Quantifying Capital Flight
We have last talked about the flight of depositors from banks in peripheral countries as well as the refusal of the private sector in surplus countries to finance the current account deficits of the periphery in mid April in 'Inner-European Capital Flight'. The topic has continued to crop up frequently with the beginning of act three of the Greek crisis and the worsening crisis conditions in Spain.
Recently Citigroup analyst Matt King has attempted to quantify the extent of private sector capital flight by backing out TARGET-2 and other official inflows (such as SMP purchases and 'troika' loans) from the total foreign inflows according to the balance of payments data of the countries concerned. Note that private sector capital flight comes in different shapes – for one thing, bank deposits are withdrawn or sent to the 'euro area core' or elsewhere, for another foreign investors are selling their bond holdings (the latter means of course that domestic buyers or the ECB must step in).
The upshot of the exercise is that both Spain and Italy have experienced capital flight amounting to roughly 10% of GDP each, and seem destined to go down the path of Greece, Portugal and Ireland, i.e, the process seems likely to continue. Only, it will probably happen even faster, as investors are no longer as willing to delay such decisions as they were initially when the 'GIP' trio (Greece, Ireland, Portugal) got into trouble.
As King notes, bondholders face a very asymmetric risk-reward trade-off. He compares them to a 'flock of sheep grazing on coupons'. They are normally happy to keep grazing, but they won't sit still when their capital seems suddenly at risk. Since they are faced with very limited upside but a lot of potential downside, once they begin to run from risk it becomes very difficult to get them to change direction. Sort of like a herd of sheep becoming aware of a lion in their midst.
One of the factors that is accelerating the flight of foreign bondholders are ratings downgrades and the associated change in the weightings of bonds in certain index benchmarks. We have discussed this phenomenon before in the case of Portugal, which has experienced a crash in its bonds when it lost its investment grade rating.
Spain and Italy – The state of private inflows after backing out TARGET-2 liabilities, via Citigroup - click chart for better resolution.
Unfortunately, so King, a number of data are only available with a considerable delay, especially data on foreign bond holdings (we have previously noted that data from the euro area are often quite late. One must wait for up to two months to be able to calculate changes in the true money supply, and the same holds for various other central bank data such as TARGET-2 balances. The lag is even worse in the case of foreign bond holdings and the data tend to be inconsistent to boot). However, one can still infer with the above mentioned method how big the flight of private capital is in toto, and the speed and direction of the trend. Since the 'GIP' trio is further ahead in the process than both Spain and Italy, and there is an observable and easily explained trend – namely the fact that worried bondholders tend to continue their flight once it has begun, one can also infer where exactly Spain and Italy are probably headed in this context.
Holders of Spanish assets, disaggregated: 'MFI' stands for 'monetary financial institutions'. As can be seen, both foreign and domestic non-MFI depositors are fleeing Spain's banks, while foreigners have been net sellers of government bonds and RMBS, replaced by domestic buyers (mainly the banks – which in turn have been financed by the ECB's LTRO's and lately ELA) - click chart for better resolution.
The same exercise in the case of Italy. As can be seen, only foreign depositors have reduced their exposure – domestic depositors retain large exposure to Italy's banks – so far, anyway - click chart for better resolution.
Data on foreign bank deposits in the 'GIP' countries, as well as private sector government bond holdings in Greece and Ireland - click chart for better resolution.
The next chart shows the percentage by which foreign bond holdings have declined in the 'PIIGS' from their peak since the month the peak in foreign holdings was reached. This is the chart that allows one to extrapolate what will likely happen in Spain and Italy going forward. The main difference being that the amounts involved are truly staggering compared to the fairly small 'GIP' countries.
A chart that shows by how much foreign bond holdings have fallen in the PIIGS since the month of their peak. As can be seen, the trends are closely tracking, regardless of how far along the process each country is - click chart for better resolution.
As King notes, judging from the average decline in foreign bank deposits and bond holdings in the 'GIP' trio (52% and 33% respectively), a further €215 billion resp. €214 billion are left to go in the case of Spain and Italy, skewed toward deposits in the former and toward government bonds in the latter (note here that euro area banks together hold some €1.2 trillion in Italian government bonds – a sizeable chunk of these holdings has now shifted from non-Italian to Italian banks following the government's debt auctions and refinancings after LTRO 1 and 2 funding was received by the banks.
