Remittances and Competitiveness: The Case of the Philippines

World Development Vol. 39, No. 10, pp. 1834–1846, 2011 Ó 2011 Elsevier Ltd. All rights reserved. 0305-750X/$ - see front matter www.elsevier.com/locate/worlddev

doi:10.1016/j.worlddev.2011.04.019

Remittances and Competitiveness: The Case of the Philippines VERONICA BAYANGOS Bangko Sentral ng Pilipinas, Philippines

and KAREL JANSEN *,  International Institute of Social Studies of the Erasmus University in Rotterdam, The Netherlands Summary. — The paper looks at the impact of migration and workers’ remittances on the competitiveness of the home economy. It extends existing research that concentrated on the exchange rate effects of remittances, the so-called Dutch disease effect, by adding labor market effects. The results show that the labor market effects of emigration and remittances have a significant impact on competitiveness that goes beyond the traditional exchange rate effect. Ó 2011 Elsevier Ltd. All rights reserved. Key words — remittances, Dutch disease, competitiveness, labor market, monetary policy, Asia, Philippines

1. INTRODUCTION

inflows is associated with an outmigration of workers so that the domestic labor force declines. Moreover, households receiving remittances may use the higher income to reduce work effort and increase leisure or education, which will further reduce the labor supply. The reduction of the labor supply may lead to an increase in the wage level, which will increase production cost and reduce competitiveness. This paper will analyze these various channels along which migration and remittances can affect the competitiveness of the economy and will use the experience of the Philippines to assess the importance of these effects. The Philippines is an interesting case study. A significant and increasing number of workers move overseas and workers’ remittances are a major source of foreign exchange receipts. Since the mid-1990s the annual deployment of workers has doubled and the remittances inflows have more than doubled as percentage of GDP. Over the same period the export performance of the Philippines has been lacklustre, actually the exports to GDP ratio declined. It is thus relevant to investigate whether there is a link between migration, the inflow of remittances and export performance. There are a few studies that analyze the link between remittances and the real exchange rate for the Philippines (e.g., Tuan˜o-Amador, Claveria, Co, & Delloro, 2007) but, to our knowledge, no studies that present a complete analysis of the various channels along which migration and remittances affect competitiveness.

Workers’ remittances are a financial inflow of increasing importance for many developing countries (World Bank, 2006). In recent years the amounts remitted have increased sharply. Estimates for 2008 in Global Development Finance 2009 show that workers’ remittances are the most important financial flow to developing countries after FDI: remittances outnumber official flows and private debt flows (World Bank, 2009). Behind this sharp increase in remittances is a surge in migration. The Global Economic Prospects of 2006 reports that the stock of immigrants to high-income countries increased at about 3% per year from 1980 to 2000 and that their share in the total population of these countries has increased (World Bank, 2006, see also the Human Development Report 2009, UNDP, 2009). In this paper we are concerned with the macroeconomic impact of migration and workers’ remittances and in particular with how they affect the competitiveness of the home country of the migrant. The surge in migration and remittances has received increasing attention in development research. Much of this research is focused on households: what motivates migration and remittances, how receipt of remittances affects poverty and household decisions, for example, with respect to education and health care but there is also attention for the macroeconomic impact of migration and remittances (for recent surveys of the literature see Fajnzylber & Lopez, 2008; Farrant, MacDonald & Sriskandarajah, 2006; Loser, Lockwood, & Balcazar, 2006; Rapoport & Docquier, 2005; World Bank, 2006). In most of this literature the analysis of the effect on competitiveness is limited to the impact of remittances on the exchange rate. Remittances constitute an inflow of finance and this may lead to an appreciation of the real exchange rate undermining the competitiveness of the traded-goods sector and, in particular, of exports; the so-called Dutch disease effect. However, an increase in remittances may affect competitiveness also through other channels. An increase in remittances

* We are grateful for the detailed and helpful comments from four anonymous referees. We also thank Professor Arjun Bedi and Professor Mansoob Murshed for comments on an earlier draft. We are solely responsible for any remaining errors. The views expressed in this paper are ours and no responsibility should be attributed to the BSP and the ISSErasmus University. Final revision accepted: February 21, 2011.   Karel Jansen passed away on 28 April 2011. His research focused mainly on international finance and development. He made significant contributions on the effects of macroeconomic policies, foreign direct investments and remittances on the stability of developing economies. He will be deeply missed but many will continue the work in his spirit. 1834

REMITTANCES AND COMPETITIVENESS: THE CASE OF THE PHILIPPINES

In this paper we will develop a quarterly macroeconometric model to capture the ways in which migration and remittances affect the economy and to simulate the impact of changes in remittances on competitiveness and export performance. The rest of the paper is organized as follows: the next section reviews the literature on the macroeconomic impact of migration and remittances. In section three we present the main recent trends in emigration from and remittances to the Philippines. The fourth section presents the main features of the macroeconomic model that we apply to test the impact of remittances and section five presents the results of the simulations of the model. The final section presents our conclusions. 2. LITERATURE REVIEW McCormick and Wahba (2000) present a theoretical model in which migration and remittances are endogenous: the worker faces two decisions: (1) whether to migrate and (2) if migrating, how much to transmit back home. Both these interrelated decisions are made so as to maximize household utility (i.e., utility of the worker and of the dependents who remain home). The migration decision depends on differentials in wages and non-traded goods prices between the home and destination countries and on unemployment in the home country. The optimal amount of remittances follows from the maximization of the household utility function. The model shows that one additional migrant reduces employment, output and demand in the home country. But if the amount of remittances exceeds the lost production at home (and with high unemployment and underemployment this is highly likely), net demand increases and the price of non-traded goods at home rise, that is, an appreciation of the real exchange rate; the so-called Dutch disease effect. This effect has received considerable attention in the literature on the impact of remittances (see Acosta, Lartey, & Mandelman, 2009; Amuedo-Dorantes & Pozo, 2004; Bourdet & Falck, 2006; Lartey, Mandelman, & Acosta, 2008, Loser et al., 2006, Vargas-Silva, 2009; Wahba, 1998). An increase in the inflow of remittances, first of all, increases the supply on the foreign exchange markets and may thus lead to an appreciation in the nominal exchange rate. However, increased liquidity on financial markets as result of the inflow of remittances may reduce the domestic interest rate, which may soften the appreciation of the nominal exchange rate. The jump in remittances is followed by an increase in spending by the households receiving the transfers on traded and on non-traded goods and, as the supply of non-traded goods is constrained in the short-run, this will lead to an increase in the price of non-traded goods or an appreciation of the real exchange rate (defined in this case as the relative price Pt/ Pn). This appreciation may be stronger in the short run, when the production factors are fixed or when the emigration of workers has even reduced output in the sector. In the medium and longer run, labor can shift toward the non-traded sector and new investment will expand production capacity in the sector. Econometric testing confirms the existence of the Dutch disease effect; Bourdet and Falck (2006) use co-integration analysis to assess the long-term relationship between remittances and the real exchange rate in Cape Verde and their basic model shows that a 10% increase in remittances leads to an appreciation of the real exchange rate of 1.2%. Amuedo-Dorantes and Pozo (2004) pool annual data for 13 Latin American and Caribbean countries and find that a doubling of remit-

