needs of populations. As a result, the demand for health workers in the public sector often fluctuates with government (or facility) fiscal conditions, and this has been well documented empirically in both developed and developing countries. As noted in the Introduction section, the fact that addressing the needs-based shortage of health workers in sub-Saharan Africa would require more than 2.5 times the projected financial resources set aside for health worker salaries helps explain why there are so few health workers (relative to need).
The nonprofit sector operates similar to the public sector, except that specific agencies will focus on particular diseases, populations, or geographic areas. In developing countries this is important because the nonprofit sector is often a major employer of health workers, especially in very poor countries. Moreover, specific to developing countries, if significant levels of donor assistance for health are channeled through nonprofit organizations with little coordination with the government, this further increases the demand for health workers within the nonprofit sector.
In the for-profit sector, the demand for health workers is driven by profit maximization. Among individuals, the demand for health workers is influenced by the demand for health-care services, which is driven by a person’s health status, ability to pay, and other factors. In many developing countries, the individual-level market for health services is large, and as a result, individuals and households are a significant source of direct demand for health workers. In a sample of 15 countries in sub-Saharan Africa, for example, outof-pocket payments accounted for a low of 6% of the total health spending (Namibia) to a high of 62% (Chad).
The supply of labor in the health sector can be defined as the total amount of labor, or in the simplest sense the number of health workers, willing to work at current wages, holding constant other important variables like working conditions. A more refined definition could incorporate various aspects of effort, including productivity (e.g., hours worked, number of patients treated) and quality (e.g., care provided according to treatment guidelines). It is important to highlight several key decisions that influence the supply of health. These include the migration decision (whether to stay in the country), the labor force participation decision (whether to work or not), and the health-care labor force participation decision (conditional on working, i.e., whether to work in the health sector or in some other field).
Migration of health workers is an important issue in many countries, but especially in developing countries. As much as 70% of the medical workforce in sub-Saharan African countries eventually migrate. High rates of migration are also found in other regions. Many view migration as the single biggest challenge to strengthening health systems in developing countries. The health-care labor force participation decision is often overlooked by policy makers, yet has important implications. Several studies have shown that even small changes in the health-care labor force participation rate have important effects on the supply of health workers.
Migration and labor force participation decisions determine the supply of health workers within a country. Beyond that, the internal migration decision (which geographic area to work in), the sector decision (whether to work in the public or private sector), and the ‘effort’ decision (productivity and quality) influence the supply of labor in various settings of interest (e.g., a rural public clinic). When delineated this way, it is clear that there are intervention points to influence the supply of health workers that go well beyond simply adjusting enrollment levels within education institutions. Too often in developing countries, policy makers overlook several of these critical labor supply decisions.
Just as with the demand for labor, a host of factors unrelated to the health-care needs of the population influence the labor supply decisions of health workers. Migration decisions are influenced by relative wages, working conditions, individual and family characteristics, and preferences . Labor force participation of health workers depends on factors such as wages and working conditions and family income, and the health-care labor force participation decision is influenced by the wages and working conditions of jobs in the health sector compared to relevant jobs outside the health sector. All of these labor supply decisions are also heavily influenced by demographic characteristics such as age, gender, family size, and parental education levels.
It is clear that both the demand for and supply of health workers in developing countries are influenced by a complex set of factors that are unrelated – or at best loosely related – to the health-care needs of the population. This has an extremely important implication: the health labor market – even at market-clearing employment and wage levels – will not necessarily generate health worker employment outcomes that meet the needs of the population. From the policy maker perspective, this provides the rationale for intervening in the health labor market to influence demand, or supply, or both, to move employment outcomes closer to those that promote society’s goals with respect to health outcomes.
The Developing Country Context
Several aspects of the health labor market that are, if not unique, at least particularly relevant in the developing country context are worth discussing.
Remuneration Is Highly Regulated
In settings where health workers are employed as civil servants, remuneration levels are highly regulated and must be set within civil service regulations. Market forces do not exert a strong influence on health worker remuneration in such settings. Health worker salaries are rarely adjusted in response to actual or projected shortages or surpluses. Rather, they are set relative to other occupations (e.g., teachers) and relative to historical levels. For example, instead of being a function of market conditions, wages for one level of nurse are often set relative to a higher or lower level nurse or relative to another civil service worker with the same number of years of training and experience, such as a teacher. The empirical evidence suggests that remuneration regulations in developing countries – for both legal and political reasons – constrain health worker remuneration changes. In settings where health worker remuneration has been decentralized or removed from the overall civil service, there is much more autonomy for facilities to adjust remuneration in response to market signals.
