 Good morning, everyone, and thank you for coming back here. Sorry we're a little bit late, but we will probably make it up, or we hope so. So we come to the second half of our Banking Supervision Forum today, and we will get right into it with a keynote speech by Michael Barr, Vice Chair for Supervision of the Federal Reserve. And Mr. Barr, will you be ready to take some questions at the end of his speech so you can already prepare in your head what you would like to ask him. Mr. Barr, please. Good morning, everybody. I was joking a little bit earlier that my remarks as prepared for delivery are released by the Federal Reserve at the time of this event, but it's noted as 3 a.m. So hopefully you're all a little bit more awake than that. My remarks today are, of course, only my remarks and not those of the Federal Reserve Board or of the Federal Open Markets Committee. Thank you very much for the opportunity to speak today. I'm delighted to be able to celebrate the retirement of Andrea Anria, my dear friend and colleague, who has done so much to strengthen the supervision and regulation of European banks throughout his career. In my remarks, I would like to provide perspective on some of the lessons learned from the banking stress experienced in the United States last spring for both banks and their supervisors. In particular, I will focus on how banks manage liquidity risk, the role of the central bank's discount window lending in this process, and the importance of robust liquidity planning for good times and bad. Last March, several large U.S. banks faced acute liquidity pressures when uninsured depositors looked at the bank's balance sheets and judged that they would be insolvent if the banks needed to liquidate their securities portfolios to meet potential outflows. The bank's poor interest rate and liquidity risk management triggered a crisis of confidence in their uninsured depositors, resulting in liquidity crises at these banks. In short, they faced old-fashioned bank runs, the speed of which was anything but old-fashioned. Despite their compliance with our capital rules, these banks lacked enough capital to reassure uninsured depositors that they had sufficient resources to weather this liquidity storm. In addition to our domestic strains, Credit Suisse came under renewed pressure in March 2023 after a long period of liquidity pressures that had been acute since the fall of 2022. Of course, Credit Suisse had been a troubled bank for some time with doubts about its future viability after the Archaegos and Greensales scandals had tarnished its reputation. These concerns became reality when the firm was forced to announce that its internal controls over financial reporting were ineffective and had been for several years. Credit Suisse was acquired by UBS in a deal that involved triggering of Credit Suisse's contingent convertible capital instruments, a severe dilution of shareholders, removal of senior bank management, as well as emergency liquidity support and extraordinary loss sharing from the Swiss government. While there is more that regulators and supervisors can do to help ensure bank's interest rate risk management and capital bases are sufficiently calibrated to the risks of their business models, today I will focus most of my comments on liquidity risk management and operational readiness for firms in the United States to utilize the Federal Reserve's discount window. This is not a new topic and I have spoken about lessons from March and the importance of bank's preparedness to tap Federal Reserve facilities previously. Today I will revisit those themes and provide some additional observations about the March stress events, including the importance of discount window preparedness relative to some specific liquidity risk factors such as uninsured deposits. A striking feature of recent U.S. experience with bank stress was that Silicon Valley Bank and Signature Bank struggled to cope with unprecedented deposit outflows arising from a loss of confidence of their uninsured depositors. While ultimately the amount of their outflows made it not possible for these banks to continue operating, these banks started from a state in which they were not sufficiently equipped to manage liquidity risk, including by being adequately prepared to tap the Federal Reserve's discount window. Banks that experienced spillovers during this period also struggled with insufficiently robust liquidity risk management. It is crucial that banks have a diversified range of liquidity options that they are able to access in a variety of conditions. And in the case of banks that are eligible to borrow from the Federal Reserve, discount window barring should be an important part of this mix. There are a few major reasons for this conclusion. First, the discount window provides funding at a predictable interest rate, namely the primary credit rate, which does not fluctuate daily, unlike interest rates on private sector liquidity facilities. More specifically, the discount window is always available as a standing facility, with a primary credit rate currently set at the top of the Federal Fund's rate target range. This rate, along with that of the standing repo facility, also set at this level, should serve as a backstop rate to short-term market interest rates and provide eligible firms with certainty about the highest rate that they will face to gain liquidity against eligible assets at any given time. The discount window provides ready access to funding that can help banks manage their liquidity risks. The ability to access funding at a predictable rate throughout through the discount window should figure importantly into banks' liquidity risk management plans under a range of scenarios. The discount window is also an important tool for monetary policy. The primary credit rate set at the top of the target range is a component for how we can