73744 THE WORLD BANK Migration and Development Brief Migration and Remittances Unit 17 December 1, 2011 Outlook for Remittance Flows 2012-14 Remittance flows to developing countries exceed $350 billion in 2011 By Sanket Mohapatra, Dilip Ratha and Ani Silwal 1 • Officially recorded remittance flows to developing countries are estimated to have reached $351 billion in 2011, up 8 percent over 2010. • For the first time since the global financial crisis, remittance flows to all six developing regions rose in 2011. Growth of remittances in 2011 exceeded our earlier expectations in four regions, especially in Europe and Central Asia (due to higher outward flows from Russia that benefited from high oil prices) and Sub-Saharan Africa (due to strong south-south flows and weaker currencies in some countries that attracted larger remittances). By contrast, growth in remittance flows to Latin America and Caribbean was lower than previously expected, due to continuing weakness in the U.S. economy and Spain. Flows to Middle East and Africa were also impacted by the “Arab Spring�. • Following this rebound in 2011, the growth of remittance flows to developing countries is expected to continue at a rate of 7-8 percent annually to reach $441 billion by 2014. Worldwide remittance flows, including those to high-income countries, are expected to exceed $590 billion by 2014. • However, there are serious downside risks to this outlook. Persistent unemployment in Europe and the U.S. is affecting employment prospects of existing migrants and hardening political attitudes toward new immigration. Volatile exchange rates and uncertainty about the direction of oil prices also present further risks to the outlook for remittances. • Remittance costs have fallen steadily from 8.8 percent in 2008 to 7.3 percent in the third quarter of 2011. However, remittance costs continue to remain high, especially in Africa and in small nations where remittances provide a life line to the poor. • There is a pressing need to improve data on remittances at the national and bilateral corridor level. This would make it possible to more accurately monitor progress towards the ‘5 by 5’ remittance cost reduction objective. Officially recorded remittance flows to developing countries are estimated to have reached $351 billion in 2011, an 8 percent increase over $325 billion in 2010 (table 1 and figure 1). 2 This rate of increase is higher than our earlier forecast of 7.3 percent growth (see Migration and Development Brief 1 We would like to thank Hans Timmer for extensive discussions, Indira Chand and Sonia Plaza for useful comments, and Uranbileg Batjargal for assistance with updating remittances data. 2 Data for 2010 were revised compared to the estimates reported in Migration and Development Brief 16, to reflect smaller inflows reported by Romania, Serbia and Sudan. Flows to China and India were larger than previously estimated. 1 16 released on May 23, 2011). Worldwide recorded remittance flows, including flows to high-income countries, are estimated to have reached $483 billion in 2011. Remittance flows grew at a faster pace in four of the six developing regions in 2011 than earlier forecasts, with the exception of Latin America and the Caribbean and the Middle East and North Africa (table 1). Remittance flows to Latin America and the Caribbean are estimated to have increased by 7 percent in 2011 after remaining almost flat the previous year. Flows to countries in Asia were buoyed by high oil prices and increase in remittance outflows from Russia to Central Asia and from the Gulf Cooperation Council (GCC) countries to South and East Asia. Remittances to East Asia and Pacific region are estimated to have grown by 7.6 percent in 2011, to Eastern Europe and Central Asia by 11 percent, and to South Asia by 10.1 percent. The depreciation of the currencies of some large receiving countries (including Mexico, India and Bangladesh) created incentives to send remittances to take advantage of the “sale effect� on local currency assets. However, flows to countries in the Middle East and Africa grew the slowest (2.6 percent) because of civil conflict and unrest related to the “Arab Spring�. Despite the crises in North Africa and the difficult economic situation in Europe, remittance flows to Sub- Saharan Africa are estimated to have increased by 7.4 percent in 2011. Even though remittances to developing countries grew in 2011, they are vulnerable to the uncertain economic prospects in the migrant destination countries. In line with the World Bank’s latest outlook for the global economy, remittance flows to developing countries are expected to grow by 7.3 percent in 2012, 7.9 percent in 2013 and 8.4 percent in 2014, to reach $441 billion by 2014 (see Migration and Development Brief 12 for the methodology for these forecasts). These forecasted rates of growth are considerably lower than those seen prior to the global financial crisis, when the annual increases in remittances to developing countries averaged 20 percent during 2003-08 (see figure 1). Worldwide remittance flows, including those to high-income countries, reached $483 billion in 2011 and are expected to increase to $593 billion by 2014. The top remittance recipient developing countries in nominal terms and in relation to the GDP are listed in box 1. Issues