Issuu on Google+

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright


Author's personal copy Economic Systems 34 (2010) 386–396

Contents lists available at ScienceDirect

Economic Systems jo urnal homepage: www.elsevier.com/loc ate/ecos ys

An empirical analysis of international labour migration in the Philippines Frank W. Agbola a,*, Angelito B. Acupan b a

Newcastle Business School & Centre for Institutional and Organisational Studies, Faculty of Business and Law, The University of Newcastle, Australia, 1 University Drive, Callaghan, NSW 2308, Australia b Strategy, Policy and Communication Office, Small Business Guarantee and Finance Corporation, the Philippines

A R T I C L E I N F O

A B S T R A C T

Article history: Received 27 October 2008 Received in revised form 26 February 2010 Accepted 2 March 2010

This study empirically investigates the impact of economic, demographic, and political factors on the size of emigration from the Philippines. In 2007, overseas workers from the Philippines sent remittances in excess of US$14 billion annually to their families back home. Although these remittances are an important source of foreign exchange and play an important role in economic development, the determinants of emigration in the Philippines are not well established. A simple unrestricted error correction model of migration was specified and estimated using data spanning the period 1975–2005. Results indicate that the level of unemployment, adult literacy and population density are the key determinants of emigration in the Philippines. The result also indicates that government instability impacts negatively on emigration in the Philippines. The policy implications of the results are discussed. ß 2010 Elsevier B.V. All rights reserved.

JEL classification: F20 F22 J61 J68 Keywords: International labour migration Unit root Cointegration Error correction model Philippines

1. Introduction For centuries, people have migrated in search of opportunities for economic prosperity or political freedom. Over the last two decades, international migration has grown rapidly, mainly due to the globalisation of economic activities and its ensuing effect on labour migration (UNPF, 2006). These recent episodes of rising international migration illustrate the challenges facing developing and developed countries in a globalised economy. International migration, mainly from developing

* Corresponding author. Tel.: +61 249212048. E-mail addresses: frank.agbola@newcastle.edu.au (F.W. Agbola), aacupan@sbgfc.org.ph (A.B. Acupan). 0939-3625/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ecosys.2010.03.002


Author's personal copy F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

387

countries to developed countries, has generated significant improvements in the lives of migrants and their families. These international migrants receive higher wages and their families who are left in their country of origin benefit through the remittances that they receive from the family member working overseas. Countries of destination of migrants on the other hand have benefited from the increased supply of skilled and unskilled labour while countries of origin saw the easing of their domestic labour market pressures (World Bank, 2006). In the last two decades or so, the dramatic growth in international migrant remittances has captured the attention of governments and international agencies because of its potential to contribute to the growth and development of national and regional economies. This is in addition to its potential to reduce poverty (Blackwell and Seddon, 2004). As highlighted by the International Monetary Fund (IMF), remittances have the capacity to act as a cushion for economic shocks and provide direct benefits to households (IMF, 2006a). The World Bank reports that workers’ remittances to developing countries were US$240 billion in 2007, and this has served as an important source of external funding for developing countries and now acts as a crucial and stable source of foreign exchange earnings for developing countries (World Bank, 2008). Given that the estimate primarily reflects official remittances and fails to account for those sent via unofficial channels, the actual amount of remittances is likely to be higher (see also Puri and Ritzema, 1999). It is for this reason that remittances have been regarded as the third pillar of development, together with foreign direct investment and overseas development assistance (Alfieri et al., 2005). The Philippines stand prominent among remittance-receiving and labour-exporting countries. With US$14.4 billion in official remittances in 2007 (BSP, 2008), the Philippines was one of the developing world’s largest remittance-recipients in absolute value terms. In terms of labour export, the Commission on Filipinos Overseas (CFO) estimates that around 8.2 million Filipinos live and work in more than 193 countries/territories around the world (CFO, 2008). In addition, based on CFO estimates, 43% of Filipinos outside the country are permanent emigrants while the rest are temporary or contract workers. Collectively called Overseas Filipino Workers (OFWs), they constitute the third largest diaspora population in the world (Zhang, 2006). The objective of this paper is to empirically investigate the impact of economic, demographic and political factors on the size of emigration in the Philippines. A simple unrestricted error correction model of migration was specified and estimated using annual data spanning the period 1975–2005 in a time-series theoretic framework. The findings of this study provide an understanding of how the historical conditions, as shaped by socioeconomic and political systems, have reshaped international labour migration in the Philippines. This paper outlines how these findings could serve as a useful guide for government policymakers seeking to formulate and implement policies to address the issues concerning international labour migration in the Philippines. The remainder of this paper is arranged as follows. Section 2 reviews briefly the immigration policy initiatives of the Philippine government for 1970–2005. Section 3 describes the methodology employed in the analysis and the data. Section 4 reports and discusses the estimated results. Section 5 provides some concluding remarks. 2. The Philippines’ migration policy initiatives Migration is a pervasive phenomenon in the Philippines. The rapid increase in the size and composition of migrants in the last two decades or so has rekindled the debate over migration in the Philippines. International migration of Filipinos, either for temporary labour opportunities or permanent migration, has been growing for many decades now. However, it was not until the passing of the Immigration and Nationality Services Act of 1965 in the United States of America that Filipino immigration diversified. This act, also known as the Hart-Celler Act, paved the way for the abolition of the natural-origin quotas and the dismantling of the nationality-based immigration restrictions into the United States of America and allowed the mass entry of immigrants from Asia and Latin America. The Act was subsequently followed by the scrapping of the same pro-European immigration policies in other countries of settlement like Canada, Australia and New Zealand. This led to a rise in Filipino migrants to these traditional immigration countries (Asis, 2006). In addition, Filipinos also settled in non-traditional immigration countries like Japan and Germany, either through marriage or work-


