Does Capital One Deserve Valuation Multiple of Bigger Banks?

"Real estate bottomed months ago."~ Jim Cramer, CNBC

First, some background. For the past several months, we've been hearing various media pundits declaring that the worst is over in the US real estate sector -- this even as anecdotal reports from the credit channel indicate that fundamentals are continuing to deteriorate. Indeed, the divergence of opinion on the housing sector is so wide as to remind us of the late 1990s, when a growing chorus of disapproval questioned tech stock valuations, even as the true believers refused to acknowledge the realities in the marketplace. The song remains the same.

The FDIC Quarterly Banking Profile released in November reported that "the amount of loans and leases that were noncurrent (90 days or more past due or in nonaccrual status) increased by $3.4 billion (6.9 percent) during the third quarter. This is the second consecutive quarter that noncurrent loans have increased, and it is the largest quarterly increase since the third quarter of 2001."

Nowhere is the battle over the condition of the US real estate sector more intense than in the financial media. Last week, for example, we witnessed CNBC commentator Larry Kudlow and his free market sidekick, Art Laffer, dispute the notion that US real estate was on the verge of a serious correction and, even if it is, that a real estate deflation would effect the nation's economic growth.

Even as Nouriel Roubini of NYU's Stern business school reported that more than a dozen subprime mortgage lenders had closed their doors in the past year and that the subprime sector was in a state of "collapse," Kudlow and Laffer refused to concede that the speculative craze in US real estate finally might be at an end and may hurt the economy in 2007. Check out the comment Roubini posted in his blog (www.rgemonitor.com) on January 9th:

Sub-prime mortgages have been a crucial element of the real estate boom of the last few years. While in 1994 only 5% of mortgage originations were subprime, in that last two years 20% of all mortgage originations have been subprime. Note also that interest-only and payment-option ARMS – many of which are to subprime borrowers - were 2% of loan originations in 2000; while by 2006 they accounted for 40% of loan originations. So what happens in the subprime market is of great importance for developments in the housing market in 2007.

Meanwhile in Barron's, we observe another indication of the widening gulf between housing bears and bulls. The usually prescient Archie MacAllaster of MacAllaster Pitfield MacKay in NY, opining in this week's Barron's Roundtable, recommended that investors put their money into Capital One Financial (NYSE:COF), the $150 billion bank holding company which just completed the acquisition of North Fork Bank. MacAllaster waxed effusive on COF, saying that "it is deserving of a higher multiple -- say the 12 times earnings a lot of banks sell for. People are starting to look at it more as a bank."

But another Barron's Roundtable regular, Scott Black of Delphi Management in Boston, was having none of it and called COF "an accident waiting to happen" because of the high proportion of subprime loans. Noting that most of COF's mortgage portfolio is "Alt-A," meaning alternative documentation subprime paper, Black disclosed that Delphi had been a holder of North Fork, but chose to sell rather than take the COF paper into the portfolio.

Frankly we have to side with Black when it comes to COF, not just because the subprime real estate market seems to be melting down, but because the subprime loan business just does not deserve the same multiple as other banks, in any economic environment. Looking at the COF asset mix prior to the close of the North Fork transaction at the end of Q3 2006, what you find is $100 billion in assets with a gross loan yield of over 1,000 basis points, some 3.68 standard deviations ("SDs") above the large bank peer group, using data from the FDIC and calculations from the IRA bank Monitor.

The ROA of 3.5% and ROE of 20% for COF are likewise stellar and well-above peer -- several SDs above peer -- but this is precisely why we feel that MacAllaster is wrong to compare COF with other banks, say Bank of America (NYSE:BAC) or Wachovia Bank (NYSE:WB). Note, however, that the portfolio of COF rates out to a "B" because of the 236bp of defaults in the first nine months of 2007, a default rate that is 6 SDs above peer.

Below is a table showing the accumulated gross loan yield, default rates and weighted average maturity for COF and a number of larger BHCs at the end of Q3 2006. Notice that none of these organizations, including Citigroup (NYSE:C) and HSBC (NYSE:HBC), have default rates or loan yields even remotely comparable to those of COF.

