Chapter 4. What impacts does migration have on development in Armenia?

Armenia has one of the highest, and increasing, emigration rates in the world, with about 30% of the population living outside the country. In parallel, Armenia also benefits from significant and increasing levels of remittances. This chapter asks to what extent these trends are contributing to the country’s development in four sectors: the labour market, agriculture, education, and investment and financial services. Drawing on the IPPMD surveys and data analysis, the chapter finds both positive and negative effects on development.

  

Armenia’s rather turbulent recent history has meant the country has one of the highest emigration rates in the world, with about 30% of the population living outside the country. As well as witnessing a marked increase in the number of emigrants, Armenia benefits from significant and increasing remittances from migration. As discussed in Chapter 2, migration – historically and today – is a significant driving force for development in Armenia.

This chapter asks how these migration trends are affecting Armenia’s development in four policy sectors: the labour market, agriculture, education, and investment and financial services. For each sector the chapter presents the findings of the IPPMD surveys and data analysis to explore the impact of three dimensions of migration: emigration, remittances and return migration. The next chapter explores key policies in each of the focus sectors and their links to migration outcomes.

Migration and the labour market

The limited employment opportunities and the lower wages in Armenia compared to emigrants’ main destination countries are major push factors for Armenians to migrate. How does this migration affect the labour market? There are several possible avenues: the remittances sent home to migrants’ families might be spent on setting up a business, which can generate employment. On the other hand, receiving remittances can increase households’ income, reducing the need for household members to be in work. This highlights a moral hazard effect of remittances – household members can become remittance-dependant, causing them to leave their jobs or stop looking for work. This section explores these effects for Armenia.

According to data from the National Statistical Service of the Republic of Armenia, Armenia’s labour force participation rate1 was 63% in 2015: 73% for men and 54% for women (ArmStat, 2015). The rate is notably higher in rural (69%) than in urban areas (58%). Similarly, the employment rate is significantly higher in rural than urban areas (65% versus 43%) and among men than women (60% versus 44%).

The overall unemployment rate in the country is 19% (ArmStat, 2015). There are significant differences in unemployment rates by geographical location, with 7% of unemployment in rural areas, compared to 27% in urban areas. Unemployment is highest among young people. From a rate of 26% and 35% respectively for 15-19 year-olds and 20-24 year-olds, it falls to 24% and 22% respectively for 25-29 year-olds and 30-34 year-olds and then further still to 18% for 35-39 year-olds. It is 4% among 65-69 year-olds.

Services are the largest sector in terms of employment, accounting for more than 40% of all Armenian who are working. Agriculture is the second largest accounting for about 37% in total employment, but unlike services, its share is declining. Industry’s (excluding construction) share in employment increases by 0.6 percentage points between 2008 and 2016, accounting for about 11% of the employed workforce while construction’s share decreased by 5% and the sector plays a small and declining role, employing fewer than 9% of workers in 2008 to just over 4% in 2015 (ArmStat, 2016).

This national pattern is reflected in the IPPMD survey data, where the labour force participation rate among the sampled working age population (15-64 age range) is 57%: 72% for men and 45% for women. Unlike the national figures however, the rate is higher in urban (59%) than in rural areas (55%). The employment rate is 45%: 57% among men and 35% among women, and is higher in rural areas (47%) than in urban areas (43%). Around 43% of the working age population report not being engaged in paid employment and not looking for work. The difference between men and women is notable, with a significantly higher share of women (55%) being economically inactive than men (28%). The rate is higher among all individuals aged 15 and over (60%), taking the retired into account.

Remittances reduce the supply of labour

It is important to look at emigrants’ characteristics to understand the impact of emigration on the labour market. Almost all current emigrants in the survey are of working age (15 to 64). Only about 44% of the emigrants were employed in Armenia before leaving the country and 27% were unemployed which is a higher share than the average among the IPPMD partner countries. Their unemployment rate has significantly decreased since they emigrated (to 4%), implying that unemployment is one of the main reasons people leave the country.

Which sectors and occupational groups are losing most labour to emigration? The left-hand chart in Figure 4.1 displays the share of emigrants in each skills group. This shows that emigrants from Armenia are more likely to come from the lowest skilled occupational group (Level 1). The right-hand chart in Figure 4.1 compares the share of emigrants lost to the agriculture, construction, health and education sectors, revealing that agriculture and construction seem to be the most affected by emigration.

Figure 4.1. Agriculture and construction and low-skilled occupations lose most workers to emigration
picture

Note: The skills level of occupations has been categorised using the International Standard Classification of Occupations (ISCO) provided by the International Labour Organization (ILO, 2012). Skills level 1: occupations which involve simple and routine physical or manual tasks (includes elementary occupations and some armed forces occupations). Skills level 2: clerical support workers; services and sales workers; skilled agricultural, forestry and fishery workers; craft and related trade workers; plan and machine operators and assemblers. Skills level 3: technicians and associate professionals and hospitality, retail and other services managers. Skills level 4: Other types of managers and professionals.

Source: Authors’ own work based on IPPMD data.

What does this mean for households that are losing their productive labour to emigration? The effects are complicated and depend on whether the emigrant had been employed before leaving and whether he or she then sends home remittances. The literature from various contexts and parts of the world shows that receiving remittances can reduce household members’ need to work (Acosta, 2007; Amuedo-Dorantes and Pozo, 2006; Funkhouser, 2006; Kim, 2007; Osaki, 2003).

