In recent months, I've written articles on Investors Bancorp, Inc. (ISBC), Capitol Federal Financial, Inc. (CFFN) and Hudson City Bancorp, Inc. (HCBK), each of which had been MHCs. Converted MHCs have material excess capital. How good are they at deploying it?
What is an MHC? MHC stands for "mutual holding company". When a thrift converts from mutual ownership (owned by its depositors) to stock ownership (owned by shareholders), it can elect to use an MHC structure, in which a mutual holding company controlled by depositors owns the majority of the shares in the thrift, and the thrift sells shares to the public. If the MHC is unwound, the publicly-traded thrift's book value increases significantly and it is left with material excess capital.
Most MHCs are small and illiquid, with less than $10 million worth of average daily trading volume. CFFN is barely over this line, and ISBC is well below it, with about $3 million. In this article, I'll focus on the two large depository institutions that had been MHCs: People's United Financial, Inc. (PBCT) and First Niagara Financial Group, Inc. (FNFG). I'll then make some summary observations about HCBK, ISBC and CFFN.
PBCT - PBCT is based in Bridgeport, CT and has a current market cap of $4.3 billion. It went public in July 1988 and completed its conversion from an MHC structure in April 2007. That conversion boosted tangible book value per share ("TBV-PS") from $4.20 in Q1 2007 to $14.62 in Q2 2007 (tangible common equity rose from $1.25 billion to $4.40 billion), and increased tangible common equity/tangible assets ("TCE/TA") from 10.91% to 32.07%; PBCT went from being moderately overcapitalized to being hugely overcapitalized. Anticipation of this conversion drove PBCT's share price up considerably in prior years. Growth in EPS and TBV-PS played a role, too, but PBCT's price/TBV-PS multiple reached a quarterly high of 5.1x in Q4 2006, too high a price at which to expect a sale of the company, and unjustifiable on a stand-alone basis given PBCT's unimpressive 12%-ish return on average tangible common equity ("RoATCE"). PBCT's share price peaked at $21.80 on February, 21, 2007 (implying a 1.49x multiple of post-conversion TBV-PS) and with the exception of one positive blip, has steadily trended downward since. PBCT closed at $12.13 on September 19.
Prior to the 2007 boost in equity, PBCT's growth in assets was modest, at 0.43% compounded annually over the prior five years. Growth in TBV-PS was good, at 8.3% compounded annually. From 2002 to 2007, PBCT's return on average assets nearly tripled, from 0.47% to 1.18%, driven mainly by increases in net interest margin and decreases in provisions/average assets. PBCT hadn't been repurchasing shares over this period, nor had it been acquisitive.
This latter point changed in June 2007, when PBCT announced its acquisition of Burlington, VT-based Chittenden Corp for $1.9 billion, 55% of which, or about $1.0 billion, was paid in cash. According to the investor presentation for the deal, PBCT paid 4.3x TBV-PS for Chittenden, a bank which had recently been delivering an RoATCE in the neighborhood of 20%. I'm not the first person to recognize that this deal was richly priced.
The deal closed in January 2008, and helped "solve" PBCT's excess capital problem. PBCT's Q1 2008 TBV-PS fell to $11.08 from $15.07 in Q4 2007, a decline of about 26%, and TCE/TA fell from 32.28% to 18.82%. PBCT's $493 million acquisition of Danvers Bancorp, announced in January 2011 and completed in June 2011, was funded 45% with cash. And so at Q2 2012, PBCT's TBV-PS is $8.76, versus $14.62 back in Q2 2007 just after the MHC conversion was completed. Its TCE/TA is down to 11.47%, indicating that PBCT is still overcapitalized. PBCT's TBV-PS has trended down steadily for the last five fiscal years, despite earning a profit in each year.
PBCT currently trades at a 1.38x TBV-PS multiple and yields 5%. The mean 2013 sell-side EPS estimate for PBCT is $0.83, implying roughly a 9% RoATCE for 2013. Given this, PBCT appears expensive on a price/TBV-PS basis. Current short interest in PBCT shares is 4.6%, lower than its February 2011 peak of 7.6% but still well above that of many comparably-sized depository institutions.
FNFG - FNFG is based in Buffalo, NY and has a current market cap of $2.9 billion. It went public in April 1998 and completed its conversion from an MHC structure in January 2003. The conversion boosted TBV-PS from $3.02 in Q4 2002 to $9.15 in Q1 2003 (tangible common equity rose from $203 million to $605 million), and increased TCE/TA from 7.12% to 17.29%. Similar to PBCT, FNFG's stock price increased meaningfully in anticipation of the MHC conversion, with FNFG's quarter-end price/TBV-PS getting as high as 4.1x. In contrast to PBCT, FNFG's share price did eventually exceed the near-term high of $12.37 that occurred on October 15, 2002; its all-time high closing price of $18.90 occurred on September 19, 2008. As an aside, FNFG is no longer a thrift. In March 2010, it received approval from the Federal Reserve to convert from a thrift holding company to a bank holding company.
