Selected Issues

Selected Issues

## Impact of Emigration1

Emigration in search of better economic opportunities has been a widespread phenomenon in many developing economies. The reduction in labor supply, especially those of skilled workers, is a cost that source countries incur, while remittances are a benefit. This note examines the costs and benefits of emigration using a simple demand-supply model. The analysis uses a sample comprising 31 countries in Central America, South America and the Caribbean.

### A. Methodology

1. Emigration rate. The emigration rate2 in country i at time t by level of education e can be defined as –

${m}_{e,t}^{i}=\frac{{M}_{e,t}^{i}}{{M}_{e,t}^{i}+{R}_{e,t}^{i}}$

where ${M}_{e,t}^{i}$ is the stock of emigrants and ${R}_{e,t}^{i}$ are the residents in source country of emigrants.

2. Theoretical framework. The theoretical framework for this exercise is based on Mishra (2007a). Based on a simple model of labor demand and supply, countries tend to experience emigration loss. This can be understood as the net effect of a rise in wage rate in the source country due to reduction in labor supply, which entails a gain for the workers who stay behind, and a loss to the owners of other factors.

3. Emigration loss. This emigration loss can be expressed as –

$\begin{array}{cc}EL=\frac{1}{2}{S}_{L}e{m}^{2}& \left(1\right)\end{array}$

where e is the magnitude of elasticity of factor price of labor (i.e., the percentage change in wage resulting from a 1 percent change in the size of the labor force), m is the overall emigration rate, and SL is the share of labor in GDP.

4. High-skilled emigration loss. Emigration loss when only high-skilled labor moves is expressed as –

$\begin{array}{cc}E{L}_{S}=\frac{1}{2}{S}_{S}{e}_{S}{m}_{S}^{2}& \left(2\right)\end{array}$

where Ss is the skilled labors’ share in national income, es is the elasticity of factor price of skilled labor (i.e. the percentage change in wage of skilled labor resulting from a 1 percent change in the size of the labor force), and mS is the emigration rate for skilled labor force.

5. Emigration of high-skilled labor. Emigration of high-skilled labor (educational attainment higher than high-school leaving certificate or equivalent) is associated with a decrease in the productivity of the workers who have stayed behind, thereby giving rise to a negative externality and augmenting the loss.

6. Augmented emigration loss. Augmented emigration loss is expressed as –

$\begin{array}{cc}AE{L}_{S}=\frac{1}{2}{S}_{S}{e}_{S}{m}_{S}^{2}+\frac{\gamma {S}_{S}{m}_{S}\left(1-{S}_{S}{m}_{S}\right)}{\left(1-\gamma \right)}+\frac{\gamma {S}_{U}{m}_{S}\left(1-{S}_{U}{m}_{S}\right)}{\left(1-\gamma \right)}& \left(3\right)\end{array}$

where SU is unskilled labors’ share in national income, and γ is percentage change in marginal product of skilled and unskilled labor due to a 1 percent change in the stock of skilled labor.

7. Negative externality from emigration of high-skilled labor. The 2nd and 3rd terms in the above equation, respectively, capture this negative externality – the effect on the marginal product of skilled and unskilled labor who have stayed behind in the source country, arising from emigration of high-skilled labor.

8. Education expenditure. The expenditure incurred by government on the primary, secondary, and tertiary education especially of high-skilled emigrants represents an additional loss to the non-migrants. It likely manifests itself in lower provision of public goods or higher tax rates.

9. Education expenditure for high-skilled emigrants. Annual expenditure on education of high-skilled migrants (as a ratio of GDP), is given by –

$\begin{array}{cc}{E}^{m}={c}_{S}*{M}_{S}& \left(4\right)\end{array}$

where cs is the cost of educating each migrant, and Ms is number of such migrants3.

10. Measuring education expenditure. cs can be measured by dividing the government expenditure (percent of GDP) on primary, secondary, and tertiary education, by the enrollment in public institutions at the respective level.4 Now, for a high-skilled person who emigrated in 2000, the government would have financed their primary, secondary, and tertiary education roughly in 1990, 1995, and 2000 respectively. But this data on public expenditure is not available for each year for any country in the sample. As an approximation, expenditure on primary, secondary and tertiary education has been averaged over 1990 to 2005, 1995 to 2005 and 2000 to 2010, respectively, to capture the cost of educating the average number of migrants over 2000 to 2010.5 Even after relying on the above simplification, public education expenditure can be found for only 19 countries.6

### B. Data and Assumptions

11. Country-wise emigration and remittances. For some countries, notably those in the Caribbean, overall emigration rate is around 40 percent and emigration rate of high—skilled labor is around 80 percent. El Salvador exhibits higher emigration rates compared to most of its Central American neighbors.7 The personal remittances8 (percent GDP) it receives are among the highest in the region.

