Central banks that seek to promote financial inclusion often have an explicit goal to increase the number of customers (or rural customers, or poor customers) that have bank account. It’s easy to understand this policy aim. After all, bank accounts are the foundation of my financial life, and probably of yours, too. But is it always the case that a bank account is the product that the poor could benefit the most from?
The evidence suggests to me that it’s not. Exhibit A, of course, is M-PESA in Kenya. Thirteen million people in Kenya have signed up for a service that meets their need to send (and store) money. It is one of the great, runaway success stories in the history of financial inclusion, and it opened the eyes of many of us to the fact that payments are an important financial need.
Exhibit B, C, and so on are a depressing series of initiatives that have been taken around the developing world to encourage banks to open accounts for the unbanked. Whether through moral suasion or by threatening punitive measures for unsuccessful banks, there’s plenty of evidence that central banks can cause banks to open new accounts for low-income customers. These initiatives are often trumpeted as triumphs of financial inclusion. But look closer and the results have been decidedly mixed.
One carefully documented example is that of the mzansi account in South Africa. In 2004, the four large commercial banks in South Africa all introduced entry-level “mzansi” accounts to comply with their obligations under the Financial Sector Charter, which aimed to improve access to banking services. In terms of adoption, the initiative was a striking success: 6 million new accounts were opened, including 4 million by people who were previously unbanked.
Unfortunately, four years after mzansi accounts were introduced, nearly half (42%) were dormant. Moreover, among accounts that were active, the dominant use case was “dump and pull”: immediately withdrawing money that had been deposited by an employer (i.e., a salary payment) or the government (i.e., a social grant). Despite the fact that many account holders indicated that they opened the account to use it to save, the “vast majority” of accounts were not being used for saving.
This isn’t a good outcome for customers, for banks, or for government. But relatively speaking, it’s a success story compared to other “bank the unbanked” drives that are driven by policy priorities. In 2005–2006 the Reserve Bank of India exhorted banks to create and promote “no-frill” accounts, targeting 100% financial inclusion in at least one district in east state/union territory. By November 2008, 155 districts were declared to have achieved 100% financial inclusion. But a study conducted in Cuddalore district of Tamil Nadu, India, concluded that just 15% of customers were using their no-frill bank account after opening one in a registration drive.
The difference between the overwhelming demand for M-PESA on the one hand and the disappointing take up of entry-level bank accounts on the other is stark. It is even more remarkable when you consider that M-PESA is by all accounts a profitable business for Safaricom, while the evidence suggests that entry-level accounts in South Africa and India are offered to customers at a loss.
A broader definition of financial inclusion is needed. Rather than starting with bank accounts and working out how we can deliver them to the poor, a more pragmatic approach is to start by understanding their financial needs so that an enabling environment for products that meet them can be established. One of the extraordinary developments over the last several years has been the new appetite shown by non-banks to help meet these needs, by, for example, developing new payment instruments. This is a trend worthy of the support of financial regulators throughout the world, and these new services should be understood as being very much a part of financial inclusion.