Friday 27 March 2015

Perception Versus Reality: Do People Get More From Credit Unions Than Banks?

The Credit Union National Association (CUNA), the credit union equivalent to the American Bankers' Association (ABA), states that credit unions exist to serve members, returning earnings to members in the form of lower loan rates, higher interest on deposits, and lower fees.

Nearly four years ago, I tested the higher interest on deposits claim in a guest post with the exact title on The Financial Brand, an industry publication geared towards marketing executives at banks and credit unions. The reaction that I received, in person, via e-mail, and in the comments were a little sharp-edged. Clearly this remains an emotional issue.

When the going gets tough, go to the facts. In 2011, banks paid higher interest on their interest bearing deposits than credit unions throughout the measurement period. 

When I re-ran the analysis, what was true back in 2011 still holds true (see chart).

There is a difference in my analytics. I searched on banks and credit unions between $500 million and $5 billion in total assets. I took a wider swath in 2011 with institutions between $100 million and $10 billion. Today's analysis reduced the amount of very small financial institutions.

I continued to control for commercially oriented banks by limiting to banks with less than 30% commercial real estate or commercial business loans as a percent of total loans. Those banks tend to have higher level of business deposits, which tend to drive down cost of deposits. However, to further control for this, I only selected interest expense as a percent of interest bearing deposits, not counting checking accounts that pay no interest.

Based on the above, for the sum total of all interest bearing deposits, banks pay higher rates, on average.

Surprisingly, changing the institution size did tell a different story in non-interest expense to average assets, or what is termed the expense ratio (see chart).

Perhaps the financial crisis, which credit unions survived surprisingly well with the exception of corporate credit unions (similar to bankers' banks), woke up credit union leadership to scrutinize operating expense to increase profits. 

Yes, you read profits. Where do you think credit unions get their capital? If credit unions suffered a similar fate to many community banks, they couldn't back up the truck for shareholders to pony up equity to help absorb losses. Becoming more profitable was the logical solution to building up capital positions.

There are probably other reasons at work. In 2011, those expense ratios for credit unions were in the fours (greater than 4%). That was likely due to my going down to institutions with $100 million in assets. While I did the same for banks, many smaller banks are privately owned, one branch operations with very low expense ratios. By raising the bar to $500 million, my analysis likely raised bank expense ratios by excluding those hyper-efficient small banks, and reduced credit union expense ratios by eliminating very small, inefficient institutions.

As the charts show, there is little difference in expense ratios, on average, for the measured institutions. 

I think both trend charts show something that previously happened between thrifts, savings banks, and commercial banks: the homogenization of business models. In the late 1990's, many traditional thrifts entered commercial banking with both feet. The result is falling net interest margins for banks and rising net interest margins for thrifts, long term. Thrift expense ratios began to rise as they took on the more expensive commercial banking teams.

Credit unions are shedding their Select Employer Group (SEG) strategies by adopting community charters or by adding so many SEGs that nearly everyone qualifies to join. They have entered commercial banking to the extent permissible by their regulators. So I expect financial performance ratios to begin looking more and more like their bank competitors.

Except for the shareholders. And the taxes.

Do you thing credit union and bank business models getting more similar?

~ Jeff

Sunday 15 March 2015

Bank Deals: How Are They Working Out for You?


Bank mergers are picking up steam. Technological change, regulation, and scale are cited most often by sellers. Take a premium now, rather than drift slowly into the abyss of irrelevance.

But what about buyers? I have written about achieving positive operating leverage in the past. In fact, it is one of my most read blog posts. In his most recent Chairman's letter, Warren Buffett weighed in with the following about buyer performance post acquisition:

"We've also suffered financially when this mistake has been committed by companies whose shares Berkshire has owned (with the errors sometimes occurring while I was serving as a director). Too often CEOs seem blind to an elementary reality: The intrinsic value of the shares you give in an acquisition must not be greater than the intrinsic value of the business you receive.'

