Seeking Alpha
Long/short equity, value, debt
Profile| Send Message|
( followers)  

Summary

  • SC’s earnings are obscured by provisions that are much higher than normalized losses. Overtime as actual losses come in and growth stabilizes, better results will surface.
  • Volume growth can offset margin deterioration from cyclical headwinds in terms of losses and funding costs.
  • This write up provides an analysis of the key drivers (volume growth and losses), and concludes that the shares currently trade at attractive prices.

Santander Consumer USA Holdings, Inc. (NYSE:SC) is a subprime auto loan financing company majority owned by Banco Santander. It originates retail installment contracts and leases mainly through franchised dealers as they sell new and used vehicles. In 2013, the company entered into a private label agreement with Chrysler to offer prime and non-prime products under the Chrysler Capital brand. This agreement is expected to result in sharp volume gains in the coming years.

I first looked into this company as a potential short. The non-prime business is full of credit risk, legal risk, and inevitably angry customers. Loan losses and funding costs will almost certainly rise from their troughs, decreasing margins. The industry is highly fragmented and competitive. SC is also expanding into the unsecured consumer lending space, increasing its credit exposure.

What SC has on its side though, is impressive volume gains and solid reserves. The recent dividend suspension created a nice entry point to an already beat up stock price. Will these offset the negatives? Qualitative discussions can only go so far and someone has to roll up the sleeves and quantify. This article tries to do that by identifying the key variables and run scenarios around them.

Identifying Key Drivers of Economics

Table 1 below shows SC's economics for the past few years (as a percentage of average installment loans, ex-leases). In terms of ROA, it's clear that loan loss provisions are key. While funding cost will likely increase in the next few years, it is a lesser concern.

Table 1: Recent Economics

Source: Company filings.

Note that 2013 provisions were way above actual net charge-offs. The trend was even more obvious in Q1 2014, with a whopping 11% provision compared to actual net charge offs of ~6.4%. This is a combination of the company's growth as well as its reserving methodology. During the Q1 2014 call, management explained that provisions were skewed higher due to reserve build from asset portfolio growth, leading to the counter-intuitive result that loan growth actually depresses profitability.

Management confirmed the following question on the call:

"If nothing else changed and your originations went back to kind of what you had previously expected, the profitability would actually increase…"

Parent Banco Santander hinted at a more normal provision in its call:

"The higher provisioning at the Santander Consumer USA resulting from an upfront system of provisions based on expected loss, which has a big impact in periods of strong growth in new lending… Profits should recover quarter on quarter, once the provisions normalize in the SCUSA business."

This has important implications, which I will discuss later. But first, in the next couple of sections I will analyze the two drivers: 1) loan volumes, and 2) losses to arrive at different scenarios assumptions. In all scenarios, I will assume gradual worsening in loan APR and funding costs.

Driver 1: Origination Volume Retained on Balance Sheet

In February 2013, the company entered into a 10-year arrangement with Chrysler, whereby SC aims to provide an increasing percentage of Chrysler financed sales each year. The contractual targets are 31%, 44%, 54%, 64% and 65% for the 12 months ending April 2014-April 2018. This agreement will drive growth in the next few years. As table 2 below shows, assuming U.S. new car sales stay unchanged and Chrysler keeps a constant market share of ~11%, SC can increase its total origination from $19bn in 2013 to almost $35bn in 2017E, an 85% increase just by hitting its contractual penetration targets.

As SC plans to sell most of its prime loans while retaining the subprime loans on balance sheet, it is the latter that is most relevant to economics. Assuming SC sells 90% of prime originations, I estimate loans retained (excluding leases) on balance sheet will increase from ~$14.8bn to $19.4bn in 2017.

Table 2: Origination Estimates

Source: Author's work, based on company filings.

Fiat Chrysler recently unveiled a plan to grow its U.S. market share to 15.8% by 2018E. While pundits have criticized the plan for being unrealistic, it provides a context for an upside case. For the downside case, I assumed that U.S. auto sales will stagnate while Chrysler actually loses market share (down to 9.5%) due to poor products.

The table below summarizes the volume assumptions for U.S. new car sales and Chrysler market share, as well as the resulting loan growth. I expect the company to hit its contractual Chrysler penetration rates in all cases.