12. Values of parameters to compute loss. The variation in emigration loss across countries arises from variation in the emigration rate; the values of all other parameters are held fixed. The following have been used to arrive at alternative sets of values for the emigration loss -

• The elasticity of factor price of labor e as 0.3 or 0.4, based on Mishra (2007b);9

• The elasticity of factor price of skilled labor es as 0.3 or 0.4, assuming the same value as above;

• The elasticity of marginal product of skilled and unskilled labor γ as 0.05 or 0.1, based on Mishra (2007a);

• The share of labor in national income SL as 0.69,10 based on Guerriero (2012);

• The share of skilled labor in national income Ss as 0.39,11 assuming skilled labor constitutes the top 20 percent earners in the income distribution.

### C. Results

13. Emigration loss. Table 1 shows the simple emigration loss using equation (1); Dominica and Grenada show the highest loss. For 29 out of 31 countries, remittances exceed the loss.

Table 1.

El Salvador: Emigration Loss vs. Remittances

(In percent of GDP)

Sources: World Bank and author’s calculations.Notes:1. Emigration loss here is calculated using equation (1);2. Bahamas has been excluded as data on remittances is missing.

14. High-skilled emigration loss. Table 2 shows that when only high-skilled emigration is considered using equation (2). The emigration loss is higher than the one reported in Table 1, owing to higher emigration rate of skilled labor compared to overall emigration. Guyana and Barbados show the highest loss. Remittances exceed the loss in 26 countries when wages for skilled labor are less elastic (es = 0.3), and in 24 countries when wages for skilled labor are more elastic (es = 0.4). For El Salvador, the loss is higher than that for Nicaragua, Dominican Republic and Honduras but lower than that for Guatemala.

Table 2.

El Salvador: High-Skilled Emigration Loss vs. Remittances

(In percent of GDP)

Sources: World Bank and author’s calculations.Notes:1. Emigration loss here is calculated using equation (2);2. Bahamas has been excluded as data on remittances is missing.

15. Augmented emigration loss. Table 3 illustrates the emigration loss augmented for “external effects”, using equation (3). For only 16 to 17 countries (depending on the value of γ), the remittances exceed the augmented emigration loss. The drivers of net gain or loss vary across countries. In Antigua & Barbuda, Barbados, Dominica, and Trinidad & Tobago – the loss is driven by very high rates (80 percent or more) of emigration of skilled labor, and low remittances. For other countries with comparable emigration rates such as Guyana and Haiti, the remittances are substantially large (13 to 21 percent of GDP) to act as an offset. For El Salvador, Guatemala and Honduras which exhibit similar loss (lower than the Caribbean countries), remittances (8 to 16 percent of GDP) act as an offset. For Colombia, Costa Rica, Mexico, Paraguay and Peru (similar loss) with lower emigration rates, remittances are only marginally higher than the loss.

Table 3.

El Salvador: Augmented Emigration Loss vs. Remittances

(In percent of GPD)

Sources: World Bank and author’s calculations.Notes:1. Emigration loss here is calculated using equation (3);2. * Education expenditure per high-skilled migrants (% GDP) is the average of -% GDP/ enrollment in primary public institutions) 1990–2005 (secondary education expenditure, % GDP/ enrollment in secondary public institutions) 1995–2005 (tertiary education expenditure, % GDP/ enrollment in tertiary public institutions) 2000–2010.

