Policy Brief #24

Savings Policy and Decisionmaking in Low-Income Households
November 2010

Sendhil Mullainathan and Eldar Shafir

From a chapter in Rebecca M. Blank and Michael S. Barr (editors) 2009. Insufficient Funds: Savings, Assets, Credit, and Banking Amongst Low-Income Households. New York: Russell Sage.

Download printable version pdf

  • The possible perception among the poor that banks are not for them, may not be far from the truth. Banks' treatment of the rich and the poor is very different: they promote debt (which can be lucrative) over savings (which are bound to be modest) to the poor, as opposed to savings (which promise to be large and lucrative) over debt (likely repaid without penalty) to the wealthy.
  • Though the poor are often implicitly criticized for making use of payday loans and check cashing centers, is paying $20 to get a much needed one-week advance on a $200 paycheck really all that different from the wealthy who routinely pay $2, or more, to withdraw their own cash from ATMs?

Theories about poverty typically fall into one of two camps: those which assume rationality and regard the behaviors of the economically disadvantaged as calculated adaptations to prevailing circumstances, and those which view these behaviors as emanating from a unique "culture of poverty." The first view suggests no need for help, while the second suggests the need for paternalistic guidance. Mullainathan and Shafir propose an alternative perspective, largely informed by recent behavioral research. According to this perspective, the behavioral patterns of the poor may be neither perfectly calculating, nor especially deviant. Rather, the poor may exhibit fundamental attitudes and natural proclivities, weaknesses and biases, that are similar to those of people from other walks of life. One important difference, however, is that in poverty the margin for error is narrow. Amongst the poor, behaviors shared by all often manifest themselves in a more pronounced way, and can lead to worse outcomes (see Bertrand, Mullainathan, and Shafir 2004, 2006). The gist of the behavioral perspective is that while the poor often are not functioning optimally, they are no more in need of behavioral change than everyone else. Instead, it is the interaction of fundamental behavioral proclivities with the context in which they function that produces the particularly destructive circumstances in which the poor often find themselves.

The Behavioral Perspective

Deviating from more traditional economic models, the behavioral perspective advocated by the authors draws from empirical research on judgment and decisionmaking, supplemented by lessons from social and cognitive psychology. Mullainathan and Shafir explore several important insights. First, context matters. A major lesson of modern psychiatry is that human behavior is a function of both the person and the situation. Second, how we construe a situation matters. People act not based on how the world is, but on how they think it is. Third, the actual way people use mental accounting–compartmentalizing wealth and spending into distinct budget categories and viewing cash, credit and debit differently–suggests that a core economic assumption about the fungibility of money often fails. Finally, certain behaviors can be facilitated or blocked by opening or closing a channel, and seemingly minor situational changes can sometimes have large impacts. The implications of this perspective for the financial lives of the poor are explored.

Institutional Factors Affecting the Poor

A fundamental lesson of a behavioral analysis is a new appreciation for the role played by financial institutions. They should not be viewed simply from a financial cost-saving perspective, but instead should be understood to affect the lives of people by easing their planning, facilitating their desired actions, and strengthening their resistance to temptation. For instance, it is well established that defaults can have a profound influence on the outcomes of individual choices (Johnson and Goldstein 2003; Johnson et al 1993). While more affluent workers often have the option of direct deposit for their paychecks, low-wage earners rarely do, and often lack bank accounts at all. Saving is thus the default for the first group who must take active steps to withdraw cash, while cash is immediately available to the poor, who must take active steps to save it. At the same time, many low-income families are, in fact, savers. Without the help of banks, however, their savings are at greater risk and grow more slowly. Thaler (1999) documents the greater tendency to spend cash in hand compared to funds deposited in the bank. Financial institutions also provide implicit planning. Utility companies, for example, typically offer automatic debit for bill payment. This service helps customers to avoid missing a payment. The "unbanked" don't have this option, and thus are at greater risk for late fees, or worse, high reconnect fees. Those who are wealthier can afford to occasionally err or be tempted by impulse spending; the poor on the other hand may pay dearly for even minor mistakes or occasionally succumbing to temptation.

