Thank you for being here this evening. I thank Professor Scott Pardee, a Federal Reserve alumnus, for the invitation to speak with you. This lecture series honors Alan Holmes, a respected former senior official at the Federal Reserve Bank of New York. It is an honor for me to participate. It also is a pleasure, since my son Francis is a recent Middlebury graduate. We have come to know Middlebury the College, and Middlebury the town. Both are special places.

In my talk this evening I would like to look at the roles the Federal Reserve Bank of Boston, and the national Federal Reserve System, have played in helping to deal with crises. I will look at four cases. In each case I think I the Federal Reserve has performed well. Also, in each case we have learned some lessons about how to do better.

You might find it interesting to see what we have learned about crisis management. I hope you also find it interesting to see at work the different lines of business activity inside the Federal Reserve, and to see how each can contribute to maintaining financial stability and public confidence.

In describing what we do in the Federal Reserve System we often use the metaphor of the "three-legged stool" . Our three major lines of business are: monetary policy and economic research; supervision and regulation of major portions of the banking industry; and payment services, in which we process cash, checks, and electronic payments for banks, and protect and improve the national payments system for consumers and businesses. You will see that a particular crisis or threat may have exercised one or another of these lines of activity, but looking across all of the cases, all three functions had essential roles.

I would like to look with you at four experiences that have tested the Federal Reserve:

1. The failure of the Bank of New England and the collapse of the insurance fund for Rhode Island credit unions, both of which happened in January, 1991;
2. The buildup to the “Y2K” event, during 1998 and 1999 and into the new year 2000;
3. The awful events of September 11, 2001 and their immediate aftermath;
4. And, more briefly, the devastation brought by Hurricane Katrina two months ago. 

The Federal Reserve Bank of Boston was involved extensively in the first three events. We have not been as involved with the response to Katrina, but other Reserve Banks closer to the affected region of the country have been very active in response, and I want to share some of what they have done.

I will begin with a look at the banking crises the Boston Fed faced in January, 1991. The U.S. economy went through a recession in the early 1990's, the national banking system went through difficult times, and here in New England the problems were acute.

During the early 90's New England lost over 20 percent of all of its depository institutions. In most of these cases the banking regulators worked to have failing banks and thrift institutions taken over by stronger institutions.

Bank of New England's problems reflected the deteriorating national and regional conditions. They also reflected the recent history of the bank itself. Bank of New England was the second largest bank in the region, and had a 200-year heritage.

The bank had grown rapidly through the 1980's as a result of an aggressive acquisition campaign. Two fairly large banks, one headquartered in Boston and the other in Hartford, Connecticut, had combined under the Bank of New England holding company. This new organization bought several smaller banking organizations across New England to expand its presence in the region. As it turned out, some of the acquired organizations had loan portfolios with higher risks than the Bank of New England may have realized. The bank's own lending sometimes was aggressive as well. In the general economic downturn, and the sharper downturn in New England, loan losses and nonperforming loans took their toll on the bank's earnings and capital position.

The bank's difficulties came into full public view in December, 1989, when a quarterly dividend announced on November 27 was rescinded by the bank.

Taking back an announced dividend is something that happens rarely. It was a sign that bank still was in the process of realizing how serious its problems were.

In rapid succession the bank's chief executive officer resigned; other senior officials were terminated by the board of directors; an interim CEO was appointed; and six weeks later a new CEO was hired.

The Federal Reserve Bank of Boston was the supervisor of the holding company of the Bank of New England organization. The Office of the Controller of the Currency, the OCC, supervised the individual banks within the holding company. This division of responsibility reflected federal law for the regulation of nationally chartered banks.

