Energy Cooperation in South Asia: Prospects and Challenges

A, Ahmed*, C. M Shahariar* and  M. A. Rashid Sarkar**

*          Graduating term student, Department of Mechanical Engineering, BUET

**        Professor, Department of Mechanical Engineering, BUET

Abstract

Energy Security has always been a concern for all the countries in the world, as with modernization, the use of energy has been increasing rapidly and people’s life and national economy becoming dependant on the usage of energy. So, to maintain a secured future, different initiatives are being taken in the energy sector all over the world. South Asian region is vital part of this world with its huge population base and rapid growing economy marked by increasing GDP. To facilitate the objective of this region to progress towards better economy, South Asian Association for Regional Cooperation (SAARC) was formed in 1985. But, due to some lack of trust between the member states and political dilemma, this association has never flourished to its full potential. However, through building mutual trust and understanding the greater perspective of this region, SAARC Member States (SMSs) has a big scope to create a strong regional cooperation which includes bilateral and multilateral energy trade between the states and a generalized energy policy which can provide a guiding principle for energy security in the future. With the present scenario of energy reserve, production and consumption of the countries of this region, it can be seen that there is a dire need of cooperation among the countries. Observing the successful regional cooperation in other parts of the world, monitoring the benefits of this kind of cooperation, it is high time that the leadership of the SMSs proceed towards a more secured future in the energy sector by pursuing strong regional cooperation.

1.      Introduction

            According to the United Nations geographical region classification Southern Asia comprises the countries of Afghanistan, Bangladesh, Bhutan, India, Iran, Maldives, Nepal, Pakistan, and Sri Lanka. By other definitions and interpretations (see below), Burma and Tibet are also sometimes included in the region of South Asia. South Asia is home to well over one fifth of the world's population, making it both the most populous and most densely populated geographical region in the world.[3] The South Asian Association for Regional Cooperation is an economic cooperation organization in the region.

            To bring South Asian nations under one roof and increase regional cooperation in different sectors, SAARC was established in 1985. Although it covers only 3.95% of global landmass, SAARC countries are inhibited by 23.7% of global population. The South Asian Association for Regional Cooperation (SAARC), comprising Afghanistan, Bangladesh, Bhutan, India, the Maldives, Nepal, Pakistan, and Sri Lanka, is characterized by desolate contrasts. The total gross domestic product (GDP) of the region amounted to $3,860 billion in 2009, but was less than 3% of the world’s total GDP. Still it is one of the rapidly emerging regions in the world. It experienced an average annual GDP per capita real growth rate of 5.2% from 2000 to 2009. Countries such as Bhutan and India have developed at even higher per capita

            GDP growth rates of 6.8% and 5.6%, respectively [1]. This economic growth is anticipated to assist the mitigation of extensive poverty in the region. To support and quicken economic escalation in the region, one of the significant inputs needed is increasing access to energy, as there is a formidable direct relationship between economic growth and energy demand. Increasing social mobility related to faster economic growth further emphasizes the demand for energy. This puts pressure on all SAARC member states (SMSs) to make sure that an uninterrupted and dependable supply of energy in its different types.

            Even though the region as a whole is enriched with assorted energy resources, with enough potential in renewable energy field, a lion’s share of these resources are yet to be exploited, for a number of causes. Inadequate availability of indigenous energy supplies, combined with the large population base, makes the region significantly dependent on energy imports. The variation of energy resource endowment between South Asia and its neighboring regions presents prospects for interregional energy trade to obtain the optimum advantages from available resources. However, the geopolitics involved and competition from alternative energy markets makes interregional energy trade a challenging proposition.

            Regional cooperation can be a supplement for national programs and projects to boost growth and energy security. Throughout the region, poverty is still prevalent, but fast progress in some countries shows that policy atmospheres can be pro-growthand that historic difficulties can be conquered. Inside the extensive set of economic policies, energy policies must play an indispensable part in maintaining the economic progress in the South Asian region. SAARC member states (SMSs) requires ensuring energy security, plummeting the expenses of energy supplies, and protecting themselves from potential threat of oil price hikes. This involves an eclectic variety of proceedings, including providing public goods and for cross-border infrastructure.

            Regional cooperation presents a key opportunity for individual SMSs to tackle obstacles to sustainable comprehensive economic progress, particularly in the energy sector. Examples of successful collaboration in chasing reliable energy policies can be observed in different regions of the world, predominantly in sharing electricity generation using cross-border transmission interconnections. Estimates suggest that in Europe, electricity system interconnection has resulted in a 7%–10% reduction in generation capacity costs. Similar cooperation within the United States has been estimated to bring benefits in the order of $20 billion. A study of the Greater Mekong Sub-region in Southeast Asia suggests that regional cooperation in energy could reduce energy costs by nearly 20%, for a saving of $200 billion during 2005–2025 [1].

2. Energy Scenario in South Asia:

            In South Asia, the reserve and potential for energy resources comes from a wide variety of options. The contemporary practice of energy usage has been as much a result of the natural bequest as the historical progression. Initially reliant on conventional bio-mass fuels, the forward-looking sectors of the countries moved on to the use of coal. So, coal mines were one of the first symbols of modernity in colonial times. Coal was discovered first in 1774, but mining began in mid-nineteenth century, primarily from eastern Indian coal seam but also later from Assam and central India. These coals were used to fuel the new railways and industries such as the jute mills around Calcutta metropolis. As modern modes of transportation started to flourish, Oil and natural gas began to be counted as the new inclusions in the energy basket. The Assam Oil Company went operational in the first quarter of twentieth century after Petroleum was struck in Digboy, Assam in 1889 [2]. As part of multipurpose river valley projects in late 1940s, large-scale hydroelectric projects were initiated. The vital factors behind creating greater demands for modern energy sources are the Population growth and urbanization. For instance, the primary energyconsumption in India increased as much as four times between 1971 and 2000 with increasing urbanization and this demand is expected to grow further while other South Asian countries have pursued comparable course, with little variations.

            Oil reserve has always been a constraint for South Asian region which is rather inadequate to meet the oil demands of the region. So, it is clear that, the region will remain dependent on oil imports. The natural gas reserves in Bangladesh, India and Pakistan are sizeable but they are not seen by experts as a dependable source for long term planning. India and Pakistan, along with Bangladesh, have large coal reserves. India is the world’s third largest coal producer, and will continue to use coal as the primary commercial fuel for electricity generation. Pakistan’s immense coal reserves are yet untapped and estimated at 175 billion tonnes in the Thar area of Sindh province, a source that Pakistan will probably start to utilize fully in the years to come, although oil and gas are plentiful in regions to the north and west of Pakistan [2]. Pakistan also has plans to develop its lignite resources and to set up mine-head power plants. Bangladesh has limited coal reserves and plans to develop them slowly; although its one large open-cut project in Phulbari has been shelved just before beginning production due to mass demonstrations about the displacement of local communities. Sri Lanka intends to begin coal imports for power generation. Hydro-electric potentials are pretty high in this region with possible locations being primarily in India, Pakistan, Nepal and Bhutan. Two mountainous countries, Nepal and Bhutan, have the potential to produce power from hydro-electric plants which are far in excess of their current or projected demands. But, high investment required is the only barrier for them to produce surplus power. Traditional fuels such as biomass and animal waste continue to contribute handsomely in the region’s energy mix, but at the same time, nuclear sources provide increasingly sizeable portions of power (in India and Pakistan), as do solar and wind power projects in India. Renewable energy sources have become a point of interest in this region and researches are going on to extract a noteworthy amount of power from renewable energy sources. Solar and Wind power is the segment that is getting pretty high interest from the researchers, private investors and governments.

3. Country-Wise Energy Scenario    

            In different countries in the SAARC region, the energy resources differ. Similarly, the process of power extraction also differ as some countries are dependent on natural gas as their main source of energy while some others are dependent on coal. These country-wise energy scenarios are discussed below in brief to get an overview on the present situation of that country and to get an insight on why regional cooperation in energy sector is required.

Bangladesh: Improved Access to and Reduced Costs of Migrant Remittance Flow

Jamaluddin Ahmed PhD FCA

Executive Summary

This report is prepared in compliance with the Terms of Reference prescribed in the contract to investigate the data on remittance inflow to Bangladesh since the beginning to   Remittance Partnership Program (RPP) in the last quarter to 2006. The report examines the remittance flow from 2003-04 to 2007-08 and the share of State Owned Banks (SoB), Private Commercial Banks(PCB), Foreign Commercial Banks (FCB) and specialized financial institutions during the period. National growth of migrant remittance during this period 2003-04 to 2007-08 was evaluated in total. A Bank by bank growth pattern of remittance to the country, identification of remittance prone areas by banks was marked and the ranking of a banks performance was carried out. The banks were ranked according to competition in performance as well as district wise. The ICT infrastructure of the sample banks were evaluated, with regards to comparative information on paper Vs electronic remittance and partial automation of SoB remittance channeling. A score card based on an international study has been tried, in order to formulate the Bangladesh remittance transfer process. A discussion on a businesses model for remittance service providers has also been addressed in this report. In the absence of a detailed data structure and cost components of major RSPs, an international study has been used as a benchmark for the purpose of assumptions and estimations. Remittance and its impact on real exchange have been discussed and comparative foreign rate spread from the international study has been evaluated, which can be applied to the context of remittance business in Bangladesh. The regulatory environment for RSP businesses in Bangladesh have been evaluated including the compliance requirements of banks. Drawing arrangements with overseas RSPs and contents of the MoU between local banks and overseas RSPs have also been addressed. Compliances of Foreign Exchange Regulations of Bangladesh and problems arising there of have been evaluated. Government policy measures with regards to fees for remittance of foreign currency from different corridors have been looked out in this report as well as the general principles of international remittance services by WB committee on payment and settlement system. Fiscal incentives provided to promote remittance in official channel have been acknowledged. In addition to this, the report also takes a look at other issues such as: operating procedures for management of foreign remittance within bank, interbank, and via third banks; and consumer education for safe, sound, secure, low cost and faster remittance and investment opportunities available within the country.    

Summary and findings

Growth of remittance:Bangladeshis recording an increase of remittance in the formal channel. Remittance corridors are changing from Middle-East to North America and Europe. During the years 2006 and 2007, Bangladesh registered a 25% increase in remittance.

