## wpiea2022233-print-pdf

## Source details

**Canonical URL:** [wpiea2022233-print-pdf](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022233-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022233-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022233-print-pdf.pdf.json)

---

### I. Introduction and scope
- Analysis covers 365 corridors and spans from 2011Q1 until 2020Q4.
- Two central questions:
  - Has the enthusiasm about bitcoin, mobile money and remtechs’ effects on the remittances market been justified?
  - Have remtechs played a major role in driving remittance costs down?
- Data source: Remittances Price Worldwide (RPW) database augmented by external identification of remittance fintechs (remtechs).

### II. Fintech definitions and narratives
- Fintech and remtech definitions:
  - Fintech: products and companies that employ newly developed digital and online technologies in banking and financial services.
  - Remtechs (remittance fintechs): firms whose business models rely on the virtual landscape and provide remittance services via personal computers and mobile devices.
- Dominant disruption narratives reviewed:
  - Bitcoin/blockchain: claims of remittances “almost 600 faster” and “almost 250 times cheaper” via blockchain.
  - Remtechs: advertised lower fees and better exchange rates by avoiding high street premises and legacy networks (examples: Azimo, WorldRemit, Remitly, Wise/Transferwise).
  - Mobile money: expectations focused on financial inclusion rather than wholesale remittance-cost reduction.

### III. Remtech footprint, funding and market presence
- Comparator remtech group: Azimo, InstaRem, Remitly, TransferGo, Wise, WorldRemit and Xoom.
- Funding and reported volumes:
  - Wise transferred £54 billion (USD74 billion) in FY2021.
  - Remitly moved USD16 billion in a 12-month window through end-June 2021.
  - WorldRemit reported almost USD10 billion by its two brands.
  - InstaRem reports “... moves $4 Billion annually.”
- Valuations and exits:
  - Wise IPO valuation reported at USD11 billion.
  - Remitly IPO valuation reported at USD7 billion.
  - WorldRemit valued at USD5 billion in August 2021 funding round.
  - InstaRem valued above USD1 billion in July 2021 funding round.
  - Xoom acquisition price by PayPal: USD890 million.
  - Azimo acquisition valuation between USD150 million and USD200 million (midpoint used as estimate).
  - No valuation info available for TransferGo.
  - At end-October 2021, Wise and Remitly market capitalizations hovered just above USD8 billion and USD5 billion respectively, reflecting a fall of around 25% since their IPO.
- RPW-based growth and coverage (entries and corridor presence):
  - 2011Q1–2016Q1: Banks 26.8%, MTO 69.2%, PO 1.4%, Remtech 3.3% (Total entries).
  - 2016Q2–2020Q4: Banks 18.6%, MTO 78.2%, PO 1.3%, Remtech 11.1% (Total entries).
  - Corridor presence 2011Q1–2016Q1: Banks 66.5%; MTO 99.7%; PO 14.2%; Remtech 38.4%.
  - Corridor presence 2016Q2–2020Q4: Banks 68.4%; MTO 96.7%; PO 11.4%; Remtech 68.8%.
  - Interpretation: remtechs grew from ~3 percent of RPW entries to over 10 percent and expanded to almost 70 percent of corridors in the second half of the sample.

### IV. Aggregate remittance-cost trends and regional patterns
- Global trend:
  - Remittance costs fell on average from 2011Q1 to 2020Q4 but remain well above the SDG target of 3 percent; after large decreases until 2015 the reduction slowed markedly.
- Regional changes (2020 vs 2015 averages):
  - SSA and EAP experienced largest decreases but amounted to around 0.75 percentage point.
  - ECA and LAC experienced average decreases around 0.25 percentage point.
  - SAR and MENA experienced increases in costs of around 0.40 and 0.80 percentage points respectively (MENA increase driven sharply by Syria; excluding Syria, average costs in MENA would have decreased by around 0.4 percentage point).
- Distributional facts:
  - Among 105 remittance-receiving countries, around 20 percent faced costs at least as high as 10 percent in 2020.
  - Among 365 corridors, around 15 percent faced costs around 10 percent or higher; 5 percent faced costs at least as high as 15 percent in 2020.
  - 4 of the 5 most expensive remittance-receiving countries are in SSA; 65% of the 20 most expensive remittance-receiving countries and corridors are in SSA and between SSA countries.
  - Among the 20 least costly countries, El Salvador, Panama and Ecuador noted as dollarized economies reducing FX risks.

### V. Evidence on remtech presence versus costs
- High-level observations:
  - The surge in remtech corridor presence coincided with a slowdown in the decline of global remittance costs.
  - Banks are the most expensive RSP type in unadjusted series; remtechs the cheapest in unadjusted comparisons (Figure 6B referenced).
- Five-year changes (2015 to 2020) by RSP type (average-cost changes):
  - MTO and banks’ average costs decreased by 0.8 p.p. vs 0.9 p.p. respectively.
  - Remtechs’ costs decreased marginally by 0.3 p.p.
  - Bank–MTO cost gap remained largely unchanged globally.
- Corridor-level clustering by remtech presence:
  - 2015: corridors with remtech presence had average Bank–MTO gap 3.7% versus 2.9% without remtech presence.
  - 2020: corridors with remtech presence had average Bank–MTO gap 3.3% versus 3.6% without remtech presence.
  - Differences in both years are not statistically significant; endogeneity (remtechs entering higher-cost corridors first) is a possible explanation.
- Selected averages (Table 4 summary):
  - 2015 sample (172 corridors; 72 with remtech presence):
    - Remtech corridors: Bank 10.5%, MTO 6.8%, Digital 6.3%.
    - No remtech corridors: Bank 10.4%, MTO 7.4%, Digital 8.2%.
  - 2020 sample (239 corridors; 164 with remtech presence):
    - Remtech corridors: Bank 9.3%, MTO 6.0%, Digital 5.2%, Mobile Money 4.3%.
    - No remtech corridors: Bank 9.4%, MTO 5.8%, Digital 7.4%, Mobile Money 4.1%.
- Interpretive points:
  - Digital remittances in corridors with remtechs cost consistently less than in corridors without remtech presence.
  - Mobile remittances appear to be the cheapest way to remit, especially when the same operator is present at both ends.

### VI. Bitcoin/blockchain: technical and empirical constraints
- Technical and market realities:
  - Bitcoin transaction fees volatile; have reached levels "above USD60 per transaction" during high market activity.
  - Average fees in the second half of 2021 were "around USD3.0".
  - Since 2015, costs have been above "USD5.0" fifteen percent of the time — implying in a USD200 remittance transaction at least a "2.5 percent" transaction cost.
  - Transaction times sometimes exceeded two days during congestion.
  - Additional costs: conversions, crypto-exchange spreads, BATM fees (usually "five percent or higher"); BATM distribution skewed to richest regions, mostly US; poorest regions like Africa have negligible BATM presence.
  - Complexity (private keys, wallet management) and financial-literacy barriers deter uptake.
- Conclusion: bitcoin is "not economically viable" as a mainstream remittance channel; bitcoin/blockchain-based remittances are "virtually nonexistent."

