Policy Brief #14

Using Financial Innovation to Support Savers: From Coercion to Excitement

Prepared from a paper by Peter Tufano and Daniel Schneider

November 2008

Download printable version. pdf


• A wide variety of financial innovations support saving, particularly by low- and moderate-income families. Some already exist, either in the U.S. or elsewhere, and others are ideas that have potential, but thus far, limited implementation.

• Savings programs can be grouped into six categories and vary along a number of dimensions. Some change the decision making process (e.g., literally forcing people to save), others change the time and place of savings (e.g., permitting tax refunds to be transformed into savings), and still others change the cost-benefit of saving (e.g, by bribing savers or making savings exciting).

• The goal is not to identify a single “best” program, but rather to acknowledge that a number of strategies may work under different circumstances, for different people and by different partners. This range of innovations shows a role for government, business, and non-profits to support family saving, often through collaborative offerings.

Families have numerous reasons for saving: education, retirement, a car repair, and most frequently “rainy days.” However, given the societal pressure to consume and the financial realities of many low- and moderate-income families, it is not surprising that household savings is weak. While some are pessimistic about the potential to address this problem, there are a number of possible solutions, many of which show great promise in supporting household saving. These varied solutions emphasize different elements of human behavior and address different impediments to savings. Some require significant government intervention, some simple changes to current regulation, and others are completely market or social network driven. This brief, drawn from the work of Peter Tufano and Daniel Schneider, highlights a variety of different mechanisms to increase that likelihood that people will save.

The Range of Savings Programs

Proposals to stimulate saving vary along various dimensions, one of which is how the innovation affects the decision to invest. Some programs fundamentally alter the decision making process, such as by forcing or mandating savings. Others change the time and place of the savings decision, such as offering saving attached to another product or in an unusual setting. Yet others change the savings deal, altering the costs and benefits of saving by providing financial incentives, social incentives, or psychological benefits. Factors these proposals consider include the ability or motivation of the saver, the process of saving (when and by whom decisions are made), and the available products. Also of importance are the various stakeholders, or entities other than the saver, who are involved, as they often bear some cost, as well as possibly some reward, for supporting family savings. Table 1 outlines the range of potential savings alternatives: coercing saving, making it hard not to save, making it easier to save, providing incentives or bribing people to save, utilizing social support networks to encourage saving, and finally, making saving fun and exciting.

(click image for higher resolution)

Coercing Saving

This strategy is based on the assumption that without government intervention families fail to accumulate adequate savings. There are two ways governments tend to intervene - through programs that force saving by limiting spending (i.e., imposing a tax), and programs that give families additional funds that they must save. Social Security is the classic example of the first type. Workers are legally required to make regular contributions and in return will receive funds later. 1 Britain’s Child Trust Fund is an example of the second type of program. It is an involuntary program that provides savings to families rather than mandating it. Upon birth and again at age 7, every child receives a grant of at least $450 (at 1.8 pounds to the US dollar in 2005), with low-income families receiving more. The grants are in the form of vouchers that parents must invest in one of several types of approved, but privately-run, accounts. Families may add to the funds,which can be accessed by the child upon turning 18. Advocates claim the program has changed the way British families view saving for their children.2 Here in the U.S., a proposal for a similar program is currently stalled in Congress though a pilot program has been initiated by a coalition of non-governmental groups. Though “free” money rarely causes recipients to complain, this type of program is expensive from the perspective of tax-payers, and is inherently paternalistic. Also, to the extent higher income families contribute more than lower-income families, an unintended result could be a widening of the wealth gap.

Making It Hard Not to Save

A slightly less coercive approach is to make it difficult, rather than impossible, not to save. Three main types of programs fall under this category: those where saving is the default, programs that bundle saving with consumption, and those with withdrawal conditions.

Employer-sponsored 401(k) programs, in which employees are automatically enrolled upon hire and must actively opt-out if they choose not to participate, are a perfect example of the first type of program. Relying on the basic human tendency toward inertia, especially when faced with complex decisions, automatic enrollment has been found to increase program participation dramatically, particularly among low-income minority populations. A proposal for a more wide-spread program of this sort would require employers not offering retirement plans to facilitate the auto-enrollment of their employees into so-called Automatic IRAs, which they could then voluntarily opt out of. Interestingly, unless automatic adjustments are built in, the power of inertia also seems to lead employees to maintain their initial, relatively low, contribution levels - even as income rises.