National share of channeling remittance: The State owned Banks (SoB) share is declining from 2003-04: 59% to 55% in 2004-05, 2005-06: 46%, 2006-07: 39%, and 2007-08: 34%. On the contrary, the corresponding share of PCB is following an increasing trend and significantly capturing the market share, 2003-04: 37%, 2004-05: 40%, 2005-06: 49%, 2006-07: 57% and financial year 2007-08: 63%. This establishes DFID’s funded RPP effort to create awareness among the bankers to create a competition between private sector and public sector banks in channeling of migrant remittance.

Remittance Performance of Banks:The RPP project has been conducting a series of discussions with bankers since 2006 and still it is continuing which created a positive environment and consciousness among the bankers. For example RPP could sharpen the business characteristics of remittance service perception of most bankers, which previously was different. Institutional weakness in providing competitive banking services with ICT support and automation of banking operation has come to surface since the operation of RPP. The result shows that banks are becoming competitive in attracting scarce foreign currency. Since SoBs have a tendency to be institutionally weak, these with the largest branch network are losing their market share, by contrast, comparatively, those with institutional strength and a ICT backbone and automated environments are aggressively increasing their market share from 37% in 2003-04 to 63% in 2007-08 financial year and SoBS losing market share from 59% in 2003-04 to 34% in 2007-08. This structural change is making remittance faster and safer.

Top remittance receiving locations:Based on information available top remittance receiving locations were identified for Sonali, Islami, Janata, Agrani, Uttara and South East Bank have been identified. Figures reveal that 16 districts out of 64 districts attract major position of inward remittance to Bangladesh. Ranking of districts by banks processing inward remittance was done. The entire six bank’s remittance performance in Dhaka, Sylhet and Comilla ranks number one to three respectively.Chittagong, being fourth largest remittance receiving location excepting Uttara Bank all other five banks are performing, Noakhali occupies 5^th national position and excepting Uttara Bank all other five banks are operating competitively in this district. Tangail occupies 6^th national position with operation of Sonali, Islami, Agrani and Janata. The 7^th national positon goes to Feni with operation of 5banks with exception of Islami Bank. Brahmanbaria attractive 8^th largest remittance, with operation of five banks with the exception of South East Bank. Gazipur is a nearby district of Dhaka, and occupies 9^th national position in terms of remittance delivery. With the exception of Uttara Bank all other 5 banks are very much in competition. Chandpur occupies 10^th position with operations of Sonali, Islami, Agrani and Janata Bank. Moulivibazar is a migrant prone district and occupies 11^th national position and all 6 banks mentioned here are operating in this district. The National position of Mymensingh is 12^th with major contribution of Sonali, Islami, Janata and Agrani Bank. Munshigonj captured 13^th national position with operation of Sonali, Agrani, and Islami Bank. Sonali, Islami and Janata remittances channeling made Manikgonj 14^th national position. Islami, Agrani and Janata operation made Narsingdi at 14^th national position. Finally operations of Islami, Agrani, Janata and Uttara have made Narayangonj 15^th in the National position of Remittance operation.

Since RPP emphasized on the identification of remittance receiving locations, all banks including the central bank has been very much interested to take motivational approach through identification of such branches and location and declaring incentives programs, for example, Janata Bank introduced incentives to the reward those managers of their bank. More other banks are also following similar path.    

Janata Bank 5 Years Remittance Performance:Five year data of Janata Bank have been captured by source countries and by Exchange Houses to see the trend of increase from one year to the other including yearly, country, and exchange house performance and their share in the total annual remittance flow of the bank. Data up to August 2008 indicate Janata Exchnage Company Italy shared 6.47%; UAE: 40.97% comprising of Janata bank branches-27.04% and other exchange houses-13.93%. Performance and share of Janata Exchange Company located in Italy in terms of total remittance on annual basis accounts for 2003: 3%; 2004:10%; 2005:17%; 2006:26%; 2007:29%; and up to August 2008:15%.  Share of remittance from UAE source as a percentage of total was in 2003: 12%; 2004:13%; 2005: 14%; 2006:15%; 2007:25%; and up to August 2008 record is  21%. Share of other countries in total Janata remittance in 2008 accounts 52.56%. Among the other countries remittance from Kuwait is 18.28% -the largest one; Saudi Arabia:7.74%; Oman emerges as the third largest with 2.49%;  UK: 2.08%  (4^th) and Bahrain: 2.05%. By contrast, the KSA share to total remittance towards Bangladesh (2008) is 29.02%, USA: 15.56%; UK: 14.83%; UAE: 13.46%; and Kuwait: 11.39%. Remittance record from other countries showing increasing trend, for example, 2004: 16%; 2005:7%; 2006:6%; 2007:10% and 2008: (15%). Kuwait corridor recorded increase in 2004: 11%; 2005: 18%; 2006:38%;  2007: 38% and 2008 (August): (31%). UAE corridor has two sources- one is own branch and the other is exchange house. The branches recorded increase in 2004: 5.62%; 2005: 8.51%; 2006:0.51%; 2007:62.15% and 2008: (15.15%). The other exchange houses recorded increase in Kuwait corridor 2004: 15%; 2005: 6%; 2006: 27%; 2007: 62%; and 2008 (August): (15%).                  

ICT Infrastructure: Development of a national payments system establishing a Bangladesh Automation Clearing House (BACH), Bangladesh Electronic Fund Transfer Network (BFTN) and the campaign and training program for bankers under the RPP and central bank platfrom has created a positive impact on the automation of banks. The third generations PCBs are ahead in this race in particular. The first and second generation PCBs and SoB are also putting the ICT on their priority list. Sonali , Agrani, Janata and Rupali has embarked automation program during last two years . While Islamic, Duth-Bangla, South East and many others proceeded far ahead in automation. Remittance channeling marked an important aspect in such automation. State owned Banks: Sonali, Janata, Agrani and Rupali Banks are owned by the Government and have been corporatized recently, need capacity building in the infrastructure. In the process of such a transformation these banks are working in different areas. In the case of remittance delivery all these banks have a common weakness, their IT infrastructure but this is on the priority. All four commercial banks have taken up action programs for automation of branches; those are activities involved in remittance processing. For example, Sonali Bank’s Remittance transactions from December 18, 2007 to 2^nd January 2008 indicates that their remittance delivery done electronically through Instant Financial Report Management System: 18%, Remittance Management System: 67%, Third Bank: 0.77%, courier and postal services: 14.23%.

Paper Vs Electronic Remittance: Islami Bank Bangladesh Limited introduced electronic remittance service. In the month of January 2006 reduction was 20.45%, February: 14.63%, March: 18.99%, April 25.26% increase May: 0.85% increase, June: 16.70% reduction July: 20.15% reduction, August: 44.42%, September: 17.78% and October: 10.77%. In the year 2008 IBBL introduced EFI message for delivery of remittance which shows significant increase in the use of EFT for remittance delivery. The percentage increase ranges from 89.37% to 150.23% during the January – October period currently, Islamic Bank remits 65% of their transaction for remittance on the same day. The remaining 35% transaction, 18 percent are cloned by 2-3 days, 8% by  10 days, 5% by 20 days and 4% by 30 days.

Scorecard on remittance transfer: Competition between RSP/MTOs should be a contributor to the decline in transaction costs and participation of more banks and similar financial institutions as remittance players. There are nine criteria for the analysis of market performance in relation to development and consumer rights. These include, transfer fee exchange rate, commissions, mechanisms used to send the money, competitive position of the corridors, geographic coverage across corridors, levels of engagement with local consumer community, relationship with financial intermediaries, transparency and disclosing information about pricing, and compliance to regulatory rules. In Bangladesh, a detailed study is needed on the scorecard on remittance transfer.

Business model in RSP/MTO industry: Agency model employed by Western Union, Moneygram and VIGO used agent who nearly always operate other businesses in same location in the receiving and sending countries. These agents pay for all rates, personnel and other fixed and operating expenses for the location in exchange of commission. Commissions are generally a fixed percentage of MTO fee for the transfer excluding foreign exchange spreads. In the branch model used by Dolex and many of smaller MTO and small to mid size niche player, the MTO owns the sending, and often also the receiving branch and pays for all fixed and operating cost associated with each branch.

Remittance Pricing: Corridor volume, which attracts global and regional MTOs as well as small niche players that compete on price, appears to be the most significant factor driving remittance fee prices.Exclusive agency partnerships between large Global MTOs and postal systems also appear to have a major influence on remittance prices by serving as a significant entry barrier to small competitors and allowing high prices to be maintained even in the face of serious competitive challenges from other large, lower priced rivals.Restrictive regulations in the US and sending countries in the EU may also have an impact on pricing by discouraging new competition and making it easier for the large global MTOs to maintain higher prices. Other factors that appear to influence remittance pricing within a corridor include: the active participation of banks, credit unions and other non-bank financial institutions in the remittance market, migrant access to low cost alternatives offered by these organizations,   technology and product/service innovations, and the strength of the informal transfer network.  Government policy initiatives may have helped reduce remittance prices in selected corridors such as the US to Mexico, but the significance of their impact is difficult to ascertain.

Cost structure of MTO/RSPs: Agent commission is the dominant variable cost in the agency network model, which varies considerably by corridor and by company. Costs include identifiable cost comprising of variable cost mainly agency commission, fixed and depreciation cost that include marketing, general administration, depreciation, amortization and agency start up cost and some unidentifiable variable and fixed costs. To arrive at total cost of remittance, it is necessary to know the exchange rate applied to the conversion either at the remitter’s end or while making payment to the beneficiary in home country.

Competitive foreign exchange spread: Foreign rate spreads on remittance for both Western Union and Moneygram appear to be too high, even in high volume corridors, and are not transparent to the sender. Increased consumer awareness of the high level foreign exchange spreads, could encourage more exchange rate competition and help lower overall remittance costs. Detailed investigations need to be carried out in the context of exchange rate spread for Bangladesh remittance.

Variation of remittance fees:Remittance fee pricing varies significantly by competitor, corridor and channel. Three set of factors contributing to remittance prices, corridor specific, sending country specific and receiving country specific. Driving factors for remittance fee pricing include competitive factors, technology and product factors, and government policy factors.

