Description
I am recommending a short of Brookline Bancorp (BRKL), an overcapitalized, underperforming regional bank based in the wealthy Boston suburb of Brookline, Mass. The thesis is based on the bet that no bank CEO would buy out Brookline at the current price (despite the end of a regulatory moratorium on BRKL’s selling and the CEO’s age of 69) after spending more than two minutes on an analysis of buy versus build. I know it is somewhat risky to bet on the rationality of an acquiring CEO, but Brookline is big enough—any deal would be over a billion dollars—that relatively few banks could actually buy it. To put a finer point on it, I am betting that Citizens Bank (owned by RBS, which has a PE multiple of under 10 times) will not buy Brookline with its 30+ PE. However, in the unlikely event of a takeout, I believe much of the premium is already priced in the stock, making any further pop on the announcement unlikely.
As one note before you read any further, although I have provided an analysis on Brookline’s business and why I think earnings estimates are too high and of low quality, this is all largely irrelevant to the thesis, as the price is so far removed from fundamentals that missing or beating estimates by a penny or two doesn’t matter. The most important thing to remember is that no bid will materialize over the next six months. More likely is the announcement of a CEO search, which is my catalyst to propel the stock downward. While I doubt the stock will ever trade at its fundamental value of $9, the two year chart demonstrates the formation of a strong downward trend within its $16-14 trading range. Short-termers can ride it down for at least 10%.
Valuation
Before any valuation discussion, I would note that BRKL completed its second step conversion from a mutual holding company (MHC) three years ago. Most converted MHCs tend to trade at high valuations. Although I admit that I can’t make sense of many of these valuations either, BRKL’s valuation is high even compared to these stocks, more than the prospect of any takeout warrants. The best comps for Brookline include Hudson City Bancorp (HCBK), New Alliance (NAL), and Rome Bancorp (ROME), all of which recently completed second step conversions:
Price/ ROE Price/ Mkt Cap/ Mkt Cap/ Mkt. Cap
05 PE 06 PE Tang. Bk. (tangible) Sales FTE ($M) Branch ($B)
BRKL 39 31 1.7 4-5% 17.2 5.4 66.7 1.0
ROME 24.8 20.7 1.0 4-5% 7.3 1.0 33.3 0.1
HCBK 24.7 19.4 1.3 5-6% 13.9 6.4 84.5 7.1
NAL 27.5 22.8 1.6 5-6% 7.8 2.0 26.2 1.7
HU 15.7 14.7 4.2 31% 4.3 1.0 9.3 1.9
NLY NM NM 1.3 11% NM NM NM NM
RBS 9.5 8.8 3.6 32% 2.4 NA NA 94
As you can see, BRKL has valuation multiples roughly 50% higher than its peers. At $16+, BRKL trades at 39 times consensus 2005 earnings of $0.42 and 31 times 2006 consensus of $0.52 (although as noted below, consensus figures are probably too high). On a price to tangible book basis, the stock trades at 1.7 times, which sounds reasonable at first glance except when you consider that its annualized return on tangible equity was only 3.9% in the second quarter, should only improve to 5% for 2006, and will remain chronically low for the foreseeable future precisely because of its huge equity base (27% of assets). This is a company that cannot cover its cost of equity. Fundamentally, it should trade below its current and declining book value of $9.20.
Applying HCBK’s 1.3 multiple would yield a $12 stock price for Brookline. Applying ROME’s 25 PE to 2005 earnings would yield a $10.50 stock price for BRKL. NAL’s price-tangible book multiple is closer, but it also has much more operational value—market cap per employee and per branch are less than half of BRKL’s.
I include Hudson United (HU) because it just agreed to sell to TD BankNorth, which I previously believed to be the most likely acquiror of Brookline. At its takeout price, HU is trading at ½ the P/E multiple of BRKL. Yes, its P/B is high, but its 31% ROE is six times larger than BRKL’s.
NLY (Annaly Mortgage) is one of the premier mortgage REITs in the country. I include them because it is good example of what you can actually do with excess equity to make money--lever it and make money off the yield curve spread. That unfortunately has become a much more difficult thing to do as the yield curve has severely flattened in the last year. NLY is down 20% in the last month and has seen its ROE fall from near 16% a year ago to 14% in Q1/01 to 11.4% most recently, with its price-to-common book multiple compressing from over 1.5 times in June to under 1.3 currently.
In my valuation analysis, I included two rather unorthodox metrics—market cap per employee and market cap per branch. These metrics highlight how little an acquiring bank would really get in an acquisition. HCBK is the only one that is even close on these metrics and they trade at a much lower price-to-tangible book ratio, not to mention have a plan for aggressive organic growth. Most regional banks trade at a market cap per employee of under $1 million.