### VII. Mobile money: successes and limits
- M-Pesa case (Kenya):
  - 2007 creation; frontend based on simple text messaging and simple phones.
  - 2014: 55% of Kenyans had accounts at financial institutions (up from 42% in 2011); 58% had mobile money accounts.
  - 2017: 56% had accounts at financial institutions; 73% had mobile money accounts.
  - Inclusion gap widened from "3p.p." to "17p.p." (mobile money vs banking).
  - M-Pesa customer base: "51 million customers across seven African countries" and "more than 600,000 active agents."
  - In 2020, M-Pesa accounted for 98% of the mobile money market in Kenya.
- Global/mobile-money patterns and limitations:
  - 2017: around one third of SSA population age 15+ had bank accounts at financial institutions.
  - 2017: SSA hosted around "45 percent" of global registered mobile money accounts.
  - GSMA (2021): "73 percent" of countries analyzed in LAC had a Mobile Money Prevalence Index rated as low or very low in 2021; in SSA around one third of countries were rated low or very low.
  - Mobile money remittances amounted to "just over 2 percent" of the total value of mobile money transactions in 2021; "84 percent" of that took place in SSA.
  - Domestic transfers made the bulk of mobile money transactions in 2021.
- Conclusion: mobile money is a strong financial-inclusion tool in contexts like Kenya but has not scaled into a dominant international remittance channel globally.

### VIII. FX margins: measurement issues and adjusted margins
- Raw FX margins by RSP type and region (2011.Q1 – 2016Q1 and 2016.Q2 – 2020Q4):
  - Example unadjusted margins (2011.Q1 – 2016Q1):
    - Bank: ECA 0.4%, EAP 2.5%, LAC 1.2%, MENA 1.0%, SAR 1.7%, SSA 1.5%.
    - MTO: ECA 1.4%, EAP 1.9%, LAC 1.9%, MENA 2.3%, SAR 1.5%, SSA 2.4%.
    - Remtech: ECA 1.0%, EAP 1.7%, LAC 1.6%, MENA 2.7%, SAR 1.3%, SSA 2.5%.
  - Example unadjusted margins (2016.Q2 – 2020Q4):
    - Bank: ECA 0.5%, EAP 2.5%, LAC 1.2%, MENA 1.8%, SAR 1.5%, SSA 2.2%.
    - MTO: ECA 1.8%, EAP 2.1%, LAC 2.0%, MENA 2.5%, SAR 1.5%, SSA 2.7%.
    - Remtech: ECA 1.4%, EAP 1.7%, LAC 1.5%, MENA 2.2%, SAR 1.3%, SSA 2.5%.
- Measurement anomalies:
  - RPW shows many zero or negative FX margin entries:
    - Zero margins: Banks 30%, MTO 17%, Remtechs 14%.
    - Negative margins: Banks 3%, MTO 4%, Remtechs 8%.
  - Estimated FX bias (replacing zero/negative margins with average margin for that quarter net of zeros/negatives):
    - Banks 0.9p.p., MTO 0.6p.p., Remtechs 0.6p.p.
  - Implication: unadjusted RPW series understate FX margins; up to one third of bank entries questionable.
- Adjusted FX margins (after accounting for estimated FX bias):
  - 2011.Q1 – 2016Q1 adjusted margins:
    - Bank: ECA 2.6%, EAP 3.1%, LAC 2.7%, MENA 2.7%, SAR 2.4%, SSA 2.7%.
    - MTO: ECA 2.8%, EAP 2.2%, LAC 2.9%, MENA 2.5%, SAR 1.7%, SSA 3.2%.
    - Remtech: ECA 1.8%, EAP 1.9%, LAC 2.6%, MENA 2.8%, SAR 1.5%, SSA 2.9%.
  - 2016.Q2 – 2020Q4 adjusted margins:
    - Bank: ECA 2.6%, EAP 2.8%, LAC 2.7%, MENA 2.8%, SAR 2.1%, SSA 2.9%.
    - MTO: ECA 3.0%, EAP 2.4%, LAC 2.8%, MENA 2.8%, SAR 1.7%, SSA 3.3%.
    - Remtech: ECA 2.3%, EAP 2.3%, LAC 2.3%, MENA 2.4%, SAR 1.6%, SSA 2.9%.
  - Change between periods (adjusted, selected):
    - Remtech change ECA 0.4p.p., EAP 0.3p.p., LAC -0.3p.p., MENA -0.3p.p., SAR 0.1p.p., SSA 0.0p.p.
- Key implications of adjusted margins:
  - After correction remtechs have the lowest FX margins in more regions (four out of six in the first period; all regions in the second period tied in SSA in some comparisons), improving remtech standing versus unadjusted data.
  - Margin differences are often modest in magnitude relative to the second-lowest margin.
  - In SSA, remtechs do not clearly deliver largest savings; remtechs’ average FX margin is the same or higher than banks’ in some comparisons.
  - Correcting bias implies RPW underestimates remittance costs and that the SDG target is further away than unadjusted figures suggest.

### IX. Market structure, incumbents, and remtechs’ strategic behavior
- Market-entry and geographic patterns:
  - Remtechs often launched in developed, competitive, lower-cost corridors (e.g., Remitly debuted in US–Philippines; took four years to enter US–Mexico and US–India).
  - By 2011.Q1–2020Q4 many remtechs had limited meaningful presence in SSA (examples: Remitly meaningful in only four SSA corridors; TransferGo had no meaningful SSA presence nine years after starting).
  - With one exception (WorldRemit), remtechs generally not present in the 20 most expensive corridors.
- Business-model adaptations:
  - Remtechs broadened services to include cash pick-ups and home delivery by partnering with banks, traditional MTOs, mobile money operators and agents.
  - The peer-to-peer matching model did not scale across many rich–poor corridors; remtechs moved into partnerships, corporate markets, or were acquired (Xoom acquired by PayPal in 2015; Azimo acquired by Papaya Global).
- Positive roles of fintech despite limited disruption:
  - Increased competition: lower costs correlated with higher number of MTOs and higher MTO-to-bank ratios.
  - Greater pricing transparency and easier price comparison.
  - Digitalization reduces non-pecuniary costs (time, transportation) and improves safety by reducing cash handling.

### X. Why disruption did not occur and synthesis
- Principal constraints on disruption:
  - Predominance of cash at receiving end and low financial inclusion limit digital remittance penetration.
  - Infrastructure, interoperability, regulatory constraints, network effects and agent-network requirements (for mobile money and cash services) restrict scalable digital-only solutions.
  - Remtechs’ expansion often relied on incumbent networks and partnerships; scaling required non-digital footprints.
  - Bitcoin/blockchain technical limits and costs; mobile money successes context-specific and not easily replicable.
- Empirical synthesis:
  - The decline in remittance costs slowed as remtech corridor presence surged post-2015.
  - The bank–MTO gap remained largely stable from 2015 to 2020.
  - After correcting RPW measurement bias, remtechs look better on FX margins but advantages are regionally uneven and often modest.

### XI. Policy recommendations and considerations
- Policymaker stance:
  - Be skeptical portraying fintech as an easy solution to expensive financial services and low financial inclusion.
- Policy actions suggested:
  - Provide a level playing field for fintech and smaller MTO; address anticompetitive practices (e.g., exclusivity clauses).
  - Adopt proportionate, risk-based regulation that balances innovation and financial integrity objectives.
  - Facilitate interoperability, infrastructure sharing and compatibility to reduce costs of entry and scale for mobile money and digital channels.
  - Governments should invest in/upgrading financial infrastructures and be open to experimentation while preventing regulatory capture.
  - Tailor policy to acknowledge infrastructure costs and profitability challenges for agents and providers; enable viable business models for cross-border cash and digital services.

*Italic: Source — Excerpted material from IMF Working Paper "Curb Your Enthusiasm: The Fintech Hype Meets Reality in the Remittances Market" (wpiea2022233-print-pdf).*

### References .............................................................................................................

