Friday, 6 March 2009

Quantitative Easing, Asset Protection Scheme, and bank nationalisations

QE is a possible replacement (not just supplement to) for asset swaps at Central Banks' liquidity windows for banks in central banks money market operations (of the 'less open' sort). For those of you who don't read all the financial pages, the world-model for this is the Bank of England's Special Liquidity Window, SLS. This closed on 19th January, having operated since April 2008, since when it swapped about £245bn of UK banks' Asset-backed Bond Securities, ABS, (mainly mortgage loans. packaged into bond structures, and sold in Special Investment Vehicle companies, SIVs, that are firms only set up, and managed by agents, just to handle the payments and standby liquidity between banks and investors so that all of this is now 'off-balance sheet').
The £245bn or so ABS was swapped (after heavy margin write-downs and haircut fees) for £185bn of treasury bills (government bonds with less than 1 year maturity) that can then be used to replace private sector funding (and redemption of banks' maturing medium term notes and 'covered bonds' they issued) to re-fund the 'funding gap' between customer loans and customer deposits. The swapped bonds are worth far more than all the gold in the Old Lady's bottom drawer (as in our Bank of England vault pictured above). Assuming you have got that, now let's move on.
Next stop, "son of SLS", the UK HM treasury and Bank of England Asset Protection Scheme, APS.
APS is alongside, and arguably also part of the permanent "son of SLS" liquidity window at the Bank of England (BoE) that for an additional fee is available to UK banks any business day, not just periodically. The BoE might have continued with an exact repeat-copy of SLS, but it had some second thoughts - intelligent ones. The way SLS was heading was that £500bn would be swapped by end of 2009. But, was this enough if it only provides 3-year roll-over compensation for UK banks' credit crunch asset write-downs, even when that will exceed the total of the UK big banks' total equity capital? As much again would be needed. But should this still be done via treasury bills?
The problem with SLS is that the assets swapped are held as collateral to support the T-bills. The assets are held for 3 years and therefore the T-bills must be rolled-over. But, what if the banks that got the bills sell them or run into any difficulty such that the bills fall in other hands and are presented back to the BoE for payment on maturity date and not for roll-over (paid for with some new bills)? The difference between collateral held after swapping for bills also backed capitalisation investments in UK banks' preference and ordinary shares, off the Government budget, and if UK banks did not hold their bills to maturity for any reason, which in theory and for accounting purposes they did not have to do, then those capitalisation investments could be disrupted.
This then brings us to QE, Quantitative Easing.
In QE, a UK bank's bonded assets are swapped for a BoE cheque that is paper cash (just as a Bank of England banknote is a bearer's cheque) but to be held on deposit as part of the UK bank's liquid deposits (reserves) at the BoE. That way the central bank has direct liabilities (deposits) to balance-off against its bought-in assets (3 year collateral). The press comment noted, however, that the BoE will swap cheques for assets with non-banks! That's just a technicality, 'non-banks' only means from SIVs that UK banks set up to process their ABS off-balance-sheet; it does not mean 'shadow-banks' like hedge-funds (unless we count SIVs as parts of shadow-banking - and why not?). The cheques have to end up remaining at the BoE as part of UK banks' deposits there, which is not something for any actual non-banks to be able to do! The media also imagine that the BoE's MPC Committee must approve this? Well yes, but only because the academic literature defines Quantitative easing as a sort of 'printing money', actually signing cheques, which means boosting the money supply and thereby risking an inflation-response in the economy! This is just a pro-forma idea, more supposition than reality. The money supply is global as well as domestic and inflation is not currently a threat or undesirable. A dose of inflation, like negative real interest rates, or even if imported inflation, could speed up debt repayments and debt rec-cycling and might put a floor under property price falls. Property is involved in everything and banks accounts (unlike inflation-adjusted GDP growth numbers) are unconcerned about inflation so long as not making credit risks worse or squeezing net interest income. Inflation might help to draw a line under the number of banks to be saved. Why also QE is attractive is because under liquidity windows, the BoE SLS-type, the t-bills received by banks, bought with off-balance-sheet bonded loan-assets) could be sold, traded, to get cash, when BoE need the banks to hold the T-bills in capital reserves where they can then be rolled-over so the debt issuance doesn't have to appear on the Government budget like Government bonds (gilts) do. Then, if banks sold their t-bills, there is a theoretical fear of