Growth! Cultivating the Right Expansion Strategy

In recent years, most U.S. banks have chosen to expand their retail distribution networks through modest branching programs instead of the lollapalooza acquisitions that were once as ubiquitous as a banker’s blue-stripped suit. Then last October, Charlotte-based Bank of America Corp. paid $43 billion to buy FleetBoston Financial Corp., followed in January by New York-based J.P. Morgan Chase & Co.’s $58 billion buyout of Bank One Corp. in Chicagou00e2u20ac”raising the question of whether these deals signal an upsurge in takeover activity and an industrywide abandonment of branching.
The most likely answer is yes and no.

Yes, a rapidly improving economy and relatively clean balance sheets throughout the industry should set the stage for a resurgence in acquisitions as many banks look to gain retail customers and build market share. “I think that [mergers and acquisitions] are driven in large part by CEO confidence,” says Brian Sterling, a principal at investment banking firm Sandler O’Neill & Partners in New York. “And when you see [Bank of America CEO] Ken Lewis spend $43 billion, that says a lot about confidence. I think we will see some uptick in M&A activity.”

And no, the return of M&A after a three-year dry spell doesn’t automatically mean that branching is dead. As long as the retail deposit market remains strongu00e2u20ac”and it shows no clear signs of weakeningu00e2u20ac”banks will continue opening new branches in an effort to capture a larger share of low-cost funding. The widespread popularity of branching today is stunning given that many so-called experts once thought the Internet would obviate the need for a brick-and-mortar distribution system. “Just five years ago, people said there was no need for the branch,” says William F. Hickey, also a principal at Sandler O’Neill.

A small group of institutionsu00e2u20ac”including Cherry Hill, New Jersey-based Commerce Bancorp and its iconoclastic chairman and chief executive officer, Vernon W. Hill IIu00e2u20ac”by and large will only expand through branching. “You can only be a great retailer by growing de novo,” says Hill, who has expanded north to New York City and soon will head south to Washington, D.C. However, the majority of banks, including Charlotte-based Wachovia Corp., will instead use branching to fill holes in their existing retail markets. “We want to keep our network fresh, dynamic, and in a position to serve the customer very well,” says Ben Jenkins, president of Wachovia’s General Bank, which includes its retail operation.

Some institutions, including National City Corp. in Cleveland and Seattle-based Washington Mutual Inc., will combine these two expansion strategies, using a small acquisition to enter a new market and then expand their market share through branching. Indeed, this approach may be a necessity in highly consolidated markets where the available banks for sale are relatively small and may fall short of the acquirer’s market share objectives. “We do believe that a deliberate strategy of acquisition and de novo can be a successful formula for entering new markets,” says Jim Hughes, an executive vice president in charge of National City’s retail network.

Of course, banks and thrifts have been acquiring each other and opening new branches for years. But the industry’s preference for one expansion strategy or the other has often been determined by circumstance, whether it was a change in government policy or the workings of macroeconomics.

The advent of regional banking compacts in the 1980s sparked a conflagration of merger activity that lasted through the rest of the decade. It then subsided during the recessionary period of the early 1990s when a record number of banks failed, only to re-ignite a few years later. The industry’s consolidation reached its peak in 1998, when there were 504 bank and thrift deals, according to Charlottesville, Virginia-based SNL Financial. This was a historic time for a heavily regulated industry that had been prevented by state and federal law from expanding nationwide. As these barriers began to fall, active acquirers like Bank of America and Wachoviau00e2u20ac”once known as NCNB Corp. and First Union Corp.u00e2u20ac”set out to assemble regional organizations through a dizzying series of acquisitions. “If you were going to be a player in interstate banking, you had to be active when people were selling,” says Jenkins.

Branching’s popularity is a fairly new development. The 1927 McFadden Act allowed national banks to branch across state lines to the extent allowed by state law, but most states forbade out-of-state institutions from branching across their borders. McFadden was later repealed by the Riegle-Neal Interstate Banking and Branch Efficiency Act of 1994, which ushered in full interstate banking and branching throughout most of the country.

