Overview of the research
There is a reason why we freely use the phrase “purse strings” without ever having to stop and wonder what in the world it means. That’s because the idea that some people and not others can pull the strings of a purse (in other words, control how the money is spent) has been used a great deal in research to explain how husbands and wives spend their money. Researchers who generally believe in the ideas of “Resource Theory” have repeatedly shown that husbands earn more money than their wives and therefore get to pull the strings, whereas some have concluded that when a wife earns more she will automatically take control of the money. In other words, whoever earns more, pulls the strings. Fair enough. So we set out to study couples in which the wives earn at least $10,000 a year more than their husbands. In the end, we did not find these women taking control of the money.

Conducting more than sixty extensive interviews with twenty couples in nine states over two years, we set out understand how they spend and save their money. We found that in most of the cases the couples tended to have multiple bank accounts. For example, a couple may opt for three accounts, with the wife having primary control over one, the husband control over another, and the couple together managing a third. This is important because most research so far has assumed that couples have one main account from which most of their expenses are paid. Because of this assumption, researchers have always remained interested in whether the husband or the wife has a bigger say in how money from that main account would be spent. Our study, however, found that when women earn more than their husbands they avoid having to deal with which of the two of them has more power at home. Separate accounts and separate control over them allow both husband and wife to make decisions without having to worry about negotiating power for each and every decision.

The way couples split their expenses across the accounts they controlled often depended on how comfortable they were with the wife earning more than the husband. When a couple was not very comfortable with this idea, all the roles that have conventionally been played by the husband, such as managing the mortgage, were handled from his account, no matter how far apart the two incomes were.

However, when the couple was more comfortable with the idea of the wife’s higher income, the distribution of activities across accounts was more flexible. In all cases, we found, husbands and wives tried very hard to make it appear to themselves and others that they were both almost equally financially important.

Why did these couples behave this way? Were they really that different from couples in which husbands earn more than the wives? To answer these questions we conducted a second study to compare the two types of couples nationally. We found that: (a) husbands were more concerned about the meaning of their income when they earned less than their wives; (b) wives participated less in decisions around the house when they earned more than their husbands; and (c) the couples said their marriages were more stable when the wife earned more.

Significance of the research
These findings indicate that when wives have the higher income, they are possibly more aware that this could cause marital problems so they try harder to maintain balance and a sense of equity. As a result, these couples feel that their marriages are stable and there is less need to change things. Such arrangements include the wife voluntarily playing a lesser role in joint decision-making for fear of appearing to dominate as the main breadwinner. It is possible that many wives find it necessary to do so because husbands generally appear to feel that earning less than their wives puts their self-worth in question. In effect, then, even when wives own the purses, they appear hesitant to pull the strings.

Reference
Commuri, Suraj and James W. Gentry (2005), “Resource Allocation in Households with Women as Chief Wage Earners,” Journal of Consumer Research, 32 (September), 185-195.