### References ........................................................................................................................................................................ 35

### Appendices
- Appendix 1. A S hort History of Remittance Fintechs ............................................................................................ 38
- Appendix 2. An Example of Measurement Error in the RPW Database............................................................. 40
- Appendix 3. A Radiography of Remtech’s Diffusion.............................................................................................. 41

### Figures
- 1. Evidence of Public Interest on Bitcoin, Blockchain and Fintech...................................................................... 4
- 2. Remittance Fintechs: Funding and Valuatio n (USD Billion) .............................................................................. 9
- 3. Remtechs: Growth and Geographic Coverage.................................................................................................... 10
- 4. Mobile-Related Remittances: Geographic Coverage ......................................................................................... 12
- 5. A Bird’s Eye View on Remittance Costs ............................................................................................................... 13
- 6. Remittance Costs and Fintechs .............................................................................................................................. 14
- 7. The Bank-MTO Cost Gap and Remtech Presence .............................................................................................. 16
- 8. Bitcoin Transaction Fee and Price Volatility ........................................................................................................ 19
- 9. Crypto ATM Distribution by Continents and Countries .................................................................................... 20
- 10. Mobile Money and Financial Inclusion ................................................................................................................ 22
- A2.1 Measurement Errors: Unusual Cost Variations in Selected Remittance Corridors .............................. 40

### Tables
- 1. Remittances Price Worldwide: RSP Share ........................................................................................................... 10
- 2. Remittances Price Worldwide: Selected Characteristics ................................................................................. 11
- 3. The Bank-MTO Remittance Cost Gap by Regions.............................................................................................. 15
- 4. Average Remittance Cost and Remtech Presence............................................................................................. 17
- 5. FX Margins by RSP Type and Region .................................................................................................................... 24
- 6. Percentage of FX Margins Equal to Zero or Negative (2011.Q1 – 2020Q4) .................................................. 24

*Source: wpiea2022233-print-pdf - References .............................................................................................................*

### 7. Adjusted FX Margins by RSP Type and Region ..........................................................................

### 7. Adjusted FX Margins by RSP Type and Region

### I. Introduction
- Recent years saw strong public interest in bitcoin, blockchain and fintech; Google Trends indicates “What is bitcoin?” and “How to buy bitcoin” were highly searched in 2017–2018, and in early 2021 searches for bitcoin reached twice as much as those for Covid-19.
- The surge in connectivity (internet, smartphones, communications networks) underpins recent fintech developments and enabled firms with small or negligible physical footprints to offer financial services more efficiently and at lower cost.
- Fintechs are defined here as companies whose business models heavily rely on the new virtual landscape and provide financial services via personal computers and mobile devices.
- Remittances are a policy priority because:
  - Retail cross-border payments are “typically perceived as being slower, costlier and more opaque than domestic payments.” (CPMI, 2018).
  - Remittances are a lifeline for millions and can affect poverty, health and education outcomes.
  - Remittances flows can reach levels above 30 percent of GDP in some developing countries.
  - Remittances are anti-cyclical and rise during difficult times.
- A global push to decrease remittance costs culminated in inclusion of remittance costs in the Sustainable Development Goals (SDG).
- Two central questions of the paper:
  - Has the enthusiasm about the effects of fintech (bitcoin, mobile money, remtechs) on the remittances market been justified?
  - Have remtechs played a major role in driving remittance costs down?
- Scope and sample:
  - Analysis covers 365 corridors.
  - Sample spans from 2011Q1 until 2020Q4 (extended from 2018Q3 in da Silva Filho (2021)).
  - The aim is to take stock of whether initial (over)optimistic predictions about disruption have materialized.

### II. What is Fintech?
- Conceptual clarifications:
  - Fintech is short for financial technology; fintechs are firms whose business models are indissociable from that technology.
  - Fintech is not entirely new: historical examples include ATMs, credit cards, SWIFT, internet banking and online brokers; mobile banking is a recent prominent manifestation.
  - Merriam-Webster defines fintech as “products and companies that employ newly developed digital and online technologies in the banking and financial services industries.”
  - The paper differentiates fintech (technology) from fintechs (firms) and notes incumbents also use new digital technologies but are not typically labeled fintechs.
- Expectations for disruption in remittances came from two main narratives:
  - Bitcoin and blockchain narrative:
    - Claimed to enable remittances that are much faster and much cheaper by bypassing intermediaries.
    - Some claims included settlement times almost 600 faster and costs almost 250 times cheaper using blockchain.
    - El Salvador’s decision to make bitcoin legal tender was portrayed as rooted in expected remittance cost reductions.
  - Remtechs narrative:
    - Remtechs (remittance fintechs) advertised lower fees and better exchange rates by avoiding costs of high street premises and legacy networks (examples cited: Azimo, WorldRemit, Remitly, Wise/Transferwise).
    - Promises often lacked detailed explanation of how lower costs or better exchange rates were achieved.
  - Mobile money expectations focused more on financial inclusion than on remittance cost reduction.

### III. The Fintech Landscape
- Practical challenges in identifying remtechs within remittance data:
  - The Remittances Price Worldwide (RPW) database classifies RSPs in three main categories: banks, MTO and PO; it does not identify remtechs.
  - To benchmark remtechs, external identification of well-known remittance fintechs is required; RPW’s inclusion of RSP names facilitates matching to fintech entries sourced outside the database.
- Relationship to prior work:
  - This paper complements da Silva Filho (2021) by adding granularity and extending the sample period.
  - da Silva Filho (2021) mapped drivers of remittance costs and analyzed cost differences by banks, MTO and PO but did not assess fintech impacts.
- Methodological note:
  - The paper focuses on seeking evidence of disruption (presence or absence of large-scale impacts by bitcoin, mobile money and remtechs on remittance costs), not on quantifying the total role fintech may have played in cost reductions.

*Italic: Source — Content excerpt from IMF Working Paper "Curb Your Enthusiasm: The Fintech Hype Meets Reality in the Remittances Market" (selected sections I–III as provided).*

### Annex 1 provides a concise history on fintechs venturing into the remittances market. As it turns out, it does not

### Annex 1 — Remittance Fintechs: history, market footprint, funding, and cost evidence

### Comparator group and data scope
- Comparator group of remittance fintechs (remtechs): Azimo, InstaRem, Remitly, TransferGo, Wise, WorldRemit and Xoom.
- RPW database contains more than 550 MTO entries; fintechs are part of the MTO group.
- A necessary condition for an RSP to be included in the RPW survey is having at least one percent market share.
- The remtech group is a subset of the MTO group and formed by the seven firms listed above.

### Funding, valuation, and market share concentration
- Funding concentration: overwhelming share of raised resources benefited a few remtechs; among the top 7 recipients, the top 3 tapped around 80% of the resources raised.
- Reported transfer volumes (company-reported):
  - Wise transferred £54 billion (USD74 billion) in FY2021.
  - Remitly moved USD16 billion in a 12-month window through end-June 2021.
  - WorldRemit (rebranded as Zepz) reported volume by its two brands reached almost USD10 billion.
  - InstaRem (rebranded as Nium) reports that it “... moves $4 Billion annually.”
- Valuations and exits:
  - Wise IPO valuation reported at USD11 billion.
  - Remitly IPO valuation reported at USD7 billion.
  - WorldRemit valued at USD5 billion in its funding round in August 2021.
  - InstaRem valued above USD1 billion in July 2021 funding round.
  - Xoom’s USD890 million valuation reflects its acquisition price by PayPal, in 2015.
  - Azimo was recently acquired at a valuation between USD150 million and USD200 million (midpoint used as estimate).
  - No valuation info available for TransferGo.
- Note: At end-October 2021, Wise and Remitly market capitalizations hovered just above USD8 billion and USD5 billion respectively, reflecting a fall of around 25% since their IPO.

