Reducing the costs of saving
Savings offer households a buffer that enables them to invest in or respond to unexpected setbacks. Yet, as of 2017, around 1.7 billion adults remained “unbanked,” possessing neither an account at a formal financial institution nor with a mobile money provider.1 People who do not have access to formal savings may instead rely on other ways to save, but these alternatives can be riskier, more expensive, and less practical than formal accounts. For instance, 28 percent of unbanked adults reported that they had saved in the past year, often using informal savings groups, cash, or assets.2 Another strategy to manage risk or facilitate investment is to borrow with credit. While strategies exist to reduce the cost of lending, low-income borrowers typically still face high prices for credit.
While savings can be a lower-cost way to address the same financial needs, saving in a bank account can be costly for low-income individuals. Both monetary costs, such as account fees, and non-monetary costs, like distance to the nearest bank, can hinder low-income clients from opening and using savings accounts. Though saving may not be appropriate for all people in all scenarios—for instance, when they have high-interest debt to pay off—facilitating savings for some low-income clients may be an important goal of policymakers and financial institutions.
This review of fourteen randomized evaluations3 that addressed the monetary and non-monetary costs associated with savings found that reducing these costs alone increased account ownership but did not translate into regular use in most cases. This finding suggests that 1) these accounts may not meet all customers’ needs; 2) other barriers, like social pressure and lack of trust in formal financial institutions, still constrain savings behavior; or 3) people simply may not have enough income to make using their account worthwhile. However, of the ten studies that measure additional welfare effects beyond savings, seven find generally positive impacts, especially among participants who frequently used their accounts. Overall, while there is significant demand for formal savings accounts in low-income countries, offering free or low-cost savings accounts alone will likely not increase savings; instead, further research is needed on how to increase account usage through improved product design, as well as linkages to payments or other facets of economic life.
Lecciones de la Evidencia
High transaction costs may prevent people from accessing and using formal savings accounts. Typically, people pay fees to open a savings account and make withdrawals, and they must maintain minimum balances. For low-income clients, these fees may be high enough to discourage formal saving or significantly reduce saving balances. Non-monetary costs, like the distance to a bank or time to open an account, can also be large enough to deter people from opening and using one. Surveys highlight the relevance of these barriers, as around one-fifth of adults without a formal bank account cited high costs and distance (26 and 22 percent, respectively) as barriers to owning an account.4
Most low-income households open accounts when costs are reduced or eliminated. In eleven of the fourteen evaluations, clients paid little to no fees to open and use formal bank accounts, as either the bank offered free accounts or researchers provided financial subsidies and helped individuals with the paperwork to open accounts. As most individuals opened accounts when subsidized, these high take-up rates in many contexts reflect demand for low-cost formal saving products among low-income clients. For example, when monetary costs were reduced in Kenya and Nepal, over 80 percent of individuals opened accounts . On the other hand, in contexts where take-up was low, factors other than cost likely drove clients’ decision to open an account. In Chile (17 percent take-up), for example, individuals already had relatively easy access to a set of formal financial products .
However, it is harder to encourage people to use their accounts and change savings habits, as increased account ownership did not consistently translate into increased account usage. While most people opened an account when it was offered to them for free, a smaller share of individuals frequently used, or made deposits into, these accounts. In most cases, fewer than half actively used5 their account (Table 2), except for in Nepal, where 80 percent of households made at least two deposits within the first year . One concern is that even with access to free or subsidized accounts, other monetary and non-monetary costs dampen usage. For instance, distance to the bank can reduce how often individuals use their accounts, as evident in Uganda and Malawi . Likewise, high fees on withdrawals may inhibit use . Conversely, the lack of any account fees, including withdrawal fees, and close proximity to the bank contributed to the active usage seen in Nepal .
