Citi joins the European stress-test jamboree

Posted by John McDermotton Jun 23 20:32.

Roll(over) up, roll(over) up, for the latest unofficial European bank stress test.

For previous editions see here, here, and here.

This week’s contestant: Citi.

And it has given us two for the price of one: a bottom-up test and a top-down test.

There are methodological differences, but both stress tests look at core European banks and make assumptions in three areas: (1) sovereign losses, (2) asset quality, and (3) funding costs. The results are pretty similar regardless of how Citi cuts the cake.

And their conclusions are fairly clear, and Spanish. Core banks will be okay so long as contagion doesn’t reach Spain. The only big banks that need worry are in Spain, though they don’t need that much capital, so long as (you’ve guessed it) contagion doesn’t reach the motherland.

In Citi’s words:

€70bn ‘Core Coverage’ Losses – We estimate total losses for the European Core banks under Citi coverage to be €69bn in our stress scenario. Of these, €17bn would be from GR / IR / PT and €52bn due to Spain. For the European Core banking systems, total losses would be €98bn with €24bn due to the three smaller peripherals and €74bn due to Spain (Figure 2). Losses arise from direct sovereign debt exposures, customer loans exposures and higher funding costs. Greece is not the key issue for the core banking systems. It is contagion to Spain that is of most concern.

No Capital Needs for ‘Core’ Banks – Under our bottom-up stress test, no bank from the “core countries” needs capital. For the sample of 26 banks, the 2013E Common Equity Tier 1 ratio would decline from an estimated 11% to 9.7%. While no core country bank drops below 7%, three Spanish banks would need to raise a combined €2.1bn


A bottom-up stress-test

For sovereign losses Citi assumes bond price reductions of 40 per cent for Greece, 61 per cent for Ireland, 67 per cent for Portugal and 82 per cent for Spain. Timely, then, for it looks like the EBA may now after all be pressing for haircut assumptions in its next round of stress tests.

These marks are then applied to the entire sovereign books of the following 26 banks.

And unlike the EBA, Citi says it is stressing both the trading and the banking books.

Our methodology is different to EBA’s. While EBA is likely to only stress the trading sovereign book (fair value through P&L, AFS) of the banks, we also stress the banking sovereign book (AFS, HTM).

As for asset quality, Citi assumes the following loan losses over 2011-13:

To stress for asset quality, we apply the assumptions in Figure 13 below on a bank-by-bank basis. We apply a 100% increase to the currently estimated cumulated cost of risk over 2011-13 for the smaller peripheral banks; 50% for the international peripheral banks; and 25% for everyone else…. We believe this is a severe enough assumption, as asset quality has started to improve in most European countries.

And for funding stress:

Here are the results:

For the 26 banks whose potential losses we calculate explicitly, the core equity Tier 1 ratio (Basel 2) drops from an estimated 11% in 2013 to a stressed level of 9.7% (Figure 9). Calibrating to a 7% minimum, the capital shortfall is €2bn.

No banks from Core countries are affected. Three Spanish banks would need to raise capital: BBVA (€1bn), Popular (€0.8bn) and Sabadell (€0.3bn). For the local Spanish, the capital shortfall is mostly due to higher stressed loan losses. For the international Spanish banks, the driver is Spanish sovereign exposure – especially at BBVA which has higher exposure than Santander.

Under our stress test, the 26 banks experience after-tax losses of €112bn (Figure 10) – with c.15% deriving from peripheral sovereign exposures, c.50% from deteriorating asset quality (in the total book, including non-peripheral loans) and c.35% from higher funding costs

So the core gets off without any additional capital needs, according to Citi. Interestingly, unlike the Nomura tests we wrote up on Tuesday, there’s little mention of how this would change should losses come in immediately, rather than by 2013.

As for those Spanish banks, although Citi reckons its loan loss assumptions are stringent, Santander (to take one example) needs no extra capital under these conditions — in contrast to other tests under harsher assumptions we’ve looked at here on FT Alphaville. It’s amazing what happens when you make a few changes to those real estate numbers.


A top-down stress-test

For sovereign losses Citi uses the same assumptions on bond losses as above, this time giving an interesting bit of extra detail on timing of those losses:

In the top-down version we get a clearer picture of what Citi thinks are severe NPL assumptions — about a doubling of NPLs in the periphery. For those keeping score, Nomura assumed a tripling of NPLs from current levels so Citi’s version could be considered conservative, especially since there is debate as to correct current NPL levels.

For funding costs Citi uses the spikes seen after the Lehman and (2010) Greek crises. There’s a whole barrel load of mini-assumptions in this section but the two most important are probably (1) average funding costs rising from 2.03 per cent to a “stressed” level of 3.36 per cent (+133 bps); and (2) a 25 per cent reduction in MFI deposits, interbank lending and term funding “in a sovereign crisis worst-case scenario”.


Putting this all together we’re left with these aggregate results:

As always, these tests are only as good as their assumptions and their transparency. The EBA, which on Thursday finally said it was going to be tough on banks’ Greek holdings liabilities, now has a few models to go on.

Over to you, stressies.

Related links:
Credit Suisse stress-tests the French banks – FT Alphaville
An €11bn stress test scenario for Santander, BBVA – FT Alphaville
Some stresstestimates – FT Alphaville

This entry was posted by John McDermott on Thursday, June 23rd, 2011 at 20:32 and is filed under Capital markets. Tagged with , , , , , .

Post a comment or leave a trackback: Trackback URL.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: