Prize-linked savings accounts are financial products that combine standard savings accounts with lottery-like features and the chance to win significant prizes. They exist in a number of countries, including Germany, Great Britain, Sweden, Indonesia, Argentina, Mexico, and Kenya.

The authors conduct an online experiment in the United States to determine whether prize-linked savings accounts can be used to increase savings, especially among households with low income. They find that offering prize-linked savings accounts leads to a significant increase in savings that is financed by a decrease in both lottery expenditure and current consumption.

Low savings rates and insufficient household liquidity in a sizeable percentage of the population are among the more pressing public policy issues in the United States. Typical proposals to tackle the problem focus on mandating savings, providing financial incentives in the form of public subsidies, or changing the way in which saving decisions are framed.

Prize-linked savings accounts are an additional attempt to encourage savings. They have the advantage of being entirely market based, and they do not require any public expenditure. Once the appropriate legislation is put in place, financial service institutions can use the product to reach new market segments.

When assessing the market potential of prize-linked savings accounts, it is important to understand why such a combined product might be more attractive than the existing “pure play” savings account or lottery. The authors suggest two psychological explanations. First, they argue that prize-linked savings accounts may permit greater moral licensing because they are perceived as a savings product and not as gambling. Second, they refer to the anchoring and adjusting heuristic, which explains why consumers evaluate bundles differently from the underlying components. In the case of prize-linked savings accounts, consumers anchor on one feature—for example, savings—and then imperfectly adjust for the other component. As a consequence, the perceived value of the bundle exceeds the perceived value of the components.

The authors report results from an online experiment in which subjects were asked to make a series of choices on the allocation of a $100 budget. Subjects were recruited from two online panels. The first panel has a representative sample of the US population, whereas the second panel has a disproportionately large sample of individuals with low income and low savings. For the first panel, there is a 10% chance that respondents received a payoff based on one of their decisions. Otherwise, they are paid a fixed participation fee. All choices are completely hypothetical in the second panel, and participants receive only a flat fee.

In the first set of choices, respondents have the option of (1) receiving cash, (2) putting their money into a savings account that pays principal and interest 10 weeks later, and (3) purchasing lottery tickets. Money is allocated in increments of $20, and variations of the interest rate (high, medium, low) and the odds of winning the lottery (good, fair, bad) give rise to nine decision situations. These allocations provide the baseline for assessing the attractiveness of prize-linked savings accounts, which are introduced as a fourth option in the second set of choices. Variations in the odds for the prize-linked savings accounts give rise to a total of 15 decision situations.

The authors conclude based on the sample used for this study that the introduction of prize-linked savings accounts leads to an increase in total savings by 12 percentage points on average. The increase is most pronounced for participants with the lowest reported savings and is financed by a reduction in both lottery expenditures and current consumption. The authors also find that the results across the panels with actual and hypothetical payoffs do not differ significantly. They conduct a series of robustness tests, and the conclusion remains the same.

“Savings with a Thrill”—the byline chosen for the reintroduction of prize-linked savings accounts in Great Britain in 1956—adequately describes the character of the product. Because the principal must be preserved and the lottery component is thus limited to a redistribution of only the theoretical interest payments, prize-linked savings are much closer to a savings product than a lottery. Assuming that one major motivation for buyers of lottery tickets is to “win big,” the considerably smaller payoff and/or odds offered by prize-linked savings accounts are likely to be a disappointment, even in the presence of moral licensing or imperfect adjustment.