One could object that it is aggressive to use a 12x P/E when NICK has traded at a single-digit P/E multiple for many years. But as I’ve already stated, my opinion is that NICK’s past valuation is irrational. So in the interest of consistency I do not permit this objection. Besides, disagreement about the proper valuation of the company is why we are here in the first place. If you want the market to tell you what everything is worth all the time, buy an index fund.

In considering how to value NICK, I start with the knowledge that non-depository financial companies have historically traded at a discount to the market as a whole. Financial companies are cyclical, leveraged, commoditized businesses. They are also often dependent on the capital markets for ongoing funding, in the form of credit lines, warehouse lines, securitizations, debt offerings, and so on.

The future regulatory environment is also unknown, but almost certain to be stricter going forward than it has been in the past, thanks to the CFPB. And investors dislike regulatory uncertainty. (My personal belief is that the ultimate effect on the subprime auto lenders is likely to marginal, and that NICK is as well positioned as any lender in this regard.)

With all that in mind, here are a couple of ways to think about valuing NICK.

Private market value

A couple of notes: first, TFCE was distressed when it sold itself, so that sale is probably not indicative of what a healthy company will sell for. Second, the VFC and Carfinco acquisitions were platform acquisitions—the acquirers bought these companies because they wanted to expand their subprime auto lending operations, and they were looking for pre-existing foundations upon which to build. As I’ve said, NICK is unlikely to be acquired as a growth platform. As such, 15-18x earnings is likely too high a multiple for NICK.

So from a private value perspective, perhaps we can say that 7x is too low, and 15x or greater is too high.

Aborted Sale to PSEC

In early 2014 NICK agreed to sell itself to another public company, PSEC, for $16 per share, before ultimately pulling out of the deal.

NICK’s LTM EPS at the time of the announcement was $1.36 (under the old share count). Excluding one-time costs, it was about $1.50. So the P/E ratio, based on the $16 deal price, was a little under 11x. NICK’s trailing twelve month EPS—after adjusting for the reduced share count and increased interest expense from more leverage—is about $1.90. Putting a multiple slightly below 11x on these earnings gives us an implied value of $20 per share.

The P/E method is simplistic, however, as it does not account for the excess capital NICK had at the time of the proposed sale to PSEC.

One way to account for this excess capital is to use enterprise value instead of market capitalization. Enterprise value is not traditionally used to value financials, but I think it has some utility here because of NICK’s changing capital structure.

In addition, I think we are well-served to consider the value of NICK’s loan book as well as its earnings, because the loan book is less influenced by short-term factors than are a single year of earnings.

Here is how the PSEC sale looked from an EV/Portfolio perspective:

Now, what if we apply this same valuation to NICK today?

Different method, but thanks to the lower share count, the same result: $20 per share. Today. Quite a bit higher than the current stock price.

But I contend that the PSEC deal undervalued NICK.

The sale to PSEC was, in a sense, distressed. The original founder Peter Vosotas, and his original investors, the Mahan family, owned almost 20% of the company—far too much to sell economically in the open market. My understanding, from speaking to investors and a number of people in the industry, is that when Vosotas left Nicholas, he wanted out. Getting his capital back quickly took precedence over getting it back at full value. So, I think the deal from PSEC represented a low bid aimed at a very motivated seller.

My other reason for believing the PSEC bid undervalued the company is simply the implied earnings yield. At $20 (the price implied by the PSEC offer) today, an acquirer would be getting a 12.5% mid-cycle earnings yield based on $2.50 of mid-cycle EPS. That is just too high a return for an established business with fat margins, a strong position in a profitable niche, and growth ahead of it. Especially in today’s world of such low interest rates.

Public Market Value

(I’ve excluded Santander Consumer here because it has a number of businesses other than subprime auto lending, and because it has a number of off-balance sheet assets and liabilities, which muddies any comparison of balance sheet metrics across companies.)

For these businesses, the best measures of earnings quality are ROA (which I’ve derived here using taxed EBIT instead of net income—which I used earlier in the report—in order to strip out financing costs) and Assets/Equity. These metrics show, respectively, the net yield of the loan book, and the leverage used to create the earnings.

If you have a high ROA, you are finding the best loans and you can tolerate spikes in loan losses or financing costs. If you have low leverage, you are more likely to weather the storm when capital becomes difficult to find.

CACC has the industry-leading ROA, and very low leverage. As such the market values it highly, at 14x earnings (and 1.9x EV/receivables).

CPSS, by contrast, has a bad ROA, and the highest leverage. As such the market values it pathetically, at 5x earnings (and 1x EV/receivables).

SC, which isn’t in the table above, is valued similarly to CPSS. It has a better ROA than CPSS, but similar leverage.

If we look at ROA and leverage, NICK’s profile is much closer to CACC’s than CPSS’. Yet based on P/E and EV/receivables, the market is valuing NICK as if it were CPSS. This is irrational, pure and simple. NICK is a business of far, far higher quality. NICK should trade closer to CACC than CPSS. NICK’s high margins mean it is one of the industry’s low-cost producers, and its low leverage mean its earnings are much more stable then CPSS’ or Santander’s.

*             *             *

You should have enough information at this point to reach your own general conclusions about NICK’s earnings power and what an appropriate multiple is for those earnings. Personally I keep it simple and use 12x. That’s where NICK traded prior to the crisis, that’s below where CACC has traded, and that’s where a lot of financial companies have traded historically.

There are some drawbacks to NICK’s business, but also some great qualities. The company’s financial profile, when approached purely quantitatively, is extremely attractive.

In the next and last post I will discuss some of NICK’s near-term growth opportunities, as well as the CFPB.