The Impact of Existing Account Usage on the Likelihood of Switching Accounts in Malawi

Xavier Giné
Fieldwork by:
Southern Malawi
810 households
2012 - 2016
Target group:
  • Adults
Outcome of interest:
  • Savings/deposits
Intervention type:
  • Cash transfers
  • Savings
  • Monetary incentives
  • Unconditional cash transfers
  • Pricing and fees

Broadening financial inclusion is a top priority for many governments worldwide, but challenges remain to improve financial decision-making. Researchers partnered with a local bank to learn how customers chose between two account types—an account that had been offered for many years, which carried high monthly fees but no individual transaction costs, or a new type of account that charged for each withdrawal but did not impose monthly maintenance fees. Customers without an existing account recognized the advantages of the new type of account and overwhelmingly chose to open one, while many customers with the existing account type did not switch to the new one. However, of the existing customers, those who used their account more due to receiving a financial transfer in the past did switch to the new account type, suggesting that more experience using financial products can improve financial decision-making.

Policy issue

Switching from an existing product to a new one comes with costs, including financial, time, and cognitive burdens. Even if these costs are small relative to the potential gains of switching, individuals tend to value items that they already have more than items they do not own; that phenomenon is called the “endowment effect.” This psychological bias may have negative impacts on individuals’ welfare if people continue to use products that are not well suited to their needs, simply because they have grown accustomed to the product. In the context of financial inclusion, these behavioral barriers may undermine policymakers’ efforts to increase access to banking products, because individuals may not understand the features or costs associated with different accounts or products, or may select accounts that are poorly suited for their needs. If the first accounts that people open are expensive or not well designed for their financial habits, then even if better products are introduced later on, the endowment effect may bias individuals’ decision-making and lead them to stick with the suboptimal financial products or services. Can encouraging greater usage and familiarity with financial products improve financial decision making, even without explicit financial literacy training?

Context of the evaluation

In Malawi, financial inclusion is relatively weak, as 34 percent of adults have a bank account and nine percent have formal savings.1 This evaluation extended a project that began in 2012, during which researchers encouraged randomly selected households to open a basic savings account with no fees. Researchers offered these households assistance in opening the account and provided financial subsidies for almost four years to cover the costs of the required minimum balance of MWK 500 (US$1.34) and the monthly maintenance fees of MWK 400 (US$1.07).. Then, in April 2015, of 742 households that opened the subsidized savings accounts, researchers randomly selected 600 to receive a large financial transfer of MWK 25,000 (US$67), either in cash or through a direct deposit into their bank account. The transfers promoted greater usage of the bank accounts: recipients made on average 1.5 more deposits in the 12 months after the initial transfer. Subsidies for account fees ended in May 2016, at which time researchers evaluated if customers chose to keep, switch, or close their account as part of this evaluation.

Food market in Malawi
Photo: Brendan Howard |

Details of the intervention

Researchers partnered with NBS Bank, a local bank in Malawi, to learn how customers choose between two account types—one that was relatively more expensive but had been offered for many years, or a new, cheaper one. As the research team ended subsidies for households that had previously opened accounts, the bank introduced a new, cheaper savings account that was available to new customers and existing customers who wanted to switch accounts. In addition, researchers varied whether households received a bonus for showing up to the bank to open, switch, or close their accounts and varied the timeframe during which households needed to execute their decision.

Of the 742 households that opened the subsidized accounts during the previous evaluation, researchers were able to track down 594 households, and these constituted the “old” households in the evaluation. During home visits in March and April 2016, researchers informed these households that the subsidies for their accounts would end in May and presented households with three options moving forward:

  1. Continue with their existing basic savings account and pay the fees themselves: The bank would begin to deduct a monthly maintenance fee of MWK 400 (US$1.07) from their account. If the account balance fell below the minimum balance of MWK 500 (US$1.34), the bank would suspend their account. This was the default option for old account holders.
  2. Transfer their existing balance to a new account without monthly fees: The new, cheaper “Pafupi” account had no monthly fees, but charged MWK 150 (US$0.40) for each withdrawal. To open the new account, customers would also have to purchase an ATM card for MWK 1,300 (US$3.50).
  3. Close their existing savings account: Household would receive the full account balance, but would no longer have any savings account with the bank.

