دستمزدها و نگهداری شبکه بهداشت: مدارک و شواهد اصلاحات دستمزد بخش دولتی در غنا

Can governments in developing countries retain skilled health workers by raising public sector wages? We investigate this question using sudden, policy-induced wage variation in which the Government of Ghana restructured the pay scale for health workers employed by the government. We find that a 10% increase in wages decreases annual attrition from the public payroll by 1.0 percentage point (from a mean of 8 percentage points) among 20–35 year-old workers from professions that tend to migrate. As a result, the ten-year survival probability for these health workers increases from 0.43 to 0.49. The effects are concentrated among these young workers, and we do not detect effects for older workers or among categories of workers that do not tend to migrate. Given that Ghana was a major source of skilled health professional migrants during this period and that our attrition measure correlates strongly with aggregate migration, we interpret these results as evidence that wage increases in Ghana improved retention mainly through reducing international migration.

High attrition of skilled employees may generate under-staffing in the public health care systems of developing countries. Wage differentials between domestic public employment and other options are one factor that could be driving doctors, nurses, and other skilled health workers to leave the public health sector, often for jobs in high-income countries. This trend has generated concern that migration will weaken public health systems and lead to poor health outcomes (Chen and Boufford, 2005). 1 More generally, migration of highly skilled individuals generates concern that “brain drain” will lead to lower levels of human capital and economic growth in developing countries.2 The recent literature on the economics of high-skilled migration has debated whether such migration really leads to lower average levels of human capital. Neo-classical trade-theoretic frameworks emphasize the benefits of migration (e.g., Grubel and Scott, 1966), but skilled migration can lead to a net negative “brain drain” in the presence of fiscal externalities, human capital productivity spillovers, or non-neoclassical frictions (Bhagwati and Hamada, 1974 and Miyagiwa, 1991). The most recent literature has changed directions, focusing on how the possibility of migration can induce greater skill acquisition. If this effect is large enough, the opportunity to migrate can actually increase average human capital levels, leading to a “brain gain” (Mountford, 1997, Stark et al., 1997 and Stark et al., 1998). Recent theoretical works have extended this framework and debated its merits (Mountford and Rapoport, 2011 and Schiff, 2005). A large empirical literature has tested whether this “brain gain” effect appears in cross-country macro data (Beine et al., 2001, Beine et al., 2010, Beine et al., 2011, Docquier et al., 2008 and Easterly and Nyarko, 2009), and a few recent contributions test this hypothesis in micro data using natural experiments or instrumental variables (Batista et al., 2012, Chand and Clemens, 2008 and Shrestha, 2011). In general, the literature finds evidence that migration opportunities do provide an incentive to acquire more education. Thus, high-skill migration may be a “drain” or a “gain” for human capital in the sending country, depending on whether the incentive to acquire skills outweighs the direct effect of losing high-skilled migrants. The context, including the particular country and migrant group in question, may determine the desirability of high-skill migration. Beine et al. (2008) simulate the net effect of skilled migration on human capital across various countries. They find that more countries lose than gain and that net brain gain is most likely in countries with low migration rates and low levels of human capital. Bhargava et al. (2011) focus on physicians and find evidence that migration does generate an education incentive. However, this effect is too small to outweigh the direct drain effect for most countries. Similar to the results of Beine et al. (2008) with all high-skill migrants, they find that high physician migration rates are more likely to lead to “brain drain” while lower rates may sustain a “brain gain.” These results from the literature provide a useful background for the present study. We focus on the migration of skilled health workers from Ghana from 2003 to 2009. According to Bhargava et al. (2011), Ghana's physician migration rate of 38% in 2004 was the 12th-highest in the world and the 2nd highest (behind Zimbabwe) among countries with more than 4 million people. Theory and empirical work indicate that positive migration rates can lead to higher levels of average human capital; however, Ghana's extremely high physician migration rate during our sample period indicates that brain drain is more likely to be a valid concern. In such situations, public policy to discourage migration may have a productive role. However, the literature has provided less guidance on what effects should be expected when public policy levers are applied to the migration of skilled health workers. Health workers from developing countries state that wage levels play an important role in migration decisions (Awases et al., 2003 and Kangasniemi et al., 2007). However, only a few studies have attempted to measure how much actual health worker migration decisions respond to wages. As a result, we know little about whether wage increases can be a cost-effective way of reducing migration of skilled health workers. In the present study, we attempt to fill this gap by measuring how elastically the retention of Ghanaian health workers responds to wage changes. Our approach has two main strengths. First, we use detailed administrative payroll data that reports wages and retention for all publicly employed health workers in Ghana from 2003 to 2009. These data allow us to track several thousand health workers over time and measure effects that are representative of a national public health system in which migration rates had reached very high levels. Second, we observe large, plausibly exogenous changes in wages generated by the adoption of a new wage structure. While not randomly assigned, the variation in wages across time, profession (e.g., doctors, nurses, etc.), and seniority generated by this natural experiment allows us to measure the causal effect of wages on migration credibly. To identify the impact of wages we employ a fixed effects strategy, including group effects for workers in a given grade-seniority group as well as common time effects. 