achieve rate control under a range of circumstances. To achieve our monetary policy aims, the discount window needs to be readily usable. If banks do not feel free to use the discount window when private sources of funding are more expensive, the discount window will not be an effective part of our monetary policy implementation toolkit. Banks have previously said that they are afraid of receiving negative feedback from their supervisors in the event that their grounds for tapping the discount window is that it is the most convenient or cheapest form of funding immediately available to them. In light of this, we at the Federal Reserve have been underlining the point to banks, supervisors, analysts, rating agencies, other market observers, and the public through numerous channels that using the discount window is not an action to be viewed negatively. Banks need to be ready and willing to use the discount window in good times and bad. The discount window is also an important tool of financial stability, because the Federal Reserve can provide liquidity in an array of circumstances, including those when other sources are impaired. Liquidity provision through the discount window is not reliant on the smooth functioning of private sector funding markets. This is in contrast to most other options available to banks, including access in the United States to the federal home loan bank system. Moreover, banks that have pre-plagged collateral at the discount window and tested its operations are not dependent on other market infrastructure and payment systems to borrow against this collateral. These features mean that discount window preparedness provides additional diversification benefits that can meaningfully enhance effective liquidity risk management. The experiences of last March weren't reviewing again to gain better insight into how well banks are positioned to borrow through the discount window. In doing so, I will discuss some observations from March in greater depth and note some potential lessons to be drawn from that experience. I will also briefly note some observations on our work cross-border in collaborating with our international counterparts. The speed of the bank runs in the spring of 2023 were unprecedented. The failure of Washington Mutual in 2008, the largest bank failure seen in the United States before that year, was the culmination of stresses that occurred over several weeks. The deposit losses experienced by SVB were much greater in both relative and absolute terms and they occurred in less than 24 hours. Digital banking and social media were factors in the rapid escalation of SVB's problems, though they were not the underlying source of vulnerability. SVB's concentrated and highly networked depositor base of venture capital firms, portfolio companies, tech and crypto companies, and high net worth individuals communicated quickly with one another and in effect coordinated the massive and rapid run. And more fundamentally, as I have said on other occasions, SVB failed because of a textbook case of mismanagement of interest rate and liquidity risk. This mismanagement made depositors lose confidence in the bank's solvency and so they ran. The March stress also showed that contagion is possible among large regional banking organizations whose size and scope are well below those of banks that have been designated as globally systemically important. Liquidity stress at SVB quickly led other banks that were perceived to have weaknesses to experience it outflows as well. Those banks that came under greater pressure tended to have large unrealized losses in their securities portfolios to rely heavily on uninsured deposits and to have a deposit and client base focused on the tech and crypto sectors, venture capital or high net worth clients. But the contagion extended beyond such banks and threatened to cause disruption more broadly to regional and community banks with traditional business models and to the banking system as a whole. The Federal Reserve, the FDIC and the Treasury Department stepped in quickly to stop that contagion and the strategy worked. Deposit flows returned to normal and stress in the banking system slowly abated. This experience has changed everyone's perception of the possible speed of bank runs. What occurred in two or three weeks or in some cases many months in previous episodes may in the current circumstances now occur in hours. These issues are top of mind as we review and consider future adjustments to the way in which we should supervise and regulate liquidity risk. We also saw during this period that firms were not as well positioned to monetize, that is to borrow against or be ready to sell their assets as they should have been. Many banks underestimated the size and speed with which liquidity needs could appear. Banks of all sizes are expected to maintain contingency funding plans to meet potential stress liquidity outflows. In response to the global financial crisis, regulators also now require large banks to maintain a multi-week buffer of high quality liquid assets that can be easily converted to reserves during times of stress. These HQLA buffers are essential to ensuring funding resilience at large firms. The March Stress episode, however, highlighted the fact that in practice there can be operational impediments to a bank's ability to monetize its liquidity buffers in large volumes and in a rapid timeframe in acute stress. In a world in which a bank run took place over a matter of weeks, it was reasonable to assume that a sufficient volume of the most liquid securities could be monetized to meet the demand for reserves associated with deposit withdrawals. But times have changed and I see several flaws in our previous assumptions. First, it may be difficult for a firm to conduct significant asset sales in a short timeframe without becoming the subject of adverse attention or if the