06/03/2011

A Short Leash on Risk


There’s an old saying in banking that bad loans are made in good times, so it comes as no surprise that the worst credit-quality downturn in nearly two decades was preceded by one of the frothiest economic expansions in recent U.S. history.

And for an experienced banker like Bob Matthews, chief credit officer at $18 billion Commerce Bancshares in Kansas City, Missouri, it seems like a sad case of déju00e0 vu all over again. Many of the mistakes that bankers made during this most recent credit cycle were made by an earlier generation in the previous cycle, which started with the savings and loan crisis in the 1980s and concluded with the collapse of the commercial real estate market in the early 1990s. It’s as if the institutional memory of an entire industry had somehow been wiped clean.

In a sense, that’s exactly what happened. There was no institutional memory because so many seasoned bankers who had lived through those scary times when institutions were failing in record numbers reminiscent of the Great Depression had long since retired to the golf course.

“Bankers forgot that we were operating in a credit cycle, and the operative word there is cycle,” says Matthews, who has held his position since 1989 and is highly regarded by other credit executives throughout the industry. Other than a short and shallow recession in 2001, which led to a manageable spike in commercial and consumer loan losses, the industry’s last significant asset-quality downturn was nearly 20 years ago.

“A whole generation of commercial lending officers and risk managers came and left during that time,” Matthews says. “We had a generation of relationship managers that had never seen a problem loan. People thought of [the economic expansion] as the ‘new normal.’”

It is tempting to blame the industry’s current loan-quality problems primarily on the sudden collapse of the U.S. economy in 2008, which resulted in the sharpest economic downturn since the 1930s. Banks are creatures of the economy, and the profitability and balance sheet strength of even the most conservative institutions will be tested by a severe recession. But not every bank threw its underwriting standards out the window when the economy was booming, including Commerce-which has performed considerably better than the industrywide average for asset quality.

According to the Federal Deposit Insurance Corp. in Washington, D.C., which collects a variety of performance data for the entire industry and publishes it quarterly, the ratio of nonperforming loans to total loans and leases in the second quarter of 2010 was 5.21%. Commerce’s nonperforming asset ratio in the second quarter was 1.06%-well below the industry average.

Commerce prides itself on maintaining prudent lending standards and a strong credit culture and clearly did not experience the same institutional amnesia that afflicted so many of its peers. The bank’s goal is always to rank in the top quartile among its peers for loan quality, and Matthews considers this to be one of the most important quarterly performance metrics that Commerce tracks. “We incorporate asset quality as a critical measurement of our success,” he says.

Bankers all across the country are beginning to tighten up their lending practices, often under pressure from their primary regulatory agency in Washington. In some instances banks are adopting sophisticated new analytic tools that will enable them to closely monitor and manage their concentrations across various asset classes and geographies, which has been a big problem in this most recent credit downturn. But technology is not a replacement for sound judgment, and it is equally important that bankers return to the prudent underwriting practices that were all but forgotten when the economy was booming. John White, founder and chief executive officer of Moorestown, NJ-based Precision Loan Solutions LLP, which provides loan review and risk management services to banks throughout the United States, says that when the economy was growing in recent years, too many institutions took short cuts when they were underwriting loans. “In the process of approving a loan, there should always be more nos than yesses,” he says.

When the recession of 2001 officially ended in December of that year, it was followed by six years of strong economic growth, low interest rates, and barely perceptible inflation. In retrospect, the confluence of low rates and the inflow of foreign investment funds created a bubble in the U.S. housing market, which peaked in 2006 before it exploded with spectacular effect. By December 2007, the U.S. economy had plunged into a recession of such magnitude that it set off a financial pandemic that rattled national economies, capital markets, and financial institutions around the world.

According to the National Bureau of Economic Research in Cambridge, Massachusetts, the recession finally ended in June 2009, although the U.S. economy is still dogged by slow growth and high unemployment.

