Data & Methodology

Data

The dataset includes all countries, providers, and survey rounds. Please refer to the codebook for variable names and definitions

data = FileAttachment("data/all_listed_prices.csv").csv({ typed: false})

Inputs.table(data, {select: false})

button(data, 'all_listed_dfs_prices.csv')

// from https://observablehq.com/@jeremiak/download-data-button
button = (data, filename = 'data.csv') => {
  if (!data) throw new Error('Array of data required as first argument');

  let downloadData;
  if (filename.includes('.csv')) {
    downloadData = new Blob([d3.csvFormat(data)], { type: "text/csv" });
  } else {
    downloadData = new Blob([JSON.stringify(data, null, 2)], {
      type: "application/json"
    });
  }

  const size = (downloadData.size / 1024).toFixed(0);
  const button = DOM.download(
    downloadData,
    filename,
    `Download Data (~${size} KB)`
  );
  return button;
}

Download Codebook

Methodology

Overview

As part of Innovation for Poverty Action (IPA)’s Consumer Protection Research Initiative (supported by the Bill and Melinda Gates Foundation), IPA will collect and publish as a public good pricing data for a variety of common DFS transactions across 130 providers in 18 low- and middle-income countries. Utilizing automated web-scraping techniques, we will gather this data on a regular basis from leading DFS providers in each country, standardize it across countries, and where possible, disaggregate provider fees and government taxes.

This data will enable BMGF, regulators and researchers to track changes in DFS pricing over time, and investigate differences in pricing models both across countries and within countries. For example, we can explore whether countries with more competitive DFS markets tend to have lower prices or whether providers in a given market tend to converge on a single price or whether firms appear to actively compete on price.

From a consumer perspective, the lack of standardized pricing disclosures makes comparison costs across providers difficult. A “live” pricing tracker will empower users to easily compare transaction costs across providers in their markets, facilitating more informed financial decisions.

Objectives

The goal of the DFS price scraping work is to capture the price (official fee + tax) consumers face when moving money domestically. Primarily this entails:

1. Exchanging cash for e-money and vice versa (CICO)
2. Moving money between personal accounts (including within the same institution, across institutions of the same type)
3. Paying for goods or services (including merchant payments, utility payments)

This definition specifically excludes: other types of payments, including B2B and G2P payments; international transfers; account fees not associated with the movement of money (e.g., maintenance fees, penalty fees); fees for other types of financial services such as credit or insurance; unofficial fees including informal agent fees (whether illicit or not).

We aim to develop an automated, generative AI-powered tool for finding, collecting and standardizing pricing information.

Categories

1. Countries: A total of 18 countries will be included, comprising countries from the Transaction Cost Index in Phase 1 of the CPRI and the 9 BMGF priority countries. Namely, these countries are Ethiopia, Kenya, Tanzania, Uganda, Côte d’Ivoire, Ghana, Sierra Leone, Mali, Nigeria, Bangladesh, India, Indonesia, Myanmar, Pakistan, Philippines, Colombia, Paraguay, and Peru
2. Transaction types: We include transactions related to the domestic movement of money. Namely, these are: cash-in via agent, cash-in via ATM, cash-out from agent, cash-out from ATM, P2P on-network transfer, P2P off-network transfer, P2P to unregistered user (i.e. voucher code), wallet to bank (pushing funds from non-bank wallet to bank account – only applicable for non-bank providers), bank to wallet (pushing funds from bank account to non-bank wallet – only applicable for bank providers), utility payment (electricity), and payment at merchant.
3. Provider: In an ideal scenario, we want to cover providers with the largest market share in domestic transactions (CICO, P2P transfers, merchant payments), regardless of categorization (i.e. mobile banking, fintech, mobile network operator), aiming at a minimum aggregated market share of 80 percent. However, reliable market share data is often private and difficult to obtain. To address this limitation, we utilized language learning models like ChatGPT to obtain a general overview of the top providers for mobile money and mobile banking in each market. Additionally, we considered other types of financial service providers, such as “Fintech E-wallets” or “Digital Wallets”, that offer similar services but may not fall strictly under mobile money or mobile banking. If these other services are used by at least 10% of the adult population in the country, we include them in the data collection. Below are the specific prompts used:

   • Step 1. Provide a list of the three mobile money providers with the largest market share in this country. Mobile money is defined as an account that you can use your phone which is connect to a large agent network where you can use cash-in cash-out

   • Step 2: List the Top Banks for Mobile Banking: Identify the three banks with the largest market share in terms of mobile banking services in this country.

