Lagarde is right: European banks do need more capital

Well, European politicians who thought that by appointing Christine Lagarde as the head of the IMF European and the Eurozone would continue to have an easy ride received a ride shock this weekend. Not only did the reformed Lagarde say that European banks needed more capital, she also suggested that the EFSF or other pan European entity could be source of the funds (to gasps of horror in Berlin).   

European officials have been quick to dismiss these suggestions.

The Financial Times for the example reports this morning that Amadeu Altafaj-Tardio, spokesman for EU economic chief Olli Rehn, said the recently-completed stress tests for the banking sector showed that only a select group of banks were in need of capital injections and that national authorities were already dealing with those cases.

But the market prices of European banks have been market down, in some cases very sharply in recent week, signally skepticism over the value of the stress tests.  So let's look at some of the issues.

In the first case, we must remember that the European stress tests are designed to test whether the European banks have enough equity capital to protect the interests of the creditors. The stress tests are focused on the interests of creditors and have nothing to say about equity holders. If the risk of substantial loss to equity holders is high then obviously they should pay less for the equity even though the stress tests might suggests that creditors are likely to be safe.

In the second case, the European stress tests ignored the key issue facing asset values in Europe - which is the value of sovereign debt. These are assumed to have no default risk in the stress tests  - whereas the markets consider that the likelihood of a default by Greece, Portugal and Ireland is high. In recent weeks even Italian bonds have begun to discount some default risk. The stress tests are based on a ridiculous assumption.

The markets can look and should look at the quality of bank balance sheets in a different way: the amount of leverage should be measured and compared after all assets have been stated at their fair values (including sovereign debt which is trading below par) and all non-tangible assets should be written off. This includes goodwill arising on consolidation and deferred tax assets. Furthermore, all off balance sheets assets and liabilities should be brought on balance sheet and the adjustment made for differences between IFRS and US GAAP on the netting off of derivatives, setting off and off balance sheet items.

Only when this is done can we compare like with like. For the European banks we see the following values for the amount of equity held as a percentage of assets based on the latest available financial data, adjusted as described above:

Name	Equity Ratio		Tier 1
Dexia	-0.22%		13.4%
Deutsche Bank	1.12%		14.0%
UBS	1.18%		18.1%
CS	1.42%		18.2%
Santander	1.72%		10.4%
Credit Agricole	1.86%		10.5%
RBS	1.94%		13.5%
Barclays	2.13%		13.5%
Commerzbank	2.32%		11.6%
Unicredito	2.46%		9.9%
Lloyds	2.58%		11.6%
ING	2.69%		12.0%
HSBC	2.99%		12.2%
Natixis	3.32%		11.6%
Popolare	3.51%		7.9%
Soc Gen	3.57%		11.3%
MPS	3.74%		9.1%
BNP	3.82%		11.9%
BBVA	4.02%		9.8%
Raifessein	4.23%		9.7%
Intesa	4.42%		11.8%
Erste Bank	4.58%		10.5%
KBC	5.24%		12.6%
Standard Chartered	5.94%		13.9%
Sabadel	6.89%		9.8%
BPES	8.07%		9.8%

Note the wide disparity between the Tier 1 ratio and the equity ratio we have here. Note also that the Eurozone is by no means the worst in Europe: the British and Swiss figure prominently in the top half. But with the ability to issue their own currencies, the Swiss and British authorities could recapitalise the banks by printing money. Not so the Eurozone countries. Recapitalizing Eurozone banks is likely to be far more difficult. That's the problem.

 US banks are not significantly better capitalized on this basis: Bank of America has an equity ratio of 2.56% (the weakest among the major US banks) and at 4.94% JP Morgan is the strongest. Smaller US banks are all much stronger than these behemoths - CIT is a standout with an equity ratio of 21.6%.

So let's not pretend that the banks are rock sold.

UPDATE 31st August 2011

The Financial Times reports this evening that a fierce debate has broken out between the IMF and the EZ authorities over how to measure bank capitalisation. The EZ is objecting to the IMF's use of market based measures of asset value, such as CDS - as if the EZ knows better. The full article is here
(you've got to get through the FT's ridiculous firewall I am afraid)

Net interest margin is a useless concept

There's a lot of talk about banks deriving lower funding costs from depositors than from the wholesale markets. But this perception, which is widespread among bank management teams rests on a misunderstanding.

Systemically important banks benefit from sovereign guarantees to their creditors. For the most part these guarantees are implicit. Customers depositors on the other hand benefit for the most part from formal guarantees through agencies such as the FDIC.

So what the cost of a asset guaranteed by the Government? Well, for formally guaranteed debt like customer deposits. it is the return on the equivalent maturity government debt and, for informally guaranteed debt like wholesale debt, there may also be an additional spread for the uncertainty associated with the likelihood that the Government may not honor the guarantee (think Lehman, Northern Rock, Bradford and Bingley). A functioning market prices this risk daily - the LIBOR market.

