FINANCIAL STRESS-TEST-DUMMIES AND BANK-SIMULATORS

The Obama administration this week ordered the 19 biggest US banks to run stress tests to check how well they can hold up if the economy deteriorates much further, and to thereby calculate how much more funding they will need before recovery is underway. This is like asking jumbo-plane aircraft pilots to land with faulty instruments in fog safely with minimal damage. When recently a pilot with extremely well-honed skills landed in the Hudson River and all passengers and crew were saved that was sheer brilliance. Would any flight simulator have allowed the same result - no, the built-in tolerances and probabilities would have dictated failure! And, would any other pilot have succeeded - maybe a few, but most not! What made the difference between success and failure was partly the plane, largely the pilot and also largely the conditions including the availability of the liquid landing zone. In banking terms those are the conditions to be determined by Government. The stress-tests are not just to test the banks, but also their strategy, operations and risk management, and then also to provide evidence, clues, indicators, stories of events so Government can also determine what it is that it must do to manage the financial and economic conditions. These stress-tests should also stress-test government bail-out measures. The pilots of the economy are also to be thereby tested. Banks think this is about testing their balance sheets and detailed accounts as seemed to be what stress-testing is in the dealing rooms and in Basel II Pillar II banking regulations. No, this is big picture stuff, and this is always what this stress-testing was meant to be, to test how well bank managements understand the economic context in which they do business. The crash-test dummies are in the boardroom.
To take the analogy further, it is not about how many passengers are on board to be saved or even what the plane is like. This is about how good, how comprehensive, the instrument controls are, and how responsive they are? Markets may be more or less liquid and customers and counter-parties more or less risky, but how much finger-tip control is there and how intelligent and powerful is the auto-pilot and do the pilots decide to dump their fuel over the sea, or save fuel for a possible quick take-off again, or to fly another 500 miles to the next airport? Are the cockpit controls and the accounting system completely accurate and the pilots fully experienced and trained in all weathers. Or is your bank's stress-test capability actually less like a complex flight simulator and more like a car-crash-dummy-test, more like a metal cage with car seats, and safety-belted nodding dummies round the board table? If you don't have the systems, then car-crash -dummies is your bank's default stress-test mode and I don't care how clever you think your guys are with their spread-sheets; this is not a spread-sheeting exam, and not merely a self-assessment exam either! Just ask yourself, can your bank model and explain why and how asset prices, stock values, have disconnected so blatantly from income-stream valuations? Ask yourself if your risk accounting and general ledger actually validate each other and then whether you can compute dynamically and in full double-entry styles in both banking and trading books for the political-economy, global markets, domestic retail and corporate banking for changes in PD, LGD, EAD, ELGD for the quarters ahead and for over the credit and economic cycles. If you think the answer should be yes, think again! The results are expected by or before the end of April. Banks will then have 6 months to raise the extra funding they think they may need from private sources or accept federal aid with conditions attached. The Government, like an air traffic control tower, wants to avoid taking over the flight instructions, it wants to dictate height and heading and make sure there are no air-to-air collisions and that the runway is clear and well-lit up. But government does not want to take over flight-pilot controls, not of large banks. Fed Chairman Ben Bernanke says government may take substantial minority stakes in Citigroup and other troubled banks only if they can't raise sufficient new funding elsewhere. Six months from now the technical recession should be over and modest positive growth returning. But, banks will have a difficult time just to survive the turbulence until then and to steady themselves onto exactly the right path for a soft landing. For important advice on various ways of determining your liquidity risk and solvency tests see the attached lengthy comment below.
Barack Obama on Wednesday described some principles for new financial regulation, signaling that the government would oversee a much wider range of financial activity and take a stronger hand in doing so. Gordon Brown and his team in the UK are doing and saying very much the exact same, including calling for stress tests. And he has begun to use a new repeatable phrase that of 'cleaning out the banks', words that have now taken on a whole new meaning quite different from the Hole In The Wall Gang, the James brothers or John Dillinger. I can't recall any time in the past when respectable people were going to clean out the banks? Of course, if Governments want to do that they can. But President Obama says, "While free markets are the key to our progress, they do not give us free license to take whatever we can get, however we can get it," and "Strong financial markets require clear rules of the road, not to hinder financial institutions but to protect consumers and investors." Let the banks be warned; these stress tests are about whether your bank needs cleaning out in a good way. "Cleaning out" in a bad way is what variously happened to Bear Stearns, Merrill-Lynch, Wachovia, HBOS, Fortis, WaMu, Dexia, Citicorp, RBS and others, though I am impressed with the new managements at both RBS and LBG. These banks were variously shown up to be flown by poorly equipped pilots, and in some cases by dummies (HBOS) or crash-test fighter-pilots (Lehman Brothers). The cleaning out will also extend to restructuring the regulators and bringing in new much more solid expertise. This is happening in Europe, the UK and USA.
