I recently presented the first part of my telephone interview with Second Curve Capital founder Tom Brown. Below is the second part of that interview. The comments in italics are from me.

Emil Lee: Can you give me your bull case for Montpelier Re (NYSE:MRH)?

Tom Brown: Montpelier Re is a reinsurance company with an above-average exposure in the areas with the greatest losses in 2004 and 2005. The market is overreacting to these unusual years. Coming out of that, there were a number of changes -- rating-agency changes -- which forced catastrophe-loss modelers to change. A lot of companies are scared and have reduced exposures. Also, primary insurers are looking to increase reinsurance coverage. So it's a very good pricing environment [for reinsurance]. For example, coming out of the heavy hurricane season of 2005, let's [hypothetically] say pricing was up 20% on Jan. 1, [and] by July it was up 30%. The question is: Where is pricing going to be in 2007? It looks like it'll be down a little bit. Also, attachment points are up, so the risk of loss goes down.

Attachment points are the point at which reinsurers are responsible for losses in "excess of loss" contracts. The higher the attachment point, the less likely the reinsurer will have to pay out claims. After huge hurricane-induced losses in 2005, reinsurance supply has gone down and demand has gone up, and that means reinsurers can demand higher attachment points.

Emil Lee: It seems you are optimistic about the fact that Montpelier Re, as well as RenaissanceRe (NYSE:RNR), is writing much more profitable excess-of-loss contracts [in which the reinsurer agrees to pay losses over a certain threshold] versus quota reinsurance [in which the reinsurer shares a percentage of losses]. This is essentially a bet against a high-catastrophe season [since excess of loss gets triggered only in high-catastrophe situations, and quota share pays losses from the start]. This paid off in 2006, thanks to a benign hurricane season, but is this wise in the long run?

Tom Brown: The strategies of Montpelier and RenaissanceRe in 2005 were meaningfully different. Montpelier lost 70% of its book value. Now it needs to eliminate more of its downside risk, and to do that, it's willing to give up some of the upside. In a year of no [major] catastrophes, you typically would've expected Montpelier to achieve a 30% return on equity, but it'll do mid-20s instead. RenRe didn't get hit as much. [RenRe] thinks it still knows the business, so it took a different approach.

If I had to choose, we own a little more of RenRe than Montpelier. Our earnings estimates for RenRe versus Wall Street's are wider than for Montpelier. The Street expects RenRe to earn $8.50 [in 2007]. We think it'll be north of $10 [per share].

Emil Lee: How do you get comfortable with a reinsurer's underwriting discipline?

Tom Brown: They have to show the rating agencies. They tell us about their model, and they show the rating agencies [the specific policies]. We [don't get direct access to their book of policies], but we have support from the rating agencies. If the agencies are not comfortable with the downside loss, they'd downgrade [the reinsurer].

It's kind of like a bank. You talk to the company about its process, but when you talk to them, they all sound pretty good. So you look at the historic record. For reinsurers, you'd look for redundancies.

Redundancies refer to instances when a reinsurer realizes that its past loss estimates were too conservative. In general, it's much more preferable to have a conservative management that errs on the side of caution.

Emil Lee: How much do you look for an insurer with a low loss ratio versus a low expense ratio versus an insurer that has skill investing its float? How important is each aspect relative to the other, to you?

Tom Brown: Well, it's kind of like a balloon -- you push on one side, and it goes out the other. We have to look at all of these factors together and compare our view of the company to its valuation in the marketplace. Most of the time, we find the marketplace's valuation is pretty reasonable, which is why we have a concentrated portfolio.

Emil Lee: How do you value float?

Tom Brown: Every company is different. We look at the earnings potential. In the case of CompuCredit (NASDAQ:CCRT), which has excess liquidity, we ask, What's the earnings impact of buying back stock or making an acquisition, and what's the range of earnings power?

Emil Lee: Your modus operandi seems to be to buy a disciplined underwriter/operator in a beat-up sector when everyone else is scared. How do you identify these types of companies, and how do you avoid value traps?