Author's personal copy 388

F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

related migration. However, this permanent migration, according to Asis (2006), was surpassed by the large temporary labour migration to Middle Eastern countries, for example Saudi Arabia and Kuwait, which began in the early-1970s and continues up till today. Many academic scholars cite the early-1970s as the time when the Philippine government embarked on a policy of labour export as a development strategy (Lindio-McGovern, 2004; Yang, 2004; O’Neil, 2004). These authors argue that it was the government’s response to ease domestic unemployment that triggered the implementation of such a policy. What was initially meant to be a stop-gap measure to tackle high domestic unemployment and a balance of payments crisis (Opiniano, 2004), however, has steadily led to an exodus of Filipinos overseas, now numbering about 8.2 million according to the data from the Commission on Filipinos Overseas (2008). According to the National Statistical Coordination Board, temporary OFWs comprise about 13% of the country’s labour force and 5% of the national population (NSCB, 2006). This is certainly a huge increase from only 3694 Filipinos deployed abroad as part of the Philippines Overseas Employment Program in 1969 (UN, 2000). Despite the fact that a growing number of studies attribute the mass exodus of Filipino workers to the Philippine government policy, the government is cautious not to admit to having an explicit policy of overseas employment (see Opiniano, 2004, for further discussion). The presence – or absence – of a ‘‘clear’’ policy on migration notwithstanding, the country has continued to be a driving force in international labour migration, making it the third largest labour-exporting nation in Asia (LindioMcGovern, 2004). It is worthwhile to note, however, that the Philippine legislature has passed a number of laws governing the basic policy framework on overseas employment. While formulating policies that help facilitate overseas employment and providing for some safety measures as well as incentives, some laws were carefully crafted in not being explicit in stating that it is a policy of overseas migration. These policy reforms only led to ambiguity and further confusion regarding government policy towards overseas migration (Opiniano, 2004). A case in point is Section 2c and Section 2i of Republic Act 8042 or the Migrant Workers and Overseas Filipinos Act of 1995. Section 2c states that: ‘‘. . .the State does not promote overseas employment as a means to sustain economic growth and achieve national development. The existence of the overseas employment program rests solely on the assurance that the dignity and fundamental human rights and freedoms of the Filipino citizen shall not, at any time, be compromised or violated.’’ Section 2i states that: ‘‘Nonetheless, the deployment of Filipino overseas workers, whether land-based or sea-based, by local service contractors and manning agencies employing them shall be encouraged’’. Clearly, the dualistic nature of the Philippine government’s policy can also be observed in government actions towards international labour migration. As Opiniano (2004) emphasises, in various forums on migration attended by government officials, the government advances the message that it does not have an explicit policy for sending Filipino workers overseas. The common policy pronouncements are that overseas employment is purely market-driven and that it is a phenomenon which is filled with opportunities that Filipino workers have instinctively taken advantage of. The Philippine government argues that its role is limited to facilitating the flow of migrant workers to foreign labour markets where their services are needed. The actions towards the promotion of overseas employment, however, seem to belie such policy pronouncements by the Philippine government. Although the marketing missions on overseas employment by various government departments, such as the Department of Labor and Employment (DOLE), the Philippine Overseas Employment Administration (POEA), the Department of Foreign Affairs (DFA), and the Overseas Workers Welfare Administration (OWWA) are seen as promoting overseas employment, the Philippine government argues that these missions are aimed at acquiring strategic information on potential labour markets as well as maintaining discussions with host countries that help to ensure the welfare of migrant workers (Tyner, 2000). It is important to recognise that while the number of OFWs continues to grow, domestic employment has declined. By the mid-1990s, more Filipinos had actually found jobs overseas compared to those that were added to the number of employed persons in the domestic labour market