'I've yet to see an investment banker quantify this all-important math when he is presenting a stock-for stock deal to the board of a potential acquirer. Instead, the banker's focus will be on describing "customary" premiums-to-market-price that are currently being paid for acquisitions - an absolutely asinine way to evaluate the attractiveness of an acquisition - or whether the deal will increase the acquirer's earnings-per-share (which in itself should be far from determinative). In striving to achieve the desired per-share number, a panting CEO and his "helpers" will often conjure up fanciful "synergies." (As a director of 19 companies over the years, I've never heard of "dis-synergies" mentioned, though I've witnessed plenty of these once deals have closed.) Post mortems of acquisitions, in which reality is honestly compared to the original projections, are rare in American boardrooms. The should instead be standard practice."


How can Warren's words be put into practice? To exemplify, I took two of the largest acquisitions in 2011 to back-check if it improved the buying bank's EPS and efficiency ratio. I went back over three years because mergers should be considered and executed with long-term financial improvement and overall bank strategy in the forefront of the "should we buy" decision. Citing the first few "clean" quarters after closing the transaction perpetuates the short-term budgeting culture that plagues our industry and prohibits long-term investing. By looking three plus years after merger announcement, I avoid that self defeating game.

I didn't take the largest deals from 2011, which were: Capital One Financial/ING Bank ($9.0B), PNC/RBC Bank (USA) ($3.5B), and Comerica/Sterling Bancshares ($1.0B). These transactions were so large, and two were foreign banks selling US subs, that it doesn't relate to my readers. But the next two deals in the table certainly relate.

Both deals look like they improved the buyer's EPS, with People's United achieving a 10.8% annual growth rate and Susquehanna achieving an even better 19.9%. People's efficiency ratio, a measure of how much in operating expenses it takes to generate a dollar of revenue, went down slightly. Achieving economies of scale should drive down the efficiency ratio. Although People's decline in this ratio was small, the relative size of Danvers ($2.6B) was only 10% of People's size ($25.0B) at the time the deal was announced.

Susquehanna's efficiency ratio went up. This is counter-intuitive, especially since Tower's relative size ($2.6B) was more significant to Susquehanna's ($14.2B) at the time of announcement. One would think that realizing the necessary cost savings to justify paying a premium would result in a lower efficiency ratio. Susquehanna was unable to achieve this "economy of scale". It is also worth mentioning that Susquehanna's earnings were sub-par at the time they announced the Tower deal. They had a 0.32% ROA at announcement. Not something to include in the shareholders' letter. Nowhere to go but up, right?

People's, on the other hand, had a better ROA (0.84%) when they announced the Danvers deal than in the fourth quarter (0.74%). The primary culprit was a precipitous decline in net interest margin of 116 basis points (yikes!). The fact that their efficiency ratio went down tells me they hit their operating expenses hard. As Warren alluded to above, I bet that wasn't in their post-merger projections.

I don't think it's very complicated to decide to do a transaction or not. What fits your strategy? Can you build it or must you acquire it? If an acquisition, can you afford the premium for the target so your bank is better off for having done the deal than passing?

Once you land a deal, accountability should be equally uncomplicated. My firm once represented a bank that had an activist investor on the Board. All that guy wanted to talk about was the efficiency ratio. Very one dimensional. But I digress.

He did hold management accountable for achieving the cost savings in the projections. So management prepared a spreadsheet of the phase-in of cost savings and the overall cost structure of the combined bank once all synergies were achieved. It wasn't a very complicated spreadsheet, and also gave management some leeway to alter where things were cut, so long as they achieved their aggregate numbers.

At the end of your strategic measurement period post-acquisition, the value of your bank (intrinsic value mentioned by Warren) should be greater for having done the deal than if you went it alone. If People's and Susquehanna could not achieve the earnings growth in the table above, then doing those deals improved their value.

If each could have achieved those numbers on their own, and there are reasons to believe they could have, then why do the deal at all?

Is it a fair question?

~ Jeff



Saturday 7 March 2015

Say It Ain't So Doug! Square 1 Bank Sells to PacWest

In the name of head scratchers, Square 1 Financial of Durham, NC, one of the most successful startup banks in a generation, is turning over the keys to PacWest, a California bank. The deal left me scratching my head, because at first glance it made little sense that a bank with Square 1's earnings trajectory would sell.

Niche banks are a growing part of our financial institutions landscape. I often cited Square 1 for their focus and success. In their own words, "Square 1 is a financial service company focused primarily on serving entrepreneurs and their investors." A bank with a focused strategy! Brings a tear of joy to my eye.