   • Step 3: Describe Other Types of Financial Services:

     This step involves other types of financial service providers, different from the providers identified in Steps 1 and 2, available in this country that allow for any of the following:

     1. Hold funds digitally.
     2. Transfer money domestically.
     3. Make payments for goods and services.
     4. Convert cash to e-money and vice versa.

     These services may include what are known as “Fintech E-wallets” or “Digital wallets.”

   • Step 3A: Usage Estimate: Do you estimate that these other services are used by at least 10% of the adult population in this country?

   • Step 3B: Provider Type: ONLY IF THE ANSWER TO 3A IS “YES” For each provider listed in the previous step, identify the type of provider and note which of the 4 services identified about the provider offers (recall to be included, the provider must offer at least one of these services)

   • Step 4: Summarize each selected provider identified through Steps 1, 2, and 3. No new providers should be added in this step. Separate the list into these categories: (1) mobile money, (2) mobile banking, (3) other services. Under each category, 3 providers should be included. Category (3), other services, should be excluded depending on your response to Step 3A. Include a source indicating that the provider is active in the country.

When we identify duplicates in the list (i.e. those belonging to the same network), we remove them. Specifically, for mobile money and mobile banking providers, if removing a duplicate results in fewer than three providers in either category, we replace it with another provider by consulting ChatGPT again.

See Annex A for the list of 130 providers included in the study.

• Tax treatment: Where available, costs scraped will include taxes to reflect the full price incurred by consumers. The provider fees and taxes will be disaggregated in the dataset. However, since most pricing information does not explicitly include tax details, IPA cannot guarantee that the scraped prices fully account for taxes. This limitation will be clearly stated in the disclaimer. Nonetheless, IPA will continue to explore efforts within IPA and by other organizations that review tax regulations for financial services to determine if this information can be incorporated into the analysis.

• Switches: In markets where fees for multiple switches are available, we select the cheapest on the website. We exclude fees for switches that are not posted on the website.

• Account type selection: If different account types are offered (current, savings), we select the account type with the lowest E-KYC requirements, which offers all the transaction types we include in our review (CICO, on-net transfers, off-net transfers, merchant payments).

Current Beta Version

The current beta version includes 13 mobile money providers in 11 countries (Table 1).

Table 1: Providers in the Beta Version

Country	Provider
Bangladesh	Rocket
Cote d’Ivoire	Moov Africa
Ethiopia	Telebirr
Mali	SAMA
Nigeria	Paga
Pakistan	Easypaisa Telenor
Paraguay	Personal Pay
Paraguay	Claro Movil
Peru	Bim
Sierra Leone	Africell
Sierra Leone	Orange
Tanzania	Vodacom
Uganda	MTN

Reference Value

To compare fairly across countries, we computed the fees as a proportion of a reference value. The simplest approach would be to set a single USD value, as is done in similar price measurement work such as the World Bank’s Remittance Prices Worldwide. However, our intent is to replicate the experience of the typical mobile money user in each country as much as possible. Because of differences in the economic development of each country, typical mobile money transaction sizes differ significantly across countries, meaning using reference values at set USD amounts is not ideal.

Because data on the distribution of mobile money transaction sizes is not publicly accessible in most countries, we had to find a different approach to setting reference values. Using self-reported transaction data from IPA’s consumer protection surveys, we found that median transaction sizes tended to equate to approximately 15 times the mean daily income per capita for the bottom 40 percent of the population. Although this is a very rough approximation, we believe it reflects the typical transaction size more accurately than a set USD value, and we tried to recreate this for each country. We implemented this approximation in the following way.

