by Lee Kyriacou, Novantas 

Introduction: Following the financial crisis, banks may have reached a profitability plateau unless they work to rebuild longstanding pillars of their business models.

After years of struggling through massive loan losses, lawsuits, new regulations, and declining spreads, U.S. banks are at a profitability plateau, with much hope placed on rising rates to restore historical profitability levels for the industry. However, while rising rates may help, the past half-dozen years have also seen substantial structural changes to many of the underlying fundamentals of the U.S. banking industry. This has created a “new normal” of lower profitability for commercial banks and thrifts, with an average return on assets (ROA) on the order of 115 bp to 120 bp.

The major structural changes cut across the income statement and balance sheet: a long-term decline of net interest margin that goes well beyond the current concerns of spread compression, permanently lower fee revenue, significantly higher compliance and litigation costs, higher capital and liquidity requirements, and greater loan loss provisioning. On top of all those, there is the fundamental change in banking away from branches and to mobile and online channels, along with the rise of major name-brand direct banking competitors. Of greatest concern is the cumulative impact of these structural changes on classic financial intermediation – the gathering of deposits and extending of loans – which looks to be marginally profitable at best.

As rates rise, banks will see a little relief, but most cannot relax. They must tackle multiple efforts to rebuild or replace longstanding pillars of their business model – and probably without raising pricing given what will be persistent industry excess capacity. Most pressing, the branch network must be reduced to a more appropriate size at a substantially lower cost – which will allow for higher deposit rates to compete with online-only banks. Building up fee revenue sources from mortgage origination, capital markets, investments, and insurance is critical to replace lost fees and supplement spread revenue. And banks will need to invest in the analytics to support smarter loan and relationship pricing, as well as in the process redesign to bring down lending costs. The largest banks with both scale and more diverse product lines have more options available to navigate through these major changes, while regional and community banks with fewer options will continue to struggle, resulting in an acceleration of bank consolidation.

Regulators and policymakers must understand the dynamics of long-term change in the industry. Foremost, policymakers need to see the threat to traditional intermediation, especially given the almost exclusive focus of smaller banks here. But more broadly, the public debate going forward needs to include the likely impact of additional changes on the long-term profitability and health of the industry. 

Limits to Industry Recovery
Let’s do the numbers, looking at FDIC aggregate data for all bank depository institutions for second quarter of 2015. This quarter saw a strong improvement for the industry, and may be a harbinger of things to come. Loans and deposits were up 4% to 5% versus a year ago, revenue actually rose rather than staying flat, and profitability picked up nicely with an ROA of 109 bp, the highest level in two years. That’s a long way from the depths of the financial crisis, when 2009 industry pre-tax profitability was a rounding error above zero, recovering to more than $221 billion pre-tax for full year 2014. 

But when you dig into how the industry has come back, and where it is relative to pre-crisis norms, you can see the revenue and profitability obstacles facing the industry. First, virtually all of the change from 2009 to 2014 came from a $220 billion reduction in loan loss provision – i.e., all revenue and expense are a wash. While net interest income was higher in 2011 by more than $25 billion, with interest income falling by $70 billion but interest expense falling further, that was wholly offset by either non-interest revenue decline or expense increases.
Second, industry revenue until recent quarters has been stubbornly flat, peaking early in 2009 and remaining effectively unchanged for six years. And while balances sheets have grown – driven by the impact of Federal Reserve monetary policy on deposit growth – loans until recently had not kept pace with deposits, resulting in a steadily declining loan-to-deposit ratio.

Revenue has finally started growing slowly in the past few quarters, and second-quarter 2015 revenue set a new quarterly high for FDIC depository institutions. We may also have turned the corner on loan growth, which clearly exceeded deposit growth with a clear uptick in the loan-to-deposit ratio. 

Net income is also at record levels, but profitability – with ROA at 109 bp and return on equity below 10% – remains well below historical norms (see graphic). And while the pace of loan and revenue growth will accelerate gradually in 2016, profitability will not for most banks go back to pre-crisis levels. Rather, we expect average industry ROA to improve to 115 or 120 bp – the “new normal” – but we don’t expect to see the industry average rise back to the pre-crisis level of around 135 bp.