The First Rule of ELA: You Don't Talk About ELA
A recent article at Bloomberg discusses the secrecy surrounding ELA ('emergency liquidity assistance') financing and the problems created by the questions surrounding the Greek banking system. We have talked about the 'catch 22' the ECB finds itself in late last week, and this article adds some more color. Incidentally, there is a video found at that link as well, which shows an interview with Standard Chartered's chief economist.
What he says about the 'TARGET-2' imbalances is precisely what we have frequently pointed out for about the past year or so: they are not a problem, unless the euro area falls apart. Then they will become a major headache.
„The first rule of ELA is you don’t talk about ELA.
The European Central Bank is trying to limit the flow of information about so-called Emergency Liquidity Assistance, which is increasingly being tapped by distressed euro-region financial institutions as the debt crisis worsens. Focus on the program intensified last week after news leaked that the ECB moved some Greek banks out of its regular refinancing operations and onto ELA until they are sufficiently capitalized.
European stocks fell and the euro weakened on May 16 as investors sought clarity on how the Greek financial system would be kept alive. The episode highlights the ECB’s dilemma as it tries to save banks without taking too much risk onto its own balance sheet. While policy makers argue that secrecy is needed around ELA to prevent panic, the risk is that markets jump to the worst conclusion anyway.
“The lack of transparency is a double-edged sword,” said David Owen, chief European economist at Jefferies Securities International in London. “On the one hand, it increases uncertainty, but at the same time we do not necessarily want to know how bad things are as it can add fuel to the fire.”
Under ELA, the 17 national central banks in the euro area are able to provide emergency liquidity to banks that can’t put up collateral acceptable to the ECB. The risk is borne by the central bank in question, ensuring any losses stay within the country concerned and aren’t shared across all euro members, known as the euro system.
Each ELA loan requires the assent of the ECB’s 23-member Governing Council and carries a penalty interest rate, though the terms are never made public. Owen estimates that euro-area central banks are currently on the hook for about 150 billion euros ($189 billion) of ELA loans. The program has been deployed in countries includingGermany, Belgium, Ireland and now Greece. An ECB spokesman declined to comment on matters relating to ELA for this article.
The ECB buries information about ELA in its weekly financial statement. While it announced on April 24 that it was harmonizing the disclosure of ELA on the euro system’s balance sheet under “other claims on euro-area credit institutions,” this item contains more than just ELA. It stood at 212.5 billion euros this week, up from 184.7 billion euros three weeks ago.
The ECB has declined to divulge how much of the amount is accounted for by ELA.
(emphasis added)
Well, we want to repeat here that what goes for 'TARGET-2' balances also holds for 'ELA': the risk is actually not confined to the nations where it is employed. It really depends on what happens – if the euro area breaks apart, then ELA exposures will radiate core-ward from the weakest countries such as Greece. The putative exit of Greece from the euro area may soon provide an illustration of the process. This is so because ultimately the TARGET system is used to finance ELA as well – it provides a kind of unsecured credit line. Yes, there is collateral that the national central banks hold in return for providing ELA, but this collateral is highly dubious. It is as though a completely 'netted out' chain of derivatives were to end at Tchai Wallah's mud hut in Calcutta as the ultimate bearer of the underlying risk: no matter how carefully everybody has 'hedged' their exposure, the whole chain would only be as good as its weakest link (this is merely meant as an illustration – players in the derivatives markets are generally quite careful about their counterparties and about obtaining adequate collateralization – more so than they once used to be. The biggest danger in that murky space is the extent to which third party – speak customer – assets are employed for collateralization purposes).
There are a few more interesting remarks in the Bloomberg article (which describes ELA as 'part of the euro system's furniture'):
„Greek banks tapped their central bank for 54 billion euros in January, according to its most recently published figures. That has since risen to about 100 billion euros, the Financial Times reported on May 22, without citing anyone.