Like non-bank lenders in the subprime sector, the reason institutions like COF have such high returns is to build a capital cushion for when the economy slows and mortgage loan default rates rise -- a lot, some times double digit percentage defaults. COF's capital to assets stood at 14% at the end of Q3 2006, but gross defaults were 570bp in 2004 and 480bp in 2005. If the current trend continues, gross defaults for 2006 should be just over 310bp, 35% below 2005 but still many times the average of the larger BHCs.

Now Archie, answer this question: does COF's 10% plus gross loan spread make up for the fact that the gross default rate is 4x peer? We don't think so, especially if the real estate market is truly headed for a correction.

If you are going to peer COF with larger bank holding companies, the best choices in our view would be C and HBC, both of which sport large subprime lending portfolios and, as a result, lower PE multiples than less risky franchises like BAC or WB. While COF may indeed be among the best of the subprime lenders, at least in the banking industry, to us that means it deserves a lower PE multiple to compensate investors for the higher risk.

Indeed, if Nouriel Roubini is correct about the trend in the subprime sector in 2007, COF may have already seen its high market valuation for the year, directly contrary to what MacAllaster suggested in Barron's when he noted that the 52- week range of COF was $69.30-$90.04. COF closed at $79.00 yesterday in NY. Archie, hit the bid.

Comments

[FIG trader_:]

For many unfamiliar with the COF story, the above "sophisticated" (though curiously selective) analysis may make sense. But anyone covering the stock for the last ten to twelve years knows there is much more than meets the eye. I'm not sure where Scott Black got has info, but COF's mortgage portfolio prior to the acquisitions was minimal. True, the credit card portfolio may be less than "prime"', though it has been generating 20%-plus ROEs for many years (as is auto finance, many fewer years). The complexion of the company has changed, but the valuation (especially circa $70 where a previous strategy of mine posted on seeking alpha of going long COF/short AXP) reflects expected lower ROEs, substantially less than that expected by the banks. (The acuisition at substantial premium to book redced the ROE substantially,but will get back up to mid to high teens by 2008) The stock trades at 1.4 x book for X-sake, the lowest valuation (exclsuive of a few dips such as post Q2 earnings and again back in 2003). The rocket scientists crunching #'s there are equal to none in the business. Fairbanks has been shrinking the risky stuff (domestic card book), lengthening the maturity of the portfolios (credit cards are short, mortgage s are theoretically longer). They have been waiting for this accdent to happen for long time (Jim Grant too), but somehow the company keeps getting back on its feet, outperforming everybody in its class, who has to get acquired (or spun off KRB, , Discover, PVN, ). Long COF/short AXP will still work exceptionally well at least until the price/book-ROE equation (multi-year outlook) is equalized, assuming AXP doesn't have "accident" like last time. remember who was buying junk bonds for AXP advisers????

2007 Jan 23 12:29 PM

[FIG trader_:]

The author was comparing apples and oranges; the the higher loan spreads and default rates are almost exclusively due to different portfolio characteristics than that of "peers" mentioned. How can the respected author fail to notice. Ultimately, higher rewards corresponding to higher risk; Nothing revealing there? If anything, the company (pre merger) could have been compared to the BAC credit card subsidary (MBNA), JPM's or C's credit card . (Credit loss and chargeoff numbers are remarkably close for credit card subs of majors) Post-merger, the banking comparisons will at some point be valid, but only to the extent they are valid for any two banks. Are C and BAC, and WFC and WB comparable? They are comparable to the extent that they are institutions leveraged (8/10 to 1), have some mix of retail/consumer, commercial, investment banking. The company "pressed the bet" on the credit card side when they could do so with good expected returns (Nineties) and backed off (i.e.went into more retail/international banking/ and auto) this decade. It seems very sensible strategy in a higher risk credit card environment in US. In fact, the auto play also was well timed (see captives GM and F under financial stress) . No numbers on ROE's but it is not hard to imagine they are/will obliterating these two under financial stress. Surely, Fairbanks is not a magician, but who best to strategize and minimize risk while maximizing ROE. its 10% to 15% average annual return will still be near top over next five years.

2007 Jan 23 03:59 PM