Author's personal copy F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

389

every year (Go, 2002). This was earlier confirmed by Imperial (2001) in her study on the Philippine labour market situation where she found that in 1999 and 2000, more than 800,000 workers were deployed abroad while less than 200,000 were added to the local labour force. In a related study, Alburo and Abella (2002) found that the number of Filipino professional workers who went abroad exceeded the net additions to the professionals in the domestic workforce. The private sector and the government – through the Philippine Overseas Employment Administration (POEA) and the Overseas Workers Welfare Administration (OWWA) – send over half a million workers on temporary migration every year (Alburo and Abella, 2002). In light of the high unemployment rate in the country, it is clear that this provides a safety valve to what would otherwise be even higher unemployment. However, in recent times, international labour migration has become more than just a stop-gap measure to counter unemployment problems in the Philippines. It has now become a central fixture of the economy, largely due to the huge amount of remittances that migrant workers send back to the country. Recent migration from the Philippines has caused significant demographic, economic and social changes. Given that international migration in the Philippines could impact on output and incomes in the country, with the associated changes in the labour market having important implications for economic growth and development to the extent that unemployment could be affected, an understanding of the impact of changes in socioeconomic, demographic and political factors on international labour migration in the Philippines is pertinent. If we heed the ever growing empirical evidence that shows that migration is becoming an increasingly important catalyst for development, especially in relatively poorer countries (Sriskandarajah, 2005), then an understanding of the determinants of migration should provide strategic information to aid policy formulation and implementation in the Philippines. 3. The model and data A number of theories have been advanced in the economic literature to explain the factors that influence the decision to migrate. An economic theory of migration analyses the allocation of labour across international boundaries where the underlying premise is that individuals migrate because of tangible and intangible benefits. Individual decisions to migrate are, however, constrained by immigration policies and socio-economic and demographic factors of the individual. As Hoddinott (1994) emphasises, these explanations of migration based on behavioural decisions of the individual give primacy to Ravenstein’s (1885) sixth law that individuals move in order to ‘better’ themselves. This concept is espoused in the migration model proposed by Todaro (1969), where it is argued that migration is the outcome of individual utility maximisation. Several authors have extended Todaro’s migration model to include household utility maximisation. Contributions to this area of research include those by Connell et al. (1976), Rempel (1981), Stark and Levhari (1982), and Low (1986), among others. Hoddinott (1994) is critical of this approach, arguing that these theories of migration reflect an oversimplification of the behaviour of migrants. It also fails to explain the role prospective migrants play in the decision-making process. Hoddinott further argues that these migration models fail to provide an insight into how costs and benefits of migration are shared. Clearly, the motivation to migrate may be described as influenced by a combination of socioeconomic, ethnic and political factors. Borjas (1989) provides a succinct review of migration theories, which are hence not discussed here. However, a brief summary of the three main theories identified by Borjas (1989), which have shaped recent methodological frameworks for modelling migration behaviour, is presented. The first is the human capital investment theory proposed by Hicks (1939) and espoused by Sjaastad (1962). This theory posits that migration rises the higher the returns, and vice versa. The second is the asymmetric information theory which is based on the concept of ‘brain drain’. This theory posits that host countries have more information about immigration skills than the source country and are therefore capable of influencing migration patterns (see Kwok and Leland, 1982, for further discussion). The third theory, popularised by Piore (1979), is based on the conjecture that the impact of migration on natives in the receiving country is likely to be small because immigrants are forced to take jobs that natives refuse to accept. Borjas is critical of these theories though, arguing that they fail to provide a sound theoretical foundation for explaining three important questions relating to migration, namely, the factors that determine the direction, size and composition


Author's personal copy 390

F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

of migration flow, the way in which migrants adapt to host countries, and finally the impact of migration on the economies of sending and receiving countries. Based on the neoclassical economic theory of utility maximisation for individuals and profit-maximisation for employers, Borjas developed theoretical models of migration and demonstrates the usefulness of using economic theory to provide unique predictions to explain the unresolved issues of migration. Borjas concludes that neoclassical models generally provide consistent estimates of the size, composition and impact of migration flows. Given the controversy surrounding the theory of migration, the understanding of the determinants of migration becomes increasingly important. As Borjas (1989) emphasises, ‘‘given any initial sorting of the population across countries, international differences in income opportunities, political conditions and immigration policies imply that incentives exist for some individuals to migrate to other countries.’’ (p. 457). As such, it is important that theories of migration provide some insights into which way the flows go, how large the flows are and which kinds of individuals become immigrants. Clearly, there is currently no single unified theory of emigration that addresses these issues. Instead, a number of theories and hypotheses continue to emerge aimed at examining the determinants of migration. The migration model is a relationship linking migration to variables identified by economic theory. In accordance with the existing literature, a migration model is developed for the Philippines, starting from a gravity model, which is a reduced form equation derived from a system of demand and supply relationships, and assuming that supply and demand are influenced by income, which captures the capacity to migrate, population and factor endowments. This study adopts a macroeconomic perspective and focuses on domestic factors impacting on international migration in the Philippines. Following the extant literature, particularly the studies by Greenwood (1975), Borjas (1987, 1989), and Adams and Page (2003), the Philippines migration model is specified as follows: m ¼ a0 þ a1 cmt þ a2 adlt þ a3 ineqt þ a4 ineqsqt þ a5 govst þ a6 ir t þ a7 pdt þ a8 ur t þ a9 yt þ et (1) where the dependent variable m is the share of migrants in total Philippines population, cm is the cost of migrating from the Philippines to various labour-receiving countries, adl is the adult literacy rate, ineq is the level of income inequality as measured by the gini coefficient, ineqsq is the level of income inequality-squared, govs is the measure of political instability, ir is the inflation rate, pd is the population density as measured by the number of people per square kilometre, ur is the unemployment rate, y is the real per capita income. This study employs the general to specific approach of Hendry and Ericsson (1991) in the specification of the dynamic migration model with the insignificant coefficients eliminated sequentially and reparametised to obtain a parsimonious specification. Hence, the final model was estimated in first difference with the error correction term, which captures the long-run time-series properties of the data series. The data on migration were obtained from the Commission on Filipinos Overseas (CFO) and the National Statistics Coordinating Board (NSCB) in the Philippines (NSCB, 2006), and include both temporary labour workers and permanent Filipino migrants living and working overseas. Data on population density were obtained from NSCB (NSCB, 2006). Data on Philippines gross domestic product, unemployment rate, and consumer price index were obtained from the International Monetary Fund’s (IMF) International Financial Statistics CD Rom (IMF, 2006b) and supplemented with official records of the Bangko Sentral ng Pilipinas (BSP) (BSP, 2006) and the NSCB (NSCB, 2006). Data on adult literacy rate were obtained from BSP (BSP, 2006) and the NSCB (NSCB, 2006). This study utilises the Augmented Dickey–Fuller (ADF) test (Dickey and Fuller, 1979) and Phillip– Perron (P–P) (Phillips and Perron, 1988) unit root tests to investigate the time-series properties of the data series. Table 1 reports the unit root test results. From Table 1, except for the P–P test for the income variable, the ADF and P–P test results indicate that all the data series fail to reject the null hypothesis of a unit root in levels, but become stationary in first difference. Given that all the data series are of the same order of integration for the ADF tests, which is I(1), we proceed to perform the Johansen cointegration test. We start by performing a preliminary test for cointegration based on the Engel–Granger test (Engle and Granger, 1987). The results led to the rejection of the null hypothesis of no cointegrated series and


