Working Paper The wage structure of overseas Filipino workers. Discussion paper // School of Economics, University of the Philippines, No.

Transcription

1 econstor Der Open-Access-Publikationsserver der ZBW Leibniz-Informationszentrum Wirtscaft Te Open Access Publication Server of te ZBW Leibniz Information Centre for Economics Tan, Edita A. Working Paper Te wage structure of overseas Filipino workers Discussion paper // Scool of Economics, University of te Pilippines, No. 25,3 Provided in Cooperation wit: University of te Pilippines Scool of Economics (UPSE) Suggested Citation: Tan, Edita A. (25) : Te wage structure of overseas Filipino workers, Discussion paper // Scool of Economics, University of te Pilippines, No. 25,3 Tis Version is available at: ttp://dl.andle.net/1419/46635 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenscaftlicen Zwecken und zum Privatgebrauc gespeicert und kopiert werden. Sie dürfen die Dokumente nict für öffentlice oder kommerzielle Zwecke vervielfältigen, öffentlic ausstellen, öffentlic zugänglic macen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt aben sollten, gelten abweicend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewärten Nutzungsrecte. Terms of use: Documents in EconStor may be saved and copied for your personal and scolarly purposes. You are not to copy documents for public or commercial purposes, to exibit te documents publicly, to make tem publicly available on te internet, or to distribute or oterwise use te documents in public. If te documents ave been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise furter usage rigts as specified in te indicated licence. zbw Leibniz-Informationszentrum Wirtscaft Leibniz Information Centre for Economics

2 UNIVERSITY OF THE PHILIPPINES SCHOOL OF ECONOMICS Discussion Paper No. 53 Marc 25 Te Wage Structure of Overseas Filipino Workers by Edita A. Tan* *Professor Emeritus, Scool of Economics University of te Pilippines Note: UPSE Discussion Papers are preliminary versions circulated privately to elicit critical comments. Tey are protected by te Copyrigt Law (PD No. 49) and not for quotation or reprinting witout prior approval.

3 Abstract. Labor-importing countries adopt differing immigration policy on foreign workers. Tey all restrict te number allowed entry and many set wage ceilings at levels below te wage paid te native workers. Te differing restrictive immigration policies result in te segmentation of te world labor market and large inequality in wage rates for any one occupation or skill. Te paper presents a simple model of wage and employment determination for observable variants of immigration policy followed by te major destinations of OFWs, in particular Nort America, Asian tiger economies and te Middle East. It estimated wage functions by regression using individual overseas workers data wit uman capital (scooling and experience), sex, occupation and destination as arguments. Excepting for completed college, scooling is found not to be a significant explanatory variable. Te oter variables exerts teir expected influence in a significant way but destination proves to ave te strongest influence on foreign wage. Te paper concludes tat would-be OFWs are unable to coose te best destination partly because of immigration restrictions and partly because tey do not meet te qualifications required. Examples are and nursing occupations. Tere is a proliferation of academic programs for tese occupations but teir generally poor quality disables te country from responding to te foreign demand for tese skills.

4 Te Wage Structure of Overseas Filipino Workers Introduction Edita A. Tan * Close to eigt million Filipinos resided abroad in 23, 2.87 million as permanent emigrants and 4.9 million as migrant workers, popularly referred to as overseas Filipino workers or OFW. (Appendix Table 1) Te two groups comprised about 9% of te population. Te permanent migrants ave settled mainly in te US and Canada wic ave absorbed 2.34 Million wit te rest in oter advanced economies suc as Japan, Australia and Europe. Te OFWs were employed in varied occupations and in varied destinations encompassing all te world s continents. Tey formed two big groups, te land-based and te seamen wo respectively numbered 4.68 Million and.216 Million. Tis large stock of Filipinos overseas resulted from an accumulation of continuous emigration since te 193s for eiter permanent settlement or for temporary employment tat ad been extended for several years or even canged to permanent status. In te beginning of te last Century, agricultural workers left for Hawaiian and California plantations. In te immediate post-wwii, engineers and oter skilled labor found employment in American military bases in te Asia Pacific region suc as Guam and Okinawa. Te liberalization of te US immigration policy in 1965 as allowed a regular, albeit restricted, entry to America for family reunification and employment in selected occupations. Subsequently Canada and Oceana began to allow entry of a relatively small number of skilled Filipino workers. Te outflow to tese destinations as been restricted to only about 5, per year in te past decade. It comprised a small fraction of te total outflow, largely of migrant labor, wic as reaced 868, in 23. Large scale labor migration of labor began in te mid 197s wit te opening of te Middle East market. (Appendix Table 2) From 36, in 1975, te number of workers departing eac year continually increased as tey found employment in more varied occupations and in more varied destinations. Wen te construction boom in te Middle East started to slow down, jobs were obtained in te East Asian tiger economies wose labor markets were experiencing tigtening labor markets. Currently tey are reported to be scattered in some 19 country destinations. Te Middle East as, owever, remained a most important destination. In 21, 34.4% of te outflow was for te Middle East, 32.3% for Asia and 23.5% for various international sipping companies. Only 2.% was for America and 5.1% for Europe. Note tat te stock of OFWs is close to 4 times te yearly outflow since a large proportion of tem ave been able to renew teir sort-term contracts, usually two years, once or more times. In fact over te 2-23 period, less tan 4% of departing OFWs were new ires, te rest were reires or returning to teir foreign jobs. Our survey sows tat on average, te latter ave worked abroad for an average of 7 years. Bagasao, et al. (24) found te * Professor Emeritus, UP Scool of Economics. Two people collaborated in te researc: Ms. Helen Barayuga, Director of POEA Electronics Data Processing Office, developed te sampling design and supervised te surveys and encoding and Ferdinand Co, Information Tecnology Officer of te UP Scool of Economics, elped process te raw file and run te regressions. Gloria Lambino rendered er usual excellent secretarial assistance. PCED provided financial support especially for te surveys and data processing. 1