1835

tances per capita raises the real exchange rate by 23%. Lartey, Mandelman & Acosta (2008) use panel data for 109 developing and transition countries and using Generalized Method of Moments (GMM) estimation method conclude that a 1% point increase in the remittances/GDP ratio leads to an appreciation of the real effective exchange rate of between 20% and 40% (depending on the regression model used). They also regress the composition of output (the ratio of traded over non-traded output) on remittances and find an 1% point increase in the remittances to GDP ratio leads to an 1% point fall in this output ratio: remittances have thus a persistent impact on resource allocation. Lopez, Molina, and Bussolo (2008) run a cross-country analysis in which they regress changes in the real effective exchange rate on changes in the ratio of remittances to GDP and a set of control variables. In the base run, they find that a doubling of the remittances ratio leads to an appreciation of 5% but in other runs, using other estimation techniques they come to larger impacts. Acosta et al. (2009) follow a different approach; they use a dynamic stochastic general equilibrium model fitted on El Salvadorian data. They explore three cases: one where remittances are exogenous, one where they are endogenous and driven by altruistic motives and thus counter-cyclical, and one where they are endogenous but driven by self-interest and thus pro-cyclical. In all three cases they observe that an increase in remittances leads to an appreciation in the real exchange rate and a decline in the labor supply. Vargas-Silva (2009) develops a model incorporating remittances and exchange rates as determinants of money demand in Mexico. Using variance decompositions and impulse response functions derived from a structural vector error correction model, he observes that positive shocks to remittances lead to an appreciation of the Mexican peso. Tuan˜o-Amador et al. (2007) find that the increase in remittances to the Philippines in recent years has been associated with an appreciation of the real exchange rate and with a shift in output from the traded to the non-traded sector. These studies suggest that the Dutch disease effect is significant and often substantial. The effect of remittances on the real exchange rate is an important channel through which the competitiveness of exports is reduced but it is not the only channel. Emigration implies a withdrawal of workers from the local labor market and may have an effect on local wages. Yabuuchi and Chauduri (2007) use an analytical model with three sectors and skilled and unskilled workers. Their interest is primarily in wage inequality but the model also shows that an emigration of either unskilled or skilled workers leads to an increase in the wages of both skilled and unskilled workers. Mishra (2006) estimates wage equations for the various skill categories of the Mexican labor force which include a variable measuring the proportion of the category that emigrated and he finds that a 10% reduction in the labor supply in a skill-group leads to a 4% increase in the wages in Mexico. The increase in wages will undermine the competitiveness of the traded sector and will, moreover, reduce the return on capital in the sector and this is likely to reduce investment in the sector. There is a further labor supply effect. Households receiving the remittances may supply less labor or reduce work effort (Chami, Fullenkamp, & Jahjah, 2003). Acosta et al. (2009) test two specifications of their general equilibrium model: one with and the other without labor supply response to remittances and they find that the version with the labor supply effects better fits with the data for El Salvador. Yang (2008) finds, on the basis of analysis of household data that the increase in remittances to the Philippines during the Asian crisis had no significant effect on the total number of hours worked but he did

1836

WORLD DEVELOPMENT

find that the increase in remittances meant that more children were kept in school and children spent less time working (Yang, 2008). Our own time-series estimations show that there is significant effect of remittances on labor supply. The increase in the wage level could be compensated by an increase in labor productivity. Most studies find, as we do, that an increase in remittances leads to an increase in aggregate demand and output but a decline in employment. These two trends imply that, on aggregate, labor productivity rises. There are a number of channels along which labor productivity will be affected. The people who migrate may leave positions of unemployment or low-productivity employment in the informal sector. This would imply that average labor productivity will increase. The rise in aggregate demand and the appreciation of the real exchange rate due to remittances will generate a shift of resources from the traded to the non-traded sector (see Lartey et al., 2008). It is generally assumed that the non-traded sector is more labor-intensive and has lower labor productivity and lower productivity growth compared to the traded sector. If that is so the shift or resources will reduce average labor productivity and also reduce the growth rate of productivity. On the other hand, the rise of wages will induce firms to invest in labor-saving techniques which will increase productivity. Such investment would also benefit from reduced credit cost due to the fall in the market interest rate and from the reduced cost of imported capital goods due to the appreciation of the exchange rate induced by the inflow of remittances. The investment effect is likely to be different in the traded and in the non-traded sector. The traded sector faces higher wage cost but it cannot adjust its prices in the face of international competition; in fact the appreciated exchange rate is already undermining its market position. In these firms profits are under pressure and this discourages new investment but, at the same time, productivity gains from new investment may be the only way to defend their market position. The non-traded sector is sheltered from international competition although it will still face competition on the local market. The increased wage cost may be compensated by the falling cost of imported inputs and even when total production cost increase its stronger market position would enable the sector to shift the higher cost into product prices. The stronger demand for its products increases sales and profits, and this would induce investment, but the increased wage cost may endanger profitability. This could, in itself, become a reason for investment: to reduce labor intensity and increase labor productivity. Migration and remittances may also affect labor productivity in more direct ways. Returning temporary migrants re-enter the labor market with skills acquired abroad making them more productive and some remittances may be directed at investment opportunities in the home country and this will contribute to an increase in productivity. There are also studies that suggest that remittances are used to finance investment in human capital (Rapoport & Docquier, 2005). Calero, Bedi, and Sparrow (2008) use household survey data for Ecuador to show that remittances are used to finance schooling, particularly of girls and in rural areas, and help to prevent children from dropping out of school when the family is hit by negative shocks. This investment in human capital could increase productivity in the longer run. But Beine, Docquier, and Rapoport (2008) suggest that there may be a more immediate effect as well. They observe that educated workers have a greater chance to migrate and that this prospect of migration invites investment in education. They use immigration data for OECD countries to calculate migration rates by educational level for a large number of

source countries and they use these rates to show that skilled migration significantly increases the ex ante (pre-emigration) human capital stocks. Of course, most of this human capital disappears with emigration but some remain in the home country. Simulations in Beine et al. (2008) suggest that, in absolute numbers, there is a small positive effect on the number of skilled workers that are available in the source countries and this could stimulate productivity. Taken together, remittances could thus affect the competitiveness of exports through their impact on the nominal and real exchange rate but also through their effect on the labor supply and the work effort, wages and labor productivity. In the medium term the effect of changes in aggregate demand, wages, interest rates, and the real exchange rates on investment will further affect productivity and competitiveness. It should be recognized that competitiveness is a complex concept. In this paper we focus on the impact of remittances on the real exchange rate, on wages, and on labor productivity. Together these factors determine the dollar unit labor cost of exports. Unit labor cost is clearly only one of the determinants of the competitiveness of a country. The World Economic Forum publishes an annual competitiveness ranking of countries which is based on 12 aspects: Institutions, Infrastructure, Macroeconomic Stability, Health and Primary Education, Higher Education and Training, Goods Market Efficiency, Labor Market Efficiency, Financial Market Sophistication, Technological Readiness, Market Size, Business Sophistication and Innovation. 1 We may assume that many of the aspects are determined by longer-term processes and will change only slowly and are not much influenced by fluctuations in remittances in the short and medium terms. It has also been observed that some countries face rising unit labor cost but still show a strong export performance (see e.g., Fagerberg, 1988). Fagerberg, Knell, & Srholec (2004) argue that the competitiveness of a country is determined not just by cost but also by factors like technology, the capacity to exploit technology, and the production structure. Technology can be either imitated (catching up) or developed through innovation. Countries differ in their ability to exploit existing or new technology. This capacity can be built through human capital, infrastructure, etc. Moreover, the production (or export) structure of a country may consist of products with high income elasticity (new, high-tech, and high quality products) or low elasticity goods. The empirical work of Fagerberg et al. (2004) shows that rich countries tend to grow slowly as they are already at the technology frontier and further technological advance depends on innovation but, on the other hand, they have a high capacity to exploit technology and they concentrate on high elasticity products. Developing countries are catching up on technology and have little own capacity at developing new technology and often their capacity of applying and exploiting new technology is low. Moreover, their production and export composition is biased toward products with lower income elasticity. For these developing countries, therefore, the cost competitiveness is crucial. 3. RECENT TRENDS IN MIGRATION AND REMITTANCE FLOWS IN THE PHILIPPINES Migration and remittances started to grow in the 1970s when oil price hikes created demand for overseas workers in the Gulf and over time they have become more important to the Philippine economy. Official statistics estimate the number of Filipinos abroad at 8.2 million by the end of 2008 (compared to the Philippine 2008 labor force of 37 million).