Structural Drivers of Lower Profitability in Banking
Bankers all bemoan the inexorable downward march of net interest margin (NIM), which has fallen roughly 75 bp in the past five years, and most expect that impending interest rate increases will improve balance sheet spreads. While some relief is expected, prior instances of rising rates after the recession have been accompanied by further declines in NIM, not increases. That is because rates do not rise as fast as existing loan yields are replaced by newer lower rate loans. This pattern may not occur after the next rate rise, simply because rates have been so low for so long that loan yields have already given up most of their pre-crisis spreads. That said, we expect this rate rise will see more rapid rises in deposit yields (measured as a “beta” of deposit yield increases relative to federal funds) because of tougher competition from online-only direct banks. And new lending is now accompanied by normal loan loss provision levels, which have ratcheted up to incorporate crisis-type losses into credit risk models.

Stepping back, there is also a longer-term structural issue with NIM. Looking across the economic cycles, NIM over the last 25 years has risen at the start of a recession and then given back all the increase and more until the next recession. This pattern reflects an overall tightening of competition for traditional balance sheet intermediation, and the availability of capital markets and non-bank sources of lending. This longer-term pattern will continue to bear down on spreads, so what relief comes from rising rates will not revive profitability to pre-crisis levels.

Regulatory change has altered banking business economics in a variety of ways as well. Non-interest revenue has taken several hits, early on in credit card fees, then debit interchange via the Durbin Amendment, then overdraft changes, and most recently trading revenues from the Volcker Rule. Collectively, these non-interest revenue hits amount to roughly $25 billion and have reduced the share of revenue from non-interest sources, even during a period of declining NIM. Beyond revenue, regulation has increased compliance and litigation expense, raised capital requirements, and created a liquidity coverage ratio requirement that leaves less room for lending and places lower value on non-core deposits.

And if this were not enough to deal with, bankers also face the one-two punch of mobile and other technology enabling the untethering of banking from paper, and of the consumer shifting transactions away from the branch at an accelerating pace. The replacement of paper or physical media by e-delivery, which made stores obsolete overnight for travel services, film development, video, and now books, is taking place for bank branches. Branch transaction counts are falling now in the mid- to high-single digits annually, and bank branches are eerily empty.

The story for smaller banks is worse still, because they rely more heavily on traditional intermediation of deposit gathering and lending. The fraction of non-interest revenue has fallen quickly for smaller banks, which have fewer fee-generating options and rely on deposit and loan fees. ROAs are commensurately lower for the smallest tier of banks, and while larger tiers are more evenly profitable, we expect the larger banks to return to higher ROAs over their smaller counterparts in the next year or two as litigation dies down.

What Banks Must Do 
The tasks facing banks are manifold and each daunting, and these issues will need attention at the same time that most banks are still struggling to adapt their products, modeling, and compliance for new regulation. Raising rates and fees will not likely be very fruitful. With thousands of banks left over from pre-interstate banking still trying to survive, there is significant excess capacity that will continue to put downward pressure on commercial and consumer lending. And with “free checking” in consumer payments, the entire industry watched when one of the major banks tried to impose a modest debit card monthly fee, and then quickly retracted in the face of expected switching. 

Foremost among these major industry tasks is reducing the size of the branch channel while increasing investment and functionality in the digital channels, particularly mobile. Getting the transition right is critical – we have seen banks close local branches without losing deposits and others that take significant deposit hits. 

The correct balance will vary by market and requires careful measurement and trade-offs between the number and size of branches, ATMs, signage, local advertising, and other elements of market presence, as well as the timing of lease expiration and the ability to cover restructuring costs and write-offs. In the end, we see dramatic declines in square footage dedicated to branches and a significant but more modest decline in branch counts; in turn, the brand is rapidly replacing the branch as the means of differentiation, with a decided advantage to national banks and household brand names. 

