Capitol Federal Financial, Inc. (CFFN) is a Topeka, KS-based thrift with a current market cap of $1.9 billion. It went public in April 1999, under a mutual holding company structure, and completed its second-stage conversion to a fully public company in Dec 2010.
What makes me believe CFFN is so overvalued? Since 1997, CFFN has had a horribly low return on average tangible common equity ((ROATCE)) - always below 10%, and occasionally below 5%. If calculated on an annual basis, CFFN's ROATCE averaged 5.54% from 1997 to 2010. It never got higher than 9.25%. When you see an institution with such a low ROATCE, the usual explanation is high tangible common equity/tangible assets ((TCE/TA)). This is not the explanation for CFFN; its TCE/TA from 2001 to late 2010 was in the 10%-12% range. High, but not high enough to ruin ROATCE.
The real culprit is CFFN's low return on average assets (ROAA). From 1997 to 2010, CFFN's ROAA exceeded 1.00% (roughly the industry average) on an annual basis four times, but its overall average was 0.65%. The reason isn't high overhead expense; CFFN's overhead expense/average assets ((OH/AA)) ratio has consistently hovered around 1.00%, which is extremely low. There are two explanations: the first, more important one is CFFN's abysmally low net interest margin (NIM). In the late 1990s, it had been as high as about 2.50% (which is still very low), but since 2005, it has rarely cracked 2.00%; it fell to 1.28% in Q3 2007. NIM was 2.06% in Q1 2012. The second reason for low ROAA is that CFFN generates virtually no fee income. CFFN's ROAA is not depressed by high loan loss provisions; provisions/average assets were only 0.06% in Q1 2012. CFFN's non-performing assets and charge-offs are low too.
Now, as bad as ROATCE had been through Q3 2010, the corporate reorganization that accompanied the second stage conversion in Dec 2010 made it worse. Common equity increased by $1.1 billion, and without a corresponding increase in shares outstanding, tangible book value per share ((TBV-PS)) increased from $5.74 to $12.05.
That increase in TBV-PS was clearly a win for shareholders, right? Sure; all other things being equal, I'd pay more for a thrift with higher TBV-PS. And so would those who owned CFFN shares in 2010. From 2001 to Q3 2010, CFFN's shares traded at an average price/TBV-PS multiple of 2.6x. Given CFFN's mediocre ROATCE, the only justification for such a lofty multiple was the expected TBV-PS increase from the second-stage conversion. When it happened, CFFN's price/TBV-PS multiple fell to 0.99x.
So CFFN investors assumed that this higher TBV-PS belonged to them, and paid for CFFN accordingly. What was CFFN's plan for all this new capital? The infusion pushed CFFN's TCE/TA from 11.3% to 20.6%. CFFN had grown assets at a compound annual growth rate of only 0.27% (!) in the 10 years ending Q3 2010. So CFFN found it challenging even to lever its old, smaller capital base appropriately. Almost exclusively, CFFN makes senior 1-4 family mortgage loans, not surprising for a thrift. However, loans comprise only 56% of earning assets, and deposits comprise only 61% of liabilities. Both figures are low, and that's a problem. If a thrift must borrow wholesale and buy securities with the proceeds in order to fully lever its equity, that won't be good for net interest margin, ROAA or ROATCE. Hence CFFN's future EPS growth isn't expected to be all that high.
What has CFFN's ROATCE been since its capital base expanded? An average of 3.5% (!) over the last four quarters. Sell-side equity research analysts don't expect it to be much higher than 4% even in 2013.
So what should CFFN do? Return excess equity capital to shareholders, since CFFN clearly cannot earn an adequate rate of return. My view is that a bank/thrift's ROATCE should be at least 14-15%; a 1.00% ROAA and 7% TCE/TA translates into an ROATCE of about 14%. Not all banks are over this line right now, but many are, and a lot that aren't fall short only because provisions/average assets are high. This ratio will fall for these lagging institutions over time, and when it does, their ROAA and ROATCE will improve. Alas, this will not help CFFN, as this ratio is already low.
What's the best way to return this capital to shareholders? I prefer share buybacks to special dividends if and only if a bank's management has good reason to believe its shares are undervalued. Yeah, I know, buybacks generate EPS accretion. I don't care. If I were a CFFN shareholder, I would want CFFN to buy back shares rather than give me a special dividend only if the post buyback share price could be expected to rise by at least the amount of the special dividend. If CFFN's shares are overvalued today, why will EPS accretion push share price even higher?
However, it looks like CFFN will repurchase shares. In Dec 2011, CFFN's Board of Directors approved a new stock repurchase plan, authorizing the repurchase of up to $193 million of common stock, or about 10% of shares outstanding.
Are CFFN shares overvalued? CFFN closed on Wed at $11.64. With Q1 2012 TBV-PS of $11.53, that implies a price/TBV-PS multiple of 1.01x. In my opinion, that's a very high price given the 3.5% ROATCE. Some investors believe that, barring material asset quality problems, banks/thrifts shouldn't trade below TBV-PS, because the bank could always liquidate and pay out TBV-PS, and banks are obviously worth more dead than alive. I am unaware of any bank that has liquidated; I'm pretty sure if liquidation were attempted, it wouldn't be cost-free. If someone came to me with a business plan for a 3.5% ROATCE bank, I wouldn't fund it. And simply calling some amount of CFFN's capital "excess," and valuing it at 100 cents on the dollar, seems crazy.
A lot of capital is trapped inside CFFN, and it is earning a minuscule return. Shareholders aren't indifferent between having this money in their pockets and having it inside the company. The longer CFFN keeps it, the less it is worth.
The shorts recognize the above. Short interest in CFFN shares is currently low, at 1.4%, but had gotten as high as 5.7% in August 2010; CFFN shares had been trading as high as $13-14 around this time.
I can't say that CFFN is the single most overvalued depository institution out there, because there are so many tiny ones with worse performance records than CFFN's. But among the 100 largest public depository institutions, CFFN definitely stands out. CFFN's mediocre operating performance isn't the fault of management, it's caused by too much capital combined with too few attractive lending opportunities in its core business.
Will management admit these problems are beyond its control, and behave appropriately (which to me means paying large dividends to shrink equity capital), or will they reduce shareholder value by undertaking illogical share buybacks and possibly even pursuing acquisitions? Time will tell if CFFN becomes even more overvalued.