Federal Reserve Bank of Philadelphia
10/05/2026 | Press release | Distributed by Public on 10/05/2026 08:57
Why Couples Overspend Even When Both Partners Are Rational Savers
ost economic theory treats households as if they're run by a single decision-maker with unified preferences. But anyone who's ever shared finances with a partner knows that's not quite right. As research has consistently shown, household members have different preferences and care more about their own consumption than their partner's. What happens to our understanding of a household's saving and spending decisions when we take that reality seriously?
To find out, economic advisor and economist Andrew Hertzberg of the Federal Reserve Bank of Philadelphia developed a model that reveals a surprising problem: When both household members have perfectly rational, standard preferences about the future, the household as a whole becomes "time-inconsistent" - that is, it systematically spends too much today and saves too little for tomorrow. Hertzberg reports his findings in his recent working paper, "Time-Consistent Individuals, Time-Inconsistent Households."
The Core Problem: A Commons Tragedy
His key insight is that shared household savings creates what economists call a "dynamic commons problem." When two people share a bank account, each person can unilaterally decide to spend money on themselves. Since each member cares more about their own consumption than their partner's, they both have an incentive to take a bit more than they should from the shared pot. Even though both members would agree in advance on an ideal savings plan, once they're actually making day-to-day decisions, each has an incentive to spend a little more on themselves at the expense of shared future savings.
This problem gets worse when household members are more self-interested and when they place less value on household "public goods" such as housing, appliances, and especially children. (The shared concern for children helps align interests because both parents value spending on kids.)
Hertzberg calculates how much wealth a household would give up in exchange for access to a device that makes them stick to their optimal plan. For a household with moderate self-interest and moderate concern for public goods, this device is worth about 3.5 percent of total wealth. This amounts to a substantial inefficiency.
The Separate Accounts Solution (Sort Of)
One obvious solution would be for household members to keep their money in separate accounts. This helps, but it doesn't eliminate the problem. Even with separate accounts, household members remain connected through altruism and joint consumption of public goods such as housing and children.
These connections mean that wealthier members may voluntarily transfer money to their partners in the future, or both members may jointly purchase public goods. Anticipating these future scenarios re-creates much of the original problem. If you know your partner might help you out later, or that they'll contribute to buying something you both want, your personal incentive to save today is reduced.
The behavior of households with separate accounts depends heavily on the distribution of wealth between members. When wealth is relatively equal, both members contribute to public goods, each accounting for what the other will contribute. When one member has significantly more wealth, that person may become the sole provider of public goods. And when the wealth gap becomes very large, the wealthier member may actually transfer money directly to their partner.
Interestingly, the wealthier member sometimes transfers money early in the relationship as a commitment device. By reducing the expected wealth gap in the future, this makes the poorer member less likely to overconsume, which ultimately benefits both partners.
It Matters Who Controls the Money
A practical implication of this research concerns government transfer programs. Many such programs have found that giving money to different household members produces different outcomes. This paper offers an explanation for why this is so.
When households save separately, the marginal propensity to consume - defined as how much gets spent versus saved from an additional dollar - depends not just on total household wealth but also on who receives that dollar and how the wealth is distributed. In some scenarios, giving an extra dollar to the wealthier member increases total household consumption more than giving it to the poorer member because it increases the likelihood of future transfers and thereby reduces both members' incentive to save.
This has implications for policy design. If a program aims to increase immediate consumption, targeting the wealthier household member might be more effective. Conversely, if the goal is to increase savings, targeting the member with less wealth could work better, although this effect holds only within certain wealth distributions.
Choosing Between Joint and Separate Accounts
When would rational household members choose to pool their money despite the overconsumption problem? The answer involves weighing risk-sharing benefits against savings discipline.
Joint accounts provide insurance against individual income shocks. If one partner loses their job or has unexpected medical expenses, they automatically share in the other's resources. Separate accounts provide only partial insurance through voluntary transfers. On the other hand, separate accounts improve each person's incentive to save.
The model predicts that households will choose joint accounts when they face larger risks to relative wealth and when their initial wealth levels are relatively equal. This matches survey evidence showing that roughly half of married U.S. households share all financial wealth while the other half maintains at least some separate accounts.
Real-World Relevance
This framework helps explain several real-world phenomena. For example, it explains why a rule requiring spousal consent for any withdrawal from a retirement savings account increases household saving. (This rule has been mandated by U.S. law for most retirement accounts since 1984.) The framework also sheds light on why commitment savings devices - such as rotating savings and credit associations in developing countries - are popular: They limit one spouse's ability to raid shared savings.
The paper's insights extend beyond married couples to any household arrangement where members share some resources, consume some public goods together, and care about each other imperfectly. For example, it helps explain why extended families might undersave: When adult siblings anticipate helping each other, they face a similar incentive problem even when they maintain separate finances.
Importantly, this time-inconsistency problem doesn't require any individual to have irrational preferences or poor self-control. Both household members can have standard, rational preferences about the future. The problem emerges purely from the strategic interaction between imperfectly altruistic people who share resources.
Hertzberg's paper suggests that household financial decision-making is considerably more complex than standard economic models acknowledge. Simple interventions - such as changing how accounts are structured, who receives transfers, and whose approval is required for withdrawals - can have a meaningful effect on household welfare by changing the strategic environment in which members make decisions.
Federal Reserve Bank of Philadelphia published this content on October 05, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on October 05, 2026 at 14:57 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]