Regulation of MTO/RSP business:Bangladesh Bank plays important role in remittance business. Commercial banks are to comply with regulatory requirements. The trial and error techniques applied  in Bangladesh for promoting RSP business in attracting safe secured, at lower cost still needs to be re-casted compared to the international practices.

Problems faced by the beneficiary of migrant through the channel:Cost of sending the remittance is based on per transaction so the small remitter have to pay the same cost of a large remittance. Sometimes the distance of migrant work and dueling place with the exchange house/bank are so far that they can not come to remit money frequently.  Major portion of the migrants have little educational backgrounds and so they face problems with language and the exchange house/banks formalities. As most of the migrant remitters go abroad from the rural area the beneficiaries reside in the same area, very few of them have knowledge about the banking system. As a result, they face problems completing the banking formalities. Non-availability of bank account of the beneficiary and the remitter is another deterrent. Some times the beneficiaries are harassed and caused delay by the corrupt bank officers.

Government Policy measures to influence the flow of remittance: Remittance transactions are inherently private, and as such, regulation does not address in any way the allocation of remittance funds, which receivers clearly have the freedom to spend or invest as they choose. Within this scope, regulatory concerns are normally aimed at facilitating the provision of formal remittance service at the lowest cost possible to as many users as possible, while maintaining a high level of security in the system. By nature, remittance involves operations in various jurisdictions, under different regulatory framework. After reconciliation of the overall objectives of high security with low costs remains a major challenge. The first objective of regulation is enforcing security in remittance services from misuse for illegal transactions including financial terrorism. The second broad objective of regulation refers to facilitating the reduction of the prices of remittance. For immigrants sending money home, remittances services have traditionally been expensive, with fees of up to 20 percent of the principal sent depending on the size and type of transfer to the destination. Authorities have shied away from imposing direct price control on remittance services, favoring mechanisms aimed at increasing transparency, enhancing competition in the system, and, in some cases, reducing barriers for users to access a wider range of services providers.

National Payment System:For safe secured, at lower cost and speedy remittance to the rural people, like other countries, Bangladesh needs a national strategy for Payment System. The National Payment System needs cooperation among banks who are in constantly engaged in business competition. Out of 6500 bank branches over 60% owned by State Owned Banks who lacks automation and inter branch connectivity. Operation of National Payment System supported Automatic Clearing House (ACH), Electronic Fund Transfer Network (EFTN) and Automation of Cheque Processing can support the speedy, safe, secured, and cheaper remittance processing to the poor of the rural Bangladesh.   Central Bank as supervisory authority of the national payment system, clearing operation, and cheque modernization process for the commercial banks must emphasis priority on the ICT application in its own operations. Central Bank should review its current operational and priority areas and outsource non-core functions to the other agencies. 

Recommendations

1.       In the absence of correct price and cost data on remittance, corridor specific study is needed to evaluate the data and then recommend for further action plan.

2.       Awareness program on cost, prices, foreign exchange regulation, AML, and remittance linked new banking products should be promoted among bankers

3.       Awareness program on fiscal incentives for migrant remittance should be carried forward before they proceed for their work and homecoming time in the airplane journey and TV drama specially chalked out for migrants 

4.       Readiness assessment of individual banks to implement the cheque modernization, BACH, and BFTN under the RPP. Separate study need to be conducted immediately to determine the eligibility participating banks with BACH and BFTN.

5.       Central Bank sitting at the steering of the national payment system should embrace stakeholders of the payment system, BACH, BFTN, and cheque modernization process to make it a success.

6.       Central Bank must take its pioneering role by accepting the genuine problems with foreign exchange regulation those hinder the smooth delivery of migrant remittance and create friendly environment to make lower cost, faster, safe and secured delivery.

7.        Central Bank should monitor the price and cost remittance on timely interval to have control over the activities of banks and remittance service providers.

Table of Contents

Section One: Background and Introduction. 8

Section Two: Basics of Remittance Operation. 9

Section Three: Five Performance of Bangladesh Remittance13

Section Four: Remittance Fees and Pricing23

Section Five: Cost of Remittance35

Section Six: Remittance Regulatory Environment in Bangladesh. 43

Section One: Background and Introduction

The DFID funded Remittance Partnership Program (RPP) started working at Bangladesh Bank since October 2006 to promote cheaper cost, legal means, and faster migrant remittance processing to rural Bangladesh. RPP components include development of National Payments Strategy followed by National Payments Laws and regulations, establishing modernized cheque processing system, automatic clearing and electronic fund transfer network and introduction of innovative banking products by challenge fund besides the research and promotional works to promote legal remittance to Bangladesh. Objective of this report is to make an evaluation of the current state of remittance processing in terms of volume increase, price and cost in Bangladesh compared to the findings of other studies and readiness of banks in terms of ICT application for speedy remittance. Literature on price and cost on remittance is not easily available in the absence of empirical studies in this particular area and case of Bangladesh study in this regard have not  been conducted  as it is very much difficult to get the information on cost and price.  This report is structured into seven sections. Section two starts with basics of remittance operation with a brief on remittance transactions, structure, players, instruments, and value added by the remittance firms. Section three provides 5 years data on remittance towards Bangladesh channeled from different corridors and by banks and Remittance Service Provider firms over the period 2003-2008. Performance of banks, identification of top remittance prone areas by banks and five year data of Janata Bank from RSP and corridors have been disclosed. ICT infrastructure of local banks have been evaluated. Section four initiates discussion on business models for RSPs including cost structure. Section five details on remittance fee pricing. Section six details on the cost of remittance transactions. Section seven provides a picture on the regulatory regime on RSP in Bangladesh including fiscal incentives and Section eight provides a summary and recommendation.

Section Two: Basics of Remittance Operation

The basics of a remittance operation:  A remittance transfer is a cross-border retail payment with special access requirements on thesending and receiving side.

A remittance operation is initiated by a remittance service provider capturing funds from a remitter using any of a variety of instruments, either directly or through a capturing agent. The most normal instrument is cash; other instruments in use are direct debits, cheques, money orders, and credit cards. There is a large number of possible instruments; one example is stored value accounts associated with cell phones. The capture channel can be physical, such as a grocery store operating as agent for a remittance service provider, or virtual, such as the internet or a call center. An ideal capture channel is close to the remitters, trusted, can handle cash, and can easily be linked to the remittance service provider. The limitations on available capture channels tend to be associated with credit risk, compliance risk, and operating costs. For example, the internet is cost effective but it cannot handle cash, and it poses compliance challenges. An external agent may have good compliance procedures but can be costly and pose a credit risk. The capturing agent transfers the funds to the remittance service provider (or to a designated third party) using mechanisms such as a domestic payment system, transport of cash or written instruments, or the creation of liabilities to the remittance service provider. The communication between the agent and the remittance service provider can take place through a dedicated computer system, email, fax, or telephone. Credit and compliance risks are addressed at some point in this process, and the data supplied by the remitter is validated.

When the remittance service provider’s policies on transaction validation are met, the disbursement process starts. The remittance service provider orders a disbursement agent to make funds available to the recipient. This message goes through channels such as SWIFT, a proprietary database, fax, email, or phone, and it is originated either by an entity in the country where the remittance is initiated or by a partner or subsidiary of the remittance service provider in the country of destination. In many cases, the time from funds capture to funds availability is close to instant. In other cases, the transfer can take up to several days. Often, the determinants of speed are the instruments used for capture and the means of messaging. Compliance requires the remittance service provider to check recipients against a list of barred recipients, which is a process that must be computerized in order to allow for instant transfers. The disbursement agent may use a variety of instruments. A bank or non-bank account is credited, cash is picked up, or a cheque or money order is issued.What legitimizes the remitter to pick up the remittance varies. It may be an official ID showing that the name and other identifying information is the same as in the remittance record, it can be a code communicated from the remitter to the recipient, and it can be a code sent to the recipient’s cell phone by the service provider.

How the remittance service provider settles across borders varies with the nature of the operation. Smaller service providers typically use a settlement agent, such as an international bank or a bank with correspondent relationships, to wire funds to the service provider’s account in the receiving country, or, alternatively, to its disbursement agent in that country. Remittance firms can also rely on disbursing partners that have accounts in the sending country. In these cases no international settlement is required from the perspective of the remittance service provider. Larger providers are able to use corporate treasuries to settle on their own books. International banks and banks with correspondent relationships settle through their usual means.

A stylized remittance transaction—structure, players, instruments

Atypical remittance transaction takes place in three steps: (1) initiation of remittances by a migrant sender using a sending agent, (2) exchange of information and settlement of funds, and (3) delivery of remittances to the beneficiary. In step 1, the migrant sender pays the principal amount of the remittance to the sending agent using cash, cheque, money order, credit card, debit card, or a debit instruction sent by e-mail, phone, or Internet. In step 2, the sending agency—which may be an MTO, bank, or other financial institution, money changer, or merchant (gas station, grocery store)—then instructs its agent in the recipient’s country to deliver the remittance. In step 3, the paying agent makes the payment to the beneficiary. In most cases, there is no real time fund transfer; instead, the balance owed by the sending agent to the paying agent is settled periodically according to a mutually agreed schedule. Settlement usually occurs through commercial banks acting through the national clearing and settlement system. A portion of informal remittances is settled through goods trade.

 Remittance securitization structure

Remittance senders

Beneficiary’s

account

Remittance

payments

(foreign currency)

Issuing bank credits

beneficiary’s account in domestic currency

Correspondent banks

Message

Issuing bank

Remittance flows (foreign currency)

Trustee collateral account

Excess cash

(foreign currency)

Issuing bank’s

account at correspondent bank

Debt service payment

International investors

Offshore   Domestic

The value added by remittance firms: A remittance firm transfers funds from one person in one country to another person in another country. Often, the international remittance is the monetary fruit of migrant labor that is being transferred back to the migrant worker’s family. This makes the remittance service a crucial component of the logistics of migration. There is a multitude of companies providing remittance services, of all sizes and using a large variety of technologies. They mainly add value in the following four areas, where their services are unique.