Author's personal copy F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

391

Table 1 Unit root test results. Augmented Dickey–Fuller test

Phillip–Perron test

Variable

Level

First difference

Level

First difference

m cm adl ineq ineqsq govs ir pd ur y

2.437 1.668 0.968 1.992 1.513 1.712 3.030 1.503 2.711 0.626

3.838 4.010 5.523 7.232 7.291 6.762 6.756 8.127 6.809 5.369

2.321 0.264 1.196 2.298 1.742 1.853 3.028 1.309 2.698 2.357

5.776 9.900 5.723 4.562 4.439 6.657 16.900 13.439 7.514 2.123

Note: The McKinnon (1991) critical values are 2.623 for 10%, 2.967 for 5% and 3.679 for 1% level.

the acceptance of the existence of at most a single cointegrated vector given that the t-statistic of 4.485 of the coefficient of the lagged residual in the test equation is greater than the critical values of 3.98, 3.42 and 3.13 at the 1%, 5% and 10% levels, respectively. However, as Agbola and Damoense (2005) note, the Engel–Granger approach assumes the existence of at most a single cointegration vector and is also sensitive to the choice of dependent variable. To overcome these limitations, we proceed to determine the presence of an equilibrium-type relationship between the variables using the Johansen’s multivariate test (Johansen and Juselius, 1990). Table 2 reports the Maximal Eigenvalue and Trace tests results for cointegration. The results of the Maximal Eigenvalue and Trace tests indicate that we can reject the null hypothesis of no cointegration. The Maximal Eigenvalue test indicates five cointegrated vectors, while the Trace test indicates that there are eight cointegrated vectors at a 5% significance level. Furthermore, we perform the bounds test approach suggested by Pesaran et al. (2001). This technique is superior to the Engel and Granger and Johansen cointegration tests because it can be performed regardless of whether the underlying order of integration of the data series are of the same order or not. For the Pesaran bounds test, we reject the null in favour of the alternative hypothesis given that our sample test statistic exceeds the relevant upper critical value. We conclude that there exists a long-run relationship between all the variables in the Philippine migration model. Data problems continue to constrain studies on migration and this study is no different. These data problems relate to the way in which data related to international migration are collected and compiled. It is therefore important to spell out how the data were gathered. The data used in this study are annual data from 1975 to 2005. The cost of migration was estimated as the average cost of a oneway airline ticket from the capital of the Philippines to the top 24 country-destinations of Filipino workers. It is important to note that the use of one-way airline tickets may underestimate the real costs of migration, given that it does not take into account other costs of migration, such as the cost of job placement and the cost of travelling from the place of residence in the Philippines to the country’s Table 2 Maximal Eigenvalue and Trace test results for cointegration. Hypothesized no. of CE(s)