So a customer deposit that is formally guaranteed by the Government has the same cost to a bank as its earns on overnight deposits with the central bank. And every customer with a deposit in a bank can access close to the rates earned by banks when they make deposits at the central bank  - if they choose to do so.

A quick survey of rates actually paid by banks throughout the world however suggests that in most places and most times, customers actually earn less that OIS rates and they seem to be content with this.

Why? Quite simply because a customer deposit with a bank gives a customer access to the payment system and providing this service has a cost to banks for which they can charge depositors a fee. That fee is usually in the form of a lower interest rate than OIS but it could equally be an explicit fee or in a few cases a charge based on deposit values (i.e. negative interest rates).

So when the market looks at net interest margins of banks, what is it doing? It is netting off three different things - gross interest receipts, gross interest expenses and fees charged by banks against interest rates for access to the payments system. but if different banks charge for the payments system in different ways (as they do) how useful is the concept of net interest margin in bank analysis?

Frankly not much.

If Greece leaves the Euro

I have read so much stuff from well established economists who argue that the best solution for Greece would be to default, leave the Euro and devalue.  Some of those arguments can be found here as well as in the writings of Lombard Street Economics.

But much of this is pie in the sky stuff, which underestimates the impact of such a step for Greece. (Things are different for the rest of the Eurozone as I will discuss in another post).

The biggest holders of Greek Government bonds ("GGB"s) are of course the Greek banks, so any serious probability of a default combined with a withdrawal from the Euro would almost certainly a trigger a massive run of the Greek banks, with depositors seeking to transfer their savings to other Eurozone banks.

The combined impact of a run on the banks together with a sovereign default would almost bankrupt the banks. This would lead to a serious credit squeeze in the economy as one sector after another would collapse, raising unemployment and reducing output.

This is of course very different from the UK in 1992, the memory of which is seared into the collective conscious of London based economists. There withdrawal from the ERM was followed by a reducing in interest rates, credit growth and economic expansion.

Despite low Eurozone interest rates, Greece faces very high interest rates, with 2 year spreads over Bunds of over 40% pa at the time of writing, so in principle if withdrawing from the Euro would allow Greek interest rates to fall, this would be a positive development.

But how likely is that?

Greece's basic problem is that it has weak political institutions. Tax evasion is rife, corruption is said to be widespread and confidence among Greek's in their own Government is weak.  In these circumstances, the willingness of Greeks to hold drachmae would be low and this would almost led to a collapse in the currency, with a high risk of hyperinflation. You can see how any attempt to recapitalize the banks with dramchae might led to depositors withdrawing their deposits and exchanging them for Euros or dollars. To contain these risks, it is easy to see that interest rates may actually rise rather than fall from current levels. That is likely to squeeze the economy even further.

Even if Greece were able to stabilize itself in the short term (with EU or IMF support post Euro withdrawal), in the medium term it is hard to see how any savings would be retained in drachma, so business investment would either have to come through foreign currency borrowings in the form of Euros or dollars and that would make any further devaluations very costly for businesses. And no country in this position can run a current account deficit - which imply capital imports.

Much of this has been rehearsed before, for example in Mexico in 1994 or Indonesia in 1997. In Mexico part of the solution was to allow Mexican banks to offer dollar deposit accounts to local depositors and in both these cases as well as in Thailand, Russia, Argentina in the years that followed capital controls were imposed to limit capital inflows. But if Greece were to do this it would have to leave not only the Euro but also the EU - that is assuming it has the institutions to enforce capital controls.

The Greeks may be facing a bleak few years but its probably nothing compared with what they would face outside the Euro.
 

Banks' analysts please wake up

What is it with banks' sellside analysts and the focus on the Tier 1 ratio or Core Tier 1 ratio as a measure of capital adequacy? I read this stuff and despair..

Take the example of say Beta Bank and Gamma Bank. Both have $1 of equity. Beta borrows $2 of debt and invests $1 in the UST market and lends $1 to Monopoly Electric Co. Gamma borrows $9,999 and invests $9,999 in the UST market and lends $1 to Monopoly Electric Co. The UST assets of both balance sheets, according to the Basel metrics, have a zero weight, while let's assume that the loans to Monpoly have 100% weighting. In other words, both banks have the same Core Tier 1 ratio.

But are they leveraged in the same way from the point of view of stockholders? Obviously not. Both banks earn OIS on their UST assets but pay LIBOR on their borrowings and - given LIBOR spreads over OIS - it is obvious that the earnings of Gamma are likely to be significantly more volatile than those of Beta. Even if Gamma has a higher ROE than Beta, how can it command a premium rating over Beta if they both earn the same returns on loans for Monopoly Electric Co? 

It shouldn't. So come on guys, take a closer look at the drivers of the volatility .