27/12/2014: Geography of the Euro Area Debt Flows

The debate about who was rescued in the euro area 'peripheral' economies banking crisis will be raging on for years to come. One interesting paper by Hale, Galina and Obstfeld, Maurice, titled "The Euro and the Geography of International Debt Flows" (NBER Working Paper No. w20033, see http://www.nber.org/papers/w20033.pdf) puts some facts behind the arguments.

Per authors, "greater financial integration between core and peripheral EMU members had an effect on both sets of countries. Lower interest rates allowed peripheral countries to run bigger deficits, which inflated their economies by allowing credit booms. Core EMU countries took on extra foreign leverage to expose themselves to the peripherals. The result has been asset-price bubbles and collapses in some of the peripheral countries, area-wide banking crisis, and sovereign debt problems."

The causes explained, the paper maps out "the geography of international debt flows using multiple data sources and provide evidence that after the euro’s introduction, Core EMU countries increased their borrowing from outside of EMU and their lending to the EMU periphery."

So braodly-speaking, core euro area economies funded excesses. Hence, in any post-crisis rescue, they were the beneficiaries of transfers from the 'peripheral' economies and taxpayers.

Some details.

According to Hale and Obstfeld, "one mechanism generating the big current account deficits of the European periphery could be summarized as follows: after EMU (and even in the immediately preceding years), compression of bond spreads in the euro area periphery encouraged excessive borrowing by these countries, domestic lending booms, and asset price inflation. We further argue that a substantial portion of the financial capital flowing into the European periphery was intermediated by the countries in the center (core) of the euro area, inflating both sides of the balance sheet of the large financial institutions in the euro area core."

So, intuitively, lenders/funders of the asset bubbles should be bearing some liability. And it would have been the case were the funds transmitted via equity or direct asset purchases (investment from the Core to the 'periphery' in form of buying shares or actual real estate assets). Alas they were not. "These gross positions largely took the form of debt instruments, often issued and held by banks. Thus, EMU contributed not only to the big net deficits of the peripheral countries, but to inflated gross foreign debt liability and asset positions for nonperipheral countries such as Belgium, France, Germany, and the Netherlands – countries that all experienced systemic banking crises after 2007."

Debt, as we know it now, has precedence over equity when it comes to taking a hit in a crisis, and debt is treated on par with deposits. Hence, "the tendency for systemically important banks to increase leverage in line with balance sheet size …implied a substantial increase in financial fragility for these countries’ financial sectors."

In the short run, prior to the crisis, leveraging up from the Core into the 'periphery' had a stimulative effect on asset bubbles. "Four main factors contributed to the suppression of bond yields in the European periphery after the introduction of the euro.
- First, the risk of investing in the European periphery declined with the advent of the euro due to investor assumptions (perhaps erroneous) about future political risks, including the possibility of official bailouts.
- Second, transaction costs declined and currency risk disappeared for euro area investors investing in the periphery countries.
- Third, the ECB’s policy of applying an identical collateral haircut to all euro area sovereigns, notwithstanding their varied credit ratings, encouraged additional demand for periphery sovereign debt by euro area financial institutions, which, moreover, were able to apply zero risk weights to
these assets for computing regulatory capital. The EU’s recent fourth Capital Requirements
Directive continues to allow zero risk weights for euro area sovereign debts, even though the borrowing countries cannot print currency to pay their debts.
- Fourth, financial regulations in the EU were harmonized and the euro infrastructure implied a more efficient payment system though its TARGET settlement mechanism."

Crucially, all four factors combined to reinforce each other giving "…core euro area financial institutions a perceived comparative advantage in terms of lending to the periphery, and this would also likely have affected financial flows from outside to both regions of the euro area.

In line with the above, the authors find:
- "...strong evidence of the increase in the financial flows, both through debt markets and
through bank lending, from core EMU countries to the EMU periphery."
- "… that financial flows from financial centers to core EMU countries increased, but predominantly due to increased bank lending and not portfolio debt flows.
- "In addition, …evidence from the syndicated loan market that is broadly consistent with the core EMU lenders having a comparative advantage in lending to the GIIPS."

Net conclusion: "The concentration of peripheral risks on core EMU lenders’ balance sheets helped to set the stage for the diabolical loop between banks and sovereigns that has been at the heart of the euro crisis."

Authors quote other sources on similar: “German banks could get money at the lower rates in the euro zone and invest it for a decade in higher yielding assets: for much of the 2000s, those were not only American toxic assets but the sovereign bonds of Greece, Ireland, Portugal, Spain, and Italy. For ten years this German version of the carry trade brought substantial profits to the German banks — on the order of hundreds of billions of euros ... The German advantage, relative to all other countries in terms of cost of funding, has developed into an exorbitant privilege. French banks exploited a similar advantage, given their major role as financial intermediaries between AAA-rated countries and higher yielding debtors in the euro area.” (From Carlo Bastasin, Saving Europe: How National Politics Nearly Destroyed the Euro, Washington, D.C.: Brookings, 2012, page 10.)