None At most At most At most At most At most At most At most At most

1 2 3 4 5 6 7 8

Maximum Eigenvalue test

Trace test

Statistic

0.05 Critical value

Statistic

0.05 Critical value

266.59 93.82 62.15 48.76 38.94 21.98 19.90 15.13 2.87

58.43 52.36 46.23 40.08 33.88 27.58 21.13 14.26 3.84

570.13 303.54 209.72 147.57 98.81 59.87 37.89 17.99 2.87

197.37 159.53 125.62 95.75 69.82 47.86 29.80 15.49 3.84


Author's personal copy 392

F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

capital, as well as administrative costs, which can be enormous (Taylor, 2006). However, due to the lack of reliable data on the other costs of migration, this study uses the one-way airline ticket as the proxy for the cost of migration. The proxy used to represent the cost of migration only represents, at best, the minimum costs associated with migration. Hence, the impact of this variable on migration should be interpreted with caution. It is hypothesised that a negative relationship exists between the cost of migration and the number of migrants. Based on human capital theory, the more educated a person is, the more likely it is that they will migrate. This study aims to investigate this hypothesis for the Philippines. Data on the number of people with primary or secondary education are not readily available, particularly for the study period. The adult literacy rate is therefore used as a proxy for the level of education of Filipinos. This study hypothesises a positive relationship between adult literacy rate and migration. Previous studies have found the income of migrants to impact significantly on migration. This study incorporates income measured as per capita gross domestic product in purchasing power parity terms. A negative relationship is hypothesised between income and migration. This study also incorporates inequality, measured as a Gini coefficient and inequality-squared to capture the Roy effect in the migration literature. The Roy effect is based on the premise that differences in the return to skills will select migrants from different parts of the skill distribution (Clark et al., 2003). The inflation rate and unemployment rate variables are included in the model to capture the effects of domestic economic activity on migration flow. A high inflation rate, which is indicative of economic malaise, is expected to impact negatively on migration flow. A high unemployment rate is expected to exert an upward pressure on migration flow as people migrate in search of jobs. This study includes the policy variable of government stability as measured, which was derived as the frequency of change in government leadership and the number of attempted and successful uprisings against the government. 4. Results and discussion Table 3 reports a simple unrestricted error correction model of migration for the Philippines by ordinary least squares method in the Eviews 6.0 econometric package. While the Johansen test suggested the existence of more than one cointegrating vector, the general to specific approach favoured a more parsimonious model with one error correction term and tests for the adequacy of this parsimonious model supports this. The error correction term (ECTt1), which accounts for the shortrun adjustment of the cointegrated variables, is negative and estimated to be 0.77, indicating a convergence towards a long-run steady state in a little over a year. Most of the estimated parameters Table 3 Unrestricted ECM of migration for the Philippines. Variable

Coefficient

t-Statistic

Prob.

Constant Dcm Dadl Dineq Dineqsq Dgovs Dir Dur Dpd Dy ECTt1

1.6350 0.0017 0.6590 171.195 165.927 0.4820 0.0358 0.2361 0.2541 0.0002 0.776

3.573 0.429 5.225 1.594 1.323 3.444 1.327 4.629 2.756 0.293 4.224

0.020 0.673 0.000 0.127 0.202 0.003 0.200 0.000 0.013 0.773 0.000

Diagnostics R2-adjusted = 0.63 F-value = 5.848 (p = 0.000) Breusch–Godfrey serial correlation LM test: x2 = 1.099 (p = 0.294) Ramsey RESET mis-specification test: x2 (1) = 0.010 (p = 0.919) Jarque–Bera normality test: x2 (1) = 4.962 (p = 0.084) ARCH test: x2 (1) = 0.312 (p = 0.577)


