Why do health labour market forces matter?
Barbara McPake a, Akiko Maeda b, Edson Correia Araújo b, Christophe Lemiere b, Atef El Maghraby c & Giorgio Cometto d
a. Institute for International Health and Development, Queen Margaret University, Edinburgh, Scotland.
b. The World Bank, 1818 H Street, NW, Washington, DC, 20433, United States of America.
c. African Development Bank, Tunis, Tunisia
d. Global Health Workforce Alliance, World Health Organization, Geneva, Switzerland.
Correspondence to Akiko Maeda (e-mail: email@example.com).
(Submitted: 10 March 2013 – Revised version received: 12 June 2013 – Accepted: 13 June 2013.)
Bulletin of the World Health Organization 2013;91:841-846. doi: http://dx.doi.org/10.2471/BLT.13.118794
Health workforce challenges
Human resources for health are central to any health system insofar as health workers perform or mediate most health system functions. They make treatment decisions at the point of service and their actions determine how efficiently other resources are used.1 Health-care delivery is highly labour intensive. To be effective, a health-care system must have the right number and mix of health-care workers and it must ensure that they possess the means and motivation to skilfully perform the functions they are assigned. Many countries are facing a “crisis in human resources for health” that involves three dimensions: availability, which relates to the supply of qualified health workers; distribution, which relates to the recruitment and retention of health workers where their presence is most needed; and performance, which relates to health worker productivity and to the quality of the care that health workers provide.
Traditional approaches to resolving human resource constraints in the health sector have relied primarily on workforce planning, i.e. the practice of estimating health workforce requirements based on a country’s epidemiological and demographic profile and of scaling up education and training capacities to narrow the gap between the existing number of health workers and the number required. However, focusing narrowly on the production of health workers results in the neglect of other important factors that influence human resource capacity, such as labour market dynamics and the behaviour and preferences of the health workers themselves. Thus, despite the extensive published literature on the human resource crisis in the health sector, few analyses have been conducted using labour economic frameworks and the dynamics of labour markets remain little known or understood, especially in low- and middle-income countries.
The health status of a population, its health-care needs and its requirements in the area of human resources for health are linked in complex ways.2 For example, the employment opportunities available to health workers and the type of employment conditions that health workers prefer are not always aligned with priority health-care needs. Health workers may be attracted to positions that do not respond to such needs or may choose to migrate in search of alternative employment opportunities. Sometimes a paradoxical situation arises: vacancies in high-priority positions in the public sector coexist with high unemployment rates among health workers. This paradox is explained by the labour market failure to match the supply and demand for health workers. For instance, several African countries (e.g. Kenya, Mali and Senegal) are experiencing acute under-employment among doctors and nurses, yet they are simultaneously investing substantial public funds in producing more health workers. This worsens underemployment and reduces the efficiency of government expenditures. The system of posting health workers to rural areas further illustrates the limitations of traditional workforce planning. When health workers are officially assigned to a remote rural area, they often find unofficial ways to evade the assignment and find employment in an urban area. These examples highlight the inadequacy of a human resource strategy focused exclusively on the needs-based production of health workers.3
Labour market conditions such as low salaries and a lack of other economic benefits are known to influence employment processes, but their influence on the planned allocation of resources is less widely recognized. Thus, an analysis of the labour market is essential to achieve a better understanding of the forces that drive health worker shortage, maldistribution and suboptimal performance and to develop policies and interventions tailored to different labour market conditions.
What is a labour market?
A market is any structure that allows buyers and sellers to exchange goods, services or information of any type. A labour market is the structure that allows labour services to be bought and sold.4 In a labour market, those who seek to employ staff are the “buyers” and those who seek employment are the “sellers”. A labour market can be delineated according to different criteria: geographical (national or international); occupational (specialized or unspecialized); and sectoral (formal or informal). A special feature of labour markets is that labour cannot be sold but only rented. Furthermore, conditions of employment (e.g. adequate infrastructure, supportive management, opportunities for professional development and career progression) are as important as prices (wages) in determining labour market outcomes. Thus, these outcomes are driven by the behaviour of employers and workers in response to changes in the terms of employment (wages, compensation levels and working conditions).