5 average stay abroad of teir sampled OFWs wo were ere for te Cristmas olidays in 23 was about 5 years. Filipinos enjoy freedom to migrate internally and externally. Labor migration is essentially a family decision aimed at maximizing expected economic benefits or returns. Some economists (Stark 1991) argued tat labor migration migt be partly a riskavoidance strategy as families try to diversify te sources and locations of teir income. Migration, owever, entails its own risk as evidenced by te experience of OFWs. According to te Pilippine Overseas Employment Administration, POEA wic keeps track of problems encountered by OFWs, tere were significant occurrences of contract violations, fraudulent job placement, pysical violence and poor working conditions. (Tan, 22, 24a). Migration may also be a means of diversifying investment. Generally, it entails substantial costs for job placement, transport and various departure fees. Many migrant families are reported to ave sold assets and/or borrowed to finance migration cost. Te wole family bears te psycic cost of separation and sares in any problem encountered by its migrant member. Foreign wage, working condition and quality of social and pysical environment for eac occupation differ between destinations, not necessarily in line wit competitive compensating wage differential for work azard. Foreign wage may be negatively related to work azards and positively to quality of work and living environment. Te US, for example, offers te igest wage and te best working and social conditions wile te reverse olds for some Middle East destinations. Te structure of foreign wage and working conditions are taken to be te result of foreign labor market segmentation tat arises from restrictive immigration policies of labor-importing countries. Countries adopt different forms of restrictions on foreign workers tat act as barriers to labor mobility and competitive compensation setting. Eac government decides on te number of foreign workers of specific skills to be allowed entry, teir wage and weter or not to extend to tem te guarantees of uman rigts and te protection of domestic laws. Only a andful of labor-importing countries ave ratified te ILO convention on migrant labor. Immigration barriers tend to give labor-importing countries some monopsony power over foreign workers. Te power goes beyond tat created by geograpic barriers tat isolate small distantly located labor markets, e.g. te market for nurses. (Sullivan 1989, Hirsc and Scumacer 1995). Te paper presents a simple model illustrating te monopsony impact of restrictive immigration policies on employment and wage rate. Variation in immigration policy leads to variation in wage rate paid to a given skill category and to net returns to migration among migrants. Two surveys were conducted to obtain data on wage rates of land-based OFWs, one on newly ired OFWs or new ires, anoter on OFWs wo were returning to teir foreign jobs or reires. For new ires, te data was obtained from te information seet tat eac departing OFW as to fill up wen is foreign employment contract is processed for approval and certification by POEA. 1 A certificate of POEA approval of is 1 Te POEA is carged wit responsibility for developing policies and strategies aimed at protecting OFWs in teir job placement, immigration procedure and at teir work place and promoting foreign employment. Te POEA sets standard placement practices, placement fee ceilings and minimum foreign wage rates by occupation. It screens placement agents before tey are registered and investigates and punises agents for recruitment abuses filed against tem. Foreign employment contracts ave to meet minimum standard wage rates and working 2

6 contract is needed for immigration purposes and for registration as an OFW. Registration provides im wit deat and disability insurance and oter protective services. A random sample of 12,679 information forms was drawn from a population of 25, forms filed by new ires wo left te country in January to Marc 23. For reires, a sample of 1,988 was drawn from tose waiting for teir POEA certificate. Te survey was conducted during te second and tird week of May 24. Te sampled reires were asked to give more information tan was asked in te information seet for new ires. (Appendix A for te questionnaire and Appendix B for te sampling). Te paper is organized as follows: Section 2 integrates migration into education/training decision in an open labor market. Earlier migration teory mainly looks at migration decision by itself. (Erenberg and Smit 23, Mincer 1978, Stark 1991) Section 3 discusses a model of monopsonistic employment and wage setting for foreign workers. Section 4 presents descriptive statistics from te surveys and Section 5, te regression results of te wage function. Section 6 concludes te study wit some policy prescriptions especially on quality of education/training. Section 2. Scooling and Migration Decisions Te Pilippines labor market may be reasonably caracterized as an open one. Tere are minimal restrictions on labor outflow except for protective reasons wen risk to life as become palpably ig as in Iraq presently. (October, 24 ) 2 Te scale of outflow and stock of OWF ave for years been large enoug to be perceptible to large segments of te population, making foreign employment a relevant option in teir job and skill coices. Skill is defined ere as an occupational skill obtained troug education/training and experience. On te aggregate, foreign employment as raised returns to education since most foreign jobs require ig scool or iger levels of education. Te large wage and cost differentials in foreign employment across skills and destinations ave canged te relative returns to categories of education/training. Returns depend on foreign-domestic wage differentials, migration-related costs, duration of foreign employment and te probability of foreign employment. Te formula for estimating returns to migration is as follows. t= ( P ) + C T T Dt Pf Y ft Dt Pd Ydt + C + s 1 t= f p Were W f = foreign wage W d = domestic wage P f = probability of foreign employment P d = probability of domestic employment conditions. POEA certification of approval of te contract is required for immigration purposes. Te certification process allows te POEA to keep track of outgoing OFWs tus letting it provide relatively accurate information on te outflows. However, it as not kept track of returning OFWs. Te Department of Foreign Affairs Commission on Filipinos Overseas makes te estimates on stock. 2 Since Marc 24, te government as proibited deployment of labor to Iraq. However, some OFWs in neigboring countries are reported to transfer to Iraq were tey are offered iger wages. OFWs may also get a contract for anoter destination but proceed to Iraq via tis destination. News media report tat some 6, OFW are now in Iraq. 3