It had one banking office (in Durham), and twelve loan production offices located in key innovation hubs across the US. Its Chairman and CEO, Doug Bowers, was a 30-year BofA vet and more recently a member of a private equity firm. So the niche Square 1 adopted made sense.

But why sell Doug?

An industry reporter hypothesized that it was the price... 22x earnings, 262% of tangible book... c'mon?! But that was close to where Square 1 was trading at announcement. So there was no price premium. In fact, the below chart demonstrates that if Square 1 remained independent, their stock price would soar past the value received in this merger.

Like most projected performance, the devil's in the details. What I did was assume Square 1's 3-year compound annual growth rate in EPS (86%) linearly came down to earth to 12% by the end of the projection period, which is PacWest's 3-year EPS CAGR. I assumed PacWest's 12% would continue throughout the projection period. If all were true, it would have been more beneficial for Square 1 to go it alone. It is what I term "earning their right to remain independent."

So if future valuation wasn't the reason, then why? Perhaps they are receiving an outsized portion of the resulting bank than their current contribution. As I mentioned above, Square 1 did not receive a price premium from PacWest. So their pro forma ownership of PacWest is pretty much in line with their contribution (see table). Usually in a merger the seller receives a larger pro forma ownership stake because they receive a premium on their stock and they are relinquishing control. Not so, in this case.


So why did they sell? Here is what Bowers said in the press release: "Joining PacWest will be a terrific opportunity for our clients, employees, and stockholders. Square 1 offers PacWest a complementary line of business and significant core deposit growth. As part of PacWest, we will maintain our steadfast commitment to the entrepreneurial and venture communities, will be able to offer clients a wider array of products and will be well-positioned to continue to serve them through all stages of their growth."

That seems to tell us why PacWest bought Square 1, not why Square 1 sold to PacWest. So with Doug silent on the issue, here are my opinions on why one of my darling niche banks turned over the keys:

1. Institutional Ownership - Square 1 went public last March, raising $52 million at $18 per share from primarily institutional owners. The company was 70% institutionally owned with such names as Patriot Financial Partners, Castle Creek Capital, Endicott Opportunity Partners and other notables. Some had 5%-10% stakes, or about two million shares. Square 1 traded about 30,000-40,000 shares per day until around February 24th, when volumes soared (a fact that will not be lost on FINRA, although increased volumes prior to a merger announcement are not uncommon due to speculation). With such significant institutional ownership and relatively light normal trading volume, it would have been very difficult for those investors to lock in the trading gains experienced by Square 1 from November-February. How do you lock it in.... sell. Even if you are paid no premium. You can still lock in the price appreciation since you bought into the IPO.

2. Law of Large Numbers - As Square 1 grew larger, it would have to generate larger and larger amounts of business volume just to keep pace. For example, they had a $1.3 billion loan portfolio, the vast majority of which was commercial business loans. If 25% of that portfolio turned over every year, and I suspect it was more because business loans churn faster than commercial real estate loans, they would have to originate >$400 million of new/renewed loans per year to keep pace. Never mind growth. Which brings me to my third potential reason for selling...

3. Growth Trajectory - Square 1 was trading at 22x earnings when they sold. Banks their size typically trade around 13x-14x earnings. The premium was most likely the result of their balance sheet and earnings growth. Perhaps Doug and his senior management team were staring down the barrel of normalized growth. As investors began to recognize the slower growth, multiples would intuitively come down to the planet earth, suppressing stock price appreciation until the multiple normalized. That could have meant trading in a tight price range for a number of years. Why not lock in your tremendous gain since the IPO, and move on?

Square 1 was truly an extraordinary financial institution and I am sorry to see them go because I held them up as a premier example of how focused effort can lead to superior results.

If Doug Bowers and team were facing normalized growth and stock price appreciation, they could have decided to "cash cow" the bank, turning over a significant part of their earnings to investors in the form of dividends. In 2014, they enjoyed a 1.25% ROA and a 12.85% ROE. A great candidate for a cash cow. 

But alas that ship may have sailed when they backed up the truck to the institutional investor loading dock. They were numbers on a spreadsheet and were supposed to deliver the fund managers a big win. 

They did.

What else could Square 1 have done to satisfy their investors?

~ Jeff