More recently, the growth of branching has largely been driven by a surge in retail deposits following the downward spiral of the stock market in 2000, when many investors sought a safe haven for their funds. “Historically, most large banks rarely thought of retail as a growth business,” says Gordon J. Goetzmann, a managing vice president at First Manhattan Consulting Group in New York. Nowadays, many large banks see retail as their fastest-growing business and are therefore expanding their branch networks.

According to the Federal Deposit Insurance Corp., U.S. commercial banks and savings institutions had a total of 87,768 offices throughout the country as of June 30, 2003. Although this data includes such nonbranch facilities as headquarter buildings and loan production offices, it’s still an effective measuring stick for the growth in the number of branches. This compares to 86,055 offices on June 30, 2001, for a net increase of approximately 2%.

While the industry’s branch count was going up, the pace of consolidation was slowing down considerablyu00e2u20ac”there were only 231 bank and thrift acquisitions in 2002, according to SNL Financial. Last year was shaping up to be even slower: There had been just 206 acquisitions as of early December. A mild recession and drawn out economic recovery, which stretched from 2000 to well into 2003, drove down the stock prices of many potential acquirers and deprived them of a strong currency with which to do deals.

There’s little doubt that as the economic recovery takes hold and bank stock prices rebound, many institutions will start looking again for acquisitions in geographic locations where they want a stronger retail presence. “I think that low-premium deals in the large-cap space might happen,” says Denis Laplante, a senior bank analyst at Keefe Bruyette & Woods in New York. “It’s still a consolidating industry. But I think that most of the activity will be in the small- and mid-cap area.”

“The reality is that large organizations are always out there courting smaller players,” adds Goetzmann.

The advantage of an acquisition-driven expansion strategy is that the buyer can gain a commanding market share in a desired location from the moment the deal closes. For example, Bank of America is now the leading retail bank in New England, a position that would have been impossible to build through a branching program. But acquisitions can end up costing more than their purchase price, as Bank of America found out when its stock dropped from approximately $82 a share to around $73 on the news that it had acquired FleetBoston. The purchase price was a hefty 43% premium to FleetBoston’s stock prior to the announcement, which led some institutional investors to dump BofA’s stock.

“Frankly I think they overpaid,” says Laplante. “Much of the value transfer is going to Fleet shareholders.” Bank of America has said the deal will reduce profits by 2% this year, and boost earnings by 1% in 2005.

What may have bothered Wall Street was the deal’s similarity to a number of overpriced acquisitions in the late 1990s that never paid off for investors. When an acquirer overpays, it generally has three ways of offsetting the premium: either significantly boost the combined organization’s revenue, drastically reduce its costs, or fashion a more modest combination of the two. Many large acquirers end up pushing too hard for overhead reductions, resulting in customer service snafus that actually cost them market share.

Bank of America has said publicly that it will eliminate $1.1 billionu00e2u20ac”or 6%u00e2u20ac”from the merged bank’s cost structure by 2005. The bank also has said that it will combine the banks’ operations as quickly as possible, while being sensitive to customer service issues. The danger is that Bank of America will rush the integration and end up losing customers. “They may get those [cost saving] numbers, but it’s borderline aggressive,” says Laplante.

Bank of America should not have been surprised by Wall Street’s rude reception of the Fleet acquisition. In January 2003, Winston-Salem, North Carolina-based BB&T Corp., which enjoys the reputation of being one of the industry’s most skilled acquirersu00e2u20ac”announced a $3.4 billion acquisition of First Virginia Banks in Falls Church, Virginia, its largest deal ever. In a little over two months, BB&T’s stock dropped from around $39 a share to approximately $30.50 before gradually climbing back to $39.69 by Dec. 1. But when BB&T announced a small $436 million deal for a bank in Florida just two days later, its stock dropped again, although this time the decline was a little over $2 a share.

The message is clear: Banks may continue to expand their retail networks through acquisition, but they should expect wary institutional investors to scrutinize their deals closely. Even those acquirers with a long track record of success may find themselves sitting in Wall Street’s penalty box until investors see benefits from the deal flowing through their income statements.

Branching, by contrast, is less costly than making acquisitions because it is executed on a more modest scale. But it takes longer to grow market share through branching precisely because it proceeds at a more measured pace, and the strategy does entail some degree of financial risk. It generally takes three years for a new branch to break even, and banks that are pursuing multiple-year programs end up sacrificing some of their profitabilityu00e2u20ac”particularly in year three when all of their new branches are still losing money.