firm is large without affecting market prices with fire sale effects and potentially leading to broader contagion. Second, the March Stress underlined the possibility that private repo markets may not be a viable financing channel for banks that need to rapidly ramp up access, especially for banks that may not regularly transact in these markets. Even if such repo markets can be a viable source of liquidity for banks that regularly tap such markets and have more gradual funding stress. Sharp shifts in calls on private repo market capacity, particularly by firms experiencing stress, may not be easily met. In addition, it proved especially difficult to monetize assets in those cases in which firms held substantial amounts of longer dated securities with significant amounts of unrealized losses as we saw in the experience of SVB. The experience showed that when these securities are sold and losses are realized, it understandably may send a negative signal to the market about a bank's viability. A compounding factor in these situations is that if the securities that a bank needs to sell are inheld to maturity portfolios, the sale under acute stress can have a severe effect on the bank's balance sheet. While this discuss the discount window provides ready access to funding to help banks manage liquidity risks in normal conditions, it is also incredibly important that it be available to banks dealing with idiosyncratic or market-wide stress events. When other forms of funding or related market infrastructure are not immediately available, readiness to borrow via the discount window, including with preposition collateral is even more crucial. In contrast to private sector liquidity sources, the Fed can provide immediate liquidity against a wide range of collateral to one or a number of depository institutions simultaneously. This function provides additional time for more orderly monetization. An update on contingent liquidity readiness that the bank regulatory agencies published in July also makes it clear that all of the agencies understand the importance of a bank's readiness to make use of the discount window. The banks that failed because of the stress event that began in March had access to and utilize the Federal Reserve's discount window. And their failures were the result, not a lack of access to the discount window, but a basic mismanagement of interest rate and liquidity risk that left them effectively insolvent in the eyes of uninsured depositors and unable to stem bank runs. That said, they also faced internal operational challenges in quickly identifying and moving collateral that would have provided them additional borrowing capacity at the discount window. While this additional borrowing capacity would not ultimately have saved these banks given the speed and severity of their deposit outflows, the lessons from managing their stress events can help others facing less acute events. Greater operational readiness can provide for greater opportunity and optionality when a bank hits about of turbulence. Ready access to sufficient liquidity provides breathing room for a bank to determine and execute its path forward. There are two key aspects of discount window readiness. Preparedness to access the discount window and prepositioning adequate amounts of collateral. In terms of preparedness, the majority of banks that are eligible to borrow from the discount window now have the legal agreements in place. This is the first step. However, many of these banks had not recently tested their discount window access prior to the stress event. Engaging in testing through actual transactions at regular intervals is a key component of operational readiness. In the case of some banks, the amount of collateral prepositioned was also a tiny fraction of potentially flight prone liabilities going into the stress event. This lack of prepledging is a concern for several reasons, including that certain collateral types can require more time to pledge. Less liquid collateral can take longer to be assessed and valued at the discount window, meaning that banks should not expect that they can gain immediate liquidity against these assets. But these are the very assets that would be best pledged to the discount window so that more liquid assets are held for other uses. The existing standards that require banks to hold HQLA buffers for self-insurance are an essential element of the regulatory framework. Requirements applying to large banks like the liquidity coverage ratio and the net stable funding ratio have meaningfully increased the resiliency of the banking system to liquidity stress and position large banks to deal better with a 30-day stress period. This type of self-insurance is critical to bank resilience, and to the robustness of the financial system. However, these requirements may not on their own be sufficient to stem a rapid run. The speed of bank runs and the impediments to rapidly raising liquidity in private markets that may be needed in hours rather than days suggest it may be necessary to reexamine our requirements, including with respect to self-insurance standards and to discount window preparedness. The lessons from March also indicate that some forms of deposits, such as those from venture capital firms, high net worth individuals, crypto firms and others, may be more prone to faster runs than previously assumed. As I have emphasized today and in two previous speeches, the discount window is a tool that banks can and should use to help them respond to a wider rate of conditions and provide additional time for orderly monetization of liquidity buffers in private markets, but only if banks are prepared to use it. Given these dynamics, we are currently studying what the lessons from March and the variability in discount window preparedness across eligible banks mean for the safety and soundness of individual banks and for the