Ireland’s central bank said last year it received “formal comfort” from the country’s finance minister that it wouldn’t sustain losses on collateral received from banks in return for ELA.
“If the collateral underpinning the ELA falls short, the government steps in,” said Philip Lane, head of economics at Trinity College Dublin. “Essentially, ELA represents the ECB passing the risk back to the sovereign. That could be the trigger for potential default or, in Greece’s case, potential exit.”
[…]
A Greek departure could spark a further flight of deposits from banks in other troubled euro nations, according to UBS AG economists, leaving them more reliant on funding from monetary authorities. Banks in Greece, Ireland, Italy, Portugal and Spain saw a decline of 80.6 billion euros, or 3.2 percent, in household and corporate deposits from the end of 2010 through March this year, according to ECB data.
“ELA is a symptom of the strain in the system, and Greece is the tip of the iceberg here,” Owen said. “As concerns mount about break-up, that sparks deposit flight. Suddenly we’re talking about 350 billion, 400 billion as bigger countries avail of ELA.”
ELA emerged as part of the euro system’s furniture in 2008, when the global financial crisis led to the bailouts of German property lender Hypo-Real Estate AG and Belgian banking group Dexia. While the Bundesbank’s ELA facility has now been closed, Dexia Chief Executive Officer Pierre Mariani told the bank’s shareholders on May 9 that it continues to access around 12 billion euros of ELA funds.
ELA was a measure that gave central banks more flexibility to keep their banks afloat in situations of short-term stress, said Juergen Michels, chief euro-area economist at Citigroup Global Markets in London.
“It seems to be now a more permanent feature in the periphery countries,” Michels said, adding there’s a risk that “the ECB loses control to some extent over what’s going on.”
The ECB was forced to confirm on May 17 it had moved some Greek banks onto ELA after the news leaked out, roiling financial markets. The ECB said in an e-mail that as soon as the banks are recapitalized, which it expected to happen “soon,” they will regain access to its refinancing operations. The ECB “continues to support Greek banks,” it added. By approving ELA requests, the ECB is ensuring that banks that would otherwise not qualify for its loans have access to liquidity.
“The ELA is a perfect life-support system, but it’s not a system for what happens after that,” said Lorcan Roche Kelly, chief Europe strategist at Trend Macrolytics LLC in Clare, Ireland. “What you need is a bank resolution mechanism, a method to get rid of a bank that’s insolvent. In Ireland, and perhaps in Greece as well, the problem is that you’ve got banking systems that are insolvent.”
(emphasis added)
As far as we're aware, the FT is correct with its €100 billion ELA estimate for Greek banks, as the Bank of Greece has recently applied for an increase from €90 billion to €100 billion. However, we probably will have to revise our own estimate of the likely TARGET-2 liability of Greece's Central Bank in light of the above mentioned size of ELA in January. Since that was roughly €54 billion, we would now estimate the latest TARGET-2 balance to be in the region of €150 billion rather than €135 billion. This can be inferred from the balance sheet the Bank of Greece published in January and the new ELA estimate.
In any case, as Mr. Kelly from Trend Macrolytics points out above, 'it is a perfect life support system' for zombie banks, but 'not a system for what happens after that'. As to the ECB trying to pass the risk buck back to the sovereigns that as a rule 'guarantee' the IOU's the commercial banks issue as collateral to their NCB's in ELA operations, this is completely moot in the case of insolvent sovereigns or in case the guarantee itself is what ultimately bankrupts the sovereign. Whoever extended the funds will end up eating the losses.
Stop the Presses: Losses for Bank Bondholders Are Now Part of the Plan? Well, Sort Of….
The following has received little attention, but it appears that some in the euro area have decided that inflicting losses on the unsecured senior bondholders of insolvent banks may after all not be such a bad thing.
Wow, capitalism for bank bondholders – this is an almost revolutionary concept these days. Will the world keep turning? Well, that is really the question – since these 'bail-in powers' will only be at the 'disposal of regulators' from 2018 onward – and as everybody knows the world will end in 2012. Wait a moment though – say what? 'Bail-in powers'?