Throughout 1989 and 1990 we worked closely with the OCC, and also with the Federal Deposit Insurance Corporation, the FDIC, which insures bank deposits, to protect the public interest as Bank of New England grappled with its problems. A particular challenge we faced was how to provide liquidity to the bank so that it could continue to operate and pursue every prudent means to resolve its problems. A sudden failure of a bank of that size, during an economic decline, not only would cause problems for the bank's customers and creditors. It also could shake public confidence in the banking system, and could cause a domino effect with consequences for other major banks.

On the other hand, liquidity provided by the Federal Reserve Bank of Boston, through the discount window, had to be fully collateralized, so that the Reserve Bank, and ultimately the taxpayer, would not lose money if Bank of New England were unable to repay its loan from us. The most significant unencumbered asset that could serve as collateral was the loan portfolio, but recent events had shown that Bank of New England had been underestimating the deterioration of its loans. Moreover, transporting all of the legal documentation necessary to perfect our interest in the loans to our vault in Boston would have been a logistical nightmare.

Our solution was twofold: to create a schedule of deeply discounted "haircuts" that resulted in conservative collateral values for various categories of pledged loans, and to establish what we called "field warehouses" . We sent teams of people from our bank to six Bank of New England locations in Massachusetts and Connecticut, to gather, inventory, and take physical custody of the documentation for billions of dollars worth of loans.

Each evening, the Boston Fed compared Bank of New England's request for borrowed funds with that day's valuation of collateral in our possession, and authorized the lending accordingly.

These field warehouses were a huge undertaking. They required staffing that far exceeded the size of our lending function. We recruited people from just about every department of our Bank to help. About 125 employees, or nearly 10 percent of our total workforce, helped with the warehouse work during several months of operation.

The field warehouse program was highly unusual for us, but it worked. We lent funds extensively to Bank of New England, fully collateralized and with our own knowledge, from our hands-on familiarity with the collateral, that the taxpayer was protected. Bank of New England did not have a sudden, disruptive collapse. Its management used the year 1990 to explore every feasible means to keep the bank going. In the end, there was no significant improvement in the economy, and the bank's problems were just too pervasive for a return to health.

At the end of the first week of January 1991, the banking regulators arranged for the bank to close, but have its business activities carry on in receivership under the FDIC, so that savers and borrowers could continue to do business. Healthy banking organizations were invited to bid to acquire the business of this "bridge bank" , with the bad loans and the nonperforming assets set aside for liquidation. Fleet Bank was the winning bidder. Years later, in 2004, Bank of America acquired Fleet.

By the time Bank of New England closed, on Sunday, January 6, 1991, the Federal Reserve Bank of Boston already was dealing with another local banking crisis: the sudden closure of 45 institutions in Rhode Island.

Many of the credit unions and other thrift institutions in Rhode Island were insured by a private insurance fund, not by federal insurance. Late in 1990 the private insurance fund became insolvent, for reasons that turned out to include apparent fraud at one of its insured institutions.

On New Year's Day, 1991, a new governor took office in Rhode Island. He had been briefed by the Boston Fed and other regulators on the situation, and in his first hour as governor he ordered the closing of 45 institutions, until they could obtain federal insurance.

These institutions held 300 thousand accounts for consumers, and for some businesses. In a state with a population of 1 million, the closings had a dramatic effect.

The biggest concern, of course, was what would happen to the savings of the deposit holders whose closed credit unions basically had no insurance. Some of the institutions were not able to reopen, and this problem took years to resolve. With state and federal funding, depositors eventually got most of their money back.

There were other issues which threatened to have immediate adverse effects on the broader population, and on the broader banking system.

People in Rhode Island and in southeastern Massachusetts were shaken by the dramatic holiday closing of these institutions. They wondered whether their own credit unions and banks were safe. Some people started to take their money out of the banks. This became a news story. When still other people saw news reports with people streaming into their banks to withdraw their savings, some of them started to do the same thing.

The Rhode Island story attracted national coverage. One report on national television had the reporter standing in front of a bank, just as a background image. That bank was not one of the closed institutions. However, its name was visible behind the reporter, and that news report started something of a run on that bank.