In a well-functioning labour market, wages or “compensation” – which can be understood as the overall return received for employment in a particular post and not only the financial component of that return – act as the mechanism whereby the intentions of buyers and sellers are reconciled. The demand and the supply of labour tend towards equilibrium. A point is reached in which the amount of labour supplied equals the amount demanded at the going level of compensation. Labour markets are said to “clear” when the supply of labour matches the demand for workers. However, labour markets do not always “clear” in this way. When they fail to do so, they exhibit either labour surplus or unemployment (i.e. people seek jobs at the going rate of pay but cannot find them), or labour shortage (i.e. employers seek to fill posts at the going rate of pay but cannot find people to fill them). These possible scenarios are illustrated in Fig. 1.
Fig. 1. Possible labour market scenarios
A health labour market is a dynamic system comprising two distinct but closely related economic forces: the supply of health workers and the demand for such workers, whose actions are shaped by a country’s institutions and regulations. Traditionally, analyses of human resources for health have been framed as a supply crisis, with demand-side factors receiving scant attention. The demand for health workers in a country is determined by what government, private sector and international actors, such as donors and multinational corporations, are willing to pay to hire them. “Willingness to pay” is dependent on the level of health-care financing and the willingness to hire health workers depends on the money available for doing so. “Willingness to pay” marks a distinction between demand and “need”. A mismatch commonly exists between the financial resources available – and hence, willingness to pay – for employing health workers and the number of workers needed to cover the health-care needs of the population.2 The supply of health workers is influenced by the level of remuneration and by many other factors that are economic, social, technological, legal, demographic and political. Fig. 2 illustrates the dynamics of the health labour market.
Fig. 2. Framework for analysis of health workers labour market dynamics
Markets fail to “clear” for two reasons. Either prices are not flexible (e.g. wages may be fixed by legislative or bureaucratic process or tied to civil service schedules that are insensitive to health market conditions), or demand or supply does not adjust to price signals. This may be because “demand” is predominantly defined by government and driven by legislative or bureaucratic process rather than by market forces, or because supply is regulated. An example of such regulation is offered by graduates who are “bonded” and constrained from exiting the labour market on their own volition. In health labour markets both types of rigidities are common. The market clearing position may be politically unacceptable. It may, for example, result in unaffordable health services and, in this respect, failure to “clear” may be a foreseeable result of price control. In such cases, health labour market analysis will allow a forecast of the resulting difficulty in filling available posts.
To overcome these constraints, health labour markets may require regulatory or institutional intervention to achieve socially desirable and economically efficient outcomes. Health labour markets can be regulated through a wide range of interventions: licensing professional occupations, accrediting universities and institutions that offer professional degrees, subsidizing medical education, restricting entry to the market and creating coercive measures (e.g. bonding and compulsory service) to direct health workers to rural and underserved areas.6 The selection of the appropriate balance of regulations and policies requires a solid understanding of the dynamics of the health labour market.
When conducting health labour market analysis, it is also crucial to take into account market structures – i.e. the degree of concentration on the demand and supply sides. The organization of health professionals through institutions such as labour unions or professional associations creates a degree of monopoly in the supply of health labour through collective bargaining. Medical professional associations may play a role beyond that of a union and often take on internal regulation of health workers by setting the requirements for obtaining a licence, defining minimum quality standards and determining the number of workers entering the profession.6 Restricting supply in this way results in higher wages and introduces inflexibilities in the labour market. On the demand side, where potential employers are well organized, for example, in the form of a single or dominant public sector employer, a health worker may have to accept the terms on offer or leave the sector altogether. These conditions can explain why markets fail to “clear”, as the dominant roles of unions, professional associations and public sector employers or single payer employers set conditions that are often driven by political goals rather than market conditions.