Unlike the previous asset-quality downturn in the early 1990s, which was concentrated largely in the commercial real estate sector, this nosedive was widely felt throughout the banking industry, including residential mortgages, home equity loans, and credit cards. “This one was more broadly based over most of the asset classes,” says Dave Wilson, deputy comptroller for credit and market risk at the Office of the Comptroller of the Currency in Washington. Wilson believes that because the U.S. economy enjoyed nearly 15 years of exceptional growth, interrupted only by that brief recession in 2001, bankers and regulators were slow to react to a gradual deterioration in the industry’s loan underwriting practices-a condition he referred to as the “frog-in-a-pot” syndrome. “I think that was an issue for us at the OCC, for the lenders, and for directors,” he says.

Wilson also believes bank risk management professionals failed to challenge lenders over issues such as the sharp increase in leverage by many business borrowers, which left them unable to service their debt when the economy turned south.

“People who bring in the business tend to get the upper hand when the numbers are looking good,” says Wilson. “Risk managers have a hard time holding the line because the pushback is always, ‘Well, look at our balance sheet.’”

According to the most recent FDIC data, the industry’s overall asset-quality profile may have bottomed out in the fourth quarter of 2009, when the nonperforming asset ratio for all the loan categories it tracks hit 5.43% before dipping down to 5.40% and 5.21%, respectively, in the first and second quarters of 2010. However, the overall real estate sector continues to cause the greatest asset-quality headaches-particularly construction and development loans, where nonperformers surged to 16.87% in each of the first two quarters of 2010-and has either increased or remained steady for 16 straight quarters going back to the fourth quarter of 2006, when the nonperforming rate was just 0.79%.

One bitter lesson that bankers learned-or relearned-in this credit cycle is the importance of avoiding excessive concentrations in a single asset class. In the late 1970s and early 1980s, many Midwestern banks got into trouble when they loaded up on agricultural loans, only to see the farm economy collapse a few years later. Around the same time, this boom-and-bust underwriting mentality was mirrored by the large U.S. money-center banks that saw many of their developing country loans turn into losses during the Latin American debt crisis. When global oil prices collapsed in the mid-1980s, many oil and gas companies in the American Southwest were wiped out, contributing to the failure or forced sale of virtually every large bank in the state of Texas. Moving ahead, commercial real estate-particularly high-rise office towers in overbuilt markets like New York and Atlanta inflicted punishing losses on many U.S. banks in the early 1990s when that market tanked during a recession.

Real estate concentrations have been a big problem in this down cycle as well, although the big bugaboo this time around has been residential construction and development loans-many of which turned into losses when the heavily overbuilt new housing market collapsed in 2006.

“I’ve been in the business 37 years, and it’s amazing how we can keep making the same mistake,” says James M. Nelson, director of credit risk management at The Risk Management Association in Philadelphia.

Commerce Bancshares made its share of residential C&D loans as well, and according to Matthews this sector has been the single largest source of problem credits and losses at the bank. But the problem has been under control in large part because the bank has always strived to maintain a diversified loan portfolio with a 50/50 balance between consumer and commercial lending-and within the latter, a 50/50 split between commercial real estate and commercial and industrial loans. So while it did get burned when housing prices in its primary Kansas City and St. Louis markets collapsed, the losses were manageable because the asset class wasn’t permitted to dominate the overall portfolio.

“We have always worked to maintain a very balanced approach to loan portfolio asset generation and management,” says Matthews.

Maintaining this kind of portfolio balance can be especially challenging for community banks in suburban markets that may have trouble finding the kind of loan diversity that larger banks enjoy. A mainstay for such institutions has been local commercial real estate loans, including those for small warehouses, strip malls, and freeu2013standing retail establishments, according to Matthews.

“This is where community banks in suburban areas have made their living in the last 10 years.”

It can also be difficult for smaller banks to monitor their loan portfolio concentrations since they lack the sophisticated information technology capabilities of larger institutions, although this may now be changing as new automated solutions start to come on the market. Harland Financial Solutions, a Lake Mary, Florida-based technology company that focuses on financial services, markets an integrated credit management system called CreditQuest that includes a portfolio management module that provides senior management with a complete view of the entire loan portfolio, including asset concentrations, industry sector exposure, and customer relationship profitability. Quite often this data is locked up in a community bank’s core processing system and can’t easily be pulled out, and Harland helps the client identify the necessary information and migrate it to a new database.