### RPW-based presence and growth of remtechs
- Entries and corridor presence (Table 1 summary):
  - 2011Q1–2016Q1: Banks 26.8%, MTO 69.2%, PO 1.4%, Remtech 3.3% (Total entries).
  - 2016Q2–2020Q4: Banks 18.6%, MTO 78.2%, PO 1.3%, Remtech 11.1% (Total entries).
  - Corridors coverage:
    - 2011Q1–2016Q1: Banks present in 66.5% of corridors; MTO in 99.7%; PO 14.2%; Remtech 38.4%.
    - 2016Q2–2020Q4: Banks 68.4%; MTO 96.7%; PO 11.4%; Remtech 68.8%.
- Interpretation:
  - Remtechs grew from around 3 percent of RPW entries in the first half of the sample to over 10 percent in the second half.
  - Geographic coverage of remtechs expanded markedly: in the second half remtechs were present in almost seventy percent of corridors (a percentage similar to banks).
  - The increase in remtech participation occurred largely at the expense of banks’ share of entries, while banks’ geographical presence slightly increased.

### Mobile-related remittances and digital measures
- RPW tracking changes:
  - RPW started tracking costs associated with digital channels since December 2020.
  - RPW tracked remittance costs associated with Mobile Operators since June 2016.
  - Since March 2021 RPW tracks a “digital-only MTO index” composed by five fintechs: InstaRem, Remitly, Wise, WorldRemit and Xoom (noting the digital qualifier refers only to the sending channel).
- Table 2 summarized selected characteristics (2013Q3–2016Q1 vs 2016Q2–2020Q4):
  - Total entries: Digital 10.7% (2013Q3–2016Q1) and 16.9% (2016Q2–2020Q4).
  - Mobile Operator: 0.1% and 0.6%.
  - Mobile Money: not surveyed in the earlier window; 0.8% in 2016Q2–2020Q4.
  - Mobile Wallet: not surveyed in earlier window; 1.9% in 2016Q2–2020Q4.
  - Corridors: Digital 44.4% (2013Q3–2016Q1) and 72.8% (2016Q2–2020Q4).
  - Corridors: Mobile Operator 4.8% and 8.2%; Mobile Money - and 11.7%; Mobile Wallet - and 31.9%.
- Evidence on mobile-related remittances:
  - Mobile money operators (MMO) meaningfully operate in only a few corridors.
  - MMOs are absent or negligible in Latin America (LA) and Europe and Central Asia (ECA).
  - Mobile-related remittances are present in a very limited number of corridors and are absent in LA and ECA in RPW coverage (i.e., not significant enough to enter RPW database).

### Remittance costs — aggregate trends and regional patterns
- Aggregate trend:
  - Global remittance costs have been falling on average from 2011Q1 to 2020Q4, but remain well above the SDG target of 3 percent.
  - After a large decrease until 2015, the reduction in costs has slowed markedly.
- Regional changes (2020 vs 2015 averages):
  - SSA and EAP (the two most expensive regions) experienced largest decreases but amounted to around 0.75 percentage point.
  - ECA and LAC experienced average decreases around 0.25 percentage point.
  - SAR and MENA experienced increases in costs of around 0.40 and 0.80 percentage points respectively (MENA increase driven sharply by Syria; excluding Syria, average costs in MENA would have decreased by around 0.4 percentage point).
- Distributional insights:
  - Among the 105 surveyed remittance-receiving countries, around 20 percent faced costs at least as high as 10 percent in 2020.
  - Among the 365 surveyed corridors, around 15 percent faced costs around 10 percent or higher, while 5 percent faced costs at least as high as 15 percent in 2020.
  - 4 out of the 5 most expensive remittance-receiving countries are in SSA; 65% of the 20 most expensive remittance-receiving countries and corridors are in SSA and between SSA countries.
  - Among the 20 least costly countries, presence of El Salvador, Panama and Ecuador (dollarized economies) noted as reducing FX risks.
  - Among the least costly corridors, 6 involve India (4 as destination, 2 as source) and 3 have Bangladesh as destination.

### Fintech presence and effects on remittance costs
- High-level observation:
  - The surge in remtech corridor presence coincided with a slowdown in the decline of global remittance costs.
  - Figure 6B evidence: banks are the most expensive RSP type, remtechs the cheapest.
- Five-year (2015 to 2020) changes by RSP type:
  - MTO and banks’ average costs decreased by a similar, small magnitude: 0.8 p.p. vs 0.9 p.p.
  - Remtechs’ costs decreased marginally by 0.3 p.p. over the same period.
  - The bank-MTO cost gap remained largely unchanged globally, undermining a straightforward disruptive-fintech narrative.
- Interpretation:
  - Despite remtech growth and digital channel expansion, remtechs’ disruptive power on costs appears limited in the aggregate; cost dynamics diverged across regions.

### Bank–MTO cost gap: regional and corridor-level evidence
- Table 3: Bank and MTO averages and Bank–MTO gap (regional averages)
  - 2015 averages:
    - Bank: EAP 11.8%, ECA 9.2%, LAC 7.3%, MENA 9.3%, SAR 7.1%, SSA 14.9%
    - MTO: EAP 7.2%, ECA 6.8%, LAC 6.6%, MENA 6.5%, SAR 5.4%, SSA 8.9%
    - Bank–MTO Gap (2015): EAP 4.6%, ECA 2.1%, LAC 0.9%, MENA 2.1%, SAR 2.1%, SSA 5.1%
  - 2020 averages:
    - Bank: EAP 10.8%, ECA 8.5%, LAC 6.3%, MENA 7.6%, SAR 6.6%, SSA 13.6%
    - MTO: EAP 6.0%, ECA 5.5%, LAC 6.0%, MENA 6.6%, SAR 4.3%, SSA 7.8%
    - Bank–MTO Gap (2020): EAP 4.9%, ECA 2.3%, LAC 0.6%, MENA 1.2%, SAR 2.5%, SSA 5.9%
  - From 2015 to 2020 the Bank–MTO gap increased in four out of six regions (EAP, ECA, SAR, SSA), decreased in two regions (LAC and MENA).
  - Largest gaps in both 2015 and 2020 were in SSA.
- Corridor-level clustering by remtech presence (Figure 7):
  - 2015: corridors with remtech presence had average Bank–MTO gap 3.7% versus 2.9% for corridors without remtech presence.
  - 2020: corridors with remtech presence had average Bank–MTO gap 3.3% versus 3.6% for corridors without remtech presence.
  - Differences in both years are not statistically significant; endogeneity is a possible explanation (remtechs may have entered higher-cost corridors first).
- Table 4: Average remittance cost and remtech presence (selected rows)
  - 2015 (sample of 172 corridors, 72 with remtech presence):
    - Remtech corridors: Bank 10.5%, MTO 6.8%, Digital 6.3% (Mobile Money not surveyed).
    - No remtech corridors: Bank 10.4%, MTO 7.4%, Digital 8.2%.
  - 2020 (sample of 239 corridors, 164 with remtech presence):
    - Remtech corridors: Bank 9.3%, MTO 6.0%, Digital 5.2%, Mobile Money 4.3%.
    - No remtech corridors: Bank 9.4%, MTO 5.8%, Digital 7.4%, Mobile Money 4.1%.
- Notes and data issues:
  - Remtech cost series show suspicious sharp changes in 2014.Q2 (sharp decrease) and 2018.Q4 (large increase) linked to unusual large variations in a few corridors; initial observations for some newly added corridors in 2014.Q1 were sharply at odds with subsequent data points.
  - RPW methodological notes: digital remittances series pre-2016.Q2 were derived using simplifying assumptions and may behave non-intuitively; RPW does not allow identification of mobile operators’ entries; better documentation/support needed.