First,there is no global retail payment system. Global payments are largely effectuated by banks with correspondent relationships, or, in a variation of this, through a system of payment card processors that effectuate the retail transactions using correspondent banking relationships to settle balances. Bank wire transfers are often expensive, and the use of payment cards for remittances is in its infancy. Therefore, remittance firms perform a unique function in the cross border retail payment infrastructure. Second, the senders of remittances often need an interface to access the banking system. Many do not have bank accounts, and banks’ opening hours are often not conducive to migrant work schedules. Cultural and language barriers can be significant, as can reluctance to engage with large financial institutions. Concerns about immigrant status can also prevent some migrants from using banks for international transfers. Similarly, migrants often have access to a more limited number of payment instruments, frequently only cash and money orders. Whereas a person with access to bank accounts, internet, payment cards, and cheques can initiate payments in a large number of ways, a person with access only to cash and not to a bank account has a limited number of options. Remittance firms provide access to payment services for individuals that would otherwise have been excluded from such services. Third, remittance firms offer delivery of funds through instruments and channels that the recipient can access, such as cash from a disbursing agent that is located not too far from where the recipient lives or that offers courier delivery. Since domestic payment systems in many receiving countries are underdeveloped, agent networks provide a substitute and prevent recipients from having to travel to places covered by the banking system. As the case with the senders, remittance firms increase access to payment services for the recipients. Fourth, remittance firms offer opportunities for cost savings. International wire transfers through banks are costly and slow. Remittance firms ‘bundle’ a number of transfers, send the bundled funds through the banking system, and ‘unbundle’ the funds at the other end. In this way, the settlement charges are spread over many remittance transactions. In this way, remittance firms reduce the cost of transferring funds.

Section Three: Five Performance of BangladeshRemittance

National Share of Remittance: Five years data ranging from 2003-04 to 2007 – 08 have been collected from different banks and analyzed those in different perspectives total remittance flow to Bangladesh during 2003 – 2004  to 2007 – 2008 have been captured. Information revealed that in 2003 – 04 total remittance to the country was USD 3,371.79m and  share of SOB are 59%, private commercial Banks: 37%, specialized financial institutions: 1%, and Foreign commercial Bank: 3% of the total remittance channeled to Bangladesh. During 2004 -05 financial year total remittance channeled to the country was USD 3848.29 m in comprising of SoB: 55%, Private sector commercial Banks: 40%, Specialized institutions: 4% and FCBs: 1% of total remittance flow. In the year 2005 – 06 total remittance flow was USD 4801.87m shared by SoB : 46%, PCBs: 49%, FCBs: 4% and specialized institutions: 1%. In this year the private commercial banks crossed the national share of SoB remittance for the first time. In year 2006 – 07 total remittance flow to the country was USD 5978.47m with share of SoB: 39% (further reduction) PCBs: 57%, FCBs: 3%, and specialized institutions : 1%. In the financial year 2007 – 2008, total flow of remittance to the country was USD 6430.93m with reduced share of SoBs : 34%, PCBs: 63%, FCBs: 2% and specialized institutions: 1%.

Performance Rating on remittance

Annual performance of all banks in remittance channeling has been analyzed. Remittance performance of Banks varied from year to year and month to months.In 2003 -04 total remittanceflow of USD 3371.97m share of Sonali Bank was 31% topping the legue table, Agrani Bank: 17%, Janata Bank : 11%, Pubali Bank: 11%, Islami Bank : 9%, Uttara Bank: 8%, National Bank : 3%, Standard Chartered Bank: 1%, Citi Bank N A: 1%, AB Bank : 1%, and other banks: 7%.

In 2004 – 05 remittance flowof USD 3848.287m was channeled by Sonali Bank: 30% with percentage point reduction from last year, Agrani Bank : 16%    with reduction of 1%  from the previous year, Islami Bank 12% marking 3% increase from earlier year, Uttara Bank 10% marking 2% increase from previous year, Janata Bank : 9% accounting reduction of 2% from earlier years. Pubali Bank captured 6^th position with 6% share, National Bank: 4% marking 1% increase from last year, Citi Bank NA: 2%, The City Bank Ltd. : 1%, AB Bank: 1% and other banks: 10% and the Standard Chartered Bank lost their position among top 10 banks channeling  remittance.

Financial year 2005 – 06 recorded USD 4801.87m remittances to Bangladesh where Sonali Bank shared 25% losing 5% compared to last financial year. Islami Bank captured 13% by increasing 1% from last year. Agrani Bank went down to 3^rd position with 11% share marking reduction of 5% share in the same year, Uttara Bank: 9% recording reduction of 1% from previous year, Janata Bank: 8% recording 1% reduction this year. National Bank accounted their share at 5% by increasing 1% from the last year. Pubali Bank recorded 4% share with reduction of 1% from last year, Citi NA: 3% with 1% increase from last year. Both Dhaka Bank and South East Bank entered in the list of top 10 Banks with 2% market share each.

Financial year 2006 -07recorded USD 5978.47m retaining  Sonali Bank position as number one with 20% market share and reduction of 5% share compared to previous financial year. Islami Bank recorded 16% share with 3% increase from the previous year. Janata gained 4^th position with 7% share toppling, Uttara Bank 7% little margin. The National Bank share increased 6% from 5% in last year. Pubali Bank recorded 5% share by increasing 1% from last year. BRAC Bank joined the top 10 league by gaining 3% share and the same applies to PRIME Bank Limited recording 3% share. South East Bank also recorded 3% share by increasing 1% from last year. The other banks also recorded their share 21% with 3% increase from the previous year.

Financial year 2007 – 08recorded USD 6430.93m. Islami Bank recorded 20.6% share by capturing top position in remittance channeling by increasing 4.6% compared to last financial year. Sonali Bank dropped to second position with 17% share by reduction of 3% from last year. Agrani Bank recorded 9% share with reduction of 1% from last year. National Bank recorded 7% by increasing 1% share. Janata Bank gone down to 5^th position with 6% share by reduction of 1% from previous year. Uttara Bank share was 6% with reduction of 1% from last year. Pubali Bank retained 5% and BRAC Bank increased its share to 5% from 3% of last year. Prime Bank retained its 3% market share and Eastern Bank graduated to top 10 banks in remittance channeling with market share of 2% and South East Bank was dropped from top 10 list this year.  

Top remittance receiving locations: Based on information available top remittance receiving locations were identified for Sonali, Islami, Janata, Agrani, Uttara and South East Bank have been identified. Figures reveal that 16 districts out of 64 districts attract major position of inward remittance to Bangladesh. Ranking of districts by banks processing inward remittance was done. The entire six bank’s remittance performance in Dhaka, Sylhet and Comilla ranks number one to three respectively.

Chittagong, being fourth largest remittance receiving location excepting Uttara bank all other five banks are performing Noakhali occupies 5^th national position and excepting Uttara Bank all other five banks are operating competitively in this district. Tangail occupies 6^th national position with operation of Sonali, Islami, Agrani and Janata. The 7^th national positon goes to Feni with operation of 5banks with exception of Islami Bank. Brahmanbaria occupies 8^th largest remittance, with operation of five banks with the exception of South East Bank. Gazipur is a nearby district of Dhaka, and occupies 9^th national position in terms of remittance delivery. With the exception of Uttara Bank all other 5 banks are very much in competition. Chandpur occupies 10^th position with operations of Sonali, Islami, Agrani and Janata Bank. Moulivibazar is a migrant prone district and occupies 11^th national position and all 6 banks mentioned here are operating in this district. The national position of Mymensingh is 12^th with major contribution of Sonali, Islami, Janata and Agrani Bank. Munshigonj captured 13^th national position with operation of Sonali, Agrani, and Islami Bank. Sonali, Islami and Janata remittances channeling made Manikgonj 14^th national position. Islami, Agrani and Janata Operation made Narsingdi at 14^th national position. Finally operations of Islami, Agrani, Janata and Uttara have made Narayangonj 15^th in the National position of Remittance operation.

Janata Bank 5 Years Remittance Performance: Five years data of Janata Bank have been captured by source countries and by Exchange Houses to see the trend of increase from one year to the other including yearly, country, and exchange house performance and their share in the total annual remittance flow of the bank. Data up to August 2008 indicate Janata Exchnage Company Italy shared 6.47%; UAE: 40.97% comprising of Janata bank branches-27.04% and other exchange houses-13.93%. Performance and share of Janata Exchange Company located in Italy in terms of total remittance on annual basis accounts for 2003: 3%; 2004:10%; 2005:17%; 2006:26%; 2007:29%; and up to August 2008:15%.  Share of remittance from UAE source as a percentage of total was in 2003: 12%; 2004:13%; 2005: 14%; 2006:15%; 2007:25%; and up to August 2008 record is  21%. Share of other countries in total Janata remittance in 2008 accounts 52.56%. Among the other countries, remittance from Kuwait is 18.28% -the largest one; Saudi Arabia: 7.74%; Oman emerges as the third largest with 2.49%; UK: 2.08% (4^th) and Bahrain: 2.05%. By contrast, the KSA share to total remittance towards Bangladesh (2008) is 29.02%, USA: 15.56%; UK: 14.83%; UAE: 13.46%; and Kuwait: 11.39%. Remittance record from other countries showing increasing trend, for example, 2004: 16%; 2005:7%; 2006:6%; 2007:10% and 2008: (15%). Kuwait corridor recorded increase in 2004: 11%; 2005: 18%; 2006:38%;  2007: 38% and 2008 (August): (31%). UAE corridor has two sources- one is own branch and the other is exchange house. The branches recorded increase in 2004: 5.62%; 2005: 8.51%; 2006:0.51%; 2007:62.15% and 2008: (15.15%). The other exchange houses recorded increase in Kuwait corridor 2004: 15%; 2005: 6%; 2006: 27%; 2007: 62%; and 2008 (August): (15%).                  

Janata Bank Limited: Remittance Performance 5 Years data

ICT Infrastructure:

For safe, secure, and faster remittance delivery, the ICT infrastructure is an important element in the banking industry in addition to other institutional issues. At the moment there are approximately 6,500 bank branches, comprising of 4691 government owned banks, 1760 private commercial banks and 49 branches of foreign banks, with the exception 4 state owned and few specialized banks, private commercial and foreign banks do not have their branches at head quarters of 64 districts. Still automation has taken significant position in the banking industry for survival in the competition and to meet the needs of national payment system of the country.

