Imagine this scenario: Two households with stock portfolios of identical worth each see their investments appreciate by 10 percent one year. The only difference is that one household's wealth grows entirely through capital gains; the other receives its additional wealth in the form of dividends. Which household is likely to spend more, and which one accumulates more wealth?
According to new research by Stefan Nagel of Stanford GSB, the household receiving the dividends is likely to run out and spend the cash. The one with the capital gains is more likely to reinvest or save them. In addition to understanding consumer behavior, the researchers say the information could even be used to draft economic policy.
The differing consumer activity is based on the theory of mental accounting first articulated by Richard Thaler in 1980. He posited that people instinctively group their assets into different categories that they then spend in certain prescribed ways. Mental accounting is the basis for the popular advice to "consume income, not principal." It argues that consumers do not see capital gains and dividends as interchangeable, but rather as two different kinds of funds earmarked for two different purposes.
"Over the years, companies have learned — sometimes painfully — that investors often prefer dividends," says Nagel, associate professor of finance. "It's not particularly rational. When companies decide to do stock repurchases rather than dividends, they are still returning money to investors. But investors don't like that very much." Consumer behavior is consistent with a widely held perception that dividends, unlike capital gains, represent a "permanent kind of income," says Nagel. "They don't necessarily expect capital gains every year, but they do come to expect dividends."
Previous research could not convincingly establish whether the mix of dividends and capital gains influences consumption because it relied only on aggregate data. But Nagel's research — done collaboratively with Malcolm Baker of Harvard University and Jeffrey Wurgler of New York University — was the first to look past aggregate data to how dividends affected consumption.
Nagel and his associates used two data sets that measured dividends and expenditures at the household level. The first was the Consumer Expenditure Survey (CEX), which provided data on expenditure measures and self-reported dividend income and capital gains (or losses) for a representative sample of households between 1998 and 2001. The second data set included the trading records of tens of thousands of households with accounts at a large discount brokerage from 1991 through 1996. Although this portfolio data does not contain explicit amounts of expenditures, it complemented the CEX by allowing the researchers to accurately measure net withdrawals from portfolios.
There are implications for the economy as well as for consumers and businesses. At the consumer level, it's clear that consumers "do have this general rule of thumb of not touching their capital," says Nagel. "In some ways this might not be rational. It might even be sub-optimal for their overall wealth." On the other hand, he says, life is complicated enough. "There's only so much time you can spend thinking about these things," he says. "Economic rules of thumb can work very well for that reason."
At the corporate level, companies are well advised to understand investors' attitudes or risk drawing the wrath of shareholders. And for the overall economy (although the study didn't specifically investigate that angle) "there is the possibility that understanding this consumer behavior could be used to stimulate the economy," says Nagel. After all, if taxes on dividends are cut, and companies respond by paying out more dividends on which investors pay fewer taxes, there could be a significant increase in consumption. "If you're looking to stimulate the economy, there's the possibility that this could have that effect," says Nagel.
On the other hand, because consumption is the flip side of saving, this means that in years when dividends are high, consumers save less. "If an investors' portfolio increases through capital gains, the money tends to stay in the brokerage account; if dividends are paid, it tends to be withdrawn, and not put back," says Nagel. "Thus if the government were trying to promote savings, it could arguably encourage corporations not to give out dividends."