*Italic: Source: IMF Working Paper (Annex 1 material extracted from wpiea2022233-print-pdf).*

### 2015. In 2020, 25 corridors also included mobile money remittances. Of those, 14 were in corridors

### wpiea2022233-print-pdf - 2015. In 2020, 25 corridors also included mobile money remittances. Of those, 14 were in corridors

### Evidence on costs, remtech presence, and digital/remittances channels
- 2020: "25 corridors also included mobile money remittances. Of those, 14 were in corridors with remtech presence."
- When bank or MTO costs are not surveyed in all quarters of the year for each corridor the average uses the available data. Averages are calculated over individual entries.
- Findings on average costs (2015 vs. 2020):
  - Banks: both in 2015 and in 2020 banks’ average costs were the same in corridors with or without remtechs presence, even though average costs fell.
  - MTOs: in 2015 average MTO costs were lower in corridors with remtechs; in 2020 they stood slightly above (unexpected).
  - Digital remittances:
    - 2015: digital remittances in corridors with remtechs cost, on average, "1.8p.p. lower" than in those without remtech presence.
    - 2020: the gap increased; digital remittances in corridors with remtech presence cost consistently less than in corridors without remtech presence.
- Possible explanations for lower digital remittance costs where remtechs are present:
  - Aggregation effects as the number of corridors with remtechs increased sharply from 2015 to 2020 and digital remittances via remtechs are cheaper than via banks and MTO.
  - Digital remittances via traditional RSP (remittance service providers) may be costlier because they cross-subsidize other services (e.g., cash), which entail extra operational costs.
  - Endogeneity: corridors without remtech presence are probably more costly.
- Mobile remittances appear to be the cheapest way to remit; related factors:
  - Mobile money is a textbook example of leapfrog technology and a cheap way to attract customers to mobile operators’ core business.
  - Cheapest particularly when the same operator is present at both ends of the transaction.
- Explanatory factors for slowdown in decline of remittance costs and lack of convergence between banks and MTO:
  - Complexity of remittance market with interacting factors.
  - Stricter AML/CFT regulation caused correspondent banks to leave some corridors, decreasing competition and making compliance more costly.
  - Potential limits to how much costs can fall given current conditions.
  - Expansion of services offered by remtechs may have played a part.
- Central issue highlighted: remtechs and fintech have not delivered the disruption many expected in the remittance market.

### Why Are There No Signs of Disruption?
- Overall claim: While remtechs and mobile money remittances are, on average, cheaper than traditional RSP, there are no signs that they have disrupted the remittance market.

#### The Bitcoin/Blockchain narrative and evidence
- Predicted advantage: use bitcoin to bypass middlemen and remit easily, instantly, and at negligible cost.
- Empirical/technical counterpoints:
  - Technical limitations: high energy intensity for mining; blockchain congestion as transactions grew.
  - Transaction times: sometimes exceeded two days during congestion.
  - Transaction fees:
    - Fees have shown great volatility and have reached levels "above USD60 per transaction" during high market activity.
    - Average costs in the second half of 2021 were "around USD3.0".
    - Since 2015, costs have been above "USD5.0" fifteen percent of the time — implying in a USD200 remittance transaction at least a "2.5 percent" transaction cost.
  - Additional costs not captured by network fees:
    - Receivers must convert bitcoins to domestic currency, incurring additional fees from crypto exchanges and BATMs.
    - Coinbase: "includes a spread in the price when you buy or sell cryptocurrencies or in the exchange rate when you convert cryptocurrencies."
    - Bitcoin ATMs (BATM) fees are usually "five percent or higher".
  - BATM distribution is heavily skewed toward richest regions and mostly in the US; poorest regions such as Africa have a negligible number of BATM.
  - Financial literacy and complexity: public/private keys, wallet addresses, and technical risks deter uptake.
  - Conclusion: bitcoin is "not economically viable" as a mainstream remittance channel; bitcoin/blockchain-based remittances are "virtually nonexistent."
- Case of El Salvador:
  - Late 2021: El Salvador made bitcoin legal tender to decrease remittance costs; claim of potential savings of "USD400 million" by the president.
  - Counter-evidence: in 2020 remittance costs to El Salvador were "just above 3 percent", making it the lowest cost country among 99 remittance-receiver countries (Figure 5D).

#### The Mobile Money experience
- Mobile money as genuine fintech innovation: enables digital financial transactions without a bank account.
- M-Pesa case study:
  - Created in 2007; frontend technology centers around simple text messaging and simple cell phones.
  - Initial purpose: facilitate repayment of microloans for the poorest population; evolved into a financial inclusion tool.
  - Impact on Kenya:
    - 2014: 55% of Kenyans had accounts at financial institutions (up from 42% in 2011); 58% had mobile money accounts.
    - 2017: 56% had accounts at financial institutions (stagnated); 73% had mobile money accounts.
    - "Inclusion gap" widened from "3p.p." to "17p.p." (difference between inclusion via mobile money and via banking system).
  - M-Pesa customer base: "51 million customers across seven African countries"; M-Pesa network has "more than 600,000 active agents."
- Evidence of mobile money benefits in the literature:
  - Mobile money helps smooth consumption (Jack and Suri, 2014), increases household consumption (Munyegera and Matsumoto, 2016), decreases probability of extreme poverty (Suri and Jack, 2016), decreases share of households with very low food security (Wieser et al., 2019), and increases health care access (Ahmed and Cowan, 2019).
- Global patterns and limitations:
  - 2017: only around one third of the SSA population age 15+ had bank accounts at a financial institution.
  - 2017: SSA hosted around "45 percent" of global registered mobile money accounts.
  - Mobile money prevalence is uneven: in LAC and MENA mobile money is still uncommon despite low financial inclusion; GSMA (2021) found "73 percent" of countries analyzed in the LAC region had a Mobile Money Prevalence Index rated as low or very low in 2021.
  - In SSA, around one third of countries were rated as low or very low in the Mobile Money Prevalence Index in 2021.
- Mobile money remittances remain marginal:
  - Mobile money remittances amounted to "just over 2 percent" of the total value of mobile money transactions carried out in 2021.
  - Of that amount, "84 percent" took place in SSA.
  - Domestic transfers made the bulk of mobile money transactions in 2021.
- Conclusion: mobile money is a successful financial inclusion instrument in some contexts (e.g., Kenya) but has not scaled into a dominant mode for international remittances globally.

#### The Remtech hype (introductory observation within provided text)
- The text identifies Wise (formerly Transferwise) as the poster child of disruptive remtechs, but the document proceeds to analyze why remtechs have not produced the expected market-wide disruption.

*International Monetary Fund — IMF Working Papers: Curb Your Enthusiasm: The Fintech Hype Meets Reality in the Remittances Market*

### 2011. Its mission reads “Money without borders –   instant, convenient, transparent and eventually free.”

### wpiea2022233-print-pdf - 2011. Its mission reads “Money without borders –   instant, convenient, transparent and eventually free.”