relating to the timeliness and quality of remittances data are discussed in box 2. Figure 1: Remittance flows to developing countries have recovered after the global financial crisis, but are forecast to grow at a slower pace in 2012-14 Percent Growth of $ billions 30% remittances inflows Remittance inflows 400 (right scale) 20% 300 10% 200 0% 100 -10% 0 e = estimate; f = forecast Source: Same as table 1 Risks to the forecasts: There are several sources of vulnerability to the forecasts for remittances to developing countries. The ongoing debt crisis in Europe and high unemployment rates in high-income OECD countries are adversely affecting the economic and employment prospects of 2 migrants. These persistently high unemployment rates have created political pressures to reduce the current levels of immigration (see discussion later). There are risks that if the European crisis deepens, immigration controls in these countries could become even tighter. This would affect remittance flows to all regions – especially to countries in Eastern Europe and Central Asia (e.g. Romania, Bulgaria, Moldova) and North African countries that have a large share of their emigrants in Europe (see figure 2 on remittances by source). A deepening and spread of the European debt crisis will also pose risks for oil prices, which could in turn reduce demand for migrant workers and depress remittances flows to Asian countries. Volatile and unpredictable exchange rates present further risks to the outlook for remittances. Box 1: Top recipients of remittances among developing countries The new estimates show that the top recipients of remittances among developing countries in 2011 were India ($58 billion), followed by China ($57 bn.), Mexico ($24 bn.), and the Philippines ($23 bn.). Other large recipients in US dollar terms include Pakistan, Bangladesh, Nigeria, Vietnam, Egypt and Lebanon (box figure 1). However, small and low-income countries such as Tajikistan, Lesotho, Nepal, Samoa and Tonga tend to receive more remittances as a share of their gross domestic product (GDP) (box figure 2). Box figure 1: Top 10 recipients of migrant Box figure 2: Top 10 recipients of migrant remittances remittances as a share of GDP (US$ billion, 2011e) (% of GDP, 2010) 58 57 31 29 25 23 21 20 20 20 17 16 24 23 12 12 11 9 8 8 Remittance inflows and gross domestic product (GDP) in box figure 2 are for 2010. See box 2 on issues with remittance inflows data for Lebanon. Source: Same as table 1. Recent trends in remittance flows to developing regions The distribution of emigrants across the destination countries and the economic and employment prospects for migrants in these destinations influence remittance flows to developing countries. Figure 2 presents updated estimates of “bilateral remittances� for 2010 from five main source regions to the six developing regions. 3 Expectedly, the US is by far the largest source of remittances for developing countries in the Latin America region, with Western Europe, primarily Spain, coming a distant second. 3 See Ratha and Shaw 2007 “South-South Migration and Remittances� World Bank Working Paper No. 102. 3 East Asia is relatively more diversified in terms of migration destinations and sources of remittances, while a significant share of South Asia’s remittances come from the six GCC countries. Figure 2: Sources of remittances for developing regions in 2010 6% 4% 7% 6% 5% 13% Developing 14% 11% 34% 11% 8% 45% 9% Other high 28% income 14% 53% 41% GCC 6% 10% 80% 40% 39% Western Europe 12% 33% 28% US 15% 18% 11% EAP ECA LAC MNA SAS SSA EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and Caribbean; MNA = Middle East and North Africa; SAS = South Asia; SSA = Sub-Saharan Africa Source: Model-based estimates of bilateral remittances using methodology of Ratha and Shaw (2007) and data from World Development Indicators and Migration and Remittances Factbook 2011. And while Western European migrant destinations are the major source of remittances for the Europe and Central Asia region, a third of the region’s remittances originate in developing countries such as Russia. Western Europe is also the most important source of remittances for developing countries in the African continent and the Middle East, reflecting their geographical proximity to Europe, and the GCC countries account for more than one-quarter of inflows to the Middle East and North Africa. Flows to Latin America and Caribbean recover Flows to Latin America and the Caribbean region have resumed growth in 2011 after a sharp decline during the global financial crisis and after remaining almost flat in 2010 (figure 3). Available data until the third quarter of 2011 for Mexico, Colombia, El Salvador, Guatemala, Honduras, Jamaica, and Nicaragua - which together account for three-quarters of remittance flows to the Latin America and Caribbean region - show that remittance inflows grew by nearly 7 percent in the first three-quarters. Remittances to Mexico surged by 11 percent in the third quarter, in part because of the depreciation of the Mexican Peso relative to the US dollar (see next section and figure 8). New housing construction in the US, traditionally a large employer of Latin American migrants, appears to have stabilized after a steep fall since 2006 (figure 4a). 