Although this complex picture makes it challenging to isolate individual effects, the IPPMD data do shed some light on this matter. Figure 4.2 compares the average share of working household members in non-migrant households, emigrant households not receiving remittances and those that are receiving remittances. The graph shows that remittance-receiving households have the lowest share of working adults, suggesting a link between receiving international remittances and the need to seek work. It also appears that women in households with emigrants but not receiving remittances are more likely to work than those in households without emigrants. Given that 77% of current emigrants in the sample are men, this implies that the women left behind may have to compensate with their labour especially if they do not receive remittances.

Figure 4.2. Households receiving remittances have fewer working members
Share of household members aged 15-64 who are working (%)
picture

Note: The sample excludes households with return migrants only and immigrants only.

Source: Authors’ own work based on IPPMD data.

What does regression analysis tell us about this relationship?2 The analysis in Box 4.1 confirms that households reduce labour supply when they receive remittances (Table 4.1). This effect seems to hold for both men and woman in rural areas. Having an absent member in the household does not seem to affect the labour decision of households.

Box 4.1. The links between migration and employment

To investigate the link between migration and households’ labour decisions, the following regression models were used:

picture (1)

picture (2)

picture (3)

where picture signifies households’ labour supply, measured as the share of household members aged 15-64 who are working. picture is the share of male household members that are working among men and picture for female household members. picture represents a variable with the value of 1 where a household has at least one emigrant, and picture denotes a household that receives remittances. picture stands for a set of control variables at the household level.a picture implies regional fixed effects and picture is the randomly distributed error term. The models were run for two different groups of households depending on their location (rural or urban). The coefficients of variables of interest are shown in Table 4.1.

Table 4.1. Remittances and migration seem to reduce labour market participation

Dependent variable: Share of the employed among household members aged 15-64

Main variables of interest: Having an emigrant/receiving remittances

Type of model: OLS

Sample: All households with at least one member working

Variables of interest

Share of the employed household members among:

(1)

All

(2)

Men

(3)

Women

rural

urban

rural

urban

rural

Urban

Household has at least one emigrant

-0.024

(0.045)

-0.021

(0.041)

-0.045

(0.060)

-0.046

(0.061)

-0.004

(0.052)

-0.030

(0.051)

Household receives remittances

-0.159***

(0.047)

-0.114***

(0.044)

-0.120*

(0.066)

-0.102

(0.068)

-0.138**

(0.054)

-0.049

(0.054)

Number of observations

730

657

599

495

718

638

Note: Results that are statistically significant are indicated as follows: ***: 99%, **: 95%, *: 90%. Standard errors in parentheses. The sample excludes households with return migrants only and those with immigrants only.

a. Control variables include the household’s size and its squared value, the dependency ratio (number of children 0-15 and elderly 65+ divided by the total of other members), the male-to-female adult ratio, family members’ mean education level, its wealth estimated by an indicator (Chapter 3) and its squared value.

Remittances and return migration encourage self-employment, but only in rural areas

The literature suggests that as remittances raise household income they can provide those left behind with the capital they need to start up a business and boost self-employment (Mesnard, 2004; Dustmann and Kirchkamp 2002; Woodruff and Zenteno, 2007; Yang, 2008). The IPPMD survey data also suggest that remittances play a role in boosting self-employment, as the share of self-employed is higher among people from households receiving remittances (43%) than among those not receiving them (34%).

Similarly, 44% of return migrants are self-employed, compared to 34% of non-migrants. Return migrants’ savings accumulated abroad can be used as a resource for working for themselves. Growing evidence from the literature suggests that return migrants and their household members tend to be self-employed or establish their own businesses (Ammassari, 2004; De Vreyer et al., 2010; Giulietti et al., 2013).

These patterns are seen in Armenia, too and confirmed by regression analysis (Box 4.2). Table 4.2 shows the results of the analysis and suggests that both receiving remittances and having a return migrant are positively associated with self-employment. Women in rural areas seem to engage more in self-employment when they receive remittances. Having a return migrant in rural households also tends to increase the probability of being self-employed. These effects, however, were not found in urban areas.

Box 4.2. Migration boosts self-employment in rural areas

To further analyse how receiving remittances is associated with the employment choices of household members, two probit models were used in the following form:

picture (4)

picture (5)

where picture represents whether an employed individual i is self-employed, picture signifies that a household receives remittances and picture denotes that a household has at least one return migrant. picture stands for a set of control variables at the individual level and picture for household level controls.a picture implies regional fixed effects and picture is the randomly distributed error term. Table 4.2 shows the computed marginal effects of the main variable of interest on each employment type for the two models.

Table 4.2. Migration boosts self-employment in rural areas

Dependent variable: An individual is self-employed (binary variable).

Main variables of interest: The individual belongs to a household receiving remittances / The individual belongs to a household with at least one return migrant

Type of model: Probit

Sample: Employed people of working age (15-64).

Variables of interest

All

Men

Women

rural

Urban

rural

urban

rural

urban

Household receives remittances

0.076**

(0.034)

0.013

(0.030)

0.066

(0.049)

-0.003

(0.049)

0.094**

(0.044)

0.025

(0.035)

Number of observations

1 407

1 095

827

611

580

448

Household has a return migrant

0.125***

(0.027)

0.023

(0.021)

0.141***

(0.036)

0.021

(0.033)

0.120***

(0.040)

0.019

(0.026)

Number of observations

1 404

1 050

826

582

578

435

Note: Results that are statistically significant are indicated as follows: ***: 99%, **: 95%, *: 90. Standard errors in parentheses. The second model (return migration) excludes households with immigrants only.

a. Control variables include age, sex and education level of individuals and their households’ size and its squared value, the dependency ratio, its wealth estimated by an indicator and whether it is in a rural or urban location.

