Bank stocks have been rising well internationally. Wednesday (yesterday) was especially good for US financial stocks as investors expressed relief the capital shortfalls identified by the government’s “stress tests” at large banks such as Citigroup and Bank of America were not as big as feared.
The bank rally occurred as news of the capital needs of the 19 banks involved in the tests leaked out during the day, ahead of the official release of the results on Thursday.
Citi, BofA and Morgan Stanley were among the big names that will have to raise equity following the completion of the tests, while JPMorgan Chase, Goldman Sachs and American Express are among those that will not need additional capital, people familiar with the situation said.
Citi and BofA emerged as the banks with the biggest capital shortfalls, with Citi’s equity needs projected to be more than $50bn and BofA requiring about $34bn in fresh equity.
However, BofA’s capital deficit is more pressing because Citi has already agreed to bolster its balance sheet by converting preferred shares owned by the government and other investors and selling non-core businesses.
People close to the situation expect Citi to have to raise no more than $6bn through the expansion of its planned conversion of preferred shares – less than the $10bn-plus the market had feared. The move could decrease the government’s stake in Citi, which was expected to be 36 per cent to about 33 per cent.
New equity ratio rule
US banks will be required to hold enough equity to ensure that they would still have a common equity ratio of at least 4% of risk-weighted assets at the end of 2010 even if the adverse scenario set out in bank stress tests were to materialise, the authorities said on Wednesday 6th May. I suggest that they should consier applying an equivalent of the Dutch National Bank’s Liquidity directive.
Banks are required to report on a consolidated level on their liquidity position to the DNB monthly, on the basis of the liquidity supervision directive. The liquidity directive seeks to ensure that banks are in a position to cope with an acute short term liquidity shortage under the assumption that banks would remain solvent. In principle, the DNB liquidity directive covers all direct domestic and foreign establishments (subsidiaries/branches), including majority participations. The regulatory report also takes into consideration the liquidity effects of derivatives and the potential drawings under committed facilities.
The directive places emphasis on the short term in testing the liquidity position over a period of up to one month with a separate test of the liquidity position in the first week. For observation purposes, several additional maturity bands are included in the liquidity report (one to three months, three to six months, six months to one year and beyond one year).
Available liquidity must always exceed required liquidity. Available liquidity and required liquidity are calculated by applying weighting factors to the relevant on- and off-balance sheet items, i.e. irrevocable commitments. The liquidity test includes all currencies. Compliance reports concerning liquidity requirements of foreign subsidiaries are submitted to the appropriate foreign regulatory authorities as required. At a consolidated level, and in every country in which a bank operates, the banking group must adhere to the liquidity standards imposed by the applicable regulatory authorities.
The use of a common equity ratio – even if as a benchmark rather than a formal future standard – is a departure from normal bank regulation. It is intended to ensure that banks have good quality capital, providing permanent capacity to absorb losses and flexibility over cash distributions.
The US authorities also said that the 19 biggest US banks will also be required to hold enough overall tier one capital to ensure that they would still have tier one capital equal to at least 6% ratio to RWA. This allows their reserve capital to operate as a cushion that can depress by half.
Banks will have 30 days to present a plan to meet the demands identified by regulators. They will also be required to outline steps they will take to “address weakness, where appropriate” in their own processes for capital planning. They are tasked to outline how they will, over time, repay existing government capital injections (e.g. in payment or in preferred shares) and reduce reliance on debt issued under a government-guaranteed programme. Up to 10 of the 19 banks are likely to need fresh equity, according to various unnamed sources. Morgan Stanley is said to need $1.5bn in new equity, as a result of its buying a majority stake in Citi’s Smith Barney brokerage.
Tim Geithner, the US Treasury Secretary, said last night on PBS: “I think the results will be, on balance, reassuring.” Investors concur, with shares in BofA, Citi and Wells Fargo – expected to need $10bn-plus in new equity – rising by over 15%.
The tests aim to ensure that even in an adverse economic scenarios, such as obviously the current crisis, banks can retain tier one capital of at least 6% of risk-weighted assets RWA – loans and other exposures less collateral, with both the assets and liabilities risk-weighted and risk-graded) and tangible common equity of at least 4% at the end of 2010 (when economic recovery should be 1 year old). Each bank told to raise additional equity has until June 8 to present a plan to regulators explaining how it intends to do so. The concern here is to have clarity in time for planning the administration’s budget for budget year starting 1 October. This is required even though further government financing will be off-budget via Federal Reserve balance sheet of T-bills, and only on-budget if requiring longer term Treasury Bond issues.