Since when are special powers for regulators required to make an unsecured creditor eat a loss? Does the term 'unsecured' mean anything at all? Is this how far the banking cartel has insinuated itself into the legal machinery that in order to let a bank fail, 'special powers' need to be invoked these days?
Bloomberg reports:
“The European Union will seek to give regulators the power to impose writedowns on senior unsecured creditors at failing banks as part of measures to prevent taxpayers from footing the bill for saving crisis-hit lenders.
The writedown powers would apply to senior unsecured debt and derivatives, while some other claims, including secured debt and deposits that are protected by government guarantee programs, would be shielded from the losses, according to draft plans obtained by Bloomberg News. Regulators would have the so- called bail-in powers from the start of 2018.
The cost of bank funding may increase as investors take on board the plans “but this is a natural process of getting taxpayers off the hook,” John Vickers, the chairman of the U.K.’s Independent Commission on Banking, said in a speech in Brussels yesterday. “It’s not sinister that funding costs go up if it’s for that reason.”
EU Financial Services Commissioner Michel Barnier had delayed proposing the law, which was originally scheduled to be released in September 2011, because of market turbulence. The Bloomberg Europe Banks and Financial Services Index has fallen 35.8 percent in the past year on concerns lenders have been weakened by the European sovereign-debt crisis.
Any bail-in of bank debt would “be accompanied by the removal of the management responsible for the problems of the institution,” according to the draft proposals, prepared by Barnier’s staff at the European Commission. The bank would face restructuring “in a way that addresses the reasons for its failure.”
(emphasis added)
Readers may recall we have mentioned Mr. Barnier before – and rarely in a favorable manner. For instance, he was at the forefront of attempting to ban 'naked' CDS trading – meaning, trading would only be possible by 'appointment' and with his bureaucracy's placet. He was also calling for short selling bans on financial shares. In other words, he wanted to remove the possibility for anyone to hedge themselves against profligate governments and floundering banks. It doesn't seem to have occurred to him that in a derivatives transaction such as a CDS contract, at the very minimum one of the parties is always 'naked' (if a holder of a government bond wants to buy insurance, the seller of the insurance does not need to have a position in the market concerned – he must only be willing and able to bear the risk).
But the proposal to not only let unsecured creditors of banks assume the losses that are rightfully theirs but in addition to actually 'remove the management responsible for the problems of the institutions' concerned, why that is… it leaves one almost speechless.
Imagine that! Management actually accepting responsibility for running a bank into the ground by resigning! Why has nobody thought of that before? And from 2018 onward to boot, that is practically only a few heartbeats away! Barnier the anarchist!
Presumably, until then, one has absolutely nothing to fear as an unsecured bondholder of an insolvent bank, while management can rest assured that it will simply remain in place to continue to administer the carcass if need be and grab as much bonus loot as possible while the getting is good. This is like a heads-up to the rats on the sinking ship, telling them that six years from now, they may well be facing a life-raft ban. We'll check back on this in 2018 and will let you know what happened.
To be fair, the article does mention that there are a few courageous souls demanding that these new rules be instated right away.
Credit Market Charts
Below is our customary update of credit market charts, including the usual suspects: CDS on various sovereign debtors and banks, bond yields, euro basis swaps and a few other charts. Charts and price scales are color coded (readers should keep the different price scales in mind when assessing 4-in-1 charts). Where necessary we have provided a legend for the color coding below the charts. Prices are as of Friday's close.
A few CDS spreads have moved to new highs for the move (e.g. those on Italy and Greece), while others have retreated somewhat – so last week ended on a mixed note. There were noteworthy moves in CDS on several CEE nations.
5 year CDS on Portugal, Italy, Greece and Spain - click chart for better resolution.
5 year CDS on France, Belgium, Ireland and Japan - click chart for better resolution.
5 year CDS on Bulgaria, Croatia, Hungary and Austria - click chart for better resolution.
5 year CDS on Latvia, Lithuania, Slovenia and Slovakia - click chart for better resolution.
5 year CDS on Romania, Poland, the Ukraine and Estonia - click chart for better resolution.
5 year CDS on Germany (white) , the US (orange) and the Markit SovX index of CDS on 19 Western European sovereigns (yellow) - click chart for better resolution.