Brookline also pays a quarterly dividend of $0.09 per share (about 100% of earnings) plus two $0.20 special dividends each year. It translates into a rather attractive yield of 4.6%. The only problem is that it is not being paid out of earnings, but rather from Brookline’s excess equity. As a result, tangible book value per share has declined from over $10 per share a year ago to $9.20 currently and will likely be below $9.00 by the end of the year.
I listened to one bull try to argue the valuation by splitting its equity into the amount necessary to support the balance sheet plus the excess. With a $2.2B balance sheet, necessary equity is roughly $150M. The excess is about $430M. By applying a common book takeout multiple of 3 times to normally capitalized banks plus the excess at 1 times, we get to a valuation of $880M ($150M*3 + $430M), or $14.40 per share. The only problem with this analysis is that the roughly $22M in annualized earnings that Brookline is earning would tank if we excluded the benefit from the excess equity. If we assume that the $430M in excess equity is earning 2.5% after tax, Brookline’s core earnings drop to $11M. That would translate into an ROE of 8% and a PE of 38 times at the assumed $450M takeout. Paying 3 times book for a bank with an 8% ROE is unreasonable. Besides, the bull’s math led to a price target 10% below the current price anyway.
Takeout Risk
Brookline completed its second step conversion from a mutual holding company just over three years ago, which means that the three-year moratorium on selling the bank (standard with second step conversions) ended in July. Combined with CEO Richard Chapman’s age of 69 (along with most other senior executives in their 60s) and no designated successor, there is a high degree of speculation that a sale of the bank is imminent.
In today’s flat yield curve environment, an acquiror’s strategy of trying to lever BRKL’s excess equity is a losing proposition. With short term funding rates approaching 4% and bond yields well under 5%, the returns would be uneconomic. Two or three years ago, in a steep yield curve environment, this strategy could have bailed out an acquiror, but not today. It makes an acquisition of Brookline that much less likely.
As the final kicker, I believe there is so much acquisition hype priced in the stock that I think the longs will be disappointed if an announcement ever does materialize. We only need to look at Zion’s (ZION) recent acquisition of Amegy (ABNK) to see that some stocks can actually decline on the day of the announcement. (ABNK was down 5% on the day.)
A Common Sense Analysis of Why Acquiring Brookline Is a Bad Idea
A bank that wanted to replicate BRKL’s 180 employees/15 branch franchise could do so at a fraction of the cost to buy Brookline. Platinum plated branches with flat screen tv’s, leather couches, marble pillars, Turkish rugs, free Wall Street Journals, and free Starbuck’s lattes could be built in Brookline for a very generous $5M a piece. That’s $75M for 15. A bank could also offer large signing bonuses ($100K?, $200K?) and premium salaries for 150 of the best branch managers, bank tellers, loan officers, and back office staff in the region. That might cost $30M or so. The point of this exercise is to put some tangible figures on what it would actually cost to duplicate BRKL’s physical franchise from scratch. It is something like $100M (probably less), a discount of over $900M to what buying BRKL would cost. Sure, it would take a couple of years to build up loans and deposits, but the cost differential easily offsets this. It is this analysis that has me convinced no bank will acquire Brookline at anything close to its current prices.
Possible Acquirors
Given BRKL’s $1B market cap, I am willing to bet that only a bank with a market cap of $4B or more would be able to actually buy Brookline without completely destroying and diluting its existing franchise. That limits the field of potential acquirors drastically to a handful of New England banks and large foreign banks with U.S. operations. My list:
Royal Bank of Scotland’s U.S. subsidiary is Rhode Island-based Citizens Bank, the sixth largest bank in the country with operations through New England and the Midwest. It has been an active acquiror for years, and late in 2004 closed its Ohio-based Charter One acquisition for $10B and 13.7 times forward earnings. (Incidentally, RBS’ market cap declined by $3B on news of the announcement indicating that investors thought that was expensive. Second, RBS has started to generate significant capital and may begin buying back stock as soon as 2006. It does not “need” BRKL’s excess capital.) CEO Larry Fish does live in the town of Brookline and it has been surmised he wants the number one market share in his hometown. This sounds plausible. If anyone has any good insights on Mr. Fish, I would love to hear them. Despite my relative ignorance, I am willing to bet they won’t buy given shareholders’ rather harsh reaction to the Charter One deal and generally stringent return requirements on the part of its British parent. Besides, Citizens already has a significant Boston presence and taking out BRKL doesn’t exactly eliminate a tough competitor.