### Wise / remtech narrative and claims
- Wise’s mission: “Money without borders –   instant, convenient, transparent and eventually free.”
- Core narrative elements:
  - Emphasis on hidden fees represented by typical FX margins; argued the only fair exchange rate is the mid-market rate.
  - Claim of superior, more efficient business model based on matching flows (peer-to-peer), reducing need for money to cross borders and thereby lowering cost and time of transactions.
- Critical observation: The email analogy (“How is it that it’s free to send an email and it costs  to send money?”) is problematic because sending money internationally is a much more complex task than sending emails.
- The analysis uses Wise’s narrative as a proxy for remtechs’ broader narrative, while noting differences across business models.

### Empirical evidence on FX margins (RPW database)
- Table 5: FX Margins by RSP Type and Region (2011.Q1 – 2016Q1 and 2016.Q2 – 2020Q4)
  - Regions: ECA, EAP, LAC, MENA, SAR, SSA
  - 2011.Q1 – 2016Q1 Period (FX margins)
    - Bank: 0.4% (ECA), 2.5% (EAP), 1.2% (LAC), 1.0% (MENA), 1.7% (SAR), 1.5% (SSA)
    - MTO: 1.4% (ECA), 1.9% (EAP), 1.9% (LAC), 2.3% (MENA), 1.5% (SAR), 2.4% (SSA)
    - Remtech: 1.0% (ECA), 1.7% (EAP), 1.6% (LAC), 2.7% (MENA), 1.3% (SAR), 2.5% (SSA)
  - 2016.Q2 – 2020Q4 Period (FX margins)
    - Bank: 0.5% (ECA), 2.5% (EAP), 1.2% (LAC), 1.8% (MENA), 1.5% (SAR), 2.2% (SSA)
    - MTO: 1.8% (ECA), 2.1% (EAP), 2.0% (LAC), 2.5% (MENA), 1.5% (SAR), 2.7% (SSA)
    - Remtech: 1.4% (ECA), 1.7% (EAP), 1.5% (LAC), 2.2% (MENA), 1.3% (SAR), 2.5% (SSA)
  - Change Between Periods (absolute)
    - Bank: 0.2p.p. (ECA), 0.1p.p. (EAP), 0.1p.p. (LAC), 0.8p.p. (MENA), -0.2p.p. (SAR), 0.7p.p. (SSA)
    - MTO: 0.4p.p. (ECA), 0.2p.p. (EAP), 0.1p.p. (LAC), 0.2p.p. (MENA), 0.0p.p. (SAR), 0.3p.p. (SSA)
    - Remtechs: 0.3p.p. (ECA), 0.0p.p. (EAP), -0.1p.p. (LAC), -0.5p.p. (MENA), 0.0p.p. (SAR), 0.0p.p. (SSA)

- Key empirical findings from Table 5 and descriptive text:
  - Contrary to expectations from remtech narrative, banks practice the lowest margins in general, having the lowest margins in four out of six regions across the two periods.
  - Remtechs have the lowest margins in the two remaining regions.
  - In SSA (the highest cost region), remtechs had the highest margins in the first period and were costlier than banks in the second period.
  - The observed pattern raises questions about why remtechs do not uniformly show much lower FX margins.

### Measurement issues: zero and negative FX margins, estimated FX bias
- Table 6: Percentage of FX Margins Equal to Zero or Negative (2011.Q1 – 2020Q4)
  - Zero: Banks 30%, MTO 17%, Remtechs 14%
  - Negative: Banks 3%, MTO 4%, Remtechs 8%
  - Estimated FX Bias: 0.9p.p. (Banks), 0.6p.p. (MTO), 0.6p.p. (Remtechs)
- Notes and implications:
  - The RPW survey shows many entries with zero or negative FX margins, which is not credible and indicates mismeasurement.
  - Zero/negative entries affect as much as one third of bank entries and around 20% of MTO and remtech entries.
  - The estimated FX bias is calculated by replacing zero and negative margins with the average margin for that quarter net of the zero/negative entries.
  - The estimated bias is sizable: 0.9 p.p. for banks and 0.6p.p. for MTO and remtechs.

### Adjusted FX margins after correcting bias
- Table 7: Adjusted FX Margins by RSP Type and Region (after accounting for estimated FX bias)
  - 2011.Q1 – 2016Q1 Period (Adjusted margins)
    - Bank: 2.6% (ECA), 3.1% (EAP), 2.7% (LAC), 2.7% (MENA), 2.4% (SAR), 2.7% (SSA)
    - MTO: 2.8% (ECA), 2.2% (EAP), 2.9% (LAC), 2.5% (MENA), 1.7% (SAR), 3.2% (SSA)
    - Remtech: 1.8% (ECA), 1.9% (EAP), 2.6% (LAC), 2.8% (MENA), 1.5% (SAR), 2.9% (SSA)
  - 2016.Q2 – 2020Q4 Period (Adjusted margins)
    - Bank: 2.6% (ECA), 2.8% (EAP), 2.7% (LAC), 2.8% (MENA), 2.1% (SAR), 2.9% (SSA)
    - MTO: 3.0% (ECA), 2.4% (EAP), 2.8% (LAC), 2.8% (MENA), 1.7% (SAR), 3.3% (SSA)
    - Remtech: 2.3% (ECA), 2.3% (EAP), 2.3% (LAC), 2.4% (MENA), 1.6% (SAR), 2.9% (SSA)
  - Change Between Periods (Adjusted)
    - Bank: -0.1p.p. (ECA), -0.3p.p. (EAP), 0.0p.p. (LAC), 0.1p.p. (MENA), -0.3p.p. (SAR), 0.2p.p. (SSA)
    - MTO: 0.2p.p. (ECA), 0.2p.p. (EAP), -0.1p.p. (LAC), 0.3p.p. (MENA), 0.0p.p. (SAR), 0.2p.p. (SSA)
    - Remtechs: 0.4p.p. (ECA), 0.3p.p. (EAP), -0.3p.p. (LAC), -0.3p.p. (MENA), 0.1p.p. (SAR), 0.0p.p. (SSA)
  - Difference: Remtech vs. Second Lowest Margin
    - 2011.Q1 – 2016Q1: -0.8p.p. (ECA), -0.2p.p. (EAP), -0.1p.p. (LAC), 0.2p.p. (MENA), -0.2p.p. (SAR), 0.2p.p. (SSA)
    - 2016.Q2 – 2020Q4: -0.3p.p. (ECA), -0.1p.p. (EAP), -0.4p.p. (LAC), -0.4p.p. (MENA), -0.1p.p. (SAR), 0.0p.p. (SSA)
    - Full Sample: -0.4p.p. (ECA), -0.1p.p. (EAP), -0.3p.p. (LAC), -0.3p.p. (MENA), -0.1p.p. (SAR), 0.1p.p. (SSA)

- Interpretation and implications of adjusted margins:
  - Correcting for FX bias improves remtechs’ standing: remtechs have the lowest margins in four out of six regions in the first period and in all regions in the second period (tied with banks in SSA).
  - However, the margin differences are often small relative to the second-lowest margin.
  - In SSA, remtechs’ average FX margin is either the same or higher than that of banks, disappointing given SSA’s status as the most expensive corridor.
  - Remtechs’ FX margins increased between periods in ECA, EAP and SSA, despite being the lowest in both periods in some regions.
  - Adjusting for bias also implies that the RPW database underestimates remittance costs, meaning the SDG target is further away than suggested by unadjusted figures.

