54002 Finance & PSD Impact MAY 2009 The Lessons from DECRG-FP Impact Evaluations ISSUE 5 The fifth in our impact series presents results on a careful evaluation of a financial literacy program in Indonesia. Despite the recent hype of the benefits of financial literacy programs in both the media and policy-making circles, this research is the first systematic and scientific evaluation of one such program. Valuing Financial Literacy: Evidence from Indonesia Bilal Zia Financial literacy has come to play an financial literacy may also have lower levels of increasingly prominent role in financial education, be less interested in financial reform in both developed and developing matters, be poorer, or have different discount countries, and is portrayed in global policy rates. circles as a panacea for many recent crisis- related financial ills. Many developed and A randomized experiment finds no impact developing countries, have set up special of financial literacy training for the Financial Literacy Panels that are charged with general population developing financial literacy programs for the To estimate the causal relationship general population. between financial literacy and demand for But do financial literacy programs actually financial services, we conducted a field work? That is, do such programs enable experiment in Indonesia alongside the households to enter the formal financial household survey. This enables us to system? A new nationally representative investigate whether financial literacy programs survey coupled with a randomized experiment do lead to greater financial access. in Indonesia provides a first step towards We study use of one of the most basic, answering these questions. but perhaps most valuable financial services, savings accounts. Half the unbanked Financial literacy is strongly correlated households from our survey were offered free with use of financial services, savings and financial literacy education, focusing on the retirement planning. costs and benefits of opening a bank savings A compelling body of survey evidence account. These literacy seminars are 2-3 hours from developed countries shows that in length, are group based, and are offered at a households with low levels of financial literacy convenient village location. tend not to plan for retirement, borrow at Although take-up of the financial literacy higher interest rates, acquire fewer assets and program is high, the results of the experiment participate less in the formal financial system show that the program has no effect on the relative to their more financially-literate likelihood of opening a bank savings account counterparts. for the general population. Our study, conducted in collaboration with the World Bank Jakarta office, is the first The financial literacy program only has a such comprehensive survey of financial positive effect for uneducated and literacy in the developing world. In Indonesia financially illiterate households. we find that financial literacy is indeed a Although the effects of the financial powerful predictor of demand for formal literacy program are negligible for the general savings, loans and insurance services. population, we do find that ex-ante uneducated Yet, as with any observational study, it is and financially illiterate households do benefit certainly possible that other factors explain from such programs, in that providing such some or all of these observed relationships. households with financial literacy education For example, individuals with lower levels of does induce them to open formal bank Do you have a project you want evaluated? DECRG-FP researchers are always looking for opportunities to work with colleagues in the Bank and IFC. If you would like to ask our experts for advice or to collaborate on an evaluation, contact us care of the Impact editor, David McKenzie (dmckenzie@worldbank.org) accounts: the likelihood of unschooled and Specifically, based on our current research financially illiterate household heads opening results, the value of the financial incentives a bank savings account as a result of financial serve as an upper bound on how much should literacy training is 12% and 5%, respectively. be spent on financial literacy programs if the goal is to improve banking access. For our Financial incentives have a much larger study, we find financial incentives are more effect than financial literacy programs for than two-times more cost-effective than the the general population. financial literacy program in inducing As a secondary experiment, we provided financially illiterate households to open bank randomly selected households a small savings accounts. financial incentive (ranging from US$ 3 to $14) if they opened a bank savings account. Policy Implications We find that even the smallest incentive 1. Financial literacy programs should be payment has a positive effect on the carefully targeted. likelihood of opening bank savings account The greatest bang-for-the-buck is when for the general population. financial literacy programs are targeted towards households that are uneducated or Does this mean financial literacy financially illiterate. Careful attention should programs simply do not work? be paid to targeting financial literacy programs While our results show a very weak causal towards such populations. link between greater financial literacy and better financial access, our metric of 2. Focus on alternative means of measurement may be somewhat imperfect. improving access, such as providing Indeed, better financial literacy may make financial incentives. households better financial planners, make We find that demand for bank accounts is them save more informally, and may lead to highly sensitive to small financial incentives. improvements in welfare that are not captured This is consistent with the observation that in our study. banks in the United States offer cash gifts or In order to measure the external benefits presents to those opening a new account, and of financial literacy programs, we are suggests that efforts to reduce the price of repeating our research exercise in India with a financial services, for example through detailed follow-up survey which will allow us encouraging competition, may be effective in to measure such benefits. facilitating financial development. Another aspect to consider is that perhaps a more comprehensive financial literacy 3. Little is known about the consumer program is necessary, one delivered over protection aspect of financial literacy several weeks rather than a few hours. This training, so more experiments are needed. kind of in-depth and prolonged focus may be The aspect of financial literacy training that necessary to change behavior among has attracted most interest in the current households. financial crisis is its potential to protect It is important to keep in mind, however, consumers from undertaking opaque and that longer courses will be more expensive to risky transactions. Little is known about the provide, and getting people to attend the full efficacy of programs designed to protect course will require additional logistical consumers in this way. resources. Doing so may not be the most efficient use of available financial resources. For further reading see: Cole, Shawn, Thomas Sampson and Bilal Zia, "Financial Literacy, Financial Decisions, and the Demand for Financial Services: Evidence from India and Indonesia", Harvard Business School Working Paper 09-117. Recent impact notes are available on our website: http://econ.worldbank.org/programs/finance/impact Impact Note 1: Which Microenterprises have high returns to capital? ­ David McKenzie Impact Note 2: Does Business Registration Reform increase entrepreneurial activity? ­ Miriam Bruhn Impact Note 3: The Promise of Index Insurance ­ Xavier Gine Impact Note 4: Why is More Capital not enough to Grow Female Businesses? ­ David McKenzie