In the five years prior to the conversion from an MHC, FNFG grew assets by about 20% per year. About half of this growth came from three institutions FNFG acquired for cash. After the conversion, FNFG became highly acquisitive, beginning with the $348 million acquisition of Troy Financial Corporation, announced in August 2003, and culminating with the $1.5 billion acquisition of NewAlliance Bancshares, announced in August 2010. FNFG completed a total of five acquisitions beginning with Troy, comprising $2.8 billion of aggregate deal value. About $570 million of this was paid in cash.
It's worth spending a minute on the NewAlliance deal. FNFG paid 1.65x tangible book value for NewAlliance, or 23x estimated earnings pre-synergies. Its TCE/TA ratio at announcement was 11.11%, and its RoATCE had been in the range of 6-8% in recent quarters. If you assume that a TCE/TA of 8% is normal, and that any excess capital above this should be paid for on a dollar-for-dollar basis, then FNFG paid 1.9x TBV-PS for the "slimmed-down" (without excess capital) NewAlliance, which I estimate would have had an RoATCE of 9-11%. It appears that FNFG paid a generous price.
While FNFG's TBV-PS hasn't fallen steadily post-conversion the way PBCT's has, its Q2 2012 TBV-PS of $5.30 is still below the Q4 2003 high of $9.25. TCE/TA now stands at 5.69%, so FNFG is no longer overcapitalized. Similar to PBCT, FNFG hasn't had an unprofitable fiscal year since its MHC conversion. FNFG closed at $8.25 on Tuesday, September 19, implying a price/TBV-PS multiple of 1.56x. FNFG currently yields 3.9%. With a mean 2013 sell-side EPS estimate of $0.78, FNFG's 2013 ROATCE should be in the neighborhood of 14%, so its current valuation appears reasonable. Current short interest in FNFG shares is 2.4%, considerably below the high of 11.8% reached in March 2011.
HCBK - As I've discussed HCBK before, I'll be brief. HCBK's stock price continued to rise after it completed its conversion from an MHC in June 2005. It grew assets rapidly and also repurchased shares aggressively. Lastly, in 2006 HCBK paid $265 million in cash to acquire Sound Federal Bancorp. The cash expenditures looked smart until HCBK's share price cratered. HCBK's August 24th closing price of $6.44 (its price on the day prior to the M&T deal announcement) was a price shareholders had seen prior to the MHC conversion. With that said, HCBK's TBV-PS has fallen only by about 7% since the MHC conversion was completed, which is far superior to the 40% TBV-PS decline experienced by PBCT and is roughly in line with the decline experienced by FNFG, at least until FNFG's Q2 2012 loss.
ISBC - I'll be brief on ISBC and CFFN as well. ISBC hasn't yet converted from an MHC, and while some knowledgeable parties think it might convert, who knows if and when it will. ISBC does not currently have excess capital. ISBC is currently trading at about 2.1x TBV-PS, despite having an RoATCE in the 10% range, which implies to me that investors are anticipating a conversion. I believe that ISBC overpaid in its recently announced $135 million cash acquisition of Marathon Banking Corp. It might be tempted to overpay for other institutions as well should it convert.
CFFN - CFFN's MHC conversion was completed in December 2010. The conversion nearly doubled TBV-PS, from $5.74 to $12.05, and increased TCE/TA from 11.33% to 20.60%. CFFN's shares had traded as high as 3.8x TBV-PS prior to the conversion. They now trade at about 1x TBV-PS. Strangely, CFFN's high closing share price of $22.14 occurred on September 23, 2008, well before the MHC conversion took place; by the time the conversion was complete, CFFN's share price had fallen to $11.65. Its current share price is $11.97. CFFN has yet to announce an acquisition. It has recently begun repurchasing shares.
The increase in TBV-PS that investors in MHCs realize is a windfall. That doesn't justify it being wasted. Investors who buy shares in anticipation of conversion are paying for the post-conversion TBV-PS. They assign value to excess capital under the assumption that it will be deployed at an attractive rate of return. This can turn out to be a costly assumption.
Banking institutions need to grow TBV-PS for share price to increase. Acquisitions are supposed to make TBV-PS grow, and buybacks can make it grow too, if struck at reasonable prices. Big decreases in TBV-PS are indicative of flawed management strategy, bad execution, or both. And neither management nor shareholders of institutions whose TBV-PS declines can reasonably expect an acquirer to pay a price that pretends mistakes weren't made.
Both PBCT and FNFG saw their CEOs depart in recent years. PBCT CEO Philip Sherringham resigned in April 2010. A Reuters article cited the Chittenden deal and a failure to deploy remaining excess capital quickly enough as two possible (conflicting?) reasons for Sherringham's departure. FNFG CEO Paul Kolkmeyer departed in December 2006. The reasons for his departure were less clear, according to a Buffalo Business First article. But note that both PBCT and FNFG inked sizable deals after these departures. It's hard to view the CEO departures as an admission of prior strategic missteps.
So to answer the question I started with, a depository institution's excess capital may be worth a lot less than face value. Buyers of to-be-converted MHCs felt rich because of excess capital but didn't end up becoming rich. Here's hoping that PBCT, CFFN, and other depository institutions either find good uses for their excess capital or choose to return it to shareholders in the simplest, easiest to value way: by paying higher dividends.