WORLD POLICY JOURNAL

ARTICLE: Volume XXI,  No 2, Summer 2004	Print
	Friendly

Remittances: The Perpetual Migration Machine
Michele Wucker*

In the late nineteenth century, wealthy British families exiled their wayward second sons to far-off colonies with the promise of a monthly allowance. In return, these black-sheep “Remittance Men,” immortalized by Mark Twain in his travelogue, Following the Equator, pledged to stay out of trouble and out of the way. Today’s remittance men (and women) are not the black sheep of their families but the workaday heroes of the developing world. Families in developing nations send their best and brightest members abroad to find jobs and send money home. In countries like Mexico, India, Turkey, and the Philippines, the mantra is no longer, “Go away and stay away,” but “Go and don’t forget us.”

Prime Minister Rajiv Gandhi once called Indians abroad a bank “from which one could make withdrawals from time to time.” Today, developing countries are making withdrawals from the “bank” of global migrant worker remittances at an increasing rate. In 2003, migrant workers living abroad sent more than $100 billion in remittances to their home countries—a source of funds that is being hailed as the great new hope for the developing world.¹

Last year, migrant worker remittances, which are growing at a double-digit annual rate, amounted to nearly twice the amount of global development aid ($56 billion). The $38 billion in remittances sent to Latin America in 2003 were nearly equal to all foreign investment in private companies in the region. Remittances account for more than 10 percent of the gross domestic product (GDP) of six Latin American nations, and for over 29 percent of Nicaragua’s GDP. In Haiti and Jamaica, remittances generate more revenues than does foreign trade. In Honduras and Nicaragua, average annual remittances per household amount to more than double annual per capita GDP; in Haiti, they amount to more than three times per capita GDP. Nor is this a phenomenon seen only in poor countries; in the late 1990s, émigrés owned one of every five dollars in the Portuguese banking system.

By comparison to the notoriously volatile capital markets, remittances are remarkably steady. Indeed, they often rise during economic crises, when receiving countries are having the hardest time attracting capital. These funds often support not just the person who receives them, but entire families, and local communities benefit from the multiplier effect as money circulates among businesses.

Ironically, the better off immigrants are, the less likely they are to send remittances. Thus, it is the poorest workers who are most responsible for keeping their home countries going. The moral implications of this situation are complicated further by the practices of the financial firms that skim sometimes excessive commissions and fees from this flow of money.

Money transfers from émigrés—what Bangladesh calls “manpower exports”—are helping the least developed countries stay afloat, for now. Yet remittances have side effects that may create other problems in the future. For one thing, although many development economists see these transfers as a stable source of income,² the flow of remittances is vulnerable to political and economic shocks abroad. Even more worrying is the fact that developing nations may see this flow of cash as an excuse for not undertaking needed development projects.

But the biggest side effect of remittances on policy considerations is this: the money eventually stops coming unless people continue to emigrate. By adding a huge financial incentive to the traditional function of emigration as a social safety valve, remittances discourage governments from making the changes needed for people to stay home—and, indeed, reward them for encouraging people to leave. In other words, remittances feed a perpetual migration machine.

This machine can rob a country of its ability to improve itself if the best and brightest of its citizens go abroad. It is crucial, therefore, that receiving countries handle remittance flows wisely in order to be able to end their reliance on manpower exports as a means of survival and induce skilled émigrés to return home.

Flying Money

The oldest references to remittances come from the late Tang dynasty (A.D. 618–907), when Chinese tax collectors devised the fei qian “flying money” system. Under this system, government liaison offices in the capital would issue certificates to provincial merchants who sold their goods in the capital. The merchants could later redeem these certificates for cash when they returned home. This system meant that merchants did not have to worry about being robbed of their profits on the perilous journey home. A similar system, chop, is named for the stone seal dipped in red ink and used to stamp a document that is then torn in two. One of the pieces is sent to a broker in the receiving country and the other to the intended recipient of funds. When the latter takes his half to the broker with the matching half, he is able to collect his money. Today, the hundi (India), hawala (Pakistan and various Muslim nations), and poey kuan (Thailand) systems all employ a code or pass-document to match the sender and the receiver of funds.