Speaking after a meeting with key lawmakers and senior advisers, Obama said his government needs to overhaul the frayed patchwork of regulatory agencies that oversee the financial industry. He expressed support for increased consumer protections. He indicated he favoured empowering the Federal Reserve as a regulator of systemic risk (gosh, where was that responsibility hitherto?), with authority over any financial firm or instrument that could destabilize the economy. That's the key question stress-tests have to answer, who's next for a military-style haircut and strict exercise regime? The European Commission and ECB want similar answers, the IMF too, and, in the UK, the Bank of England. This is not an academic research project taking 4 years. The answers to these complex questions are required in 4 weeks. But improving on stress-test models and cockpit risk management control systems will require banks to focus on these in the boardroom regularly for at least 4 years (cost across all banks in Europe and USA = $8bn estimate). The US Administration will present a more detailed plan for regulatory reform in early April, when leaders of the G20 wealthiest nations, each with their own stress-tested plans, meet in London. By then, round the same table, all will know the global depth of the banking industry's political-economy crisis, and no dummies any longer tolerated there. That's my expertise, all of this, and I and my colleagues at asymtopix.eu and union-legend.com and what we know of the companies we talk to, none any longer can afford to under-estimate the challenges involved here.
WHAT IS THE CHALLENGE?
If you are in or close to the executive and full boards of a major bank, let me say categorically: It is not bottom up deal by deal stress-test technology or risk models and risk policies from the dealing rooms. It is not top down, comprehensive, national and global economic modeling. It is not stressing of risk factor ratios or moving up and down risk grade tables. It is not reliant on what we already know, not about adding up all risk buckets and business lines, to collate and correlate. It is not about dumping spreadsheets and getting in the super-computers and a new general ledger system, new software, years of new external and internal data, new definitions and holistically integrating taxonomies for all credit risks and all market risks. It is not about liquidity risk and risk concentration versus asset class and economic sector diversification.
It is all of this, a triangulation, top down, bottom up and sideways along, outside and not just inside the bank, and a damn sight more than all of this too covering what's happening to the whole of banking, and in the markets and among government policy options and the bank's business strategy options, and can't be done by persons who only have silo understanding and not a full grasp of the totality. It cannot be done except by those who thoroughly and professionally understand the totality of a bank, in finance, accounting, risk and economics, and the totality of Basel II Pillar II and the coming of Basel III, and why even Pillar I implementations are deeply flawed, and why Pillar II is a dangerous distraction and probable big mistake! If you understand where this shopping list of a rant comes from, then you are halfway there and should be advising the top of your bank or already in the boardroom! Retail giants such as Bank of America, HSBC, RBS, Citigroup and specialty, less retail, and non-retail banks such as JPM, State Street, Morgan Stanley, Goldman Sachs are all now required to measure their likely losses in the event that the credit and economic gets worse before it gets better. Officials say the tests will assume that the unemployment reaches 10.3% by the end of 2010 (currently 7.6%), average home price fall from bubble-peak hits 47% (27% so far). Mortgage defaults are over 9%, sub-investment grade corporate bonds heading there, consumer debt maybe 6%, prime quality bank debts maybe still pretty low at variously around 3%. But what if default rates for many broad classes trend to some similar higher rates as the domino or systemic impacts pull down good quality with the bad?
What if it doesn't matter if your are a creditor or debtor nation, all lose similar growth, and what if most or all businesses, most or all creditors and debtors, good as well as bad banks, all suffer significantly? What if it is not about whether your balance sheet is 5%, 10% or 25% better quality than the next bank, because that will matter a lot less in 6 months time than it matters now? The test will also assume that economic activity as measured by gross domestic product shrinks by 3.3% in 2009 before recovering slightly in 2010. Great, but does your bank know how to translate that into anything meaningful. Is 3.3% inflation-adjusted GDP, 2% or 6% before subtracting for inflation? Can your bank survive better if other banks survive better, or better if your bank does better than others who are then doing significantly worse? Is it better to take government investment on board and central bank funding and insurance guarantee programmes or better to keep going without that - better for whom - your bonuses, your shareholders, your customers, and or the economy? Will your bank do better if it can somehow deleverage and behave pro-cyclically or better if it can grow lending and act anti-cyclically, better if it delivers higher profits from traditional banking or lower profitability from traditional banking?