While this problem was developing, we had another one coming. January 3 was the pay date for Social Security recipients. Most of them received their monthly payment by electronic direct deposit into their bank accounts - or their credit union accounts.

With 45 closed institutions, thousands of recipients in Rhode Island would have no access to their Social Security payments, and most recipients really needed that money.

These sudden problems tested our payment operations at the Boston Fed. We supply cash, that is, currency and coin, to the banks across New England . The incipient runs on Rhode Island institutions meant that some of them were running out of cash. Even the safest and soundest banks only hold amounts of cash on hand that are small relative to their total deposits. A run of any size can exhaust a bank's supply of cash.

In an environment with a lot of nervous people, any news that even one bank has run out of cash might start a panic.

Our Cash people arranged emergency deliveries of currency to numerous Rhode Island institutions. However, the situation was so tense that we had to do more.

So, we transported $60 million in currency to an armored truck company's storage facility in Rhode Island, so that it would be in the state, and immediately available if needed. This was highly unusual. It was a lot of money, for which we were responsible, outside of our highly secure vaults in Boston. We arranged for the Rhode Island State Police to guard the warehouse while we had our money there for several days.

Some of the scenes during those several days would remind you of an old movie about the Depression, with armored trucks pulling up to a bank, and uniformed couriers carrying in bags of money while people crowded the lobby. It was not quite that dramatic, but it was close to it. People needed to know that they could get their money if they needed it. Our response, with an abundance of cash, helped them to have that confidence.

It took nearly two weeks, but people calmed down, and business returned to normal at the healthy institutions.

Meanwhile, we did not have much of a solution for the Social Security recipients. We offered to work with the U.S. Treasury and the Social Security Administration to identify the affected recipients and see if Social Security could issue checks to them as soon as possible. However, Treasury and Social Security asked us to try harder.

It was important to them that the thousands of recipients get their money on time, or as close to on time as possible. Also, it was important that the integrity of electronic direct deposit be maintained. The Treasury, Social Security, and the Federal Reserve have been advocates for a more electronic payments system in this country for many years. Now, in 2005, it is happening in a big way. Then, in 1991, it was not nearly as far along. If the problem had to be resolved in a way that made electronic direct deposit look more risky than getting a paper check through the mail, that could have set back the progression toward electronics.

So, we tried harder. We asked if a commercial bank in Rhode Island would volunteer to work with us, to pay out money to Social Security recipients who were not the bank's customers. We said we would work with the bank, and with Social Security, to reconcile it all later. This was not an appealing opportunity for a bank. However, Citizens Bank stepped forward.

Newspapers and television told the recipients who had accounts at the closed institutions that they could come to Citizens Bank, show evidence that they had direct deposit at a closed institution, and get their monthly payment, in cash.

It worked. The recipients got paid. Direct deposit did not get a bad name. It took Citizens Bank and us six months to untangle and reconcile all of the back-room accounting involved, but it was worth it.

Overall, our bank did well in dealing with the Bank of New England and Rhode Island credit union crises. We took a couple of important lessons from them, too.

In hindsight, we thought we should have done more, earlier, to realize how badly Bank of New England's condition was deteriorating prior to late 1989. We recognized that the separate responsibilities of the Federal Reserve and other regulators, and our normal ways of focusing on our own responsibilities, might have led to some balkanization of information that we all needed to see and assess together. We have worked more closely with the other regulators since then.

We also were discovering that a key success factor in crisis management is to have the gumption to do whatever we have to do to get to a good outcome. This means being willing to do things we never did before. It means finding a way. We did that when we set up the field warehouses for Bank of New England's collateral.

We did that when we took cash out of our vault and stored it secretly in a Rhode Island warehouse. We had to be pushed to do that for the Social Security recipients. When pushed, we found a way. That taught us to push ourselves harder to find creative ways to do what is needed to avoid or resolve a crisis.