4 The number of permits issued for new privately-owned housing grew in October 2011 by 17.7 percent from a year earlier and by 10.9 percent from a month earlier. 5 Migrant employment in construction experienced a seasonal recovery during summer, but remains well below the pre-financial crisis levels (figure 4b). By comparison, migrant employment has been relatively stable in wholesale and retail trade, and in restaurants and hotels, but has declined in manufacturing (with a small uptick in September 2011). Since the start of the financial crisis in September 2008, the employment of migrants in the US has declined less (-3.7 percent) than natives (-4.1 percent). 4 The US hosts some three-quarters of emigrants from countries in the Latin American and Caribbean region. 5 US Bureau of Labor Statistics (Nov. 17, 2011) 4 Remittances to Latin America have also been affected by the slowdown and spike in unemployment in Spain, which hosts about one-tenth of all Latin American migrants, including from Ecuador, Colombia, Bolivia, Argentina and Peru. However, remittances outflows from Spain have remained fairly resilient (see discussion later) as migrants have cut into savings and even consumption in order to send remittances, and perhaps to prepare for an eventual return. Figure 3: Remittance flows to Latin America and Caribbean have resumed growth, but are vulnerable to prospects for economic recovery in the US and Spain Mexico percent* El Salvador Honduras Guatemala Dominican Rep. 25 Jamaica 15 5 -5 -15 -25 Mar-09 Sep-09 Mar-10 Sep-10 Mar-11 Sep-11 *Year-on-year growth of 3-month moving averages. Source: Central banks of the respective countries and World Bank staff estimates The prolonged economic downturn in the US up until the second quarter of 2011 has resulted in several US states implementing legislation to control illegal immigration, most notably Arizona and Alabama. The stock of Mexican migrants in the US has reached a plateau, with smaller numbers of Mexican immigrants entering the US, and more returning home, in part because of tighter border enforcement and increasing deportations by the border patrol agency. 6 Figure 4a: Construction sector in the US has stabilized US housing starts, annualized (Millions units) 2.0 1.5 1.0 0.5 0.0 * New residential housing construction started, 3-month moving average Source: US Census Bureau 6 “Immigration from Mexico in fast retreat, data show,� Los Angeles Times, Nov. 15, 2011. 5 Figure 4b: Migrant employment in the US appears to be recovering Migrant employment in US recovering faster US migrant employment by sector* than natives* millions millions millions 24 3.1 125 Wholesale and Migrants 2.9 retail trade Natives (right scale) 120 (left scale) 23 2.7 2.5 Manufacturing Restaurants 115 22 2.3 and hotels 2.1 Construction 110 21 1.9 Sep-08 Mar-09 Sep-09 Mar-10 Sep-10 Mar-11 Sep-11 Dec-09 Jul-10 Feb-11 Sep-11 *3-month moving averages Source: Current Population Survey *3-month moving averages Source: US Current Population Survey, US Census Bureau Box 2: Need for improving remittance flows data There is a pressing need to improve data on remittances at the national and bilateral corridor level. Remittances data at the national level is of relatively poorer quality in many countries than data on foreign direct investment and official flows. Remittance inflows often include other items other than those sent by migrants, such as small value trade and payments transactions and in some cases, even transfers to non- government organizations (NGOs) and embassies. The World Bank, IMF, OECD and Eurostat have worked together with central banks and national statistical offices to introduce new definitions of “Personal remittances� in the sixth edition of the IMF’s Balance of Payment Manual (BPM6). However, some countries still compile remittances data using the Balance of Payments Manual (BPM4), which can result in significant underreporting of inflows. For instance, while Ghana reported remittance inflows of $136 million to the IMF Balance of Payments in 2010, Ghana’s central bank reports inflows in excess of $1.8 billion. There are also issues with consistency of definitions and coverage of migrant remittances, especially during times of crisis. For example, preliminary Balance of Payments data on Lebanon’s workers’ remittance inflows report a significant decline in 2010, but statements by central bank officials suggest that these flows have remained relatively resilient. In the absence of more definitive information, we have kept remittance flows to Lebanon at the same level in 2010 and 2011. Bilateral remittances data often do not exist. In a handful of cases where data on the sources of remittances are published by central banks (for example, by the Philippines), these flows are often attributed to international clearing centers through which remittances are routed, instead of the countries where the remittances actually originate. Improving remittances data will help in stimulating market competition and more accurate monitoring of progress towards internationally-agreed remittance cost reduction objectives. Better data will also help in evaluating the market for innovative financing instruments such as diaspora bonds. The World Bank is leading a Global Remittances Working Group which includes improving remittances data as one of its four thematic areas. Source: International Transactions in Remittances: Guide for Compilers and Users, International Monetary Fund 2009 and authors’ elaborations. 