We used World Bank data on the daily mean income per capita for the bottom 40 percent of the population to approximate the median transaction size. We converted this value to local currency in 2017 using World Bank’s PPP conversion factor for 2017, then inflated it to current local currency using local CPI levels. Lastly, we multiplied this income per capita in current local currency by 15 to obtain our reference value.

Table 2: Reference Values

Country	Reference Value 1
Bangladesh	BDT 1,700
Colombia	COP 120,000
Cote d’Ivoire	XOF 12,000
Ethiopia	ETB 800
Ghana	GHS 150
India	INR 900
Indonesia	IDR 320,000
Kenya	KSh 1,600
Mali	XOF 8,000
Myanmar	MMK 23,000
Nigeria	NGN 8,000 2
Pakistan	PKR 3,500
Paraguay	PYG 350,000
Peru	PEN 180
Philippines	PHP 1,500
Sierra Leone	SLE 270
Tanzania	TZS 24,500
Uganda	UGX 35,000

Annex A

Table 3: List of providers

Country	Provider	Type of Provider
Bangladesh	Rocket	Mobile Money
Bangladesh	Bkash	Mobile Money
Bangladesh	Nagad	Mobile Money
Bangladesh	BRAC Bank	Mobile Banking
Bangladesh	Dutch Bangla Bank	Mobile Banking
Bangladesh	Islami Bank Bangladesh Limited (IBBL)	Mobile Banking
Colombia	Daviplata	Mobile Money
Colombia	Nequi	Mobile Money
Colombia	Movii	Mobile Money
Colombia	Bancolombia	Mobile Banking
Colombia	Davivienda	Mobile Banking
Colombia	Banco de Bogota	Mobile Banking
Colombia	Rappi Pay	Other types of FSPs
Colombia	Billetera PSE	Other types of FSPs
Colombia	TuDinero	Other types of FSPs
Côte d’Ivoire	Moov Africa	Mobile Money
Côte d’Ivoire	Orange	Mobile Money
Côte d’Ivoire	MTN	Mobile Money
Côte d’Ivoire	Societe Generale de Cote d’Ivoire (SGCI)	Mobile Banking
Côte d’Ivoire	Banque Atlantique Cote d’Ivoire (BACI)	Mobile Banking
Côte d’Ivoire	Banque National d’Investissement (BNI)	Mobile Banking
Ethiopia	Telebirr	Mobile Money
Ethiopia	M-BIRR	Mobile Money
Ethiopia	HelloCash	Mobile Money
Ethiopia	Commercial Bank of Ethiopia	Mobile Banking
Ethiopia	Awash Bank	Mobile Banking
Ethiopia	Dashen	Mobile Banking
Ghana	Vodafone	Mobile Money
Ghana	MTN	Mobile Money
Ghana	Airtel Tigo Money	Mobile Money
Ghana	Ecobank Ghana	Mobile Banking
Ghana	Absa Bank Ghana	Mobile Banking
Ghana	Access Bank Ghana	Mobile Banking
Ghana	Zeepay	Other types of FSPs
Ghana	Hubtel	Other types of FSPs
Ghana	ExpressPay	Other types of FSPs
India	Paytm	Mobile Money
India	PhonePe	Mobile Money
India	GooglePay	Mobile Money
India	ICIC Bank	Mobile Banking
India	HDFC Bank	Mobile Banking
India	State Bank of India	Mobile Banking
India	MobiKwik	Other types of FSPs
India	FreeCharge	Other types of FSPs
India	AmazonPay	Other types of FSPs
Indonesia	GoPay	Mobile Money
Indonesia	OVO	Mobile Money
Indonesia	DANA	Mobile Money
Indonesia	BRI	Mobile Banking
Indonesia	Bank Madiri	Mobile Banking
Indonesia	Bank Central Asia	Mobile Banking
Indonesia	LinkAja	Other types of FSPs
Indonesia	ShopeePay	Other types of FSPs
Indonesia	Blibli Pay	Other types of FSPs
Kenya	M-PESA	Mobile Money
Kenya	Airtel Money	Mobile Money
Kenya	T-Kash (Telcom)	Mobile Money
Kenya	KCB Bank	Mobile Banking
