Bank Retailing
Plus: Wise, Danske Bank, Happy Holidays
Welcome to another issue of Net Interest, where I distil 25+ years of experience analysing and investing in financial stocks into a weekly newsletter. This week we see if retail business models can be applied to banking, using the UK’s Metro Bank and its US cousins as case studies. It’s the last Net Interest until after the holidays but if you’ve enjoyed what you’ve read this year, why not join some of the most net interesting people on the planet as a paying subscriber. As well as supplementary content each week, you will get access to a bustling archive of over 100 issues (and my email address). Thanks!
Every hedge fund manager remembers their first short. Mine was a stock called Commerce Bancorp. At the time, it was one of the fastest growing banks in America. Under the stewardship of its founder-CEO, Vernon Hill, the bank had spent thirty years expanding its branch network beyond its original base in southern New Jersey. By 2006, it had over 400 offices across eight states and Washington DC.
Vernon Hill treated his branches as stores; his product: deposits. “From our inception, Commerce has refined a business model patterned after the great retailers of America, i.e., Home Depot, WalMart, McDonald’s, Starbucks, rather than the typical bank or financial institution,” he told shareholders in his 2000 annual report.
His thinking was that there were two sides to banking: lending and deposits. With most banks focused on lending, his time was better spent focusing on deposits: “Banking is, essentially, a government license to borrow money cheaply. Anyone can make loans, but only licensed, government-sanctioned banks can accept deposits. That’s why the legal and economic value of a bank is in gathering deposits from loyal customers.”
Deposit gathering became the bank’s primary mission. While the average American bank branch grew deposits at a rate of $1 million per branch per year, Hill boasted that Commerce grew at a rate of $18 million annually. In the five years up to the end of 2006, overall deposit growth ran at 32% a year.
The market loved the strategy. “Commerce’s retailing strategy has driven the company’s strong growth and set it apart from the crowd,” wrote one analyst a few years earlier. “With a focus on growing deposits first, then revenues – Commerce has posted a compound annual Earnings Per Share growth rate of 18% over the past five years. Management has ingrained a strong sense of customer focus as well as a team spirit and employee pride in the franchise that pervades nearly every aspect of the company.”
In the ten years before I took an interest, the stock ran up over 7x.
But clouds were beginning to accumulate. All of those deposits needed to be put somewhere and Commerce had built up a large securities portfolio to house them. When the yield curve was steep, this was a profitable activity: In 2002, the bank’s net interest margin – the key driver of its profitability – was 4.69%. As longer-term rates fell, though, and the yield curve flattened, the net interest margin came under pressure; by 2006 it had fallen to 3.35%. Earnings growth slowed to around 5% a year. The stock’s valuation, meanwhile, remained stubbornly high.
It turned out to be a reasonable short, but not for the reasons I anticipated. As well as running Commerce Bank, Vernon Hill retained a keen interest in real estate. His first job after graduating was in property development – in fact, it was his work with McDonalds, for which he developed over 100 sites, that inspired his retail approach to banking. But Hill began to conflate his interests. In early 2007, regulators announced that they were investigating Commerce Bank over real estate transactions conducted with entities that he controlled. Buying or leasing real estate from your own CEO represents a conflict of interest that regulators don’t take too kindly to. And if that wasn’t enough, Commerce was also awarding design and architectural contracts to Hill’s wife, Shirley, and her store design agency, InterArch, on an exclusive basis. In 2005, the bank paid her firm $7.5 million for these services. Regulators slapped the bank with a consent order with instructions to tighten up its governance; the stock price collapsed.
By the summer, Hill had been ousted. Faced with declining net interest margins, longer timelines on new branch openings due to the regulatory actions, slowing “same store” deposit growth and turnover in management, the board of directors put the bank up for sale. Four banks expressed an interest, and two put in a bid. The highest bidder was TD Bank of Canada.
Fortunately, we’d closed our short by then. TD’s bid was worth $42 a share, more than the $41 all-time-high the stock had hit in May 2006 and around 35% more than the low after the regulatory probe was announced. No-one says shorting is easy. If anything, the environment made our original thesis more compelling and sure enough, during due diligence, TD tried to reduce its price to between $35 and $38 a share because of “potential losses in Commerce’s securities portfolio as a result of recent sharp declines in the bond markets and slower low-cost deposit growth in Commerce’s stores”. But by then it was too late.