Author's personal copy F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

393

have the expected signs and are statistically significant. The R2 goodness-of-fit measure is adequate as it explains 63% of the variation in migration flow in the Philippines. The calculated F-statistic is 5.848 (p-value = 0.000) and indicates that the explanatory variables are jointly significant in influencing the Philippines’ migration pattern and that the model is useful. The null hypothesis of no model misspecification could not be rejected as the estimated Ramsey RESET test statistic of 0.010 has a p-value of 0.919. The null hypothesis of no serial correlation was not rejected as the Breusch–Godfrey LM test statistic was estimated to be 1.099 with a p-value of 0.294. The null hypothesis of normality of the residuals was not rejected as the Jarque–Bera test statistic was estimated to be 4.962 with a p-value of 0.084. The CUSUM and CUSUMSQ plots (not reported) reveal no evidence of random or systematic movements in the estimated coefficients over the study period. According to the results of the diagnostic tests, the migration model is well behaved and well specified. We proceed to discuss the empirical results. The cost of migration, per capita income and population density, known as gravitational demographic variables, were included in the estimated migration function. Table 3 shows that the coefficient of the cost of migration (cm) variable is negative but statistically insignificant in influencing migration in the Philippines. It is important to note that although the cost of migration is not statistically significant it has a negative impact on migration flow, as expected. As pointed out earlier, the use of one-way airfare as a proxy for the cost of migration may not fully capture the total cost incurred by migrants, hence the statistical non-significance of the coefficient of the cost of migration in the migration model. The negative impact is similar though to those of Adams and Page (2003) and Clark et al. (2003), who also found that the cost of migrating to a labour-receiving region reduces the share of international migration in many low-income and middle-income developing countries. The population density (pd) variable is found to be negative and statistically significant in influencing the migration flow in the Philippines. The negative impact of the population density variable on migration suggests that as the population increases in the Philippines the migration flow falls. This is contrary to expectations, as an increase in population is expected to result in an increased flow of migrants. For the Philippines government, this finding is important, as it demonstrates that there is an inherent tendency of the economy to discourage out-migration as the population grows. While this is difficult to rationalise, a possible explanation is that an increase in population will be associated with an increase in the demand for goods and services produced domestically. This will stimulate growth within the economy. As the welfare of individuals within the economy improves, they are unlikely to migrate. Further research investigating the reasons for this result is, however, warranted. The per capita income (y) variable, a proxy for the level of economic growth and development and absorptive capacity of the Philippines economy, is found to have a negative but statistically insignificant impact on migration flow out of the Philippines. This finding is consistent with the argument put forward earlier that as the economy grows it has the potential to discourage migration out of the Philippines. This finding has important implications as it suggests that as the Philippines government pursues a policy of promoting growth within the economy, this has the potential to curb migration out of the country. The implication is that it is important that the Philippines government implements sustainable developmental initiatives such as raising entrepreneurial awareness and skills as well as improving the quality of education, as these are capable of absorbing the excess labour force and thus improve the living standards, thereby discouraging emigration in the Philippines. The coefficient of the adult literacy (adl) variable is negative and statistically significant at a 5% level in influencing the migration flow in the Philippines. The result indicates that the migration flow in the Philippines is highly sensitive to the adult literacy rate. The negative sign indicates that an increase in adult literacy discourages emigration out of the Philippines, as expected. This result is important as it demonstrates that human capital development, as captured by the adult literacy rate variable, does play an important role in curbing the migration flow out of the Philippines. This finding is contrary to the positive effect of human capital on the migration flow as postulated by the human capital theory. A possible explanation for this result is that Filipino migrants abroad often work in relatively unskilled positions so that being highly educated may not be necessary for these tasks and hence the observed negative relationship between human capital development and migration. This perhaps reflects the composition of migrants from the Philippines – they are mostly unskilled workers.


Author's personal copy 394

F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

Clark et al. (2003), on the study of migration theory, emphasise that differences in the return to skills will select migrants from different parts of skill distribution. This was earlier articulated by Borjas (1987, 1991) and is referred to as the Roy effect. The relationship between migration and skill distribution in the Philippines is captured by the inclusion of the Gini coefficient – an inequality measure – and a statistic that describes the distribution of household income and skill distribution within the Philippines. Table 3 indicates that the coefficient of the inequality (ineq) variable and the inequality-squared (ineqsq) variable are statistically insignificant in influencing migration in the Philippines, thus demonstrating that the Roy effect does not hold for the Philippines. Quite important in developing countries, such as the Philippines, is the political climate and the regularity of transfer of power in the country. This study follows the approach proposed by Gastil (1987), whereby a dummy variable is used as a measure to rate the political and civil liberty within the economy, with a higher rating indicating less freedom. In this study, the number of coups and coup attempts is used as a proxy for political instability. The coefficient of the political instability (govs) variable is negative and statistically significant at a 5% level, as expected. The result indicates that increased political instability reduces emigration from the Philippines. This finding is consistent with those of Karemera et al. (2000), who also found that increased political instability, which is often accompanied by increased restrictions on domestic travel, reduced international migration to North America. However, it is important to note that for the Philippines, there was no official restriction on residents’ overseas travel even during the politically unstable years. The result indicates that as political stability deteriorates the level of migration declines. While our data does not permit us to move beyond speculation, the result appears consistent with events of the day. During the highly politically unstable years of 1983, 1986, 2001, 2005 – when there was growing restiveness and clamour amongst the general populace for a change in government – migration as a percentage of the population declined markedly. The periods immediately following these tumultuous times caused migration to rise, reflecting a renewed sense of national pride and hope (Spaeth, 2006). While other factors may also have influenced the decline in migration in those years, the result of this study adds some credence, at the very least, to the negative impact of political instability on the migration flow in the Philippines. The explanatory variables of inflation rates and unemployment rates are included in the model to capture the effects of domestic economic activity on migration flow. Now, a high inflation rate is indicative of economic malaise and high cost of travel. The coefficient of the inflation rate (ir) variable is negative but statistically insignificant in influencing migration. Although statistically insignificant, the negative sign suggests that an increase in the inflation rate in the Philippines could lower migration out of the Philippines given that potential migrants may not have the capacity to migrate as inflation tends to raise the cost of migration. The negative but statistically insignificant sign perhaps reflects substitution effects of composite economic events that are captured by the inflation rate variable but are not linked to emigration (see, Karemera et al., 2000, for further discussion). The coefficient of the unemployment rate (ur) variable is positive and statistically significant at a 5% level. The result indicates that a rise in the unemployment rate leads to an increase in emigration. The statistical significance of the unemployment rate variable perhaps reflects the existence of a relatively large unemployed labour force in the Philippines. The result does provide some empirical evidence to suggest that growing unemployment in the Philippines could serve as a trigger for the government to further pursue the policy of labour export as a developmental strategy to stabilise the level of unemployment in the Philippines. 5. Conclusion The international labour migration of Filipinos, either for temporary or permanent labour opportunities or permanent migration, has been growing for many decades now. In the present study, we investigate the determinants of migration in the Philippines during the period 1975–2005. The results indicate that the key determinants of migration are adult literacy rate, unemployment rate, population density and political instability. An important finding is that human capital development negatively impacts on emigration in the Philippines, contradicting the human capital theory prediction. The level of economic growth and development as captured by per capita income and