The Bank of New England experience was about working hard to keep a crisis from happening. The Year 2000, or Y2K, experience also was about working hard to keep a crisis from happening. Once again the Federal Reserve Bank of Boston, and the national Federal Reserve System, performed well, and learned more lessons about crisis management.

Looking back from October of 2005 to the advent of the year 2000, with the knowledge that the U.S. and the world had no significant problems, it may be hard to believe that any kind of crisis was possible. Believe me, though, it was.

You may know that the essence of the Y2K problem was that many computer systems would be unable to recognize the change from 1999 to 2000. More specifically, they were not programmed to recognize any year that did not begin with "one nine" . The origins of the problem were in software conventions that originated in the 1960's and 1970's, when computer memory was an expensive and therefore a scarce resource. To conserve memory, each year of the calendar was represented by just the last two digits: 72 for 1972, for instance. Remarkably, this convention endured long after memory became relatively cheap. Many older software programs also endured into the late 90's.

As a result, tens of thousands of systems had to be modified or replaced before the year 2000 began. While this problem was well known in information technology circles for a long time, it did not get real attention and action until the late 90's. It was one of those things that had to be done eventually, but would cost a lot of money without much return, and could be put off for a good while. So, it was put off, until, by about 1998, organizations could not put it off any longer.

We know now that here in the U.S., and worldwide, the necessary remedial work was done, and the world moved pretty smoothly into the Year 2000. That, of course, was the objective of all the hard work people did to prepare their computer systems.

In the Federal Reserve System, with our work on Y2K we found ourselves moving steadily outward: from remediation of our internal systems; to working with all the banks in America to make sure they would be prepared; to providing factual reassurance to the public. Protection of public confidence, in the banking system and in the broader economy, became our challenge as we entered 1999.

By the beginning of 1999 we knew that our essential Reserve Bank systems were ready for Y2K. Also, we had required banks to test the interfaces between their computer systems and ours. That testing was underway and going well.

Moreover, the Federal Reserve and the other bank regulators had a comprehensive supervisory initiative in full force to examine the banks' preparations for Y2K. From our work as a regulator we knew that the banks were well along with their preparations.

Nevertheless, there was public anxiety about Y2K, as it pertained to banking services, and more broadly. A fundamental problem was that intensive work on the Y2K problem began so late; in most organizations it began in 1998, or maybe 1997. So, as Y2K became more of a popular news story in 1998, most firms could only say, "We're working on it" . It would have been better if they could have said, "We've already fixed it" , but they had not. And their lawyers may have been advising them not to provide a lot of reassurance, because of legal liability if it turned out otherwise.

In fact, the Congress passed legislation in 1998 to shield firms from liability if they made positive statements about their Y2K preparations, in good faith, and then something went wrong. That law passed because the public needed reassurance and was not getting it.

Instead, the public was reading and viewing colorful news stories about people who were moving into the forest to prepare to live without electricity; or people who were stockpiling bottled water, beef jerkey, and Spam; or people who were taking their money out of the bank and hiding it at home to keep it "safe" .

There were stories about the dangers of airplanes not being able to stay in the air after midnight on January 1, 2000 . There were predictions, by reputable economists, that the U.S. would experience enough disruption from Y2K to bring about a recession. Most of the news about Y2K induced worry, even fear. We had been reluctant during 1997 and 1998 to move beyond getting our own systems ready, and doing everything we could to ensure the banks would be ready. By the end of 1998, however, we realized that we had to do more. We had to stick our necks out. We had to bolster public confidence. We had to counter fear with facts.

The Presidents of the 12 Reserve Banks, and the Governors on the Federal Reserve Board in Washington, launched what for us was an unprecedented public information campaign, with the Presidents, the Governors, and others going out personally to meet with newspaper, television, and radio editors and reporters and seek coverage of the facts about Y2K.

We are not people experienced or at ease with news media. This did not come naturally to us. We just steeled ourselves to do it, because that was what it would take to help to preserve public confidence, especially in the banking system.