Tom Brown: To be honest, we've bought crappy companies, too. I made a lot of money personally in the '90s, when banks had significant real estate loan problems. I ended up owning the ones that had the worst underwriting but offered the most leverage to improvement.

Emil Lee: Was this a NAV play?

A net asset value, or NAV, play -- made famous by Ben Graham -- refers to a company selling below its net asset value, or liquidation value.

Tom Brown: If you believed the banks would survive, and make a 15% return on equity and trade at 10-12 times earnings ... [you'd make a lot of money]. It was that sort of upside analysis.

Emil Lee: In the case of those troubled banks, what gave you the conviction that they'd survive?

Tom Brown: You have to look at the capital reserves, the flow of NPA [non-performing assets], and the inflows and outflows. Total NPAs will peak after the inflow. It's helpful to monitor the inflow.

Emil Lee: Can you tell us about the bull case for CompuCredit?

On bankstocks.com, Tom Brown describes CompuCredit as the single most compelling investment opportunity in the financial-services sector.

Tom Brown: CompuCredit's primary business is the issuance of subprime borrowings. [It has] two main products. One is a fee-based card with a $300 line of credit, where it makes more money off fees than the net interest income. The other is at the upper end of the subprime market, where CompuCredit makes more money off the balance with a $1,000 to $1,500 line of credit. That business has grown nicely as a result of portfolio acquisitions, and primarily on the basis of organic growth.

CompuCredit has emerging businesses -- such as payday lending, subprime auto financing, and debt collection. Those businesses are growing wildly, and they're turning profitable. The real wild card is the $700 million of liquidity -- $100 million of cash and $600 million they could draw down. What are they doing to do with that liquidity? If, in the next six months, they don't find anything, my prediction is they'll buy back more stock. Management and the board own 60% [of the company].

Emil Lee: It seems CompuCredit might be underleveraged.

Tom Brown: Tangible equity to assets is at about 33%; 20% is where they feel comfortable.

The equity-to-asset ratio measures how leveraged a company is. In this case, take the inverse -- for a 33% equity-to-asset ratio, we take 1 divided by 0.33, which equals 3. This implies that CompuCredit has $3 in assets for every $1 in tangible equity, which excludes goodwill. A 20% E/A ratio implies $5 in assets for every $1 in equity (1/0.2). In general, as long as it's prudent, leveraging up allows a company to increase its profitability.

You can buy this company at eight times 2008 [Second Curve's estimated] earnings -- plus the excess liquidity. CompuCredit has managed earnings and GAAP earnings. This year, the consensus earnings estimate is around 4 bucks; that's the GAAP estimate. The managed [earnings] estimate, which is what Bloomberg uses, is $3.63 per share. My feeling is next year it'll do $4.50 to $5.00 per share.

In general, the difference between GAAP and managed earnings is that the managed earnings acts as if securitizations stay on a company's books. This gives a more holistic view of the company. Tom advises paying attention to managed earnings.

Emil Lee: It seems as though CompuCredit has a built-in hedge -- when credit is tight, and advance rates on securitization and warehouse facilities dry up, that means favorable pricing for the debt-collection segment. What are your thoughts on this?

Tom Brown: I must be careful what I wish for! [In those circumstances], it'll be the best time for CompuCredit to be making acquisitions, and for the debt-collection segment to be making credit card receivable acquisitions [in other words, buying up bad debt] -- but it'll be the worst time for CompuCredit's stock price.

Emil Lee: Why don't other credit card lenders do this?

Tom Brown: There's reputational risk. With two-year delinquent borrowers, you have to get aggressive. That person may never pay you. Lousy collection practices affect the performing business because of word of mouth and media headlines. Another issue is that large card issuers can book gains [from selling charged-off debt] right up front. It's a way of shoring up weak earnings.

Emil Lee: Why is CompuCredit able to exploit the subprime segment better than its competition? It seems that CompuCredit's 22% net interest margin would attract competition, so is there a long-term competitive advantage?