Health labour market analysis and better policies
Despite the limited number of studies on the health labour market dynamics in low- and middle-income countries, recent analyses of underlying market scenarios are beginning to reveal the importance of understanding such dynamics and of tailoring policy responses to the unique conditions of each country. To illustrate this point, we now turn to some examples of analyses that provide useful insights into the dynamics of the health labour market.
In a comparative study, Vujicic et al. (2009)7 analysed the implications of government wage bill policies in the Dominican Republic, Kenya, Rwanda and Zambia for the health workforce. All four countries were implementing general government wage bill restriction policies and the study’s purpose was to explore the effects of those policies on the strategy for maintaining or expanding the health workforce. Their research suggests that in Rwanda the health sector wage bill was maintained despite the general wage bill restriction and that the health workforce was successfully expanded in line with the country’s health strategy. In Kenya, on the other hand, the wage bill for the health sector was reduced in line with the general restrictions and this limited expansion of the workforce. In Zambia, the main obstacle to the expansion of the workforce was not deemed to be the wage bill restrictions, but the difficulties in filling budgeted posts. In the Dominican Republic, wage bill restrictions constrained growth in salaries but not in the number of health workers, which resulted in a contraction of the hours worked in the public sector and an increase in the prevalence of dual practice. Health workers responded to their reduced pay by allocating more time to other occupations. These differential responses illustrate how four countries that introduced similar wage bill policies faced different health worker responses because of their very different market situations. Supply responded flexibly to labour market conditions in the Dominican Republic and Rwanda. In the Dominican Republic, the supply of health workers fell in response to declining pay, whereas in Rwanda the supply increased in response to non-wage measures that supported expansion. Despite having a shortage of health workers relative to need, Kenya has a pool of long-term surplus in human resources for health (i.e. unemployed health professionals). The “shortage” is thus generated by inadequate demand – employment opportunities with adequate working conditions. By contrast, in Zambia the health labour market faces a long-term shortage, such that an increase in the demand for health workers did not increase employment, since there was an insufficient pool of unemployed or under-employed workers wanting to take advantage of better conditions.
Another example is that of Malawi, which faced a dire shortage, maldistribution and outmigration of health workers in the early 2000s. Malawi has subsequently succeeded in reversing a negative trend through a combination of demand- and supply-side interventions. A 50% increase in training output for priority cadres was accompanied by a 52% salary top-up to enhance deployment and retention. This led to a significant improvement in health workforce availability, as health worker density rose from 0.87 to 1.44 per 1000 population between 2004 and 2009.8
A widely promoted solution for increasing the availability of human resources for health is to expand training and increase funding for public sector employment. But this requires funds, largely from the public purse. Countries such as Ethiopia and Niger, whose macroeconomic conditions prevented them from implementing this approach, chose to invest in community-based health workers, who undergo shorter training and require less pay. In early experiences, these cadres have played a significant role in improving service coverage and health outcomes in underserved communities.9,10 Similarly, experiences in Mozambique11 and elsewhere show that mid-level cadres respond differently to health labour market conditions and are more easily retained in rural areas than physicians.
These examples highlight why it is important to understand underlying market conditions when introducing human resource policies in the health sector, as summarized below:
- where constraints to supply are most important, policies such as expanding training opportunities may be appropriate;
- where constraints on demand are most important, policies such as increasing the funding available for the health workforce are likely to be appropriate;
- an effective strategy will more often address both supply and demand constraints simultaneously; and
- in some cases, solutions may require structural changes to the labour market, such as the reorganization of the service delivery system and changes in the skills required of health workers (e.g. greater use of mid-level health workers), which in turn will change the labour market dynamics.
