Auto ABS Shows Market Weakness

There was some troubling news in the auto sector this week. Ford abruptly fired its CEO, which is never a good sign. But probably more disturbing, Moody's released a report on Santander , one of the largest subprime auto lenders. Due to some new disclosure regulations, issuers of auto loan asset-backed securities (ABS) now have to give more detail about their underwriting statistics. These disclosures showed that Santander only actually verified income on 8% of the borrowers inside an ABS deal they recently bundled. For many investors, this recalled the ugly days of no doc loans during the 2000s housing bubble, which allowed many borrowers to fraudulently buy multiple houses for speculative purposes. That was one of the primary causes of the housing bust. Meanwhile, iShares (a unit of Blackrock) announced it will be launching an ETF backed by consumer ABS -- auto loans included. Could this be a way to bet that the auto lending business will suffer the same fate as housing during 2008? Our firm invests broadly in ABS, so this is a situation we are watching closely. Here are some things to consider. Is Auto Lending Getting Dicey? The answer to this is an unequivocal yes. The question is, how dicey? Over the last five years or so, there have been many new entrants into the subprime auto lending business. Many of these are startups funded by hedge funds or private equity, with no other business but to do auto loans. In lending, it is always and everywhere true that the only way to get market share is to lower your lending standards or provide more favorable terms. So any time there is a lending segment where there are lots of new entrants, you can almost guarantee that lending standards are declining. We are already seeing the results of this, in general. Auto delinquency rates are now higher (seasonally adjusted) than any time since the recession. Logically, that shouldn't be happening right now. Unemployment is at a paltry 4.4%. Wage growth is finally picking up. Based on macro indicators, this should be a time when delinquencies are falling, not rising. Is Santander's Disclosure Evidence of Weak Underwriting? In this case, not really. Auto lending generally is not as exacting as home lending. There are a lot of reasons for this, but two main ones. First, cars are much more liquid than houses. It can take a year or more to foreclose on a house and get it sold, especially during a soft market. Cars are relatively easy to repossess and used cars are sold at large-scale auctions all the time. The second big reason is that auto loans are much shorter-term than home mortgages. If a borrower defaults after just a year on their house, they essentially haven't repaid anything in principal. But in the Santander ABS this week, the deal will have repaid 14% of principal in just year one. That is a lot of protection against losses should a borrower default quickly. In the case of Santander, it generally only require formal income verification (which would involve something like producing a pay stub) on their weaker borrowers. Hence, we do not view this new disclosure as a sign of weaker underwriting standards, rather a formal revelation of their general lending practices. You can argue about whether they should have tighter standards. My perspective is that if a firm has consistent standards, then their historic loss curves should be relevant. If they aren't lending the same way they were in the past, the history isn't relevant. This is what happened during the housing bubble. Loans without full documentation went from extremely rare to quite common. So we couldn't use historic loss rates as any guide about how subprime mortgages would perform. In the case of autos, we have a larger amount of subprime borrowers, which will logically lead to larger overall loss rates whenever this cycle turns south. But if you control for the underwriting buckets, I don't think you should expect subprime itself to perform dramatically worse than it has historically. How Has Santander Performed in the Past? Speaking more generally, if you look at various underwriters, you can see a clear pattern for losses. Santander's deals generally have higher loss rates than AmeriCredit (another subprime lender). DriveTime (provider of those awful nope commercials that have been running lately) has higher loss rates than Santander. If you are an ABS investor, the absolute loss rate isn't very important. It whether enough cushion is built into the ABS structure to protect you against losses. In the case of Santander's new deal, their AAA-rated tranche started with 50.7% hard credit protection. This is the sum of all assets in the pool of car loans that would take losses before that tranche. Because the loans are relatively short term and all principal goes to pay off the top tranches first, that credit protection will rise from 50.7% to 74.1% after one year and will be paid off by the second full year. The A-rated tranche started with 27% subordination, which will rise to 40% after year one and 53% after year two. Full disclosure, we bought the A-rated tranche from this deal and have participated in various Santander deals in the past. Santander's 2007 vintage loans, i.e., the ones written just before the Great Recession, experienced 22%-to-24% cumulative losses over the life of the loans. No BBB or higher rated deal took a loss. More-recent deals have experienced cumulative losses between 10% and 14%. So you could take a pretty big multiple of historic loans and never have these higher-quality tranches get hit. What Does This Mean for the Consumer Loan ETF? I think auto lending will see serious losses in coming years. I suspect that subprime lending standards will deteriorate such that more and more of the total auto lending market is subprime. That will mean total auto losses may be higher than we saw in the Great Recession. But in terms of investment-grade-rated auto loan deals, there is so much cushion that it is unlikely to matter. I'm sure there will be some poorly constructed deals from the weaker underwriters than wind up taking losses. But if you pick your spots and stay up in quality, you should be safe. Shorting this ETF would be a mistake. I don't think you should ignore the weakness in auto lending, though. It could certainly be a negative for the equity of auto lenders -- not just Santander but also Ally Financial , DriveTime, Carmax , and so on. It will also matter for auto manufacturers, parts makers and the like. It is a big issue, I just don't think the issue matters for ABS.

Auto ABS Shows Market Weakness

But even if auto lending losses rise, it won't affect the new iShares consumer loan ETF.

Columnist Conversation

This Weekend on Real Money

Morning Fibocall for 7/10/2020

Fibocall: The INDU and Our LT View

COLUMNIST TWEETS

Please Join or Log In to manage and receive alerts.