7 C = fixed cost of migration for transport, placement services, immigration processing cost C js = cost of job searc = s(1-p f ) C p = psycic cost and risk premium assumed fixed T = is duration of foreign employment, t is time period D t = discount factor, = 1/(1+r) t d is domestic, f is foreign Generally P f, P d > P f >. Returns increase as wage differential, probability of foreign employment, P f, and its duration, T, increase, and fall as s and any of te fixed cost components increase. An increase in te probability of foreign employment increases wage gain and reduces te cost of job searc. Te probability of foreign employment may be inferred from te number of new ires or total deployed relative to te labor force and market studies. Tere are many skills tat at tis time ave close to zero foreign employment cance, e.g., legal skills, but tere is supposed to be a large demand for or even a sortage of nurses and ig-level skills especially in ig-wage economies suc as te US and Canada. (Kadria, 21, Biao Xiang 21,Sields 24) But wile ig wages may be earned in tese countries, te probability of employment for most oter skills is small. Te Middle East may pay lower wages but employs more OFWs. Hence te probability of employment tere may raise te expected returns to levels competitive wit te iger-paying destinations. Wen foreign jobs become available for a particular skill, tose in te labor force wit te skill may directly decide to migrate. Tose in te labor force witout te appropriate skill may consider investing in te skill. Te yout, on te oter and ave to invest first in education/ training, ET, before tey can respond to foreign demand. ET decision is more complex as te coices are wider, not just to migrate or not to migrate but wic skill to invest in. A person s ET coice set is bounded by financial constraints and is ability and personal taste (or inclination for different types of occupations, e.g., service oriented, sciences, arts and wat e can afford to pursue). He is assumed to assess or estimate returns to eac skill in te set and coose te one tat promises te igest returns. Te above formula is applicable to te coice of education/training categories and even to alternative foreign destinations. Te returns to ET i relative to ET j depend on teir wage differential in foreign and domestic employment, in te probability of teir foreign employment and in teir ET costs. 3 Te supply of skills to foreign markets as reservation wage on te vertical axis wic is interpreted ere to be te second term in te returns formula. Foreign wage must at least equal foregone domestic income and all migration-related costs plus risk premium and psycic cost in order to make migration wortwile. Psycic cost is assumed to differ between individuals. Variation in psycic cost will result in a positively sloped supply curve to foreign employment. 3 Formula for education coices i, j T T T Dt PfiY fit Dt PfjY fjt t= t= t= D P t di Y dit T t= D P t dj Y djt ET ET [ C + s(1 P ) s(1 P )] [ C C ] fi fj i j 4

8 Te responsiveness of te domestic market to foreign demand for a given skill will depend on te size of te stock of te skill and te degree of flexibility of te education-labor market in producing it. Clearly te larger te stock, te larger te potential supply for foreign markets. In te intermediate term, te labor force witout te appropriate skill may decide to invest in acquiring te skill demanded troug additional education/training. Acquiring a new skill entails additional ET cost wic as to be added to te total cost of migration. Tere may even be de-skilling wic as appened wit engineers accepting tecnician level foreign jobs, new medical graduates taking nursing courses in order to qualify for foreign nursing jobs and teacers working as nannies. Sifts between related fields suc as from medicine to nursing will be less costly tan sifts between broadly differentiated skills suc as law to medicine or teacer training to engineering. Te parameters of te long-run supply of a skill its position, size and elasticity will reflect te relative cost of te ET it entails, te relative scarcity of te ability and desirable traits it requires and te level of prestige society places on it. Te supply parameters are likely to differ between skills. Te more costly te ET of a skill and te iger te ability it requires, te smaller te size and elasticity of te supply curve. On te oter and, te demand curve for eac skill is assumed to follow marginal productivity teory. Section 3. World Wage Structure and Wage Function Consider a ric but closed economy. Its restrictive immigration policies confine te country s labor market to its own domestic demand-supply conditions and sield it from foreign labor market influences. It determines its own wage and employment structures. Suppose it decides to import labor of particular skills, say experts and ousemaids. Figure 1a and 1b give te ric economy s and te world s initial supply and demand curves for workers and for ousemaids. Te vertical axis of bot markets as reservation wage for te particular skill. Te world market for skill is taken to be competitive mainly because te supply is relatively small and not very elastic. education/training is relatively costly and requires relatively ig ability. World ict ict employment is and wage rate is and te ric country s employment is ict N d N w and wage rate is ict Wd Ww. Te ric country s desire to increase its employment by AC sifts te world demand curve to te rigt by AC in Figure 1a. World wage rate goes up ict Ww 1 to. Te ric country s new equilibrium employment is at wic includes bot ict N d 1 labor imported and additional domestic workers. It may pay world wage for wage efficiency reasons. All workers enjoy te iger wage. ict Ww 1 or iger Te market for ousemaids is different - te ric country faces a small domestic supply and a very large world supply. Witout labor imports, demand and supply would intersect at and employment is at. Te world supply is S S. Te ric country as tree immigration and wage options 1) allow as many ousemaids as demand warrants at te world reservation wage,. Te employment of foreign maids would be at. Tis decision would eliminate te employment of domestic ousemaids. 2) restrict entry to say, AB. Te supply curve sifts outward by te number allowed entry to N d 1 Wd S d 2S d 2 N d Ww Ww w w1. If te government lets te immigrants compete in te 5