Author's personal copy F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

395

domestic activity as captured by inflation rate were found not to impact on emigration in the Philippines. Another important finding is that the Roy effect does not hold for the Philippines. In the last decades or so, international labour migration has emerged as one of the main factors in social transformation and development in the Philippines. Its significance looks set to grow as the factors affecting the flow of international labour migration change in response to changing economic conditions domestically and abroad. A review of migration policy reforms in the Philippines suggests the lack of effective monitoring of international migration in the Philippines and the extent to which economic factors influence the decision of migrants. This is due in part to the failure on the part of the government to acknowledge migration as a developmental strategy and make a concerted effort to ensure the smooth and efficient flow of migrants. The results of this study underscore the importance of the need for the Philippine government to pursue policies of ensuring effective monitoring and collection of comprehensive statistics on international migration in order to assess the impact of changing economic and political conditions on migration in the Philippines. The recent establishment of bilateral agreements by the Philippine government with migrant receiving countries on the regulation of migration and the rights and condition of Filipino migrants could help to enhance protection and greater social security of Philippine citizens working abroad. In the end, the Philippine government should look into implementing policies and programs that will develop a more viable and vibrant domestic economy that is capable of absorbing the labour force in the Philippines and avoid the reliance on the unpredictable demand by migrant receiving countries.

References Adams, R.H. Jr., Page, J., 2003. International migration, remittances and poverty in developing countries. World Bank Policy Research Working Paper 3179, World Bank, Washington, DC. Agbola, F.W., Damoense, M.K., 2005. Time-series estimation of import demand for pulses in India. Journal of Economic Studies 32, 146–157. Alburo, F.A., Abella, I.D., 2002. Skilled labour migration from developing countries: study on the Philippines. International Labour Office International Migration Papers 51, International Labour Office, Philippines. Alfieri, A., Havinga, I., Hvidsten, V., 2005. Definition of remittances and relevant BPM5 Flows. United Nations, available at: http:// unstats.un.org/UNSD/tradeserv/TSGdocuments/issuePaper1.pdf. Asis, M.B., 2006. The Philippines’ culture of migration. Migration Information Source, available at: http://www.migrationinformation.org/Profiles/print.cfm?ID=364. Blackwell, M., Seddon, D., 2004. Informal Remittances from the UK: Values Flows and Mechanisms. UK Department for International Development, Norwich. Borjas, G.J., 1987. Self selection and the earnings of immigrants. American Economic Review 77, 531–553. Borjas, G.J., 1989. Economic theory and international migration. International Migration Review 23, 457–485. Borjas, G.J., 1991. Immigration and self-selection. In: Abowd, J., Freeman, R. (Eds.), Immigration, Trade and the Labour Market. University of Chicago Press, Chicago, IL, pp. 29–76. BSP, 2006. Overseas Filipino workers’ remittances by country and by type of worker. Bangko Sentral ng Pilipinas, Philippines, available at: http://www.bsp.gov.ph/statistics/keystat/ofw.htm. BSP, 2008. Overseas Filipino workers’ remittances by country and by type of worker. Bangko Sentral ng Pilipinas, Philippines, available at: http://www.bsp.gov.ph/statistics/keystat/ofw.htm. CFO, 2006. Stock estimate of overseas Filipinos. Commission on Filipinos Overseas, Philippines, available at: www.poea.gov.ph/ docs/ofwstock2003.doc. CFO, 2008. Stock estimate of overseas Filipinos. Commission on Filipinos Overseas, Philippines, available at: www.cfo.gov.ph/ Stock%202006.pdf. Clark, X., Hatton, T.J., Williamson. J.G., 2003. What explains cross-border migration in Latin America? Harvard Institute of Economic Research Discussion Paper No. 2012, available from http://www.economics.harvard.edu/faculty/williamson/ files/Cross-Border.pdf. Connell, J., Dagupsta, B., Laishley, R., Lipton, M., 1976. Migration from Rural Areas: The Evidence from Village Studies. Oxford University Press, Delhi. Dickey, D.A., Fuller, W.A., 1979. Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association 74, 427–431. Engle, R.F., Granger, C.W.J., 1987. Co-integration and error correction: representation, estimation, and testing. Econometrica 55, 251–276. Gastil, R.D., 1987. Freedom in the World. Greenwood Press, Westport, CT. Go, S.P., 2002. Remittances and international labour migration: impact on the Philippines. In: Metropolis Inter-Conference Seminar on Immigrants and Homeland, Dubrovnik, May 9–12. Greenwood, M., 1975. Research on national migration in the United States: a survey. Journal of Economic Literature 13, 397– 433. Hendry, D.F., Ericsson, N., 1991. An econometric analysis of UK money demand. American Economic Review 81, 8–38. Hicks, J.R., 1939. The foundation of welfare economics. The Economic Journal 49, 696–712. Hoddinott, J., 1994. A model of migration and remittances applied to Western Kenya. Oxford Economic Papers 46, 459–476.