The chit, (the word is a diminutive of the Hindi chiiti, or certificate), was introduced to China by British colonialists in the nineteenth century. British workers wanting to buy goods from local merchants would write chits that merchants could cash with the Chinese comprador who managed an escrow account into which British workers’ salaries had been deposited.³

Remittances, which are now less often sent by such informal systems, have become a big and highly lucrative business, with profit margins as high as 30 percent. Migrant workers now rely on money transmitters like Western Union and MoneyGram, smaller regional or national transfer companies, and, to a small but growing extent, banks. The principle remains the same: in return for a fee, the sending company accepts the money the worker wants to send, then notifies its representative in the recipient’s locale, who pays the recipient in local currency.

Until the last few years, most banks handled mainly large flows of funds, leaving the small remittances to the money-transfer firms. But, eyeing a market that the financial services market research firm Celent Communications estimates will generate more than $15 billion in annual revenues by 2006, banks are upgrading their technology to compete with money-transfer companies. Their twenty-first-century version of chops and chits is the ATM card, which migrant workers can send home to allow family members to withdraw funds from cash machines. Gwenn Bézard, an analyst at Celent Communications, projects that by 2013 there will be 13 million holders of remittance debit cards.⁴

Heavy Burden, Slight Shoulders

A global money-transfer industry has arisen from the diligence and loyalty of migrant workers, who as recently as the mid-1990s paid fees as high as 30 percent of the funds transferred to send money home. Although fees fell to an average of 15 percent by the end of the decade, they continue to take a significant bite out of the money sent back to developing countries. Today, the average cost to send $200 or less—a typical amount—to Latin America is 7.6 percent. The percentage charged generally falls as the amount of the transfer increases, but most émigrés are unable to send enough to lower the bite significantly.

Only 19 percent of workers who earn more than $50,000 a year send remittances, while 46 percent of those who earn less than $30,000 do so.⁵ Thus, generally speaking, individuals who send money home are the very people least able to bear the transaction costs: low-income migrants who are more likely to lack legal documentation and less likely to understand how the financial system works. In a 2001 study, only one in three Latino immigrants in the United States knew that some of the money they were sending to their families was being eaten up by fees, commissions, and often unfavorable exchange rates.⁶

The flow of remittances from low-income senders may be disrupted by short-term shocks, as happened when over a million (mostly Arab and South Asian) foreign guest workers fled the Middle East with the onset of the first Gulf war in 1990. In the wake of September 11, senders in the United States faced unemployment amid the ensuing backlash against immigrants and an economic slump. In 2002, transfers via the thousand-year-old hawala transfer mechanism were cut off abruptly when the U.S. government banned unlicensed moneytransfers. The new rule also shut down al-Barakat, one of three transfer firms that were keeping Somalia from going under with an estimated $500 million in annual remittances. Government officials alleged that al-Qaeda had used al-Barakat to finance terrorist activities.⁷ Such disruptions in the flow of funds show how variable remittances can be as a source of income.

In the United States, according to a study by the Pew Hispanic Center and the Multilateral Investment Fund, 6 million people, or 42 percent of adult, foreign-born Latinos send remittances home. Moreover, as the study shows, this creates a self-reinforcing pattern: in Mexico, for example, 26 percent of those surveyed who received remittances were considering migrating themselves, compared to 17 percent of those who did not receive money from abroad. In Ecuador, 83 percent of respondents agreed with the statement, “One of the principal reasons that people leave is so that they can send money back to their families.”⁸

While remittances provide an incentive to leave, they are not a permanent solution to home-country financial hardship. Over the course of decades remittance flows tend to slow as families in the home country die or emigrate, and ties weaken from lack of family contact. About half of Latino immigrants who have been in the United States for ten years or less send money home, according to the Pew study. For those who have been here for 20 to 30 years, that number drops to 23 percent. As migrants integrate into the host country, their earning power passes to the second generation, and greater portions of savings remain in the host country.

