Have Consumers Been Deleveraging?
Meta Brown, Andrew Haughwout, Donghoon Lee,
and Wilbert van der Klaauw
Since its peak in summer 2008, U.S. consumers’ indebtedness has fallen by more than a trillion dollars. Over roughly the same period, charge-offs—the removal of obligations from consumers’ credit reports because of defaults—have risen sharply, especially on loans secured by houses, which make up about 80 percent of consumer liabilities. An important question for gauging the behavior of U.S. consumers is how to interpret these two trends. Is the reduction in debts entirely attributable to defaults, or are consumers actively reducing their debts? In this post, we demonstrate that a significant part of the debt reduction was produced by consumers borrowing less and paying off debt more quickly—a process often called deleveraging.
The source of this evidence is a special new credit data set [see a recent Federal Reserve Bank of New York (FRBNY) staff report for a description of the FRBNY Consumer Credit Panel] that allows us to look carefully on a timely basis at what consumers are doing with both their mortgage and non-mortgage debt.
First, we can easily see that in 2009 families actively shrank their non-mortgage debt, such as credit cards, student loans, and auto loans. The chart below shows the annual change in non-mortgage debt after stripping out charge-offs. Before 2009, consumers increased their non-mortgage debt obligations each year. But in 2009 they stopped adding to their debt. Their net borrowing (other than mortgages) turned negative (by $13 billion) for the first time since at least 2000. Since consumers had been increasing their non-mortgage borrowing by an average of over $200 billion per year between 2000 and 2007, the recent data strongly point to a change in behavior during 2009. In 2010, households resumed non-mortgage borrowing, raising it $35 billion. But the scale of this net borrowing remains far below the levels of 2000-07.
Second, turning to mortgages and Home Equity Lines of Credit (HELOCs), we can see a similar pattern of more paydowns, although the analysis is more complicated. The complexity arises because it is hard to strip out the impact of mortgage charge-offs.
Unlike what happens with other forms of debt, after a mortgage charge-off and foreclosure, there is typically a house that can be resold, albeit often at a discounted price. So we can’t simply add up what is charged off by some customers without taking into account what happens to the house afterward. Take the example of a borrower who defaults on her $100,000 mortgage, and the lender repossesses her house. The lender then resells the house to a new buyer, who pays $80,000 for the property, making a 20 percent down payment and financing the remaining $64,000 with a thirty-year mortgage. So, although the borrower defaulted on a $100,000 loan, the net change of total mortgage indebtedness from this series of events is only -$36,000 (=$64,000–$100,000).
Thus, in order to focus on the active borrowing and repayment behavior of mortgage borrowers, we break down the change in balances into three categories. The first two reflect the buying and selling of houses and foreclosures, while the last one measures the behavior of consumers outside of these transactions.
- Changes in mortgage debt related to housing transactions, shown in blue in the chart below, include the payoffs of mortgages associated with the “normal” (i.e., outside of foreclosure) sale of a house from one owner to another, and the opening of new first mortgages for the purpose of buying a home, whether it is for sale by its previous owner or a lender. As expected, this series fell sharply between 2007 and 2009 as the value of housing transactions declined. In this calculation, we exclude the reduction in debt attributable to charge-offs.
- For convenience, we show the gross value of charge-offs as the red line. Here we see clear evidence of the foreclosure crisis, as charge-offs on mortgage debt exceeded $1 trillion between 2007 and 2010. Note that the 2010 figure is preliminary.
- Our main series, in green, shows the combined impact on debt of cash-out refinances of first liens, changes in junior-lien balances, including HELOCs, and regular amortization of first-lien balances. While first-lien amortization reduces balances at a fairly steady pace, the other components have declined sharply since 2007. We interpret this component of balance changes as indicative of consumer responses to economic and financial conditions. While consumers were on average extracting equity and increasing their mortgage debt until 2007, they have started to pay down debt since then. Between 2000 and 2007, consumers increased their indebtedness by an average of $130 billion per year. In 2008, this series turned negative. Consumers paid down $140 billion in mortgage debt in 2009 and $220 billion in 2010.
Taken together, the new mortgage and non-mortgage credit data we have at the New York Fed indicate that consumers have changed their behavior in ways not limited to missing payments and defaulting more often. Between 2000 and 2007, consumers’ borrowing added an annual average of about $330 billion to the cash they could spend; by 2009, consumers were diverting $150 billion away from potential spending in order to reduce the debts they had built up. This represents a remarkable $480 billion reversal in cash flow in just two years.
We care about this change in behavior because it affects American families’ balance sheets and because it affects macroeconomic patterns like those that we track as part of our policy responsibilities here at the New York Fed. The switch toward paying down debt is probably reflected in restrained growth in aggregate consumption in the United States. But the link between debt paydown and spending is complicated, depending among other factors on exactly who is paying down and who is increasing debt, a point emphasized by a recent working paper from Gauti Eggertsson and Paul Krugman.
So, U.S. consumers have been deleveraging. Holding aside defaults, they have indeed been reducing their debts at a pace not seen over the last ten years. A remaining issue is whether this deleveraging is a result of borrowers being forced to pay down debt as credit standards tightened, or a more voluntary change in saving behavior. There is evidence on both sides of this question. For example, the Fed’s Senior Loan Officer Opinion Survey indicates that credit standards were tight through much of 2007-10. On the other hand, the reduction in housing and stock values over the same period may have led families to want to reduce their debts, in an effort to restore their net worth. We hope to discuss these questions in more detail in later posts. Meanwhile, we will continue to post updates of household credit conditions on our website.
Data for the above calculations are here:
The views expressed in this post are those of the author(s) and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the author(s).