Finding the right balance also requires redesigning the customer cross-sell process, which has relied heavily on branch visits for incremental selling “at bats” and must change in order to reach customers who only rarely come into a branch. In the near term, as revenues rise, expect to see many banks take restructuring charges and goodwill write-downs, though this will not likely be a once-and-done charge but rather an ongoing series of them.

Getting higher non-interest revenue is, for the most part, a longer-term issue. The larger banks are well ahead of the game here, with banks greater than $100 billion in assets up around 40% of revenue from non-interest sources, while banks with less than $10 billion have only 25%. The largest banks dominate capital markets revenue, and many have national “monoline” businesses for credit cards and mortgage origination. Some regional and smaller banks have grown their way into significant mortgage origination or investment revenue volumes relative to their size, while others have acquired insurance agencies to build a revenue stream. Commercial lending for regional and larger community banks must beef up the quality of treasury management and capital market products sufficient to cross-sell effectively to their mid-tier business customers.

Apart from branch restructuring and fee revenue sources, banks must also continue to deepen analytic capabilities and streamline processes. Foremost on improving analytics is pricing, which most banks have turned to as a way to wring extra basis points and fees from both deposit gathering and lending. The next wave of analytics will go beyond product pricing and begin to optimize relationship pricing, taking into account local market position and competition as well as indicators of customer price sensitivity.

On the process side, loan origination for both commercial and consumer lending is ripe for improvement. On the consumer side, new compliance requirements and heightened scrutiny are driving caution but also ought to be driving process redesign. On the commercial side, compressing margins are forcing banks to look for efficiencies, and on the small business lending side new non-bank lenders with instant approval are forcing banks to redesign for time as well as credit quality.

In the end, not all of these multiple forces are easily fixed, and the collective weight of them will simply accelerate industry consolidation. The smaller banks are going away at rate that now tops 300 per year – so with total banks and thrifts at around 6,500, this pace will reduce the number of banks by roughly 5% a year. But we should now expect consolidation among the larger regional banks as well, who need to diversify revenue and deposit sources, and more importantly, develop the strong brands that are now needed to compete with the national banks and Internet-only banks with household names. One exception will be niche players in credit card, wealth management, and business owner private banking, who can successfully carve out a defensible position.

What Policymakers Need to Understand about Banking
Turning to public policy implications, the critical need is for policymakers to take a step back to see the long-term trends affecting the industry. Many of these trends are pushing the U.S. banking industry inevitably toward a larger scale: the waning importance of the branch and rising value of brand, structural spread compression and the deeper analytics that can partially offset the intense competition, and regulatory requirements that limit revenue sources and raise expenses. 

To keep things in historical perspective, it was regulation that restricted interstate banking and created the atomistic structure of the U.S. banking industry – with almost 18,000 banks and thrifts in 1985, which market forces have been restructuring for the past three decades. Consolidation continues to undo the consequences of prior interstate banking limits and generally has business economics strongly behind it.

Importantly, policy attention needs to be paid to the financial health of financial intermediation, particularly in light of the long-term slide of NIM. For the banking sector to provide liquidity for the U.S. economy, there needs to be sufficient spreads to attract depositors, cover expense and losses, and provide risk-adjusted yields. This is especially so for smaller banks that do little else other than local financial intermediation. 

With brand-name, online-only banks squeezing deposit gathering, and non-bank and monoline lenders doing the same in lending and payments, there is greater market pressure than ever that is breaking apart financial intermediation into component markets. Uneven non-bank versus bank regulation for consumer lending and deposit fees, as well as liquidity and capital impacts and money market mutual fund regulation, all have unintended consequences that cumulatively make banking more difficult and less profitable (and might also hinder macroeconomic policy). 

The larger banks will use scale, analytics, diversification, and national branding as offsets, though long-term profitability may well still suffer. Local community banks, and even many regional banks, have fewer options, other than consolidation. Policymakers need to grasp this threat to traditional intermediation. And more broadly, the public debate going forward needs to take into account the likely impact of additional change on the long-term profitability and health of the industry.