### Overall takeaways
- The remtech narrative—promising much lower costs and greater transparency—finds only partial empirical support once measurement issues are addressed.
- Measurement problems in the RPW survey (zero/negative FX margins) materially affect conclusions and must be corrected to assess true FX margins.
- Even after bias correction, remtech advantages are regionally uneven and modest in magnitude in many cases; in SSA remtechs do not clearly deliver the largest savings.
- The RPW survey likely underestimates remittance costs; correcting bias increases estimated margins and extends the distance to the SDG cost-reduction target.

*Source: wpiea2022233-print-pdf - 2011. Its mission reads “Money without borders –   instant, convenient, transparent and eventually free.”*

### Section 4 unveiled two key pieces of evidence that were not supportive of the disruptive fintech narrative. First,

### Section 4–7 synthesis: Evidence on fintech and remittances

### Evidence inconsistent with a disruptive remtech narrative
- Two prima facie findings:
  - The decline in remittance costs started to slow down precisely when the number of corridors with remtech presence began to surge (post-2015 timing emphasized).
  - The bank–MTO gap – the average difference in costs between banks and MTO – remained largely stable from 2015 to 2020.
- Conclusion: prima facie evidence was not supportive of a disruptive remtech narrative.

### Bitcoin/blockchain: expectations versus reality
- Core reasons why the bitcoin/blockchain disruptive narrative failed:
  - Bitcoin transaction fees are usually volatile and can reach very high levels due to “blockchain congestion”, which would become chronic if bitcoin were used widely for remittances.
  - Most remittance participants are unbanked and many transactions involve cash (mainly at the receiving end), making bitcoin infeasible as a vehicle currency.
  - Narratives ignored full lifecycle costs: network-related fees, third-party fees, conversion back to cash, and BATM fees (bitcoin ATM fees) which are typically high; many locations lack bitcoin ATMs.
  - Cryptocurrencies’ complexity and risks reduce convenience and appeal, even for the financially literate.
- Empirical outcome:
  - Use of bitcoin to send remittances has been virtually non-existent and is unlikely to change materially.
  - Many “blockchain-based remittances” start-ups cited as notable examples proved to be ordinary crypto exchanges, closed, or pivoted away from bitcoin-based remittances.

### Mobile money and M-Pesa: successes, limits, and replication challenges
- Mobile money can advance financial inclusion and has clear domestic benefits (e.g., Kenya’s M-Pesa).
- Key contextual drivers of M-Pesa’s success in Kenya:
  - Safaricom’s very large market share: 65% in 2007; 64 percent in 2021.
  - Large physical agent network: in 2021, more than 28 million active M-Pesa users and almost 250 thousand M-Pesa agents in Kenya.
  - By contrast, DRC, Egypt, Ghana, Lesotho, Mozambique, and Tanzania had around 18 million active M-Pesa users and 300 thousand agents combined, despite a combined population almost six-fold that of Kenya’s.
  - In 2020, M-Pesa accounted for 98% of the mobile money market in Kenya.
  - Enabling regulatory approach (Central Bank of Kenya’s light touch, trust-account requirement) and specific historical events (e.g., 2008 election violence) aided rapid uptake.
- Replication barriers and limitations:
  - Mobile money diffusion is constrained by infrastructure, regulation (e.g., bank-led regulatory models that required bank partnership in Ghana until 2015), interoperability challenges, and market structure (e.g., preexisting banking access in South Africa).
  - Mobile money’s role in international remittances has been limited despite being typically less costly.
  - Mobile money’s success often depends on large non-digital footprints (agent networks), undermining the simple “digital disruption” narrative.

### Remtechs: business models, market entry patterns, and partnerships
- Nature of remtechs:
  - Remtechs are largely business-model innovations (digital channels, pricing transparency, price comparison), not necessarily groundbreaking financial-technology innovations.
  - Their predominantly digital business model is simultaneously a strength (convenience, lower non-pecuniary costs) and a constraint (limited ability to handle cash transactions).
- Market-entry and growth patterns:
  - Many remtechs launched in more developed, competitive, and lower-cost corridors (e.g., Remitly debuted in US–Philippines; took four years to enter US–Mexico and US–India).
  - From 2011.Q1 to 2020.Q4 Remitly reached at least 1 percent market share in only four SSA corridors.
  - Nine years after starting, TransferGo had no meaningful presence in SSA; currently, according to its website, it can remit to only three SSA countries.
  - Wise reached the 1 percent threshold in only five SSA corridors during 2011.Q1–2020.Q4; currently, according to its website, it can remit to only seven SSA countries.
  - Xoom was meaningfully present in only five SSA countries during the period above.
  - With one exception (WorldRemit), remtechs are generally not meaningfully present in the 20 most expensive corridors.
- Strategic pivots and partnerships:
  - To scale, remtechs broadened services to include cash pick-ups and home delivery, often by partnering with banks, traditional MTO, mobile money operators, and agents—the incumbents they initially sought to disrupt.
  - Examples: WorldRemit partnerships with banks and mobile operators; Remitly and Xoom partnerships with Ria to offer cash pick-ups; Wise partnering with France’s second largest bank and integrating with Brazil’s PIX.
  - The peer-to-peer (“matching flows”) model touted by Wise proved unsustainable as remtechs scaled into “lopsided” corridors; peer-to-peer matching did not scale across many rich–poor corridors.
  - Some remtechs moved into corporate markets or underwent consolidation: Xoom was acquired by PayPal in 2015; Azimo was acquired by Papaya Global in a deal valued between USD150 and USD200 million (previous valuation around USD90 million).

### Positive roles of fintech despite limited disruption
- Areas where fintech has materially helped:
  - Increased competition: costs are strongly negatively correlated with the number of MTO and the ratio of MTO to banks at the corridor level.
  - Greater pricing transparency: remtechs raised awareness about hidden FX margins, improving end-user decision-making.
  - Easier price comparison and predominantly digital models: enable more efficient competition and push incumbents toward digitalization.
  - Digitalization lowers non-pecuniary costs (time, transportation) and increases safety by reducing the need to carry cash.
  - Mobile-money remittances are usually cheaper; on average remtechs charge less than incumbents, partly due to service type differences.

### Conclusions and policy implications
- Overall assessment:
  - There is a large gap between fintech’s hyped disruptive effects and the empirical reality in the remittances market.
  - Structural factors—especially low financial inclusion and the predominance of cash in remittance flows—severely limit digital disruption.
  - Fintech is unlikely to disrupt the remittance market in the foreseeable future given persistent frictions (financial inclusion, infrastructure, interoperability, network effects, regulatory constraints).
- Policy recommendations and considerations:
  - Policymakers should be skeptical of portraying fintech as an easy solution to deep-rooted problems like expensive financial services and low financial inclusion.
  - Clear, unambiguous policy measures include providing a level playing field for fintech and smaller MTO, addressing anticompetitive practices (e.g., exclusivity clauses), and adopting proportionate, risk-based regulation.
  - For mobile money and interoperability, policy design must balance fostering innovation and fostering competition, acknowledging infrastructure costs and profitability challenges.
  - Governments can play an important role by building/upgrading financial infrastructures, setting rules conducive to infrastructure sharing and compatibility, preventing regulatory capture, and being more open to experimentation.