As a bank, what is it you are legitimately doing apart from providing a transmission mechanism as a public service? Is your idea of success and benefits the same as the government's that has to care about the whole economy as you should be too? Is a much higher financial margin profitability of the whole financial sector better or worse for the economy right now? Does your economic capital model tell you anything to adjudge your answers to these questions? The answers to these questions in the longer term are no problem, but for the next 1-2-4-6 years that is another matter altogether? The Government does not care about individual players; it wants to see the whole picture of what banks need to be that fits intelligently with what the economy needs them to be right now? Do banks, or your bank, know how to think about that? Yes is easy, elaborating on that answer is not. The Government is your bridging loan, but only if your bank can determine how to be a bridging loan in turn for the economy, and maybe in several economies? There is also so much peer-review going on that looking for a presentational answer that cannot be validated in depth won't wash. Confidence is more than skin-deep, and loss of confidence is fatal. See: www.union-legend.com/index.php?page=references There is a certain rhetorical quality to these questions. Where they are not rhetorical is where we really do not have the political-economy macro-answers, and worse where not even the biggest banks have the data or the systems and completed models to find their own reliable answers with any substantial % confidence whatsoever? It is a wholly different game to forecast via scenario stress-tests for some theoretical set of future-shocks at some indeterminate time over the horizon. What can be done and how is to be done when already in the middle of the distress, in the middle of the fat-tail unexpected risk distribution, that's quite different and not to be understood in the Basel II and Solvency II regulatory advice notes, or even from the very best academic papers - they're being thoroughly revised and rewritten - who in your bank reads them?
Regulators are in the business of providing clues and indicators, not glide-paths for exactly how to do any of this. It is nothing like regulatory risk model equations. For example, regulators have calculated that there is roughly a 10% chance that economic conditions would reach say a 3.3% real GDP fall in 2009 based on a consensus of economic forecasts, which has never been much of a reliable benchmark. Therefore, prominent economists are saying the probability is much higher than 10% and banks and government should test for a more catastrophic scenario. I agree, do you? But, some analysts say the worst projections are not much more dire than what many private forecasters already expect e.g. 5% real GDP fall in 2009. And, anyway, the worst scenarios are not that, not how bad will 2009 be, it is whether the economy might remain flat for a few years thereafter i.e. not a V- or U- shaped recession but an L-shaped one. That will test banks capital reserves to destruction beyond the 200% wipe-out currently being faced. It would not be hard to show a bad year wiping out reserves by another 50% and 3 years flat growth by over 100% again! According to the new Treasury Department guidelines, the banks would have to assume that the economy contracts by 3.3% in 2009 and flat in 2010, and housing falls 22% generally on top of the 15%+ so far, then maybe recovery is regionally patchy after that, especially across the US States in 2010, while unemployment goes up to 9% then over 10% in 2010. Does your bank know what PDs and LGDs could become in these years. Can you even dynamically model for that either at portfolio, aggregated bucket, level or for every account in the books? Have you got integrated accounting and computing power to process all that in less than one month? If not, how else will you get a realistic approximation - try talking to Lloyds Banking Group, or RBS, or central banks or others?
"I don't think they (the stress conditions) are harsh enough," said David Hendler, analyst at CreditSights, who says the dire projection is too optimistic about how much growth will be generated from President Barack Obama's fiscal deficit stimulus programme. "That would be a pleasant outcome, but you have to plan for the worst." I agree. The average outlook of private-sector forecasters envisages the economy shrinking by 2% this year and unemployment peaking just below 9% in 2010. But, even the biggest banks are regional not just national and regional variations in a country the size of the USA are significant. Big banks are also internationally exposed, and globally in respect of wholesale financial markets, which hardly anyone has ever been able to soundly model in respect of national or international or global economic models, and of which there are not many around. The trading books have small RWA calculations but we know these are now unreliable. Can your bank stress-test collateral, security, hedges and cross-border financial flows, corporates and industry sectors, and interbank credit in the trading book as readily as credit risk and collateral in the banking book? The stress-tests must adjust for changes in market liquidities. Do you know how to measure the liquidity of markets, never mind liquidity risk of your own bank. By 'model' I and regulators and economists do not mean 'make assumptions' but integrate all major factors into accounting models over time whereby factors constrain each other? Is this a high-level project with top quality people do the work or a low-level delegated work relying on simply processing every combination of likely factor values?