REMITTANCES AND COMPETITIVENESS: THE CASE OF THE PHILIPPINES

Overseas Filipinos (OF) are classified into three categories. Permanent migrants are legal permanent residents abroad. These may be Philippine passport holders or naturalized citizens in the host country. Temporary migrants are Filipinos whose stay overseas, while regular and properly documented, is temporary based on the employment-related nature of their status in the host country. Temporary migrants include contract workers, intra-company transferees, students, trainees, entrepreneurs, businessmen, traders, and others whose stay abroad is 6 months or more. Irregular migrants are those whose stay abroad is not properly documented. They do not have valid residence and work permits; they can also be overstaying workers or on tourists visa in a foreign country. Data on migration and overseas workers are problematic as different agencies follow different ways to collect their data. The statistics in Table 1 are constructed by the Commission on Filipino Overseas on the basis of embassy reports. The Philippine Overseas Employment Administration (POEA) reports data on the annual deployment of temporary overseas workers based on registration by recognized employment agencies (see Figure 1): they show the number of temporary OF with a contract going out each year (both new hires and re-hires). Table 1 shows that in 2008 there were 3.9 million Filipinos that had permanently settled abroad. That is a sharp increase from the 2.3 million of 1998: over this decade about 150,000 persons are leaving each year. The group of permanent migrants include workers but also dependents; available statistics suggests that about half of this group are workers. The temporary and irregular migrants have gone out to work abroad, in most cases without dependents; their number has sharply risen: since the mid-1990s the annual outflow has doubled. Of course, these temporary workers will eventually return to the Philippines and rejoin the labor force but their temporary absence, and the fact that in recent years ever greater numbers go abroad implies a reduction the labor supply at home. The annual outflow of emigrants and the rising number of deployed temporary workers are a considerable drain on the

1837

labor market. Data from the National Statistical Office show that over the last decade the population of working age increases by about one million per year and the labor force by about 600,000 per year. The permanent emigration and the rise in the number of deployed temporary workers may withdraw between 150,000 and 200,000 persons of working age from the Philippines each year. This represents a considerable labor market effect. With the growth of migration we see a rise in remittances receipts. Remittances in this paper cover transfers sent by both Filipino migrants and overseas workers. In the Philippines, remittance data are sourced from the balance of payments statistics. Table 2 shows data on remittances as recorded by the central bank of the Philippines, Bangko Sentral ng Pilipinas (BSP). Table 2 shows that the magnitude of remittances to the Philippines has been significant, both in absolute terms and as a percentage of GDP and other economic indicators. In 2009 remittances reached US$ 17.4 billion and the latest available data for 2010 (January to June) showed remittances at US$ 9.1 billion. 2 It is important to note that this record is incomplete: the central bank records the remittances that flow through the banking system but there are many other channels along which the transfers are made. Vos (1992) uses data from household surveys to estimate the unrecorded remittances and concludes that recorded remittances cover around half of the total. The National Statistics Office (NSO) runs an annual survey amongst OF to trace the channels along which remittances are made and comes to the same conclusion: the officially recorded remittances account for about half of total transfers made. 3 The data in Table 2 may thus give an incomplete picture of the level of remittances received but they do give a correct impression of the trends. Recorded remittances were around 3% of GDP in the early 1990s and increased significantly in the second half of the 1990s to a level of around 10% of GDP in recent years. A small part of this increase is due to

Table 1. Stock estimates of overseas Filipinos as of end-December 2008 Region/Country World Total Africa Asia, East and South Hong Kong Japan Malaysia Taiwan Asia, West Kuwait Saudi Arabia United Arab Emirates Europe Germany Italy Spain Americas/Trust territories Canada United States Oceania Australia Region unspecified

Permament

Temporary

Irregular

Total

3,907,842 1,986 247,097 23,507 141,210 26,002 8,100 4,599 500 351 713 294,987 44,619 27,003 32,435 3,101,941 533,826 2,552,034 257,232 233,943 –

3,626,260 44,303 581,330 125,810 60,020 89,681 83,070 2,144,625 136,018 1,072,458 541,666 299,468 8,075 77,087 14,190 250,595 73,632 128,616 44,325 23,926 261,614

653,609 8,265 256,622 6,000 30,700 128,000 2,885 112,700 10,000 20,000 32,000 98,624 2,100 13,000 4,055 166,163 6,135 155,843 11,235 7,975 –

8,192,650 54,554 1,085,049 155,317 231,930 243,683 94,055 2,261,924 146,518 1,092,809 574,379 693,079 54,794 117,093 50,680 3,518,699 613,593 2,836,493 312,792 265,844 266,553

Permanent—immigrants or legal permanent residents abroad whose stay do not depend on work contracts. Temporary—persons whose stay overseas is employment related, and who are expected to return at the end of work contracts. Irregular—those not properly documented or without valid residence permits or who are overstaying in a foreign country. Source of data: Commission on Filipino Overseas.

1838

WORLD DEVELOPMENT 1,600,000

1,400,000

1,200,000

1,000,000

800,000

600,000

400,000

200,000

20 08 20 09

20 06 20 07

20 04 20 05

20 02 20 03

20 00 20 01

19 98 19 99

19 96 19 97

19 94 19 95

19 92 19 93

19 90 19 91

19 89

19 88

19 86 19 87

19 84 19 85

0

Figure 1. Deployment of temporary contract workers, 1984–2009. Source: POEA.

Table 2. Relative size of Overseas Filipino (OF) Remittances: level, growth rate and as a percentage of selected economic indicators Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 (Jan–Jun)

Level (US$B)

Annual Growth (%)

As % of: GDP

XGS

FD

GIR

DSB

4.3 5.7 7.4 6.0 6.1 6.0 6.9 7.6 8.6 10.7 12.8 14.5 16.4 17.4 9.1

11.3 33.3 28.3 18.3 0.5 0.3 14.2 10.1 12.8 25.0 19.4 13.2 13.7 5.6 3.8

5.2 7.0 11.3 8.9 8.0 7.9 8.4 8.9 9.1 10.8 10.9 10.0 9.9 10.8 10.3

28.9 36.9 35.7 20.8 17.1 20.8 21.3 22.7 23.1 23.9 24.1 24.4 28.3 36.4 30.4 (Jan–May)

118.9 681.1 365.5 544.9 270.1 3,092.8 448.6 1,543.4 1,242.7 578.5 438.9 495.5 1,063.9 890.6 1,667.9 (Jan–May)

36.6 65.3 68.0 40.0 40.2 38.4 42.1 42.4 52.7 57.8 55.6 42.8 43.7 39.2 18.6

85.7 102.6 144.6 91.5 96.6 92.4 88.7 95.3 118.5 140.2 157.7 188.2 223.0 253.6 227.5 (Jan–May)

Source of data: Department of Economic Statistics, Bangko Sentral ng pilipinas. Refers to cash remittances passing through the banking system. GDP, gross domestic product; XGS, exports of goods and services; FDI, foreign direct investment; GIR, gross international reserves; DSB, debt service burden.

the growing share of remittances going through official channels but the main part is a genuine increase in transfers. Tuan˜o-Amador et al. (2007) presented three major factors behind the uptrend in OF remittances since 1996. One, there is a trend rise in the number of deployed workers and emigrants. For another, there has been a change in the skill composition of Filipinos workers and migrants. From 1995 to 2007, there was a significant rise in the number of deployed Filipino workers in the services and professional categories. And thirdly the measures adopted by the BSP and the banks to encourage OF to channel their remittances through the financial system are essential. The BSP’s initiatives are geared toward enhancing transparency and promoting competition in the remittance market; improving access to financial services, especially the transfer of funds to beneficiaries in remote areas of the country; encouraging OF and their families to increase

savings and investment; and increasing financial literacy among OF and beneficiaries. The OF surveys of the NSO suggest that the share of remittances channelled through the banking system increased from 46% in 1995 to 56% in 2009, probably as a result of these reforms. As remittances have been growing, the performance of Philippine exports has been comparatively modest as Table 3 shows. The table shows growth rates of exports, measured in dollars. Many East and Southeast Asian countries follow an export-driven development strategy reflected in high growth rates for exports and high ratios of exports to GDP. The Philippines cannot match that success. Exports of goods and services as percentage of GDP stand at 37% in 2008, against much higher ratios for example, Malaysia (110), Thailand (77), and Vietnam (78). The Philippine export ratio had slowly

