Household deleveraging: Evidence of behavioural change in US households

Following the huge build-up of household debt and leverage before the crisis, economists of all persuasions have been discussing the unwinding of this process. This is the known as household deleveraging and it is of particular importance to the economic recovery of countries like the US and the UK. There is a sense that private consumption growth in these countries will not return to historically normal patterns until this process is complete.

It is not just the prospects for economic recovery that make household deleveraging so important. It is also highly relevant to economic theory and future economic models. The reason for this relevance is the notion that households and consumers changed their behaviour following the financial crisis. In a lot of economic models and theories, consumers and households do not change behaviour and their preferences are held constant. While economists have suspected that households in the US did change their preferences following the crisis, the clearest evidence so far has been provided by a recent research note from the Federal Reserve Bank of New York (FRBNY).

In their Current Issues in Finance and Economics series, the FRBNY have looked at trends in household debt and credit.[1] Emphasising the importance of good data, the FRBNY analyse a new dataset of individual credit accounts to explore the causes of the reduction of household debt by around $1.3 trillion. While they note that some of this has been achieved by a historic increase in consumer defaults, they do find evidence that households actively reduced their debt by paying down loans and reducing new borrowing.

The dataset

The enabler of this analysis is the FRBNY consumer credit panel dataset – a panel of quarterly data on individual and household debt. It represents a 5% random sample of US individuals with a credit file as well as their family members. The credit report data contains detailed information on mortgage loans, such as current balance, scheduled payments and current status. Furthermore, there is data on other types of loans, such as car loans, student loans and credit cards. This allows information to be analysed such as the number of accounts that an individual has, the credit limits and the total balance on each of these accounts.

The findings

On the non-mortgage side of things, there seems to be evidence that consumers have been actively reducing their debt. The chart below, taken from the research note, shows non-mortgage debt changes after stripping out charge-offs. It shows that up to 2009, these loans were growing strongly, but then things changed after the financial crisis. Other than student loans, there were negative changes in 2009 and 2010, revealing that shift in consumer’s reliance on this type of credit.

non-mortgage debt changes

It is a little more complicated to decipher these trends on the mortgage side of the consumer balance sheet. This is due to the issue of foreclosed homes and the fact that the banks resell these to recover some of the lost assets. Without property level data, the researchers are unable to see how much a debt write off from a foreclosed house is recovered with a resale. Nevertheless, they can still look at active borrowing and repayment of mortgages over this period.

The chart below decomposes the changes in mortgage balances into three components:

  1. The dashed blue line shows changes in mortgage debt related to housing transactions, including payoffs of mortgages associated with ‘normal’ sales of houses from one person to another – hence, it excludes changes in debt related to charge offs.
  2. The solid orange line shows charge offs and the effect of the foreclosure crisis.
  3. The dashed green line combines things like cash-out refinances of first lien loans and regular amortizations of first-lien balances.

Decomposition of mortgage changes

Particularly in reference to the dashed blue line representing changes in mortgage debt related to housing transactions, the authors interpret this sharp change as consumers reacting to the economic and financial conditions around them. As can be seen in the overall red line of changes in mortgage debt, consumers were extracting equity and increasing debt up to 2007, but started to pay down this debt after that. Taken with the changes in pattern of non-mortgage debt, the authors claim that this indicates a major change in consumer behaviour following the financial crisis.

A voluntary or forced change in saving behaviour?

The chart below helped the researchers infer an answer to the question of whether this change in consumer behaviour was forced or voluntary. The chart shows the total number of new and closed accounts, as well as credit account inquiries. The blue and green lines are instructive here, showing the account openings and credit inquiries respectively. If the change in consumer preferences was completely forced by changing lending standards, one might expect that the level of inquiries for credit accounts would have stayed the same following the crisis. However, the chart shows that the account inquiries tracks new openings closely, suggesting that consumers have voluntarily reduced their demand for credit.  

new, closed accounts & inquiries

Finally, the authors also present evidence that credit limits were also reduced following the financial crisis. This is an important point, as it shows that the lenders also played a role in reducing consumer credit by reducing the limit that households could borrow.


Changing household preferences for credit obviously has major implications for private consumption growth in the US. With a modest housing recovery seemingly underway, this process of deleveraging may be slowing down and approaching its end.

The implications for economic theory are also very important. What US consumers did not do during the crisis was attempt to smooth their consumption – as unemployment increased and incomes dropped, consumers did not try to increase their credit to maintain a smoothed level of spending. What they did do was change their behaviour in relation to the environment around them. As economists, therefore, we need models that endogenise human behaviour, with the hope that it will help us to understand when consumer preferences will change and how long the change will last for.

[1] Brown, Haughwout, Lee, & van der Klaauw (2013). The financial crisis at the kitchen table: trends in household debt and credit. Federal Reserve Bank of New York.


1 Comment

Filed under Uncategorized

One response to “Household deleveraging: Evidence of behavioural change in US households

  1. Pingback: Modelling heterogeneous consumers: De Grauwe versus Woodford | globalmacromatters

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s