3 This approach tests whether the groups of health workers who received the largest raises had their attrition rates fall the most. We relax the common time trends assumption that fixed effects requires in three main ways. First, we allow for time fixed effects that differ across three groups: doctors, nurses, and other health workers. Second, we control for observed individual demographics and concurrent policies affecting migration of health workers both out of Ghana and into the UK. In our preferred specification, we find that wages have a strong negative effect on attrition with a 10% wage increase leading to a 1.03 percentage point decrease in annual attrition. This implies an increase of the 10-year survival rate from 43% to 49%. Third, we check whether wage increases are targeted at particular groups of health workers in a manner that would bias our estimates. We demonstrate that, if anything, large wage increases are targeted toward groups of workers with higher than average attrition trends. As a result our (negative) point estimates will be biased toward zero, if anything. Controlling for linear time trends that are specific to each grade-seniority group generates similar, though noisy, results. We find evidence that the effect of wages on attrition is concentrated among early-career workers with no effects on older health workers. We also find evidence that doctors respond more strongly to wage increases than do other health workers. There is some evidence that men also respond to wage increases more elastically than do women, though this result is not particularly robust. Finally, we demonstrate that the effect we measure is concentrated among workers in occupations that tend to migrate and is not apparent for other workers. We take this as further evidence that wages affect attrition mainly through reducing migration. Our results most directly relate to two strands of the existing literature. First, Bhargava and Docquier (2008) investigate the determinants of physician migration with cross-country panel data and find that migration rates are negatively associated with physician wages. Also, Okeke (2009) finds that physician migration rates respond to changes in GDP brought about by rainfall shocks. The present study is best interpreted as complementary to this existing work. Cross-country empirical studies confirm that wages and growth are correlated with physician migration across a large span of countries and time, though credibly measuring causal effects can be difficult. By focusing on one particular instance in which detailed micro data and plausibly exogenous variation in wages are available, we are able to confirm that the cross-country correlation between physician migration and wages has causal content in this instance. As noted above, we do so in a context where high migration rates make measuring the effectiveness of such a policy particularly useful. Our estimates indicate that raising wages can lead to a large reduction in health worker attrition during times of heavy migration. Second, our results relate to the much broader literature on the role of wages in determining migration decisions for all individuals, not just doctors. Classic theory indicates that low home wages “push” migrants abroad (Borjas, 1987). However, higher income may also relax credit constraints, leading to higher migration rates (Lopez and Schiff, 1998), and increased migration has been observed in response to rainfall-induced income shocks (Yang and Choi, 2007) and randomized Progresa transfers (Angelucci, 2004). The empirical literature is large, making a full review impossible,4 but the present study is most similar to a pair of studies that use micro data to more credibly measure the effect of compensation on migration decisions. Yang (2006) uses variation in the real wage differential between home and destination generated by exchange rate volatility to measure the effect of real wages on return migration of temporary workers from the Philippines. He finds evidence that migrants respond to the classic “push” and “pull” effects of wages, rather than aiming for investment targets. Gibson and McKenzie (2011a) narrowly focus on the “best and brightest” from three Pacific nations to ease concerns about bias from positive selection. They argue that other factors (e.g., risk aversion, family considerations) matter more than income for the migration decisions of these individuals. We add to this literature by examining a new and interesting population: highly skilled health workers from Ghana. In our sample, low home wages do appear to generate a strong “push” effect. These results run counter to those of Gibson and McKenzie (2011a), demonstrating that wage considerations can be important not just for the return migration of low-skilled individuals (e.g., Yang, 2006) but also for highly-skilled first-time migrants. The remainder of the paper is as follows: Section 2 describes Ghana's health sector and the 2006 wage reforms; Section 3 provides a short discussion of the theory; Section 4 describes our identification strategy; 5 and 6 describe the data and the results. Section 7 tests the robustness of our empirical strategy to group-specific time trends and errors in linking wages across the pre-reform and post-reform periods. Section 8 concludes.