REMITTANCES AND COMPETITIVENESS: THE CASE OF THE PHILIPPINES

1839

Table 3. Annual export growth of selected Asian countries (%), 2000–2009 Year

Philippines

Malaysia

Indonesia

Thailand

Singapore

Korea

China

Vietnam

9.06 16.16 9.87 2.72 9.78 3.79 15.56 6.42 2.54 22.26 1.62

17.04 10.62 6.14 12.44 20.78 22.16 13.42 13.51 13.21 21.10 8.70

27.64 12.30 3.14 8.36 10.39 22.93 19.00 13.99 18.03 13.82 9.74

19.58 7.09 4.71 18.21 21.64 14.97 17.15 17.05 34.21 13.87 12.66

20.50 10.77 3.05 14.88 24.32 15.45 18.46 10.10 14.26 21.09 8.92

21.22 14.04 7.88 20.73 30.63 12.13 14.84 14.21 14.38 15.78 10.62

27.95 6.80 22.39 34.58 35.39 28.50 27.17 25.63 17.29 15.88 20.98

25.20 4.01 11.17 20.61 31.45 22.51 22.74 21.93 29.09 9.80 17.89

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Average

Source of basic data: BSP 2009 Selected Philippine Economic Indicators.

Table 4. Pairwise Granger Causality Test Direction of Causality. Sample: 1999Q1 2009Q1 Pair of variablesa

REMT and FXR REMTandQSE1P REMT and LF REMIT and REERMA REMT and DEPLOY a b c

Observations quarters

35 35 35 35 35

Lags (in quarters) One

Probability

Two

Probability

Three

Probability

Single Bi-direction Single Single Bi-direction

b

Singleb Bi-directionb Singlec Singleb Bi-directionb

b

Single Single Single Single Bi-direction

b

b c b b

b c b b

b b b b

Variables are in logarithm. At 5% level of significance. At 10% level for significance.

increased over time to reach a peak of 55% in 2000. Since then the ratio has declined to 37%. 4 These two facts: the migration and the consequent doubling of the remittances to GDP ratio since the mid-1990s and the decline in the exports to GDP ratio since 2000 may well be related. It is the aim of this paper to analyze the channels of interaction between migration, remittances, and exports. To get a first insight into the relationships between the variables we run Granger causality tests from 1999 to 2009 on the following indicators: overseas Filipino workers remittances (REMIT), labor force (LF), a proxy for wages, real compensation for non-agriculture workers (QSE1P), number of workers deployed overseas (DEPLOY), nominal peso-dollar exchange rate (FXR), and real effective exchange rate of major trading partners (REERMA). 5 These variables were taken from the BSP database. At 10% level of significance, the (Granger) causation runs from remittances to nominal peso-dollar rate, labor force, real compensation for non-agriculture workers, the number of deployed workers, and real effective exchange rate of major competitor countries. Table 4 also shows that there is bidirectional causality between remittances and nominal compensation for non-agriculture workers and the number of deployed workers. This analysis shows that remittances are an important force in the Philippines with impacts on exchange rate, wages, and employment. Of course the Granger tests conducted here deal only with bi-variate relationships and, in fact, relations are much more complex. The impact of migration and remittances on the real exchange rate, wages, and labor productivity are part of a complex set of interactions in the economy. Existing studies often apply a single-equation regression to test the impacts but that may offer only limited insight in complex adjustment processes. In our approach we use a more complete macroeco-

nomic model that includes the relevant institutions, markets, and agents and the various interactions between them. We believe that this gives a better insight into the impact of migration and remittances. When studying the impact of migration and remittances we have to deal with the issue of endogeneı¨ty. Remittances are part of GDP as they immediately are reflected in expenditure and this leads to a positive correlation that does not mean very much. Most studies use the two-stage least squares (instrument variable) and the GMM approach. In our model, we used the same methods to address the issue of endogeneı¨ty. In the next section, we introduce a full-fledged macroeconomic model in which the main facts around migration and remittances are integrated. 4. REMITTANCES AND COMPETITIVENESS In the second section of this paper we identified various channels along which remittances could affect the competitiveness of the Philippine economy. To test the relevance of these channels and to estimate their importance we use a quarterly macroeconometric model. 6 (a) The nature of the model The underlying analytical structure of our model shares features with the New Keynesian model of Ball (1999). The Ball (1999) model is designed to analyze monetary policy under inflation targeting and assumes that inflation and output are backward-looking, thus it deliberately abstains from any optimizing foundation. Central to this model are important nominal rigidities in describing the macroeconomy. In addition, there are lag effects in the transmission mechanism.

1840

WORLD DEVELOPMENT

We assume that aggregate output is demand-determined in the short to medium run. Goods markets are monopolistically competitive (Blanchard & Kiyotaki, 1987), leading to profits for firms that charge non-competitive sticky prices (Calvo, 1983), which clear all of domestic production to satisfy demand (net of imports) for consumption, investment, government spending, and exports. Firms use a mark-up when setting prices which is responsive to demand and monetary conditions. Meanwhile, households and firms negotiate a non-competitive real wage, engaging in sticky nominal contracts (Calvo, 1983). Asset markets are imperfect. The nominal exchange rate is allowed to transitorily deviate from purchasing power parity (PPP) so that movements occur in the real exchange rate. In addition, the nominal short-term interest rates play the leading role as the instrument of monetary policy, with the money supply having a limited role in describing the monetary stance. The model is designed to trace how the Philippine economy reacts to shocks and to policy interventions. Fiscal policy is introduced in the model through discretionary changes in government spending. The main attention in the model is for monetary policy as this is crucial for the changes in interest rates and the exchange rate which are so central in the analysis of competitiveness. The main features of the monetary block of the model are the following: (1) the policy interest rate of the BSP follows a policy rule that responds to inflation, output gap, and exchange rate pressures; (2) changes in the BSP policy rate affect the nominal exchange rate based on the uncovered interest parity (UIP) condition; and (3) the nominal peso-dollar rate is an effective transmission mechanism, as both direct and indirect passthrough effects to inflation are above average. The innovation of the model used in this paper is that we have explicitly introduced labor supply and remittances as endogenous variables with a number of impacts on the macro economy. As argued above, overseas Filipinos and their remit-

tances are very important and short-term fluctuations in the level of remittances have significant impacts on the economy. Figure 2 provides a schematic overview of the model and of the interactions amongst the variables. The 67 equations are grouped into seven major blocks: monetary sector (bottom left), public sector (bottom right), prices (middle left), expenditures including balance of payments (middle right), production (upper right), and the labor sector (upper left). As noted the model is largely demand driven. In the expenditure block, private consumption expenditures are mainly determined by disposable income. Private investment is driven by output growth, interest cost, and the exchange rate and export demand is determined by foreign income and the real exchange rate. Total domestic expenditure plus the balance of trade in goods and services reflects the aggregate demand in the economy, and is equal to gross domestic product (GDP). Aggregate demand translates into demand for labor. On the labor market this demand is confronted with supply to determine unemployment and wage pressures. Output gap estimation is based on Dakila (2001): it is expressed as the difference between the log of one quarter moving average of GDP (GDP from the supply side, deseasonalized series) and potential output. The output gap then feeds into the wholesale price index. The wholesale price index is affected by the average prices of merchandise imports in pesos, the excess liquidity as indicated by money supply relative to gross domestic product, wages in industry and services, and the output gap. This specification makes the pricing decision based on a flexible mark up. The main link between monetary policy and wholesale price index, and consequently inflation, is the output gap. Hence, there is an impact of monetary policy on expenditure. In addition, the real money supply strengthens the link to price level and consequently between monetary policy and the production sector.