One of the harshest critics of the banking industry’s merger practices over the past several years has been hedge fund investor Thomas K. Brown, president of New York-based Second Curve Capital (and a regular columnist for Bank Director). As it happens, Brown isn’t enthusiastic about most branching programs either. “Acquisitions can be done well, but most aren’t,” says Brown. “Branching [programs] can be done well, but most aren’t.”

The biggest problem, according to Brown, is that many organizations lack the underlying retail banking skills to make their new branches successful. “It’s analogous to the restaurant business, where you open a new restaurant and people don’t come in because the food isn’t any good,” he says. “So instead of opening more restaurants, fix the food.”

This is essentially the same conclusion that First Manhattan came to last year when it studied the branching phenomenon. The consulting firm found there was considerable variability between the performance of new branches that had been open for five years, with the top 20% performing 10 times better on average than the bottom 20%. And the most effective predictor of which quintile an organization’s de novo branches are likely to fall is the same-store sales performance of its mature branches.

“If your existing branches aren’t doing that well, opening new ones isn’t going to help,” says Goetzmann, who helped run the study. Or as Brown puts it, “Don’t open more restaurants if your same-store sales aren’t very good.”

Although a long list of superregional banks announced branching programs last year, including the likes of Bank of America, Wachovia, Washington Mutual, and U.S. Bancorp., First Manhattan found that the 30 largest banks actually closed twice as many branches as they opened from 1997 through 2002. Many of those closures resulted from post-acquisition downsizing, when large branch networks were rationalized to save money. Now, many of those same banks have been examining their existing networks and opening new branches in locations with the greatest growth potential.

One advantage of branch expansion programs is they create far less distraction for the entire organization than a large acquisition. They also do not require the conversion of an acquired bank’s culture since branching seeks to propagate the existing culture. But industry orthodoxy holds that branching is primarily used to fill out an existing franchiseu00e2u20ac”and rarely to expand into a new market.

Vern Hill at Commerce thinks that’s a bunch of hogwash. He started Commerce in 1973 with just one branch after earning an MBA from the Wharton School at the University of Pennsylvania and later started a successful property company that developed commercial sites for major retailers. Today, Commerce has more than $21 billion in assets, with some 260 branches spread from New York City south into Delaware.

Although Commerce has done a few very small mergers, its branch network has been built almost entirely through de novo expansion. Hill chose not to expand through acquisition because he does not believe that you can build a great retaileru00e2u20ac”and the operative phrase here is great retaileru00e2u20ac”by converting someone else’s culture. “No great retailer has ever been built through acquisition,” Hill says. “They’ve all been builtu00e2u20ac” going back to J.C. Penneyu00e2u20ac”one store at a time.”

In fact, Commerce does not describe itself as a bank, but as a “growth retailer” that sells convenience. One of the most important management metrics at Commerce is same-store sales, which is the year-over-year growth in loans and deposits at each branch. This is an important measurement in retailing, but less so in banking. “It’s the thought process that’s important,” says Hill. “Guys talk about retail banking, but they don’t know what it means. All [same-store sales data] tells you is whether you have a differentiated model.”

There’s no question that Commerce operates under a distinctly different business model than most other banks today. Says Hill, “We think the value of a bank is in its core deposits.” Hill also believes that most consumers are willing to trade higher interest rates on deposits for a great retail experienceu00e2u20ac”and that’s exactly what his bank works hard to provide. Commerce is a big proponent of free checking, and its branches are open seven days a weeku00e2u20ac”including several hours on Sunday. Branch personnel are exhaustively trained in the bank’s own facility in Cherry Hill. Consumers must like what they see, because loans and deposits have grown 26% and 32%, respectively, over the past five years.

One of Hill’s strengths is that he understands the power of core deposits. With a cost of funds of just over 2%, Commerce can afford to underwrite credit conservatively and still make a decent spread while safeguarding its asset quality. The bank’s business model has turned out to be highly profitable: Earnings-per-share growth over the last five years has averaged 31%. “I actually think [Commerce] is the best bank in America right now,” says Brown, whose hedge fund maintains a position in Commerce’s stock.