6 Oil-driven economic activities in the Gulf and Russia are providing a cushion for remittances to Asia High oil prices, which have hovered over $100 a barrel in recent months (figure 5), continue to provide a much-needed cushion for migrant employment in, and remittance flows from, the GCC and Russia. Oil driven economic activities and increased spending on infrastructure development are making these destinations attractive for migrants from developing countries. Remittances from the GCC countries to Bangladesh and Pakistan (where the GCC countries account for 60 percent or more of overall remittance inflows) grew by 8 percent and 31 percent respectively in the first three quarters of 2011 on a year-on-year basis. 7 Flows from the high-income countries in the Middle East to the Philippines also grew by nearly 8 percent during this period. 8 Similarly, outflows from Russia, mainly to Central Asian countries, have increased with the recovery of oil prices, but appear to have become more volatile in the post-crisis period (figure 5). Figure 5: Oil prices remain a key driver of remittances to Central Asia $ billions $/barrel 140 8 120 Outward remittances 100 6 from Russia 80 4 60 40 2 Crude oil price 20 (right scale) 0 0 Source: Central bank of Russia, World Bank commodity price database. As a result of the increased demand for migrant workers, migrant deployments from Bangladesh grew strongly, by 37 percent, in the first three-quarters of 2011 (after registering a 20 percent decline the previous year). Remittances also grew in the double digits during this period, but this growth appears to have decelerated after the Eid ul-Fitr festivities in August, but from a high base (figure 6). Although oil driven economic activities have provided a cushion for remittances to Asian countries, remittance flows to India (the largest recipient among developing countries) appear to have been relatively more affected by the weak employment in the US and by the debt crisis in Europe. Quarterly data show that private current transfers – composed mostly of migrant remittances – increased by 33 percent in the first quarter of 2010 in a recovery to the pre-crisis levels of early 2008 (figure 7). However, growth of remittance inflows has been anemic in the subsequent period, averaging just 4 percent since the second quarter of 2010. However, remittance flows, especially from the GCC, are reported to have 7 The Pakistan Remittance Initiative (PRI), a joint initiative of the central bank and Pakistan’s government, has been working actively with commercial banks and money transfer operators to lower the cost of inward remittances and to improve the payments systems and delivery channels in order to bring a larger share of remittances into formal channels. 8 While the Middle East countries accounted for only 16 percent of remittance inflows to the Philippines in 2010, the large share of deployments of overseas Filipino workers (OFWs) in the Middle East (more than 60 percent of all land based workers in 2010) makes it likely that actual volume of remittances is likely to be higher. The sources of remittance flows are sometimes attributed to international banks and clearing centers through which flows are routed, instead of where they originate (See box 2 on improving bilateral remittances data). 7 surged in the second half of 2011 because of a weakening rupee (see next section) which may have prevented remittances from slowing even further in the second half of 2011. Figure 6: Migrant deployment from and remittance flows to Bangladesh have resumed Thousands Year-on-year growth (%)* 80 50% 70 40% 60 Remittances 30% (right scale) 50 20% 40 10% 30 0% Migrant deployment 20 -10% Nov-08 Apr-09 Sep-09 Feb-10 Jul-10 Dec-10 May-11 Oct-11 * Growth of 3-month moving averages. Source: Bangladesh Bank Figure 7: Remittance flows to India appeared to have plateaued in the first half of 2011 India - private transfers US$ bn. Percent Quarterly private transfers 15 60% 10 40% 5 20% 0 0% -5 Year-on-year -20% growth (right scale) -10 -40% Source: Reserve Bank of India. Currency depreciation in many recipient countries is increasing the migrants’ incentive to remit After several years of appreciation, the sharp depreciation of the currencies of some major recipients relative to the US dollar – and the currencies in remittance source countries in the GCC that are linked to the dollar – in recent months is creating a “sale effect� in the assets of the countries of origin (figure 8). The higher purchasing power of each dollar of remittances may increase the incentive to remit, to take advantage of the higher purchasing power in the home country (see also Migration and Development Brief 11). The Mexican Peso depreciated by nearly 14 percent between July and November 2011 (see figure 8), while remittances to Mexico increased by 11 percent in the third quarter on a year on year basis (and by 22 percent in September) more than doubling from a 5 percent average growth in the first half of the year. While a part of this increase is because of the economic recovery in the US (see section on remittances to Latin America), this is also likely to be related to the increased purchasing