A second strategy that makes it hard to avoid saving involves bundling saving with something consumers are already buying. One example is Bank of America’s “Keep the Change” program in which every debit card purchase is automatically rounded up to the next dollar and the difference is transferred from the user’s checking to savings account. Another example is Upromise, whereby rebates of up to 10% on purchases from designated partner companies are automatically invested in 529 college saving plans. These types of programs seem to have had some success, with enrollment in each growing rapidly.3 At the same time they raise some concerns. Programs like this may encourage people to spend more - good for the sponsoring companies, but quite perverse if the goal is increased saving. Moreover, people may ignore the need for more substantial savings, assuming that they are doing enough.

To the extent that self-control is a barrier to increased saving, commitment savings products like CDs or IRAs with early withdrawal penalties make it harder for people to dis-save. At the same time, there is evidence to suggest that people are willing to put more into retirement accounts with withdrawal conditions if they are allowed to take loans against them. It may therefore be possible to increase saving by adding some commitment to otherwise liquid products and some liquidity to otherwise illiquid products.

Making It Easier to Save

Under this scenario, individuals must consciously decide to save. This can be facilitated by making savings products available when and where people have the opportunity to save: at the workplace, at tax preparation sites, and at retail points of sale. At the workplace, for example, workers could commit to saving a portion of future raises. The effectiveness of such programs comes, presumably, from the tendency of individuals to treat future cash-flow differently than current dollars.4

Similarly, tax refunds seem particularly savable. Preparers could allow filers to precommit to saving their refund, and provide them with access to savings products before the check arrives. Programs run by both H & R Block and at Volunteer Income Tax Assistance (VITA) sites have demonstrated that many low-income filers are indeed interested in opening savings accounts.5 Unfortunately, the filing process itself can be an impediment. Only recently has the IRS allowed refunds to be split between accounts, making it more likely that at least some portion will be saved. Also, since many low-income individuals are unable to open bank accounts, an inexpensive and low-risk alternative is necessary. Savings Bonds would seem to fit the bill, and in fact, until the 1960s, tax filers could buy them directly on their tax forms. Recent pilot programs through H & R Block and VITA sites suggest that this may still be a good option.6

Saving might also be easier if there were not only point of sale debit cards, but also saving cards, with which consumers could “buy” savings. Pre-paid cards currently exist in the form of gift cards and prepaid rechargeable stored value cards which can be used at ATMs. These cards, such as the NetSpend cards, function as payment products for the unbanked. One can imagine a pure alternative that emphasized saving, paying a bit of interest and restricting withdrawals. Such a card would make it just as easy to save as to consume, and a few entrepreneurs are working on launching these savings cards. Alternatively, one could combine savings with existing stored value cards, as is done in the NetSpend National Savings Program which allows prepaid card holders to dedicate a “savings pocket” on the card that earns interest.

Providing Incentives, or Bribing People to Save

Promotions and discounts are two common ways the private sector “bribes” potential customers. Financial incentives can also be provided to help spur low income asset building. The Retirement Savings Contribution Credit, or Saver’s Credit, is a program that provides a tax credit for contributing to an IRA or other recognized retirement account. Unlike most federal tax incentives, which disproportionately benefit higher-income filers, the Saver’s Credit is progressive. However, take-up is currently low and it is not refundable, making its benefit trivial to those with no tax burden. A simplified pilot program administered by H & R Block, however, suggests that the presence of a match (i.e., credit) can indeed raise IRA contributions7, and a Universal 401(k) program has been proposed that would match contributions to retirement accounts through a fully refundable tax credit deposited directly into the tax filer’s retirement account. Individual Development Accounts (IDAs) are another type of matched savings account currently available to some low-income families. The match is provided through a variety of government and private sector sources, withdrawals are restricted to asset-building purchases, and participation often requires case management and financial education.

In addition to creating positive incentives, it may also be possible to increase saving by eliminating disincentives. The most obvious of these are the strict asset tests associated with most public-assistance programs. While intended to ensure that only the poorest of the poor receive assistance, they essentially make it impossible for recipients to have emergency savings, let alone build up wealth.