I was reminded of the Commerce Bank episode this week. As a large shareholder, Vernon Hill netted around $400 million when the bank was sold to TD. He decided to reinvest some of the proceeds in a new banking venture, in London, which he called Metro Bank. In essence, he replicated the Commerce Bank model in the UK. The core idea to go after deposits was the same, the branding was the same, the focus on customer service was the same. Even the governance failings were the same. This week, the Financial Conduct Authority issued a final notice in relation to breaches of regulatory rules in 2019. Like Commerce before it, Metro Bank pursued a strategy modelled around retailing. The question: does it work?
The Rise and Fall of Metro Bank
Before 2010, no new high-street bank had been founded in the UK for 170 years. Vernon Hill saw a market where existing banks under-serve, overcharge, and under-invest in their continuing business. “They have a uniquely British philosophy that, ‘We’re doing you a favor by letting you bank with us (or buy from us)’,” he writes. To break into the market, he borrowed from his Commerce playbook.
The building block was customer service centred around the branch. Like at Commerce, the model was influenced by fast food retail – “from making all channels accessible to consumers, to saturating communities with stores for their convenience and treating every customer, large or small, like royalty.” The bank even launched the concept of drive-thru banking in the UK.
But since leaving Commerce Bank, Hill had found a new model to mimic: the Apple store. “Metro Bank sees its banking business as a retail operation more akin to Apple than the Bank of England; we open stores, not branches. We recruit colleagues, not employees.” His book, from which many of the quotes here are taken, uses Apple as a case study throughout. (“We Apple-ized the banking world in Britain.”)
Part of the evolution reflects a difference between the UK and the US: “You don't have suburban sprawl, particularly in Greater London,” he told investors on an earnings call a few years ago. “You’re going to get less stores but they’re going to be much higher deposits per store. When I look back at the 450 to 500 stores in America…I often think [of] the mistakes I made. When I want to go into an area so bad that instead of waiting for the A plus site…I took the A minus site. We did okay but not as good as we should have done in that market. And I swore to myself I’m not going to do that again. So our theory is: the best site in the best town is always worth the money. And it's always worth waiting.”
Customers responded well to the proposition. In its first ten years of business, Metro Bank established 77 stores across the UK and accumulated £15.6 billion of deposits. But prime sites are expensive. As at the end of June 2022, the net book value of Metro’s freehold land and buildings racked up to £311 million. The bank retains freehold over 28 of its stores, equivalent to around 37% of its portfolio. In addition, it has £221 million worth of right-of-use assets on its balance sheet associated with long leases across the rest of its network (“really, really long leases,” according to the bank’s CEO). Together, these assets represent almost three quarters of the group’s tangible equity, a much higher share than at other banks.
A lot of money also goes into fitting the branches out. Many stores have double height entrances, requiring the upper floor to be ripped out. According to Hill, the bank spends £3-4 million renovating each site it acquires. Here, the bank imported some bad habits from the US: it hired Shirley Hill to do the design. Before shareholders revolted in 2019, Metro Bank paid her firm £25 million in fees. 1
Combined with staffing, where costs amount to close to £250 million a year, the branch network is very expensive to manage. Hill recognised this – his philosophy was to deploy a superior customer experience to attract stickier deposits. And he was leery of other banks that achieve profitability via cost control. “There are two ways to make money,” he says. “Grow your top line – that’s the best way. Cost cutting is your way to extinction.”
But the model only works in a high interest rate environment. For most of Metro Bank’s history, rates have been low so the margin on the deposit book has also been low, much as it was towards the end of Hill’s tenure at Commerce Bank. In addition, the competitive environment became more intense on the asset side, as the UK’s “ring fencing” regulatory structure trapped excess liquidity domestically causing big banks to compete on price for new mortgage lending. As a result, Metro was never able to generate the revenues to cover its costs. Last year, its net interest margin was only 1.40% and, since launch, cumulative losses have amounted to £1 billion.