Table 3: Volume Assumptions

Source: Author's work, based on company filings.

Driver 2: Analysis of Losses on SDART (Santander Drive Auto Receivables Trust) Securitizations

This section derives an estimate of lifetime cumulative net loss (CNL) for new and recent productions. We can then ballpark net charge-offs given CNL and weighted average life (WAL) estimates. For example 15% CNL spread over two years of WAL amount to 7.5% of net charge offs per year. Loans in the SDART securitizations had roughly 2.2 years of average life and that is the number I ended up using for my scenarios.

For simplicity, I grouped the loan data into three buckets of similar performances: 1) 2003-2005 vintages, 2) 2006-2007 vintages, 3) recent 2012 and after vintages.

Source: Author's work, based on company filings, and Manheim Consulting website.

As can be seen in chart 1 above, losses from recent vintages are still running well below pre-crisis 2003-2005 levels, and nowhere near the 2006-2007 vintages. Delinquencies (chart 2) tell a similar story. Extrapolating the data, I estimated that the 2013 vintage will experience about 15% CNL. This is reasonable as new originations have a higher mix of new vehicles and higher credit scores.

The 2006-2007 vintages frame the worst case scenario. These loans ended with ~23% cumulative net loss, as they were originated during the credit bubble and suffered through 2008-2009 when the Manheim index (chart 3), an indicator of used car prices and thus a proxy for loss severity, went to 1995 levels. A CNL of 23% would roughly translate to ~10% per year of net charge-offs. If provisions are supposed to approximate normalized losses, then management's 9.2% provision rate (LTM) appears to be very conservative.

Given 1) my estimate that the 2013 vintage will experience ~15% CNL and 2) expectations that credit standards will deteriorate and used car prices gradually decline, I expect CNL for new vintages to increase the next few years.

Loss assumptions:

  • Base Case: Recent and new originations will average 16.5% cumulative net loss for the next few years (or 7.5% net charge-offs per year assuming 2.2 years of average life)
  • Upside: 15% CNL (6.8% net charge-offs)
  • Downside: 18% CNL (8.2% net charge-offs)

Focus on Expected Losses as Opposed to Accounting Provisions

As an investor, I'm concerned with what this company is worth, not what the GAAP income statement will show. When an accounting line item does not reflect underlying performances, I adjust it with something more appropriate. As I estimate forward earnings for valuation, I used normalized charge-offs in place of provisions. This is because 1) provisions will artificially obscure profitability in the near term due to SC's reserving policy, as discussed earlier, 2) the vintage analysis above shows that actual losses will come in much lower, and 3) overtime, the two will converge as growth stabilizes. Point 3 may take a while to realize, but the value is still there.

I can do this because the company is not under-reserved. The below table shows that despite its high delinquency rates, SC has great reserve coverage metrics compared to its peers.

Table 5: Reserve Adequacy Vs. Peers

Source: Company filings.

Scenarios and Results

Table 6 presents a summary for the inputs and outputs of each scenario. At the bottom left I used a 2016E P/E multiple for target price calculations. For context, competitors CPSS trades around 7x 2015E earnings (2016E consensus not available), ALLY trades around 12x, and CACC trades ~10x 2015E. SC warrants a premium over CPSS, the closest comparable in terms of exposure to subprime credit risk. This is because SC is much larger, has better funding costs, and has the volume growth to mitigate higher losses.

Table 6: Scenario Assumptions and Results

Source: Author's work, based on company filings.

These earnings estimates are at the high end of analyst range, but then again I'm simply using a normalized charge-off percentage through the time period, while sell side analysts have to try and outguess management on provisions quarter to quarter (whether that number will reflect expected losses or not).

Conclusion

The market is discounting a level of risk that cannot be substantiated by available data. The upside is substantial versus the downside. None of this changes the fact that SC is not for the faint of heart - it's a tough business in a cyclical industry, operating with high leverage. There are also other risk factors not discussed here due to space constraint. However, there's a right price for everything, and an analysis of key drivers tells me the current price is right.

Disclosure: I am long SC. 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.

Source: Santander Consumer USA - The Price Is Right