Although automation is a nightmare in Bangladesh, there still there attempts to automate. Other than a few foreign banks automation is misnomer and needs definition. In true terminology, the automation of banks still has far to go, both in private and public sector banks. Bangladesh Bank being the regulator of the banks is still far behind the commercial banks. However, current status of IT application in the commercial banks can be categorized into: fully automated, partially automated and computerized. Fully automated banks include: AB Bank, BRAC Bank. Citi Bank NA, Dutch Bangla, Eastern Bank, First Security Bank, HSBC Bank, One Bank, Prime Bank, Shahjalal Bank, Social Investment Bank, South East Bank, Standard Chartered Bank, Trust Bank, Woori Bank, Commercial Bank of Ceylon, Dhaka Bank and Bank Asia. Only Prime Bank and BRAC Bank are in the top ten banks of remittance service provider.

Partially Automated Banks include Al-Arafa Islami Bank, Exim Bank, Habib Bank, IFIC Bank, Islami Bank, National Bank of Pakistan, NCC Bank, The City Bank, Oriental Bank, Premier Bank and United Commercial Bank. The computerized banks include BCI Bank, Jamuna Bank, Mercantile Bank, Mutual Trust Bank, National Bank, Pubali Bank, Standard Bank, Uttara and State Bank of India.    

State owned Banks: Sonali, Janata, Agrani and Rupali Banks are owned by the Government and have been corporatized recently, need capacity building in the infrastructure. In the process of such a transformation, these banks are working in different areas. In the case of remittance delivery all these banks have a common weakness, their IT infrastructure but this is on the priority. All four commercial banks have taken up action programs for automation of branches, these are activities involved in remittance processing. For example, Sonali Bank’s remittance transactions from December 18, 2007 to 2^nd January 2008 indicates that their remittance delivery done electronically through Instant Financial Report Management System: 18%, Remittance Management System: 67%, Third Bank: 0.77%, courier and postal services: 14.23%.

Paper Vs Electronic Remittance: Islami Bank Bangladesh Limited introduced electronic remittance service. In the month of January 2006 reduction was 20.45%, February: 14.63%, March: 18.99%, April 25.26% increase May: 0.85% increase, June: 16.70% reduction July: 20.15% reduction, August: 44.42%, September: 17.78% and October: 10.77%. In the year 2008 IBBL introduced EFI message for delivery of remittance which shows significant increase in the use of EFT for remittance delivery. The percentage increase ranges from 89.37% to 150.23% during the January – October period currently, Islamic Bank remits 65% of their transaction for remittance on the same day. The remaining 35% transaction, 18 percent are cloned by 2-3 days, 8% by  10 days, 5% by 20 days and 4% by 30 days.

Section Four: Remittance Fees and Pricing

Two basic business models are used in the MTO industry: The “agency� model employed by Western Union, Moneygram and Vigo, who usesagents who nearly always operate other businesses in the same location in thereceiving and sending countries. These agents pay for all rents, personnel and other fixed and operatingexpenses for the location in exchange for a commission. Commissions are generally a fixed percentage of the MTO fee for the transfer,excluding foreign exchange spreads. However, in some countries, includingMexico, commissions to paying agents are negotiated at a fixed price pertransaction.In the “branch� model used by Dolex and many of the smaller regional MTOs and small to mid-sized niche players, the MTO owns the sending, and often also the receiving branch and pays for all fixed and operating costs associated with each branch.

Detailed cost data is not available, but some components of the cost structure for major MTOs using the agency model can be estimated:  Agent commissions, the dominant variable cost in the agency network model, vary considerably by corridor and company. Total commissions for both sending and receiving agents appear to be in the range of 40% to 60% of the remittance fee (excluding foreign exchange commission) for most MTOs, and 25% to 45% of the premium priced remittance fees charged by Western Union.Marketing costs for First Data’s Payment Services reporting segment, which is nearly all Western Union, were disclosed as 7%-8% of segment revenue, approximately $300 million. This number is included in selling, general and administrative costs, which totaled 16% of total revenue at the corporate level.Depreciation and amortization on capitalized costs, including the costs associated with building or purchasing software and systems to handle transactions and transfers, is approximately 3%-4% of total cost.Agency start-up costs have been estimated by Piper Jaffray at $1000-$1500 per newAgency. Licensing and regulatory compliance costs were not quantified in the study, but were considered a major problem by the small MTOs interviewed in the World Bank Andreassen study.

A model was developed to estimate Western Union’s cost structure and capacity to lower remittance fees:  Information disclosed in SEC filings and the agency cost component estimates were used develop a rough model of the cost structure for the international portion of Western Union’s MTO activity that is most relevant to worker remittances.A simple model (assuming that a 35% commission on fees is the only variable cost and that all other costs are fixed) was used to illustrate the impact that transfer fee price reductions would have on Western Union’s international operating margins.This model was also used to demonstrate how costs could be lowered by operating low cost regional hubs and increasing transaction volume

Estimated Cost Structure for Western Union International Transactions Related to Remittances

Remittance and Real Exchange Rate: Literaturesupported that remittance number of beneficial effects for the welfare of receiving countries, higher remittances inflows tend to be associated with lower poverty indicators and higher growth rates. Beyond these typical income dimensions of welfare, remittances seem to reduce output volatility to some countries and some socio-economic groups. For a while remittance inflows may ease external financing constraints and therefore hold the potential for higher investment by developing countries. Workers’ remittance can be viewed as capital inflow and therefore the theory of Dutch Disease phenomenon associated with a surge in inflows can also be applied in this context. Remittance additional demand for non-tradables in the domestic economy due to additional demand or so called spending effects.  Remittance is considered to have impact on the resource movement effect and output growth in non-tradables production more profitable. Output growth in the non-tradables pushes factor demands. The price shift and resources relocation in favor of non-tradables erode the competitiveness of export oriented sectors and hurt import competing sectors. The final result of this real exchange rate appreciation is normally increased import flows and lower export sales. There are number of connected  macroeconomic effects  that can result from  a real exchange rate appreciation associated with remittance flows which include; (a) adverse effects on the tradable sectors; (b) widening the current account deficit;  and (c) weaker monetary control, inflationary pressures and sectorial allocation of investments. Thus remittance can potentially affect the real exchange rate through three main channels. First: remittance may affect the external equilibrium of the economy by raising the net foreign asset position of the country. Second: remittance can also affect internal equilibrium of the economy in a situation where domestic capital and labor are efficiently utilized. Third: for remittance to affect the real exchange rate is through their impact on growth.   

Exchange Rate: A remittance transfer will usually involve a foreign exchange transaction, typically conversion from the currency of the remitting country to the currency of the receiving country.  Sometimes it may involve more than two currencies in case the remitter is located in a under developed market from where he converts his local currency into US Dollar and thereafter it is converted again into Taka at the Bangladesh end as direct quotations between the local currency of the remitter and the Taka may not be available in many centers in Africa, Central Europe etc. To arrive at the total cost of the remittance, it is necessary to know the exchange rate applied to the conversion either at the remitters’ end or while making payments to the beneficiary in home country.

Typically in some of the centers in the Middle East, where Exchange Companies are engaged in making such remittances, the local currency gets converted into home country currency for payment through the Vostro accounts of the concerned Exchange Companies. It will depend on the competitiveness of the market at that centre, whether the remitter is getting a market related competitive price. The remitting agency adds a margin over the interbank rate while quoting the price to the remitter. Such margins will vary depending on the uncertainty about the inter bank rates available to the remitting agency. Small Money Transfer Operators (MTOs) may not be able to access competitive rates for covering their remittance. However in case of remittance through larger banks information on margins, fees etc. are always made available to the customer. It is very difficult to compare the exchange rates between various remittance service providers and therefore arrive at a very definite cost for remittances.

At the home country end, both for inward and outward remittances, while special interbank linked rates are being quoted by the remitting bank for valued customers, however in a majority of small value remittances, the “Card Rate� of exchange is being used. The margin could range from 0.25% to 1% in such cases. However, if the remittance amount is small, the exchange cost in absolute terms would be marginal.

Comparative Foreign Exchange Spreads: Actual daily foreign exchange spreads (for Western Union transfers to 22 countries from the US and the UK and for Moneygram transfers from the US to the same countries) were calculated for four consecutive business days early in June 2005.Spread calculations for each corridor were based on the difference between theaverage wholesale commercial market exchange rates for that corridor published daily on Bloomberg and the actual exchange rates charged by Western Union and Moneygram. Four-day average spreads were used in the analysis to smooth out the daily fluctuations, such as the ones illustrated below:

Western Union Exchange Rate Spreads As % Of Market Rate for selected US Corridors (%)

Janata Bank: Exchange Rate given to Exchange Houses

To summarizethe Foreign exchange rate spreads on remittances for both Western Union and Moneygram appear to be too high, even in high volume corridors, and are not transparent to the sender.Increased consumer awareness of the high level of foreign exchange spreads, could encourage more exchange rate competition and help lower overall remittance costs.

Float: Findings of Ole E. Andreassen (June 2006) sponsored by World Bank on Remittance Service Providers in the United States: How remittance firms operate and how they perceive their business environment found that thelevel of float in a remittance system depends on the structure of how funds flow through the system. The average transfer time in our sample is 14.72 hours; the median is 3, and the funds move between several institutions in the course of a transfer. This means that there is little opportunity for a net float income. This is consistent with firms settling cross-border on average once per weekday. The more funds move from institutions to institutions, the less opportunity there is for creating float. In a best-case scenario, where a firm has the control of funds for the entire process and funds are available to the recipient in 12 hours, a firm with the average volume per month from our sample, $36.3mn, which could earn 10% on float, will have a float income of only around $3,500 per month.Of the firms interviewed, 16.33% report some float income, 83.67% report that they have no float income, and 73.47% report float loss. Of the firms that report float income, one is very small and settles with its overseas partner every month, one has an agreement with its partners allowing it a day of float, and the majority are subsidized by their owners in that the owner is in charge of disbursement and collects money less frequently than technically feasible, thereby carrying the cost of working capital on the disbursement side. The net effect for the latter type of operation depends on the cost of capital in the receiving country versus the return on capital in the sending country.

In most cases, pre-funding takes place by transferring money to a bank account in the receiving country, ready for disbursement. This means that any float is most likely to accrue to the financial institution providing the account, bar a possible interest rate on the bank account. The only real option for investing float in such a case, is where the bank owns, or is the same entity as, the remittance operation.