These banks will also be required to remedy any “weaknesses” (governance, systems, data quality, reporting process, risk management controls, and risk mitigation measures) in their internal capital planning (liquidity risk management) and to outline how they eventually intend to buyt hemselves out of government-sponsored aid.
The revelation that BofA needs about $34bn in extra capital will increase pressure on Ken Lewis, its embattled chief executive. The banks and the regulators declined to comment to news sources such as FT or Reuters or were unavailable.
The stress tests overseen by the FDIC involved more than 150 senior bank supervisors, analysts and economists being interrogated (subject to Geneva Convention or Basel Accord Rules) in the top 19 US banks about their likely losses in the event of a deeper-than-expected recession. This is somewhat complicated and relieved by FASB relaxaion of fair value mark-to-market valuations that takes the illiquid temporary turbulence effects out of market pricing. It is also helped that 3 month LIBOR has fallen to its lowest level since the ‘80s even if longer term funding from private sources remains a desert.
Following the policy (announced February 10) these banks were asked to estimate their short term future losses against capital to meet those losses, under adverse conditions that reflect a view of current conditions shaded on the relatively more benign side of the possible range of outcomes, as defined by the FDIC.
The teams of bank supervisors – specialising in specific types of assets, bank earnings capacity and reserves – then evaluated the banks’ submissions and asked for further information. They tested banks’ projections against independent benchmarks of their own tailored to the portfolio mixes of each bank. The problem here is that the authorities do not have complete macoeconomic models with sufficient financial sector detail to produce and adequatelyrobust and complete benchmark. It is easier to benchmark the banking sector to the economy than to directly correlate any one bank to the economy. The latter is almost impossible. The problem therefore is that whatever macro-prudential and macro-economic modelling the supervisors have to check with is not availabl to the banks themselves. There is, of course, considerable scope for judgement in these matters, but the all tolled the liklihood of the banks assessments agreeing directly with that pre-calculated by the supervisor team is zero! Hence, there will be considerable interative toing and froing between supervisor team and the bank. It may take until end of June to reach consensus agreement.
Taking into account likely losses, operating earnings (repeatable and one-offs) and own-capital reserves, the supervisors make a judgment about on whether each bank requires additional capital to guard against the risks represented by the stress test, with particular concern for whether that capital may be obtained from private sources or must be sought from the Federal Reserve, US Treasury, FDIC, and/or Congress and the White House.
There was both an overall tier-one capital test (which virtually every bank met easily) and a common equity test (which identified the need for many banks to hold economic-capital cushions against wider market and economy risks).
Supervisors did at least tell the banks about their assessment headline terms shortly in advance of visits, i.e. only last week – giving them a few days to crunch through their spreadsheets and challenge the findings. One immense difficulty of all this is that spreadsheets are not powerful enough to turn the handle on all of banks books and do not facilitate audit-trail drill-downs. We are dealing with macro- views top-down applying credit risk, market risk and liquidity risk assessments of large portfolios, large asset classes, where the magins for error greatly exceed the margins that the supervisors ant to have precision on i.e. increments of 0.25% of ratio to RWA or 0.125% of assets. Liquidity funding covers typically 30% of assets (the funding gap) and this is a larger ‘call’ on authorities financial resource than incremental top-up to Tier capital.
With assets of say $30tn of which $20tn are less than AAA, the funding gap is say $6-9tn that will require most of it rolling over perhaps within 6 months. Federal Reserve, US Treasury and FDIC balance sheets probably already have secured half of the requirement. The government is anxious not to let the banks deleverage to satisfy the remainder, and yet reluctant to be seen delivering yet more support in the $1+ trillion region, which would severely damage confidence just when the recession is hopefully in its last 2 quarters. The FDIC had also said in advance that relaining TARP and other sources of $150bn should be more than enough to see out this year. Hence, there is concern to keep the outcome of new funding requirement to below this, hence too Geithner’s plan to leverage the hedge funds into this role by offering them cheap loans to buy toxic assets. The hedge funds know they have political leverage here despite their general unpopularity (only a few notches adrift of the mistrust of the banks, except that at least bank shares are currently rising). This leverage resulted in their generous ‘exit’ clauses should they sometime down the track doubt their ability to make double-digit returns from buying up to $500bn of the banks ‘impaired’ assets.
Bankers say the authorities were willing to update their assessments in the light of changes to asset portfolios from the fourth quarter to the first quarter (reflecting FASB dispensation), but refused to put much weight on very strong first-quarter operating earnings when reaching their judgment on future revenues (which neatly agrees with maket sentiment that at first welcomed the record profits before recognising that these were mainly accounting changes and one-off gains).
Today’s announcement will in effect put banks into one of four categories: those (if any) that need extra tier-one capital; those that need to increase their equity buffer; those that have an adequate capital buffer under the stress test standard; and those that have surplus capital and could be eligible to repay bail-out funds subject to further conditions.
