Are transfers from members living abroad a stimulus to a key sector such as schooling in Albania? In the present article I summarize results of a study investigating which factors are able to boost the household expenditure in education. Among these factors, a large emphasis is given to the amount of remittances sent to Albanian families from migrants living abroad. High rates of unemployment and the severe poverty in Albania have induced strong migration pressures. These large migration flows provide Albania with an important source of capital in terms of external remittances. The two main empirical findings are that the household income has a positive and well determined impact on education expenditure, whereas international transfers do not influence education spending.
Keywords: Migrant Remittances, Engel Curves, Education expenditure
JEL classification: R, J
See other useful links
Suggested citation: Cattaneo, Cristina, Educational Expenditure and Remittances: Is there a Link? (February 27, 2014). Review of Environment, Energy and Economics (Re3), http://dx.doi.org/10.7711/feemre3.2014.02.002
“Education is the most powerful weapon which you can use to change the world” (Nelson Mandela). Education is a right, like the right to receive adequate food or adequate housing and shelter. Education is also a means to human development, by expanding opportunities and freedoms. Peace, democracy and economic growth are fostered through education. Education contributes to health improvements and poverty reduction. The objective of the present analysis is to detect which factors are able to boost the investment in education in Albania. In particular, the analysis aims to investigate the link between international remittances and education. Are transfers from members living abroad able to stimulate a key sector such as schooling?
High rates of unemployment and the severe poverty experienced by the households have induced strong migration pressures in Albania. Albanians, among all other transition populations, are the most inclined to leave their country. Although statistics on migration are of poor quality, partly due to the irregular nature of much of the migration, rough estimates suggest that 40 per cent of people have some relatives settled outside the borders of the country (UN, 2002). Recently, the government of Albania published figures that quantify the number living abroad as over one million in 2005, representing 30 per cent of the total population. These large migration flows provide Albania with an important source of capital in terms of external remittances. In 2007 workers’ official transfers represented 27 per cent of national GDP and exhibited a stable increase over time (WDI, 2008). Despite the impressive size of the remittance flows, little is known about the use of these transfers by Albanian families. In particular, it is still a matter of speculation whether this money is channelled towards investments in human and physical capital or is spent on consumer goods.
Economists are often sceptical regarding the capacity of remittances to sustain economic development and are equally critical of the use of remittances for consumption, with no funding left for saving and investment. For example, an empirical analysis conducted in Mexico reports a negative effect of migration on educational attainment (McKenzie and Rapoport, 2006). The negative effect may arise because of parental absence which, due to migration, can reduce parental inputs into education acquisition, or it may induce the children to meet eventual labour and cash shortages through working and household activities. These effects more than compensate for the positive effect of remittances, which relieve family credit constraints. In El Salvador, remittances do not significantly enhance investment in human capital (Acosta, 2006).
The majority of studies, however, report a positive impact of remittances on educational expenditures. For example, Adams (2005), comparing the expenditure behaviour of migrant and non-migrant households in Guatemala, reports that households receiving remittances spend much more on secondary education than do non-migrant families. Cox-Edwards and Ureta (2003) find that remittances in El Salvador reduce hazard of dropping out of school. Lopez et al. (2007) find for 6 out of 11 Latin American countries a positive effect of migrant transfers on education attainment. López-Córdova (2005) analyses the effect of remittances on school attendance among Mexican children aged between 6 and 14 years old. He reports that a small increase in the proportion of recipient households is responsible for a moderate increase in school attendance. Calero et al (2009) investigate the effect of remittances on the probability of being enrolled in school in Ecuador. They find a positive effect of remittances on school enrolment especially for girls, and children living in rural areas of Ecuador.
2. The Albanian context
During the communist regime the access to basic schooling in Albania was free and universal. Enrolment rates in basic and secondary education were generally higher than those in other planned economies in the region (World Bank, 2001). However, since the transition, gross enrolment rates declined in almost all levels of education. Not only the enrolment rates sharply declined, but Albania has at present the lowest educational attainment compared to most transition countries (World Bank, 2005). A dramatic decline was associated in particular with the vocational enrolment rate. The transition to a market economy was generally associated with increasing uncertainty about future jobs. This widened the relative risk of return for a highly specialized education, such as vocational degrees and dampened the expected return to vocational education. The perceived benefits of education, evaluated in terms of earnings or quality of jobs were low, and this undermined the long-term incentives for families to invest in education. Moreover, after transition, labour markets offered rising income opportunities, which increased the opportunity costs of staying at school.