Making Saving a Group Activity (Utilizing Social Support Networks)

Another way to increase savings voluntarily is by leveraging the psychological and social power of groups, acknowledging that the social context of savings matters a good deal. Rotating savings and credit associations (ROSCAs), and the America Saves! campaign are two examples of peer support encouraging saving. Members of ROSCAs meet on a regular basis, and contribute funds which are then pooled and given to one member on a rotating basis. ROSCAs appear to help people save not only by establishing a group norm of a regular contribution, but also by giving the members a valid excuse for withholding income from demanding spouses or extended family members. While such associations are found primarily in developing countries where formal financial institutions are less readily available, similar associations might also work in this country, particularly in low-income and immigrant communities that are relatively homogenous, isolated, and stable.8

Also relying on peer support, or pressure, America Saves! is a different type of program that encourages individuals to make savings pledges through citywide campaigns that provide education, encouragement, and savings clubs. For low-income families, saving clubs perform many functions, including support, education, reduced demands on the family saver, social reward, and through the pooling of resources, access to financial choices and institutions that might otherwise be unavailable. Existing social networks, particularly tight-knit faith-based groups, have the potential to be ideal settings for such efforts.

One traditional, but often overlooked, opportunity for social savings is to tap into existing family structures. Sales of savings bonds to be gifted to children demonstrate that this old-fashioned motive for savings can be tapped to build savings.9

Making Saving Fun and Exciting

The final approach on the continuum is aimed at getting people who otherwise might not be interested in saving to want to save—not necessarily because it is the “right” thing to do, but because it is fun. The idea of lottery-linked savings goes back over 300 years to 17th century England. Prize bonds, each representing a chance in a monthly drawing, are still offered in the UK and are owned by one-quarter of British households.10 Private financial institutions, particularly in developing countries, also market prize-linked savings products that offer daily prize drawings which are compensated for by somewhat lower-than-usual interest rates, but no risk of principal loss. Given the extensive evidence that lower-income families play lotteries and that they believe they have a better chance at getting rich through playing than from saving,11 prize-linked savings options might be an attractive option in the U.S. as well.

Many people are also motivated by the concept of collectibles (stamps, Beanie Babies, beer labels, etc.). Tufano and Schneider argue that the U.S. could create a collectible savings program in which each increment to savings was marked by a physical object, with the goal being to “collect them all.” By setting concrete, incremental, and achievable goals, the authors note, families will have a physical marker of success that makes it possible to easily track their savings progress.

From Ideas to Action

There are a wide range of potential solutions to the problem of low savings, particularly among lower-income families. Some solutions are best suited to government action (child trust grants, savings bonds at tax time), others to the private sector (collectibles, point of sale savings), and still others to social groups or non-profits (peer support saving circles). If there are so many good ideas, Tufano and Schneider ask, why don’t we see more of them in practice? Virtually all of the examples cited are from the real world, though often on a small scale, or in other countries. In certain circumstances relatively simple changes to existing laws and regulations is all that would be required to expand these programs. In other cases, educating the private sector about these products and how they stand to profit could lead to new offerings (compliance with antidiscrimination laws is often improved, for example, when companies begin offering default 401(k) programs). Programs requiring large government expenditures seem less likely to succeed in the short-run, due to the typical American objection to social welfare spending. This resistance might be reduced if the programs are framed as universal (all families stand to benefit) and restricted to “worthy” purposes (saving for retirement or education). Whatever innovations are considered, more research is needed. For example, what will be the impact of these programs on total savings—will it increase, or will families just reshuffle existing assets? Research will be needed to understand the impact of new options on other financial decisions as well—presumably families should not borrow at high rates in order to invest at low ones. Finally, more information is needed on how much saving and what type is optimal for families, being sensitive to the fact that for low-income families, concerns about short-term emergencies are just as legitimate as the need to plan for a retirement that is decades away.

1. Saving is defined as the deferral of consumption today to enable the use of funds later, whether decades away as in retirement, or just weeks away as when the water heater breaks.
2. White, 2007
3. Choi, Laibson, Madrian, and Metrick, 2004; Madrian and Shea, 2001
4. Madrian and Shea, 2001; Bernartzi and Thaler, 2004
5. H & R Block, 2007a, 2007b; Rhine, Su, Osaki and Lee 2005; Beverly, Tescher and Romich, 2004; Tufano 2008b
6. Maynard, 2007; Tufano 2008b
7. Duflo, Gale, Liebman, Orszag, and Saez, 2006
8. Biggart, 2001
9. Tufano 2008b
10. Tufano 2008a
11. Holton, 2000; Consumer Federation of Americas, 2006

About the authors

Peter Tufano is the Sylvan C. Professor of Financial Management and Senior Associate Dean for Planning and University Affairs at Harvard Business School.

Daniel Schneider is a Research Associate at Harvard Business 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 NPC is directed by Sheldon H. Danziger.

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.