Figure 2. Overseas Filipino remittances and the monetary policy transmission mechanism.

REMITTANCES AND COMPETITIVENESS: THE CASE OF THE PHILIPPINES

Changes in wholesale prices drive prices of the industry and services sectors, and finally the final demand prices. Final demand prices are contained in the implicit GDP deflator. This then is the basis of headline inflation. Because of the forward-looking nature of inflation targeting, the role of inflation expectations becomes crucial. Indicators of inflation expectations include the 2-year ahead inflation forecast. The estimation of long-run inflation expectations follows a hybrid structure that contains both forward-looking and backward-looking expectations. The structure includes rational component of inflation, indicated by the medium-term (3–5 years) inflation target announced by the Government and contemporaneous and inertial components indicated by current and past inflation rates. The rational component is based on Demertzis’ and Viegi’s (2005) work on inflation targets as focal points for long run inflation expectations. The idea is that in the absence of concrete information of inflation expectations, the only information that agents have is the quantitative inflation target announced by the Government. (b) Migration, remittances, and the economy All the interactions described so far are rather standard for this type of model. The innovation of our model is that we have made remittances endogenous and have traced the channels along which migration and remittances affect the economy. The labor market part of the model is relatively simple. The migration decision is exogenous: Filipinos move overseas and this reduces the domestic labor supply. The labor force is determined by demographic trends (population of working age) and the level of wages. As noted above, every year a significant number of people leaves the country reducing labor supply. The reduction of the labor supply may be concentrated in some segments of the labor market, for example, migration may lead to shortages of skilled or semi-skilled labor but our model does not present this segmentation and just focuses on the average wage level. In another paper (Bayangos & Jansen, 2009) we explored the dynamics of remittances in the Philippines and we concluded that remittances are pro-cyclical: remittances are positively correlated with (lagged) home GDP, primarily with the consumption component of GDP, and positively correlated with the (lagged) GDP of major host economies. Rt ¼ a þ dC t þ bðrt  rft Þ þ hY ft þ et :

ð1Þ

Remittances are positively related with consumption C, indicating that remittances do not stabilize consumption as found in some other studies. 7 We look at interest rate differential between local and international rates (r  rf) to determine whether investment considerations are at play. In addition, we look at the income of host countries Yf to capture the cyclicality of remittances with the income of host countries: remittances vary with the business cycle of the host countries; in good times, employment opportunities and wages are better allowing migrant workers to transfer more. Y dt ¼ ð1  tÞY t þ Rt :

ð2Þ

Following Chami et al. (2003), Eqn. (2) shows that remittances R add to disposable income Yd and, through this, to real private consumption expenditure Ct in Eqn. (3). Disposable income is computed as income less tax. C t ¼ a þ dY dt þ kM t  hðrdt  pet Þ þ et :

ð3Þ

1841

Consumption demand is also determined by the money supply (M) and the real interest rate ðrdt  pet Þ. Part of the increased consumption demand is for imported goods, and affects the trade balance, while the rest is on domestic goods and may push up their prices. The impact of an increase in remittances on investment demand is complex. The increased demand for domestic goods would lead to an increase in investment demand but remittances also affect variables like wages, interest rates, and exchange rate that also play a role in the investment decision. The increase in wages induced by remittances will reduce profitability and thus reduce investment demand. Remittances also affect market interest rates and thus the cost of investment. The appreciation of the exchange rate will make imported capital goods cheaper, which would push investment demand. The net outcome of all these factors is difficult to predict but in our model runs the net effect of an increase in remittances is a rise of investment. It is sometimes suggested that remittances may be used to finance residential construction but we did not find econometric evidence for this effect for the Philippines. The impact of remittances is seen as directly affecting monetary conditions: an increase in remittances is reflected in an increase in deposit liabilities Dt in Eqn. (4). Dt ¼ a þ dY t þ kRt þ et :

ð4Þ

Deposits are also driven by output Y. Total domestic liquidity is determined by adding deposit liabilities and currency in circulation. Thus, any change in remittances will have an impact on the money supply. As noted above, migration reduces the labor force. But there is a further labor market effect. As suggested by the studies reviewed in Section 2, remittances receiving households may use the higher income to opt for more leisure or more schooling, which will further reduce the labor supply. Our regression results show that variations in remittances have a significant effect on the labor force. Lt ¼ a þ -A  dRt þ kW t þ et :

ð5Þ

Eqn. (5) shows that an increase in remittances R will have a negative effect on labor force supply L. In addition, working age population (ages 15 years old and above) A and wages W will have positive impact on the labor force. The increase in remittances is thus associated with an increase in aggregate demand and thus also of demand for labor but with a decline in the labor supply, a fall in unemployment and, since the level of unemployment is an important determinant of wages, an increase in wages. Wages feed into production cost and sectoral production prices and through these to wholesale and retail price inflation. In developing countries like the Philippines there may be an excess supply of labor and it could be argued that people migrate because there is not much to do at home. If that is so, the reduction of the labor force need not increase wages. However, there are three arguments against this. First, as there is no social security unemployment is not an option: every person has to do something to make a living, even if only a low productivity informal sector job. They migrate not because there is nothing to do but because they can earn more abroad. Moreover, as indicated earlier, emigration from the Philippines is quite substantial relative to the labor force. And, thirdly, emigration increasingly concentrates on skilled workers and does create shortages in these segments of the labor market. There is thus good reason to expect an impact of migration on wages and that is what we found in our econometric analysis. As argued in Section 2 the rise in wages may

1842

WORLD DEVELOPMENT

be compensated by increased productivity due to better use of labor, investment in labor-saving technology, and improved skill levels. The Dutch disease effect is also captured by the model. ent ¼ a  bT þ dðrft  rt Þ þ et :

ð6Þ

The estimation of Eqn. (6) follows the convention in which an increase in the nominal peso-dollar rate en corresponds to depreciation of the peso. The BSP maintains a floating peso, whose value is determined, to a great extent, by supply and demand factors. The nominal exchange rate is sensitive to the level of the current account balance T and to the interest rate differential ðrft  rt Þ. The direct effect of an increase in remittances on the current account is positive leading to an appreciation, but there are also indirect effects on exports (through the impact of remittances on competitiveness) and on imports (through the increase in domestic demand that remittances generate) and the overall impact of an increase in remittances could be a deterioration of the current account balance which would lead to a depreciation of the nominal exchange rate. There is also an indirect effect through the financial markets. The remittances inflow increases liquidity on financial markets, which exerts downward pressure on market interest rates. But the remittances also lead to an increase in spending, and thus a fall in the output gap, and in an increase in inflation. These changes induce the central bank to increase the policy rate and this pushes up the market interest rate. The net effect is an empirical matter. If, on balance, the market rate rises the gap between local and global interest rates gap rises and the nominal exchange rate appreciates. The change in the real exchange rate is determined by the change in the nominal exchange rate and the change in relative prices. As the increase in remittances lead to higher demand pressures, local prices increase, bringing an appreciation of the real exchange rate. We can now see how remittances complicate policy making, in particular monetary policy. The BSP uses the overnight reverse repurchase rate (RRP) rp as its policy instrument and follows a policy rule to anchor inflation in the long run (Clarida, Gali, & Gertler, 2000). The overnight RRP adjusts to inflationary pressure measured by the difference between the inflation forecast and the inflation target announced by the Government, the output gap and the expected exchange rate gap (the difference between the expected exchange rate and realized exchange rate). This is seen as, rpt ¼ a þ bðpft  pt Þ þ qðqt  qt Þ þ ðEt entþ1  ent Þ þ et ;

ð7Þ

where rp is the RRP, a connotes the neutral monetary policy stance, pf is the one-quarter ahead inflation forecast, p* is the medium-term inflation target announced by the Government, q is the real output, q* is the potential real output, Et entþ1 is the expected nominal peso-dollar rate at time t, ent is the realized nominal peso-dollar rate and the error term is et. The RRP rate is transmitted to the benchmark interest rate, r, through the natural arbitrage condition. In this model, the benchmark interest rate is the 91-day Treasury bill rate. rt ¼ a þ brpt þ qpt þ crft  #M t þ et ;

ð8Þ

Eqn. (8) states that the benchmark interest rate is higher, the higher the RRP rate, the higher the inflation rate, the higher the foreign interest rate, and the lower the level of money supply. In this equation, there is a direct channel from the BSP’s policy rate to the benchmark interest rate. Remittances affect the benchmark rate as they affect liquidity on financial markets and thus the money supply.