Migration and agriculture

Agriculture plays an important role in Armenia’s economy. Apart from a sudden increase in its share in the sector’s value-added in GDP in the early 1990s due to a contraction of the non-agricultural sector (Makaryan and Galstyan, 2013), it has remained rather stable since around 2000. In 2015, its value-added as a percentage of GDP was 19% (World Bank, 2017a). Agriculture also employs an important share of the country’s labour force. In 2013, 36% of the employed population worked in the agricultural sector (FAO, 2016b). However, this was the fourth lowest share amongst IPPMD partner countries, and reflects the share of the population living in rural areas (37%; UN, 2014).

The sector is highly characterised by small and subsistence farmers with low productivity, few resources and difficult access to markets (Oxfam, 2016); nearly half of all farms in Armenia are considered to be subsistence level (Mnatsakanyan et al., 2015). Only a very small group of larger farmers and commercial entities account for about 98% of agricultural output in the country (FAO, n.d.). Prospects are, however, looking up; a production per capita index measured at 100 over 2004-06 had increased to 130 by 2013 (FAO, 2016a), the second highest amongst the IPPMD partner countries.

Following the dissolution of the Soviet Union and the privatisation of land in Armenia, most of the rural population began looking to the capital, Yerevan, or outside the country for employment, mainly in Russia. As most of the emigrants were young people, rural areas were left with elderly people unable to carry out many of the heavy activities associated with agricultural work (Mnatsakanyan et al., 2015).

Of the 2 000 households interviewed, 1 001 (50%) had agricultural activities at the time of the interview. These include arable farming (384 households, 38%), animal husbandry (32 households, 3%) or both (585 households, 58%). This section, focusing on these households, asks whether remittances and return migration are helping to modernise and increase productivity in the agricultural sector.

Agricultural households do not channel remittances into productive agricultural investment

According to the IPPMD data, agricultural households in Armenia are more likely to be receiving remittances than non-agricultural households; the difference is statistically significant for remittances originating from any source (29% vs. 21%). Looking specifically at households with current emigrants, the gap remains in favour of agricultural households, with 23% of agricultural households receiving remittances compared to only 16% for non-agricultural ones. Amongst agricultural households with emigrants, 77% receive remittances, compared to only 64% of non-agricultural households.

In theory, remittances can be invested in productive assets such as machinery, barns, fencing, feeding mechanisms, irrigation systems and tractors (Mendola, 2008; Tsegai, 2004). The productive investment of remittances can also help households move from labour-intensive to capital-intensive activities (Lucas, 1987; Taylor and Wouterse, 2008; Gonzalez-Velosa, 2011), or to specialise in a certain type of farming (Böhme, 2015; Gonzalez-Velosa, 2011).

The IPPMD survey included a question on how much households spend on agricultural assets on average over a typical year3 in the previous 12 months. Only 163 agricultural households (18% of those that provided an answer to the question) claimed to have done so. Looking more closely at these 163 households, those receiving remittances were more likely to have made such expenditures (22% vs. 16%, a statistically significant difference). However, they spent less on average than those not receiving remittances (AMD 51 490 vs. 19 940,4 not statistically significant) (Figure 4.3). Households that receive remittances may also choose to spend their additional income on either specialising in one activity, such as farming or animal rearing, or diversifying by doing both. The data suggest little difference between households receiving or not remittances here (59% vs. 58%) (Figure 4.3). In addition, both types of household showed little difference in their degree of agricultural specialisation (not shown).

Figure 4.3. Households receiving remittances spend more on agricultural assets
Household expenditures and business ownership, by whether household receives remittances
picture

Note: Statistical significance calculated using a chi-squared test for differences in shares and a t-test for differences in the amounts spent, is indicated as follows: ***: 99%, **: 95%, *: 90%. Statistical significance in first figure to the left is related to the share of households with agricultural asset expenditures, and not the amount spent.

Source: Authors’ own work based on IPPMD data.

Remittances might also be used to finance entrepreneurial non-farm activities that require capital, such as a retail business or transport services (FAO and IFAD, 2008). This would be consistent with the gradual move away from agricultural dependence occurring in Armenia. This has been the case in Albania, for instance, where remittances have been negatively associated with both labour and non-labour inputs in agriculture (Carletto et al., 2010). However, and contrary to the theory, the data suggest that few households financed non-farm activities and those receiving remittances are slightly less likely than households not receiving remittances to own such a business (1% vs. 3%) (Figure 4.3).

Regression analysis was used to probe further whether households receiving remittances invest in or out of agriculture (Box 4.3). The results largely confirm the patterns suggested above: there is a strong link between a household receiving remittances and investment in agricultural assets – the coefficient is positive and statistically significant (Table 4.3, row 1). However, based on the 163 households that did spend money on agricultural investments, the amount of remittances received seems to be negatively correlated with investments in agricultural assets (Table 4.3, row 1). There also does not seem to be any statistically significant link between the amount of remittances received by a household and having both agrarian farming and animal rearing activities. The regression results also suggest that there is a strong negative link between receiving remittances and ownership of a non-agricultural business (Table 4.3, row 1), as suggested in the descriptive statistics of Figure 4.3, but no link with the amount of remittances sent (Table 4.3, row 2). Overall, remittances seem to have a positive effect on the probability of investing in agriculture, but little effect on any other outcome.

