Senders now have more ways than ever to move money across borders, yet costs remain high and vary widely across corridors. This column argues that a missing piece in explaining these high costs is the exchange rate risk borne by the intermediaries that move the currency, the money transfer operators. Across 339 remittance corridors worldwide, fees rise when the recipient country’s currency becomes more volatile, and this currency risk adds about as much to the price of a transfer as operators’ markups. Competition alone cannot lower these costs.
The recent introduction of a 1% tax on cash transfers from the US has drawn renewed attention to the financial support migrants send to their families back home.1 International remittances, larger than foreign investment and aid combined,2 help households meet daily needs, pay for education, and insure against hard times (Yang 2008, Riley 2018, Dinkelman 2020, Khanna et al. 2026). Recognising their importance, the UN Sustainable Development Goals call for remittance costs to fall below 3% of the transaction amount by 2030. Yet the average fee for an ordinary transfer still stands at around 6%, showing how far that ambition remains from families’ everyday experience.
Consider a worker in New York wishing to send money to relatives in Lahore. A common option is to visit a local shop offering money transfer services, such as a grocery or a convenience store, and hand over US dollar banknotes to the agent. The recipient will then collect Pakistani rupees, in cash, at a local agent in Lahore. Transferring money this way can cost around 8–14% on a US$200 transaction,3 or between $17 and $27.4
In a column on VoxEU more than 15 years ago, Beck and Martínez Pería (2009) asked why sending money abroad was so expensive. Their answer pointed to the structure of the retail market, suggesting that policy should focus on promoting competition. Compliance costs and consumer information also received attention (Cecchetti and Schoenholtz 2018), while Bettin et al. (2017) and Beck et al. (2022) documented the role of financial access and the prevalence of cash.
Senders now have more ways than ever to move money across borders: new digital providers have challenged incumbents,5 while mobile money (Lee et al. 2021, Wieser et al. 2019)6 and stablecoins (Auer et al. 2025, Du et al. 2026) offer further alternatives. Real-time payment infrastructures are spreading worldwide (D’Andrea and Limodio 2024), with projects under way to link national systems.7 Moreover, price-comparison websites8 make it easier to evaluate different options and shop around (Nakasone et al. 2024).
Yet costs remain high and vary widely across corridors.9 In new research (Flori and Limodio 2026), we argue that a missing piece in explaining these high costs is the exchange rate risk borne by the intermediaries that move the currency, the money transfer operators (MTOs).
To see where this exposure arises, consider our earlier example. The intermediary in New York quotes an exchange rate and commits to delivering a fixed amount of Pakistani rupees. The recipient is paid from balances held with the local agent, which the operator must later replenish by buying rupees. If the rupee strengthens against the US dollar before that purchase, the settlement becomes more expensive. To protect against potential losses, operators typically build a margin into the exchange rate they offer customers. In highly uncertain environments, however, setting that margin becomes harder, and larger exchange rate swings make mistakes more costly.
Where financial markets are deep, with well-developed instruments to facilitate cross-currency transactions, the risk of holding foreign currency need not show up in prices. But remittances flow overwhelmingly towards currencies whose markets are thin and underdeveloped. Figure 1 shows that the depth of the foreign currency market (measured by its over-the-counter trading volume) rises steeply with the economic development of a country (measured by its log income per capita). For most remittance-receiving countries, the financial markets through which operators can mitigate currency exposure are shallow or virtually absent.10
Figure 1 FX markets are thin outside a handful of major currencies
Because this risk enters the marginal cost of delivering a transfer, competition cannot remove it. Competition squeezes markups, but a corridor can still be both highly competitive and very expensive.
We build a theoretical model and test its implications using the World Bank’s Remittance Prices Worldwide database on the fees charged by money transfer operators for 34,680 transfers across 339 corridors between 2011 and 2024.
Figure 2 shows that sending money to richer countries does not necessarily cost less. Transfers do, however, tend to be more expensive when the destination country’s currency fluctuates more.
Figure 2 Remittance fees vary with currency volatility, not GDP
We examine this relationship further by tracking fees within each corridor over time and comparing what the same operator charges across the corridors it serves. Fees rise as the exchange rate between the sending and receiving countries becomes more volatile, and during episodes of exceptional instability they rise by about $1 on a $200 transfer, on average.
To establish causality, we exploit the Swiss National Bank’s unexpected decision to abandon the franc’s floor against the euro on 15 January 2015.11 Before that date, a popular investment strategy consisted of borrowing francs at low interest rates and investing in currencies offering higher yields (such as the Norwegian krone). When the floor was removed, the franc surged, turning previously profitable trades into losses. The rush to reassess these positions triggered sharp swings in the currencies in which traders invested, raising volatility in exposed corridors by 32–42%.
This case study allows a sharp comparison: transfers to the same destination – say, Pakistan – from a country whose currency was exposed to the franc shock (such as Norway) versus one that was not (such as the US). Importantly, the shock arose for reasons unrelated to anything that would make migrants in Oslo send more money than those in New York.
Prices responded immediately (Figure 3). In the quarter of the shock, the cost of a $200 transfer from exposed countries rose by almost $1.60 relative to the control group, or 11% of the average cost ($14). The effect is stronger for thinner currency markets and for less geographically diversified operators, consistent with our conjecture that limited hedging opportunities make currency risk more costly. We also find that, following a currency shock, operators make lower profits and are more likely to exit the corridor. Currency instability thus hits migrants twice: through higher prices and, over time, reduced competition.
Figure 3 Remittance prices around the 2015 Swiss franc shock
To quantify how much currency risk adds to the fees ultimately paid by senders, we estimate a model of the remittance market.12 Comparing estimated costs with a counterfactual in which the exchange rate is fixed, we find that currency risk contributes to remittance prices about as much as markups, on average. Its share is highest in South America (24%) and sub-Saharan Africa (22%), where large devaluations and widespread currency illiquidity, respectively, make this risk particularly costly (Figure 4).
Figure 4 Markup and currency risk as a share of remittance fees, by region
Efforts to lower remittance costs have rightly targeted competition and transparency: these measures can reduce markups, but they do not eliminate the risk arising from conversion into local currency. Policy should therefore also promote deeper foreign exchange markets, expand access to affordable hedging instruments, and ensure liquidity is available during periods of market stress.
Reaching the 3% target will require looking at the currency markets on which every cross-border transfer ultimately depends.
Source : VOXeu
The theory of optimal currency hedging has been well established for decades. This column uses…
The human and economic costs of extreme weather events are already large and will grow…
Geopolitical tensions are changing the nature of supply-chain risk. This column presents new survey evidence…
Raising educational attainment is the canonical way skills keep pace with technology. This column argues…
Firms’ access to external finance differs widely across Europe. This column uses data from the…
Whether inflation can ease public debt burdens is an increasingly important question as debt sustainability…