Remarks by …

                                        Remarks by
                                       John C. Dugan
                                Comptroller of the Currency
                                         Before the
                                Florida Bankers Association
                                       Miami, Florida
                                      January 31, 2008


       I appreciate the opportunity to be here today with this large group of Florida Bankers.

While we appear to have a good mixture of national and state banks, I think it's fair to say

that most of you here today represent community banks. It's both my responsibility and my

pleasure to interact frequently with community bankers, both to make sure you hear what's

uppermost in our minds, and just as important, to keep my finger on the pulse of the industry.

You've heard me say it before, but it bears repeating: community bank supervision is critical

to the OCC's core mission and comprises nearly two-thirds of our employees, living in every

region of the country. We take a back seat to no one in recognizing the importance of

community banking to the health of the U.S. economy.

       In that context, let me return my focus today to a topic that I've addressed a number

of times during my tenure as Comptroller. It's one that's critical for many community banks,

especially here in Florida, and it continues to be very much on our minds at the OCC. I'm

referring, of course, to loans related to commercial real estate or "CRE." And in particular,

I'm referring to the challenges we face ­ both community banks and the OCC ­ from the

intersection of two inescapable facts: significant community bank concentrations in

commercial real estate loans, and the declining quality of a number of these loans, especially

those related to residential construction and development. Today, I'd like to talk briefly

about both of these facts, and then turn to our supervisory perspective and expectations.




                                               1
       Let me start with CRE loan concentrations. The banking agencies issued guidance in

late 2006 to remind everyone of the increased risk that arises from these concentrations, and

to reiterate our expectations when evaluating this increased risk. Those of you with long

experience in banking probably don't need to be reminded of real estate concentration risk.

It's a first principle of sound banking and bank examination that, the higher a bank's

concentration in a particular category of loans, the greater the risk that losses from that asset

class will affect the financial soundness of the bank. That principle proved to be a harsh

reality in the late 1980s for banks with concentrations in commercial real estate loans,

especially those in the Southwest and Northeast.

       But I can tell you from our discussions around the country that, as the result of a

prolonged period of exceptionally benign credit conditions, many community bankers had

become a little too complacent regarding the potential for significant stresses in these

markets. Lending growth was historically high in commercial real estate, especially in the

regions of the country that enjoyed an extraordinary boom in the housing markets. The result

was that CRE concentrations rose significantly in many banks around the country, even as

the quality of risk management practices lagged ­ which is why we felt the need to issue the

guidance.

       For example, looking at community banks as a whole, the ratio of commercial real

estate loans to capital has nearly doubled in the past six years, to 285 percent. Even more

significant than this overall industry statistic is the number of individual banks that have

especially large concentrations. Over a third of the nation's community banks have

commercial real estate concentrations exceeding 300 percent of their capital, and almost 30

percent have construction and development loans exceeding 100 percent of capital. Here in




                                                2
Florida, as in other states where housing is so important to local economic growth, the

concentration levels are more pronounced. Over 60 percent of Florida banks have CRE loans

exceeding 300 percent of capital, and more than half have C&D loans exceeding 100 percent

of capital.

        I guess it wasn't surprising, then, that the notice and comment process for the

concentration guidance prompted a full-throated chorus of criticism from community banks ­

many of you were very concerned that examiners would interpret the numerical thresholds in

the guidance as hard limits that would jam the brakes on CRE lending, which has been such

an important source of profit for so many community banks around the country.

        I spent a lot of time during this period explaining to bankers that this was not our

intent; that we simply would not implement the guidance in that way; and that our

watchwords would be "balanced supervision." I'm not sure I really convinced our skeptics at

the time, but I think a year of experience under the guidance has gone a long way towards

doing just that. In my many meetings around the country and in Washington during the last

year, bankers have come to recognize that we have not used the guidance to halt sound CRE

lending. Instead, we've used it to emphasize the higher risk that attends concentrations and

to stress the need for enhanced risk management to address that elevated risk.

        So, you may ask, if the guidance has been implemented more smoothly than many

had feared, why am I talking about it again here today? The answer is really pretty simple:

against the backdrop of elevated credit risk caused by CRE concentrations, we're now

entering a stage of the CRE credit cycle where problems have started to surface and losses

have started to increase ­ which is the second inescapable fact I mentioned at the outset.

During the last six months, we've witnessed unprecedented mortgage market disruptions and




                                                3
declining house and condo prices in many parts of the country. Indeed, as was widely

reported last week, the median sales price for an American single-family home fell in 2007,

the first such annual decline since at least 1968, and possibly since the Great Depression.

Price declines have been most acute in states that currently have the worst general economic

conditions ­ such as Michigan and Ohio ­ as well as those that experienced the greatest

house price appreciation in recent years, such as California, Arizona, Nevada, and, as I'm

sure you know better than I, Florida.