Box 4.3. The links between remittances and investing in farming

To estimate the probability that an agricultural household has invested in an asset or activity, the following regression model was estimated:

picture (6)

where the unit of observation is the household hh and the dependent binary variable agri_exp in equation (3) represents the probability that the household had agricultural expenditures in the previous 12 months and takes on a value of 1 if the household spent money and 0 otherwise, picture represents the fact that the household receiving remittances, picture stands for a set of household-level regressors while picture represents regional-level fixed effects. Standard errors, picture, are robust to heteroskedasticity.

A second OLS model was also estimated:

Ln(picture (7)

where agri_exp represents the logged amount of the agricultural expenditures. All other variables are as defined in equation (6).

Table 4.3 presents the regression results. Column (1) presents results on whether the household has made agricultural asset expenditures, column (2) on the amount spent on agricultural assets in the past 12 months, column (3) on whether the household has activities in both farming and animal rearing, and column (4) on whether the household operates a non-agricultural business. The table also presents results for three variables of interest, estimated in separate models. The top rows present results related to the fact that the household received remittances in the past 12 months, while the middle rows present results related to the logged amount of remittances received by former members of the household in the past 12 months, limiting the sample to those that received remittances.

Table 4.3. Remittances increase the probability of spending on agricultural assets

Dependent variable: Investment outcomes

Main variables of interest: Household received remittances/amount of remittances received by household

Type of model: Probit/OLS

Sample: Agricultural households

Variables of interest

Dependent variables

(1)

Household typically makes agricultural asset expenditures (equation 6)

(2)

Logged amount typically spent on agricultural assets in a 12 month period (equation 7)

(3)

Household has activities in both farming and animal rearing (equation 6)

(4)

Household operates a non-agricultural business (equation 7)

Household received remittances in the past 12 months

0.068**

(0.032)

-0.366**

(0.182)

0.007

(0.039)

-0.024***

(0.008)

Number of observations

893

163

988

891

Logged amount of remittances sent from former household members

0.008

(0.029)

-0.033

(0.156)

0.001

(0.035)

0.004

(0.004)

Number of observations

165

42

182

174

Note: Statistical significance is indicated as follows: ***: 99%, **: 95%, *: 90%. Coefficients resulting from probit models reflect marginal effects. Standard errors are in parentheses and robust to heteroskedasticity.

Return migration can also affect the agricultural sector in many of the same ways as remittances, since the migrants may return with financial, human and social capital, i.e. savings, their own labour, new skills and contacts. However, fewer farming households have return migrants compared to non-farming households, according to the IPPMD data. Of the 509 households with return migrants, 252 (25%) were returning to farming households, while 257 (26%) were from non-farming households, rates that are rather similar. However, and more strikingly, when looking specifically at migrant households (those with current emigrants or return migrants), fewer farming households had return migrants than non-farming households (50% vs. 57%), a statistically significant difference.

In fact, looking at whether having a return migrant in the household is linked with the same outcomes as listed in the analysis on remittances above, the results suggests that return migration does not lead to better investment outcomes for agricultural households (not shown, more details in OECD, 2017). Moreover, the difference between return migrant and non-return migrant households was only slightly more in favour of return migration in terms of operating both arable farming and animal husbandry (59% vs. 58%), nor is return linked with one specific activity or the other (not shown). Finally, households with return migrants were also only slightly more likely to be operating a non-agricultural business than those without a return migrant (3% vs. 2%).

The fact that remittances are being channelled into agricultural asset expenditures is promising, but perhaps not enough. While the literature finds that remittances and return migration are typical vectors for investment and to revitalise the sector, this seems to not be the case in Armenia in general apart from the link between remittances and the probability of spending on agricultural assets. For instance, the amounts spent are negatively correlated with remittances and there is very little evidence of diversification into various agricultural activities or non-agricultural ones. This may be a more general issue, as the rate of non-agricultural business ownership by agricultural households is also remarkably low in the IPPMD dataset.

Some of these results may be linked to the fact that it is mainly the poor who are migrating; although their investment capacity may be increased through remittances and return migration, the amounts may be too low to invest, compared to other, perhaps richer, households. The results suggest that as Armenia transitions away from agricultural dependence, perhaps in tandem with emigration, the sector may suffer. However, migration in Armenia is highly seasonal, particularly to Russia. There is therefore a possibility that some remittances are being brought back by hand and that the links with seasonal migration to Russia are actually reinforcing the agricultural sector, as households exploit differences in seasonal demand between the two countries. This may explain why Armenia seems to be firmly set on an agricultural value-added in GDP of around 20%, as mentioned above.

Migration and education

In 1992, the Armenian higher education system went from a “free of charge” to a partly fee-paying system as professional institutions introduced fee-based education alongside the free-of-charge state-financed education. Reforms were also introduced to meet the European standards related to the Bologna process. These changes concerned pre-professional education, vocational education, higher and post-graduate education, and meant that the number of private educational institutions increased (Makaryan and Galstyan, 2013). Another important change was the transition from the Soviet-era 10-year schooling system to the current 12-year education system (UNICEF, 2010).

Today Armenia’s National Curriculum for General Education is based on a 12-year programme, which consists of compulsory primary (grades 1 to 4), compulsory lower secondary (grades 5 to 9) and upper secondary (high school, grades 10-12). There are also alternatives to high schools in the form of vocational and technical-professional schools.