power of remittances. Bangladesh has experienced a similar phenomenon of a depreciating currency, but over a 8 longer period, and the growth of official remittances appears to have decelerated possibly because of a diversion to informal channels (see previous section). Figure 8: Sharp depreciation of currencies in some major recipients of remittances has increased incentives for sending remittances Local currency per US$ (July 2010=100) 110 105 Indian Rupee 100 Mexican Peso (right scale) Bangladeshi Taka 95 90 Source: World Bank staff calculations. The Indian rupee steadily appreciated (by about 15 percent) against the US dollar between November 2008 and June 2011, but had more than reversed the gains by November. Anecdotal reports from money transfer companies suggest that remittances from the GCC countries surged in the third and fourth quarters of 2011 because of the weak Rupee. 9 A similar weakening of the Kenyan Shilling has been attributed to a surge in remittance inflows to Kenya in August-September. 10 Diversification of migration destinations and demand for seafarers has lent resilience to remittances to the Philippines In the Philippines, remittance inflows grew steadily by over 7 percent during the first three-quarters of 2011. This is largely because demand for overseas Filipino workers (OFWs) has remained remarkably stable: preliminary data show that OFW deployments increased by nearly 20 percent between 2008 and 2010 during a time of global economic crisis. While deployment of land-based OFWs to the Americas and Europe declined by 12 percent between 2008 and 2010, this was more than offset by an increase in deployments to the Middle East (8 percent increase, mainly to Saudi Arabia and the UAE), Asian countries (28 percent increase) and Africa (53 percent increase). Among Asian countries, Hong Kong and Singapore have seen large increases in OFW deployments. The deployment of seafarers, who account for one-quarter of overall OFW deployments, increased by about 33 percent between 2008 and 2010 and is expected to increase by a further 15 percent in 2011. 11 The rise in seafarers and the increasing diversification of Filipino migrant destinations has provided resilience to remittances. However, indigenization programs being considered or implemented in the GCC countries (for example, the Nitaqat program in Saudi Arabia) have raised concerns of adverse implications for future remittances to Philippines, Pakistan, Bangladesh and other migrant-sending countries (see box 3). There 9 “Dirham flirts with Rs14; Remittances surge� Nov 20, 2011. Emirates 24/7 News. 10 “Weak shilling boosts value of remittances� Business Daily Africa, November 9 2011 (http://www.businessdailyafrica.com/Weak+shilling+boosts+value+of+remittances/-/539552/1269628/-/1421x6yz/-/) 11 “400,000 Pinoy seafarers deployment seen� in 2011, Philippines Information Agency (PIA), Nov 20, 2011 (http://www.pia.gov.ph/?m=1&t=1&id=64840) 9 are also concerns about the minimum wages and welfare of female domestic workers who account for a large share of OFWs in Saudi Arabia. 12 While the indigenization may not affect remittances in the near term, they highlight the importance of destination country policy changes for the future sustainability of remittance flows to developing countries. Box 3: Indigenization programs in the GCC are likely to have minimal implications for remittances Saudi Arabia's recent indigenization effort titled "Nitaqat" came into effect on September 10, 2011. Saudi firms have been color coded to four categories - Red, Yellow and Green, and Excellent. Firms labeled "Red" will not be able to renew their foreign workers' visas and were given a November 26, 2011 deadline to improve their status by hiring more native Saudis. "Yellow" firms have until February 23, 2012 to improve their status and will not be allowed to extend their existing foreign employees' work visas beyond six years. "Green� firms with high Saudization rates will be allowed to offer jobs to foreign workers that are employed by firms in the Red and Yellow categories and transfer their visas. And firms in the highest “Excellent� category will enjoy the ability to hire workers from any part of the world using a web-based system with minimal clearance. “Saudization� or the policy of encouraging employment of native Saudis in the private sector is not new. Similar efforts since at least 2004 have met with little success, with firms using various ruses or “wasta� to meet the quota requirements. Although the current Nitaqat program is better formulated, with quotas that take into account the existing employment patterns, such indigenization efforts are unlikely to adversely affect the longer-term employment prospects of migrants in the Gulf. Construction firms with among the lowest quota will still find it difficult to comply as Saudi natives would be reluctant to work under the harsh conditions and low wages that migrants willingly accept. There would also likely be a dearth of Saudi natives willing to work as assistants and cashiers in retail stores. Domestic household workers, mainly from the Philippines and from Sri Lanka, have been exempted from the Nitaqat program. At the other end, financial firms and banks will find it difficult to reach the high skill requirements. In the near term, the program could