Low account usage suggests that barriers other than cost may hinder saving. Behavioral biases, such as inattention and temptation, can limit savings. These issues are prevalent in both high- and low-income countries, with extensive evidence particularly on retirement savings in the United States.6 Another barrier likely more unique to low-income settings, social demands from others in one’s household or broader community can erode savings or prevent individuals from saving in the first place. For example, in Kenya, individuals with low bargaining power (mostly women) used their accounts less after being offered free ATM cards . The ease of withdrawal made the money more vulnerable to be spent by their husbands or other family members. Additionally, limited access to information and financial knowledge may lead people to save less. However, more often than not, the impacts of financial education interventions are minimal, particularly in low- and middle-income countries.7 Expanding access to products designed to overcome barriers that clients face may increase account usage. Product design features that can offset barriers to save include accounts with reminders; commitment savings devices, which restrict access to savings until meeting a specific goal; group-based savings models; automatic deposits; accounts linked to payments; or private accounts.
But not all barriers can be addressed through design features. For those who do try to open accounts, regulatory barriers may hinder access with unnecessarily stringent Know Your Customer requirements. Trust is also an issue: low-income customers may opt out of saving formally because they do not trust financial institutions. This was a factor in Kenya, where a history of banking scandals in the 1980s and 1990s contributed to low trust in financial institutions .
Account features and other ways that make it easier to make deposits can increase account usage. Offering ATM cards or door-to-door deposit collection services, where bank officers visited households regularly to collect deposits, encouraged account holders to make more transactions and save more . Weekly deposit collection services in Sri Lanka that reduced travel time and distance to make deposits nearly doubled monthly bank savings . In addition to addressing travel costs, deposit collection services address other barriers to saving: they combat behavioral biases because deposit collectors regularly encourage account holders to make deposits, and the face-to-face contact with deposit collectors may impose social pressure.
Limited, but promising, evidence suggests that digital financial platforms make it easier for savers to open and use accounts in some cases. Digital technology offers a promising way to promote financial inclusion, as two-thirds of individuals who do not have a formal bank account own a mobile phone.8 Studies in Mozambique and Sri Lanka measure the impacts of offering individuals mobile-linked savings accounts. Those offered mobile-linked accounts increased deposits into these accounts, but only in Mozambique did this lead to an increase in overall savings . In contexts where mobile money is relatively well-developed and low-cost banking options already exist, reducing transaction costs further through digital platforms may not encourage additional formal savings, as was the case in Sri Lanka . Further research is needed to understand how digital financial platforms can also address other barriers to saving, in addition to reducing costs.
Incentivizing savings with a modest increase in interest rates had a limited effect; much higher, above-market interest rates did increase savings but are likely not viable for a bank to offer. Offering higher interest rates is another way to potentially encourage savings, but interest rates that are financially sustainable for a bank seem insufficient to increase savings. In the Philippines, increasing the interest rate from 1.5 to 3 percent for a commitment savings account had no impact on clients opening an account or saving more . On the other hand, consumers did respond to substantially higher rates of up to 20 percent in Kenya and Mozambique, but these rates were far beyond what typical banks would offer without subsidies  .
Reducing the costs of saving led to positive household and business impacts in several instances, with impacts varying greatly depending on the context and target population. Ten studies measure additional welfare effects beyond savings (Table 2). In six studies, reducing the cost of savings helped individuals manage household finances, including increasing household income , spending more on household expenses like education or food , borrowing less from the bank or paying off loans , and becoming less dependent on family and more financially supportive to peers . Especially when savings products are targeted to clients with businesses, e.g., farmers or entrepreneurs, there is some evidence that reducing the cost of, or incentivizing, saving can also facilitate business investments . Effects were quite large across these studies. For example, entrepreneurs in Kenya increased their business investment by 60 percent and personal spending by 38 percent, with even greater impacts for people who actively used their accounts . Despite some of these positive effects, low rates of account usage mean that reducing transaction costs alone has limited effects overall.
For some people, it simply may not make sense to save in a formal account. For instance, individuals may need to spend everything when their income is quite small or to pay off expensive outstanding debt. Surveys demonstrate the relevance of these barriers: over 80 percent of respondents in Uganda and Malawi cited being too poor as a reason for why they did not use their account, even when offered nearly for free . Globally, two-thirds of unbanked adults reported having too little money as a reason they did not have an account.9
A few studies suggest that people may shift savings away from other sources and into a formal account rather than saving more overall, though the challenges of measuring all forms of savings make it difficult to be conclusive. Eight studies attempt to measure total savings, including in formal accounts and from other informal sources, like cash, durable goods, or with friends and family. Only in four studies did access to an account increase overall savings . There is some evidence that increased savings in a formal account may lead to reductions in other forms of savings, like informal savings . Yet measuring savings, as well as financial behavior and outcomes, is inherently difficult such that some variation in outcomes may be a result of measurement differences. Future research should emphasize the need to better understand where increased savings balances come from, either by spending less, taking on loans, shifting from one form of savings to another, or working more. A novel study from Sri Lanka utilized high-frequency surveys to capture the source of participants’ increased savings and found that households indeed worked more .