At the Boston Fed, our President, Cathy Minehan, led this outreach for New England. I helped. I made several media visits, including two in Burlington, Vermont - close to Middlebury of course. One morning I spoke with the editors at the Burlington Free Press, and they ran a thoughtful, positive story the next day. Then I went to the Channel 3 television studio, expecting to give the station manager and one or two of his colleagues a briefing. Instead, they brought me into the newsroom, sat me down with a reporter, brought over a cameraman, and asked me to do an interview on camera. We do not look for opportunities like that. In fact, we do not regard them as opportunities. I was nervous doing it, but I did it.

Of course, in television the news is visual, and the station manager knew that a taped interview he could run on the evening news was the best way to deliver the message.

We Federal Reserve people were not the only people who went public with positive messages about Y2K. We may have been the first to do so in a coordinated national effort. It certainly was a first for us. And we believe it helped the public.

We took other unusual initiatives to keep Y2K from becoming a crisis. Cash is the backup means of payment when other means, such as credit cards, debit cards, and electronic transfers, do not work. It was highly unlikely that there would be a failure of any of those mechanisms. Still, it was a possibility on the public's mind, and the possibility of other banking problems also was on people's minds. We decided to have extra supplies of currency in our vaults, just in case, and to tell the public about it.

Moreover, in an initiative that recalled what the Boston Fed had done in Rhode Island in early 1991, all Reserve Banks stored supplies of cash in commercial bank vaults across the country, to have it immediately accessible if needed.

It was not needed, and we brought the cash back into our vaults in early January 2000. However, it was a good insurance measure.

In monetary policy the Federal Reserve also took a measure of insurance. The U.S. had a very strong economy in the late 1990's. During 1999 the Federal Open Market Committee, or FOMC, had been raising the target Federal Funds Rate pretty steadily. After an increase at the November FOMC meeting, the Committee did not raise the target rate at its December meeting. This was at least mildly counter to the FOMC's trend during that period, and I think we can read it as allowing a little bit of a cushion for the Y2K event.

I believe the Federal Reserve contributed materially to keeping Y2K from becoming a crisis, particularly for the banking system. We contributed effectively in the ways we had planned: by fixing our own systems, by testing with the banks, and by active supervision. We wound up contributing in ways we had not imagined, with a publicity campaign and with stockpiles of cash across the country.

One lesson learned was that sometimes it is not enough for the Federal Reserve to have done its own work well. Sometimes it is not enough even to have made sure the banks did their work well. Sometimes the Federal Reserve has to put its credibility to use, and be more public, more extroverted, than is customary for us, to serve the public interest.

Another lesson learned was the importance of cash: making sure it is available, and making sure banks and the public know it is available. This prominent role for cash would come up again in future crises.

Once more, we were learning the importance of doing whatever it takes, to ward off a crisis, or to resolve one.

The most devastating American crisis in our lifetimes was the attack on our country on September 11, 2001.

I would like to tell you how the Boston Fed and the national Reserve System responded to this crisis, which exercised all three legs of the stool: payment operations, bank supervision, and monetary policy. Here again, I believe we performed very well. Here again, we learned some lessons about crisis management.

Shortly after the second plane hit the World Trade Center, we in Boston started to take measures to protect our people and our facilities. We suspended all access by vehicles to our building. We installed at our entrances metal detectors and x-ray machines so that any people entering the building and anything they carried would be tested. We asked our staff who work on higher floors to relocate to the low-rise wing of the building, which has four floors above ground and additional space below ground.

Our working conditions were hectic for awhile. We were trying to determine what we needed to do locally to maintain essential operations. Simultaneously, we were participating in numerous conference calls involving management at multiple levels from all twelve Reserve Banks and the Board of Governors, to understand what was happening in each region of the country and work together to keep essential activities in operation. We also wanted to allow many of our people to leave Boston and go to their homes, and it took some time to sort out where we could operate without staff, where we needed to keep staff with us, and where we needed additional staff to shore up essential activities.