Changes in the benchmark interest rate are then carried over to the changes in the other market interest rates, such as savings and lending rates through the natural arbitrage condition. It is also assumed that in the short-run domestic inflation is relatively sticky, indicating that inflation expectations for the short term are similarly sticky. This implies that by controlling the nominal overnight RRP rate, the BSP can also affect the short-term real RRP rate. Through market expectations of future real rates, longer real rates (that is, longer than overnight rates) also are affected. Thus, the lowering of the overnight RRP is expected to lower short and longer real interest rates, and consequently affect economic activity. We can conclude that the endogenous remittances lead to a complex adjustment process and set considerable policy challenges. An increase in remittances pushes up private consumption, has an ambiguous effect on investment, and a negative effect on exports. On balance, aggregate demand increases and the output gap narrows. The narrowing output gap generates inflationary pressures which are aggravated by the impact of migration and remittances on wages and the impact on the money supply but are mitigated by the impact on the interest rates, which reduces interest cost, and on the exchange rate, which reduces the cost of imports. But again, on balance the inflationary pressures increase. The narrowing of the output gap and the increase in inflationary pressures force the central bank to increase the policy rate and the increase in the rate needs to be more than it would have been in the absence of remittances because of the downward impact the increasing remittances have on the market interest rates. On the other hand, the fact that remittances lead to an appreciation of the real exchange rate and a loss of competitiveness may make the central bank reluctant to raise the policy rate. The higher interest rates attract more remittances (Eqn. (1)) and a further appreciation of the exchange rate (Eqn. (6)). A central bank trying to stabilize the output gap, inflation, and the exchange rate will find it difficult to achieve all the objectives when remittances are substantial. (c) Structure of the model, diagnostics, and model solution 8 The analytical framework developed in the previous section yields a system of simultaneous equations. The model consists of 67 equations, with 29 simultaneous equations. There are 32 recursive equations largely estimated using ordinary least squares, and the remaining 6 are identities. The 29 simultaneous equations are estimated using single-equation methods: four are estimated using generalized method of moments, 13 are estimated using two-stage least squares and the remaining 12 equations are estimated using ordinary least squares. The choice of instruments for the two-stage least squares is assumed to be all the lagged endogenous variables and all current and lagged exogenous variables in the whole system. These equations are largely overidentified, while the rest are identified. It is argued that there is nothing wrong with overidentified equations since the statistical fit is never perfect anyway (Greene, 2003). Each of the 29 simultaneous equations is assessed for basic and higher-order diagnostic tests. The signs and magnitudes of individual coefficients in each equation, such as t statistics, the adjusted R2, Durbin Watson, and F statistics are all examined. In general, all of the behavioral equations pass these tests. In particular, the adjusted R2 values for all equations are greater than 60% and values in all equations suggest there is no penalty for the number of explanatory variables used. All calcu-

REMITTANCES AND COMPETITIVENESS: THE CASE OF THE PHILIPPINES

lated F values are higher than the critical values, at the 5–10% level of significance, thereby indicating a significant degree of reliability of coefficients of determination. 9 Results of higher order test statistics of residuals are similarly examined. Higher order diagnostic tests start with the Jarque-Bera test. This test is designed to ascertain whether the series is normally distributed. Results show that all of the series are normally distributed. With a lag order of up two and at a 5–10% level of significance, Breusch-Godfrey results show that not all equations exhibit serial correlation. There are equations which initially exhibit serial correlation but for which additional lags are incorporated to make the residuals stationary. White’s heteroskedasticity test in the residuals is also used. White’s test is a test of the null hypothesis of no heteroskedasticity. Using a 5–10% level of significance and up to two fitted items, RESET results reveal that there are no specification errors in equations. J-test is also checked for four equations estimated using GMM at 5–10% level of significance. Results show that the four equations are over-identified, but are valid equations. Solving a system simultaneously is indeed difficult. Both deterministic and static simulations are performed using the Fair-Taylor method. 10 This is an iterative algorithm, where each equation in the model is solved for the value of its associated endogenous variable, treating all other endogenous variables as fixed. Meanwhile, terminal conditions are assumed to hold in a specified time period. Put simply, this means that the values contained in the actual series after the end of the forecast sample are used as fixed terminal values. Forward solution is similarly used for equations that contain future (forward) values of the endogenous variables. To gauge the simulation and forecasting performance of the model, the mean absolute percent error (MAPE) of selected endogenous variables is computed. As a general rule, the smaller the MAPE the better the fit of the model to the actual data. MAPE (which is unit free) is computed as follows:  X  1 P A  100; ð9Þ MAPE ¼ n A where A refers to the actual value, P is predicted or simulated by the model and n is the number of periods covered by the simulation. The model’s forecasting performance over parts of the sample period and the simulated response to some exogenous changes in policy variables are assessed. The simulation period extends from the first quarter of 1999 to the fourth quarter of 2003. In our model, the major macroeconomic variables can be predicted within reasonable error margins. Using generalized method of moments, two-stage least squares and ordinary least squares, about 89% of the MAPEs of our dynamic models fall below 10%. These include key variables in the monetary and real sectors, like the consumer price index (CPI2000), remittances (REMIT), labor force (LF), 91-day Treasury bill (TBR91), and the nominal peso-dollar rate (FXR). For instance, CPI2000, LF, TBR91, REMIT, and FXR have a MAPE of, respectively, 0.91%, 1.43%, 5.12%, 5.15% and 8.11%. These results indicate that the simulation properties of the model are reasonable. 5. REMITTANCES AND COMPETITIVENESS: MODEL SIMULATIONS The strategy we follow to assess the impact of remittances on competitiveness is straightforward. We simulated a sus-

1843

tained one billion US$ increase in remittances on the estimated macro model from first quarter 1999 to fourth quarter 2003. Table 2 shows that this is a substantial increase; it raises remittances inflows in these years by about 15%. Of course, remittances are endogenous in our model; we generated the increase in remittances by increasing the (exogenous) income of host countries (see Eqn. (1) above). Annualized quarterly growth as well as volatility using the coefficient of variation (CV) is computed. Volatility is a measure of how wild or quiet an indicator is relative to its history. The CV is a comparative measure defined as the ratio of the standard deviation to the mean. Table 5 presents the results of the simulation. The simulation shows that the increase in remittances leads to a decline in export growth: there is thus an apparent loss of competitiveness. The simulation results show the various channels through which this takes place. First, labor force growth declines as workers emigrate and reduce work effort. This has a positive effect on unemployment but leads to a growth in wages (the non-agricultural compensation index). There is a significant appreciation of the nominal exchange rate. This is caused by the inflow of foreign exchange but also by the interest rate effect. The inflow of foreign exchange does, in the first instance, increase liquidity on the money market and this places a downward pressure on the interest rate. But there is a monetary policy effect as well: the increase in remittances stimulates aggregate demand. As a result the output gap falls and inflation accelerates. The central bank is following a Taylor policy rule and thus responds by increasing the policy rate and this pushes up the market interest rates. As Table 5 shows this policy effect is stronger than the liquid-

Table 5. Impact scenario: A sustained one billion US$ increase in overseas remittances Economic indicators

Percent change from baseline model 1999–2003 Average

CV

2.56 1.28 0.13

0.39 0.22 0.34

Total exports (growth) Merchandize exports of goods (growth) Non-merchandize exports of goods (growth)