9 domestic labor market, wage will equilibrate at for bot domestic and foreign workers. Te wage rate will be iger tan te world wage rate for ousemaids. 3) A tird tack is to restrict entry to say AB, pay domestic workers market equilibrium wage and set te wage of foreign workers at world level or sligtly iger at say. Wd 1 Wd 1 W f 3 Wage Dd 1 D d S d Wage Dw 1 D w S w Wd 1 Wd A B C Ww 1 Ww A C S d N d N d 1 D d D 1 d S w N w N w 1 D w Dw 1 N w Figure 1a Figure 1b Wage S d S d 2 Wage S w 1 Wd D d D w W d1 A B W f 3 S d S d 2 C D D d Ww D w S w Nd 3 Nd Nd 1 N d N w Figure 2a Figure 2b 6

10 Tere will be excess demand CD. Tis policy results in tree wage levels for maids, one for domestic or native maids, one for imported maids and one for world supply. Te disequilibrium condition may be sustained by imposing rules suc as penalizing illegal immigration and employment of illegally entered foreign workers. Te Pilippine Commission on Filipinos abroad reports about 1.9 Million irregular emigrants wit majority in te US and oter advanced economies. Te point is tat te importing country as te power to decide on bot te number of foreign workers to employ and te wage rate to pay tem. Labor importing countries ave actually made different decisions on tese two variables. Te US and Canada restrict te number of foreign workers of selected skills to let in but once inside, te foreign workers are allowed to compete wit te natives for wage and oter terms of employment. Singapore, Taiwan and oter countries, in contrast, restrict bot te number and te wage rate of foreign workers. Saudi Arabia approximates tack 1 were it relies on foreign workers for a number of occupations wic it pays wage rates tat may be close to teir reservation levels. No country as followed policy 1) were entry is totally free, and only te US, Canada and Australia follow te second policy. Apparently most labor-importing countries adopt variants of te tird policy. Immigration barriers tend to preserve te wage structure of eac labor-importing country. Our surveys sow tat in fact wage rate varies across destinations not so muc because of te variation in teir per capita income but because of variation in immigration policy. (Appendix Table 3) Tis leads us to posit a wage function wit destination as an explanatory variable. (Section 5) Section 4. Te Structure of OFW Wage Rates Our two surveys give te wage structure of OFWs. Table 1 presents te mean wage of new ires and of reires by major occupations and destinations. For eac occupation, wage rate varies widely across destinations. For nurses, wage rates range from $458 in Taiwan, $56 in Saudi Arabia to $2,75 in te UK and Ireland and $3,359 in te US and Canada. Te wage rates for Oter professionals range from $478 in Singapore and $46 in Oter Americas to $2,415 for UK and Ireland and $3,323 in te US and Canada. Tere is as wide variation in all oter occupations excepting for entertainers wo are mainly employed in Singapore and Japan teir respective wage rates being $1, and $1,733. Housemaids earn as low as $189 in Brunei and $24 in Saudi Arabia and as ig as $1,744 in te UK and Ireland and $2,114 in te US and Canada. 7

11 Table 1: Average Montly Wage of New Hires by Occupation and Destination DESTINATIONS Saudi Arabia Brunei OCCUPATIONS Not Stated Nurse Oter Professionals Sales Clerk Domestic Helper Entertainer Oters TOTAL ,898 2, Oter Europe + Spain + Cyprus Oter Middle East ,271 2,477 Oter Asia , Papua New Guinea + Cina East Timor Oter americas (cuba) Taiwan ,855 1, ,5 479 Hong kong Africa , Singapore , , Trust Territories + Diego Garcia ,163 1,163 Israel ,212. 1,369 1,68 1,733 1,54 1,732 Japan ,35 3 4,45. 2,75 2,415. 2,62 2,146. 1,744 1,992 United Kingdom + Ireland , ,359 3,323. 3, ,114 3,23 United States + Canada ,249 1, , , TOTAL ,295 4,62 5,817 12,679 Note: First row for mean wage, second, for standard deviation and tird, for number of observations