Close to half the Armenian adult population (47%) have completed post-secondary education – the second highest share in the IPPMD sample after Georgia (at 49%). The pupil-teacher ratio is also relatively low compared to the other IPPMD countries, at 19 students per teacher in 2007 (OECD, 2017). Armenia has made efforts to achieve universal primary school enrolment in the past decade, and primary net education enrolment rates increased from 84% in 2007 to 96% in 2015 (UNESCO, 2016). Although many reform efforts are still underway, the quality of education needs significant improvement, and the curriculum for higher education needs to be better adjusted to the demands of the labour market (Makaryan and Galstyan, 2013). Unemployment is high, especially among the young (as discussed in the labour market section above), and many youth see labour migration as an alternative when they face difficulties finding a job matching their education and skills levels (Makaryan and Galstyan, 2013).

Remittances encourage investments in education

Education is fundamental to individual and national development in both developed and developing economies. Migration and education are closely interlinked in several ways. On the one hand, remittances can alleviate households’ credit constraints and enable households to invest in educating children. Return migrants may also bring back funds to finance education of household members. However, emigration, and in particular parental migration, may have negative consequences for the family left behind. The majority of emigrants from Armenia are men, leaving their wives and children behind to work in the fields in order to keep the households intact. This could lead children to drop out of school, while the absence of parents may also cause emotional deprivation (Manasyan and Poghosyan, 2012).

Existing research shows that an important reason for emigrating from Armenia is to earn enough to cover children’s educational expenses, including hiring tutors, university fees or supporting young people who have moved to another town in order to pursue further education (Makaryan and Galstyan, 2013). Another study – by the International Labour Organization (ILO) – found that one in five Armenian emigrants in the sample aimed to use the income gained abroad to pay for the education of children (ILO, 2009). Better educational prospects abroad are another pull factor for Armenian young people to emigrate (Makaryan and Galstyan, 2013).

The IPPMD data show that overall, about one in ten remittance-receiving household use remittances to pay for a member’s schooling. As shown in Chapter 3, female-headed households are more likely to invest in a member’s schooling (14%) than male-headed households (8%). Households receiving remittances are also more likely to spend a larger share of their budget on education (Figure 4.4). Remittance-receiving households with children of school age (6-20 years old) spend on average 3.2% of their yearly budget on education, compared to 2.7% of households not receiving remittances. The difference is however not statistically significant. The pattern is reverse for return migration: households with a return migrant spend a lower share of their budget on education (2.7%) than those without a return migrant (2.9%).

Figure 4.4. Households receiving remittances spend a larger share of their budget on education
Share of total budget spent on education (%), by migration status
picture

Note: The sample only includes households with children in school age (aged 6-20 years).

Source: Authors’ own work based on IPPMD data.

The association between migration, educational expenditures and school attendance was investigated further using regression analysis, controlling for relevant household characteristics (Box 4.4). The results show that remittances are positively associated with educational expenditures, both in absolute amounts and as a share of household total budget (columns 1 and 2). This may reflect an increase in educational expenses such as extra tutoring or education fees. On the other hand remittances are not associated with higher educational enrolment rates of youth in the age group 15 to 22. No link was found between the probability of attending school and having an emigrant or receiving remittances for men in this age group (column 3). Having an emigrant in the household is negatively associated with school attendance by women in the same age category (column 4, second row). Similarly, return migration is negatively associated with youth school attendance (though the sample was too limited to perform separate analysis for women and men) (lower part of the Table 4.4). This indicates that even though remittances can stimulate more investments in education, migration may have disruptive effects on youth schooling, especially for girls.

Box 4.4. The links between migration educational expenditures and school attendance

A regression framework was developed to estimate the effect of migration and remittances on education expenditures using the following equation:

picture (8)

picture (9)

picture (10)

where the dependent variables picture in equation (8) and picture in equation (9) represent household educational expenditures measured in absolute (logged) values or as share of total household yearly budget respectively. picture represents a binary variable for whether an individual is attending education; picture represents a binary variable for households receiving remittances, where “1” denotes a household receiving remittances and “0” if not; while picture takes on value “1” if the household has at least one emigrant and “0” if not; picture and picture are two sets of observed household characteristics influencing the outcome;a δr represents regional-level fixed effects, standard errors, picture, are robust to heteroskedasticity. As a robustness check, the analysis was also performed excluding immigrants from the sample, which did not alter the results.

In the lower part of the table, the binary variable for remittances is replaced by a binary variable for households having a return migrant.

Table 4.4. Remittances stimulate investments in education, while emigration and return may have the opposite effect

Dependent variable: Educational expenditures (values and share of household budget)

Main variables of interest: Receiving remittances/ having an emigrant/having a return migrant

Type of model: OLS, probit

Sample: All households/only households with children in school age (aged 6-20)

Variables of interest

Educational expenditures

School attendance

Yearly amounts

Share of household budget

Men aged 15-22

Women aged 15-22

Household receives remittances

0.394**

(0.172)

0.011**

(0.004)

-0.043

(0.091)

0.097

(0.075)

Household has at least one emigrant

-0.304*

(0.179)

-0.006

(0.006)

-0.008

(0.092)

-0.127*

(0.069)

Number of observations

406

1 733

422

444

Household has at least one return migrant

-0.049

(0.131)

-0.001

(0.003)

-0.089***

(0.029)

Number of observations

406

1 773

866

Notes: Results that are statistically significant are indicated as follows: ***: 99%, **: 95%, *: 90%. Standard errors are in parentheses. The sample includes immigrants/immigrant households. Excluding immigrants and households with immigrants does not change the results.