For remittances to stay at the same levels, therefore, people have to keep leaving their homelands. There is every reason to believe that they will continue trying to do so, because as remittances come in, more people want to leave. Part of the reason for this is the keeping-up-with-the-Joneses effect that remittances create: they fuel inflation locally, especially of real estate prices, and so it becomes even harder for a family to survive if it does not receive money from abroad.

This is the catch-22 of remittances: they boost the economies of the labor-exporting countries, which gives their citizens the hope of being able to remain at home, yet at the same time they contribute to the pressures that make people leave.

Who Stays and Who Leaves?

Remittances are also problematic because of a phenomenon that the economist Philip Martin has called the “migration hump.”⁹ Emigration increases once per capita income rises enough (often through remittances) for people to afford to get out. Emigration does not slow again until per capita income reaches the other side of the “hump,” that is, until it increases to a level at which citizens no longer expect moving to another country will improve their quality of life enough to be worth the trauma of dislocating. This phenomenon means that a developing country is exporting its nascent middle class.

The economist B. Lindsay Lowell has calculated that losing high school– and college-educated nationals slows economic growth, albeit slightly. On the other hand, the emigration of those with only a grade school education accelerates a country’s economy—though again, very slightly. This data suggests that ensuring the return of educated émigrés would benefit labor-exporting nations. “Employing developing country nationals (or returnees) at a higher wage would induce return migration, increase the permanent skill level of the source country, and increase source country growth,” Lowell argues.¹⁰

The impact of remittances and migration on brain drain varies considerably among countries. The World Bank’s Richard Adams, Jr., has argued that for most countries, migration does not cause a significant brain drain; less than 10 percent of the college-educated populations in labor-exporting countries have emigrated overall. For a few countries, however, especially those with wealthy neighbors, high proportions of the most highly educated leave (these include the Dominican Republic, El Salvador, Guatemala, Jamaica, Mexico, Morocco, Tunisia, Turkey, and Sri Lanka).¹¹

Getting over the Hump

The challenge for sending countries with respect to the migration hump is twofold: first, governments need to channel remittances in such a way that they improve the quality of life at home; second, they need to figure out how to lure educated émigrés to return home and use the skills they gained abroad to help the economy grow.

“Channeling” remittances, however, is problematic, for these funds are and should remain private. Some countries, in trying to use remittances to strengthen their hard currency reserves and at the same time lure workers back, have committed spectacular policy blunders. Bangladesh, South Korea, Pakistan, and the Philippines all have made attempts to require workers to send mandatory amounts home through formal banking channels. Only the Korean government has succeeded, and only because so many Korean migrant workers were employed by Korean companies in the Middle East. In return for the government’s assistance in winning contracts abroad, these companies deposited wages in Korean banks.¹² In the 1970s, Turkey encouraged its guest workers in Germany to keep their savings in deutschemark accounts in Turkey, but when the Turkish economy collapsed, the government would only convert the funds into Turkish lira at an unfavorable rate, decimating the workers’ nest eggs.

Perhaps the most notorious case of a government manipulating remittances for its own purposes involved the 4.5 million bracero agricultural guest workers who came to the United States in a special governmentsponsored program to deal with a manpower shortage after the outbreak of the Second World War. The United States withheld 10 percent of these workers’ wages, to be paid once they returned to Mexico.

Somehow, the money disappeared. The United States says Mexico was at fault, but there is some evidence that at least part of this money never made it out of the United States. Workers and their descendents are still fighting to win back $500 million in wages and interest, and in March 2001 they filed a lawsuit against the U.S. and Mexican governments, the Wells Fargo bank, and three Mexican banks. In August 2002, a San Francisco judge threw the case out, partly on statute-of-limitation and jurisdictional grounds. But even the not necessarily immigrant-friendly California State Legislature thought this was unjust and in January 2003 it extended the statute of limitations to allow the workers to press their case.¹³

At the opposite end of the spectrum of failed policies, some countries have barred émigrés from holding home-country bank accounts. Colombia, for example, did not end this practice until late in 2003.¹⁴

Policymakers have learned that there are ways to increase the impact of remittances on the home country without killing the goose that lays the golden egg. The most successful programs are a mix of public and private initiatives. Morocco, for example, subsidizes the cost of sending remittances through its state-owned Banco Popular. This gives migrant workers an incentive to use the formal system, but workers actually use it only because they have confidence that the system will work—unlike the disastrous Turkish example. Remittance flows thus create a positive feedback loop that can encourage governments to maintain responsible monetary policies.