increase wages and reduce productivity and competitiveness as the labor market goes through a period of adjustment. This makes it likely that the Nitaqat program deadlines would be extended and the rules made more flexible. The indigenization program is likely to have limited implications for remittances in the medium term. The magnitudes of intended and expected impact of programs for job creation for Saudi nationals is relatively small compared to the migrant worker population in Saudi Arabia. Saudi Arabia has nearly six migrant workers for each Saudi native in the private sector labor force. The Saudi private sector will continue to predominantly rely on foreign labor in the foreseeable future, and the opportunity of substitution between Saudi job seekers and migrant workers in the production process continues to be heavily constrained. Indigenization efforts will displace only a small fraction of foreign workers, if at all. Source: Saudi Arabia Ministry of Labor; authors’ elaborations. Difficult economic situation in Europe and tighter immigration policies are affecting remittances Migrants in several large destination countries in Europe had significantly higher unemployment than natives in 2010 (see figure 9), in part because their unemployment rate increased more than natives in most European countries during the crisis (with the exception of Germany and UK). This trend 12 Philippines and Saudi Arabia have seen contentious negotiations over the minimum wages paid to female Filipino domestic workers (“Phl rejects appeal from Saudi Arabia to cut minimum wage for Pinoy maids� May 21 2011 http://www.philstar.com/Article.aspx?publicationSubCategoryId=63&articleId=688220) 10 is likely to have continued into 2011. High domestic unemployment rates have created additional pressure to reduce the levels of migration and to impose more restrictive migration policies. The UK has imposed tougher admission criteria for non-EU migrants, with the objective of reducing annual immigration from the hundreds of thousands to tens of thousands. Employers seeking to hire non-EU migrants are now subject to an annual quota and the “shortage occupation list� has been shrunk. Policies have also been introduced that have made it difficult for foreign students to come to the UK as students. The UK Home Office is also considering legislation that would prevent foreign workers earning below a certain threshold to bring their families to the UK. 13 However, employers and universities warn that such restrictions may reduce future productivity and growth. 14 Moreover, many unskilled or semi- skilled migrants (mainly from Eastern Europe) voluntarily left after the crisis, in part because of better prospects in their own countries and the appreciation of their currencies relative to the British Pound, which reduced their wages in home currency. Figure 9: Migrants had higher unemployment rates in European countries in 2010, after experiencing steeper increases during the global financial crisis Unemployment rate in 2010 (%) Increase in unemployment rate between 2008 30 15 and 2010 (%) 25 Native 10 Native 20 Foreign-born Foreign-born 15 5 10 0 5 0 -5 Source: OECD Since the crisis, Spain has also introduced new policies that have made burdensome the process of hiring foreign workers by employers, in particular new minimum salary requirement and discontinuation of an expedited immigration processing option for large businesses. 15 Spain has also seen migrants return to their home countries since the end of the construction boom in 2007. These include migrants from Ecuador, Colombia, Argentina and Peru as well as low-skilled migrants from North Africa who have moved on to other European countries. 16 Despite the high levels of unemployment of migrants in Spain, remittance outflows grew by about 15 percent in the first half of 2011, as migrants cut into their savings (and even consumption) to be able to send remittances, and also to prepare for a potential return if the crisis deepened further. Quarterly remittance outflows data up until the second half of 2011 suggests that the growth of remittance outflows from Western European countries have increased from the levels seen at the 13 “UK family migrations proposed,� BBC, Nov. 16, 2011. 14 “Student visa curbs are damaging our reputation, Universities UK warns,� The Guardian, Nov. 1, 2011. 15 “Spain: New Regulations Cause Processing Difficulties for Employers,� Totallyexpat.com, Nov. 8, 2011. 16 “Spain sees exodus as crisis deepens,� Financial Times, Oct. 17, 2011. 11 height of the global financial crisis in 2008-09, but there has not been a sustained recovery because of the weak economic and employment situation in these countries (figure 10). The UK’s recent restrictions on immigration of non-EU nationals, as well as the return of many Eastern European migrants, may be a factor in the weak recovery of remittance outflows from the UK. Outward flows from UK, Italy and Spain are still well below the levels prior to the crisis, while outflows from France have been declining consistently since the last quarter of 2008. Figure 10: Continued economic uncertainty has led to weak recovery of outward flows from Europe Percent 40% 20% 0% -20% -40% France Spain UK Italy Source: IMF Balance of Payments; World Bank staff calculations. The “Arab Spring� has affected migrants within the region and in Sub-Saharan Africa The full impact of the “Arab Spring� is not yet visible in data. But the crisis in Libya has resulted in mass departures of migrants to their home countries or to neighboring countries. Hundreds of thousands of Egyptians have returned home since the crisis began, causing a deceleration in the growth of remittances to the country in 2011. 