37 Fintech, mobile money and payments

Chapter 37 of 60 in the Bangladesh 2036 research base. Contents of the series.

The rail won its decade and now has its law; the decade ahead is decided by whether payments become finance or stay a pipe

Chapter 07 carries the aggregate record and names the rail regulation decision as one of its four; this chapter opens the machine that decision runs on. The scale is settled: registered mobile financial services accounts grew from 125,506 in 2011 to 238.68 million at end 2024, 88.89 million of them active, served by 822,726 agent outlets against 19,829 automated teller machines [IMF FAS 2024]. The fiscal year print is still rising: mobile financial services transactions reached 1,817,408.1 crore BDT in FY25, up 18.3 percent on the FY24 total of 1,536,170.0 crore on chapter arithmetic from the same table [BB Econ 2025, Table IIG]. The thesis of this chapter is that the 2026 to 2036 window the chapter 15 scenarios price is decided by conversion, not construction: the Payment and Settlement System Act 2024 gave the central bank its first statutory footing, the interoperable platform and the licence categories exist on paper, and whether the 13 provider rail becomes a savings, credit and merchant float system or remains a cash in, cash out pipe with a rising fraud bill is decided by four rule books, interoperability enforcement, digital bank licensing, conduct supervision and the admissibility of payment ledgers as credit data, that Bangladesh Bank and the providers write in this window. Two measured facts discipline the claim: the rail holds 0.26 percent of GDP in balances against 34.7 percent of GDP in annual transactions [IMF FAS 2024], a pipe by design, and the central bank's own subscriber series broke by 40.3 percent in a single month of 2025 with no footnote [BB Econ 2025, Table IIG], so the decade's payments policy will run on a measurement system chapter 53 already distrusts.

The record: a thirteen provider rail moves a rising share of GDP and holds almost none of it

The account stock tells a construction story that ran in two bursts. Registered accounts grew steadily from 125,506 in 2011 to 99.34 million in 2020, then jumped 70.9 percent in 2021 to 169.72 million on chapter arithmetic, the pandemic year in which government assistance and garment wage payments moved onto wallets, and kept compounding to 191.06 million in 2022, 220.46 million in 2023 and 238.68 million in 2024 [IMF FAS 2024]. Activity lagged and then caught up: active accounts fell from 37.51 million in 2018 to 32.33 million in 2020 while registrations rose, the dormancy gap chapter 07 reads as one flow accounts, then tripled to 88.89 million by 2024, a 37.2 percent active share on chapter arithmetic [IMF FAS 2024]. Agent outlets grew from 2,551 in 2011 to 822,726 in 2024 [IMF FAS 2024], and the central bank's own fiscal year count of the network, 1.8 million agents and 232.4 million registered accounts at end June 2024 against 1.6 million and 207.3 million a year earlier, is larger than the survey count because it counts every sub agent point [BB AR 2024].

Value moved the same way with a fiscal year acceleration into the present. The Financial Stability Report's calendar series records transaction volume of 7,701 billion BDT in 2021, 10,159 billion in 2022, 13,529 billion in 2023 and 17,452 billion BDT in 2024 [BB FSR 2024], while the Financial Access Survey puts 2024 at 17,374,107 million BDT, 1,737,411 crore, 34.7 percent of GDP against 0.004 percent in 2011 [IMF FAS 2024]. On the fiscal year basis the central bank settled around 6,784 million transactions worth 15,416.35 billion BDT in FY24, up 22.5 percent in volume and 26.6 percent in value on FY23 [BB AR 2024], and 1,817,408.1 crore in FY25, up 18.3 percent on chapter arithmetic [BB Econ 2025, Table IIG]. The first two months of FY26 run at 14.89 percent year on year value growth, with August 2025 alone at 151,123.6 crore BDT [BB Econ 2025, Table IIG]. The usage mix explains both the scale and the ceiling: in June 2024, cash in was 30.4 percent of transaction volume, cash out 30.1 percent, up from 28.3 percent a year earlier, and person to person transfers 26.7 percent, while inward remittance was about 1 percent [BB AR 2024]. The rail is still, by its own measured mix, mostly a machine for converting cash to electronic value and back.

Reach outruns the formal system it sits beside. The 2022 census found 38.53 percent of households holding a mobile financial services account [BBS Census 2022]; a companion census figure for financial institution account holding carries no data points in the data lake and could not be confirmed here. Findex measures a fall in overall adult account ownership, from 52.8 percent in 2021 to 43.3 percent in 2024 [WB Findex 2024, account.t.d], a formal sector retreat that sits beside the mobile account count's continued rise and belongs on the same watch list as the fraud and measurement risks below. The card and machine estate deepened on the same schedule: 39.57 million debit cards and 2.67 million credit cards at end 2024, 19,829 automated teller machines at 15.87 per 100,000 adults [IMF FAS 2024], while bank branches per 100,000 adults peaked at 9.28 in 2022 and fell to 9.14 by 2024 [IMF FAS 2024].