        The combination of these conditions is putting considerable stress on one particular

category of commercial real estate lending: residential construction and development ­ and

other categories of CRE loans will feel similar stress if general economic activity slows

materially. Indeed, given the traditional cyclicality of the economy and real estate markets,

and the relatively long lags between economic cause and real estate effect, prospects for a

commercial real estate project can change considerably between conception and completion.

Even a well structured and soundly underwritten project can become a problem during the

periodic overbuilding cycles that characterize these markets.

        This is hardly news here in Florida, given the well publicized slowdown and stresses

in commercial real estate markets in many parts of your state. But let me give you some of

our perspective based on our supervision of community national banks doing business here.

In particular, I'd like to focus on construction and development loans, a good chunk of which

relate to residential real estate.

        As I mentioned previously, banks in Florida have relatively larger C&D portfolios

than the national average, which caught my eye because of trends we're seeing at the national

level. While overall commercial real estate performance has been sound, weaknesses have




                                               4
plainly begun to emerge in C&D loans, primarily related to the slowdown in residential home

sales. Indeed, during the past year national community banks have experienced a significant

increase in nonperforming C&D loans. As of Sept 30, these loans amounted to 1.96 percent

of total C&D loans, a rate that was more than twice that of a year earlier. Although starting

from an admittedly very low baseline, an increase like this ­ over 100 percent in a single year

­ is clearly a trend that we need to monitor closely. And in Florida, that trend is even more

pronounced. While nonperforming loans a year ago were 40 basis points less than the

national average, the figure has increased to 3.34 percent of total C&D loans. That's 70

percent greater than the national average and an almost eight-fold increase in one year.

       Of course, this reported data is not our only measure of declining credit quality.

Those of you who are national bankers know very well that commercial real estate

concentrations have been a topic of discussion at your examinations for some time. Over the

past four years our examiners have assessed your risk management practices in an effort to

ensure they are appropriate for the level of risk you have assumed. These efforts continued

all over the country in 2007 and have extended into 2008.

       For consistency in these assessments of existing practices, we have brought together

teams of highly experienced examiners to look horizontally across banks with higher CRE

concentrations. These "horizontal reviews" allow us to identify and convey best practices

more effectively, and provide consistent advice on any additional measures that we believe

should be taken. They also allow us to drill down in a consistent way on loan files around

the country to make our own assessments about declining asset quality in the current

environment.




                                               5
        So let me tell you what we're learning, both in terms of risk management practices

and asset quality, starting with the better news. When we started these horizontal reviews,

we found that many banks lacked robust risk management practices and sufficient

management information systems to adequately identify risk exposures. Over time, we

began to see improvements, especially in risk management policies and in boards of directors

articulating the level of risk they believe is appropriate to take, by product, loan structure,

and overall borrower creditworthiness. We have also seen more use of stress testing, at least

through the analysis of the effect of different interest rate scenarios. That has all been to the

good.

        But we've also continued to observe a number of risk management deficiencies that

are a cause for concern. For example, despite our previous guidance, a number of banks with

CRE concentrations have not extended their stress testing of income-producing properties

beyond interest rates to other business variables that affect risk, such as vacancy rates, lease

rates, and expense scenarios ­ not only at the time the loan is made, but also periodically

throughout the life of the credit relationship. The potential for rapid deterioration in this

business is simply too great not to conduct such testing on an ongoing basis.

        Another issue that has surfaced in the horizontals involves real estate appraisals. We

have seen an increasing number of instances in which appraisals on file have become

outdated with respect to current market conditions. That in turn can make it very hard to

assess the true credit quality of loans on the books.

        In terms of asset quality, our horizontal reviews have indeed confirmed a significant

increase in the number of problem residential construction and development loans in

community banks across the country. Not surprisingly, but perhaps not as visibly as has been




                                                 6
the case with larger builders, declining residential sales and housing starts have taken a toll

on smaller builders. As sales slow, holding periods have extended both for raw land and for

homes under construction. This has had a direct effect on collateral values and is forcing

many lenders to seek guarantor support and additional collateral, at a time when neither may

be willingly offered. Likewise, more prospective buyers are abandoning planned purchases

of homes and condos from developers where the general decline in housing prices has made

it more difficult to obtain mortgages. The combination of these circumstances has begun to

translate into more problem assets and increased provisions for loan losses.

       In saying this, I recognize, and I think we all need to recognize, that the recent

increase in nonperforming CRE loans must be viewed in context. The baseline from which

the increase began was an exceptionally low level of problem assets that resulted from the

long period of benign credit conditions. An increase in nonperforming loans from such a low

baseline is hardly unexpected, and thus far overall nonperforming CRE loans, even in the

area of residential construction and development lending, are a long way from approaching

historical peaks. Nevertheless, the trend is unmistakable, and the potential consequences are

magnified in this credit cycle by the fact that so many community banks have CRE

concentrations that are so much higher than has ever been the case in the past.