Knowledge and research gaps
One reason that labour market analysis has received little attention in the debate surrounding human resources for health is that data in this domain are scarce in low- and middle-income countries. Some critical areas in terms of data and research are highlighted below:
- Central to any labour market analysis is an understanding of the absolute and relative levels of health worker remuneration from all sources. This is a very difficult variable to measure in the health sector and doing so requires considerable effort. Pay of health professionals, especially in low- and middle-income countries, consists of multiple components including salaries, informal payments and bonuses and allowances that can vary considerably among individual health workers. Furthermore, health professionals often hold multiple jobs or generate income outside their primary employment. Availability of more comprehensive data on their pay levels will be essential for understanding the dynamics of the health labour market.
- Technological changes require transformation in the health system and are important determinants of labour market evolution, although little research is available to guide government policies and investments. One important implication for labour markets of the growth of medical technologies over the last decade is the growing demand for highly skilled workers that cost more to produce and employ.12,13
- Few studies have been conducted in low- and middle-income countries to measure the “elasticity” (responsiveness) of the supply of health workers to changes in the wage rate. One study conducted in China suggests that the elasticity of supply may be considerably higher in that country than in high-income countries, and the authors conclude that increasing health worker pay may be a more cost-effective strategy to expand the health workforce than expanding training programmes.14 More empirical studies will be needed to establish whether this is equally applicable to low- and middle-income countries other than China.
- Little research has been conducted on the responsiveness of health worker quality to economic variables. Among the well-known human resource problems in the health sector are low productivity, effort and morale. There is a need to evaluate the impact of changes in salaries, training availability and other working conditions on health worker performance. Some work in this area has already begun.15-18 Such studies will help to generate hypotheses about the impact of pay and institutional variables on health worker effort and will inform the effectiveness and sustainability of pay for performance and other financial and non-financial incentives to elicit more effort and greater productivity from health workers.
- Health worker preferences and responses to market conditions are also beginning to attract some research in the context of health worker recruitment and retention in rural and remote areas. Discrete choice experiments have been conducted in several low- and middle-income countries to elucidate workers’ preferences in terms of job characteristics and assess their willingness to be deployed to remote and rural areas.19,20
We have described how labour market analysis can enhance our understanding of the factors that constrain human resources for health and result in more effective policies and interventions to address these. We have also described the health labour force analytical framework and explained that labour markets are eminently shaped by supply and demand and only indirectly by need. Although any policy conclusions derived from labour market analysis are tentative at this stage because of gaps in data and research, several country experiences point to important challenges in the health labour market and, depending on the country context, such challenges should be tackled by considering market forces from both a supply- and demand-side perspective. The identification of appropriate policy options will require further research and evaluation of effective strategies, as well as a deeper understanding of the underlying labour market conditions affecting the health worker situation in a specific country.
A better understanding of the impact of health policies on health labour markets and, subsequently, on the employment conditions of health workers would be helpful in identifying an effective strategy for the progressive attainment of universal health coverage. The human resource challenges faced by the health sector should therefore be addressed within a country’s broader development framework, where the factors affecting the dynamics of the health labour workforce – from education to regulation, incentives and the fiscal space for the wage bill – can be addressed in a holistic, integrated manner.
The authors are grateful to Christiane Wiskow (International Labour Organization) and the participants of the African Regional Workshop on Health Labour Market Analysis held in Hammamat, Tunisia, in March 2013.
- Anand S, Bärnighausen T. Health workers at the core of the health system: framework and research issues. Health Policy 2012; 105: 185-91 http://dx.doi.org/10.1016/j.healthpol.2011.10.012 pmid: 22154420.
- Vujicic M, Zurn P. The dynamics of the health labour market. Int J Health Plann Manage 2006; 21: 101-15 http://dx.doi.org/10.1002/hpm.834 pmid: 16846103.
- Hongoro C, McPake B. How to bridge the gap in human resources for health. Lancet 2004; 364: 1451-6 http://dx.doi.org/10.1016/S0140-6736(04)17229-2 pmid: 15488222.