12 Table 2: Average Montly Wage of Reires by Occupation and Destination OCCUPATIONS DESTINATIONS Oter Domestic Nurse Professionals Sales Clerk Helper Entertainer Oters TOTAL 769 1, Saudi Arabia , Brunei. 1, Oter Europe + Spain + Cyprus , Oter Middle East 52 1, Oter Asia +. 1,941 4, ,124 1,289 Papua New Guinea + Cina. 1,158 6, ,833 1,915 + East Timor Oter americas (cuba) Taiwan Hong kong Africa Singapore Trust Territories + Diego Garcia Israel Japan United Kingdom + Ireland United States + Canada TOTAL. 1,381. 1, ,4 1,362. 1, , ,76. 1, , ,134 1,28. 1, , ,61 1, ,8 1, , , ,865 6, ,272 5,899 1,64 7, ,38 9, ,452 3, ,16 1, ,143 1, ,65 6,886 1,667 1, , ,227 1,988 9

13 Experience 4 is compensated by iger wage in virtually all destinations and occupations as sown by te iger wage rates earned by reires tan by new ires. (Table 2) In Saudi Arabia and oter Middle East countries, reires earned muc iger wages tan new ires in all occupations. In Singapore, experienced Oter professionals and ousemaids earned more tan double new ires. An exception is te entertainer group in Japan. Entertainers usually work on 3 to 6 monts contracts. Reired entertainers earned less tan new ires peraps because tey may no longer be working as regular entertainers. For tis occupation, novelty rater tan experience, tends to ave a iger market value. Excepting for entertainers, te ratio of mean wage of new ires to mean wage of reires exceed 1. and ranges from 1.5 for clerical workers to 2.9 for nurses. On average, reires ave worked 7. years abroad, longest in Saudi Arabia at 9.3 years and 5.7 years in oter Middle East countries. Tis region as te longest istory of large scale labor imports. Lengt of experience in oter destinations suc as Hong Kong, Oter Europe and Oter Asia was also relatively long, at least 4 years. (Table 3) 4. OFW Wage Function Te following wage function was estimated by regression: Ln wage = a + b 1 S + b 2 E +b 3 D + b 4 O + b 5 G + e were: W = montly wage in US$ S = education classified into elementary, incomplete ig scool, ig scool graduate, some college, college graduate and vocationaltecnical training E = experience, for new ires tis is proxied by age = 23 birt year for reires and E= 23 year of first foreign job. D = country of destination O = occupation G = gender 4 Te survey questionnaire for reires tried to asked for a istory of foreign employment but too few answered te question. We resort to measuring experience abroad as year 23 less year of first foreign employment. For new ires experience is proxied by 23-birt year.

14 Table 3: Average Montly Wage (US$) of New Hires and Reires and Average Years of Experience of Reires by Occupation OCCUPATION Not stated Nurse Oter Professionals Sales Clerk Domestic Helper Entertainer Oters TOTAL New Hires Wage (US$) Wage (US$) Reires Experience (No. of Years) Ratio of Reires to Reires ,63 3, , , , , , , , ,817 1,233 1, , , ,679 1,996 1,998 Note: First row for mean wage, second, for standard deviation and tird, for number of observations Regressions were run on 4 sets of data. Te first two sets consist of te observations of new ires and reires tat ave complete information on te education variable; te oter two sets ave te full samples including tose wit incomplete education answers. In te second two sets, unanswered education question was treated as a separate education category. Te results of te regressions tat ave complete information on all variables education, occupation, destination, age/experience and sex using individual new ires are given in column 1 of Table 4. Te results are quite robust wit a ig R 2 of.89. Only completed college is found to exert a significant and expected influence on foreign wage. Having completed college increases wage by 12.1% relative to elementary educated. All te oter occupations including Oter professionals earn less tan nurses. Housemaids earn te lowest at -68.2% less tan nurses, sales next lowest at -.6%, and Oters wic largely comprise skilled manual workers, %. Te entertainer occupation stands out for not being a significant variable. Wage increases by.85% for every year of age and being female reduces wage as compared to male by 1%. Destination contributes te largest variation in wage rate. Using Saudi Arabia as te base, te igest wage is earned in te US, next UK and Ireland, ten Japan and Hong Kong. Wage is iger by 193.6% in te US and 152.9% in te UK and Ireland as compared to Saudi Arabia. Te only destination tat pays less tan te latter is Brunei. 11

15 Table 4: Regression Results (ln wage = a + b 1 S + b 2 E + b 3 D + b 4 O + b 5 G + e) Vocational H.S. Level H.S. Graduate College Level VARIABLES College Graduate/Higer Oter Professionals Sales Clerk Domestic Helper Entertainer Oters NEW HIRES REHIRES MODEL 1 MODEL 1 MODEL (dropped) (dropped) (dropped) (dropped) * 3.29 * 3.27 * * * * * * * ** * * * * -6.1 * * Brunei * Oter Europe + Spain + Cyprus Oter Middle East * 1.64 * 1.6 * Oter Asia + Papua New Guinea Cina + East Timor * 4.92 * 5.6 * Oter americas (cuba) * Taiwan * 2.8 * 2.21 * Hong kong * 2.63 * 2.8 * Africa * 2.95 * 3.8 * Singapore * 3.49 * 3.54 * Trust Territories + Diego Garcia * * Israel * 1.95 * 2.23 * Japan * United Kingdom + Ireland * * * United States + Canada * 4.8 * 4.18 * Age * 4.89 * 5.2 * Experience * Female * * * Constant * * * R Adjusted R No. of Observations 7,162 1,587 1,587 Note: 1. Base variables for: Education Elem. Level and Graduate, Occupation Nurse and Destination Saudi Arabia 2. First row is for te estimated coefficient and second, for te t-statistic: ** - significant at.1 level * - significant at.5 level 3. Dropped means no observations