1.04 1.01

0.42 0.47

Labor sector indicators (%) Labor force (growth) Non-agriculture compensation index (growth) Unemployment (growth)

0.33 1.68 0.70

0.06 0.32 0.31

Financial indicators (%) Money supply (year-on-year growth) RRP (%) 91-day treasury bill rate (%) Nominal peso-dollar rate (growth)

1.17 0.12 0.09 0.75

0.41 0.17 0.13 0.56

Macroeconomic indicators (%) Real GDP (growth) Output gap (growth) CPI-inflation CPI-inflation forecast (2 years ahead) CPI-inflation expectations (long run)

0.82 0.25 0.12 0.04 0.03

0.21 1.59 0.21 0.08 0.05

GDP components Personal consumption (growth) of which: disposable income (growth) Gross capital formation (growth)

1844

WORLD DEVELOPMENT

ity effect and the market rate rises. As this widens the interest rate gap with the international interest rate this leads to a further appreciation of the nominal exchange rate. The increase in remittances translates into a significantly faster growth of private consumption expenditure, which is also fuelled by the increasing wages. In Section 2 it was noted that rising wages imply a lower return on capital and thus a reduced investment demand. The increase in the market interest rates will further discourage investment. Still, in the simulation investment grows. This is due to the aggregate demand effect. The inflow of remittances and the resulting higher domestic spending invites capital formation to create the necessary production capacity. The appreciation of the exchange rate makes (imported) capital goods cheaper which also helps to increase investment. The faster growth of consumption and investment demand exerts pressure on prices; there is a slight acceleration of inflation. The acceleration of inflation means that the real exchange rate appreciates more than the nominal one. We can bring these various channels through which remittances affect competitiveness together in the unit labor cost. ULC ¼

WL 1 W 1  ¼  VA=P e a=P e

ð10Þ

Unit labor cost (ULC) in US dollar is the product of the nominal wage rate (W) and the (inverse of) physical labor productivity (employment, L, over the volume of output, Value Added (VA) deflated by the value added deflator, P), and the exchange rate, e, the number of local currency units per one dollar. The growth of unit labor cost is then the sum of: gULC ¼ gw  ga þ gP  ge

Our estimate of the Dutch disease effect is comparable to those found in the literature. As noted in Section 2, Bourdet and Falck (2006) found that a 10% increase in remittances led to a real exchange rate appreciation of 1.2%. AmuedoDorantes and Pozo (2004) estimated that a doubling of remittances led to an appreciation of 23% and Lartey et al. (2008) found that a 1% point increase in the remittances to GDP ratio (which at the sample averages implies an increase in remittances of about 60%) leads to an appreciation of between 20% and 40%. In our results an increase in remittances of about 15% leads to an appreciation of the real exchange rate by almost 1%. Acosta et al. (2009) use a different type of model than we do here but they cover the same channels and come to similar conclusions for the impact of remittances on the El Salvadorian economy: an increase in remittances leads to an appreciation of the real exchange rate but also to a fall in the labor supply and an increase in wages, leading to a decline in exports. In the literature on remittances and Dutch disease it is suggested that the impact of remittances on the real exchange rate is the main channel by which remittances can undermine export performance. Our results show, however, that the labor market effects are as important.

ð11Þ

where gi is the growth rate of wages, labor productivity, prices, and the exchange rate, respectively. 11 This equation allows us to assess the relative importance of the various channels linking remittances to competitiveness. The first two terms (gw  ga) capture the change in labor cost: migration and remittances increase wages and affect labor productivity. The second term (gp  ge) captures the real exchange rate effect. The increase in prices reduces competitiveness and the change in the nominal exchange rate will add to this: a depreciation would compensate for the price increase but an appreciation would further erode competitiveness. Table 5 shows that the increase in migration and remittances results in a rise in nominal (and real, nominal corrected for inflation) wages and a decline in the labor force. Overall labor productivity is rising (measured as the difference between the real output growth and labor force growth). As noted in Section 4, output is demand determined and as remittances stimulate aggregate demand output rises despite the fall in labor force growth. The nominal exchange rate is appreciating and, as inflation accelerates, the real exchange rate appreciates even more. Using the numbers of Table 5, we see that gULC ¼ ðgw  ga Þ þ ðgP  ge Þ ¼ ð1:68  1:15Þ þ ð0:12  ð0:75ÞÞ ¼ 0:53 þ 0:87 ¼ 1:40: The unit labor cost increases significantly as a result of the increase in remittances. There is a significant increase in the wage level which is not fully compensated by the increase in labor productivity. The nominal exchange rate appreciation is again quite strong. The real exchange rate appreciates a bit more than the nominal exchange rate as inflation accelerates.

6. CONCLUSION AND POLICY IMPLICATIONS Remittances have a positive impact on many economic indicators. The simulation of an increase in remittances inflows in this paper shows that it will increase consumption, investment, labor productivity, and economic growth. But it also leads to a change in the economic structure, in particular a decline in traded goods production and exports and this implies that the dependence on remittances increases. We have been able to trace the channels along which migration and remittances affect the competitiveness of the economy. Our findings on the impact on the nominal and real exchange rate confirm the findings of other studies but our results show that the labor market effects are also highly significant. Emigration cuts into the labor force and the receipt of remittances further reduces labor supply. There is a strong effect on wages. The impact of the higher wages on competitiveness is mitigated by the increase in labor productivity. As the labor force falls and output rises there is a more intensive use of labor through a decline in unemployment and underemployment. In the literature it is suggested that migration and remittances may also induce an increase in the investment in human capital—which could also increase productivity—but our model does not include these investment and so we have not been able to trace this effect. It is clear that it is important to include the labor market channel in the analysis of competitiveness. In our simulation the unit labor cost increase by 1.4% and about 40% of that increase is due to the labor market effects. Studies that exclusively focus on the Dutch disease channel thus miss an important other channel affecting the export performance. It should be noted that a great part of the increase in unit labor cost is due to the appreciation of the nominal exchange rate and that this increase is partly driven by the monetary policy response. The output gap is falling and inflation is accelerating and the policy response of the central bank reinforces the appreciating impact of remittances. Remittances have many positive effects for the Philippines. The loss of competitiveness is, however, a drawback with possible long term negative effects. It explains the comparatively poor export performance and may, in the longer term, limit the diversification of the economy and productivity growth.

REMITTANCES AND COMPETITIVENESS: THE CASE OF THE PHILIPPINES

It is, therefore, understandable that authorities are looking for ways of mitigating this effect. Tax policy could be used to soften the impact of rising wage cost but the scope for reducing payroll taxes is limited in the Philippines: personal income tax revenue accounts for only 1.5% of GDP. Monetary policy makers should consider the exchange rate effects when setting the policy interest rate. The increase in remittances narrows the output gap and this pushes the inflation up. The central bank reacts by increasing the policy rate but the increase in interest rates attracts more remittances and other capital inflows and results in a further appreciation of the nominal exchange rate. At present, the BSP also looks at the exchange rate movements when setting the policy rate but, it claims, only to smoothen exchange rate shocks and not to target the exchange rate. If the central bank decides to target the exchange rate to mitigate the appreciation, it would increase the policy rate by

1845

less in response to an increase in remittances. This may help to reduce the appreciation of the nominal exchange rate but it will also keep inflation at a higher level and so appreciate the real exchange rate. An alternative strategy for the central bank would be to intervene in the market to buy up foreign exchange and so regulate the nominal exchange rate. This would lead to an accumulation of international reserves and an expansion of the money supply. Usually central banks sterilize these money supply effects to control inflation but such an approach tends to be quite costly as the earnings on the international reserves are less than the cost of sterilization. Standard macroeconomic policy may thus be rather ineffective in dealing with the competitiveness problem. There is a need for more structural policies to enhance productivity, including investment in infrastructure and education, reforms to increase competition on domestic markets, and export promotion.