- Fields G, Andalón M. A toolkit for analyzing labor markets in the health care sector in Africa: health, nutrition and population, AFTHD Africa Region. Washington: The World Bank; 2008.
- Soucat A, Scheffler R, Ghebreyesus TA, editors. The labor market for health workers in Africa: a new look at the crisis. Washington: The World Bank; 2013.
- Nicholson S, Propper C. Medical workforce. In: Pauly MV, McGuire TG and Barros PP, eds. Handbook of Health Economics 2. Elsevier; 2012. pp. 873-925.
- Vujicic M, Ohiri K, Sparkes S. Working in health: financing and managing the public sector health workforce. Washington: The World Bank; 2009.
- Management Sciences for Health. Evaluation of Malawi’s emergency human resources programme. 2010. Available from: http://www.msh.org/news-bureau/msh-publishes-evaluation-of-malawi-human-resource-program.cfm [accessed 7 July 2013].
- Wakabi W. Extension workers drive Ethiopia’s primary health care. Lancet 2008; 372: 880 http://dx.doi.org/10.1016/S0140-6736(08)61381-1 pmid: 18795419.
- Amouzou A, Habi O, Bensaïd K, Niger Countdown Case Study Working Group. Reduction in child mortality in Niger: a Countdown to 2015 country case study. Lancet 2012; 380: 1169-78 http://dx.doi.org/10.1016/S0140-6736(12)61376-2 pmid: 22999428.
- Pereira C, Cumbi A, Malalane R, Vaz F, McCord C, Bacci A, et al., et al. Meeting the need for emergency obstetric care in Mozambique: work performance and histories of medical doctors and assistant medical officers trained for surgery. BJOG 2007; 114: 1530-3 http://dx.doi.org/10.1111/j.1471-0528.2007.01489.x pmid: 17877775.
- Denton FT, Gafni A, Spencer BG. The SHARP way to plan health care services: a description of the system and some illustrative applications in nursing human resource planning. Socioecon Plann Sci 1995; 29: 125-37 http://dx.doi.org/10.1016/0038-0121(95)00004-6.
- Sales CS, Schlaff AL. Reforming medical education: a review and synthesis of five critiques of medical practice. Soc Sci Med 2010; 70: 1665-8 http://dx.doi.org/10.1016/j.socscimed.2010.02.018 pmid: 20359806.
- Qin X, Li L, Hsieh CR. Too few doctors or too low wages? Labor supply of health care professionals in China. China Econ Rev 2013; 24: 150-64 http://dx.doi.org/10.1016/j.chieco.2012.12.002.
- Das J, Hammer J. Money for nothing: the dire straits of medical practice in Delhi, India. Washington: The World Bank; 2005 (World Bank Policy Research Working Paper 3669).
- Das J, Sohnesen TP. Variations in doctor effort: evidence from Paraguay. Health Aff 2007; 26: 324-37 http://dx.doi.org/10.1377/hlthaff.26.3.w324.
- Leonard KL, Masatu MC. Using the Hawthorne effect to examine the gap between a doctor's best possible practice and actual performance. J Dev Econ 2010; 93: 226-34 http://dx.doi.org/10.1016/j.jdeveco.2009.11.001.
- Maestad O, Torsvik G, Aakvik A. Overworked? On the relationship between workload and health worker performance. J Health Econ 2010; 29: 686-98 http://dx.doi.org/10.1016/j.jhealeco.2010.05.006 pmid: 20633940.
- Ryan M, Kolstad J, Rockers P. How to conduct a discrete choice experiment for health workforce recruitment and retention in remote and rural areas: a user guide with case studies. Geneva: World Health Organization; 2012. Available from: www.who.int/entity/hrh/resources/DCE_UserGuide_WEB.pdf [accessed 2 July 2013].
- Araújo E, Maeda A. How to recruit and retain health workers in rural and remote areas in developing countries. Washington: The World Bank (World Bank Health, Nutrition and Population Discussion Paper). Forthcoming.