To compare this decision made by existing customers to a similar one made by individuals without a bank account, researchers randomly selected 216 “new” households from comparison villages that had not been offered the free savings account in the previous evaluation. These households had three similar options:

  1. Open a basic savings account with the same terms as above,
  2. Open a Pafupi account with the same terms as above,
  3. Turn down both offers and open neither account, which was the default option for new households.

Because the bank required households to visit the bank in person to open, close, or transfer accounts, researchers asked households to visit the bank by the end of May 2016 to report their decision. Researchers randomized three aspects of that visit. First, they randomly selected some households to receive a show-up bonus for visiting the bank; households were randomly assigned to receive either no payment, MWK 500 (US$1.34), or MWK 1,000 (US$2.68). Second, to study whether inattention or forgetfulness led people to keep their old accounts or not switch to the new account, researchers varied when households were asked to visit the bank. Among those households who received some financial incentive to visit the bank, the bonus was conditional on coming in a specific week, either within one week or after three weeks. Finally, at the bank, researchers randomly gave some households a larger than expected cash bonus of MWK 1,500 (US$4.02), which was sufficient to pay for the ATM card required to open a Pafupi account.

Researchers conducted household surveys and collected data from administrative records and observations at the bank to measure how previous experience owing and using savings accounts affected respondents’ choices over real financial products.

Results and policy lessons

While less than half of existing account holders switched to the cheaper, Pafupi account, all of the new households that opened accounts selected this cheaper option, indicating an endowment effect for existing customers such that they preferred their old account, even if more expensive. However, those existing account holders who had been encouraged to use their account more frequently after receiving a financial transfer were just as likely to select the cheaper accounts as the new households. This suggests that greater familiarity with the bank account can mitigate the endowment effect and therefore improve financial decision-making.

Most households visited the bank to report their decision, while offering a show-up bonus increased the likelihood of visiting: Seventy-five percent of old households and 66 percent of new households visited the bank. Offering a show-up bonus also increased the likelihood of visiting the bank. Offering individuals MWK 500 (US$1.34) increased the likelihood of visiting the bank by 26 percentage points from a base of 50 percent (a 53 percent increase). Doubling the show-up bonus to MWK 1000 (US$2.68) increased the probability of showing up by an additional 8 percentage points. However, extending the timeframe during which individuals could visit the bank had no impact on the likelihood they actually visited, which implies that forgetfulness or inattention did not prevent households from making decisions about their accounts.

Existing account holders were less likely to switch to the cheaper account, compared to new customers: Of the old households who already had the basic savings account, only 49 percent switched to the cheaper, Pafupi option. Conversely, among new households who previously had no account, 56 percent opened accounts and all of them chose the cheaper accounts. This contrast indicates an endowment effect; old clients preferred to keep their existing accounts, despite being more expensive.

However, familiarity with the account mitigated the endowment effect: Old account holders who received the transfers were nearly as likely as new customers to chose Pafupi accounts, thus offsetting the endowment effect. This was likely because the financial transfers increased the number of times households used their savings accounts. For each additional bank transaction encouraged by the financial transfer, old households were 18 percentage points more likely to switch the cheaper Pafupi account.

For policymakers interested in promoting financial inclusion, these results suggest that while financial literacy trainings are often ineffective, hands-on experience might lead households to make better financial decisions. Providing transfers directly into individual’s bank accounts can encourage usage, increase familiarity with banking products, and improve financial awareness and decision-making.

Demirgüç-Kunt, Asli, Leora Klapper, Dorothe Singer, Saniya Ansar, and Jake Hess. 2018. The Global Findex Database 2017: Measuring Financial Inclusion and the Fintech Revolution. Washington, DC: World Bank. doi:10.1596/978-1-4648-1259-0.