Charts below summarise flows from Core markets to 'peripheral' markets

CPIS is stock of portfolio debt claims from CPIS data in real USD:

BISC is stock of total international bank claims from consolidated BIS data in real USD:

BISL Flow is valuation-adjusted flows of total cross-border bank claims from locational BIS data in real USD:

And conclusions: "Not only did peripheral countries borrow more after EMU; in addition, financial institutions in the core of the euro area expanded their balance sheets to facilitate peripheral deficits, thereby increasing their own fragility. This pattern set the stage for the diabolical feedback loop between banks and sovereigns that has been such a powerful driver of the euro area's recent crisis."

So next time someone says that 'periphery' is to be blamed for the causes of the crisis, send them here. for in finance, like in dating, it takes two to tango…

30/8/2013: Hypo-Depfa Saga

One of the best articles on Hypo-Depfa fiasco I've seen to-date anywhere!

http://www.irishtimes.com/business/sectors/financial-services/near-collapse-of-german-bank-and-its-irish-subsidiary-shrouded-in-mystery-1.1509849r via @DerekinBerlin and @IrishTimes

Key quotes (for me, personally):

"Sitting in his Frankfurt office, in the shadow of Deutsche Bank’s twin towers, the 75 year-old Bavarian says the mainstream view in German finance circles – that Depfa sank HRE – is not strictly correct. “HRE would have gone down on its own, because of its own business,” he says, a view he formed during a hectic year studying HRE’s books."

It was clear from the beginning of the Hypo collapse that aside from Depfa, German lender was all over the shop in terms of loans it was issuing, its own funding was no different from the market consensus model, which relied on a toxic mix of medium term and short term funding sources exacerbating maturity mismatch risk with liquidity risk.

"Six months before the end, in early 2008, German financial regulator BaFin asked experts at the Bundesbank to conduct a full audit of all HRE operations. Its final audit report warns about “serious deficits” right across the group’s structures – from Dublin to Munich – particularly in the division supposed to assess risk of investments worth around €400 billion. ...employees were often unable to answer questions about the bank’s activities. ...HRE executives had no idea what was going on at their bank – either in Germany, or in their Dublin subsidiary, Depfa – because they had “no adequate, timely presentation of the actual financial situation”."

You have to just love the incompetence of the German financial authorities. Having received a report listing 49 breaches of regulations by Hypo-Depfa, in July 2008 (amidst already raging liquidity crisis worldwide), German BaFin "demanded quarterly progress reports on remedial action. Six weeks later, HRE and Depfa went over the edge."

A sense of BaFin being run by Dublin's FinReg or their equivalent is ever present.

17/6/2013: Deutsche, AIB and Cypriot Banks: 3 links

Back in 2011, I wrote about the extreme leverage ratios in some of Europe's top banks: http://trueeconomics.blogspot.ie/2011/09/13092011-german-and-french-banks.html. Deutsche Bank was at the top of the list. Now, 19 moths later it seems others are catching up: http://www.reuters.com/article/2013/06/14/financial-regulation-deutsche-idUSL2N0EO1D220130614.

And while on topic of banks, let's check this one for the record: http://www.independent.ie/business/irish/aib-will-not-repay-35bn-cash-it-owes-to-the-state-29337833.html. I wrote about this in Sunday Times last weekend, in passim, but this is more comprehensive article.

Another link of worth on the topic of banks is Cyprus banks fiasco history from ZeroHedge: http://www.zerohedge.com/news/2013-06-17/guest-post-real-story-cyprus-debt-crisis-part-1

29/10/2012: German banks exposure to GIIPS

Here's a pic courtesy of Morgan Stanley Research of the German system exposure to GIIPS - pre-crisis, and now:

That is a massive deleveraging (if you look at now, through September), but it is also a massive exposure. Few years ago I said that German banks didn't have a property bubble at home. The had one abroad (in GIIPS) instead. Now, with their banks close to 2005 levels of exposure, it is pretty clear where the Euro area policy train is heading before the station 'Legacy Debts' is reached.

9/2/2012: Interesting chart on Euro area deposits

Here's an interesting chart from Credit Suisse via zerohedge:

Now, it's not the blue line that worries me and the others (EAP5=Euro Area Periphery states or PIIGS). It's the massive dip in the grey line. Given there's little deleveraging of consumers and corporates in France and Germany and that there is little it terms of concerns for stability of German banks (whether or not this sense of security is justified or not), the chart suggests that deposits are flying out not just in fear of local banks risks, but in fear of the euro risks.