Thanks to the Bank's great people, we managed to remain open and keep all of our essential operations intact. As the day went on, we began to see where we had opportunities to avoid disruptions by taking some initiatives. I mentioned that we had suspended all vehicular access to our building during the morning. Within a couple of hours of doing so, we began to hear expressions of concern from some banks about obtaining adequate supplies of cash from us. When armored carriers could not come into our building, they also could not leave our building with new supplies of currency. Some banks had an immediate concern that if this restriction remained in effect for even a day or two, they might not have sufficient cash to stock ATM machines for customers and make cash available to businesses.

When we realized that this initiative we had taken to protect our people and facilities could have adverse effects, we worked with our Protection Department to set up the capability to inspect armored vehicles on the sidewalk before they entered the building, and then, upon successful inspection, to allow them to enter. We sent a broadcast message electronically to the depository institutions of New England to advise them that deliveries and shipments of cash had resumed, and we would be providing cash services as usual. We believe this quick modification helped to avoid any cash shortages, or even any concerns about cash shortages, in New England.

The government suspended all air transportation immediately after the attacks on the World Trade Center. Among other effects, this meant that many checks could not be collected, because the Reserve Banks and correspondent banks use air transportation to collect checks which must travel beyond certain distances to reach the paying banks. We in Boston, and the Federal Reserve nationally, anticipated that delays in check collection, which could have lasted for an indefinite period, could have disrupted life for consumers and businesses. If banks knew that they could not collect checks, and had no sense of when normal collection might resume, some banks might have discouraged deposits of checks from their customers. In turn, this might have discouraged some merchants from accepting check payments. People who received paychecks might not have been able to cash them.

With this concern in mind, the Federal Reserve Banks announced that they would continue to accept checks as usual, including checks from banks which did not normally deposit with the Federal Reserve, and would pass credit according to normal funds availability schedules. We knew that we would not be able to collect many of these checks, but we passed credit anyway, and absorbed the float, so that the check payment system would not be disrupted, and consumers and businesses would not suffer.

Our bank supervision people were contacting banks across New England throughout the day on September 11, and during the days thereafter as well, to assess their conditions. A major concern was that banks in New York would not be able to complete transactions with the banks in New England and elsewhere, causing funding shortages. The corporate customers of banks also could be strapped for funds if transactions between banks could not flow.

A by-product of this outreach was that our bank supervisors became "information central" for banks around New England. We received calls from banks in most New England states, looking for all sorts of information. They had called the state banking commissioners' offices in their states, but some of those offices had closed and sent everyone home because of the crisis. The word was out that the Boston Fed was open, so we became the place to call to get information or help.

Boston's bank supervisors conducted a broad swath of surveillance and research activities in conjunction with the Board of Governors and other Reserve Banks.

An example of this cooperation was extensive attention to the insurance companies impacted, including their financial condition, loss expectations, capital levels, and ultimately which banks had elevated exposure to these insurers.

While insurance companies are outside of the Federal Reserve's supervisory responsibility, they are a significant and interconnected portion of the financial services landscape. Part of crisis management is to put your organization in a position that allows for a rapid and informed response to a situation.

One of the operations we provide in Boston is what we call offline funds and securities transfer, and during that week it really helped banks to complete transactions and get money where it was supposed to go. While most Fedwire Funds Transfers and Securities Transfers are completed electronically, through computer connections, some institutions, particularly very small institutions with relatively few transfers to do, prefer to execute their transfers "offline" , by telephoning us with their transfer order and then having us call them back to verify their order. The Reserve Banks have consolidated these offline operations from twelve to two Banks, including Boston. The half of the country we cover includes New York .