Kenya	Co-operative Bank of Kenya	Mobile Banking
Kenya	Equity Bank	Mobile Banking
Kenya	M-Shwari	Other types of FSPs
Kenya	Tala	Other types of FSPs
Kenya	Branch	Other types of FSPs
Mali	SAMA	Mobile Money
Mali	Moov Africa	Mobile Money
Mali	Orange	Mobile Money
Mali	Malitel Money	Mobile Money
Mali	Banque Internationale pour le Commerce et l’Industrie du Mali (BICIM)	Mobile Banking
Mali	Banque Malienne de Solidarité (BMS)	Mobile Banking
Mali	Banque Sahélo-Saharienne pour l’Investissement et le Commerce (BSIC)	Mobile Banking
Myanmar	Wave Money	Mobile Money
Myanmar	M-Pitesan	Mobile Money
Myanmar	KBZPay	Mobile Money
Myanmar	KBZ Bank	Mobile Banking
Myanmar	CB Bank	Mobile Banking
Myanmar	AYA Bank	Mobile Banking
Nigeria	Paga	Mobile Money
Nigeria	PalmPay	Mobile Money
Nigeria	Opay	Mobile Money
Nigeria	Access Bank	Mobile Banking
Nigeria	Zenith Bank	Mobile Banking
Nigeria	Guaranty Trust Bank	Mobile Banking
Nigeria	Kuda	Other types of FSPs
Nigeria	Remita	Other types of FSPs
Nigeria	GTBank’s GT World	Other types of FSPs
Pakistan	EasyPaisa Telenor	Mobile Money
Pakistan	Jazz	Mobile Money
Pakistan	Ufone	Mobile Money
Pakistan	Habib Bank Limited	Mobile Banking
Pakistan	Meezan Bank	Mobile Banking
Pakistan	United Bank Limited	Mobile Banking
Paraguay	Personal Pay	Mobile Money
Paraguay	Claro Movil	Mobile Money
Paraguay	Tigo Money	Mobile Money
Paraguay	Banco Itau	Mobile Banking
Paraguay	BNF Banco Nacional de Fomento	Mobile Banking
Paraguay	BBVA Paraguay	Mobile Banking
Peru	Bim	Mobile Money
Peru	Movistar	Mobile Money
Peru	Claro	Mobile Money
Peru	BBVA	Mobile Banking
Peru	BCP (Banco de Credito)	Mobile Banking
Peru	Interbank	Mobile Banking
Peru	Yape	Other types of FSPs
Peru	Tunki	Other types of FSPs
Peru	Kuspit	Other types of FSPs
Philippines	PayMaya	Mobile Money
Philippines	GCash	Mobile Money
Philippines	SmartMoney	Mobile Money
Philippines	UnionBank	Mobile Banking
Philippines	BPI	Mobile Banking
Philippines	BDO	Mobile Banking
Sierra Leone	Africell	Mobile Money
Sierra Leone	Orange	Mobile Money
Sierra Leone	Splash Money	Mobile Money
Sierra Leone	Sierra Leone Commercial Bank	Mobile Banking
Sierra Leone	Union Trust Bank	Mobile Banking
Sierra Leone	United Bank of Africa	Mobile Banking
Tanzania	Vodacom	Mobile Money
Tanzania	Airtel	Mobile Money
Tanzania	Tigo Pesa	Mobile Money
Tanzania	CRDB Bank	Mobile Banking
Tanzania	NMB Bank	Mobile Banking
Tanzania	Standard Chartered Bank Tanzania	Mobile Banking
Uganda	MTN	Mobile Money
Uganda	Airtel	Mobile Money
Uganda	Wave Money	Mobile Money
Uganda	Stanbic Bank Uganda	Mobile Banking
Uganda	Equity Bank Uganda	Mobile Banking
Uganda	I&M Bank Uganda	Mobile Banking

Footnotes

1. Local CPI data for 2024 were used, except for Ethiopia, Indonesia and Myanmar, which used the most recent available data from 2023, 2023 and 2019, respectively.

2. The World Bank does not have data for Nigeria on income per capita for the bottom 40 percent of the population, so we adopted Ghana’s income per capita (bottom 40 percent) in the calculations as it is the most similar economy in terms of GDP per capita.