stability of the financial system more broadly. Since March, some banks have been assessing their operational readiness to tap the discount window, particularly relative to their run-about liabilities such as uninsured deposits, and have been taking measures to test and pre-pledge assets where possible. Banks have also been reassessing their assumptions on the liquidity value of held to maturity securities given the experiences of March. Let me wrap up with some brief observations with you on international spillovers to liquidity stress events. Episodes of financial stress have in some instances gone hand-in-hand with stress in U.S. dollar funding markets. Foreign banks, which have more limited access to dollar deposits than U.S. banks, and rely more heavily on dollar wholesale funding for their operations, are particularly vulnerable to dollar liquidity strains. Of course, these banks have an important role in providing credit in the U.S. economy and strains that they experience can affect U.S. businesses and households. In addition, dollar funding markets are global, and so strains in one segment can have broader repercussions for market functioning. Foreign banks that have branches in the United States have access to the discount window. Outside the United States, some of these firms also have access to dollar liquidity from their own central banks. The Federal Reserve maintains swap lines with foreign central banks and also maintains the foreign and international monetary authorities, or FEMA repo facility, to provide dollar liquidity to foreign official counterparts. These facilities have proven effective in damping pressures in U.S. dollar funding markets when adverse pressures have emerged, pressures that could exacerbate strains in broader U.S. and global financial markets. Indeed, during the global financial crisis and the COVID-19 pandemic, the peak outstanding amounts for the swap lines were a little above and a little below 500 billion, respectively, making them among the most used of our liquidity facilities. As with all backstop facilities, however, it is not just their use that helps the smooth functioning of markets, but banks and central banks knowing that liquidity will be available when needed, that helps to prevent liquidity hoarding and precautionary sales that can contribute to stresses. In part for this reason, the Federal Open Markets Committee decided to make the swap lines and FEMA repo standing facilities after initially setting the operations only up as temporary. These observations underscore the importance of our ongoing communication and collaboration with our counterparts, both other central banks and regulatory authorities, in order to ensure that we have a comprehensive understanding of evolving financial market dynamics and cross-border linkages that can affect our respective financial markets and financial institutions and that we can work to address gaps before pressures emerge. In my experience, these cross-border collaborations have been essential in enhancing the resilience of individual banks and the robustness of the financial system. I am grateful to my colleagues for their collaboration and again wish to express my admiration for Andrea Anria and on his retirement. Thank you very much. We have a little bit of time for questions. I'm happy to take questions from all of you. Yes. Thank you. Simon Haynesworth, Moody's. Central banks have struggled for decades, and possibly longer, with the problem of stigma in banks accessing central bank facilities at times of stress. I hear what you're saying about exhorting banks to make greater use of facilities and the recommendations you're making to others to understand that that usage is not necessarily a bad thing. But is there a more fundamental redesign required of how the liquidity facilities work at central banks like the Federal Reserve? We know the Bank of England made attempts to do that in terms of delineating its facilities more clearly to reflect which are those which are more LOILR, which are more standard use, is just saying the bank should use it is not going to be enough to deliver your goal. It's a great question, and I think the answer to that is we're not sure. It is an area that we're exploring very actively. I do think it helps to have senior officials say very publicly and repeatedly that the discount window is available for use and should be used. We've also done that on an interagency basis. All three agencies have gotten together and explained that we expect discount window usage. We will not criticize discount window usage. We are also conveying that to supervisors in the field. And as I mentioned briefly, to analysts, to market participants, to rating agencies, to make it clear that we do not view discount window usage negatively in any way. Yes, sir. Thank you. Alistair Arn, Bank of America. Very clear speech. Thank you. So just banks have often used a surplus over their liquidity coverage ratio of 100% as a shorthand for showing you how liquid they are. What you've described is quite a different picture that banks may have better or less good access to the discount window, better quality collateral, better prepositioning. So is it wrong for the markets who have been thinking that 130 LCRs are great and 105 is bad? Because from your perspective, it might be quite different. The banks might have a very different other that's important for you. Thank you. It's a terrific question. So I do think that it's appropriate for the market to look to the LCR as an important signal of the liquidity of the bank. And so banks that have a significant buffer above their acquired LCR are more liquid. And certainly the experience with Credit Suisse in the spring suggests that that institution had more time because it had buffers in excess of the acquired LCR, the minimum LCR I should say. And that's true for U.S. firms too. So we do