Some Non-Institutional Factors Affecting the Poor

The notion of "economic slack," or the ease with which one can cut back on consumption to satisfy an unexpected need, is central to the lives of the poor. Whereas a rich person can cut back on more frivolous spending, a poor person faced with a financially demanding situation is forced to cut back on essential expenditures. One common finding in the literature is that poor people frequently make late payments, which are often accompanied by penalty fees. These fees can be extraordinarily high: a 5% late fee for a monthly bill is effectively a 100% APR on a loan. In short, not paying a bill is equivalent to borrowing at a very high rate. High-interest borrowing, however, may be the least costly consequence of late payments. What makes lack of financial slack particularly onerous are the indirect consequences. If a phone is cut off, a large lump-sum is typically required to get reconnected. Acquiring this large sum is particularly difficult on an already stretched budget. More importantly, the lack of phone has other consequences, such as limiting the ability to search for work, or be contacted by an employer. Households struggling with a chronic lack of slack are always at risk of becoming ever more destitute. The poor, it turns out, must be better planners than the well-to-do. Similarly, the notion of "buffer-stock savings," or lack thereof, is especially acute among the poor who arguably need it more. Due to their constant and highly volatile struggle with the moment, saving for the future is naturally left until some better time.

Decisions of the Poor, from a Behavioral Perspective

A little over 10% of American households are unbanked, and as such often have no access to credit or formal borrowing instruments. Instead, they rely on often expensive alternatives. This non-participation in the formal sector is often attributed to a rational decision (the small amount of saving is not worth the cost of the account), a hassle-factor (banks rarely locate in low-income neighborhoods), or cultural factors (distrust of mainstream financial institutions or a preference for living day-to-day). A big problem–millions of unbanked households–is explained by big factors. The behavioral perspective suggests, in contrast, that even in the context of big problems, small factors may sometimes play a decisive role. The status quo, bolstered by indecision and procrastination, has a force of its own (Samuelson and Zeckhauser, 1988). The mere perception that banks are intended for people of greater wealth reinforces the impression that banks are not meant for people of lesser means. Moreover, the default option is often different for the poor and the rich. The government could play an important role in encouraging the opening of bank accounts among the poor by encouraging the automatic transfer of tax refunds to bank accounts. This in turn, according to the behavioral view, would lead to positive effects on savings, since a "good" default (a bank account) has replaced a "bad" one (money on-hand).

A fundamental service that financial institutions provide is to allow for the gradual transformation of small amounts of cash, which are easier to come by, into larger sums, which can be hard to attain. The mechanism for transforming relatively small amounts of cash into usably large sums is straightforward according to traditional economic theory: individuals simply save until they have accumulated enough, or if credit is available, they borrow against future income. Since the poor often do not have access to credit, they have to save, but the psychology of planning and self-control suggests that this may be more difficult than traditional theory is prone to assume. Many apparently less-than-rational solutions popular among the poor can be viewed as reasonable alternatives in a world of imperfect planning and limited control. Lottery tickets, for example, which the poor are more prone to buy (Blalock, Just, and Simon, 2007), especially those with moderate ($200 - $500) payouts, can be viewed as "deposits" which eventually lead to a win and the ability to buy an expensive item. The cash is not accumulating and providing a recurring temptation, and the lottery tickets serve as a commitment device, albeit an expensive one.

Payday loans and check cashing centers are other financial vehicles commonly used by lower-income households. Both are costly options that are sometimes used to point out the myopia of the poor. Several basic observations, from a behavioral perspective, can be made about these widespread institutions. A typical payday loan involves receiving an advance on one's paycheck at a steep price (an effective interest rate that can translate to over 7,000% APR). For the credit constrained without a buffer stock, a lack of cash at a crucial time can result in disastrous and mounting consequences. In such a situation, even–or perhaps especially–the farsighted would take out a loan at a high interest rate. From a policy perspective, payday loans may be the lesser evil compared with policies to cap interest rates (and thus drive out payday lenders), unless such caps were accompanied by policies that solve the fundamental lack of a buffer stock among the poor. Similarly, check-cashing centers provide a service that the more well-to-do get for less. For those unable to adhere to a bank's minimum balance requirement, check-cashing arrangements may in fact prove to be the lower-cost option. And even when there are other reasonable options, a behavioral analysis suggests that it is not the mere existence of good alternatives that makes the greatest difference. Rather, there also need to be effective channels. The case of the First Account program implemented by the Center for Economic Progress in the Chicago area provides a good example. The goal of this program was to entice an unbanked, lower-income population that was mostly dependent on check-cashers to open low fee accounts at a local bank. Workshops were held to provide information (mechanics of opening an account, basic bank products, personal budgeting, etc.). Participants in the control group were given a letter of reference to take to the bank, while a smaller subset met directly with a bank representative at the workshop who helped them fill out the necessary paperwork. From a purely economic perspective, the mere presence of the bank representative should have little effect on take-up, as it does not alter the cost-benefit analysis at the core of the decision. From a behavioral perspective, however, this small change had a large positive effect on take-up.