short-term crowding-out of private capital by government capital (actually the opposite happens, but no mind this is more than half-respectable Monetary Theory)- but it's not that really anyway; that's just very hard to prove theory, Keynesians say unproven and wrong because treasury bills and bonds are nearly 100% liquid instruments. Worth repeating: the BoE (ECB too in Euro zone) fear that if sold on (possibly by forced-sale due to cash-flow embarassment or takeovers) these t-bills reappear to be presented for payment, when central banks want to be sure they can simply roll them over i.e. not redeem them, not for 3 years, the period they will hold the swapped assets. The central banks don't want to redeem the t-bills until they've made money from the collateral assets, and from the difference between the face and write-down values that they've used to back other treasury spending on capitalisation of banks. Hence the safest way to proceed from BoE (ECB and Federal Reserve too) perspective with "son of SLS" is QE. And, yes, having fully liquid deposits at BoE means the UK banks have that as backing for 'cash' payments i.e. so-called 'printing money', but actually no different than t-bills except t-bills have to be sold on, or borrowed against as short=term collateral, to get the money and could come back to embarrass the BoE. Cash-flow support via QE does not come back to haunt the BoE i.e. like people demanding gold for their grubby notes. And, yes, there is a money supply and inflation aspect, but also consumption spending and higher imports and wider trade deficit. BoE might like to be telling the banks to use the money to support loans to infrastructure, fixed investment customers e.g. property developers, to keep the building of new housing stock built, or better to assist industrial manufacturers to make exportable products, even if only building inventory ahead of recovery? I spoke about all this on Sky News today. The same is true, however, in the USA.
What is happening in the USA is also so in the UK. An important context for the Lloyds deal is the rise in CDS spreads this month that echo the jump in spreads in September when Lehman Brothers crashed that led in the USA, UK and other countries to bank bailouts. CDS spreads are risk price indicators for the cost of interbank funding, but also reflect risks within the net CDS exposures of banks because of the huge $50tn size of the CDO and Synthetic CDO markets that greatly exceed the underlying assets to be insured. Got that?
What was the lesson of Lehman Brothers? It followed from Bear Stearns, Northern Rock, and other reminder lessons that the most important asset a bank has is confidence, just as the most important lesson of a whole market is also confidence, direct confidence, not just confidence about the underlying assets or the underlying economy. If people are confident in a bank, it can continue to do business; if not, it can’t. For the last six months, where has that confidence been won or lost? Not in banks’ balance sheets, which is a big concern for audit firms as well as banks. The balance sheets are mistrusted when they absolutely should not be! Is confidence won or lost in the various bouts of capital and targeted asset guarantees provided by Treasury and the Fed, HM Treasury and Bank of England, ECB and others? These support a widespread assumption that the government will not let the creditors of large banks lose money, out of fear of repeating the Lehman debacle; only shareholders are at high or near-absolute risk.
Let’s say that AIG, Citigroup, RBS, or LBG are restructured (less ruthlessly than say Fortis) via government conservatorship so that creditors did not get all their money back (Unlike the 100% guranatee in Ireland or as applies to FDIC-insured deposits in the US or to recently-issued senior debt explicitly guaranteed by the government), they may be forced to convert debt for equity, higher risk/higher return, or be stiffed! A concern is that this could undermine further confidence in banks. According to the FT, bank bonds are one quarter of all U.S. investment-grade corporate bonds. I suspect the % in Europe is similarly high. Losses would be spread far and wide, hitting other banks, pension funds, insurance companies, hedge funds, and so on. If banks cannot support their derivatives exposures, then institutions that bought CDS protection from face further losses. (Banks were net buyers of CDS protection i.e. they invested in them for speculative gain more than just for default insurance.) The fear is that it will be impossible to predict how these losses will be distributed and who else might be dragged under. I'm indebted here to 'Contrahour' from
Lehman did not seem to force any other major firms into bankruptcy, although it may have severely wounded AIG since it was counterparty to a dominant share of CDS contracts. Once investors figure out that bank debt is not safe generally, they will refuse to lend to banks, and we are back in September all over again, and this is what CDS spreads are now telling us is the case today. But, now, this is despite governments' massive delivery of liquidity into their banking systems to keep banks functioning? So what changed this week?