Source: Authors’ own work based on IPPMD data.

a. The set of household and individual explanatory variables included in the model are the following: household size and household size squared, household dependency ratio (defined as the number of children and elderly in the household as a share of the total adult population), mean education level of the members of the household, number of children in the household, binary variables for urban location and household head being female, and finally an asset index (based on principal component analysis) that aims to capture the wealth of the household (for all three equations). In addition, the model for school attendance also includes a control for age of the youth.

Return migration has limited impact on human capital accumulation

Whether or not migrants acquire education and skills in the destination country affects the economic payoff of migration (Dustmann and Glitz, 2011). Migrants who acquire education abroad and return with new skills can help increase human capital back home. The extent to which this will happen depends on the degree to which emigrants improve their skills during their migration period, and whether migrants return to their origin countries or not. To enable the use of education and training acquired abroad, the skills need to be validated in such a way that they can be recognised in the domestic labour market. This has previously shown to be a barrier for Armenian return migrants. One study showed that only 7% of Armenians who acquired training abroad received a certificate (ETF, 2013).

Armenian migrants are fairly well educated on their departure. However, few of them acquire education in the country of destination, and even fewer return with new skills. Only 6% of all return migrants in the sample had obtained additional education in the country of destination: 4% of men and 9% of women (Figure 4.5). Compared to other IPPMD countries, this is a low rate (OECD, 2017), but is confirmed by another study which also found that only 6% of return Armenian migrants had acquired additional education abroad (ETF, 2013). This was also a considerably lower rate than the other two countries in the study: Georgia and Morocco. Thus human capital transfers from migration appear to be limited in Armenia.

Figure 4.5. Few Armenian emigrants acquire additional qualifications overseas
Share of current emigrants and return migrants who acquired education abroad (%)
picture

Source: Authors’ own work based on IPPMD data.

Migration, investments and financial services

Armenia has one of the most open investment regimes among the emerging market countries, and is ranked 33 worldwide on the 2017 Economic Freedom index (The Heritage Foundation, 2017). The World Bank also placed it among the top 40 countries worldwide in their latest ease of doing business ranking, and at number 9 in the world for starting a business (World Bank, 2017b). In 2015, the Eurasian Economic Union trading block came into being, grouping Armenia, Belarus, Kazakhstan, Kyrgyzstan, and Russia into a single economic market of 176 million people. Despite these achievements, Armenia still faces challenges in its investment climate. These include its small market size (with a population of less than three million) and its closed borders with Turkey and Azerbaijan.

Armenia has a relatively small but growing financial system, dominated by its banking sector. In 2015, close to 90% of financial assets were held by 21 commercial banks (Central Bank of Armenia, n.d.). The non-bank sector in the country is underdeveloped, with a small but growing insurance sector and a very limited capital market. Developing this sector is a priority for the authorities. A pension reform and other policy measures are currently underway to attract institutional investors and promote financial innovation.

Migration and remittances do not seem to stimulate investments in productive capital

Migration can affect long-term investments in the country of origin in various ways:

  • Migrants can accumulate savings and for example start and run businesses in the country of origin while abroad and on their return

  • Remittances can fund investments in productive assets such as real estate assets

  • Return migrants can bring funds, entrepreneurial skills and valuable networks back to their country of origin

Surprisingly, given the country’s healthy business climate assessment by the World Bank and Hertiage foundation, the level of entrepreneurship in Armenia is relatively low. Following the global crisis in 2008-09, new business registration in Armenia declined and job creation became more challenging. According to the 2010 Life in Transition Survey, only a small share of Armenia’s labour force (12%) has ever attempted to start a business, and among those who do, only a very small share succeeds (6%) (EBRD, 2010).

The share of households owning a business in the IPPMD sample is also very limited, at less than 4%. The difference between households with migration experience (receiving remittances or having a return migrant) and those without is also small, though statistically significant (Figure 4.6).

Figure 4.6. Real-estate ownership is higher among remittance-receiving households and return migrants
Share of households owning a business or real estate, by migration status
picture

Note: Results that are statistically significant (calculated using a chi-squared test) are indicated as follows: ***: 99%, **: 95%, *: 90%. Real estate includes non-agriculture land and housing other than the house or apartment in which the household currently lives.

Source: Authors’ own work based on IPPMD data.

On the other hand, real-estate ownership among households in the sample is higher. This includes non-agricultural land and housing other than the household’s own dwelling. Twenty-five percent of households with a return migrant own real estate, compared to 21% of households without a return migrant (a statistically significant difference). The difference between households with and without remittances is similar (24% vs. 21%), but not statistically significant.

Despite the patterns shown above, the results from the regression analysis, controlling for household characteristics, show no statistically significant links between emigration, return migration, remittances and ownership of a business or real estate. The link between remittances and business ownership is negative, in line with the descriptive statistics in Figure 4.6, indicating that remittance-receiving households are less likely to run a business, although the relationship is not statistically significant. The same holds for real estate, despite the reverse pattern shown in the descriptive statistics in Figure 4.6. Separate analysis for rural and urban households for real estate ownership was also performed (not shown), but did not show any statistically significant relationships.

Hence, despite being ranked highly on global indices for investment climate and business start-ups (as reported above), entrepreneurship is low in Armenia. Migration and remittances do not seem to be promoting productive investments. Potential reasons could be the low financial inclusion (Chapter 5) and underdeveloped financial markets, which make access to loans limited. For example, small and medium-sized companies often lack the necessary skills to be considered credit worthy (IMF-WB, 2012). These limitations are discussed further in Chapter 5.