Tom Brown: Right now, that's one of the thing I like about CompuCredit. To issue Visa and MasterCard (NYSE:MA) cards, you have to be a member of the network. CompuCredit works with three different banks that house new credit card receivables. The banks that are issuers of the cards, such as JPMorgan (NYSE:JPM) and Bank of America (NYSE:BAC), are heavily regulated, and the regulators don't like the subprime business -- that walls off competition. It's very difficult [for competitors] to serve CompuCredit's customers.

Emil Lee: CompuCredit says its loans must make sense for the borrower. Is this warranted, given the seemingly heavy fee structure and 22% net interest margins? Do you see CompuCredit's customers climbing the FICO ladder -- and does CompuCredit retain the customers that are able to improve their credit scores?

Tom Brown: CompuCredit is expending an increasing amount of its effort to retain those "moving up" customers. [CompuCredit] would really like to migrate customers from fee-based cards to balance-driven cards. It has specific rewards programs. For example, if a borrower makes timely payments, [CompuCredit] will increase the line of credit. The program works to increase timely payments.

Emil Lee: CompuCredit has only a 10-year operating history. Is that enough credit cycles to judge the company?

Tom Brown: Senior management came from the collections business, so management has a longer history of dealing with collections. When you're lending money to subprime borrowers, underwriting is more important than the economy.

Emil Lee: How do you judge whether a merger will create real synergies or not?

Tom Brown: First, we have to understand the buyer. Does [the merger] make sense? What [is the buyer] trying to do, and does what the seller wants to do make sense? I go in knowing historically that two-thirds to three-fourths [of acquisitions], from the acquirer's standpoint in banking in corporate America, destroy value. The burden of proof is on the acquirer. We know most [acquisitions] fail, so let's figure out why this is going to fail.

Emil Lee: How do you value the residual value a company retains in securitizations?

Tom Brown: We go through the assumptions -- there's enough information in the trust, such as prepayment [assumptions]. Sometimes, we need to know from management what their models are, the lumpiness with prepays -- [such as] right after graduation -- but there is data to track the actual performance of trusts versus the gain on sale calculation.

Emil Lee: CompuCredit is expanding into other subprime segments, such as auto financing and payday lending. How does this build the company's moat?

Tom Brown: Well, that's what we'll see in 2007 -- which could cause CompuCredit to be a home run. It's testing in a Texas storefront location whether it can cross-sell these services.

Emil Lee: It seems CompuCredit has recently had an uptick in charge-offs and delinquency. Is this a cause for concern? Do you think defaults will stay reasonable?

Tom Brown: I think defaults will go up. The fastest-growing part of its portfolio is the fee-based card, which has higher delinquencies but significantly higher revenue. The fee-based card is more profitable. Overall, it's a positive for profitability.

Emil Lee: In the end, it comes down to this: The stock market thinks CompuCredit is vulnerable to a credit crunch when an increasingly debt-laden borrower starts to default, spurred by things such as a possible housing downturn or a weakening economy. What would your rebuttal be?

Tom Brown: I would respond that underwriting is more important than the economy. CompuCredit is going to a higher level of earnings growth and profitability throughout the economic cycle.

In conclusion
Speaking with Tom was a tremendous and humbling learning opportunity. It was interesting to hear how he has developed his contrarian view, how he looks at the unlocked earnings potential in CompuCredit, and how he has thought through the different angles of his investments. For readers who want to know more, check out bankstocks.com!

Go back to the beginning of this conversation:

Montpelier Re is a Motley Fool Hidden Gems recommendation. MasterCard is an Inside Value pick. JPMorgan Chase and Bank of America are Income Investor selections. Check out any of these Foolish newsletter services free for 30 days.

Fool contributor Emil Lee is an analyst and a disciple of value investing. He owns shares of Montpelier Re and appreciates hearing your comments, concerns, and complaints. The Motley Fool has a disclosure policy.

This article represents the opinion of the writer, who may disagree with the “official” recommendation position of a Motley Fool premium advisory service. We’re motley! Questioning an investing thesis -- even one of our own -- helps us all think critically about investing and make decisions that help us become smarter, happier, and richer.