Currency conversion:Andreassen (June 2006) study indicated a sample, 8.82% of the firms disburse in USD only, 48.53% disburse in local currency only, and 42.65% disburse in both currencies, normally at the discretion of the recipient. This means core component of most remittance services includes a currency conversion. Remittance firms solve this in different ways. Only 19.70% of firms do these themselves; in 62.12% of the cases, the overseas partner (or parent/subsidiary/sister company) does the currency conversion. In 7.58% of the cases, the currency conversion is outsourced to a third party. The average spread that a remittance firm charges over the rate it obtains is 1.63%, the median is 1%. The firms that exchange currency themselves gain a 3.31% spread on average, the firms that have the partner exchange currency gain a 1.24% spread, and the ones that outsource gain 1.01%. Note that this rate is the spread that accrues to the remittance firm; it is not the spread over the wholesale rate. Since the firms typically do not do the currency conversion themselves, it is likely that the entity that does the conversion charges a spread over the wholesale rate, which means that the exchange rate spread over a wholesale rate faced by the remitter, is higher than the numbers reported above. Where the remitter is quoted a rate, the remittance service provider runs the risk of covering that rate. Many firms take on such risk, and assume that the spread they quote will cushion them from currency fluctuations, but there are many variations. One firm, for example, notifies agents of the rate its disbursing partner is able to obtain several times per day, but if it turns out the agent promises a rate that the disbursing partner is not able to cover, the agent carries the risk.

Firms that consider their comparative advantage in commercial relationships and strong finances tend to also have a higher spread, perhaps related to their bargaining power in the relationship. They tend to settle internationally more frequently, which also allows them to match their buy and sell rates better, and they use banks and credits to payment cards more frequently than the average company for disbursement. Higher spreads are significantly correlated with the perception that exchange controls are an obstacle to doing business, which might indicate that exchange controls give opportunities for charging higher spreads.

The exchange revenue is shared in 29.09% of the cases. There is no significant correlation between the size of the spread and whether it is shared or not. Firms that perceive an increasingly competitive environment are more likely to share their foreign exchange revenues than the sample average. Sharing the exchange gain with a partner can be a way of buying oneself out of trouble. Firms that share foreign exchange gain have fewer problems with corruption and reporting requirements abroad. The firms that share the foreign exchange need on average 20.13 hours to make the funds available to the recipient, the firms that do not share exchange gain need 9.59 hours on average. A possible explanation of this is that the gain is realized immediately during the transaction (which takes time) rather than being settled after the transaction has taken place.

Transfer time:Andreassen (June 2006) study findings reported that companies spend on average 14.72 (median: 3) hours getting a remittance to the recipient. Transfer times vary within the different products that a company offers. The minimum transfer time reported by companies is 7.34 hours on average (0.58 hours median), while the longest is 32.06 hours on average (5 hours median). The average difference between the longest and shortest transfer time reported is 25.17 hours (1 hour median). When we look for factors in the firms’ business environment that might be correlated with longer transfer time, we find that firms with longer transfer times perceive competition both in the U.S. and abroad as a higher obstacle than other firms. We also find that longer transfer time tends to be associated with higher average transaction sizes. There can be several explanations for this. Remitters may send transactions less frequently if it takes longer, or firms that principally send higher amounts may not face the pressure to transfer fast. There is no apparent relationship between transfer time and fee or exchange rate spread, even if we control for destination region.

Remittance fee pricing: This exercise is both complex and non-uniform and remittance fee pricing varies significantly by competitor, corridor and channel. Major fee pricing differences can also be observed between different sending countries-even for the same competitors sending remittances to the same receiving countries. Fees also vary with the size of the remittance. Fees can also vary significantly for a single competitor by type of service offered, sending city, and individual sending agent.

Fees by Competitor, Corridor and Channel for Sending $200 from US (NYC)

Variation of Remittance Fees:Remittance fee pricing varies significantly by competitor, corridor and channel. Major fee pricing differences can also be observed between different sending countries-even for the same competitors sending remittances to the same receiving countries. Fees also vary with the size of the remittance. Fees can also vary significantly for a single competitor by type of service offered, sending city, and individual sending agents.  Comparative Remittance fees varies largely  across different corridors, channels and competitors described insuggest that remittance fees are higherthan necessary in the EU, the higher pricedcorridors within the US and, with a few exceptions,for the services offered by Western Union. Small remittance transfers are particularly costly dueto the fee pricing policies of most MTOs

Major Global MTO Fees For Sending $200 To MexicoFromSelected US Cities

Sending $200 To Mexico From Selected US Cities

Factors Contributing to Remittance Prices: The potential impact of three sets of factors on remittance prices were analyzed quantitatively wherever possible, and qualitatively using anecdotal Examples, Corridor-Specific Factors,  Sending Country-Specific Factors and  Receiving Country-Specific.

Corridor-Specific Pricing Factors:  Corridor volume appears to be the single most important factor contributing to remittance prices.The next most influential price determinant is the extend to which global MTOs’ partner with national postal systems to expand their agency networks in either the sending or receiving end of the corridor and other factors that appear to have an impact on corridor pricinginclude: active participation of banks, credit unions or other non-bank financial institutions in the remittance market; cultural and geographic commonality with group of countries that includes one highly competitive, high volume corridor with lower prices;   the strength of informal transfer network in the corridor and  government policy initiatives within a corridor Factors.

Sending Country-Specific Pricing Factors:  A restrictive regulatory climate may be discouraging competition from smaller MTOs and enabling large players to maintain higher fee prices in the EU and, to less of an extent, the US corridors.Other sending country factors that appear to have an impact in pricing include: migrant access to low cost alternative remittance options through banks and non-bank financial institutions and the level of competition from the informal transfer network within the sending country

Receiving Country-Specific Pricing Factors:  There are several specific cases in which the access to modern, efficient, low cost transfer, payment and clearing systems in the receiving country may have been a factor in remittance prices, LikeAnalik, Bancomer, Credit union access to payment/clearing systems.  In receiving countries with high levels of mobile phone, Internet,ATM and POS usage, product and service innovations utilizingthese technologies are beginning to offer low cost remittancealternatives that could have a significant impact on remittancepricing in the future are Smart communication and Xoom.

To summarizecorridor volume, which attracts global and regional MTOs as well as small niche players that compete on price, appears to be the most significant factor driving remittance fee prices.Exclusive agency partnerships between large Global MTOs and postal systems also appear to have a major influence on remittance prices by serving as a significant entry barrier to small competitors and allowing high prices to be maintained even in the face of serious competitive challenges from other large, lower priced rivals.Restrictive regulations in the US and sending countries in the EU may also have an impact on pricing by discouraging new competition and making it easier for the large global MTOs to maintain higher prices. Other factors that appear to influence remittance pricing within a corridor include: the active participation of banks, credit unions and other non-bank financial institutions in the remittance market, migrant access to low cost alternatives offered by these organizations, technology and product/service innovations, and the strength of the informal transfer network. Government policy initiatives may have helped reduce remittance prices in selected corridors such as the US to Mexico, but the significance of their impact is difficult to ascertain.

Remittance Fee Pricing Trends and Driving Factors:  Remittance fee prices have declined dramatically in corridors such as the US to Mexico and Hong Kong to the Philippines in recent years. The study findings were used to identify the competitive, technological and policy factors that appear to be most influential in driving these price reduction trends and to make some observations on the future direction of remittance fee.

Competitive Factors:  The proliferation of new, small and mid-sized niche MTOs that competeby offering the lowest prices.  Intensified competition in high volume corridors between global andregional MTO’s seeking additional volume and increased market share.Increased competition in emerging corridors, such as the Gulf to SouthAsia, as global, and some large regional MTOs implement aggressive expansion programs in these areas. The active participation of banks such as Bank of America, and ICICIoffering low or negligible transfer fees to attract migrant accounts.  The activities of banks such as Bancomer and Analik, that focus on low priced remittance services as a major source of revenue.  The participation of credit unions and other non-bank financial institutions offering low fees for their members

Technology and Product Factors:  Investment in low cost transfer, processing and settlementsystems employing modern technology.  The introduction of innovative, low cost alternative remittanceproducts and services using the internet, mobile phones, ATMs,credit/debit cards or POS devices, such as those offered by Xoom, Ikobo, Smart Communications and others.

Government Policy Factors:  Migrant education programs that include information on remittance options and their costs, such as those provided by the Philippines Government.Unilateral policy initiatives in receiving countries, such as Mexico’s “Matricula Consularâ€� identity card program that provides migrants with access to lower cost remittance services offered by banks and other financial organizations. Bilateral policy initiatives such as the joint US-Mexican program that linked the automated clearinghouse (ACH) systems used by their central banks to reduce the cost of remittance transaction

Future Fee Pricing:  After a rapid rate of decline from 1999-2003, remittance fees in the US-Mexico corridor appear to be stabilizing, and major decreases are not expected in the near future.  The remittance industry appears to be consolidating.  Global MTOs are acquiring large regional players (WesternUnion’s pending acquisition of Vigo).   Regional players are being acquired as growth platformsby large financial service organizations (Global Paymentsacquisition of Dolex).  Mid-sized MTOs are merging with or acquiring other midsizedfirms and small niche players in high volumecorridors such as the US to Mexico.  Industry consolidation could significantly reduce the numberof small to mid-sized competitors and lead to less price

To Summarize   Competition within the higher volume corridors, particularly among the small to mid-sized niche players that compete by offering the lowest prices, is most responsible for the trend toward lower remittance fees. Technology and product /service innovations are expected to have a major influence on prices in the future, but so far have only had a significant impact in a few areas. Policy initiatives appear to have been helpful in some corridors, but the significance of their impact is difficult to access. After any major drop, prices in most high volume corridors are likely to stabilize, particularly if the trend toward industry

Remittance fees are high, regressive, and nontransparent

Remittance fee pricing is complex, and rarely are   senders   informed   about   the   full   and precise price of a remittance transaction. Fees may be as high as 20 percent of the principal, depending on the remittance amount, channel, corridor, and transaction type. The average  price  is  reported  to  have  been  around 12  percent  of  the  principal  in  2004  (Taylor 2004;  Kalan  and  Aykut  2005).  Prices are believed to have declined recently but are still very high in low-volume corridors. Currency- conversion  charges  are  even  less  transparent than  remittance  fees;  they,  too,  vary  depending  on  the  competitor,  corridor,  and channel, ranging    from    no    charge    in    dollarized economies   to   6   percent   or   more  in   some countries   (Orozco   2004;   Hernández-Cos  2004; Kalan and Aykut 2005).