Prior to its recently announced $1.9B acquisition of New York-based Hudson United, a take-out from BankNorth was the number one risk in my book. In 2004, the Maine-based bank sold 51% to the Canadian based bank, Toronto Dominion. The move on TD’s part was explicitly to serve as a vehicle to expand further into the U.S, which the HU deal certainly achieves. However, the timing of the HU deal is curious—right before the moratorium on BRKL was to expire. I think it is an indication that BNK is uninterested in Brookline. In any event, the relatively large HU deal takes BankNorth out of the M&A game for at least a year.
Jay Sidhu of Sovereign is a shrewd, hard-nosed negotiator who has stated that deals will only be done if they can be accretive to EPS in year 1. That’s not possible with BRKL. His acquisition of several divested Fleet/Bank Boston branches four years ago was brilliant and has given Sovereign a significant presence in the Boston area. He does not need Brookline as a fill-in.
Brookline’s Business
Brookline has $2.2B in assets of which $1.1B are mortgage loans (split roughly equally between commercial real estate, single family, and multifamily apartment buildings), $415M are indirect auto, $75M are traditional C&I, and $450M are in short term investments and longer term fixed income securities. On the other side of the balance sheet, it has $69M in non-interest bearing deposits, $101M in NOW accounts, $159M in savings accounts, $274M in money market accounts, $531M in cd’s$416M in borrowings, and $615M in equity. The company has 15 branches and roughly 200 employees (including part timers) located throughout Brookline and neighboring Middlesex County, in the metro suburbs of Boston. Because of its relatively recent conversion from a mutual bank (owned by the depositors), it is massively overcapitalized with a tangible equity to asset ratio of 27%. (Most banks are near 6%.) Finally, because it can’t effectively compete in the commercial loan market (look how small the balances are and they have been declining), Brookline decided two years ago to begin putting its excess equity to work by funding indirect auto loans. These are generally regarded as very low quality because the yields are thin, credit risk is an issue, and there is minimal customer interaction with which to leverage a relationship. Rapid growth of these loans (expect $600M in total balances by the end of 2006) should not be mistaken for genuine business growth.
Recent Results
Brookline’s second quarter results were sub-par and missed expectations. EPS came in at $0.09, missing consensus’ $0.10 target, mostly because of integration costs related to Mystic. The net interest margin actually compressed by 5 basis point from Q1/05 to 3.26%, amazing given the rising interest rate environment and BRKL’s high degree of no cost equity. Average loan balances, excluding the indirect auto portfolio, were up only $3M (less than 1%) from the first quarter to $1.597B. Core deposits similarly showed no growth from Q1, remaining flat at $604M. In talking with company management, you will hear that their competition is being irrational in pricing both loans and deposits and that customers are not particularly loyal if some other bank offers a high rate cd. Brookline has declined to compete at current pricing except when is has to, which is why there has been such little organic growth. Competition will likely not let up soon.
Consensus estimates are too optimistic and have only come down modestly despite the weak second quarter. The Ryan Beck analyst lowered his 2005 and 2006 numbers 10% from $0.42 to $0.38 and from $0.53 to $0.48, respectively. The only reason there is any growth projected in the earnings anyway is that short term interest rates have been rising. Because BRKL has a large amount of equity, which has an interest cost of zero, its interest rate spread is widening. There is no fundamental business reason why their earnings are increasing. As soon as the Fed’s rate increases stop, which will be soon, earnings growth will disappear.
Mystic Acquisition
In another move that destroyed shareholder value, Brookline acquired Mystic Bancorp early this year in a $65M deal. Acquisitions paid for in cash make a great deal of sense for a company that is overcapitalized, but Brookline issued $38M in equity (2.5M shares) to pay for the deal. No doubt, Mystic shareholders insisted on the stock component, but that is just plain crazy. As a result, tangible book value per share declined 7.5% from $9.89 in Q4/04 to $9.15 in Q1/05.
Final Points of Interest
The short interest in the stock is high with a short ratio of 25 days and short interest as a % of float at 8.4%. Clearly, this is a crowded short, but with good reason.
BRKL has significant and committed institutional backing—the top 6 institutions own 36% of the stock and include Third Avenue, Mac-Per-Wolf, Private Capital, Advisory Research, Barclays, and Putnam. Directors and insiders also own 9%. While it may seem foolish to bet against such a group, the decision to sell on any of their parts could spark a serious decline in the stock.
The $0.20 special dividend record date was July 28th. The stock can now be shorted without worrying about an extra dividend for six more months.
Catalyst
Announcement of a CEO search. The realization that this franchise is not saleable will occur when BRKL announces that it searching for CEO Dick Chapman’s replacement. Our conversations with CFO Paul Bechet have hinted that such an announcement may occur as soon as this fall.
Technical downward move within its two year trading range of 14-16.