5 year CDS on Bahrain, Saudi Arabia, Morocco and Turkey - click chart for better resolution.
Three month, one year, three year and five year euro basis swaps – following a relentless plunge throughout most of May, the longer dated basis swaps finally manage to bounce - click chart for better resolution.
Our proprietary unweighted index of 5 year CDS on eight major European banks (BBVA, Banca Monte dei Paschi di Siena, Societe Generale, BNP Paribas, Deutsche Bank, UBS, Intesa Sanpaolo and Unicredito) – in consolidation mode - click chart for better resolution.
10 year government bond yields of Italy, Greece, Portugal and Spain - click chart for better resolution.
Austria's 10 year government bond yield (green), Ireland's 9 year yield (white), UK gilts (yellow) and [the Greek two year note yield (orange) – please ignore that one – we couldn't persuade the system to spit out the correct yield. It really stands at 225%]; but UK gilt yields and Austrian 10 year yields at new lows are certainly noteworthy - click chart for better resolution.
5-year CDS on Australia's 'Big Four' banks. Still reflecting 'risk off' mode in financial markets, which is to say, risk for them is increasing – click chart for better resolution.
Addendum 1: Greek Police Enters Deposit Flight Debate
The latest organization deployed to dissuade Greek citizens from getting their money out of the banks post-haste is apparently the Greek police. The argument appears to center on the possibility that withdrawn money could become the object of desire of enterprising thieves. Opportunistic crisis-profiteers, so to speak.
“Police are urging Greeks to keep their money in bank accounts rather than putting it at risk of theft, amid further uncertainty about whether the austerity-struck country will remain in the eurozone.
Greece's banks are likely to be shored up on Friday or Monday with €18bn (£14bn) of bailout funds they have been due to receive for weeks but which were held up by political uncertainty caused by inconclusive elections. Greece goes to the polls again on June 17, further stoking fears about its future within the euro.
The scale of withdrawals from Greek banks – almost 25% of deposits have been taken out in the past two years – and fears that other countries may suffer mass withdrawals has led to speculation that a eurozone-wide guarantee is needed to maintain confidence in the banking system.
Greece's national police spokesman, Thanassis Kokkalakis, told Reuters: "Many people have withdrawn their money from the banks fearing a financial crash, and they either carry it on them, find a hideout at home or in storage rooms.
"We urge people to trust the banking system, leave their money there, or at least in a safe place, not hide it at home, where they must anyway take the basic security measures."
(emphasis added)
Well, dear Greek citizens, as anyone in Argentina can tell you: it is so much better and more fun when the banks or the government steal your money! At least it will then have been used for the noble cause of propping up the banks and their lackeys in politics – what more can you ask of your deposits?
Addendum 2: Dallara's Latest Apocalyptic Predictions
IIF chief Charles Dallara was let loose from his cage for a few moments last week in order to inform us that a Greek exit from the euro would still – even after the PSI deal – be the equivalent of Lehman squared. Or perhaps even Lehman cubed.
He actually has a point insofar as all the debts contracted in euro terms would need to be sorted out after a Greek exit from the euro, a task that would most likely see countless defaults occur (not to mention that it sounds like a legal nightmare of vast proportions). Also, French banks in particular continue to be exposed to Greek debtors in the private sector, so there would presumably be quite a few direct losses to banks in addition to the several hundred billion euros various public sector entities (the 'troika') would likely have to write off.
Dallara's favorite number is €1 trillion 'or more' in total losses. It's a nice, round, fat and scary number. He even says that 'the ECB will be rendered insolvent', which we rather tend to doubt, since it is in possession of a printing press. This is like saying that the Pakistani mafia will run out of faked Indian rupees. Not likely going to happen.
Dalllara thinks therefore that someone should quickly throw an additional €10 billion at Greece to 'help its economy' (that someone being the euro area tax cows outside of Greece). How that would alter the situation (except that another €10 billion would go down the drain) is not made quite clear. Curiously he sounds rather more sanguine about Spain, although 'contagion effects' are a big part of his loss estimate.
Charts by: Citigroup, Bloomberg