The agent banking channel, the banks' answer to the same last mile, is the record's cautionary counterpoint, and the FY25 print deepens the turn chapter 07 dated. Outlets grew from 448 in 2015 to 21,601 in 2023, then fell for the first time to 21,248 in 2024 [IMF FAS 2024]. Deposits kept rising, from 156.02 billion BDT in 2020 to 417.59 billion BDT in 2024 [BB FSR 2024], but transaction value fell 5.7 percent in FY25 to 785,177.1 crore BDT from 832,532.9 crore in FY24 on chapter arithmetic [BB Econ 2025, Table IIG]. The bank owned channel stalled while the wallet channel grew 18.3 percent in the same year [BB Econ 2025, Table IIG], and internet banking transactions ran 41.90 percent ahead of the previous year in July and August 2025, at 111,667.1 crore BDT in August alone [BB Econ 2025, Table IIG]. Internet banking's own annual volume grew from 1,373.84 billion BDT in 2021 to 11,092.78 billion BDT in 2024, up 707.5 percent on chapter arithmetic [BB FSR 2024], the fastest compounding digital channel on this chapter's ledger even before the FY26 acceleration Table IIG records. Money is moving to rails owned outside the branch and the bank agent, faster than the banks' own digital estate, and the deposit franchise chapter 06 watches is being skimmed at the margin.

One measurement fact belongs in the record because the decade's policy will be set against it. The central bank's e-banking table reports mobile financial services subscribers of 240.47 million in February 2025 and 143.63 million in March 2025, a 40.3 percent single month drop on chapter arithmetic, with no footnote in the table [BB Econ 2025, Table IIG]. Either the providers deduplicated accounts or the definition changed; no source available to this chapter says which, the series before and after the break each continue smoothly, and the resolving source is the Bangladesh Bank Payment Systems Department. Chapter 53's argument about statistical credibility has a payments instance.

Mechanism: flows, agent economics and a regulatory design that stops the rail becoming a bank

Three flows built the rail, and each was a policy decision before it was a technology adoption. First, government to person transfers: the state moved social safety net payments onto wallets, 69.87 percent of beneficiaries surveyed in 2022 collected payments through mobile money agents [Harvard Dataverse 2022], and the central bank records the government's stated preference for mobile financial services in safety net delivery alongside the pandemic cash assistance programme that moved through four providers [BB AR 2024]. Second, remittance formalisation: the wallet channel carries part of an inflow last confirmed at 27.52 billion USD for WDI year 2024 [WB WDI 2026], the price mechanism chapter 21 documents; a calendar 2025 total could not be confirmed from the sources this chapter draws on. The central bank counts export oriented wage payments through bank or mobile wallet accounts among the flows the rail must carry [BB AR 2024]. Third, the agent commission itself: cash in and cash out at a margin made every village store a bank branch surrogate, which is why cash conversion and person to person transfer combined for 87.2 percent of measured usage volume in June 2024, cash in 30.4 percent plus cash out 30.1 percent plus person to person transfer 26.7 percent on chapter arithmetic [BB AR 2024].

The 2021 registration jump has a named driver of the same kind. The pandemic stimulus and wage digitisation moved enrolment faster than any marketing campaign, 99.34 million to 169.72 million registered accounts in one year [IMF FAS 2024], and the active count's later catch up to 88.89 million [IMF FAS 2024] tracks the flows recurring monthly rather than the accounts sitting open.

The design that keeps balances at 0.26 percent of GDP is the second half of the mechanism, and it is deliberate. The Mobile Financial Services Regulations 2022 and their predecessors hold customer funds in scheduled bank accounts rather than on provider books, the providers intermediate payments and not savings, and the result is a rail with the transaction scale of a payment system and the balance sheet of a pass through [BB AR 2024]. Bank deposits stand at 37.89 percent of GDP against the wallet float's 0.26 percent [IMF FAS 2024]. The same settlement architecture runs underneath the banks: the Bangladesh Automated Clearing House cleared 18,755 thousand regular value cheques worth 9,294.97 billion BDT in FY24, down 4.6 percent in volume and 7.7 percent in value, while the electronic funds transfer network settled 256,659 thousand credit transfers worth 7,253 billion BDT, up 11.75 percent and 30.25 percent, and real time gross settlement carried 9,763 thousand transactions worth 55,510.72 billion BDT [BB AR 2024]. The national payment switch moved 79,365.65 thousand automated teller machine transactions worth 682.6 billion BDT, 13,002.14 thousand point of sale transactions worth 47.4 billion BDT and 41,194.53 thousand interbank fund transfers worth 2,030.1 billion BDT in FY24, the last up 130.6 percent in value in one year, and the switch now routes BanglaQR merchant payments [BB AR 2024]. Paper is dying on a measured schedule and the whole electronic estate is compounding, digital payments up 21.5 percent in volume and 4.6 percent in value in FY24 [BB AR 2024].