NOTES 1. In the 2008/09 ranking the Philippines is ranked 71 out of 134 countries, below regional neighbours like Korea, Malaysia, China, Thailand, Indonesia, and Vietnam. (see WEF, 2008). 2. We have decided to use the BSP data on remittances in this paper and in our model. There is alternative series put together by the World Bank, based on Balance of Payments data but using different methodology from the BSP (World Bank: World Development Indicators). The estimates in the World Bank series tend to between 10% and 20% higher than the BSP data but the trend over time is the same. 3. Results of the survey can be seen at http://www.census.gov.ph/data/ sectordata/datasof.html 4. The data on the export ratios are drawn from the online version of the World Bank’s World Development Indicators. 5. The REER with Philippine major trading partners series (with 1980 as base year) is reported by BSP. This series is computed by deflating the nominal effective exchange rate index (NEERI) using the consumer price index of the major trading partners weighted by the total trade (export plus import) of the Philippines to these countries. Major trading partners of the Philippines include the United States, Japan, European Union (EMU), and the United Kingdom. If REER goes up, this signals a decrease in its value or an appreciation in the real exchange rate making Philippine exports less price-competitive. On the other hand, if REER

goes down, this signals a depreciation in the real exchange rate which makes Philippine exports relatively more price-competitive. 6. Our study builds on Bayangos (2007) and Bayangos and Jansen (2009) dynamic, structural and quarterly macroeconometric model for the Philippines. Our dataset covers the period from March 1999 to June 2009. 7. The literature on the cyclicality of remittances flows is inconclusive with some studies finding evidence for countercyclical patterns and others for procyclicality (see Bayangos and Jansen (2009)). 8. The complete specification of the basic version of the model is found in Bayangos (2007), Chapter 5. For this paper the model has been extended to include the various channels along which remittances affect competitiveness. 9. Exact collinearity is similarly checked. Highly collinear regressors lead to spurious estimates. There are a few cases though where exact collinearity is encountered especially when dummy variables are used, however, a re-specification of some of these equations are done. 10. In technical terms, this is called the Gauss-Seidel algorithm method. 11. The nominal exchange rate is defined as the number of local currency units per USD. Growth of the nominal exchange rate is thus a depreciation of the currency and a decline an appreciation.

REFERENCES Acosta, Pablo. A., Lartey, Emmanuel. K. K., & Mandelman, Federico. S. (2009). Remittances and the Dutch disease. Journal of International Economics, 79, 102–116. Amuedo-Dorantes, C., & Pozo, S. (2004). Workers’ remittances and the real exchange rate: A paradox of gifts. World Development, 32(8), 1407–1417. Ball, Laurence N. (1999). Policy rules for open economies. In J. B. Taylor (Ed.), Monetary policy rules (pp. 127–156). Chicago: The University of Chicago Press. Bank, World. (2006). Global economic prospects; Economic implications of remittances and migration. World Bank: Washington DC. Bank, World. (2009). Global development finance. World Bank: Washington D.C. Bayangos, V. (2007). Inflation targeting and exchange rate uncertainty. Maastricht: Shaker Publishing.

Bayangos, V. & K. Jansen (2009). The Macroeconomics of Remittances In the Philippines, Working Paper nr 470, Institute of Social Studies, The Hague. Beine Docquier, Michel Fre´deric, & Rapoport, Hillel (2008). Brain drain and human capital formation in developing countries: Winners and losers. The Economic Journal, 118, 631–652 (April). Blanchard, Olivier. J., & Kiyotaki, N. (1987). Monopolistic competition and the effects of aggregate demand. American Economic Review, 77, 647–666. Bourdet, Y., & Falck, H. (2006). Emigrants’ remittances and Dutch disease in cape verde. International Economic Journal, 20(3), 267–284. Calero Carla, Arjun S. Bedi and Robert Sparrow (2008). Remittances, Liquidity Constraints and Human Capital Investments in Ecuador, The Hague, Institute of Social Studies, Working Paper nr 458. Calvo, Guillermo (1983). Staggered prices in a utility maximizing framework. Journal of Monetary Economics, 12(3), 383–398.

1846

WORLD DEVELOPMENT

Chami, R, C. Fullenkamp & S. Jahjah (2003). Are Immigrant Remittances Flow a Source of Capital for Development?, Washington DC: IMF Working Paper 03/189. Clarida, R., Gali, J., & Gertler, M. (2000). Monetary policy rules and macroeconomic stability: Evidence and some theory. Quarterly Journal of Economics, 115, 147–180. Dakila, Francisco Jr. (2001). Alternative Monetary Policy Rules for the Philippines, Philippine Review of Economics, Vol. XXXVIII, No. 2, December. Demertzis, Maria and Nicola Viegi (2005). Inflation Targets as Focal Points, De Nederlandsche Bank (DNB) Working Paper, No. 17. Amsterdam: De Nederlandsche Bank. Fagerberg, J. (1988). International competitiveness. The Economic Journal, 98(391), 355–374 (Jun., 1988). Fajnzylber, P., & Lopez, J. H. (Eds.) (2008). Remittances and development, lessons from Latin America. Washington DC: World Bank. Fagerberg, J, Mark Knell & Martin Srholec (2004). The Competitiveness of Nations: Economic Growth in the ECE Region, Oslo: Centre for Technology, Innovation and Culture, University of Oslo, Working Papers on Innovation Studies nr 20040223. Farrent, M., A. MacDonald and D. Sriskandarajah (2006). Migration and Development: Opportunities and Challenges for Policymakers, Migration Research Series, no 22, International Organization for Migration (IOM). Greene, William (2003). Econometric analysis (5th ed.). Prentice Hall. Lartey Emmanuel K.K. & Federico S. Mandelman, & Pablo A. Acosta (2008). Remittances, Exchange Rate Regimes, and the Dutch disease: A Panel Data Analysis. Atlanta: Federal Reserve Bank of Atlanta, Working Paper 2008-12. Lopez, J. H., Molina, L., & Bussolo, M. (2008). Remittances, the real exchange rate, and the Dutch disease phenomenon. In P. Fajnzylber, & J. H. Lopez (Eds.), Remittances and development, lessons from Latin America. Washington DC: World Bank.

Loser, C., C. Lockwood, A. Minson and L. Balcazar (2006). The macroeconomic impact of remittances in Latin-America—Dutch disease or Latin cure?. Inter-American Dialogue. McCormick, B., & Wahba, J. (2000). Overseas employment and remittances to a dual economy. The Economic Journal, 110, 509–534. Mishra, Prachi (2006). Emigration and Wages in Source Countries: Evidence from Mexico. IMF Working Paper WP/06/86. Rapoport, H. and F. Docquier (2005). The Economics of Migrants’ Remittances, IZA Discussion Paper No. 1531, Institute for the Study of Labor (IZA), Bonn. Tuan˜o-Amador, Ma. N. Cyd, Claveria, Racquel A., Co, Ferdinand S., & Delloro, Vic K. (2007). Philippine overseas workers and migrants remittances: The Dutch disease question and the cyclicality issue. Bangko Sentral Review, XI(1), 1–23. UNDP (2009). Human development report 2009: Overcoming barriers: Human mobility and development. New York: UNDP. Vargas-Silva, C. (2009). The tale of three amigos: Remittances, exchange rates, and money demand in Mexico. Review of Development Economics, 13(1), 1–14. Vos, R. P. (1992). Foreign private asset accumulation, not just capital flight: Evidence from the Philippines. Journal of Development Studies, 28(3), 500–537. Wahba, J. (1998). The transmission of Dutch disease and labour migration. The Journal of International Trade & Economic Development, 7(3), 355–365. World Economic Forum (2008). Global Competitiveness Report, WEF, Geneva. Yabuuchi, Shigemi, & Chaudhuri, Sarbajit (2007). International migration of labor and skilled–unskilled wage inequality in a developing economy. Economic Modelling, 24, 128–137. Yang, Dean (2008). International migration remittances and household investment: Evidence from Philippine migrants’ exchange Rate shocks. The Economic Journal, 118, 591–630 (April).