Encouraging workers to send remittances through formal financial channels can help reduce costs to the workers themselves as well as increase the receiving country’s hard currency reserves, strengthen local financial systems, and make more credit available to nurture local economies. Encouraging competition among money transmitters and promoting better use of technology will further reduce the costs of transfers and thus increase the amount of money available to developing countries.

According to the Inter-American Dialogue Task Force on Remittances, “the single most important task, for both governments and non-governmental agencies, is to encourage senders and recipients to make use of banks and other financial institutions.”¹⁵ The task force recommends that governments open the remittance market to such financial institutions as credit cooperatives and micro-finance organizations, publish regular analyses of services provided by firms so that users can compare costs and choose accordingly, and compile market research to help develop strategies to increase participation in formal financial services.

Until 2001, when the United States banned unlicensed money transmitters, just over half of the remittances sent from the United States went through informal money transfer systems, which often were much cheaper than formal sending mechanisms.¹⁶ Before September 11, 2001, hawala fees, for example, were typically between 0.5 percent and 3 percent of the money sent. The trade-off now is higher fees for individuals in return for greater protections against money laundering and terrorist financing, and stronger banking systems in the receiving countries. In view of the meager resources of many sending migrants, however, fees need to come down.

Luckily, as banks have begun to compete with money transfer companies, remittance costs have fallen considerably. They vary by country and are as low as 4 percent in El Salvador and 5 percent in Mexico. However, the fees to remit funds to the Dominican Republic and Jamaica, the highest in the Caribbean region, average 8 percent and 12 percent, respectively.¹⁷

The International Remittance Network (IRNet) run by the World Council of Credit Unions (WOCCU), charges between $6.50 and $10 for transfers of up to $1,500. At present, only members of credit unions may send money, but WOCCU has proposed to the U.S. Congress that credit unions be allowedto provide such services to nonmembers.¹⁸

Financial institutions have also been devising new ways to send money, which they often advertise as lowering the costs of doing so (although it can be difficult to calculate the true costs of sending money). The SafeSend card from Bank of America, for example, is a stored-value card that can be sent to relatives or other recipients in Latin America and used to withdraw cash from automatic teller machines (ATMs) or to make purchases in stores. The bank offers lower sending fees to customers who also open bank accounts, including a new low-cost package designed for customers who typically do not maintain balances high enough to avoid maintenance fees on regular accounts. American Cash Exchange, based in New Jersey, has announced its “Poni PIN Card,” which works in ATMs. The purchaser of a Poni card scratches the back of the card to reveal a personal identification number (PIN), which he relays to the intended recipient by means of a free five-minute phone call that comes with the card; the recipient can access the funds by using a similar card obtained locally and the transmitted PIN. The company charges a $12 commission for the card, which on its smallest denomination, 1,000 pesos (roughly $90), is still high.

Last year, Democrat Charles Schumer introduced the Money WIRE Act in the Senate, which would force companies to disclose their fees on receipts and in their advertising. Luis Gutierrez, a Democrat from Illinois, has introduced a similar bill, the Wire Transfer Fairness and Disclosure Act, in the House of Representatives. Both are intended to keep immigrants from being overcharged. Neither has yet passed.

Meanwhile, officials in both sending and receiving countries have been trying to lower costs for money transfers in the hope that transmitters will pass on the savings to their customers. The Inter-American Development Bank has been helping the Dominican Republic, El Salvador, Jamaica, and Mexico put in place electronic systems that will help local banks link up to their foreign counterparts, lowering the costs of transferring money.