Salary Is A Dominant Remuneration Method
The way doctors and nurses are paid can provide strong incentives for improving health worker productivity and quality of care. In many low-income countries, health workers in the public sector receive most of their compensation in the form of a salary. Along with weaknesses in governance, this is an important factor contributing to the significant level of health workforce absenteeism and low productivity many developing countries experience. Alternative types of payment mechanisms have the potential to provide stronger incentives to health workers and thereby improve performance and efficiency. Developed countries have a long history of alternative payment mechanisms, including fee-for-service, capitation, and performance-based pay, but only recently have developing countries experimented with innovative compensation policies. The benefit of performance-based pay is that it aligns the incentives and rewards to health workers with the particular objectives of the district or facility where health workers are employed, and several empirical studies have demonstrated this.
Remuneration Is Fragmented
Allowances are often a significant component of health worker remuneration. For instance, allowances account for 45% of the overall health wage bill in Kenya and 14% of the overall wage bill in the Dominican Republic. However, allowances are often fragmented and are not used strategically. In Kenya, for instance, the more lucrative housing allowance that accrues to doctors in the Nairobi area has created a disincentive to locate in remote areas. In the Dominican Republic, after a health worker leaves a location, the geographic allowance he or she was receiving turns into a permanent component of the worker’s wage. The allowance structure in many developing countries is often not designed to raise remuneration levels in less desirable work settings relative to remuneration in desirable settings. These compensating differentials in remuneration are necessary to recruit health workers to less desirable settings such as rural areas.
Donor Assistance For Health Is Significant
In many developing countries, particularly in sub-Saharan Africa, there are significant external resources devoted to investments in the health workforce. For example, approximately one-fifth of the UK’s support for the health sector in developing countries is channeled to health workforce activities. Although most of these health workforce resources are used to finance in-service training of health workers, agencies such as the Global Fund to Fight AIDS, TB and Malaria and the Global Alliance for Vaccines and Immunization devote significant resources toward health worker remuneration. When there are significant levels of donor assistance for health that are not fully coordinated with the government (through a national health strategy), this poses a challenge for health workforce policy. Nongovernment organizations and other nonprofit organizations are not subject to the same regulations as the government and, as a result, offer terms of employment that are very different than what is available to health workers in the public sector. This can generate significant movements of health workers across different sectors and can influence greatly the allocation of health workers across various priority programs.
There Are Administrative Inefficiencies In Key Management Functions
Owing to various reasons, including a centralized hiring process, the recruitment process in developing countries is often subject to significant delay and is not targeted to areas with the highest need for staff. For example, in Kenya in the late-2000s, it took an average of 10 months to fill a vacant position once a suitable candidate was found. With reforms to the hiring process, this has recently been reduced to an average of 3 months. In many developing countries, salaries follow people rather than remaining tied to a particular position. In other words, when health workers transfer or move, they often retain their remuneration level. This poses a significant challenge in that it limits the extent to which remuneration can be linked to a specific position (rather than person) and, therefore, the ability of policy makers to generate compensating differentials to attract health workers to less desirable settings. Decentralization, under certain conditions, has the potential to significantly reduce many of these inefficiencies in administrative procedures. For example, when Rwanda devolved remuneration authority to the local level, facilities were able to adjust payment levels to attract health workers to some of the hardest-to-fill positions.
Dual Job Holding Is Extremely Common
In developing countries, health workers often hold more than one job. For example, more than half of doctors in South Africa have additional employment outside of their primary practice. Often, the primary job is a salaried position within the public sector and the second job is after-hours in a private clinic. Although some governments explicitly allow dual job holding through part-time contracts (e.g., Dominican Republic), it is often poorly monitored and regulated. The challenge that dual job holding poses is that it limits the influence policy makers have on total remuneration and, therefore, the incentive structure health workers face within the entire healthcare system. Vietnam is a useful illustrative example. In Vietnam, salaries of physicians working in the public sector are set according to Ministry of Health policy and are deliberately set higher in rural areas in order to make rural postings more attractive. This is a sound strategy on the part of the Ministry of Health. However, when all sources of income are taken into consideration, including earnings from dual job holding, the total remuneration in urban areas ends up being much higher than in rural areas. In fact, the effective hourly wage in the second job (in the private sector) is almost double the primary job in the public sector. As a result of dual job holding, there is a considerable earnings disadvantage to locating in rural areas of Vietnam that the Ministry of Health’s salary structure did little to reverse.