Nouriel Roubini and others, who with the credit ratings agencies, share one outlook in common, which is that there are no penalties just now for being excessively gloomy, advocates in good populist manner the dstruction of some banks to allows others to flourish in their place, a kind of fast-forward Darwinianism. Roubini complains that the tests have no precise insolvency threshold. In my view this is only sensible. Insolvency for banks is not a fixed moment in time event and there are at least six different ways of assessing insolvency. And, in any case, when government stands behind the banks, more so when it owns them, the normal regulatory and insolvency critieria no longer apply as before! Quoting the IMF estimates of losses, which actually added nothing we did not know already, Roubini reiterates that since losses have doubled in six months, the stress test results cannot “be credibly interpreted as a sign of bank health”. My response is, no of course not; these are not yet healthy patients. That is not the issue. The question is have they died or are the vital signs good enough, with evidence of at least near-normal brain-function not to turn off the life support?
Roubini is like a camp-doctor. These banks are cripples, let’s top them if they need “extra capital have, in fact, failed. As a result, these institutions will not be able to raise outside capital and will immediately require government help. Once again, the question will be how the near-insolvent banks can be kept afloat, to avoid systemic risk. But the question we really should be asking is: why keep insolvent banks afloat? We believe there is no convincing answer; we should instead find ways to manage the systemic risk of bank failures.” Good luck to that as a policy prescription: the complexity is mind-boggling.
Like a Eugenicist, Roubini says simply, “Trillions of dollars have been thrown at the system, just so that we can avoid the natural process of creative destruction that would take down these institutions’ creditors. Why shouldn’t the creditors bear the losses?”
He cites as counter-arguments, the “Lehman factor” and counterparty risk, the fear of being on the other side of a transaction with a failed bank”. But this includes the fact that Government is now a major counterparty and cannot risk the banks going under and all those T-bills being presented suddenly for redemption payments. Perhaps mindful of that, he suggests “a new insolvency regime for systemically important financial institutions is passed on a fast-track basis by Congress. Problem nearly solved.” Fast-tracking anything through Congress is an operational risk. I see the problem also that instead of allowing the vacuum of major failed banks to be filled by those with unreconstructed ethics, the government and regulators should continue with supporting the troubled banks in order to clean them up and ensure the country has better, ethically sounder banking. This is economically as important as anything else for the long term. Roubini has a fond belief that there is sufficient spring in the rest of the financial markets for them to rebound even if those big banks fall over. This I very much doubt. I judge that the interconnectedness of financial institutions is far far more complex a network than Roubini realises. Finance is not a competitive market mainly. The banks chased alpha unrealistically, fooling themselves into believing they are their on performance drivers. The difference in performance among banks is minor. The major factors are the economy, nationally and globally. I think Roubini should abandon a micro-theory view and get back to macro-economics.
The stress-tests, from which bankers have to learn this is how they must manage in the future, concern managing through the vicious circle between economic and financial weakness plaguing the economy, paving the way for a reliable recovery. Once that is achieved then other tinkering, Shumpeterian perhaps, is open to debate, not before.
Just now the focus is on narrowing the information gap between what bankers think banks are worth and what investors think banks are worth, they hope to enable many banks to raise the equity they need from the private market. Banks must clean up their balance sheets to attract more equity but not at the expense of the underlying economy.
Officials privately agree that the stress test is not a worst-case assessment but they argue that by intervening relatively early by the standards of past crises they can help ensure such an outcome does not materialise. The assessments are, however, based on a banker’s rather than an economist’s analysis of losses and that is their gross realism failing. They should though avoid the accountant’s snapshot view of solvency by allowing banks to value assets held to maturity at higher than current market prices, and gives them credit for their ability to earn their way back to health over time, taking account of time to generate debt recoveries.
The Federal Reserve tries to narrow the gap between the banker’s view and the economist’s view (two disttinctly different cultures) by taking a broad view of risks and a more forward-looking assessment of loan losses and recoveries. If the strategy succeeds in restoring confidence and drawing in additional equity, it should result in a decline in risk spreads on bank debt, reducing their funding costs and encouraging them to be more aggressive in lending to households and businesses. This correlates well with UK policy that has exceeded even the total support for banks of the rest of the EU. The other OECD countries will be under pressure to follow suit with publishing stress tests of their own. The EU is seeking stress tests on 47 banks, now reduced to the top 25. But these are late. This is because everyone under-estimated the difficulties of assessing scenarios and building models across so many national economies, and the lack of models appropriate to doing this work at a pan-EU level as well as individual country levels.