Cash Transfers Get an Upgrade

In a recent post, Jeremy Shapiro points to the Findex database, using it to suggest that recipients of government transfers are more likely to have a financial account. The Findex data, however, is only a starting point. It doesn’t tell us anything about particular programs that bundle government transfers with financial inclusion. Do programs like this exist? If they do, where are they? What types of services do they offer and do they enhance welfare for recipients?

We went looking and it turns out that at least 39 of these programs exist across Africa, Asia and Latin America, mostly offering financial inclusion through a digital delivery mechanism. But there wasn’t enough data to answer the big question: do these programs help recipients more than the alternative of just giving a larger transfer?

Government transfers are historically delivered in cash. Recently they’re being delivered through or alongside a financial service (e.g., mobile money or access to a microloan), making them what we call a Financially Inclusive Program (FIP). To find which transfers were FIPs, we looked at every cash transfer program in two recent reports, the World Bank State of Social Safety Nets (2015) and ODI Cash Transfers: What does the Evidence Say? (2016).

From this list of 212 cash transfer programs, we found that only 39 were delivered through or alongside a financial service. The remaining programs in the sample offered only a cash transfer and we could not consider them financially inclusive. Most of the 39 FIPs were identified through their delivery mechanism; they are sent by digital means such as mobile money, or through formal financial mechanisms such as a bank transfer.

Digital transfers are becoming increasingly popular with governments and donors as they offer cost savings for both senders and receivers, as shown by Aker et al. in a randomized trial in Niger. In these cases, the addition of a financial product may not require a tradeoff between just offering more cash or bundling with a financial service; digitization drives the total cost of the program down. To take advantage of this cost saving, we imagine that programs not identified among the 39 could easily become financially inclusive by digitizing the delivery mechanism.

However, there may be even better welfare increasing effects of FIPs beyond just cost saving. So far, we know that both cash transfer programs and financial inclusion have been independently shown to enhance welfare. There is also strong evidence from graduation programs that combining transfers with a large bundle including financial services, training, and a productive asset is welfare enhancing. We are interested in the potential of a simpler bundle, the FIP. Can welfare gains beyond more efficient delivery be won from the right combination of financial services and transfers?

Here’s how it could work. Cash transfers automatically increase incomes. Financial services help people make better investments and protect themselves from shocks. Provided together in a bundled program, cash and financial services could amplify the impact of offering either independently. In short, FIPs could make cash go further.

At the moment this is a theory that lacks evidence to back it up; only a handful of programs are currently designed in this way. If bundling cash transfers with financial services increases the cost of the program beyond the recipients’ total benefit, then governments are better off just giving a larger transfer. A major challenge we faced in establishing a causal link between FIPs and welfare outcomes was the lack of information available. It was often difficult to tell how the transfer was delivered or if it was bundled with any other financial services. Only the CEPAL database, reporting on just Latin America, identified the delivery mechanism for the transfer.

But even CEPAL has drawbacks – there is still no standard definition for the variety of financial services used to deliver transfers. One program may list a ‘debit card’ as the transfer mechanism while another lists a ‘magnetic card’. Are these the same thing? It’s unclear. Determining the relative ability of different delivery mechanisms and financial services to enhance transfers is challenging without clear, universal definitions.

What is clear, however, is that FIPs are an exciting opportunity for cash transfers to deliver cost-effective poverty alleviation. While we may not know the extent of poverty alleviation they may offer, we do know that government transfers are associated with increased use of financial services, digital delivery reduces costs, and a larger bundle of services produces sustained welfare gains. Figuring out the best combination of services will only happen through more research of existing and future programs. We recommend the following:

1. Cash transfer programs should offer financial inclusion through digital delivery mechanisms, taking advantage of likely cost savings and potential for huge welfare gains.
2. Existing and future programs should be incentivized to clearly state if they have a financial inclusion component, allowing them be easily identified and studied.
3. Key stakeholders including donors, governments and financial service providers should pin down key definitions relevant to FIPs and integrate them into standard reporting protocols for programs to get the ball rolling.
4. A comprehensive global database should be created to aggregate FIP data allowing for evaluation of particular cash transfer and financial inclusion combinations, guiding future program design.

Given the evidence that’s out there, and the potential for these programs, we think this idea holds a ton of promise. With a bit of work and some additional convening power we can transform correlations in the Findex data into causal analysis for increasingly effective policy design.