In an initiative modeled on U.S.-Canadian arrangements, the U.S. Federal Reserve Bank and the Bank of Mexico in 2003 created an automated clearinghouse that will reduce the costs of sending the roughly $14 billion that Mexicans in the U.S. wire to relatives back home each year—Mexico’s second-largest source of foreign exchange after oil. And under Treasurer Rosario Marin—who emigrated from Mexico to the United States as a child—the U.S. Treasury Department has partnered with Mexico’s federal savings bank, BANSEFI, to create an internet-based “People’s Network” program to channel remittances inexpensively to 3.8 million rural Mexicans. The program has also promoted micro-finance initiatives, another way of leveraging remittances to promote economic growth. In Mexico, it appears that remittance receivers are more likely to use the formal financial system. According to the Pew survey, 33 percent of remittance receivers in Mexico have bank accounts, as opposed to 22 percent of the general population.

Efforts to bring immigrants into the financial mainstream are under way in the United States as well, where as many as half of Latino immigrants do not have bank accounts. Unfortunately, these efforts have become highly politicized, especially for undocumented immigrants. Since 2002, when the Mexican government redesigned its matricula consular identification cards issued to Mexicans abroad to make them more secure, roughly 300 U.S. banks have begun to accept them as proof of identity for opening a bank account. This aroused the ire of House Immigration Reform Caucus chairman Tom Tancredo, a Republican from Colorado, who claimed that the ID cards legitimized illegal immigrants. In May 2003, Colorado became the first state to bar the use of Mexican consular cards and sparked a national anti-immigrationist furor that prompted congressional hearings on the issue. Last fall, the U.S. Treasury reendorsed the use of matriculas, but the controversy has continued as many states debate possible laws to allow or deny the practice.

In an environment that is often hostile to immigrants, the question inevitably comes up: does it hurt U.S. communities when remittance money leaves the country? In short, no. Though the funds are significant for receiving economies, they are a drop in the bucket in the overall U.S. economy, roughly equal to the dollar value of shares traded on the New York Stock Exchange in just two days. Moreover, remittances pale in comparison to the benefits to the United States from the lower wages paid to immigrant workers and the steady supply of labor they represent. To be sure, the money sent home might instead be used to better the lives of immigrants here. However, according to a new study by the Inter-American Development Bank, 93 percent of the $450 billion that U.S. Latino immigrants earn is spent here.¹⁹

The economist J. Edward Taylor has argued that those who send remittances are likely to be able to afford to do so. For one thing, he has found that families with children in the United States are likely to send less money back home than workers without children.²⁰ He also notes that home-country family ties provide a social safety net. This suggests remittances are well spent, in the sense that they reduce the cost to the U.S. economy when an immigrant facing health problems or other difficulties can return to his family in his home country.

Savings, Investment, and Community

Economists estimate that 90 percent of remittances are spent on consumer goods, and because of the multiplier effect that occurs as this money passes from one business to another, the impact of remittances on economic growth is probably close to double the actual amount sent. While this appears to be good news, many experts would not agree. This is because they say that money spent on consumer goods would be better spent on economic development.

However, the problem may not be as bad as it appears. First of all, a growing number of recipients surveyed by Pew— between a quarter and a third—put some portion of the money they receive in remittances into savings or small investments. Second, by some estimates, as much as a third of remittances go to housing construction, which is really investment, not merely consumption. Finally, policymakers are finding ways to use remittance flows— even those used for consumption—to open up sources of credit. The Inter-American Development Bank, for example, is testing a program that would allow applicants to count remittance income in their loan applications and thus increase their eligibility for credit. In several countries, remittance flows are adding to the amount of credit available to local businesses and consumers by allowing the banks that process them to borrow from the international capital markets at lower interest rates. By making use of a legal mechanism called securitization, which essentially guarantees that the hard currency generated by remittances will be dedicated to repaying investors, banks in Brazil, El Salvador, and Mexico have already raised hundreds of millions of dollars less expensively than they would have been able to do otherwise.²¹

Another question related to remittances is How can developing countries harness the resources available to émigrés on the higher end of the income scale, who are less likely than low-income migrants to send remittances but who are likely to be interested in investment or philanthropy for community development? Governments should—and are beginning to—make it easier for these immigrants to invest in their communities back home.

