Finance: Research, Policy and Anecdotes

In 2011, I published a paper with Sole Martinez Peria on the drivers of remittance costs across countries, with data on 119 corridors and for 2009. Thanks to amazing data collection effort of the World Bank on Remittance Prices Worldwide, Roland Kpodar, Mathilde Janfis and I expanded the previous exercise to a panel of corridors over time, even exploiting variation across different providers within a corridor, now published as IMF Working Paper as well as a summary on VoxEU.

We document a decline in remittance prices, with the median remittance fee decreasing from 7.7 percent in 2011 to 5.7 percent in 2020. Nevertheless, most of the corridors still have median remittance cost above 5 percent in 2020, far from the 2030 Sustainable Development Goal target of average fees of less than 3 per cent and no remittance corridors with costs higher than 5 per cent. The data also reveal significant variations in remittance fees within the same corridor (across firms), between the same sending country and different receiving countries, or between the same receiving country and different sending countries. Similarly, remittance fees vary with the payment instrument, the access point, and the speed of the transfer.

What factors explain the variation in remittance costs? Five results stand out:

• Higher GDP per capita in the sending country and easier geographic access to financial institutions is associated with lower fees, especially for banks.
• Scale economies matter: a larger market for remittances (as proxied by closer economic ties and a larger migrant population from the receiving in the sending country) is associated with lower costs as is a shorter distance between sending and receiving countries.
• The market structure is important: banks charge higher fees than money transfer operators (MTOs), but a larger share of banks among remittance service providers is also associated with higher fees charged by MTOs. Unlike banks, MTOs’ fees react to competitive pressures, with more market players being associated with lower MTO but not bank remittance fees.
• In corridors where the sending country has a pegged exchange rate, both banks and MTOs charge lower fees.
• There is some evidence that cash payments attract higher fees, while payments over the Internet are charged lower fees. Nonetheless, there are no conclusive results regarding the impact of the regulatory framework, which leaves unsettled the debate that compliance to regulatory requirements (e.g., on AML/CFT issues) heightens the cost of remittance services.

Overall, these findings point to both cost- and risk-based constraints and market structure as barriers to lower remittance fees. Higher transaction costs as result of a more rural population in the sending country and lower scale can explain high remittance fees in some corridors. These factors are largely structural, implying a limit to the extent to which remittance fees can be lower with policy actions.

However, decisive policy actions are needed in areas that are directly under the control of policy makers and where the yield from reforms can quickly materialize. For instance, stronger competition through easing entry to the remittance market, especially for non-bank providers, and digitalization might help reduce remittance costs. Similarly, exchange rate stability (or better hedging possibilities) might contain remittance costs, more so because exchange rate margins (often not fully disclosed by remittance service providers) can make up a significant portion of the remittance fees in corridors where exchange risks are the highest. The one area where our analysis cannot really make any clear inferences (due to endogeneity concerns) is on the role of the regulatory framework in its direct effect on remittance costs.

Overview