Author's personal copy 396

F.W. Agbola, A.B. Acupan / Economic Systems 34 (2010) 386–396

Imperial, G.S., 2001. A Study of Active Labour Market Policies and Programs in the Philippines. Draft Paper Prepared for ILO. Institute for Labour Studies, Philippines. IMF, 2006a. International Financial Statistics. International Monetary Fund, Washington, DC. IMF, 2006b. Remittances statistics. International Monetary Fund, Washington, DC, available at: http://www.imf.org/external/ np/sta/bop/remitt.htm. Johansen, S., Juselius, K., 1990. Maximum likelihood estimation and inferences on cointegration—with applications to the demand for money. Oxford Bulletin of Economics and Statistics 52, 169–210. Karemera, D., Iwuagwu, V., Davis, B., 2000. A gravity model analysis of international migration to North America. Applied Economics 32, 1745–1755. Kwok, V., Leland, H., 1982. An economic model of brain drain. American Economic Review 72, 91–100. Lindio-McGovern, L., 2004. Alienation and labour export in the context of globalization: Filipino migrant domestic workers in Taiwan and Hong Kong. Critical Asian Studies 36, 217–238. Low, A., 1986. Agricultural Development in Southern Africa: Farm Household—Economics and the Food Crisis. James Currey, London. McKinnon, J.G., 1991. Critical values for cointegration tests. In: Engel, R., Granger, C.W.J. (Eds.), Long-Run Economic Relationships: Readings in Cointegration. Oxford University Press, New York, pp. 266–276. NSCB, 2006. Labour and employment. National Statistical Coordination Board, Philippines, available at: http:// www.nscb.gov.ph/secstat/d_labor.asp. O’Neil, K., 2004. Labour export as government policy: the case of the Philippines. Working Paper, Migration Policy Institute, Washington, DC, available at: http://www.migrationinformation.org/feature/display.cfm?ID=191. Opiniano, J.M., 2004. Our future beside the exodus: migration and development issues in the Philippines. Working Paper, Institute of Migration and Development Studies, Philippines. Pesaran, H.M., Shin, Y., Smith, R.J., 2001. Bounds testing approaches to the analysis of long-run relationships. Journal of Applied Econometrics 16, 289–326. Phillips, P.C.B., Perron, P., 1988. Testing for a unit root in time-series regression. Biometrika 75, 335–346. Piore, M.J., 1979. Birds of Passage: Migrant Labour and Industrial Societies. Cambridge University Press, New York. Puri, S., Ritzema, T., 1999. Migrant worker remittances, micro-finance and the informal economy: prospects and Issues. Social Finance Unit Working Paper 21, Enterprise and Cooperative Development Department, International Labour Office, Geneva. Ravenstein, E.G., 1885. The law of migration. Journal of the Royal Statistical Society 48, 162–227. Rempel, H., 1981. Rural–Urban Labour Migration and Urban Unemployment in Kenya. International Institute for Applied Systems Analysis, Laxenburg. Sjaastad, L.A., 1962. The cost and returns of human migration. Journal of Political Economy 70, 80–93. Spaeth, A., 2006. Glory days. Time Asia, 20 February, available at: http://www.time.com/time/printout/0,8816,5010602271161331,00.html#. Sriskandarajah, D., 2005. Reassessing the Impacts of Brain Drain on Developing Countries. Discussion Paper, Migration Policy Institute, Washington, DC. Stark, O., Levhari, D., 1982. On migration and risk in LDCs. Economic Development and Cultural Change 31, 191–196. Taylor, E.J., 2006. International migration and economic development. In: Paper presented at the International Symposium on International Migration and Development, Turin, 28–30 June. Todaro, M.P., 1969. A model of labour migration and urban unemployment in less developed countries. American Economic Review 69, 486–499. Tyner, J.A., 2000. Global cities and circuits of global labour: the case of Manila Philippines. Professional Geographer 52, 61–74. UN, 2000. Youth in the Philippines: a review of the youth situation and national policies and programmes. In: Economic and Social Commission for Asia and the Pacific, United Nations, New York. UNPF, 2006. A passage to hope: women and international migration. United Nations Population Fund, available at: http:// www.unfpa.org/swp/2006/pdf/en_sowp06.pdf. World Bank, 2006. Global Economic Prospects 2006: Economic Implications of Remittances and Migration. The World Bank Group, New York. World Bank, 2008. Migration and Remittances Factbook 2008. The World Bank Group, New York. Yang, D., 2004. How remittances help migrant families. Working Paper, Migration Policy Institute, University of Michigan. Zhang, K., 2006. Recognizing the Canadian diaspora. Canada Asia Commentary No. 5, Asia Pacific Foundation of Canada, Vancouver.


An empirical analysis of international labour migration in the Philippines