Market structure is the part of the mechanism the machine cannot yet measure. Thirteen providers operated at end June 2024 [BB AR 2024], alongside 8 licensed payment service providers and 10 payment system operators, and the market is widely described as a bKash and Nagad duopoly with Rocket and Upay behind; the provider level shares and audited financials could not be confirmed here, the providers' audited accounts and Bangladesh Bank returns being the resolving sources. Employment in the channel is measured: about 1.8 million agents [BB AR 2024], a labour force chapter 08's informal economy count carries.

The decade ahead: an act, a switch and a licence class meet the conversion problem

The legal construction is new and datable. Before 2024 the rail ran on the Payment and Settlement Systems Regulations 2014, the Mobile Financial Services Regulations 2022 and circulars; the Payment and Settlement System Act 2024, gazetted with presidential assent on 4 July 2024, gave the central bank statutory power over integration, supervision and regulation of payment systems and customer protection [BB AR 2024]. Bangladesh Bank restructured its Payment Systems Department on 15 February 2024 into a separate Payment System Oversight Division with onsite and offsite supervision of banks, mobile financial services providers, payment service providers and payment system operators, and opened a Regulatory Fintech Facilitation Office as the sandbox front door [BB AR 2024]. Conduct instruments followed: the Guidelines for Merchants Acquiring and Escrow Services of 26 September 2023 and the Guidelines for Issuing Prepaid Instruments by Non payment System Entities of 6 June 2024 [BB AR 2024]. The framework, in short, was completed in the eighteen months before this chapter's baseline, and everything in the decade is execution of it.

Interoperability is the first execution test. The Interoperable Digital Transaction Platform, Binimoy, has a fee schedule the central bank set on 10 November 2022 [BB AR 2024], and a 2026 assessment by the state owned infrastructure facilitator still records that interoperability among mobile financial services remains absent [IIFC 2026]. The platform's transaction volumes could not be confirmed here, Bangladesh Bank's Payment Systems Department being the resolving source. The mechanism matters more than the date: without wallet to wallet settlement across providers, the rail is thirteen closed loops, each wallet's cash out margin protected, and the merchant payment and government collection rails fragment the same way.

Merchant payments and QR are the second test and the float is the prize. BanglaQR merchant transactions now route through the national payment switch [BB AR 2024], merchant accounts number about 1.3 million against 231.1 million personal accounts [BB AR 2024], and merchant payment was a 2 percent share of June 2024 volume against 30 percent for cash out [BB AR 2024]. Merchant float, the balances a QR accepting economy accumulates, is the raw material of wallet based savings and working capital products, and its volume could not be confirmed, the central bank's merchant payment series being the resolving source.

Digital banks are the third test and the least measured. The annual report's only digital bank references in FY24 are a regulatory framework training topic [BB AR 2024]; the licence count, the applicants and the launch dates could not be confirmed, with Bangladesh Bank the resolving source. The mechanism the licence class tests is whether a deposit taking institution with no branches can hold the 37.2 percent active account base [IMF FAS 2024] at lower cost than the branch estate, or whether digital bank licences become the holding company structure for the incumbents' wallets.

Cross border rails are the fourth test, and the base year comparison is stark. In December 2016 the financial intelligence unit recorded remittance inflow through mobile financial services of 81.2 million BDT, 0.11 percent of that month's remittances, in its investigation of hundi networks running through wallet agents [BFIU 2017]. A decade later the formal remittance flow the wallet channel sits beside is last confirmed at 27.52 billion USD for WDI year 2024 [WB WDI 2026], a WDI vintage figure not joined to Bangladesh Bank's fiscal year sums; a calendar 2025 total could not be confirmed from the sources this chapter draws on. Real time gross settlement settles US dollar transactions, 744 thousand worth 25,895 million USD in FY24 since the September 2022 opening [BB AR 2024], and the current share of remittances arriving through the wallet channel could not be confirmed, Bangladesh Bank's remittance channel tables being the resolving source. Chapter 21's price mechanism governs: the float closed the hundi wedge, and the wallet channel's share of the formal flow is the retail half of that settlement story.