16 Te regression results for reires wic ave an alternative proxy for experience (year 23 less year of first job foreign job) are given in Columns 2 and 3. Age and foreign experience were alternatively used in te wage regression. For new ires every year increases wage by.85%. For reires age as a smaller coefficient value tan foreign experience, 1.25% vs. 2.68%, but bot are significant and of te expected sign. Tis is to be expected since aving worked abroad likely reflects longer tenure wit te current employer, ence more specific training and better personal relationsip. Tan (24b) observes tat advertisements for foreign jobs place muc importance on experience but te age variable in te survey appears not to capture well te experience desired by foreign employers. Te quadratic form wit age 2 or experience 2 is not supported by te data. (Appendix Table 4) As in te case of new ires, completed college is te only education category tat exerts a positive and significant effect on wage, increasing wage by 22.7% relative to elementary education. Te wage rate of Oter professionals is not significantly different from tat of nurses. Also not significant is te entertainer category. Tere is a larger difference in te wage rate of clerks, ousemaids and oter skilled manual workers relative to nurses among reires tan among new ires. It is not unreasonable to expect more opportunities for on-te-job training and terefore iger returns to experience of nurses as compared to tese tree occupations. Excepting for Oter Americas and Japan, all destinations exert significant influence on wage rate relative to Saudi Arabia. However, te coefficients for UK and Ireland and for te US and Canada are smaller among reires tan among new ires. Alternative specifications were run to take account of possible multicolinearity especially between occupation and education and occupation and destination. Te skills demanded tend to be destination specific except for te Middle East countries wic employ varied skill categories including sales. (Appendix Table 4) Nort America and te UK import mostly nurses and oter professionals wile Japan, factory workers and entertainers. One specification excludes education, anoter excludes occupation. Te alternative specifications marginally lowered R 2 by 2 percentage point from.89 to.87. Most coefficients retained teir significance level and value. Te coefficient of college education increases wen occupations are excluded in te regressions. Destination remains a strong determinants of wage rate. Tere was a large number of non-response about education, 42% of te sample of new ires and 15% of te sample of reires. We ran regressions on te full sample and treated no stated education as anoter education category. Tose wo did not state teir educational attainment ave more tan elementary education as tey are distributed across te various occupations. Te results are given in Appendix Table4. For te regression of new ires, tis education category like all oter education categories excepting completed college is insignificant. Te coefficients of te occupation categories are sligtly larger in te full sample tan in te small sample but te coefficients of te oter variables are not significantly different in te two samples and te R 2 s are te same, about.9 As far as te reires are concerned, te observations wit no stated education form a muc smaller proportion of te total sample, 15%. R 2 is.34 wic is iger tan for te smaller sample at.3. Completed college as well as no stated education are bot significant and ave positive coefficient relative to elementary education. Te categories Oter professional, Oter Americas exert significant effect on wage rate in te larger sample but not in te smaller sample. On wole te regression results are robust wit

17 relatively ig R 2 and wit most coefficients significant and of te expected sign. Eiter samples may be used. Section 5. Related information on Migration Recall returns to migration partly depend on foreign-domestic wage differentials, migration-related costs and premium to risk. Some information on domestic wage, migration-related costs and risk is presented ere. Table 5 gives te annual gross returns defined as foreign-domestic wage differential in US dollars for selected occupations for new ires and reires. Foreign wage is from our surveys. Domestic wage is obtained from te montly wage rates of employees in large establisments for selected occupations in Metro Manila. See Appendix Table 5 (Large establisments are tose wit 5 or more employees.) Montly domestic wage rates are quite low ranging from $191 for nurses and $192 for service workers oter tan ousemaids to $32 for Oter professionals. Clerical workers earn $26 and Oter skilled workers, $285. Note te rater narrow range. Compare te range to te average montly foreign wage rates of $47 for Service and Oter skilled workers to $163 for nurses. Te annual foreign-domestic wage differential ranges from $1,464 for Service and Oter skilled workers to $1,464 for nurses. Annual gross returns vary even more widely wen we consider destination. Gross returns for nurses are $3,78 in Saudi Arabia, $7,464 in Singapore and $38,16 in te US. For Oter professionals, gross returns in te respective locations are $2,544, $1,896 and $36,36. Note tat domestic wage is te average for all employees, not taking account of experience and oter relevant variables. Te annual gross returns for reires wo receive iger wage rates tan new ires are muc iger. Te average gross returns are $35,316 for nurses, $14,82 for Oter professionals, $4,536 for Clerks, $5616 for Service workers and $4,5 for Oter skilled workers. Reired nurses in Saudi Arabia gain by $6,936 and in Singapore, $9,132. Reired Oter professionals ave muc iger gross return tan new ires in Saudi Arabia, $1,344, in Singapore $19,844 and in te US, $38,