Major  MTOs  such  as  Western  Union  and MoneyGram  apparently  charge  higher  remittance   fees   than   banks   and   other   financial institutions   that   offer   remittance   services to   attract   migrant   customers. Informal channels such as hawala are reported to   be   cheaper   than   formal   services.   Some heavily traveled remittance corridors, such as United   States–Mexico   and   South   Africa– Mozambique, are much cheaper than others. Urgent  transactions  delivered  in  minutes  cost much more than next-day transfers, and electronic transfers cost more than bank checks or drafts,  because  they  also  clear  much  faster than the latter.

The fee amount also depends on the remittance amount.  Average remittance  fees,  as  a percentage  of  money  sent,  decline  rapidly  as the transaction size increases, indicating scale economies   and   the   potential   advantage   of bundling  remittances—that  is,  the  advantage of  sending  more  funds,  but  less  frequently. According to one firm’s fee schedule, the cost of  sending  money  from  Belgium  to  Africa drops  from  21  percent  to  below  4  percent as   the   transaction   amount   increases   from 40  euros  to  900  euros.  Similarly, the cost of remittances from the United States to Mexico (through the major MTOs) is more than 10 percent for $100, but less than 3 percent for $500.

In recent years, remittance fees have declined in high-volume corridors in response to several   factors.   First,   global   and   regional MTOs   have   intensified   their   competition in   mature   corridors   (United   States–Latin America, for example), as new competitors have   been   attracted   by   high   and   growing remittance   volumes.   In   the   United   States– Mexico corridor, for example, remittance fees have dropped nearly 60 percent since 1999.   Second,   Bank   of   America   and other banks in source countries are using minimal transfer fees to attract migrant accounts, while a growing number of banks in recipient countries (including ICICI and Bancomer) are competing for remittance customers.  Third, the use of Internet-based technology for messaging  and  advanced  clearing  and  settlement has  reduced  the  cost  of  remittance  transactions. In some countries, new remittance tools have   emerged   based   on   cell   phones   and smart cards. Finally, government policies to improve transparency in remittance transactions (as in the United Kingdom), provide  financial  training  to  migrants  (as  in  the Philippines),  and  establish  bilateral  initiatives (such as the Partnership for Progress between the  United  States  and  Mexico)  have  helped reduce remittance costs. These   positive   developments   remain   the exception.   In   most   corridors,   particularly the   low-volume   corridors,   remittance   fees continue   to   be   very   high.   In   the   New Zealand–Tonga   corridor,   for   example,   fees are  about  three  times  as  high  as  those  in  the United States–Mexico corridor. The wide gap between remittance fees and costs shows that both should be reduced.

Fees for transferring EURO 300 from Germany to Serbia-2005

Remittance Fees paid by Recipients in case of immediate withdrawal of funds in Serbia

Jose  De Luna Martinez, Esaku Endo, and Corrado Barberis (2006): The German Serbia Remittance Corridor-Challenges of Establishing a Formal Money Transfer System. The World Bank

The main challenge for authorities is to ensure the integrity of the system by reducing the opportunities for misuse, while also aiming at minimizing the disruption and cost of the service for bonafide participants. In March 2006, a task force including the World Bank and the Committee on Payments and Settlement Systems (CPSS) released a set of General Principles for international remittance services. These include, first: Transparency and consumer protection-the market for remittance services should be transparent and have adequate consumer protection. Second: Payment System Infrastructure-improvements to the payment system infrastructure that have the potential to increase the efficiency of remittance services should be encouraged. Third: Legal and regulatory environment-remittance service should be supported by a sound, predictable, nondiscriminatory, and proportionate legal and regulatory framework in relevant jurisdictions. Fourth: Market structure and competition-competitive market conditions, including appropriate access to domestic payments infrastructures, should be fostered in the remittance industry. Fifth: Governance and risk management-remittance service should be supported by appropriate governance and risk management practices.

Types of Remittance Service and Accessibility Considerations

Section Five: Cost of Remittance

The cost of a remittance transaction appears to be far lower than the price: Service  providers’  remittance  costs  appear  to be  much  less  than  the  fees  charged  to  customers. Domestic transfer fees are only a fraction of the cross-border remittance fees (net of the currency-conversion charge). The cost of a domestic   automated   clearinghouse   (ACH) payment in the United States is one-third of a cent.  Domestic transfers using Visanet cost 2 cents per transaction, as opposed to 51 cents per   transaction   for   international   transfers (Brocklehurst, 2004).  In  some  corridors,  fees for  international  remittances  are  as  low  as $1.80    per    transaction    (London-Manila), which  hints  at  a  falling  lower  bound  for  the cost of remittances. The fact that some banks have been offering free remittance services as loss-leaders to attract new business suggests that the actual cost of remittances is modest. Courier  services  that  offer  remittances  also charge  small  fees  for  this  additional  service. Finally,   industry   cost   estimates   as   well   as other   calculations   presented   below   suggest that remittance costs are not very high.

The cost of providing remittance services varies with the business model used by the service provider.  Western  Union,  MoneyGram, and  Vigo  use  agents  who  pay  all  operating costs  in  exchange  for  their  franchise  and  a commission  on  sales.  In  the  “branchâ€�  model used   by   Dolex   and   many   of   the   smaller regional MTOs, the fixed and operating costs associated  with  each  branch  are  paid  by  the MTO.  By  leveraging  existing  businesses  on  a commission  basis,  the  agency  model  is  much less  capital-intensive  than  the  branch  model and can be expanded rapidly through partner- ships, but it has higher variable costs.4 In both models, relatively high fixed costs are associated  with  transaction-processing  operations, compliance   with   regulatory   requirements, marketing, and administration.5

Data on MTOs’ costs of providing remittance services are hard to obtain. However, an analysis of profitability of the market leaders using  publicly  available  financial  statements suggests that remittance costs are significantly lower  than  the  fees  charged  to  customers. Western    Union    has    sustained    operating margins  that  are  at  least  50  percent  higher than  other  MTOs  and  industry  peers  in  the payments  and  electronic  processing  market  (table 6.2).7 Its operating profit per remittance transaction  may  have  averaged  $8  to  $9  in 2004. This is consistent with an earlier annual report   (Western  Union  2000)   that   put   the company’s  operating  profit  at  $684  million (or 30 percent of its $2.3 billion revenue). The operating profitability of the other major market players (MoneyGram and Dolex) has been in the range of 15–20 percent (table 6.2).  A very simple model for Western Union (which assumes that agency commission costs are 35 percent  of  revenues  after  deduction  of  fixed costs  and  that  all  other  costs  are  fixed  costs) suggests that average transaction fees could be reduced by as much as one-third while maintaining  operating  margins  within  the  same range as those of other major MTOs and peers. Reducing   these   operating   profits   to   zero would provide a rough estimate of the break- even cost for these firms. Such an exercise reveals   that   the   break-even   fee   for   Western Union is probably around $9 per transaction and  would  fall  below  $5  if  the  volume  of transactions  were  to  double  (box  6.2).  Although it would be unreasonable to suggest that any company reduce its prices to cost, this simple model does appear to indicate that there is considerable latitude for reductions in transaction fees within the higher-priced corridors. A more direct way of estimating the cost of a   remittance   transaction   in   a   hypothetical MTO is to add up plausible cost components, such  as  staff  to  process  the  transaction  and provide security, rental of the premises, fixed costs  (including  franchise  licensing),  the  cost of  network  and  technology,  and  administrative  costs  for  regulatory  compliance. This methodology yields a cost estimate of $5.50 for the first remittance transaction (table 6.3). Because   most   remittance   transactions   tend to   be   repetitive—the   same   amount   is   remitted  from  the  same  location  to  the  same beneficiary—the  cost  for  subsequent  transactions  drops  to  $3.60  (less  staff  time  is  required). It drops to under $3 per transaction if electronic processing is used.

Admittedly,  the  calculations  in  table  6.3 are  based  on  a  theoretical  model  of  a  basic remittance  transaction  that  does  not  capture the  global  network  and  diversified  services provided   by   major   MTOs.   Moreover,   the model’s assumptions are subject to considerable uncertainties, the greatest of which is that average costs would be higher if the number of transactions were smaller.  It  is  worth  noting, however,  that  many  independent  agents  provide remittances as a side business: for them, fixed and variable costs could be significantly lower  than  for  dedicated  remittance  service providers. Indeed, there may be a case for providing  free  remittance  services  in  order  to draw  customers  for  other  products  and  services, as practiced by certain banks. Remittance  costs  should  continue  to  fall under  the  influence  of  increased  competition and   better   technology.   Large   MTOs   may have   considerable   latitude   to   reduce   fees while maintaining reasonable profit margins. In  corridors  where  costs  have  already  fallen significantly,  further  decline  may  be  modest; but elsewhere there is scope for significant de- cline,  especially  with  the  volume  of  transactions rising rapidly.

A scorecard on remittance transfers:Competition between MTOs has been a major contributor to the decline in transaction costs and the emergence of many financial institutions as remittance payers. For the most part, money transfer companies have been proven to work in a competitive environment with financial institutions. In this section we examine those companies that have had the best performance within the realm of money transfers. Industry officials and observers of this market argue that it is important to look at factors other than transaction costs as indicators of a company’s importance to consumers, advocates and development players. For example, some argue that including information about geographic distribution as well as legal compliance, it is also critical in understanding performance. Our industry scorecard was based upon a quantitative framework that served as the basis of measurement for evaluating the ways in which money transfer companies respond to a range of important factors associated with remittance transfers.

The framework synthesis provided here includes nine criteria for an analysis of market performance in relation to development and consumer rights. The criteria includes transfer fees and exchange rate commissions, mechanisms used to send money, competitive position in the corridor, geographic coverage across corridors, levels of engagement with the local consumer community, relationships with financial intermediaries, transparency in disclosing information about pricing, and compliance to regulatory rules. We collected data from more than 50 money transfer companies on issues relating to costs, locations in the U.S., types of payers in Latin America and the Caribbean, consumer satisfaction with companies, and the relationship between the geographic locations of payers and of the households that receive remittances. The methodology employed was based on field work data collection on pricing, interviews with money transfer companies about their locations and payers, and surveys on consumer satisfaction. Companies were scored according to whether their activity was above or below the average. Furthermore, data was not collected for all of the indicators listed in Table 7 (see below). Specifically, data was not collected for three criteria relating to disclosure practices, compliance to the regulatory environment and corporate philanthropy, because the appropriate methodology is still being discussed with money transfer companies.