Box 4.5. The links between migration, remittances and productive investments

To analyse the link between migration and business and real-estate ownership, two probit model regression were run taking the following forms:

picture (11)

picture (12)

where picture is either household business ownership or real estate ownership (depending on the specification); picture takes on value “1” if a household owns at least one business/owns real estate and “0” otherwise; picture in equation (11) represents a binary remittance variable with value “1” for households that receive remittances and “0” otherwise; picture represents a binary variable for whether the household has a migrant or not, and picture are a set of observed household and individual characteristics that are believed to influence the outcome.a picture is a randomly distributed error term. In equation (12) picture is binary variable taking on value “1” if the household has at least one return migrant, and “0” for households without return migrants. As a robustness check, the analysis was also performed having excluded immigrant households from the sample; this did not alter the results.

Table 4.5. Migration and remittances are not linked to higher business or real estate ownership

Dependent variable: Household runs a business/ owns real estate

Main variables of interest: Amount of remittances, having an emigrant/return migrant

Type of model: Probit

Sample: All households

Variables of interest

Dependent variable

Business ownership

Real estate ownership

Household receives remittances

-0.023

(0.018)

-0.002

(0.030)

Household has at least one emigrant

-0.028

(0.019)

0.033

(0.030)

Number of observations

1 803

1803

Household has a return migrant

0.000

(0.010)

0.029

(0.021)

Number of observations

1 803

1 803

Note: Statistical significance is indicated as follows: ***: 99%, **: 95%, *: 90%. The sample includes immigrant households. Excluding households with immigrants does not change the results.

a. The set of household and individual explanatory variables included in the model are the following: household size and household size squared, household dependency ratio (defined as the number of children and elderly in the household as a share of the total adult population), mean education level of the members in the household, number of children in the household, binary variables for urban location and household head being female, and finally an asset index (based on principal component analysis) that aims to capture the wealth of the household.

Conclusions

This chapter has explored how migration affects the four sectors in Armenia: the labour market, agriculture, education, and investment and financial services. The results indicate that migration can have both positive and negative impacts on household well-being and Armenia’s national development and there remains plenty of untapped potential.

On the negative side, remittance receipts appear to reduce the incentives for household members to seek work. Having an emigrant in the household is negatively associated with school attendance by women in the same category. Similarly, return migration is negatively associated with youth school attendance. On the positive side, the financial capital brought home via remittances and return migrants seems to stimulate more self-employment, especially in rural areas. Remittances also seem to be invested in education, with remittance-receiving households spending a larger share of their budget on education than households not receiving remittances.

However, the limited link between migration and productive investment – notably in business and agriculture – is a major missed opportunity for a country that receives significant volumes of remittances. Policies to support and enable households to channel remittances towards productive use, and measures that stimulate investment by return migrants would not only benefit the household, but also the country’s development as a whole. The next chapter explores how sectoral policies influence migration.

References

Acosta, P. (2007), “Entrepreneurship, labor markets and international remittances: Evidence from El Salvador”, in International Migration, Economic Development and Policy, World Bank and Palgave Macmillan, Washington, DC, pp. 141-159.

Ammassari, S. (2004), “From nation-building to entrepreneurship: the impact of élite return migrants in Côte d’Ivoire and Ghana”, Population Space Place, 10: 133–154.

Amuedo-Dorantes, C. and S. Pozo (2006), “Remittance receipt and business ownership in the Dominican Republic”, The World Economy, Vol. 29/7, Wiley-Blackwell, Hoboken, NJ, pp. 939-956.

ArmStat (2016), Labour Market in RA, 2016, National Statistical Service of the Republic of Armenia, Yerevan, www.arstat.am.

ArmStat (2015), National Accounts of Armenia, 2003-2007, Statistical Handbook, National Statistical Service of the Republic of Armenia, Yerevan, www.arstat.am.

Böhme, M.H. (2015), “Does migration raise agricultural investment? An empirical analysis for rural Mexico”, Agricultural Economics, 46(2), 211-225.

Carletto, G., B. Davis, J. Miluka and Z. Zezza (2010), “The vanishing farms: the impact of international migration on Albanian family farming”, Journal of Development Studies, 46(1):140-161, www.tandfonline.com/doi/abs/10.1080/00220380903197978.

Central Bank of Armenia (n.d.), Statistics of Financial Organisations, database, Central Bank of Armenia, https://www.cba.am/en/SitePages/statfinorg.aspx.

De Vreyer, P., F. Gubert, and A. S. Robilliard (2010), “Are There Returns to Migration Experience? an Empirical Analysis Using Data on Return Migrants and Non-migrants in West Africa”, Annals of Economics and Statistics, Vol. 97/98, pp. 307–28, www.jstor.org/stable/41219120.

Dustmann, C. and A. Glitz (2011), “Migration and education”, In: Hanushek, E.A., S. Machin and L. Woessmann (Eds), Handbook of the Economics of Education, Volume 4, Elsevier.

Dustmann, C. and O. Kirchkamp (2002), “The optimal migration duration and activity choice after re-migration”, Journal of Development Economics, 67: 351-372.

EBRD (2010), Life in Transition Survey II: After the Crisis, European Bank for Reconstruction and Development, London.

ETF (2013), Migration and Skills in Armenia, Georgia and Morocco: comparing the survey results, European Training Foundation, Torino, www.etf.europa.eu/webatt.nsf/0/E811EB8749A0287BC1257C0700318D02/$file/Migration&skills_Armenia&Georgia&Morocco.pdf.

FAO (n.d.), “Armenia at a glance”, Food and Agriculture Organization of the United Nations, Rome, www.fao.org/armenia/fao-in-armenia/armenia-at-a-glance/en.