The link to zerohedge post is here.

13/09/2011: German and French Banks - "extreme" leverage & elevated risks

So what's the real trouble with German (and French) banks, folks? Errrm... they are kinda seriously bordering the "insolvency" territory.

The sources for this information, in addition to those cited below, include an excellent research note prepared by Peter Mathews (FG), TD for the Dail Finance Committee, the IMF GFSR and IMF WEO databases.

Deutsche Bank

Leveraged 52:1 (16 August 2011) based on a Tangible Common Equity (TCE) to Total Asset measurement.

Tangible Common Equity is a better gauge of solvency than Tier 1 capital, particularly in the midst of a liquidity crisis. Tier 1 gives no sense of a bank's ability to withstand a liquidity crunch as it includes market-sensitive instruments that are subject to liquidity and price declines risks. Tangible Common Equity is also a much better indicator of a bank's ability to raise further funds in the market as it inversely relates to the rate of assets dilution implied in any rights issuance. (1), (2)

As TCE of €36.2bn is written against €1.85 trillion of assets, DB has just 1.96% cover in form of TCE against assets it holds - a writedown of just 2% on the asset values (cross the book) will wipe out the DB TCE cushion, rendering its current equity-holders de facto bust. Even excluding derivatives, MorningStar estimated DB leverage (TCE ratio 2.1%) at 47.6:1.

DE's current leverage levels compare unfavorably against 44:1 TCE leveraging on Lehman Bros books at the time of collapse (ordinary leverage ratio in Lehman's prior to collapse was 31:1) and makes DB the second most-leveraged bank in the euro area after Credit Agricole.

To bring DB closer to sustainable levels of TCE delveraging - 8-10% reading (note this is different from Tier 1 capital) will require it raising €110-150 billion in equity (depending on specifics of risk weighting ratios) or 3-4 times the current valuation of TCE or 3-4 times the current market value of the DB. Implied dilution for current equity holders under such scenario bears the risk of 75%- 80% loss on equity.

Note that 8-10% ratios are rather conservative, considering that in 2006-2009 TCEs for countries with banking sector crisis averaged (across top100 banks) TCEs of 11.5% to 15.3%. (3) Raising TCEs to the crisis-average levels of ca 13.4% will require equity raising of ca 5.7 times current market valuations or implied dilution of current equity by 85%.

To match TCE/Total Asset leverage ratio of the most leveraged US bank, JP Morgan chase (5.58%), DB would require €67 billion of additional equity or equity raising to the tune of 1.8 times current market cap.

DE's current market capitalisation of €37 billion as of 2 September matches relatively closely tangible common equity of €36.2 billion. In previous weeks, DE market cap fell as low as €26 billion or 70% of TCE. A market capitalisation at or below TCE is a warning sign that the bank is in trouble and questions surround its solvency and stability.

Worse than that, per research from Espirito Santo, DB liquid assets as % of the short term (<1 year) funding in 2010 stood at 47%, well below global leaders Credit Suisse (82%), UBS (77%) and Barclays (59%). At the same time 2010 wholesale funding maturity requirement was 49% - in excess of the iquid assets cover. Again, Credit Suisse had 33% funding call against 82% cover.

DB is structurally important to Germany as its assets stand at around €1.85 trillion, close to 75% of Germany's 2010 GDP (€ 2.498 trillion).

DB exposure to Greek assets is €1.6 billion for the core Group components (sovereign debt only), of which €1.34 billion in Deutsche Postbank AG exposures. Under 70% haircut scenario across the entire DB Group, the total implied loss will be around 5% of TCE.

Commerzbank

Current leverage around 35:1 in TCE terms, which is elevated compared to both historical averages and 2008-2009 crisis levels for comparable institutions. Given current assets valuations at ca €700 billion, the implied TCE is 2.92%, which means that a writedown of 3% of the assets will result in a complete wipe-out of TCE.

The core problem with 35:1 TCE leveraging in the current environment of globally impaired liquidity is that any deleveraging of the balance sheet will require substantial equity rising, similar to that in DE as discussed above. Adjusting for derivatives held, TCE ratio in Commerzbank runs at 30:1 - according to MorningStar who called this leverage ratio as consistent with "extreme" risk rating.

Together with DB, Commerzbank account for ca 102% of German GDP. As German debt to GDP ratio currently stands at 83% and heading for 87% , German taxpayers can see significant adverse impact of the DB and Commerzbank recapitalizations should the calls on PIIGS come in.

