Fears that auto loans may be the next big bubble are beginning to permeate the market. The levels of care loans that have been securitized are now about $200B more than during the heights of the subprime bubble of 2006.
About 1% of car loan applications in the US contain some form of false information. That’s the same rate as existed on home mortgage applications in 2009. Point Predictive, a fraud prevention firm based in San Diego, CA, put together a group of 13 car lenders to share data in an effort to prevent fraud. The group determined that car loan fraud is often perpetrated by both consumers and car dealers.
One question that has yet to be addressed is whether Uber and Lyft could be responsible for the huge amount of fraud that is taking place on auto loan applications.
In 2016, Uber secured a $1 Billion loan from Goldman Sachs to underwrite subprime auto leases to its drivers. Much like with mortgage lenders in the run-up to the financial crisis, it seems that the meteoric rise of ride-sharing services could be fueling unscrupulous lending. Uber and Lyft are repeating history. They’ve realized that a good business isn’t good enough and that a good business that is levered to the hilt, becomes a great business.
Many people in urban centers who would not normally own and operate cars are now buying cars like never before, because they expect to make a living driving for Uber and/or Lyft, or because they believe that occasional ride-share driving will pay for the monthly payment and other expenses for the car. The rush of buying of automobiles, in order to allow people to bet on this thesis, which is commonly promoted by Uber/Lyft and many members of “Big-Auto” lobby has a big role fueling the auto loan bubble.
In many ways, an auto loan bubble is much worse than a mortgage bubble. The primary reason for this is that auto loans are collateralized by assets which depreciate, not appreciate over time. In the case of a typical mortgage-backed security, the underlying collateral typically consists of a home or other piece of real estate, which on-average (and I cannot stress ‘on average’ enough) is expected to appreciate and outperform inflation over time. With an auto loan, foreclosing on the underlying asset is a much less favorable outcome, and will often result in the recovery of pennies on the dollar of the initial loan value.
Much like in the mortgage bubble that popped 2008, it seems that the pricing of risk for auto loans is drastically mispriced and that once again, this underpricing of risk is fueled by an unusually low interest rate environment. Traditional risk models that model a given risk versus the risk-free rate struggle to work correctly when the risk-free-rate variable sits at or next to zero.
If interest rates stay low, markets stay calm, and Uber and Lyft continue to run at an eerily consistent rate of growth, then the auto loan bubble will continue to inflate. But if interest rates rise and the economy slows, it’s my belief that the auto loan bubble pop in a way that causes serious economic damage to those in the bottom income quartile in the US (particularly immigrants and young people residing in urban centers), and Uber and Lyft will almost certainly be parties which will take the blame, much as Countrywide and other mortgage lenders were blamed (if only after the fact) for the carnage of 2008.