“Institutions that are very large usually have sophisticated IT staffs, and they don’t face the same challenges of getting the data,” says Kimberly Songer, director of risk solutions at Harland. “Many smaller banks have core systems that don’t allow for this.”

Lenders and risk managers have also relearned the high positive correlation between asset quality and having a strong relationship with the customer. Greg Becker, president of $14.5 billion SVB Financial Group in Santa Clara, California, says most of the bank’s problem loans stem from situations where it did not have a close relationship with the customer. With a nonperforming asset ratio of just 0.34% in the second quarter of this year, SVB’s loan quality issues are minor compared to a great many institutions around the country. However, Becker points to one situation involving a loan participation where SVB was not the borrower’s primary bank. “We didn’t have a seat at the table with management to plan a course of action that would get the bank out whole,” says Becker. “You really need to have a relationship with the borrower.”

William L. Perotti, chief credit officer at Cullen/Frost Bankers Inc., a $17 billion institution in San Antonio, Texas, couldn’t agree more. Cullen/Frost has also come through this credit cycle in excellent shape. Its nonperforming asset ratio in the second quarter was 0.93%. But many of its problem loans have one thing in common. “We did not ask enough questions of our borrowers,” Perotti says. “We took too much at face value and didn’t probe enough.” Of course, it’s always easier to probe when the bank enjoys a deep relationship with a borrower that might be experiencing some difficulties.

“The deeper you can build the credit relationship, the fewer charge-offs you have,” Perotti says. “Being a borrower’s primary bank helps you manage risk. The biggest charge-offs often come from the thinnest relationships.”

Matthews echoes that sentiment as well. “The strategic business model for the company and for the loan portfolio emphasizes long-term and multi-dimensional relationships with our clients,” he says. “You’re more likely to have problems when you are taking a fraction of a customer’s business than all of their business.”

Banks are also feeling pressure from their primary regulators to strengthen their lending practices, which became a little haphazard during the economic boom. White at Precision Loan Review-which works with approximately 400 banks a year on a variety of credit-related issues-says the regulators themselves got a little lax in what they expected from their banks but that all changed once the industry’s asset quality began to deteriorate. “They’ve gotten much tougher,” he says. “It’s a little like they’ve gotten religion.”

And what exactly do the regulators want? “They want a better infrastructure for the whole lending process,” White explains, including the loan approval process, document flow, the bank’s risk management methodology, and how it determines loss reserves. “They want you to have policies and procedures, and you better be following them,” he says. In many instances the regulators have also been pushing for an independent loan review if they have substantive questions about an institution’s lending and risk management practices-and the regulators will ask questions about that process as well. “Are they qualified?” White says. “Did they do the review properly? And what did the bank do with the information?

Did it implement the firm’s suggestions?”

But the hardest thing many banks need to learn is how to say no when all of their competitors are saying yes. That’s not much of a problem now. With the economy still in recovery mode, highly bankable customers are still relatively scarce. But when the economy finally recovers and businesses and markets begin to grow again, will banks forget all the lessons of the past? Will they continue to make good loans, even when times are good again?

“It’s very important that banks develop truth challengers-or risk managers-who will tell management what is going on,” says Wilson at the OCC.

Oftentimes a bank’s most important lending decision is the loan it doesn’t make. Matthews says that one of Commerce’s core principles is to tread very carefully through unfamiliar territory. “We don’t undertake growth in a portfolio if we don’t understand the business, the market, or the product,” he says. For example, Commerce largely avoided the subprime and Alt-A residential mortgage markets because it didn’t like how the loans were being written at the height of the boom. In recent years those two asset categories have inflicted punishing loan losses on a number of large banks and played a leading role in pushing the U.S. economy into a steep recession.

“If you look at the performance of our loan portfolio over the last cycle, it’s because of what we didn’t do,” Matthews says.

Join OUr Community

Bank Director’s annual Bank Services Membership Program combines Bank Director’s extensive online library of director training materials, conferences, our quarterly publication, and access to FinXTech Connect.

Become a Member

Our commitment to those leaders who believe a strong board makes a strong bank never wavers.