Scoring criteria and their measurable indicators

Reducing remittance fees will increase remittance flows to developing countries

Reducing  remittance  fees  would  increase  the disposable  income  of  remitters,  encouraging them  to  remit  more.  It also might encourage smaller    and    more    frequent    remittances.

And   lower   prices   in   a   particular   channel might encourage remitters to shift from other channels—notably informal ones. The degree to which a fee reduction would result  in  an  increase  in  flows  depends  on  the purpose  of  the  remittance.  At  one  extreme, where the purpose is to meet a specific need— payment  for  tuition,  a  medical  emergency,  a social  ceremony,  or  the  purchase  of  a  gift item—the  amount  of  remittance  may  not  be sensitive  to  the  remittance  fee.  At the other extreme, remittances by a poor, cash-strapped remitter may be highly cost elastic.  Similarly, remittances meant for investment are likely to be   cost elastic. In reality,   most   remittance transactions fall between these two extremes. Even when remittances are driven by altruism, they will tend to be cost elastic, as evidenced by the literature on charity, which shows that people tend to donate more as the cost of donating declines.

In a recent survey of Senegalese migrants in Belgium, two-thirds of the migrants said they would send more if the cost of sending went down. In a survey of Tongan migrants in New Zealand, 30 percent of remitters said they would increase the amount of remittances by 0.74 percent (on average) if costs fell by 1 per- cent (Gibson, McKenzie, and Rohorua 2005). That survey found the overall cost-elasticity of  remittances with respect to the fee (averaging the  elasticity  over  those  who  would  increase remittances  and  those  who  would  not)  to  be –0.22. Based on this estimate, Gibson and others  (2005)  calculate  that  lowering  the  fixed cost  of  sending  money  through  banks  and MTOs from New Zealand and Tonga to competitive  levels  in  the  world  market  would  result  in  a  28  percent  increase  in  remittances from  existing  remitters.  It might also induce some non-remitters to start remitting.

If  the  cost  elasticity  (–0.22)  of  the  New Zealand–Tonga  study  were  applicable  to  all developing     countries,            a          reduction         in remittance  cost  from  12  percent  to  (say)  6 percent could result in an 11 percent increase in   annual   remittance   flows   to   developing countries.  One  caveat  to  this  calculation  is that  the  cost  elasticity  applies  only  to  high- cost  corridors,  which  also  tend  to  have  low volumes.  In corridors where the remittance cost is already low, further decreases may not increase flows.  For  example,  a  fee  reduction by  a  major  MTO  may  not  produce  much  effect  if  a  major  part  of  the  flows  is  already moving  through  low-cost  informal  channels. This  is  confirmed  by  the  World Bank survey of Senegalese migrants in Belgium; half of the respondents  who  paid  remittance  fees  of  20 percent or more said they would send more if costs  were  halved;  not  even  one-fourth  of those who paid less than 10 percent said they would send more. Almost 75 per- cent   of   the   Senegalese   migrants   who   send money   through   the   large   MTOs   said   that  they   would   send   more   if   the   costs   were lowered,  a  result  confirmed  by  findings  from a World Bank survey of the Nigerian diaspora in Belgium.

An  indirect  implication  for  cost  elasticity may be drawn from Yang’s (2004) finding of an  elasticity  of  0.6  for  remittance  receipts denominated  in  Filipino  pesos  with  respect to  the  peso–dollar  exchange  rate.  Applying this  elasticity  to  a  remittance  transaction  of $150, if the remittance fee were halved from (say)  12  percent  to  6  percent,  remittance  receipts  would  rise  by  3.6  percent,  or  $5.4, while  the  remittance  fee  would  decline  from $18  to  $9.31.10   If  the  same  elasticity  were to  apply  to  the  entire  flow  of  remittances  to developing  countries,  remittance  receipts,  in response   to   a   halving   of   costs  would   increase  significantly, by  more  than  $5  billion using  only  recorded  flows,  and  more  than $8   billion   using   both   recorded   and   unrecorded  flows. Reductions in remittance fees would also be likely   to   increase   other   cross-border   retail flows such as transfers from public and private institutions  to  individual  beneficiaries  (pensions,  child-care  payments),  small-value  payments in exchange for goods and services, acquisitions  of  assets,  and  debt  servicing.   In more developed countries, migrant remittances are only a small share of retail payments, which, in turn, are a fraction of wholesale payments. But in developing countries, especially in  smaller  and  poorer  countries,  remittances are a significant source of funding in relation to  the  size  of  the  economy  and,  therefore,  of the retail payment system. A reform of the retail payment system to facilitate remittances would probably benefit other (not easily quantifiable) components of retail payments.

Based  on  the  evidence  presented  above, notably  the  finding  that  the  cost elasticity  of remittances  is  negative,  policies  that  aim  to lower  remittance  costs  by  increasing  access to  banking  services,  promoting  competition, and   disseminating   information   have   the potential   to   provoke   sizeable   increases   in remittance flows to developing countries.

Policiestoreduceremittancecosts

Measures   to   reduce   remittance   costs should aim to improve the efficiency of remittance transactions by (a) enhancing market competition to reduce high profit margins; (b) helping remittance service providers’ access to new payments technology; and (c) devising ways to encourage remitters to send larger amounts.  As  a  way  to  enhance competition,   governments   can   encourage postal systems and other state-owned distribution  alternatives  to  open  their  networks  to multiple MTO partnerships on a nonexclusive basis. In addition, they should avoid overregulation,  excessive  monitoring,  or  reporting  requirements  that  could  drive  out  smaller  competitors  that  lack  the  economies  of  scale  to absorb the cost of compliance. Developing  a  shared  network  would  be  a powerful  way  to  increase  competition.  Cooperation  on  infrastructure  and  competition in  service  provision  would  allow  network benefits  to  accrue  to  the  consumer.  The technology  required  to  set  up  a  payment- processing  infrastructure  with  large  capacity is   no   longer   an   expensive   proposition.   A functioning payment infrastructure could be extended to a new country at a minimal cost and in a matter of weeks.  There have been some attempts to set up shared networks in the remittance-source countries.  Also  some  governments in remittance-receiving        countries have facilitated the establishment of payment net- works   that   are   shared   by   savings   banks, credit  unions,  and  microfinance  institutions operating   in   poor   and   remote   areas   (for example,  BANSEFI  in  Mexico   and  Apex Link in Ghana).  Another way to address the issue of high fees in the remittance industry would be to develop best-practice guidelines for remittance service providers. Several such guidelines have been is- sued  by  Credit  Union  National  Association, Inter-American Development Bank, and World Savings  Bank  Institute,  which  urge  service providers to disclose fees, exchange rates, and the time of delivery. At the end of 2004, the World Bank and the Bank for Committee on Payment and  Settlement  Systems  (CPSS)  set  up  a  task force,withparticipationfromtheIMF,todevelopvoluntaryprinciplesforremittanceservice providers,  regulators,  and  supervisors  for  im- provingtransparencyinthemarket.

Such guidelines would have to be voluntary. Central banks generally are not willing to impose such guidelines or to cap remittance fees and foreign-exchange commissions.  A  recent survey (de Luna Martinez 2005) revealed that in only 9 of 40 countries—Brazil, Bulgaria, Indonesia, Pakistan, Philippines, Russian Federation,  Thailand,  Tunisia,  and  República  Bolivariana   de Venezuela—did   central   banks even have the legal power to do so. All 40 central banks indicated that even if they had the power to limit fees, they would not do so, preferring to leave fee-setting to financial institutionsin response to market competition.

Policies to reduce costs, regulate informal providers, and provide remittance- linked financial services

Approximate cost of remitting $200: Percent of principal amount

Source:  Brocklehurst2004;Orozco2004;Gibson,McKenzie,Rohorua2005;Hernandez-Coss2004;RathaandRiedberg2005;KalanandAykut2005;Andreassenandothers2005. *WorldBanksurveyofAfricandiasporasinBelgium.  Note:  Figuresdonotincludecurrency-conversioncharge. —  Datanotavailable

Estimating the cost of a remittance transaction

Cost in dollars

Box: Estimatingremittanceindustrycosts

Remittance industry costs are difficult to obtain.

Isolatingthecostofremittanceservicesisdiffi

cultinthecaseoffinancialinstitutionsthatprovide otherservicesaswell.Estimatingcostsisnoteasy eveninthecaseofdedicatedremittanceservice providersbecauseofthedifferencesinthequality andreliabilityofremittanceservices(onlysome providersgivecustomerslegalredress).InRemit-

tanceindustrycosts,therefore,wehaveusedpublicly availableinformationonWesternUnion,thelargest MTOthatisalsoapubliclylistedcompany.

Weusedasimplemodeltoestimateabreak-even feeforWesternUnion’sinternationalmoneytransfer operations.ThemodelsuggeststhatforWestern Union’soperatingmarginsonitsinternationalmoney

Dollars

10

8

6

4

2

0

Western Union:Operating profit

break-even price vs.volume

transferstodroptothepeergroupaverageof17.8

75     100    125    150    175    200

225    250    275    300

percent(table6.2),theaveragetransactionfeewould havetobeloweredfrom$22.90to$15.30(column2 ofthetablebelow)—veryclosetothecompany’scur- rentfeeinseveralU.S.corridors.Themodelalsoindicatesthatthebreak-evenfeeatwhichtheoperating profitbecomeszerois$9.30(column3).Thispriceis inthesamerangeasMoneyGram’sstandardflat priceintheU.S.corridors.Asensitivityanalysisusing thismodelsuggeststhatthebreak-evenfeewouldbe

$6.50–$7.00ifagencycommissionswere25percent, andaround$11ifcommissionswere45percent.

Transaction volume (millions)

Source:KalanandAykut2005.

Thefigureillustrateshowthebreak-evenfee showninthetabledecreasesasthenumberoftransactionsincreases.Iftransactionvolumedoubled

fromthecurrent76millionto150million,the lowestfeeatwhichtheinternationaloperation wouldremainprofitablewouldbe$4.74