On September 11, and for the rest of that week, our offline service provided a vital backstop for some of the largest banks in the country, when they had difficulty executing transfers electronically with their contingency facilities. Our people in Boston worked long into the night to transfer tens of billions of dollars through this offline service for large institutions, which probably never had used the service previously. Normally we transfer about 2 billion dollars per day offline. For the four days, September 11 through 14, we transferred 155 billion dollars. Our ability to use this operation as a lifeline for some of the largest banks helped to keep the national payments and financial system in effective operation.

The Federal Reserve System also was highly attentive and active in multiple ways to try to ensure adequate levels of liquidity in financial markets, so that markets would not freeze in response to the disruptions that had occurred, such as the closing of the stock markets. The Reserve Banks were very active and accommodating lenders at the discount window, to help banks with funding shortages occasioned by disruptions, such as in the overnight federal funds market, which was not operating normally because so many participating firms were not in operation. On evenings when normally we might have lent a few hundred million dollars at the discount windows of all Reserve Banks combined, we lent amounts above twenty and thirty billion dollars. These loans all were with collateral, but their order of magnitude gives you some sense of how we stepped up to meet unexpected, urgent needs at many banks across the country. The Boston Fed extended substantial loans to banks in New England which had deficits related to operational problems at New York banks. The check float we absorbed also had the effect of adding liquidity to the banking system.

The Federal Open Market Committee decided on September 17 to lower its target for the Federal Funds Rate by fifty basis points. In announcing this move, the FOMC also made a point of stating that the Federal Reserve would continue to supply unusually large volumes of liquidity to the financial markets, as needed, until more normal market functioning was restored.

September 11 really tested all three legs of the Federal Reserve stool. We did well, and the continued operations of the national banking system helped to maintain public calm.

We also learned important lessons from our efforts to respond to 9-11. We learned a new dimension of contingency backup. Traditionally, backup had been about computer systems. If your primary computing location was disabled, by a power outage or a storm, you needed a backup site to keep your systems going. The experience of many banks and other organizations on and after 9-11 showed that all enterprises need to prepare for situations when their people might not be available. In the case of 9-11, many people were killed, or injured. We hope we never see such an event again. However, many kinds of events can make it impossible for people to work - Hurricane Katrina, for instance.

At the Boston Fed and at the other Reserve Banks we have augmented our backup plans to address the potential for key people not being able to get to work. Many more people now have remote computer access, so that they can work from home if conditions make our building unusable, or if travel is not safe.

In addition, for particularly vital operations we have backup arrangements with people at other Reserve Banks. They perform our duties periodically so that in the event of a crisis in Boston, they can pick up the work immediately and keep it going. In turn, Boston provides similar backup for functions at other Reserve Banks.

Another, even more vital lesson we learned is the importance of constant communication with our people during a crisis. Frankly, we did not do this well on September 11. We were very focused on keeping essential services going, and on external communications with New England banks and the other Reserve Banks.

We did some things to inform our people about what the Bank and the Federal Reserve System were doing. However, we found out later that a lot of what we thought was being communicated was not being communicated. We in senior management were not focused enough on this need. Beyond that, it was nobody's specific job to stay on top of internal communications.

This experience taught us to pay more attention to internal communications during a crisis. We designated a small crisis communications team who now work with the President of the bank and me whenever a disruption or threat develops.

We do tabletop drills regularly to test ourselves against crisis scenarios. One of the key elements in these drills always is to communicate, early and repeatedly, with our people. We have implemented an email notification system, telephone calling trees, electronic message boards, and even a new public address system throughout our building, to support broad and timely communication.

People and organizations never can be perfectly prepared for crises, but we are better prepared as a result of the 9-11 experience.

I want to conclude with a brief look at Hurricane Katrina and its aftermath. This crisis did not affect New England or the Boston Reserve Bank much. Boston provided assistance to the Reserve Banks more directly involved. The responses of those Reserve Banks are interesting in that they reflect lessons learned from the earlier crises by Boston and the Federal Reserve System. They teach us a fresh lesson, too.