Metro Bank came to the market via IPO in 2016, its model initially rewarded with one of the highest valuation multiples in European banking. Its stock is now down 95% and its market cap flounders at £190 million. Vernon Hill once observed that “the Commerce historic experience was roughly speaking our market cap was 20% of our deposits.” At around 1%, Metro trades well shy of that. Part of the reason is the bank’s lack of financial discipline and the different environment that prevails today. But part stems from the delusion that banks are retailers.
Because unlike retailers, banks have long duration assets, and regulators require them to hold capital against those assets to provide a cushion against any unexpected losses that may emerge. That capital is a drag on growth – it’s capital that can’t be invested in new branches or infrastructure. Regulators establish the amount of capital required with reference to a bank’s assets – in principle, the higher the asset risk, the more capital needs to be set aside. There are 17 categories of exposure, each with a different “risk weighting”. And that’s under a so-called standardised approach; under an advanced approach, which banks have rigorously to qualify for, calculations get even more complex. Not something Apple has to contend with.
In 2019, Metro Bank revealed that it had mis-stated the risk weighting of its assets, giving the impression it had more excess capital than it did. It was a shock to markets – the stock fell 39%, the largest single price drop experienced by a UK bank since the financial crisis. Initially, the authorities put the error down to poor systems, although this week, the Financial Conduct Authority concluded that management of the bank had known about the under-statement for some time; they and the bank have been fined accordingly.
That management is now gone. Vernon Hill left in October 2019 and his CEO left shortly afterwards. Under new management, the bank is focused on its banking challenges. Its target is to reduce costs in 2022, making it the first year in the history of the bank that costs have gone down. Higher rates will support revenues; net interest margin entered the second half of the year at around 1.85%. Meanwhile, regulators have gradually been reducing capital surcharges placed on the bank in light of its failings, and the prospect of Metro qualifying for a less capital intensive “advanced approach” looks real.
On his latest earnings call in July, the bank’s new CEO said: “We now are guiding that we will be operating at break-even in the quarter one of 2023, and it’s done through good old-fashioned banking by expanding margin and changing the nature of the balance sheet.”
But what of Vernon Hill?
Alongside his foray into the UK, he invested in a Philadelphia-based bank, Republic Bank. He gave it the same shtick. Hill worked alongside the bank’s founder, Harry Madonna, to rebuild Republic in the image of Commerce Bank. “I wanted to rebuild our bank model. I wanted to put our team back together, and I wanted to do it all over again. The object is to take the Commerce Bank market share back from Toronto-Dominion – TD Bank.” He took over as chairman in 2016 and as CEO in 2021. Shirley and her team from InterArch were in, too. “We took Republic Bank – an unsexy, commoditized business – and broke it down into retail delivery and customer-centric services.”
Unfortunately, the rebuild didn’t work. Two separate activist investors took stakes in the company and agitated for change. Among their complaints: “a retainer to pay Hill’s wife for architecture and interior design work for the bank”. The board splintered into two factions, four members backing a continuation of the expansion plan under Hill, and four backing changes under founder Harry Madonna. When one of Hill’s allies died, the Madonna faction mobilised to oust him. After a series of lawsuits, Hill finally left in August this year.
“He’s absolutely in love with a business model that doesn’t work,” said one of the activist shareholders.
Retailing is fine, but it’s not banking.
Now on its third edition, Vernon Hill’s book, Fans! Not Customers, meanders through the author’s theory of banking. The first edition was subject to an injunction brought by TD Bank, which claimed copyright protection on the basis that portions of it were written when Hill was CEO of Commerce Bank and that it was designed as a marketing tool for the bank. The injunction was overturned in 2019. According to the judge: “Hill may perhaps not be the next prize-winning, or even best-selling, business-book author. But he has a story to tell and readers eager to learn from him.” My advice: don’t bother. If you are looking for reading ideas, there are plenty in here.
The self-dealings with InterArch go the other way, too. According to a Federal lawsuit, Shirley Hill invested assets from a profit-sharing plan for InterArch employees in Vernon’s banks. At one point, 70% of the plan’s assets were invested in Metro Bank stock, and 13% of the assets were invested in Republic Bank.