       Given these circumstances, what do we see as the consequences in the coming

months? Not surprisingly, there will be more frequent interaction between supervisors and

banks with concentrations in CRE loans that are declining in quality. There will be more

criticized assets; increases to loan loss reserves; and more problem banks. And yes, there

will be an increase in bank failures. Last week we saw the first failure of a national bank in

nearly four years ­ the longest such period in the 145-year history of the OCC. I am quite




                                                7
sure that the period before the next one will not be nearly so long. As the credit cycle turns,

the failure rate is very likely to return to the more typical levels that we have experienced in

most other years.

       So as we enter this more stressful period for community banks with concentrations in

commercial real estate lending, let me flesh out our supervisory approach and expectations at

the OCC. Our fundamental message is two-fold. For our part, OCC examiners are not afraid

of banks with high levels of classified CRE assets; we have lots of experience working with

these institutions, and our basic goal is to help them manage through the attendant issues and

problems. For the industry's part, we expect bank management to be realistic about

identifying problem assets themselves, so that our examiners are not forced into the position

of having to do this for them. The idea is to recognize problems early and manage through

them, with good and continual communications between examiners and bankers, before the

problems fester and get worse.

       Why am I so focused on this aspect of the problem? To be honest, it arises from the

lessons we learned from dealing with the wave of problem CRE assets in the late 1980s.

Critics thought that the banking agencies waited too long to address mounting problems; that

by the time they did the problems were too big to be easily addressed; that the agencies then

were forced to take drastic action, including slamming the brakes on CRE lending; and that

there were too many communications breakdowns as problems intensified. While there were

many other factors at work that contributed to the severe CRE problems of that era, we agree

with the basic point that it's much better to get at problems early when they're smaller and

more easily managed, and to keep communications lines open as we do.




                                                8
       In fact, I think it's hard to argue with this basic point. But I also think there is often

more resistance and reluctance to recognize a problem in its early stages, before it gets big

and stares us in the face. That's why I want to emphasize today, as we see the clear signs of

CRE credit quality declining, that we will expect banks with CRE concentrations to make

realistic assessments of their portfolio based on current, changed market conditions. That

may require you to obtain new appraisals. For those of you in stressed markets, it will almost

certainly require you to downgrade more of your assets, increase loan loss provisions, and

reassess the adequacy of bank capital. These are normal, expected steps to take in these

circumstances, and I can't stress enough how important it is for you to make these realistic

judgments yourselves, based on sound credit administration practices, instead of forcing our

examiners to try and make these same judgments in the first instance. This includes

monitoring, and curtailing if necessary, interest reserves that might have been established at a

loan's inception. As I said, our job is to provide a balanced approach to working with banks

as they manage through higher levels of problem assets, which is doable; that job becomes

far more difficult when management fails to recognize problems in the first instance.

       So, as we continue with our horizontal reviews and other CRE supervision this year,

we won't be trying to supplant your assumptions of market trends and conditions with our

own, as long as yours appear reasonable and well supported. But we will be talking with you

about how those assumptions were developed and how they are actually materializing.

While we normally will not impose regulatory parameters on your assumptions, we will

expect banks with concentrations to have a better risk management system for that portfolio

than lenders who don't have such concentrations. We will also expect realistic grading of

credits, which means not only properly assessing the collateral support that may exist, but




                                                9
also realistically valuing that collateral. And, where changed market conditions are very

likely to have changed the value of the real estate, it will be quite important to order new

appraisals ­ even where the existing appraisal is otherwise current.

          This leads me to one final, and related, point. I hope I've made clear that banks

with increases in problem CRE assets will likely have to increase provisions to their loan loss

reserves to ensure adequate coverage. But I also know that, in recent years, some bankers

have been fearful of criticism from their external auditors that the bank's loan loss reserves

were too large. As we and other banking agencies have stressed, your methodologies for

arriving at your loan loss reserves, and the reserves themselves, need to be sound and well

documented. But I firmly believe that in this environment, increases in loan loss reserves for

many banks are both warranted and prudent. I would be extremely surprised if your auditors

disagreed with this position ­ but if they do with respect to a national bank, I urge you to

contact your examiner-in-charge or Assistant Deputy Comptroller. If we have to intervene in

this situation, we will not hesitate to do so.

        In closing, let me say that I remain hopeful that the decline in CRE credit quality is

not as great as some are predicting. But even if it is, we plan to work with banks to address

the problems as they arise. As I've tried to convey in my remarks today, that process works

best with a combination of early and realistic problem identification by bank management;

frequent and robust communication between bankers and examiners; and balanced

supervision. If we do all that I believe we'll go a long way towards achieving our common

goal: to make sure that banks remain safe and sound so you can continue to meet the needs

of your customers.

        Thank you very much.




                                                 10