The Curious Case of the Falling Remittances

Global remittance flows command a rightfully growing amount of attention. Recently Pew published a visualization of World Bank data on international remittances that helps show the scale and corridors of transfers. Of note, FAI’s Alicia Brindisi has been writing about south-to-south remittances and the huge market they represent.

Remittance flows are, of course, primarily driven by migration patterns. The largest country-to-country corridors for remittances—from the US to Mexico ($22 billion in 2012), from the UAE to India ($17 billion) and from India to Bangladesh ($7 billion) for instance—match the flow of migrants. In the US to Mexico corridor, remittance flows fell during the Great Recession as there was a net outflow of Mexican migrants. Curiously though, while migrant flows have returned to pre-recession levels, remittances have not. Meanwhile major banks, which had invested in providing remittance services to that corridor, are cutting back services.

Felix Salmon at Reuters examines the phenomena, wondering what is going on. He suggests that the drive to lower costs of remittances is at least partially at fault. His theory is that remittance services need to be sold; in other words, consumers need to be convinced to use them. As competition and regulatory efforts push down the price of remittances, providers market the services less and therefore migrants use them less, keeping more cash in their pockets rather than sending it back to family. Salmon concludes that pushing to lower the costs of remittances may therefore be counterproductive.

While this theory might make sense, I think it’s wrong. Salmon’s thinking on remittances is somewhat emblematic of the thinking about migration and remittances that Michael Clemens and I take on in our recent paper on the research agenda in this domain. Remittances are not, for the most part, an optional choice for migrants, nor a windfall to the recipients. Thinking of remittances as a member of a household simply taking home his or her pay provides a clearer picture. When a household decreases spending, our first instinct is not to suspect that advertisers are failing to properly stimulate demand. In the case of remittances, migrant breadwinners are generally going to transmit as much money as they can through the most convenient and effective mechanism they can. If remittances to Mexico are falling we should begin looking at what migrants are earning, not the behavior of large providers of financial services.

While there is room for a lot more research in this area, the best evidence we have provides two reasons to doubt Salmon’s explanation:

First, there is good evidence that suggests that migrants are at least somewhat price sensitive. When the cost of remitting falls, they remit more. See this 2010 paper from Aycinenaet al. examining the US to El Salvador corridor, and this from Dean Yang on Filipino remittance response to exchange rate shocks. If we don’t see a positive response to falling remittances prices among Mexican migrants, this evidence suggests that their ability to remit is constrained, not their interest in doing so.

Second, migrants do appear to be close to sending as much home as they can. In both of the studies looking at price senstivity, migrants react but don't fully take advantage of falling prices. That's consistent with survey evidence that migrants are sending home a large portion of the earnings and have limited ability to send more. Another piece of evidence is from the MILK project which in work with Kenyan households found that migrants who were not already remitting responded very weakly to health shocks experienced by the origin household. It’s hard to imagine any bank marketing campaign being more powerful in generating remittances than word that someone at home needs urgent medical care.

So if Salmon’s explanation is wrong what are we left with? To start, I think it’s important to separate falling remittances from the behavior of the large banks. Even in a shrinking market, it’s possible to earn profits if you’re beating the competition. The withdrawal of the large banks indicates that they have failed at capturing market share. A far more likely explanation for the big banks moving out of this market, then, is that they simply don’t understand the needs of poor households. If large banks largely fail at attracting poor households to any of their services, what reason is there to believe that they would be able to attract customers to remittance services? As Salmon has noted in his writing about payday lending, customers are willing to pay for certain features of financial services, including speed, convenience and respect. Perhaps the big banks, as in other financial services for poor households, have not figured out how to put together the value proposition. And if Mexican migrants in the US are struggling, it’s all the more likely that they’ll be turning to the providers who have.

Most of all, it's reason to continue looking for ways to lower the cost of remittances in a way that does meet the needs of migrants.