care about that level of self-insurance. It's a really important signal of liquidity at the bank. The point of my remarks is that it may not be enough if you have an acute stress period. And there an additional signal of liquidity would be the fact that a bank has pre-positioned a collateral at the discount window has tested its operational capacity to use that discount window and is part of their contingency funding plan. And so we do want banks to both have strong self-insurance and to have ready demonstrable access to emergency sources of liquidity if they need it. And it's also, as I said in my speech, good. We believe it's healthy that if we see funding pressures in the market that are causing market prices to rise above the top end of the federal funds rate for banks to use the discount window as a way of keeping monetary policy in an appropriate range. So both in good times and in emergency times, we think usage of the discount window ought to be viewed in a very positive light. Thank you very much. Morea Tadeo from Bloomberg. I'm in the next panel, but I was very interested by the amount of times that you mentioned. Credit Suisse, you also said, you also talked about the velocity of some of the bank runs. And you also just said, maybe this institution, and I just talked about it as an example, had more time. But I think everyone can agree that it got caught in this very negative sidegeist with perhaps amplified by social media in a way that it didn't apply before. So how do you end this cycle? Is there a way that you can end it? And I guess that if you go back, the bank would have said, perhaps things shouldn't turn out the way they did, but we just got caught in this sidegeist and do they have the tools to prevent a situation like that? Thank you for the question. Look, I don't want to ignore the role of social media. Social media played a role in the March stress events. They played a role with respect to Credit Suisse and the US institutions. We do social media monitoring to see how institutions are being perceived online as part of our risk assessment. So social media does matter, but the role of social media can also be overplayed. Credit Suisse had underlying fundamental vulnerabilities and they were hit with a shock, but if they didn't have those fundamental underlying vulnerabilities, they would not have disappeared. So I just think the same with SVB. SVB failed because not of social media, but because of absolutely fundamental failures of risk management with respect to interest rate risk and liquidity risk. Without those vulnerabilities, there can be social media chatter. It's not enough to kill an institution. So we have to pay attention to social media, but also really focus on the basics of banking. Thank you very much for your comments and especially just following up on that very good point, paying attention to the risk management processes and interest rate risk and liquidity risk in the banking sector. One of the themes of this conference has really been about the regulatory perimeter and the non-bank financial institution market. And the comments about how much growth there has been globally, US and Europe, in the last decade, the numbers are staggering. How are you thinking about understanding the correlation risk and the lessons from LTCM in the current context when your regulatory perimeter is over the banking sector and you don't have visibility into that interest rate risk and liquidity risk and leverage outside of that perimeter and where it might be connected back? You know, it's a perennial question in financial regulation. We always have this question of what's inside the perimeter, what's out, how should you regulate what's inside given what's out. I've taught for many, many years, banking regulation class, and this is a theme that we explore over and over again in that class. And there's no definitive answer to the question. What we try and do is, first of all, gain better insights into what's happening in the NBFI sector. Collectively, we're doing that through the Financial Stability Board. We do that in the United States at the Fed and through the Financial Stability Oversight Council. So we have better insight into that sector than we did prior to the global financial crisis. We don't have perfect visibility, but we have much better visibility than we did before, and that's helpful. We also monitor the relationship between the banking system and the NBFI sector from the bank side. So we pay attention to how banks are doing, for example, at counterparty credit risk management, which is a key feature of protecting the banking system from those kinds of spillovers. We are paying attention to the buildup of risks in some areas. We are paying attention in the United States, for example, to the growth of the basis trade in the Treasury market, which is a highly leveraged position. It has many advantages in normal times in proving price efficiency, price discovery, in providing liquidity to the market. But if there are shocks to that, it can cause a spillover. So we pay attention to issues like that. The rise of private credit was mentioned yesterday. We do pay attention to the private credit market in the United States, which is both on its own, but also funded by banks. So there's an important relationship there. We do in the United States also have a large non-bank sector providing originating mortgages. So we look at risks and the fragility in that sector as well. I could sort of go on sector by sector, but part of the answer is making sure that we know as much as we can, and that we have a robust, strong system with very strong capital in the banking system so that if problems happen in the non-banking sector, the banking system is resilient to those stresses. My clock is saying minus 2.39 minutes. So I think we're done with the Q&A, but it's been delightful speaking with you. Thank you.