One theory is that the semi-forced conversion of Citigroup, RBS and LBG of preferred into common shares is a sign that government may try to force creditors to exchange their bonds for common stock in future bailouts. Preferred shares are not, technically speaking, debt. But they are a lot like debt (fixed coupon dividends), and once you finish converting preferred into common, the next layer of the capital structure is subordinated debt. So the markets are left wondering, and we know markets don’t like large uncertainty. Another possibility is more people are thinking that the government may end up restructuring bank debt. In the FT, for example, Martin Wolf and Willem Buiter, both very serious and globally influential opinion-formers, question whether government should be protecting bank creditors. Wolf, I believe, favours yes for now, but Buiter says no.
Each time the lines on CDS charts, as above, spike up, government has to be seen taking action to imply that creditors will be protected, without making severely testable promises. Hence the bad bank APS in the UK implemented this week. Chances are we’ll see more along these lines. At some point, though, the government may lose credibility and then, disastrously, no source of confidence remains. But, next stop G20 in early April for a global confidence-building inter-government conference.
In June, Sweden takes over the EU's 6 month Presidency. One of the steps in Sweden’s sometimes-heralded bank rescue program was an explicit government guarantee on all bank liabilities, as per Ireland's response too and its nationalisation of Anglo-Irish. But, if any country could wholly-guarantee its banks, you would think it would be the U.S. But the real barrier to taking such a step is probably political - the accusation that this is the road to communistic socialist control of the commanding heights etc. No-one should underestimate US concern, sometimes bordering on paranoia, about the Federal tax-dollar, one reason why EU fiscal centralisation has not proceeded further. Usefully, to understand the technicalities of governments' current initiatives is Morgan-Stanley bank's definition of QE as follows (extract): "Unlike the Asset Purchase Facility already in operation, these asset purchases won’t be funded by the issuance of Treasury securities (which effectively withdraws money from the economy, offsetting the injection of money from the purchase of the assets). The Bank of England will buy assets in exchange for, effectively, a cheque from the Bank of England. The seller then deposits the cheque at its bank which (assuming it banks directly with the Bank of England) results in that amount being added to the commercial bank’s reserves at the Bank of England. The seller of the asset has less of one type of asset (e.g. gilts) and more cash. The commercial bank has more deposit liabilities and increased assets at the Bank of England. The Bank of England has increased liabilities in the form of commercial bank reserves and increased assets in the form of the securities it has purchased. Money has been created (‘printed’)." For ,ore see:
Additionally, it is a moot point that the bonds and bills are actually HM Treasury property via the DMO (Debt management Office), not exactly Bank of England's and therefore aspects of that can alter the accounting in respect of national debt and budget deficits. By shifting to BoE cheques instead of treasury bills we get a build-up of liabilities at BoE that can then build assets in the banks itself and take the pressure of HM Treasury! This is politically neat.