The conversion question underneath all four tests is the one chapter 13 defers to this chapter and chapter 07 frames: whether payment ledgers attached to verified identity become admissible credit data. The consumer protection base it depends on is measured and thin: 22.03 percent of government to person beneficiaries surveyed in 2022 reported fraud or harassment in the payment process, 58.19 percent knew their personal identification number, and 6.08 percent reported paying an agent an extra fee [Harvard Dataverse 2022]. Only 39.12 percent of adults secure their phone with a lock [WB Findex 2024, con18]. The fraud and cybersecurity ledger, from the 81 million USD central bank heist of 2016 to the 2024 malware that intercepted two factor authentication codes, is chapter 54's account; the payments specific point is that conduct supervision under the new Act is the mechanism that decides whether trust scales with the rail.

The decision points, with authors: interoperability enforcement, authored by Bangladesh Bank's Payment System Oversight Division, decides whether the switch or the wallets win; digital bank licensing, authored by the same central bank with the Finance Division's banking franchise interests in the background, decides whether the wallet balance sheet becomes a deposit franchise; conduct supervision, authored by the central bank under the 2024 Act, decides whether the 22.03 percent fraud incidence [Harvard Dataverse 2022] falls; and ledger admissibility, authored jointly by the central bank and the repaired banks of chapter 06, decides whether the transaction history becomes the collateral substitute the small firm credit famine needs. The chapter 15 baseline assumes for chapter 07 that none of the four market decisions lands cleanly; the reform scenario assumes the conversion instruments exist at scale by the early 2030s, and this chapter's rule books are where that assumption is executed or dropped.

Three risks print in monthly series and the upside is one interoperability order away from intermediation

Risks. First, fraud outruns trust: a rail whose measured fraud incidence among the poorest users is 22.03 percent [Harvard Dataverse 2022], whose two factor codes are already a malware target [IIFC 2026] and whose hardware is unlocked for 60.88 percent of adults on chapter arithmetic [WB Findex 2024, con18] can lose the user base faster than licensing can add it, and the revealing indicators are the fraud incidence in the next survey wave and formal account ownership against the 43.3 percent of the 2024 wave [WB Findex 2024, account.t.d]. Second, the pipe ceiling binds: balances stay near 0.26 percent of GDP [IMF FAS 2024], merchant share stays near its 2 percent of volume [BB AR 2024], the cash out margin stays protected by closed loops, and the rail saturates as a toll road on remittance and salary flows, with the agent banking stall of FY25, minus 5.7 percent [BB Econ 2025, Table IIG], as the preview. Third, measurement failure: the March 2025 subscriber break [BB Econ 2025, Table IIG] goes unfootnoted, transaction value and account counts diverge, and the what to watch indicators of this chapter and chapter 07 stop resolving to a single series, which is chapter 53's regime in miniature.

Upside. First, interoperability completed: an enforced wallet to wallet settlement order converts the rail from thirteen loops into one national system, kills the cash out toll on competition, and is the cheapest financial inclusion instrument on the table; the revealing indicator is a published interoperable transaction volume series. Second, the merchant float and the credit conversion: QR acceptance at scale plus admissible ledgers gives the repaired banks of chapter 06 a cash flow underwriting base, and the small firm credit path the reform scenario targets runs through this decision; the revealing indicators are mobile money balances to GDP and the SME loan share recovering from 5.98 percent of GDP [IMF FAS 2024]. Third, the cross border retail rail: publication of the wallet share of remittances and its growth converts the largest external flow chapter 03 carries onto a measured retail channel, deepens the formalisation the float began, and gives the decade's largest single inflow a competitive market for its last mile; the revealing indicator is the channel split the central bank publishes.