FAO and IFAD (2008), International migration, remittances and rural development”, Food and Agriculture Organization of the United Nations and International Fund for Agricultural Development, Rome, www.fao.org/3/a-ak405e.pdf.

FAO (2016a), “Gross per capita Production Index Number (2004-2006 = 100)”, FAOSTAT (database), www.fao.org/faostat/en/#data/QI, (accessed 1 October 2016).

FAO (2016b), “Employment distribution, agriculture”, FAOSTAT (database), www.fao.org/faostat/en/#data/OE, (accessed 1 October 2016).

Funkhouser, E. (2006), “The effect of emigration on the labour market outcomes of the sender household: A longitudinal approach using data from Nicaragua,” Well-being and Social Policy, 2(2), 5-25.

Giulietti, C., Wahba, J. and K.F. Zimmermann (2013), “Entrepreneurship of the left-behind”, IZA Discussion Paper No. 7270, Institute for the Study of Labor.

Gonzalez-Velosa, G. (2011), “The effects of emigration and remittances on agriculture: evidence from the Philippines”, Mimeo, University of Maryland.

Heritage Foundation (2017), 2017 Index of Economic Freedom, www.heritage.org/index/ (accessed 1 March, 2017).

ILO (2012), International Standard Classification of Occupations: ISCO-08, International Labour Organization, Geneva.

ILO (2009), “Migration and development: Armenia country study”, ILO Subregional Office for Eastern Europe and Central Asia, 2009, www.ilo.org/wcmsp5/groups/public/---europe/---ro-geneva/---sro-moscow/documents/publication/wcms_308849.pdf.

IMF-WB (2012), Republic of Armenia: Financial System Stability Assessment, International Monetary Fund and the World Bank, Washington, DC.

Kim, N. (2007), “The impact of remittances on labor supply: the case of Jamaica”, Policy Research Working Paper Series No. 4120, World Bank, Washington, DC.

Lucas, R.E.B. (1987), “Emigration to South Africa’s mines”, American Economic Review 77(3).

Lucas, R.E.B. and O. Stark (1985), “Motivations to remit: Evidence from Botswana”, Journal of Political Economy 93(5).

Makaryan, G. and Galstyan, M. (2013), “Costs and benefits of labour mobility between the EU and the Eastern Partnership Partner Countries: Armenia Country report”, CASE-Center for Social and Economic Research No. 461/2013, Warsaw.

Manasyan, H. and Poghosyan, G. (2012),”Social Impact of Emigration and Rural-Urban Migration in Central and Eastern Europe: Final country report”, Armenia.

Mendola, M. (2008), “Migration and technological change in rural households: Complements or substitutes?” Journal of Development Economics, 85(1–2), February 2008: 150-175, http://dx.doi.org/10.1016/j.jdeveco.2006.07.003.

Mesnard, A. (2004), “Temporary migration and capital market imperfections”, Oxford Economic Papers, 56(2): 242-262.

Mnatsakanyan, H., V. Urutyan and A. Yeritsyan (2015), “Country Report: Armenia,” International Center for Agribusiness Research and Education, Yerevan, www.agricistrade.eu/wp-content/uploads/2015/05/Agricistrade_Armenia.pdf.

OECD (2017), Interrelations between Public Policies, Migration and Development, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264265615-en.

Osaki, K. (2003), “Migrant remittances in Thailand: economic necessity or social norm?”, Journal of Population Research, 20(2), 203-222.

Oxfam (2016), “Strengthening Armenia’s Agricultural Sector Through Multi-Stakeholder Networking,” Oxfam Case Study, A case study on the Agricultural Alliance, Oxford, http://oxfamilibrary.openrepository.com/oxfam/bitstream/10546/620114/1/cs-armenia-agriculture-networking-aa-241016-en.pdf.

Taylor, J.E. and F. Wouterse (2008), “Migration and income diversification: evidence from Burkina Faso”, World Development, Vol. 36(4): 625-640.

Tsegai, D. (2004), “Effects of migration on the source communities in the Volta Basin of Ghana: Potential links of migration, remittances, farm and non-farm self-employment activities”, Working Paper, Economics and Technological Change, University of Bonn, Bonn.

UN (2014), World Urbanization Prospects: The 2014 Revision, UNDESA, New York.

UNESCO (2016), Institute for Statistics (UIS) database, www.uis.unesco.org (accessed 3 February 2017).

UNICEF (2010), “Education in Armenia: Education profile”, UNICEF, New York, https://www.unicef.org/ceecis/Armenia_2010.pdf.

Woodruff, C. and R. Zenteno (2007), “Migration networks and microenterprises in Mexico”, Journal of Development Economics, 82(2): 509-528.

World Bank (2017a), “Agricultural, value added (% of GDP)”, World Development Indicators (database), http://data.worldbank.org/indicator/NV.AGR.TOTL.ZS, (accessed 1 February 2017).

World Bank (2017b), Doing Business (database), www.doingbusiness.org/rankings, (accessed 3 February 2017).

Yang, D. (2008), “International migration, remittances and household investment: evidence from Philippine migrants’ exchange rate shocks”, Econ J 118(528):591–630.

Notes

← 1. Defined as the ratio of labour force to the working age population (15-64).

← 2. See Chapter 3 for methodological background on the regression analyses used in this project.

← 3. The question in the survey asked households how much they spend on average on agricultural productive assets (such as farming equipment) over the course of one year.

← 4. Using the exchange rate with the USD at 1 July 2014, the equivalent totals are USD 49 vs. 126.