In this analysis I capture education by measuring the total household expenditure related to pre-school and higher education. In Albania, more than 98 per cent of students go to public schools, and therefore the variability of education spending is related to the quantity of education demanded by the household, above the compulsory level. Given the limited enrolment in private schools, the quality of education should not affect the education expenses. The total amount of education costs is primarily made by textbooks and schooling materials, such as pens and exercise books. Other items that enter the education expenditure are gifts to teachers, tutoring, preschool, transport, uniform and lodging. The fees for private schools represent only a marginal component of the total expenses.
Figure 1 - Education expenditure and income transfers for increasing income quintiles
Not all households report some spending in education. Overall only 62 per cent of total households report positive spending on this item. There is thus a substantial censoring in the variable. The censoring is more relevant for households lying at the bottom of the income distribution and it decreases for increasing wealth, moving from the first to the fifth quintile (Figure 1). The higher the welfare, the higher the likelihood that households report a positive spending in this item. Wealth not only influences the incidence of the censoring but also the amount of education expenditure. At higher welfare, the spending on education rises. These last findings emphasize the critical role of the household budget constraint in determining education expenditure functions.
A second important measure for this analysis is the amount of remittances that the households receive from members migrated abroad. Overall, 18 per cent of the households receive transfers from abroad (Figure 1). The interesting feature is that the percentage of the recipient families decreases for increasing welfare, suggesting that in Albania private transfers represent a substantial aid to poorer households. The greater proportion of recipient households is found among the poorest families, with a percentage of 20.3. In contrast, among the richer families, only 15.2 per cent receive remittances. There is no clear link between the amount of transfers and the welfare position of the households, though richer families seem to receive larger external transfers than poorer ones. This may be due to the fact that richer households have better qualified migrants in terms of human capital and therefore these movers earn more and are able to send larger remittances. This result is also consistent with better-off families having a larger number of movers.
3. The economic analysis
Using appropriate econometric techniques, the determinants of the expenditure for education-related items are estimated, including the amount of remittances received from members migrated abroad. The empirical model reveals that the “general” total household income has a positive effect on education spending, highlighting the important role that the budget constraint plays in a family’s schooling decision. The estimated model indicates that a 1 per cent increase in the monthly household income raises monthly expenditure in education by 0.1 per cent, on average and ceteris paribus. The number of young siblings, aged between 6 and 18, exerts a positive effect on education expenditure. Male headed households do not show a differentiated consumption pattern compared to female headed families. The education of the household head is extremely important in shaping the purchases of schooling goods and the effect is more pronounced at higher levels of the head’s education. This result is consistent with existing empirical findings, which highlight the critical role of parental education on children’s schooling. Urban areas provide easier access to school compared to rural ones.
Finally, the transfer variable exerts a non-significant impact on education expenditure. This finding implies first, that incomes from different sources are not pooled together within the household. The effect of remittances in fact, statistically differs from the effect of “general” household income. This interpretation is consistent with the prediction of the so called non-unitary models. This model asserts that some family members have distinct preferences and distinct bargaining power relative to other members, and therefore the resource allocation is not determined by the household as a unique entity. The ownership of the income may affect the pattern of its use. The interests of the movers sending remittances differ from the interest of the head of the household, and this yields a null effect of remittances on educational expenditures.
But why don't transfers influence education spending? One explanation could be that migrants impose conditions on the type of spending of remittances and this use does not include education. The transfers sent home by migrants are typically used to purchase food and basic necessities, for medical expenses, investment in construction and improvement of the quality of dwellings. A second interpretation could be that movers are the more able and the more educated siblings. Therefore, the ones left behind are the less able and thus unlikely to pursue a pathway that involves education and hence do not require remittance transfers for this particular purpose. Third, the returns to education in foreign markets, such as Italy and Greece, may be limited, and siblings in migrant families may find out-migration to be a better investment than education.
A fourth motivation can be gained by an inspection of the schooling situation in Albania. The country displays the lowest educational attainments compared to most transition countries. During the transition, enrolment rates in education declined in Albania, and this effect is eventually linked to a perceived low quality of education provision. The closure of most pre-schools and secondary vocational schools, inadequate public spending, deterioration of school infrastructure as well as lack of value placed on education are all factors responsible for the reduction in school quality (World Bank, 2005). The poor quality is also due to inefficiency in the allocation of resources available to schools, with high disparities in the distribution of teaching materials across schools, and to the low academic qualifications of a large proportion of the teachers (IMF, 2006). This situation influences the perceived benefits of education, in that the bad quality may lead to low returns to education and bad employment opportunities, undermining the incentive to invest in education in the first place. Therefore, migrant families allocate remittances to alternative investments, ensuring higher return than education.
A final exercise is conducted. The effect of remittances on education spending may not be constant but it may change as one moves along the various quintiles of the distribution of the education expenditure. In other words, this second exercise tests whether the households with low educational spending receive greater benefit from migrant remittances than households with high educational spending, as the former are likely, ceteris paribus, to be the poorest households and the most constrained in their budgets. Moreover, I test if families, which allocate few resources to education, respond more to changes in total income than families which spend large amounts of money for education.