The Federal Reserve Bank of Atlanta has a branch in New Orleans, right in the middle of the downtown, with 160 employees. The hurricane did not flood the branch building, but the building had to close because of the conditions in the city. Sadly, about 70 of the branch employees lost their homes. Happily, all of them survived safely. And all of them have been kept on full pay throughout this disruption.

The importance of cash emerged again in the days after Katrina hit. People who had money in the bank, or had direct deposit of their pay coming, needed cash. Inside New Orleans, banks were closed and ATMs were not working because the city did not have electricity.

Once people made it to safety outside New Orleans, though, they went to banks and ATMs to get money. The New Orleans branch, the usual provider of currency in that region, was not able to provide cash.

After 9-11, the Reserve Banks had developed mutual backup plans for vital services, including cash. Other branches of the Atlanta Bank, in Birmingham and Nashville, and the Houston branch of the Dallas Bank, jumped in to help. It took very unusual arrangements with armored carriers, but we got the cash to where it was needed, and kept a difficult situation from getting worse. Again, our Banks and branches had the determination to do whatever it took to keep an essential service going.

The Atlanta Bank also became "information central" for many bankers in Louisiana . They could see that the Federal Reserve was operational and getting things done, and they had telephone numbers to contact us, so they did, about all sorts of issues. Often our Atlanta people became intermediaries to help bankers get in touch with state offices, or federal agencies, or other organizations. My counterpart in Atlanta told us about having a call come in on his cell phone, and using his Blackberry to make an outgoing call to help the party on his cell phone. This " clearinghouse" role reminded us of what we experienced on 9-11.

A lesson learned from Katrina has to do with what I will call " corporate self sufficiency" . Essentially, the Federal Reserve had nobody to depend upon except the Federal Reserve to keep its facilities and services going.

Cash services continued successfully because the New Orleans branch could rely on other Fed branches. If we had not had our own backup arrangements, cash services would have stopped.

On a more basic level, even though our New Orleans branch was closed, we had 15 people inside, mostly protection officers who protected the building and the money in the vault, and helped us to contact our employees who had been evacuated from the city.

They had food and water, because all Reserve Banks and branches keep supplies on hand. They had some electricity, because all our buildings have diesel generators for backup.

By the end of the first week after Katrina hit, they were getting low on food, water, and ice. They also were getting low on diesel fuel. In the city there was basically no help available.

The Atlanta office rented two Winnebagos, loaded them with supplies, and drove them to the branch in New Orleans. They delivered the needed supplies, and the Winnebagos, with air conditioning, gave our 15 people places to take breaks and get some comfortable sleep.

Our branch management contacted a former director of the branch who had connections with the fuel industry. He arranged for a resupply of diesel fuel.

With these and similar ad hoc arrangements we kept things going. Today the branch is returning gradually to normal operations and helping its people to find housing.

We realized that to keep things going, especially during that first week or so, we had to rely on our own organization. If we were depending on other entities, public or private, we would have been stymied.

Our Supervision and Regulation function in Boston engaged in similar attention to the insurance industry and its potential impact on banking sectors, as I described earlier in our response to 9-11. As before, this analysis was shared across the Federal Reserve System. I note this here because distribution of knowledge and shared support across Reserve Banks, in operating areas such as cash, but also in analytical support, must become a key part of our corporate self sufficiency.

The experience of Katrina is leading us to look even harder at our contingency plans. If we have to be self-sufficient, then we'd better prepare to be self-sufficient.

In the crises I have recalled with you this evening, the Boston Reserve Bank and the national Federal Reserve System performed well under very difficult conditions. I am proud of my colleagues.

I also take pride in our readiness to learn from these experiences, and to prepare to do even better when challenged again. We will face crises in the future. They will not be the same as those we have faced. They will teach new lessons. What we have learned, though, will serve us well. Thank you.