Hill’s success at moving into New York City and Long Island in recent years disproves the notion that you can’t successfully branch into new markets. Since the fall of 2001, Commerce has opened 25 branches in the city’s five boroughs and 14 on Long Island. Hill claims the bank’s venture has been spectacularly successful, with the annualized growth rate for all its branches there averaging around $100 million per branch.

Commerce plans on moving into the Washington, D.C. and Baltimore markets in 2005, with somewhere between 10 and 15 branches, and eventually will challenge Bank of America in Boston. All of that growth likely will come from de novo branching. “We’ve never done a big acquisition, and it’s almost impossible to see that happening.”

Even though Hill has shown that de novo branching can be done profitably on a large scale, and into new markets, most banks and thrifts will use branching to fill out their existing franchise. “We’re looking hard at how we intelligently and strategically expand our branch network,” says Jenkins. The bank wants to open between 30 and 50 new branches a year for the next five years and has selected a number of growth markets within its network for expansion, including Orlando, the Tampa/St. Petersburg metropolitan area, Atlanta, Richmond, and the Washington, D.C./Northern Virginia market. Despite its vast sizeu00e2u20ac”Wachovia is the country’s fifth-largest bank, with approximately 2,600 branches stretching along the East Coast from Florida to Connecticutu00e2u20ac”Jenkins believes that even a modest branching program is important. “I think it’s very valuable,” he says. “We’re a retailer, and you can’t imagine a retailer standing pat.”

Wachovia even went into the Manhattan market last year, opening two branches that have already hit a 20% return on investment and were generating average retail and small-business deposits of $350,000 per day at year-end 2003. “We thought there was a service vacuum [in Manhattan], and that we could fill that vacuum,” says Jenkins.

Running a successful de novo strategy requires a skill set that’s quite different from that of a successful acquirer. To be good at branching, one needs to be competent at demographic analysis, site selection, staffing, training, and business development at the street corner level. The acquisition process relies on some of those same skills, but most of the attention is focused internally on reengineering and organizational integration.

One activityu00e2u20ac”branchingu00e2u20ac”is inherently outward looking, while the otheru00e2u20ac”acquiringu00e2u20ac”primarily looks inside once a deal has been struck. “There are organizations that only do acquisitions,” says Hickey at Sandler O’Neill. “There are organizations that only do de novo. And there are organizations that do both.”

An expansion strategy that may find increasing favor in the coming years is to enter a new market through a modest acquisition that can be used as a platform for a branching program thereafter. This is precisely what Washington Mutual has done in the New York market, where it acquired Dime Bancorp a few years ago and has used that as a platform to branch throughout the city. But a dual strategy has its own risks. If a bank acquires an underperforming organization in hopes of making it better, investors may revolt. “The market hasn’t looked too kindly at those kinds of deals,” Hickey says.

Some urban markets are now so consolidated from previous acquisitions that it may not be possible for an acquirer to attain its market share objectives in just one deal. This was the situation facing National City last November when it paid $475 million to acquire St. Louis-based Allegiant Bancorp, a $2.5 billion institution that was that market’s largest remaining independent bank. Allegiant has a 37-branch network scattered throughout the city, a solid cross-section of customers, and a strong commercial banking capability. National City, the country’s 11th largest bank with $123 billion in assets, will give Allegiant a highly rated online banking capability and a more robust set of products, including its small-business banking model.

The Cleveland bank will also use Allegiant as a platform from which to expand further in St. Louis. “They have a nice cadre of branches set out around St. Louis,” says National City’s Hughes. “They’re probably not as deep as we’d like, but it offers us an opportunity to expand. We’d like to get to 50 as quickly as possible.”

Because Allegiant uses a single back-office system and has a single charter, Hughes says it should be relatively easy to integrate the smaller bank into National City’s operation. This means the branch-expansion project can occur simultaneously. “We have people looking at the demographics and looking around to see what property is available,” he says. Allegiant’s president and CEO, Shaun R. Hayes, will become president of National City’s Missouri bank, and he will assist in the branching effort. “We think we have a really good local team to help us through this,” says Hughes.

Hughes agrees that branching and acquiring rely on a different set of skills, and not all banks are good at both. But the circumstances in St. Louis are requiring National City to do both, and it’s a scenario that may confront many other banks as they attempt to expand in an industry that has already seen plenty of consolidation.