*IMF Working Paper excerpt: Section 4–7 content on fintech, remittances, bitcoin, mobile money, and remtechs.*

### References

### References

### Key literature and empirical studies
- Acharya, C. and R. Leon-Gonzalez (2014), “How do Migration and Remittances Affect Human Capital Investment? The Effects of Relaxing Information and Liquidity Constraints”, Journal of Development Studies, Volume 50, Issue 3.
- Adams, R. (2011), “Evaluating the Economic Impact of International Remittances on Developing Countries Using Household Surveys: A Literature Review”, Journal of Development Studies 47(6).
- Aggarwal, R., A. Demirgüç-Kunt, M. S. M. Pería (2011), “Do remittances promote financial development”, Journal of Development Economics, Volume 96, Issue 2.
- Ahmed, H. & B. W. Cowan (2019), “Mobile Money and Healthcare Use: Evidence from East Africa”, NBER Working Paper.
- Aker, Jenny C, S. Prina and C. J. Welch (2020), “Migration, Money Transfers, and Mobile Money: Evidence from Niger”, AEA Papers and Proceedings, Volume 110.
- Aron, Janine (2018), “Mobile Money and the Economy: A Review of the Evidence”. The World Bank Research Observer., Volume 33., No 2.
- Beck, Thorsten, M. Janfils and K. R. Kpodar (2022), “What Explains Remittance Fees? Panel Evidence”, IMF Working Paper 22/63.
- Bettin, G., A. F. Presbitero and N. Spatafora (2014), “Remittances and Vulnerability in Developing Countries”, IMF Working Paper 14/13.
- CPSS and World Bank (2007). “General Principles for International Remittance Services”.
- Committee on Payments and Market Infrastructures (CPMI) (2020), “Enhancing cross-border payments: building blocks of a global roadmap. Stage 2 report to the G20”, Bank of International Settlements.
- Jack, William and Tavneet Suri (2014), “Risk Sharing and Transaction Costs: Evidence from Kenya’s Mobile Money Revolution”, American Economic Review, Volume 104, Number 1.
- Suri T. and W. Jack (2016), “The long-run poverty and gender impacts of mobile money”, Science, Volume 354, Issue 6317.
- Wieser, Christina, M. Bruhn, J. P. Kinzinger, C. S. Ruckteschler and S. Heitmann (2019), “The Impact of Mobile Money on Poor Rural Households: Experimental Evidence from Uganda”. World Bank Policy Research Working Paper No. 8913.

### Thematic coverage in the cited literature
- Effects of remittances on household behavior and human capital:
  - Acharya and Leon-Gonzalez (2014); Terrelonge (2014); Askarov and Doucouliago (2020); Azizi (2019).
- Remittances and financial development, costs, and intermediation:
  - Aggarwal et al. (2011); Bhattacharya et al. (2018); Beck et al. (2022); da Silva Filho (2021); Watkins and Quattri (2014).
- Mobile money, branchless banking, and regulations:
  - Bank of Ghana (2008); Bank of Ghana (2015); Central Bank of Nigeria (2018); GSMA (2018a, 2018b, 2021); Reynolds et al. (2018); Aron (2018); Munyegera and Matsumoto (2016).
- Fintech, blockchain, and remittance innovations:
  - Flore (2018); McKinsey & Company (2016); Schueffel (2016); Bersch et al. (2021).
- Remittances in crisis, natural disasters, and vulnerability:
  - Mohapatra et al. (2012); Bettin et al. (2014); Donou-Adonsou et al. (2020).

### Appendix 1 — A Short History of Remittance Fintechs (summary of facts and milestones)
- Xoom:
  - Created in 2001.
  - Business model converged to a digital-only model in 2007.
  - Acquired by PayPal in November 2015.
  - Currently sends money to almost 160 countries.
- WorldRemit:
  - Founded in 2010 by a Somali refugee studying in London.
  - Early distinct feature: possibility of sending money directly to mobile wallets.
  - Early focus: Africa.
  - Currently says it can send money from over 50 countries to more than 130 countries.
- Remitly (founded as Beamit):
  - Founded in the US in 2011; service officially launched in August 2012.
  - Early investors included Eric Schmidt and Jeff Bezos.
  - Initially worked only in the US–Philippines corridor.
  - Added US–India and US–Mexico corridors in 2015.
  - Began adding non-US outbound corridors in 2016; more aggressive global expansion from 2017.
  - Currently says it supports transfers from 21 sending countries to more than 100 receiving countries.
- Wise (formerly Transferwise):
  - Founded in 2012 in the UK.
  - Promised sharply reducing remittances costs using a “peer-to-peer” approach and minimizing FX margin.
  - Peer-to-peer model did not prove sustainable due to loopsided flows.
  - Recently opened capital with a valuation close to £9 billion.
  - Currently says it can send money to 80 countries.
- Azimo:
  - Founded in 2012 in the UK.
  - Eurocentric remtech: only 3 out of the 28 sending countries outside Europe.
  - Says it can send remittances to more than 200 countries and territories.
  - Recently acquired by Papaya Global.
- TransferGo:
  - Founded in 2012; eurocentric with all sending countries in Europe.
  - Currently claims it can send money to 161 countries.
- Finablr and related developments:
  - Finablr founded in 2018 as “a global platform for Payments and Foreign Exchange solutions underpinned by modern and proprietary technology”.
  - Owned subsidiaries such as UAE Exchange, Remit2India, Unimoni, Xpress Money and BayanPay.
  - Remit2India: founded in 2001, described as “a pioneer in online remittances”; operated in four corridors with one destination: India.
  - May 2020: Remit2India interrupted its services (website non-operational for transactions originating from Australia, Canada, the UK and the USA as of 13 May 2022).
  - August 2021: Finablr was bought for one dollar and re-branded as Wizz Financial.

### Appendix 2 — Measurement error example in the RPW Database
- Figure A2.1 illustrates “Measurement Errors: Unusual Cost Variations in Selected Remittance Corridors”.
- Source: Remittance Prices Worldwide Database.
- The figure presents quarterly time series across corridors spanning 2014.Q1 through 2020.Q3 and shows cost variation scales up to 40 in selected charts (labels shown: 0, 2, 4, 6, 8, 10, 12, 14 and 0, 5, 10, 15, 20, 25, 30, 35, 40).

### Appendix 3 — A radiography of remtech’s diffusion (survey coverage notes)
- Appendix 3 maps remtech survey coverage by corridor and quarter from 2011.Q1 through 2020.Q4.
- Footnote: “Cells with zeroes indicate quarters in which remtechs were surveyed in the indicated corridor.”
- The appendix lists extensive corridor codes and quarterly markers (examples include corridors and quarterly labels such as 2011.Q1, 2011.Q3, 2012.Q1, 2012.Q3, 2013.Q1, 2013.Q2, 2013.Q3, 2013.Q4, 2014.Q1, 2014.Q2, 2014.Q3, 2014.Q4, 2015.Q1, 2015.Q2, 2015.Q3, 2015.Q4, 2016.Q1, 2016.Q2, 2016.Q3, 2016.Q4, 2017.Q1, 2017.Q2, 2017.Q3, 2017.Q4, 2018.Q1, 2018.Q2, 2018.Q3, 2018.Q4, 2019.Q1, 2019.Q2, 2019.Q3, 2019.Q4, 2020.Q1, 2020.Q2, 2020.Q3, 2020.Q4).

*Italic: References and appendices extracted from the PDF chapter "wpiea2022233-print-pdf - References".*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022233-print-pdf.pdf_