The Pan-American Development Foundation, a nongovernmental organization affiliated with the Organization of American States, works with hometown associations and other immigrant outreach groups that have been created to channel émigré dollars into development projects such as the building of churches, sanitation systems, roads, and hospitals, particularly in Haiti, El Salvador, and Mexico. Donations that immigrants make to the Transnational Community Development Fund are tax deductible in the United States, and some generate matching funds from the U.S. Agency for International Development. Similarly, the Mexican government matches, three for one, community or hometown association donations to local public works projects. So far, this governmental program has channeled $60 million into development projects.²² Under the Unidos por la Solidaridad (United in Solidarity) program run by El Salvador’s development agency, FISDL, and the government’s directorate for citizens abroad, DGACE, Salvadoran hometown associations can apply for matching federal development funds for social infrastructure projects. So far the hometown associations have donated $2.1 million for 45 projects, matched by $6.9 million in government funds.²³

To harness the migration machine for development, countries must find ways to harness the skills of better-off émigrés as well as their financial contributions, thus turning brain drain into “brain circulation.” India and China have had some success in this regard. A Public Policy Institute of California study of 2,300 foreign-born Silicon Valley engineers found that 51 percent had been involved with founding or running a start-up company in their home countries. Half traveled to their native countries at least once a year on business, and four out of five shared technology information and tips about U.S. job and business opportunities with colleagues back home.²⁴

Policymakers in other countries, like the Dominican Republic and Pakistan, are working on adapting such strategies to their own circumstances. Mexico’s “Godfather Program” seeks to lure Mexican émigré investments. Colombia’s Foreign Ministry has had some initial success with the Colombia Nos Une (Colombia Unites Us) program to promote technology transfer and investment from the 4.3 million Colombians abroad and the 440 homeland associations they have formed.

More than a dozen programs worldwide offer financial support to help returning migrants start businesses. France, for example, pays voluntary Malian returnees roughly the amount that it would have cost for the police to have deported them forcibly.²⁵ While such programs focus on deportable migrants, the principle of facilitating entrepreneurship in the homeland could be applied on a broader scale, and perhaps used to finance home-country training centers as well. This could be a cost-effective way of encouraging return migration and creating the jobs necessary for deterring future migration.

Various initiatives have attempted, with varying success, to train returning migrants to start and run businesses, and to provide financing for such endeavors. These include a joint program established in 1982 by the Sri Lankan Ministry of Labor and the Merchant Bank of Sri Lanka; an investment advisory service offered by Thailand’s Bangkok Bank in the 1980s; a technical assistance service created by the Overseas Pakistanis Foundation to help returnees channel savings, obtain credit, and navigate government bureaucracies; and the Philippine Overseas Employment Administration’s centers for training and financial support. However, these programs were successful only among returnees who had high skills and where the pool of local savings was deep enough to provide returnees with sufficient credit.²⁶ This brings us back to the point made earlier about the importance of channeling remittances into the formal financial system.

A Way Forward

Though remittances offer a powerful tool for development, they are no panacea, for they do have negative side effects. Because so many people in poor and developing countries have come to depend on them, they induce the flow of people across borders. Unless forward-looking policies are put in place, this migrant flow will continue indefinitely, along with the accompanying trauma to families and the social stresses on both sending and receiving countries.

Both sending and receiving countries must focus on the problems surrounding remittances. Governments should continue their efforts to reduce the bite that fees take out of remittances by promoting competition in the remittance market and encourage immigrants to use formal financial channels. They should dedicate more funds to matching-grant development programs and continue to subsidize home-country association efforts to improve the living conditions and economies of sending countries, so that their citizens do not feel compelled to migrate. Remittance flows can and should be used to promote savings and investment. And governments should encourage skilled migrants to return home and reverse the brain drain that occurs when countries have no choice but to send their most able citizens abroad to make ends meet.

This is the first part of a two-part article on remittances. The second part, which will appear in our fall issue, will focus on the political impact of global remittances in both home and host countries.