17 Many Tunisians have returned home from Libya, which has adversely affect remittance inflows. 18 Migrants from Africa have been particularly hit by the crisis. According to the Government of Niger, some 200,000 Nigeriens have returned to the country after the crisis began in Libya. Chad has also seen many migrants in Libya return home after the crisis, causing a burden on scarce resources in the country as well as loss of future remittances. 19 Migrants from many other Sub-Saharan African countries, often unskilled and illegal, have been stuck and vulnerable to abuse in Libya. Although most Bangladeshis living in Libya were reported to have returned home immediately after the crisis erupted in early-2011, the effect on remittances has been relatively minor as Bangladeshis in Libya were a small share of all Bangladeshis abroad (Migration and Development Brief 16). Despite the crises in North Africa and the difficult economic situation in Europe, remittance flows to Sub-Saharan Africa are estimated to have increased by 7.4 percent in 2011. Remittances from Kenyan migrants grew to $644 million in the first nine months of 2011. Inflows surged by 45 percent on a year- on-year basis in part because the weak Kenyan shilling made it more attractive to invest in local currency assets (see previous section). Remittance flows to Ethiopia are reported to have increased to more than $1.5 billion in the 2010-11 fiscal year. 20 17 "Egyptians' remittances dip due to Libya conflict" Ahram, August 11, 2011. 18 "Libyan Crisis Badly Affects Tunisian Economy" Ahram, August 11, 2011. 19 "Chad-Libya: Agencies prepare for more migrants," irinnews.com, May 25, 2011. 20 http://www.thereporterethiopia.com/News/remittance-spikes-by-88-percent.html 12 Trends in remittance costs The G8 and the G20 countries have agreed to the objective of reducing global average remittance costs by 5 percentage points in 5 years (“5 by 5�objective). 21 The World Bank’s Remittance Prices Worldwide database shows that the simple average remittance cost at the global level declined between 2008 and the first quarter of 2010, but then appears to have increased in the subsequent quarters (figure 11). 22 The average remittance cost, weighted by bilateral remittance flows, however, has registered a consistently declining trend. Average remittance costs fell from 8.8 percent in 2008 to 7.3 percent in the third quarter of 2011. There is evidence that costs have been falling in high volume remittance corridors, such as from the US to Mexico, UK to India and Bangladesh, and France to North Africa. Figure 11: Trends in global average remittance costs Average remittance costs Percent 10.0 9.0 8.0 7.0 2008 Q1 Q3 Q1 Q3 Q1 Q3 2009 2009 2010 2010 2011 2011 Simple average Weighted average* * Average of corridor level remittance costs weighted by updated bilateral remittance estimates of Ratha and Shaw (2007) for a balanced sample of country corridors starting from 2008 onwards. Country corridors and RSPs where exchange rates applied to 23 the transaction were not provided (including corridors originating in Russia) were excluded from the sample. Source: Migration and Remittances Unit calculations using Remittance Prices Worldwide (RPW) database and estimates of bilateral remittance flows from Ratha and Shaw (2007) Although the simple average cost of sending remittances to Sub-Saharan Africa is the highest among the six developing regions, the weighted average costs for the Middle East and North Africa and East Asia and Pacific are higher (figure 12). This is because these regions have several large remittance corridors with relatively higher costs. For example, it is more expensive to send remittances from France to Morocco or from US to China than it is from UK or US to Nigeria. However, Sub-Saharan Africa, excluding Nigeria, has many smaller remittance corridors with significantly higher costs. 21 At their summit in L’Aquila in July 2009, the G8 Heads of State and Governments made a pledge to achieve the objective of a reduction of the global average costs of transferring remittances from 10 percent to 5 percent in 5 years (the “5 by 5� objective). A 5 percentage point reduction of the global average cost of remittances’ flows is believed to translate into an additional US$ 15 billion annually for recipient populations. 22 The World Bank’s Remittance Prices Worldwide database monitors remittance costs in 213 (mostly North-South) remittance corridors (see http://remittanceprices.worldbank.org/). 