The context for QE includes an agreed deal already for over £325bn between RBS and BoE. Another deal is being negotiated for £258bn Lloyds Banking Group, LBG. A problem has arisen here though, it seems? The swap requires fees upfront, writedown and haircut plus interest payments. If the assets are of the riskiest sort, then the coupon and standby first-loss liquidity protection can be quite heavy and may require 'paper losses' reported in the P/L which then further postpones coupon payments on preference shares and so on. The problem may be that LBG feels suitably cash-flow liquidity constrained that it wishes to make the payments to the BoE (in effect actually to HM Treasury's DMO which manages Government bonds and bills) in shares, preference, or given the scale ordinary full-voting shares. This could then take the Government's share of LBG above 50% and possibly to 70%! What happens then if almost half of the UK mortgage market is government-owned and more than half of everybody is doing financial savings and borrowings with government-owned banks, and so on, and maybe on to a degree of government control and knowledge about everyone, directly or by cutouts such as UKFI, Big Brother state, actual or potential, just imagine the leader writers of the press starting to write "know what, good citizens, we have asked our panel of professors are we or aren't we... and they have all agreed... the UK is now a fully-fledged, centrally-controlled Camminist State in all but name, Long Live Comrade Brown!" It is bade enough that Private Eye magazine devotes a whole page fortnightly to precisely this amusing idea. Imagine all this belonging to the Party of "hard-working families" (Chairman Brown's acclaimed constituency today), the party of hard-working taxpayers and the poor, but not of course of bonus-earners, non-doms, and unearned-income shareholders. This, as much as anything else, is what lies behind deep resistance by Government/s to nationalise the banks, because it means effectively, and variously, nationalising most of everything else. banks are that important, truly the commanding heights of the economy, Clause 4 risen from the dead and straight into power! Lloyds shares were up 6% in mid-morning trade and remained over 5% up at end of day, while oddly other major UK banks' shares fell? The oddity is that ordinary shareholders in LBG are about to be diluted. Perhaps they have calculated that on balance the bank will be worth that much and more, and that the long negotiation suggests that outright nationalisation, as Vince Cable MP says seems now only sensible, will not happen and the long term faithful shareholders will not be wiped out. Short-sellers too must be steering clear, or no-one is prepared for once at this sensitive time to lend out LBG shares only to get them trashed. Both the government and Lloyds said nothing had been agreed by mid-morning, after the BBC broke the story at 9am, and silence persisted.
The BBC reported that under the plan the government "would insure" up to £258bn of assets, and would get non-voting shares in return, which actually means that the preference shares would back the standby liquidity reserve for the SIV assets. The amounts are also significant in that for both banks, RBS and LBG, they virtually cover 100% of their current year 'funding gap' re-funding requirement. But what the experts and public may not know is that in fact the assets are not just those of the two banks, but very likely will include securitised assets from other smaller banks and building societies that included large dollops of theirs in RBS and LBG (LTSB and HBOS) securitised assets in SIVs and covered bonds such as the HBOS rolling $125bn program or Lloyds other SIVs, but not including the well-known SIVs, Grampian and Cancara.
The BBC speculation is that the deal could also mean the bank swapping some of the government's existing non-voting preference shares, on which the bank currently pays 12% interest, for ordinary shares and that it is believed the interest on these shares costs Lloyds about £480m a year. One can only suppose that there is a price for converting fixed coupon shares for zero dividend shares and this will be paid for in more shares too? I find it very hard to believe the bank is so cash-flow sensitive! But, although the numbers are large in ratio to GDP, they are not if we just consider that what is happening is the public sector displacing the private sector from what in reality is lucratively good business, bank-financing. The swapping of preference shares for ordinary shares would be seen as important for Lloyds' cashflow says BBC political correspondent Reeta Chakrabarti, who added that an agreement between Lloyds and the government should have been reached a week ago, but both sides confirmed discussions remain ongoing. A spokesman for Lloyds Banking Group said conversations had not yet concluded because there was still "a good deal of detail to be worked through", and actually this is as much to do with legal drafting paperwork as it is to do with spread-sheeting the financials and doing some risk modeling. Of course, had they asked me I'd have modeled it all before lunch and structured a call-off contract arrangement with an algorithm for quarterly revisions so that a provisional agreement could be announced and let the lawyers and investment bankers turn their clocks off. A HM Treasury spokesman said in almost perfect Rumsfeldian: "Discussions are ongoing. A deal will be announced at the conclusion of those discussions." Do we take this to mean that "no deal" is "no option"? [Agreement was announced next day, Friday, resulting in government shareholding in the bank growing from 43% to 65% voting shares and 75% overall, just under the 80% threshold.]