While banks collect a wealth of administrative data that can be used to evaluate clients’ savings behavior, data from a single bank does not provide a full picture of an individual’s savings behavior. A challenge of all savings studies, administrative data only captures savings at a single bank, therefore not capturing informal savings or savings at other financial institutions. The limitation of administrative data is particularly relevant for low-income individuals, who often save informally, and cautions relying solely on administrative data to understand clients’ financial behavior.
|Country||Intervention Details||Opening Fees||Withdrawal Fees||Maintenance Fees||Minimum Balance||Interest-Bearing||Other features|
|1||Kenya||Financial subsidy: Paid opening fee and minimum balance||N||Y||-||N||N|
|2||Nepal||Financial subsidy: Offered free basic account
Information: Hosted public meeting on savings benefits and account*
|3a||Chile||Assistance with paperwork||N||Y||N||-||-||Can make transactions at retail stores|
|3b||Malawi||Financial subsidy: Paid opening and maintenance fees
Assistance with paperwork
Information: On account
|3c||Uganda||Financial subsidy: Paid opening and maintenance fees
Assistance with paperwork
Information: On account
|4||Sri Lanka||Mobile-linked bank account
Digital: Free basic mobile phone
Financial subsidy: Varied fees for deposits (0%–8%);
Paid minimum balance to open account;
Deposited funds into account to demonstrate how to use it
Financial incentive: Cash lottery (winner paid via deposit into mobile-linked account)
Information: Trained customer service representatives available at call center
Mobile- linked account
|5||Kenya||Financial subsidy: Paid opening and maintenance fees
Information: On account
|N||Y||N||-||N||Free ATM card|
|6||Kenya||Financial subsidy: Offered free ATM card that reduced withdrawal fees;
Paid minimum balance
Information: On account and ATM
|7||Philippines||Deposit collection service||-||-||-||-||Y||A few clients had commitment savings accounts that restricted access to withdrawal; most had a basic account|
|8||Sri Lanka||Deposit collection service
Financial subsidy: Paid minimum balance to open account
Assistance with paperwork
|9||Mozambique||Digital: Provided free basic mobile phone*
Financial incentive: Provide interest on savings paid in fertilizer
Information: Provided information on mobile money and fertilizer usage*
|-||-||-||-||Y||Mobile money account|
|10||Philippines||Financial incentive: Varied interest rate, either 1.5% or 3%
Account owner: Varied offer of individual account, joint account, or choice of individual or joint
|-||-||-||Y||Y||Commitment savings account with lockbox|
|11||Kenya||Financial incentive: Varied temporary interest rates from 0%–10% for six months
Financial subsidy: Paid minimum balance
|N||Y||N||N||N||Choice of opening individual and joint accounts|
|13||Indonesia||Financial subsidy: Offered subsidy ranging from $3 to $14 to open account||N||N||N||Y||Y|
Note: Dashes indicate information was not available.
*All study participants received this intervention, including the comparison group
|Account Opening||Active Usage||Bank Savings||Total Savings||Other Impacts|
|1||87%||41% made at least two transactions within first 6 months||↑||Increased business investment, food expenditure, and private expenditure|
|2||84%||80% made at least 2 deposits within first year||N/A||-||Increased expenditure on education, fish and meat, and festivals and ceremonies; Increased ability to cope with shocks; improvement on index of reported financial situations|
|3a||17%||3% made at least 5 deposits within first 2 years||N/A||N/A||N/A|
|3b||69%||10% made at least 5 deposits within first 2 year||↑||-||-|
|3c||54%||17% made at least 5 deposits within first 2 years||↑||↑||-|
|4||87%–92%||18%–26% used service at least once||↑||-||-|
|5||69%||15% made at least 5 transactions over 2.5 years||↑||-||Decreased borrowing from family;
Increased transfers to neighbors
|6||100%*||22% of opened accounts received at least 1 deposit within first 6 months*||N/A||N/A||N/A|
|7||28%||15% regularly used deposit collection service||↑||N/A||Reduced borrowing from the bank|
|8||89%||N/A||↑||↑||Increased household income, earnings, and loan payments|
|9||N/A||N/A||↑||↑||Increased fertilizer usage and knowledge, irrigation pumps ownership, and household expenditure|
|10||23%||10%–11% made more than 1 deposit||N/A||N/A||N/A|
|11||27%–58%*||16%–44% of opened accounts received at least 1 deposit within first 6 months*||↑†||N/A||Individual accounts increased income, assets, and likelihood of being an entrepreneur|
Note: Upward arrows represent statistically significant positive differences in outcomes between the treatment and comparison groups at the 90 percent confidence level or higher; dashes represent no statistically significant difference.
* Study offered couples the opportunity to open multiple accounts; account opening rate based on all offered accounts and active usage rate based on all opened accounts
† Only for higher interest rates on individual accounts
Abdul Latif Jameel Poverty Action Lab (J-PAL). 2020. "Reducing the costs of saving." J-PAL Policy Insights. Last modified July 2020. https://doi.org/10.31485/pi.2517.2020
Dupas, Pascaline, and Jonathan Robinson. 2013. “Saving Constraints and Microenterprise Development: Evidence from a Field Experiment in Kenya.” American Economic Journal: Applied Economics 5 (1): 163–192.
Prina, Silvia. 2015. “Banking the Poor via Savings Accounts: Evidence from a Field Experiment.” Journal of Development Economics 115: 16–31.
Dupas, Pascaline, Dean Karlan, Jonathan Robinson, and Diego Ubfal. 2018. "Banking the Unbanked? Evidence from Three Countries." American Economic Journal: Applied Economics 10 (2): 257–297.
De Mel, Suresh, Craig McIntosh, Ketki Sheth, and Christopher Woodruff. “Can Mobile-Linked Bank Accounts Bolster Savings? Evidence from a Randomized Controlled Trial in Sri Lanka.” NBER Working Paper No. 25354, December 2018.
Dupas, Pascaline, Anthony Keats, and Jonathan Robinson. 2017. “The Effect of Savings Accounts on Interpersonal Financial Relationships: Evidence from a Field Experiment in Rural Kenya.” Economic Journal 129 (617): 273–310.
Schaner, Simone. 2016. “The Cost of Convenience? Transaction Costs, Bargaining Power, and Savings Account Use in Kenya.” Journal of Human Resource 52 (4), 919–945.
Ashraf, Nava, Dean Karlan, and Wesley Yin. 2006. “Deposit Collectors.” Advances in Economic Analysis & Policy 6 (2): 1–22.
Callen, Michael, Suresh De Mel, Craig McIntosh, and Christopher Woodruff. 2019. “What are the Headwaters of Formal Savings? Experimental Evidence from Sri Lanka.” Review of Economic Studies 86 (6): 2491–2529.
Batista, Catia, and Pedro C. Vicente. “Improving Access to Savings through Mobile Money: Experimental Evidence from Smallholder Farmers in Mozambique.” Working Paper, September 2019.
Karlan, Dean, and Jonathan Zinman. 2018. “Price and Control Elasticities of Demand for Savings.” Journal of Development Economics 130: 145–159.
Schaner, Simone. 2015. “Do Opposites Detract? Intrahousehold Preference Heterogeneity and Inefficient Strategic Savings.” American Economic Journal: Applied Economics 7 (2): 135–174.
Schaner, Simone. 2018. “The Persistent Power of Behavioral Change: Long-Run Impacts of Temporary Savings Subsidies for the Poor.” American Economic Journal: Applied Economics 10 (3): 67–100.
Cole, Shawn, Thomas Sampson, and Bilal Zia. 2011. “Prices or Knowledge? What Drives Demand for Financial Services in Emerging Markets?” The Journal of Finance 66 (6): 1933–67.