Using A Labor Economics Perspectives To Guide Policy
The labor economics perspective suggests that to design effective health workforce policies, it is important to understand the overall labor market conditions in the health sector – namely, is the current employment level demand constrained, supply constrained, or at or near equilibrium? For example, when there are surpluses (i.e., few unfilled vacancies and unemployed health workers), it is necessary to stimulate demand in order to increase employment levels. In the public sector, this might be done through lowering wages or increasing resources available for hiring health workers. Negotiating lower wages in the public sector is difficult politically for the various reasons mentioned, but effective wages can be lowered through skill substitution (e.g., shifting tasks away from physicians toward nurses) or contracting with private agencies where total labor costs might be lower. Increasing the level of resources for salaries can be achieved through direct increases in Ministry of Health salary budgets or increased block transfers to districts or facilities (in a decentralized setting). Each strategy has its associated challenges, but there are several examples of countries that have successfully implemented these policies. Reducing the price of health services to households is also an effective way to stimulate demand for health-care and, therefore, for health workers. This can be achieved through reducing or removing user fees or other financial barriers to care. However, policies that aim to increase the supply of health workers are much less appropriate when there are labor surpluses. Increasing the number of graduates, for example, will likely increase health worker unemployment rates when employment levels are demand constrained.
When there are shortages of health workers (i.e., there are unfilled vacancies), a different set of policy options is required in order to change employment levels. In this case, the supply of health workers needs to be targeted. One option is to expand training capacity to increase the number of health workers, provided that graduates remain in the country. Higher wages, improved working conditions, and better continuing education opportunities are some of the interventions that will make jobs more attractive to health workers. Although wages tend to receive the most attention, evidence has shown that improving other job characteristics is often a more cost-effective way to attract workers to vacant posts.
Labor economics also offers some specific quantitative and qualitative analytic tools that can help generate empirical evidence to guide health workforce policy on specific issues. For example, qualitative analysis can be used to identify the critical job characteristics that influence health worker decisions to locate in rural areas and, more broadly, factors that influence health worker motivation and performance. A technique known as discrete choice analysis, in which potential workers are asked to rank jobs with different attributes (including, e.g., wage, location, and training) can be used to quantify the expected impact of alternative policies aimed at recruiting health workers to rural areas. The Government of Liberia recently implemented a rural area incentive program for nurses that directly incorporates findings from a discrete choice analysis. Labor force surveys can be used to measure current health worker remuneration differentials between different levels of care, specialties, and geographic areas, and the remunerations differentials that would be necessary to entice health workers to change job locations.
Through a better understanding of the underlying behavior of health workers and those that employ them, and how they interact in the health labor market, policy makers can more effectively design health workforce policies. The labor economics paradigm can be an important tool to help address the many health workforce challenges in developing countries and, ultimately, to improve the health of the population.
- Anand, S. and Barnighausen, T. (2004). Human resources and health outcomes: Cross country econometric study. Lancet 364, 1603–1609.
- Buerhaus, P. (2008). Current and future state of the US nursing workforce. Journal of the American Medical Association 300(20), 2422–2424.
- Buerhaus, P., Auerbach, D., Staiger, D. (2009). The recent surge in nurse employment: Causes and implications. Health Affairs w657–w668.
- Chaudhury, N., Hammer, J., Kremer, M., Muralidharan, K. and Rogers, F. H. (2006). Missing in action: Teacher and health worker absence in developing countries. Journal of Economic Perspectives 20(1), 91–116.
- Dussault, G. and Vujicic, M. (2009). The demand and supply of human resources for health. In Carrin, G., Buse, K., Heggenhougen, K. and Quah, S. (eds.) Health systems policy, finance, and organization, pp 296–303. New York: Elsevier.
- Lagarde, M. and Blaauw, D. (2009). A review of the application and contribution of discrete choice experiments to inform human resources policy interventions. Human Resources for Health 7(62), 1–10.
- McCoy, D., Bennett, S., Pond, B., et al. (2008). Salaries and incomes of health workers in sub-Saharan Africa. Lancet 371, 675–681.
- Serneels, P., Lindelow, M. and Lievens, T. (2008). Qualitative research to inform quantitative analysis: Health workers’ absenteeism in two countries. In Amin, S., Das, J. and Goldstein, M. (eds.) Are you being served? New tools for measuring service delivery, pp 271–298. Washington, DC: The World Bank.
- Vujicic, M. and Zurn, P. (2006). The dynamics of the health labor market. International Journal of Health Planning and Management 21(2), 1–15.
- Vujicic, M., Zurn, P., Diallo, K., Adams, O. and Dal Poz, M. (2004). The role of wages in the migration of health care professionals from developing countries. Human Resources for Health 2(3), 1–14.