18 Table 5 Annual Gross Returns to Migration for Selected Occupations, 22 A. Occupations 1. Nurse Montly Foreign Wage 163 New Hires Montly Domestic Wage 191 Annual Gross Return 1464 Montly Foreign Wage 3134 Reires Montly Domestic Wage 191 Annual Gross Return Oter Professionals Clerks Service workers } 5. Oters B. Selected Locations 1. Nurses Saudi Arabia Singapore US Oter Professional Saudi Arabia Singapore US Note: Gross returns are simply foreign-domestic wage differential. Domestic wage is assumed to be te mid-point of te wage range for oter professionals, service workers and tecnicians. 15

19 In Table 6, we ave out-of-pocket cost of migration for OFWs bound for Hong Kong and Italy. Te estimated cost is based on a small, albeit unrepresentative sample OFWs. Neverteless, it is insigtful. Te out-of-pocket costs are categorized into placement fees paid to recruiters, fees paid to POEA and te Overseas Workers Welfare Fund, OWWA, and miscellaneous expenses for passport, medical examination, National Table 6 Pre-departure Expenses of Migrant Women Bound for Hong Kong, Cina or Italy Lowest Range Maximum Range Placement fee (Hong Kong P3, ($581) P65, ($126) OWWA Contribution 5 (or $1) 5 (or $1 ) POEA Administrative fee 125 (or $25) 125 (or $25) Mandatory Medical fee 9 $17 9 $17 P3715 P7215 Miscellaneous expenses: Passport Medical, I clearance, Video Pre departure orientation POEA certificates of overseas Employment (COE) Agency registration fee $ $1, P P44 Total for Hong Kong P4,35 $784 P76,55 $1487 Total for Italy P8,35 $1556 P311,55 $638 Notations: In US$ total cost: Excange Rate for 22 was P51.6/$1 Source: Añonuevo and Añonuevo (eds.) 22, Coming Home, Women, Migration and Reintegration, Manila, BalikBayani Foundation and Alika Overseas Workers and Community Intralive, Inc. 85 Bureau of Investigation clearance, pre-departure orientation fee, POEA certificate fee and Agency registration fee. By POEA rule, cost of transport is to be paid for by employers but is sometimes borne by te migrant. Te total cost for Hong Kong-bound OFW ranges from $784 to $1,487 and for Italy, $1,556 to $6,38. Placement fee comprises te bulk of te cost and explains most of te cost variation. For Hong Kongbound workers, placement fee comprises from 74% to 85% of total cost, for Italy-bound workers, from 87% to 96%. Placement fee is iger for Italy tan for Hong Kong peraps because te iger wage rates in Italy attract more job applicants and tose more willing to pay iger placement fees. Te annual gross returns of all five occupational categories of new ires exceed te out-of-pocket cost of placement for Hong Kong. (Table 5) Reires do not ave to pay placement fee but only te POEA and OWWA fees and some of te miscellaneous fees, at most P1,35 or $2. If OFWs can work beyond one year, returns would be positive for all categories. Tose bound for Italy would earn at least $1, per mont or $12, per year. Net returns would be positive even if te migrant pays te maximum cost of $6,38. Te same olds for Taiwan-bound OFW. Te average montly wage for ousemaids and oter workers is $458 or $5,496 per year. A small survey by St. Cristoper Curc (catolic) in Taiwan found placement cost at P72, or $2 in

20 OFW bear tree kinds of risk risk of fraudulent placement were no job as been contracted, risk of contract violations consisting of non-payment, delayed payment or below-contracted payment of salary, pysical abuses suc as rape and injuries and even deat. New ires generally face more risk tan reires. Te latter ave acquired better information about teir foreign destination and ave built more satisfactory terms of employment and relationsip wit employer tan new ires. POEA and te Overseas Workers Welfare Administration (OWWA) ave put in place rules and offices to minimize suc abuses. POEA screens and registers placement agents before tey can recruit OFWs. It evaluates and approves employment contracts. POEA s certificate of approval is required for immigration purposes. In turn, OWWA maintains service centers in cities wit a large concentration of OFW suc as Riyad and Hong Kong. Te $25 registration fee covers deat and disability insurance, albeit of meager amounts P1, ($179) for natural deat and P2, ($358) for accidental deat, injuries up to P1,. Te medical fee of P9 ($16) covers ealt insurance of te same benefits as te national PilHealt insurance. Table 7 gives a sense of te risks involved in foreign employment. Bot POEA and OWWA receive complaints from te workers and try to settle tem troug earings, negotiations wit employers and deat and disability insurance. Te number of complaints must be seen as a ratio of total stock, total deployed or total deployed new ires. Close to 5, cases were received by POEA and OWWA in 2. New ires numbered more tan 15,. (Appendix Table 2). Exploitative and fraudulent recruitment practices were no job as been contracted regularly occur. Frequently reported were non-payment, below-contracted wage payment and delayed payment of wages. Tere were also accidental deat and disability claims wic numbered 163 and 62 in 22. (Tan, 24b) Tere ave been pysical violence including rape. Te incidence of contract and pysical violence differed by destination. POEA and OWWA do not publis tese data. But te experience by Sri Lankan workers likely applies to OFWs. Incidence is most serious for te less skilled in some Middle East destinations. (Table 8) Professional workers tend to find teir own foreign employment troug advertisements in te internet and oter sources and avoid ig placement cost. Because of te sortage of nurses in te US, recruiters are reported to earn $1, to $12, per placed nurse in American ospitals. In turn tey offer nurses free placement services and immigration and transport cost plus allowance for settling accommodation in te US. Neverteless. te excess demand for nurses tere as not abated because te great majority of nursing graduates are unable to meet its more stringent requirements: te passing of Pilippine licensure examination, experience in large ospitals and te passing of te US Council for Nursing Graduates of Foreign Scools (CNGFS) examination. Apparently relatively few meet all tese requirements 17