Meanwhile, across all channels, Liberal Democrat Treasury spokesman Vince Cable, a most honourable Member of Parliament, described full nationalisation of Lloyds as being, not just in his view, but somehow according to some Golden Calf god, "inevitable" and has called on the government to bring the whole banking sector under public control. I don't see it as technically or otherwise inevitable, whether desirable or simpler or cleaner is another matter. BBC political correspondent Carole Walker said Vince Cable is not alone in believing full nationalisation will restore confidence in the banks more quickly, saying, "There are a number of people in Westminster who believe we are getting closer and closer to full nationalisation... [They think] it would be simpler and easier for the government to nationalise the banks and that in the long-term it would save the taxpayer a lot of money." Sorry, but they are dead wrong there. Simple does not mean easier and this is about medium term not long term. Gordon brown wants to clean out the banks, but nationalisation means beginning with cleaning out the shareholders, but what problem are they, and how much less is the prospect of a profitable return for taxpayers if the banks shares are no longer quotable or able to rise on the good news in April from the G20 and possible GDP turnaround in late 2009 and in 2010?
Restoring confidence should extend to share values as a proxy signal to bondholders and potential bond investors who would profitably redeem the Government's stakes. Nationalisation removes some market signaling that in other improved future circumstances are useful, valuable. It also brings in some other problems of what is or is not in the national debt and on-budget and what is or is not in or outside regulatory supervision as well as competition issues. The Euro zone is operating on 2 year bailout timescale. The UK government is on a 3 year timeline. What happens then to level-playing field EU political-economy laws and markets? The deal is part of the Treasury's taxpayer-backed Asset Protection Scheme, APS,to "insure" banks' riskiest assets against further losses - actually much more than that and really funding gap funding and general balance sheet protection too. It was put forward by Chancellor Alistair Darling, no doubt with Mervyn King, to "restore confidence in the banking sector", which while true is a care-worn somewhat threadbare mantra after a year or more of endless repeating. the problem seems to be that Ministers cannot speak out authoritatively about banking in any detail, because who'd believe their professional credentials for doing so, and bankers cannot do so either, not even central bankers and regulators, because they are all disgraced and discredited publicly and no-one trusts them or their motives! This is why international coordination on bank regulation and fiscal responses is politically important apart from being economically important. The last bastion it is assumed where voters might venture trust and belief is when governments across the world all seem to agree. Gordon brown and his government is therefore hugely depending on the success of the G20 London Conference on 2 April. This was the essence of his speech today at the Scottish labour Party Conference in Dundee, which in turn repeated much of what he had been saying earlier this week in Washington and in addressing the Congress and Senate. He is truly a global statesman hero, Flash Gordon, and I truly hope, and believe, he can, and is, playing that role fully, with integrity, and most successfully. Conservative Shadow chancellor George Osborne, less than artlessly, continues to deliberately fail to get it, believing still that his best best is to undermine the last bastion of public trust, government handling of global economic crisis, by sniping in true ya-boo political bad-form that government efforts to get the banks funding each other again may be likened to "insuring the car after it has crashed". He is keeping pace with the Republican Party minority leader in the US senate who also says this is a time for government belt-tightening not taking the belt off! In my view both have their trousers round their ankles and should be thinking about how to restore some dignity to their politics in this time of crisis! This is absolutely not the time to be undermining confidence in governments' economic or political authority! Opposition parties should envisage the crisis as if we are in a wartime economy and value the importance of all loyally pulling together. Normal service in politics may stick with non-economic policy matters and otherwise wait until more normal political-economy conditions return.

1 comment:


Sweden's financial crisis (alongside Finland's) in the early 1990s (when Soviet Union crashed and there was an US-UK-Can-Aus severe recession followed by an EU 1-2 years later), and by many accounts did a reasonably good job of recovery. A housing bubble, fueled by cheap credit, collapsed in 1990 (same time as the much more enormous Japan crash, then US and UK for same reasons), with residential real estate prices falling more lack japan with persistence by 25% in real terms by 1995 and nonperforming loans reaching 11% by 1993, while the Swedish krona fell in value by 30%, hurting a banking sector largely financed by foreign funds. Continental Europe property values remained flat and Switzerland entered long low growth that was echoed in late 90s in Ecu snake/Euro zone area). As Urban Backstrom in a 1997 paper said, “aggregate loan losses [of the 7 largest banks] amounted to the equivalent of 12 % of Sweden’s annual GDP. The stock of nonperforming loans was much larger than the banking sector’s total equity capital.” The banking sector as a whole was broke, and so too was Finland's, Japan's, Russia's, others, as was a few years later was Mexico, and nearly all SE Asia retail banks, including state-banks even in China, and perhaps, most severely of all, the case of Ireland today that was the first to follow Sweden's example.
