An end to US deleverageing?

ScreenHunter_65 Jan. 18 08.30

From Westpac’s Elliot Clarke today:

In the past two weeks, we have highlighted the current status quo for US credit creation and household income. The take home points on the first subject were that QE has done little to stimulate greater provision of term credit to households and firms, and that a material portion of the increase in corporate credit had been used to alter the capital structure of firms (reduce equity and/or undertake M&A activity) rather than expand productive capacity. On the second topic, it is apparent that household income growth through the recovery has been as much about population and price growth as it has real hourly wage growth, meaning established households’ real income has improved little.

Following the release of the ‘NY Federal Reserve’s Quarterly Report on Household Debt and Credit’, this week we turn to recent trends in consumer debt, specifically student loans and auto financing. Historically, mortgage credit growth has been the primary driver of aggregate household leverage. But, as we continue to highlight, since the GFC, student and auto debt has provided the primary overall positive impetus, partly offsetting the persistent reduction in mortgage debt. Despite a rare increase in total mortgage debt in the September quarter, the combined increase in student and auto loans was still the primary quarterly contributor to household debt, adding $64bn versus $51bn for mortgage credit.

Relative to history, auto loan originations are now at their highest level since mid-2007. Having increased by over 20% since its June 2010 low, the stock of auto credit outstanding ($845bn) is now at the highest level in the history of the NY Fed series, which dates back to the beginning of 2003. The rise in student debt has been much larger: the current outstanding stock of $1.027trn is twice that seen in June 2007 and more than four times the level of March 2003. As by way of a comparison, according to the NY Federal Reserve, total mortgage debt (first and second lien) currently stands at $8.4trn versus a peak of $9.9trn in September 2008 and its March 2003 level of $5.2trn.

The accrual of debt in these two areas of the economy has been positive for growth to date. Auto loans have made car purchases much more accessible to those with less-than-ideal credit health, stimulating robust growth in durables consumption – this is particularly true relative to the weak pace of services consumption growth, itself supported by debt-funded education. The NY Fed reports that the average credit score for consumer credit remains just below 700, broadly unchanged from 2003.

The best way to highlight that this is a poor credit score is by referencing former FOMC Governor Elizabeth Duke’s comments from earlier this year when she noted that prime purchase mortgage origination to individuals with a credit score of between 620 and 680 had fallen by 90% between 2007 and 2012. The point to take from this discussion is that the positive impetus to consumption growth from auto loans is of questionable credit quality and therefore susceptible to a sudden shock. Further, there is a non-trivial probability that some of these loans will turn sour in time should macroeconomic conditions remain as challenging as we suspect.

Whereas the risks to growth linked to the auto loan stock lay in the future, the risks associated with student loans are already becoming clearly apparent. This is best highlighted by the material rise in the proportion of student loans 90+ days delinquent, from around 9% in June 2012 to almost 12% as at September 2013. Further, it bears remembering that the NY Fed has previously noted that a large proportion of students (44% as at December 2012) were not obligated to make repayments and are therefore not included in delinquency statistics.

Looking forward, the impact of student debt accrual on the economy will be heavily dependent on the job and income prospects of the affected cohorts. As we continue to highlight, good job opportunities remain slim, making it that much harder for individuals to pay back the debt without undue hardship and the loss of credit standing.

In a ‘Liberty Street’ blog post that accompanied the latest credit report, the NY Fed characterised the September quarter as potentially being the turning point in the household deleveraging story. While it is true that deleveraging relative to income has slowed and that household debt in dollar terms rose in the September quarter, it should be clear that the underlying sectoral breakdown of debt gives little support to the idea that the US is about to embark on a new leveraging episode. Rather, what is most likely is that total household debt will remain at or around current levels as student and auto debt and poor labour market outcomes preclude younger households from leveraging up further, and, all the while, existing mortgage borrowers continue to pay down their outstanding balances. As alluded to last week, poor income growth will make this process painfully slow.

David Llewellyn-Smith
Latest posts by David Llewellyn-Smith (see all)


  1. “.. what is most likely is that …. poor labour market outcomes (will) preclude younger households from leveraging up further, and, all the while, existing mortgage borrowers continue to pay down their outstanding balances… poor income growth will make this process painfully slow.” Spot on!

    • Janet, that’s probably the first time I’ve seen a an acknowledgement from a bank economist of structural debt (and income growth) issues. I wish they would do the same for our part of the world.

  2. Ha ha ha ha ha ha the kabuki continues unabated! Household deleveraging? Who gives a crap, has the deleveraging in money market funds or OTC derivatives ended?

    Why don’t we focus on the real transgressors instead of bashing households all the time

    • “Why don’t we focus on the real transgressors instead of bashing households all the time”

      Because that stuff only matters to the real economy when we have to bail the b….ds out!