16. Augmented emigration loss and education expenditure. The sunk cost arising from public expenditure on education of high-skilled migrants has also been included in Table 3. For the 19 countries for which education expenditure is available, remittances exceed the sum of augmented loss and education expenditure in 11 to 12 countries (depending on the value of γ). Countries in the Caribbean such as Antigua and Barbuda, Barbados, Guyana and Jamaica spend a high proportion on education especially at the tertiary level (more than 5 percent of GDP). While the remittances received by the latter two tend to offset this spending, that is not the case for the former two. For the 12 countries with missing data on education expenditure, remittances exceed the augmented loss by a magnitude (6 percent of GDP or more) in Dominican Republic, Haiti and Honduras which would likely not be offset even if the education expenditure were to be included in the loss. For El Salvador – remittances (among the highest) outweigh the sum of emigration loss (moderate high-skilled emigration) and education expenditure (lower than the sample average) by the widest margin. Other countries for which remittances outweigh the loss include Nicaragua, Guatemala, Ecuador, and Bolivia.

17. Other recent work on emigration and remittances in Latin America and the Caribbean. The above results are broadly in line with Beaton et al. (2017) who find that the joint effect of emigration and remittances on growth has likely been negative for the Caribbean and South America – the former characterized by very high emigration, but small and possibly positive for Central America which receives much higher remittances.

### D. Conclusion and Caveats

18. Remittances outweigh the sum of augmented emigration loss and education expenditure for nearly half of the 31 countries in this sample. Most of these countries are from Central America. The countries for which remittances are not able to outweigh the above sum are mostly the Caribbean ones.

19. Caveats. Of course, these results need to be qualified by the fact that there are other costs and benefits of emigration which have not been included in this framework, for e.g. the impact on trade, investment and marginal productivity of capital, and some others which are harder to measure. Other characteristics like demography, civil unrest, and natural disasters have also not been captured. The value of elasticity of factor price of skilled labor may be different from that of overall labor. Estimating country-level elasticities would better capture country characteristics, but is subject to the availability of detailed survey data.

## References

• Beaton K., S. Cerovic, M. Galdamez, M. Vaskov, F. Loyola, Z. Koczan, B. Lissovolik, J. Martijn, Y. Ustyugova, and J. Wong, 2017, “Migration and Remittances in Latin America and the Caribbean: Engines of Growth and Macroeconomic Stabilizers?,” Working Paper 17/144 (International Monetary Fund: Washington D.C.)

• Crossref
• Export Citation
• Brücker H., S. Capuano, and A. Marfouk, 2013, “Education, Gender and International Migration: Insights from a Panel Dataset 1980–2010,” Institute for Employment Research.

• Export Citation
• Guerriero, M., 2012, “The Labor Share of Income around the World. Evidence from a Panel Dataset,”

• Karabarbounis, L. and B. Neiman, 2014, “The Global Decline of the Labor Share,” Quarterly Journal of Economics, 129 (1), 61103.

• Mishra, P., 2007a, “Emigration and Brain Drain: Evidence from the Caribbean,” The B.E. Journal of Economic Analysis & Policy, 7, Article 24.

• Mishra, P., 2007b, “Emigration and Wages in Source Countries: Evidence from Mexico,” Journal of Development Economics, 82, 18099.

Prepared by Nitya Aasaavari.

Emigration rate has been taken from Brücker, Capuano, and Marfouk (2013). They use data from 20 OECD member states on the immigrant population aged 25 years and older by gender, educational level and detailed country of birth from 1980 to 2010 (at 5-year intervals).

Number of migrants has also been taken from Brücker, Capuano, and Marfouk (2013).

The data can be found from the UNESCO Institute for Statistics – http://uis.unesco.org/.

Since the relevant data for many countries starts from 2000, primary education expenditure has been averaged over 1990 to 2005 instead of 1990 to 2000, but secondary and tertiary education expenditure over the respective ten-year period.

Another way to measure cs is by using government expenditure (percent of GDP) per student enrolled in primary, secondary, and tertiary levels – found at http://uis.unesco.org/. But this series is available for fewer countries in the sample; for the ones for which it is available, the magnitude of public education expenditure turns out to be similar.

Emigration rate based on simple average is reported here; population weighted emigration rate would be likely higher, driven in part by the rise in emigration rate itself over time.

Personal remittance data from the World Development Indicators (WDI) of the World Bank comprises personal transfers and compensation of employees.

Using census data from Mexico and USA, Mishra (2007b) finds that a 10 percent decrease in the number of Mexican workers due to emigration in a skill group (defined by schooling and experience), increases the

This is the simple average of labor shares in 17 countries reported in Guerriero (2012), calculated as

Share of skilled labor = income share of top 20 percent earners * share of labor in national income; income share of top 20 percent earners is 0.56, an average across 12 LAC countries taken from WDI.