21 Table 7. Cases Filed and Settled by POEA, Welfare cases I. Cases Handled A. Cases pending at te beginning B. Cases received II. Cases Disposed III. Cases pending At te end 2. Recruitment Cases I. Cases Handled -Employment related - Recruitment regulation A. Pending cases at te beginning -Employment related - Recruitment regulation B. Cases received -Employment related - Recruitment regulation II. Cases Disposed A. Regular Dis- Position -Employment related - Recruitment regulation III. Cases pending At te end Legend: no breakdown available. Source: POEA Overseas Employment Statistics , unpublised

22 Table 8 Complaints Filed by Sri Lankan Overseas Workers by Sex, 22 A. Distribution of Complaints by Cause, 22 Female % of Total Total % of Total % of Female Violation of Contract Harassment Deat Lack of communication Stranded-lack of reception on arrival & stranded witout 5 7 employment Oter Total Complaints B. OCW Given Assistance by Complaints Female % of Total Total % of Total % of Female Harassment & non payment of Wages Harassment Sexual arassment Non-payment of salary Insane Disabled Sick Pregnancy Wit babies Oter problems Total Source: Sri Lanka Bureau of Foreign Employment Handbook, 23, Researc Division,SLBFE Table 9 Percentage Distribution of Overseas Contract Workers and Complaints 22* Country OCW Complaints % Complaints % of Total OCW KSA UAE Barain Oman Kuwait Qatar Jordan Singapore Lebanon Cyprus Malaysia Maldives Oters Total *Provisional Source: Conciliation Division Sri Lanka Bureau of Foreign Employment SLBFE Information Tecnology Division Sri Lanka Bureau of Foreign Employment SLBFE

23 since te Pilippines, despite te attractiveness of te US offer, fails to meet its demand. Less tan 2% of US job order for nurses were filled up in recent years. Te same olds for UK and Ireland. (Tan 24b) Section 6. Conclusion Returns to foreign employment are, on average, positive. Te foreign-domestic wage differential is generally ig and compensates for te monetary costs of migration. However, foreign wage rates vary widely not just across occupations but for eac occupation, across destinations. Te variation across destinations is attributed to te segmentation of te world labor market arising from immigration barriers tat eac labor importing country imposes on foreign workers. Te paper illustrates ow te more common immigration barriers provide an importing country some monopsony power over te employment and wage rates of foreign workers. It may decide to pay tem a wage tat is competitive wit te natives as in te US and Canada or it may decide to pay tem just teir reservation wage. Some ric economies like Singapore, Taiwan and Spain pay Filipino ousemaids way below wat tey pay teir lowest-wage workers. Wit eac country setting its own wage for foreign workers, te wage rates of migrant workers tend to differ across destinations. Tis is clearly evidenced by te wage structure obtained in te surveys conducted for tis study. Te wage regressions sow destination to be te strongest variable determining foreign wage rate. Experience enances foreign wage but being female reduces it. Working in America and Canada could earn double te wage rate in Saudi Arabia, and working as a nurse and in oter professions also earns muc iger wage rates tan as a ousemaid and oter occupations. Professional occupations require college education but all te oter occupations ave looser educational requirements. Excepting for completed college all education categories exert an insignificant influence on wage rate. Obviously America, Canada and a few oter OECD destinations are relatively attractive destinations. But tey ave small if not zero immigration quota for most occupations. Currently tere is a large demand for nurses and igly skilled labor. Te Pilippines as been unable to supply te demand for tese skills. Tere is a very active recruitment by American ospitals for Filipino nurses but relatively few meet teir requirement. Most foreign ospitals require a few years experience, completion of te nursing degree and passing te Pilippine professional licensure. In addition to tese, te US requires te passing of te examination by te Council for Nursing Graduates of Foreign Scools. Apparently few pass tis examination for less tan 2% of te US job orders for nurses were filled up in te last two years. Te job orders for nurses for UK and Ireland wic pay relatively ig wages ave also been partially filled. Most Pilippine nurses are employed in te Middle East were te pay is less tan alf tat of US or UK. Te OFWs enjoy substantial returns from foreign employment but on te wole tey do not maximize te gains from migration. Te poor quality of teir scooling and training prevents tem from qualifying for jobs in ig-wage destinations. Te inability of te Pilippines to supply nurses and specialists in te US and oter OECD countries may be blamed on te scarcity of good quality college education for tese fields. Tere is as well a deart of good quality teacers and engineers. Yet undreds of 2