The empirical findings are as follows. The effect of income on education spending is fairly constant across all points of the distribution. Income has the same influence on education expenditure, no matter if the family spends little or large resources in education items. Interestingly, only the households in the bottom quintile of the education distribution benefit from the international transfers. Remittances in fact positively influence education expenditure, but only among families which spend little in education. Additional interesting results are found. First, the effect of the education of the household head is magnified moving from the lower to the upper quintiles of the educational distribution. This finding suggests that the education background of the head of the family is a major determinant of increasing spending in education. Second, the number of adults in the family increases educational expenses, but only among those families spending large resources in education. Third, the number of children aged between 6 and 18 years old increases education expenditure only for low education spending.
The primary objective of this work is to identify which factors are able to boost the investment in education at household level in Albania. In particular, the analysis aims to investigate the relationship between international remittances and education. Some interesting results were obtained.
First of all, household income has a positive and well determined impact on education expenditure, confirming the critical role of family budget constraints in explaining purchasing behaviour. A second important result is that the international transfers do not influence education spending. This finding may suggest that movers have made conditions upon the use of remittances and the selected allocation does not include the education of siblings. It may be the case, for example, that the interest of movers does not match with the interest of the head of the household and this conflict produces the null effect of remittances on education expenditure. Moreover, the perceived rewards for education in the local market are low due to the poor quality of education provision, and this creates limited incentives for the investment in schooling. Low returns to education features as well in the foreign markets, such as in Italy and Greece. The income accruing from remittances may be channelled into alternative investments, such as purchases of new houses or quality enhancement of existing ones. The empirical findings are consistent with the predictions of the non-unitary models, as far as incomes from different sources, namely “general” household income and private transfer income, exert distinct effects on consumption. Finally, the education of the household head seems to be more relevant for families which spend large resources in education, whereas international transfers increase education consumption only among families which spend little in education.
Community level characteristics play a crucial role in shaping the effectiveness of migration. If migrant sending countries lack basic infrastructure, it is unlikely that remittances alone can produce any investment enhancing mechanism. Remittances are likely to boost economic development in those environments where migrants do not have to serve simultaneously as workers, savers, investors as well as producers. For this reason, improvement in school quality should be a primary target for Albania, so that migrant families can anticipate the future returns to education and higher incentives to invest in education are produced. If improvements in infrastructure and in service delivery are achieved in this field, remittances can eventually finance the increased demand in schooling.
Acosta P. (2006).'Labor Supply, School Attendance and Remittances from International Migration: The case of El Salvador', World Bank Policy Research Paper 3903, The World Bank, Washington D.C.
Adams R.H. (2005). 'Remittances, Poverty and Investment in Guatemala', in Ozden C. and M. Schiff (eds.), International Migration, Remittances and the Brain Drain, World Bank and Palgrave Mcmillan
Calero C., A. S. Bedi and R. Sparrow (2009). 'Remittances, Liquidity Constraints and Human Capital Investments in Ecuador', World Development, 37(6), pp. 1143-1154
Cox-Edwards A. and M Ureta (2003). 'International Migration, Remittances and Schooling: Evidence from El Salvador', Journal of Development Economics, 72(2), pp. 429–461
IMF (2006). Albania: Poverty Reduction Strategy paper – Annual Progress Report. IMF Country Report n. 06/23, International Monetary Fund, Washington D.C.
Lopez J. H., P. Fajnzylber and P. Acosta (2007). 'The impact of remittances on poverty and human capital: evidence from Latin American household surveys', Policy Research Working Paper Series 4247, The World Bank
López-Córdova E. (2005). 'Globalization, Migration and Development: The Role of Mexican Migrant Remittances', Economia, 6(1), pp. 217 -256
McKenzie D. and H. Rapoport (2006). 'Can migration reduce education attainment? Evidence from Mexico', World Bank Policy Research Working Paper 3952, The World Bank, Washington D.C
UN (2002). Common Country Assessment Albania, Albanian Centre for Economic Research (ACER)
World Bank (2001). Financing, Efficiency, and Equity in Albania Education, World Bank Technical paper report n. 512, The World Bank, Washington DC
World Bank (2005).Poverty and education in Albania: who benefits from public spending?, Human Development Sector Unit, Report n.31983-AL, The World Bank, Washington DC
World Bank (2008). World Development Indicators 2008, The World Bank, Washington D.C
Cattaneo, C., "Migrants’ international transfers and educational expenditure", Economics of Transition, Volume 20, Issue 1, pages 163–193, January 2012
FEEM Nota di Lavoro 1.2010