Talk of an approaching down-cycle has been increasing for some time, and the fundamentals remain under close scrutiny by all market participants. Like many, I think that a review of the last cycle, its impacts, and considering what worked and didn’t work to protect lenders last time is worthwhile. Equally as important is the need to distinguish what is different today that will make the next down-cycle different. We certainly wouldn’t send a SWAT team to handle a fire, so let’s make sure that we know exactly what we are up against before drawing up war plans.
Market round-up
Several predictive metrics have already set the table for a potential recession. In summary, the three-month yield curve is inverted, along with including the two-year and 10-year curves representing grander time frames which economists consider. GDP growth is slowing, and manufacturing is contracting (as indicated by the ISM Manufacturing PMI of 47.8, down from 4.91 in the prior reading). On the positive side, consumer spending remains strong, and unemployment remains at a 50-year low of 3.5 percent. The Fed just dropped rates by 25 basis points on October 30, 2019 which provides more support to the US economy. Close eyes remain on a trade deal with China and the EU/Brexit, which represent risks, which if successfully overcome could represent a softer landing, or in the best case – an avoidance altogether. Most economists agree that if we do hit a recessionary period that it will be shorter and less profound than the Great Recession. However you slice it, guessing whether we have a recession, when it will occur are open items that will continue to elude us until we can look at it clearly in the rear-view mirror.
Historical view of the Great Recession
The Great Recession, and its most poignant impacts on auto were:
• Toxic assets – in particular, verifications of customer provided stability information (earnings, employment, and durations) took a back seat to the desire for more volume.
• Loose credit – extending loan terms in concert with higher LTVs and PTIs taken in concert with lackluster verifications.
• Intense competition – this would be the driver behind the accumulation of toxic assets and loose credit.
• Shake out of banks and finance companies – banks fled the scene entirely and have done nothing since to signal re-entry, while many FinCo’s were completely shook out.
So, what’s different this time around?
Since the Great Recession, finance companies have happily picked up the volume that was prior captured by banks, so much so in fact that the non-prime market has continued to expand. From where I sit, I have observed some changes that will make this next cycle different. Many of these changes are economic and controls-based or operational on the part of lenders, and the main differences I see are:
• Improved verifications limiting toxicity – manually conducted lender verifications have definitely cleaned up, and lenders and dealers alike are well aware of the risks associated with trying to speed through verifications. While overstated incomes still come in for funding, it is my belief that across the industry, lenders are holding a much stronger line. The result? Much less toxicity on balance sheets.
• A focus on fraud – Taking the prior point a step further, there has been a significant advancement on the part of even smaller, less sophisticated lenders on implementing technologies that reduce through-the-door fraud including: dealer introduced fraud, fraudulent income and employment, and power-booking. For the fraud that does get through we are seeing lenders take legal action directly against dealerships, and this intelligence is being continuously fed back into the protective solutions that are out there – a nice cycle of “plan, do, check, act.” Bottom line – lenders are uniformly interested in battling fraud, and that equates to less toxicity on the balance sheets as well.
• Capital & ABS markets rolling along – both Fitch and the NY Fed indicate some points of caution with upticks in 60-plus delinquencies, however ABS issuances continue to drive a solid pipeline. While the sources of capital have moved, there is still tremendous demand for automotive assets. Investor demand behind the capital flows remains strong.
• Competition remains high – as indicated by the 2019 Non-Prime Automotive Financing Survey (produced by the National Automotive Finance Association), and S&P’s Subprime Outlook delivered by Jonathan Smoke at the 2019 National Automotive Conference – competition is driving to profit compression, and is driving some consolidation.
In summary, there are tangible and substantive actions that are supporting the underlying quality of the market, and should help in the event of a recession. This is not even mentioning the proliferation of leasing and longer-term rental options available to consumers, along with ride-sharing that provide more flexible options to consumers that either did not exist or were not as mature in the prior recession.
The brass tacks of being prepared
Assuming that the underlying asset toxicity has been largely addressed, credit standards hold strong, and competition doesn’t drive pressures on those dimensions to change, the impacts of a recession will be less profound than the last time around. Still, we will continue to keep an eye on delinquencies, and any other aspects of the consumer, structure, and collateral to ensure that we are prepared for whatever comes. I would like to highlight what I see as the most substantial actions that lenders can take to ensure that they can weather a potential storm:
• Enhance your credit program – the biggest opportunity that I see are with lenders that are operating in the absence of a statistically-driven credit model. Now, there are absolutely judgmental underwriting shops that do well and are able to make proactive and protective program enhancements. But, the basis by which these program enhancements are made should ultimately be tied back to the most powerful indicators of risk. For example, a lender may think that by limiting LTV, PTI, or term, that they are “strengthening” each of their dealer call-backs. Unfortunately this can often result in sinking your capture rate and ultimately loan volume at a time when you don’t want to contract your portfolio. A better way to approach it would be to tie program enhancements to the actual indicators that drive risk within your program, which may not be associated so much with deal structure, but more about customer stability. By attaching to the most powerful determinants of risk and not focusing solely on deal structure, lenders can improve loan quality while growing their portfolios. Additionally, what generally comes along with a statistical model is the ability to more easily make tweaks over time as historical vintages age and that data becomes available to incorporate into the model.
• Improve collections efficiencies – modest enhancements to how we approach collections can have substantive impacts on outcomes. In the above example about enhancing credit programs, the name of the game is to deploy a program that triggers purchasing decisions based on lenders’ historical loss experience. I like to think of behavioral collection models in a similar vein – I think of them as an extension of the existing credit program that brought the deal onto the balance sheet, but now appended with actual customer payment performance. I mean, you apply a model to the initial decision, it makes an awful lot of sense to apply a model to sort rank the portfolio based on likelihood of roll, and use that to direct your operational efforts to maximize collection dollars and limit the flow through to loan defaults.
• Take advantage of competitive gaps in the narket – the market for secondary contract (i.e. whole loan) sales may start to show some volatility, and produce price drops or spikes that will have beneficiary market participants. Back in 2009, when I started Pelican Auto, I was able to acquire loans with very deep discounts and high yields that I have not seen in today’s market – and that may change. To me, this all represents a world of opportunity provided the lenders that pick up volume are well informed in their decision on how and where to grow their portfolios.
So now, we wait
It has been my overwhelming experience and observation that the non-prime lending market is resilient and willing to make the necessary enhancements to ensure the longevity of the industry. The changes that I have seen since the Great Recession give me great confidence that this is exactly what is going on – from capital partners, through lenders and even consumers. We have seen the entire community address the largest area of toxicity with great seriousness: fraud. I believe that if we keep the pressure up on fraud that the next categories of risk to attack are credit and portfolio risk. I would love to see the entire non-prime lending community make substantive improvements to their ability to manage credit and portfolio risk in a way that provides options to consumers while continuing to deliver more stability to lenders so that we can weather down-cycles without significant fall-out that we saw in the prior recession.