This paper measures the impact of home wages on attrition of skilled health professionals from the public sector in Ghana by exploiting variation in wages caused by a policy-induced natural experiment. We find that a 10% wage increase reduces the annual attrition rate by about 1.02 percentage points. This corresponds to a six percentage point increase in the 10 year survival probability of a typical worker. The effect is concentrated solely among young workers age 20–35 who come from professions that tend to migrate. We take this as evidence that the effect of wages on attrition from the public sector mainly runs through migration. While we do not have truly random variation in wages available, our use of sudden, policy-generated variation in wages allows us to plausibly estimate a causal effect of wages on health worker attrition using micro-data. Additionally, the relationship between the wage changes and pre-existing trends in attrition indicate that any bias resulting from endogenous choice of wages by policymakers is likely toward zero. To our knowledge, estimating the causal effect of wages on health worker migration using detailed micro data and a sudden policy change has not been possible before. These results support the most basic economic models of migration in which individuals choose to migrate based on wage differentials between home and foreign countries. These results run counter to expectations that the impact of marginal wage changes may be small when wage gaps are very large or that higher home wages might relax credit constraints causing more migration. Context is important for interpreting our results. The literature studying whether migration levels respond to wages (or GDP) at the macro level has been mixed (e.g., Docquier et al., 2007 and Mayda, 2010). Likewise, micro-level studies that pay close attention to measuring causal effects find different results for different populations. Relatively low-skilled, temporary migrants from the Philippines respond elastically to real relative wage changes in a classic “push-pull” manner (Yang, 2006) while the “best and the brightest” from other Pacific island nations may place a greater weight on other factors (Gibson and McKenzie, 2011a). Our unique data and the policy change we observe allow us to measure the effect of home wages for the entire staff of a national public health system. In this sample of individuals who tend to migrate permanently and who are highly-skilled, we find confirmation of the classic “push-pull” effects of wages. The population in which we measure this push effect of home wages is of particular interest. Migration of skilled health professionals away from developing countries has been widely debated. To many, it is an unfortunate “brain drain” of needed health workers from places where skilled health professionals are already scarce (e.g., Chen and Boufford, 2005). As a result, many steps have been taken toward reducing such migration. For example, the UK National Health Service has voluntarily imposed restrictions on foreign recruitment via its Code of Conduct, and this idea has been taken up recently by the World Health Assembly. Others question whether medical migration leads to a “brain drain” or a “brain gain” in the first place (Clemens, 2007) and whether restrictions on migration violate human rights of migrants, ignore more important problems of health worker performance and urban–rural distribution, or are driven by recouping misguided education subsidies (Clemens, 2009). Of course, settling this debate is beyond the scope of this article. The existing literature (Beine et al., 2008 and Bhargava et al., 2011) does indicate, though, that unusually high health worker migration rates are more likely than most instances of high-skilled migration to lead to “brain drain.” That makes the situation facing the public health system of Ghana in the early 2000's particularly useful for study. In this context, our results can provide useful policy guidance. If policymakers operating public health systems in developing countries face unusually high migration rates, then they may wish to reduce this trend. We demonstrate that health workers can be retained, not just by restrictions on leaving but also by rewards for staying; salary supplements are one effective option. More generally, our results indicate that migration of health workers should be an important consideration as policymakers in developing countries contemplate public sector wage reforms. Finally, if policymakers in developed countries desire to redistribute health workers, subsidizing health worker salaries in sending countries is one means to this end that does not involve restrictions on the movement of individuals. Of course, some caveats are necessary in interpreting the results. We trade external validity for internal validity by narrowing our focus. Our use of attrition from public payroll rather than an explicit measure of migration creates uncertainty about whether we are measuring other forms of attrition. We cannot account for all factors of the necessary cost-benefit calculus, including the reality that raising wages for health workers can sometimes lead to calls for wage increases in other sectors of the public payroll. Nonetheless, we make progress toward evaluating the effectiveness of one particular means toward limiting the migration of skilled health workers. We find that wage increases can result in large reductions in the attrition of skilled health workers. If policymakers judge that an instance of medical migration is harmful, wage increases may be an effective policy tool.