Implications for the Design and Regulation of Financial Services

Our perspective suggests some potential alterations to the way financial institutions for the poor are designed. Several principles are relevant, and they often stand in contrast to classical economic assumptions and common intuition. The first, often underappreciated by program designers, is that information provided does not necessarily constitute knowledge attained. This, combined with the "curse of knowledge"–the tendency of those who know something to overestimate the probability that others do too–can result in underinvestment in outreach programs. Another principle concerns the relevance of people's construal processes. How information is framed systematically alters how it is construed. Marketing, for example, has been used profusely and effectively to target the poor on products ranging from fast food, cigarettes, alcohol, predatory mortgages, high-interest rate credit cards, etc. (see, for example, Caskey 1996; Mendel 2005). Significantly less has been done by marketing firms to aggressively promote more positive options such as healthful diets, various not-for-profit services, and so on.

Second, in contrast with the economic truism that having more options is always good, behavioral research suggests that a greater number of alternatives can increase decisional conflict and overload decisionmakers, leading to deferral, procrastination, or inferior choices (see, for example, Bertrand, Mullainathan, and Shafir 2006; Kling et al. 2008). The case of shopping for mortgages provides a good example. While at one time monthly payment might have been a good point of comparison, in today's market the number and type of loans are so varied that initial monthly payment provides very little information about the true price of the loan, and sellers of loans can, and do, take advantage of this fact (Willis 2006).

Third, existence need not imply availability. Whereas most programs focus on the options that are available, a large behavioral literature emphasizes the importance of channel factors and small costs. Specifically, take-up of a program can be influenced by the perceived nature of these small channel blockages. Related to the notion of channel factors is the distinction between intention and action. Forgetfulness, distraction, and habit, for example, can often intercede to produce observable actions that do not match underlying intentions. This phenomenon is not unique to the poor, but may be more likely to arise in the context of poverty, where superior institutional arrangements are less immediately available. Context-sensitive behavior, in other words, may run counter to people's true intentions; revealed preference fails.

Concluding Thoughts

A fundamental implication of the present perspective is that the poor are neither irrational nor in need of change, at least no more so than the rest of us. Instead, it is the context in which people function–from financial institutions, benefit programs and the design of default structures to the complexity of application forms–that merits careful attention. Taking a behavioral perspective is likely to both enrich and complicate our views of the role of institutions and regulation. If these are founded on a better understanding of decisionmakers, and generate novel policies, it clearly seems worth trying.


Bertrand, Marianne, Sendhil Mullainathan, and Eldar Shafir, 2004. "A Behavioral Economics View of Poverty." American Economic Review 94(2): 419-23.

Bertrand, Marianne, Sendhil Mullainathan, and Eldar Shafir, 2006. "Behavioral Economics and Marketing in Aid of Decisionmaking Among the Poor." Journal of Public Policy and Marketing 25(1): 8-23.

Blalock, Garrick, David R. Just, and Daniel H. Simon, 2007. "Hitting the Jackpot or Hitting the Skids: Entertainment, Poverty, and the Demand for State Lotteries," American Journal of Economics and Sociology 66(3): 545-70.

Caskey, John P.,1996, Fringe Banking: Check-Cashing Outlets, Pawnshops, and the Poor, New York: Russell Sage Foundation.

Johnson, Eric J., and Daniel Goldstein, 2003. "Do Defaults Save Lives?" Science (November 21): 1338-39.

Kling, Jeffrey R., et al., 2008. "Confusion in Choosing Medicare Drug Plans." Working paper. Cambridge, Mass.: Harvard University.

Mendel, Dick, 2005. Double Jeopardy: Why the Poor Pay More. Baltimore, MD: Annie E. Casey Foundation.

Thaler, Richard H., 1999. "Mental Accounting Matters." Journal of Behavioral Decision Making 12(3): 183-206.

Willis, Lauren E., 2006. "Decisionmaking and the Limits of Disclosure: The Problem of Predatory Lending: Price." Loyola Law School Los Angeles Legal Studies Paper 2006-27. Los Angeles: Loyola Law School.

The National Poverty Center is charged with promoting high-quality research on the causes and consequences of poverty, evaluating and analyzing policies to alleviate poverty, and training the next generation of poverty researchers.

The National Poverty Center’s Policy Brief series summarizes key academic research findings, highlighting implications for policy.

The NPC encourages the dissemination of this publication and grants full reproduction right to any party so long as proper credit is granted the NPC. Sample citation: “Title, National Poverty Center Policy Brief #x”.

Major funding for the National Poverty Center is provided by the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services.

Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the National Poverty Center or any sponsoring agency.