On their blog "House of Debt," Professors Atif Mian of Princeton and Amir Sufi of the University of Chicago postulate that spending growth over the last few years may be unsustainable at best and crash-prone at worst. Their concerns seem to be unfounded.
Mian and Sufi evaluated the contribution of overall spending growth from motor vehicles. Since 2009, motor vehicle spending has far outpaced the aggregate spending on all other goods. Furthermore, while motor vehicle spending only accounted for 1% GDP, it accounted for nearly 15% of yearly GDP growth during that time.
Since most motor vehicle purchases are made on credit, Mian and Sufi conclude that a large percentage of economic growth since the recession ended has been debt-driven. Taken one step further, since the recession was the outcome of over indebtedness, Mian and Sufi hold that the increase in motor vehicle purchases could be the precursor of a second financial crash.
The authors are correct in their assessment that spending growth is coming disproportionately from sectors that are reliant on debt. That, however, is far from a worrisome anomaly. It is the conventional way a recovery occurs following a recession.
Consumption as a Driver of Economic Growth in a Recovery
During a recession, the Federal Reserve lowers interest rates. Purchases that are highly sensitive to interest rate changes (i.e., expensive durable goods such as motor vehicles) become relatively cheaper to buy. The lower financing charges incentivize demand.
What has been an anomaly during the current recovery is that consumer demand has not been nearly as sensitive to interest rate changes as it was during previous recoveries. In other words, the drop in interest rates should have stimulated consumption and general economic growth a lot more than it actually has. This was all discussed in our December 12, 2013, article, "Quantitative Easing Has Failed to Attract Consumer Demand, but That's O.K."
As noted in that paper, we anticipate spending growth to accelerate soon as a result of the natural replacement of aging durable goods. This includes an acceleration in auto purchases to replace an aging fleet. We expect consumers to take on more debt over the next several quarters.
Consumer Debt Levels
The expected buildup in durable goods purchases is unlikely to lead to another recession for the simple reason that households have done a sizable amount of deleveraging.
Since the Great Recession ended, consumers have actively improved their balance sheet by reducing their debt ratios. The household debt-to-income ratio has nearly returned to its pre-housing bubble trend while the household debt-to-asset level is now well below it.
Both of these indicators suggest households can take on more debt and increase spending.
The household financial obligations ratio, which measures the percentage of income required to meet the minimum payments on household debt, is near its lowest level since the Fed started keeping track in 1980. The ratio is currently 2.7 percentage points below where it was in Q4 2007. In other words, a combination of deleveraging and low interest rates has freed up 2.7 percentage points of disposable income for spending.
All of these numbers point toward a consumer that is less risky today than in the previous ten years.
The Lack of Importance of Auto Loans on the Financial Sector
Mian and Sufi are putting too much credence on risky auto loans as the potential starting point for a financial crisis.
There is no doubt that financing companies have lowered their underwriting standards to make it more feasible for a riskier borrower to get an auto loan. Loans to subprime borrowers, which crashed during the recession, are quickly returning to their prerecession levels.
However, the subprime sector still makes up a relatively small percentage of total new consumer auto loans.
There are also no impending signs of a bubble forming in the auto sector. As a percentage of total household debt, consumer auto loans remain well below their historical averages. The auto debt-to-disposable income ratio is at a level seen throughout the 1980s and 1990s.
Additionally, total consumer auto loans - including both subprime and prime - are only a tiny portion of debt held by financial companies, unlike mortgages.
If 100% of subprime auto loans defaulted, financial businesses would only lose about 0.24% of their total financial assets. According to Standard and Poor's, subprime auto recovery rates averaged 39.1% from 2000 to 2008 and rose to an average of 44.1% in 2009 and 2010 from an increase in used vehicle prices. Factoring in the recovery rate, the actual loss from a 100% default of subprime borrowers is only 0.14% or approximately $110 bln of an $80 tln industry.
This is hardly enough to cause a systemic financial problem.
Economic growth in a recovery is disproportionately derived from spending through credit channels. Lower interest rates attract consumption demand and lead to an increase in debt.
It should be no surprise that one of the primary sources of the current economic recovery is the auto sector.
Yet, Mian and Sufi are worried that this could potentially lead to a disruption in economic growth and, possibly, a future financial crisis.
Household debt levels have returned more or less to pre-recession trends. Lending to consumers is less risky today than it has been over the last ten years.
Even if the most risky consumers fail to repay their new motor vehicle purchases, it should have minimal effect on the balance sheets of financial businesses. Subprime motor vehicle purchases are too small in the aggregate to cause a systemic crisis.
Consumers can now take on more debt and increase spending without creating undue financial risk. We expect consumers to increase their purchases of durable goods over the next several quarters.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.