Most people have a pretty good idea how much money they make each month, and at least a rough idea of how much they spend.
A new study suggests they are probably wrong on both counts. And the source of this insight is as interesting as the findings.
A new research institute created by JPMorgan Chase looked at income and spending patterns of 2.5 million Chase customers between October 2012 and December 2014 and found that both income and spending varied widely from month to month.
About 41 percent of households experienced at least a 30 percent variation in monthly income during the survey period. Fully 60 percent of the households experienced a 30 percent variation in consumption.
Never miss a local story.
“When you survey people as to what their income and consumption are, they don’t pay any attention to the fact that it’s not a stable number,” said Diana Farrell, the lead author of the study and the head of the new institute. “They give an average.”
The implication is that people aren’t actually aware of the monthly variation in their cash flow, and so many aren’t prepared for a bad month. A family might find, for example, that it could not pay the mortgage in a given month even though, on average, the household’s income is enough to pay its bills.
It is hardly news that many Americans lack adequate savings. A 2011 study reported about a quarter of U.S. households could not come up with $2,000 even if given a month to do so, while another 19 percent said they would need to visit a pawnshop or a payday lender.
But the new study offers a precise accounting of the shortfall. It broke households into five income brackets. The median household in each bracket did not have enough liquidity to cover the plausible range of variation in income and spending for even a single month. (The conventional advice is to maintain a rainy-day fund sufficient for three to six months.)
In the middle bracket, those households with incomes between $40,501 and $63,100, the median household had $3,000 in liquidity. Farrell and her co-author estimated that such households should have about $4,800 a month.
These are pretty basic calculations, but they are the kind that could not be made from data previously available to researchers. This detailed information was known only by you and your bank.
“Having that kind of data on individuals is potentially a really useful thing,” said Wade Pfau, a professor at the American College of Financial Services. “This seems like exactly the kind of research question a big bank could answer.”
Farrell, previously an economic adviser to President Barack Obama, said she saw the new institute as an opportunity to answer questions about the workings of the economy by tapping into the bank’s vast collection of data.
“The recession exposed a lot of questions around how the economy really operates,” she said. “Understanding the complexity, the interconnectedness, the sheer scale of the economy today would benefit from granular, on-the-ground information.”
The study aimed to identify households that bank primarily with Chase - those with at least $500 in monthly deposits, five monthly withdrawals and a Chase credit card – and then supplemented the bank’s records from other sources. Like other research that taps “big data” – vast collections of individual records – the aggregate numbers do not reveal the details of any individual household’s transactions.
There’s already a lot of evidence that income volatility has increased. A 2012 study by Karen Dynan of the Brookings Institution and two co-authors found income volatility increased by about 30 percent between the early 1970s and 2008 as fewer workers maintained steady employment at a steady wage.
The surprise in the new study was the volatility of consumption. The study found consumption is not particularly synchronized with income. Households, in other words, are quite likely to experience months in which income falls to a relative low even as consumption hits a relative high.
There are caveats to the conclusion that households should save more. Money kept in savings earns next to nothing. Every dollar not producing something is a lost opportunity.
Moreover, even if people are underestimating their liquidity needs, it’s likely some of the variability does not come as a complete surprise. Payments to the government spike in April. Spending on consumer goods peaks around Christmas. Workers paid weekly intermittently get five paydays in a month.
The study does not show how often households are actually caught short. But it does suggest that thinking about monthly cash flow as varying within a range, rather than aiming at an average, may be helpful to avoid that fate.