23 See “Trends in the Average Cost of Migrant Remittance Services� by Massimo Cirasino and Hemant Baijal, AccessFinance, Issue #35 February 2011 (http://newsletters.worldbank.org/newsletters/listarticle.htm?aid=237987 – see footnote 4 in p. 5) 13 Figure 12: Average cost of sending remittances to developing regions Weighted average cost for sending $200 in Q3 2011, by recipient region (US$) 19.0 18.0 16.6 14.8 12.8 9.6 South Asia Latin Europe & Sub-Saharan East Asia & Middle East America & Central Asia Africa Pacific & North Caribbean Africa * Average of corridor level remittance costs weighted by updated bilateral remittance estimates of Ratha and Shaw (2007). Country corridors and RSPs where exchange rates applied to the transaction were not provided (including corridors originating in Russia) were excluded from the sample. Source: World Bank staff calculations using Remittance Prices Worldwide (RPW) database Remittance costs on average continue to remain high, impacting many poor migrants and their families. Efforts to reduce remittance costs include increasing market competition in many remittance corridors and wider application of cheaper and more convenient remittance technology. Also there is a great need to improve data on remittances and migration at the national and bilateral corridor levels, for more accurate monitoring of progress towards the internationally-agreed “5 by 5� objective. G20 and Diaspora Bonds In the search for additional sources of financing for development, and in the effort to increase the development impact of remittances and diaspora contributions, diaspora bonds have received attention from the G20 and the recent Gates Report. The annual savings of the diasporas are large, more than $50 billion for Africa alone (see Migration and Development brief 14). Mobilization of these savings to finance projects such as railways, roads, power plants, and educational institutions is possible through the issuance of retail diaspora bonds. The diasporas’ desire to give back to the country of origin can potentially help lower the cost of financing. The diasporas’ home bias and a lower perception of sovereign risk can make diaspora investments more stable relative to foreign investments (see Okonjo- Iweala and Ratha 2011). 24 Several countries (including Nepal, Ethiopia and Kenya) have recently issued diaspora bonds or have earmarked a portion of their bond issuance for diaspora investors. Nigeria has indicated its intention to issue a diaspora bond for infrastructure development in 2012. Several others, including Bangladesh and Rwanda, are considering issuing diaspora bonds. 25 A better knowledge of the location of the diaspora and their profile, such as age, education, occupation, sector of work, income, wealth, savings, remittances and investments are key for appropriately structuring and marketing the innovative financing instrument. This would enable issuers to assess the potential demand for diaspora bonds prior to coming to the market. 24 “A Bond for the Homeland.� By Ngozi Okonjo-Iweala and Dilip Ratha, Foreign Policy, May 24 2011. 25 “Diaspora Bonds Gain Momentum.� (Available at http://go.worldbank.org/WC69CPEP60) 14 Table 1: Outlook for remittance flows to developing countries, 2012-14 2008 2009 2010 2011e 2012f 2013f 2014f $ billions All developing countries 324 307 325 351 377 406 441 East Asia and Pacific 85 85 94 101 109 117 127 Europe and Central Asia 45 36 36 40 44 48 53 Latin America and Caribbean 64 57 57 61 66 71 77 Middle-East and North Africa 36 34 35 36 37 39 42 South Asia 72 75 82 90 97 105 114 Sub-Saharan Africa 22 20 21 23 24 26 28 LDCs (UN-classification) 23 24 25 27 29 32 35 Low-income countries 22 23 25 28 30 33 37 Middle income 302 284 301 324 347 373 404 High income 132 123 124 132 137 144 152 World 456 429 449 483 514 550 593 Growth rate (%) All developing countries 16.4% -5.2% 6.0% 8.0% 7.3% 7.9% 8.4% East Asia and Pacific 18.8% 0.4% 10.2% 7.6% 7.3% 8.0% 8.7% Europe and Central Asia 16.3% -19.8% -0.1% 11.0% 8.8% 10.1% 11.4% Latin America and Caribbean 2.2% -12.2% 1.2% 7.0% 7.6% 7.9% 8.1% Middle-East and North Africa 12.0% -6.7% 3.3% 2.6% 5.0% 5.3% 5.5% South Asia 32.6% 4.8% 9.5% 10.1% 7.4% 7.9% 8.4% Sub-Saharan Africa 15.8% -7.0% 4.5% 7.4% 6.3% 6.8% 7.3% LDCs (UN-classification) 32.5% 2.0% 4.9% 8.5% 8.5% 9.1% 9.6% Low-income countries 32.8% 3.7% 8.9% 12.0% 9.4% 10.4% 11.5% Middle income 15.4% -5.9% 5.8% 7.6% 7.1% 7.6% 8.2% High income 14.7% -7.1% 1.1% 6.3% 4.1% 4.9% 5.7% World 15.9% -5.8% 4.6% 7.5% 6.4% 7.1% 7.7% e= estimate; f=forecast Source: World Bank staff calculation based on data from IMF Balance of Payments Statistics Yearbook 2011 and data releases from central banks, national statistical agencies, and World Bank country desks. See Migration and Development Brief 12 for the methodology for the forecasts. Remittances are defined as the sum of workers’ remittances, compensation of employees, and migrant transfers – see www.worldbank.org/prospects/migrationandremittances for data definitions and the entire dataset. Migration and Development Briefs are prepared by the Migration and Remittances Unit, Development Economics (DEC) and Poverty Reduction and Economic Management (PREM) network. These briefs are intended to be informal briefing notes on migration, remittances, and development. The views expressed are those of the authors and may not be attributed to the World Bank Group. The latest data on remittances and other useful resources are available at http://www.worldbank.org/migration. Our blog on migration titled “People Move� can be accessed at http://blogs.worldbank.org/peoplemove. Contributions, feedback, and requests to be added to or dropped from the distribution list, may be sent to Dilip Ratha at dratha@worldbank.org. 15