Everyone's recent forecasts including by the Federal Reserve and most private forecasters have undershot the severity of the downturn. Big banks — those with more than $100 billion in assets — have to carry out supervised analyses by April of how much their capital would be depleted under the Treasury Department assumptions. If U.S. banking regulators conclude that your bank may not have enough capital under those circumstances, the bank has six months to sort the problem - is 6 months realistic and practical? It is only if the tests show that your minimum regulatory capital survives at least 6 months?
The Treasury said that it would provide new capital in exchange for shares of preferred stock that could be converted to shares of common stock at a price slightly below the level at which the shares traded on Feb. 9. The UK seems to be offering similar but with conversion rates at markedly higher prices than current share prices. For many of the US big banks, 9 Feb price is slightly higher than today, but still at fire-sale levels. In a telephone conference call with reporters, officials from the Fed and OCC said there would be no simple measure for "passing" or "failing" a test. The FDIC has firmer standard tests and these might kick in on insurance guarantee rates sooner. It's not exactly clear if the FDIC and the fed are singing from exactly the same hymn sheet? "It sure sounds to me like they are designing this to make it sound like the banking system is in great shape," Paul Miller, an analyst at Friedman, Billings Ramsey, a brokerage firm that specializes in bank stocks, said to the Washington Post. This is his interpretation of officials says their goal is to increase confidence of investors and depositors in the big banks, providing tangible evidence that the institutions would have enough money, whether they had to raise it from private investors or get it from taxpayers.
What this is really about is trying to get a consensus view of the expected shape of the credit cycle and economic cycle from the perspective of banks' capital resources. Fed officials said there is no cap on how much money a single institution could get. But that is of course how banks got into funding gaps that proved hard to maintain, so maybe some ratio caps to current funding gap or deposits might seem sensible. Officials declined to estimate how much money the government would end up injecting before the crisis was over. This is a sensible answer given the variety of means available, but over $2 trillions would be the ballpark just for big banks. Banks and other "qualified financial institutions," which now includes investment banks and insurance companies that are part of bank holding companies, can start applying for more money immediately.
Christopher Whalen, MD at Institutional Risk Analytics, said Citigroup and other major banks would almost certainly become insolvent once they absorb the full brunt of losses from the economic downturn. That is absurd, since it fails to see that banks don't just stand still in their liquidity management. He said, "The stress test is about politics. The OCC and the Fed already know the answer. The answer is that we're going to have to come to a decision: are we going to put in more equity or are we going to resolve the banks through bankruptcy?" This is now classic doomster headlining for the media. This is not the question or the answer. Bankruptcy is not an option. replacing ownership and getting rid of managements and bad bank work-out funds and much else are options, active choices in play right now.
Europe is focusing hard on the G20 concerns to take urgent steps to overhaul regulation of big cross-border banks. President Obama met with Congressional legislators to discuss an overhaul of how the financial regulatory system manages risk in the future. Emerging from the meeting, Obama said the first principle of a new system should be that financial institutions "that pose serious risks to the markets should be subject to serious oversight by the government." The EU European Commission has exactly the same intent to be applied to the top 45 banks. Obama also said that the regulatory system should be strengthened to withstand major stresses and that the government should take steps to rebuild trust in markets by promoting transparency. This also means bringing off-exchange markets on-exchange. With the exception of government spending, every major component of the economy shrank in 2008.
Output fell 6.2% at an annualised rate in the 4th quarter of 2008 (also by 6% in 4th quarter in the UK), revised downward from a previous estimate of a 3.8% decline. The drop was steeper than the consensus estimate of 5.4% - far steeper than 0.5% of the previous quarter.This was enough to have immediate ripple effects globally. The economy took the biggest hits in exports, retail sales, equipment, software and residential fixed investment plus contraction in inventories (usual sign too of drop in loan financing) of unsold goods despite lower consumer sales. But this data will continue to be severely revised for another year. The trade gap that had been narrowing widened as exports fell at an annual rate of 23.6% shaving a full 1% of GDP and is a sign of downturn spreading through the rest of the world. This all bodes ill for business investment and sure enough it fell at an annualized rate of 28.8%, also a sign of bank credit being choked off or debt restructuring. deposit savings crept up but private savings will only jump significantly once the Government bond auctions for 2009 are well underway and hopefully banks will grow their loans by an aggregate of 10% minimum, what Government is hoping to see, and that means growing capital reserves proportionately too. For much more read comment below.