What to watch: five indicators whose thresholds mark the conversion

1. Mobile financial services transaction value, fiscal year. Current value 1,817,408.1 crore BDT in FY25, up 18.3 percent on FY24 on chapter arithmetic [BB Econ 2025, Table IIG]. Threshold: growth holding above 15 percent a year signals merchant and government rails compounding on top of cash conversion; a fall below 10 percent signals the cash in, cash out saturation the usage mix warns about.
2. Mobile money balances to GDP. Current value 0.26 percent in 2024 [IMF FAS 2024]. Threshold: a sustained rise through 0.5 percent signals savings and float products arriving on the rail; a flat print below 0.3 percent through the window the chapter 15 scenarios assume confirms the payment pipe ceiling.
3. Agent banking transaction value, fiscal year. Current value 785,177.1 crore BDT in FY25, down 5.7 percent on chapter arithmetic [BB Econ 2025, Table IIG]. Threshold: a second consecutive annual fall confirms the bank owned rural channel contracting under the wallet's competition; recovery above the FY24 level of 832,532.9 crore signals a deposits led agent revival.
4. Government to person payment fraud or harassment incidence. Current value 22.03 percent of beneficiaries in 2022 [Harvard Dataverse 2022]. Threshold: a fall below 10 percent in the next survey wave signals conduct supervision working under the 2024 Act; a rise above 25 percent signals fraud scaling with the rail and the trust base eroding.
5. Mobile financial services share of remittance inflows. Last measured base 0.11 percent of monthly remittances in December 2016 [BFIU 2017]; the current share could not be confirmed, Bangladesh Bank the resolving source. Threshold: publication of a channel split above 10 percent signals the retail cross border rail arriving; continued non publication of the series is itself the stall indicator.

Sources used

[IMF FAS 2024] IMF Financial Access Survey via bdpolicy.db, series: imf_fas_mobile_money_registered_accounts, imf_fas_mobile_money_active_accounts, imf_fas_mobile_money_agent_outlets, imf_fas_mobile_money_tx_value_bdt, imf_fas_mobile_money_tx_pct_gdp, imf_fas_mobile_money_balances_pct_gdp, imf_fas_agent_banking_outlets, imf_fas_debit_cards, imf_fas_credit_cards, imf_fas_atms_countrywide, imf_fas_atms_per_100k, imf_fas_commercial_bank_branches_per_100k, imf_fas_commercial_bank_deposits_pct_gdp, imf_fas_sme_loans_pct_gdp.

[BB AR 2024] Bangladesh Bank Annual Report 2023-24 via ocr_text/bb/annual_report/ar2023-2024.txt, chapter 13 Payment and Settlement Systems: MFS providers, accounts and agents at end June 2024, FY24 MFS settlement, usage mix, BACH, BEFTN, RTGS including USD leg, NPSB and BanglaQR, PSP and PSO licences, Merchant Acquiring and Escrow Guidelines 2023, Prepaid Instrument Guidelines 2024, Payment and Settlement System Act 2024, Payment System Oversight Division, Regulatory Fintech Facilitation Office.

[BB FSR 2024] Bangladesh Bank Financial Stability Report 2024 via bdpolicy.db, series: bb_fsr_mfs_transaction_volume, bb_fsr_agent_banking_deposits, bb_fsr_internet_banking_volume.

[BB Econ 2025] Bangladesh Bank Monthly Economic Trends, October 2025 issue, Table IIG e-banking and e-commerce statistics and accompanying commentary: FY24 and FY25 MFS and agent banking transaction totals, monthly MFS values July 2024 to August 2025, internet banking commentary, and the MFS subscriber series break of March 2025.

[WB Findex 2024] World Bank Global Findex database, 2021 and 2024 waves via finance/findex_bd_full.parquet, series: account.t.d, con18.

[BBS Census 2022] Bangladesh Bureau of Statistics Population and Housing Census 2022 via bdpolicy.db, series: census2022_mfs_account_national_avg_pct.

[Harvard Dataverse 2022] G2P beneficiary payment collection channel shares, Bangladesh, DOI 10.7910/DVN/V4EKDG, via bdpolicy.db, series: ssn_g2p_mobile_agent_share_pct, ssn_g2p_fraud_harassment_rate_pct, ssn_g2p_pin_awareness_rate_pct, ssn_g2p_agent_extra_fee_incidence_pct.

[BFIU 2017] Bangladesh Financial Intelligence Unit Annual Report FY2016-17, section 7.2 remittances transferred illegally through MFS, December 2016 MFS remittance share and hundi agent investigation, OCR text in the source archive.

[IIFC 2026] IIFC Insight magazine vol 1 issue 2, 2026, e-governance and digital public infrastructure essays: the February 2016 central bank heist at approximately 81 million USD, the SikkahBot two factor interception, the myLocker exposure, and the interoperability absence finding.

[WB WDI 2026] World Bank World Development Indicators, personal remittances received, via indicators/wb_full_bd.parquet, series: BX.TRF.PWKR.CD.DT; the bdpolicy.db series wb_remittance_inflows carries no data points in the data lake and is not used.

---

Verified line by line against primary sources: 43 claims checked, 4 corrected.

Previous: 36 Youth, the demographic window and ageing

Next: 38 Insurance, pensions and long-term savings

All sixty chapters and the ten narrative chapters.