Commerzbank is the most exposed of all German banks when it comes to Greek sovereign debt, with nominal exposure at of the end of Q2 2011 of €2.9 billion. Applying the market-expected mid-point writedown in the case of default of 70%, bank losses on the Greek sovereign bonds will wipe out around 19% of the bank equity.

Recent data shows deep concentrations of Greek risks exposures in German banking sector, with German commercial banks holding ca €19.3 billion in public sector debt from Greece, €2.9 billion worth of banks debt from Greek banks and €7.4 billion of corporate and private debt, to the total of €29.5 billion (per BIS data). According to Fitch research, only €13.1 billion of that is on the banks balancesheets, the rest tucked away off the books.


French Banks

Credit Agricole is leveraged 70:1 (assets €1.5 trillion), while BNP Paribas is leveraged 36:1 (assets €1.93 trillion, Common Equity Tier 1 ratio of 9.6% well below minimum standard set for SIFIs of 10%). Bank's assets to market value currently stands at 64:1. BNP's exposure to Greek debt is now at ca €4 billion. SocGen is leveraged 34:1 (assets €1.16 trillion) on TCE basis and 28:1 on ordinary basis (again, recall Lehman's numbers at the point of collapse were 44:1 and 31:1) the bank has huge short term funding requirements presently being exposed by the flight out of Europe by US money market funds and Asian investors. SocGen exposure to PIIGS debt is €4.3 billion, referring to banking book only. SocGen is also in trouble on the liquid assets side with 26% ratio of liquid assets to short-term wholesale funding calls in 2010. Worse, wholesale funding that matured within 1 year of 2010 as percent of total wholesale funding was 69% for SocGen. Three French pillar banks have assets coming in at well over 200% of French GDP.

The banks are aggressively moving out of the PIIGS with SocGen in recent note stated that it cut its exposures to the peripheral states by 23% since early June 2011, taking out $1.5 billion and $2 billion in assets from Greek and Italian books. French banks total exposure to Greece is estimated at around $89 billion.

On top of this, French banks are now becoming effectively shut out of the dollar funding markets (4). And liquidity woes do not stop there. US Prime money funds have cut their holdings in certificates of deposits from French banks by about 40% in the three months through August 11. The proportion of the remaining holdings of French banks short term funding notes maturing in less than a month increased to 56% on August 11 from 17% on June 11. (5) The banks, of course, deny this is happening.

Let's run though some grim figures for one of the French "dogs" - BNP: as of June 30, 2011, the bank had €109bn worth of sovereign bonds on its books, amounting to 190% of the bank TCE (that's JUST Government bonds!), of which €31bn (or 54% of TCE) was in PIIGS bonds split as follows: Portugal - €1.7bn, Ireland - €400mln, Italy - €23bn, Greece - €3.8bn, and Spain €2.5bn. 95% of these exposures were held on banking book. So, now, let's do the same exercise as above - apply 50% haircut to Greek bonds, 25% haircuts to Porto bonds, 15% to Spanish, Irish and Italian bonds - all below market rates of implied haircuts, but let's indulge them with this assumption. This adds up to a writedown of €6.21bn or a wipe out of 11% of TCE. Bank becomes insolvent.


Summary

To summarise the above, two of German core banking institutions are currently operating in the extremely risky environment with leverage levels that can be classified as "extreme". The French banking system is even more sick than the German one with leverage ratios close to those attained by Lehman and PIIGS exposures that are well in excess of "manageable", given already strained capital cushions.

In common parlance, if it barks & wags the tail, it's a dog... regardless of what the official stress tests and powerpoint slides say.


Notes

(1) See BASEL III: Long-term impact on economic performance and fluctuations, by P. Angelini, L. Clerc, V. Cúrdia, L. Gambacorta, A. Gerali, A. Locarno, R. Motto, W. Roeger, S. Van den Heuvel, J. Vlc_ek, BIS Working Paper 338, February 2011, http://ssrn.com/abstract=1858724

(2) Note that Tier 1 capital is classified into different Tiers of capital, based broadly on the maturity profile of the capital invested. The most stable capital is Tier 1 capital and consists of items such as paid-up ordinary shares, non-cumulative and non- redeemable preference shares, non-repayable share premiums, disclosed reserves and retained earnings.

(3) Bank Behavior in Response to Basel III: A Cross-Country Analysis, by Thomas F. Cosimano and Dalia S. Hakura1, May 2011, WP/11/119, IMF Working Paper, Table 3.

(4) http://www.forexcrunch.com/bnp-paribas-executive-admits-access-denied-for-dollars/

(5) http://www.businessweek.com/news/2011-09-13/bnp-paribas-socgen-rebound-after-rejecting-funding-concerns.html