So what did Sweden do? If the options on the table today in OECD countries are (a) additional recapitalization, (b) an aggregator bank to buy up bad assets, and (c) nationalisation, the Swedish solution included all 3. First, in late 1992, the government guaranteed all bank creditors (not shareholders), 100%. Because investors did not at the time question the solvency of government, they would continue to lend money to banks, and the central bank provided unlimited liquidity.
The U.S. today has guaranteed new debt issued by banks, and there is an implicit blanket guarantee for at least the largest banks, there is still uncertainty among bank creditors, as indicated by CDS spreads.
But, even if an insolvent bank has access to credit, it can still be an insolvent bank (capital wipe-out by credit defaults and market asset write-downs), hoping how to win time to become solvent at cycle upturn, so it’s unlikely to grow lending and may seek to shrink its loanbooks, or worse, it may be tempted to make extremely risky loans or buying distressed assets or other more distressed banks cheap, to thereby gain faster return path to solvency?
As a condition of Swedish government support, auditors reviewed the balance sheets of the all the banks involved, with the goal of taking write-downs immediately and paring back to a true statement of net book value. When it turned out that 2 big banks, Nordbanken and Gota, were totally insolvent, they were nationalized (Nordbanken was already part state-owned), giving the state control of over 20% of the banking system (by assets). Gota was merged into Nordbanken, which only held onto the “good” assets, while the “bad” assets were moved to two new entities, Securum and Retriva. S & R were capitalized by government, respectively, 21% of Nordbanken’s and 45% of Gota’s assets. This was the good bank/bad bank plan that gets much study lately. Nordbanken (the good bank) was recapitalized by the government, at 3% ratio to GDP, and became a healthy bank, while Securum and Retriva were told to get whatever value they could out of the bad assets. This model has been followed in Belgium with the toxic rump, written down severely to Eur10bn of Fortis, while the good bank part is sold to BNPP (subject to shareholder legal actions that are destabilizing government).
S & R were run like a cross between private equity firms and asset management companies, managing and improving assets and also finding buyers for them. According to the Cleveland Fed, S & R managed to return $1.8bn out of their $4.5bn in initial capital to the government, for a taxpayer net loss of $2.7bn. (Government may not have lost money on its loan guarantee).
Nordbanken, after being run by the government, was eventually privatized (government’s share is now 19.9%), and the taxpayer recovered the capital put into it.
This is generally seen as a success story, although the Cleveland Fed has a sobering conclusion:
"The cost of the crisis to Sweden was not limited to the capital spent by the [asset management companies]. There have been significant income and output losses associated with the crisis. In the early 1970s, Sweden had one of the highest income levels in Europe; today, its lead has all but disappeared."
Cerra and Saxena (2005) found the crisis caused a permanent decline in output that can explain the entire fall in Sweden’s relative income. So, even well-managed financial crises don’t all have happy endings.
The Swedish story is used as an argument for nationalisation. But another lesson you can draw is that it’s not the nationalisation per se that matters, but the pricing of the bad assets. The key was that the banks were forced to write down their assets in one shot and then to sell them to the bad banks at current market prices. That cleaned up their balance sheets and, once they were recapitalised, allowed them to operate as healthy banks.
TARP in the US was this idea as long as it paid fair value for assets and was combined with recapitalisation to fill the resulting holes in banks' balance sheets.
The same holds for an 'aggregator bank'. The problem would be letting the banks decide which assets they want to sell, and then letting them unload them on the aggregator bank at inflated prices. That in the bank of England's case was SLS, but the assets were treated as collateral and deeply discounted by 25% plus haircut plus interest margin on the treasury bills loaned in return. What all this solves is buying time through the credit and economic downturn. Thereafter, much of the defaults have hardened into realised losses less recoveries and as this takes time there is a medium term overhang during which the banks do not respond well to expand their lending in the economic upturn. Recovery growth rates are thereby low or flat! But, that problem is for a another much later day.