The EBA Stress Tests: What’s the News Value?

The Irish banks, AIB and Bank of Ireland, show up poorly on the stress test of 51 European banks (33 in the Eurozone) released Friday night. The methodology is explained on the EBA website. Briefly, there has not been a review of each bank by a team of EBA inspectors as is implied by some of the media coverage – RTE’s bulletin referred to an ‘examination’. It is a mechanical exercise based on the ‘static’ 2015 balance sheet, as published, with no adjustment for the plausibility of provisions but also with no credit for retained earnings post 2015. The ‘stress’ is essentially a GDP downturn from 2016 through 2018 resulting in a depletion of capital adequacy as against the end-2015 balance sheet number.

The scale of the depletion reflects the extent of the assumed downturn. The essential reason for the sharper loss of capital adequacy for the Irish banks is that the downturn assumed for Ireland is greater. Against a baseline, the cumulative adverse GDP shock for the main Eurozone countries included is as follows:

Belgium -7.6
Germany -6.6
Ireland -10.4
Spain -6.7
France -5.6
Italy -5.9
Lux -8.2
Neth -8.4
Austria -7.6
Finland -8.3

The adverse shock assumed for Ireland is the largest and 3.2% above the average for the others shown. There are some other factors but the EBA release makes it clear that these numbers are the main driver of the projected capital depletion. The basis for the large Irish shock is a calibration against the experience over 2008 to 2011 when the downturn in Ireland was more severe than elsewhere.

The EBA may have sacrificed plausibility to uniformity of treatment – the exercise is in any event an input into a further phase called SREP, the supervisory review and evaluation process, rather than a definitive assessment of bank capital adequacy. The Irish banks, and numerous others, may of course need to generate or raise more capital but the relative worsening in their position flows from the assumptions employed and not from any ‘news’ uncovered by the EBA sleuths.

Comments

comments

31 thoughts on “The EBA Stress Tests: What’s the News Value?”

  1. While the process may be slightly flawed hopefully it will draw renewed attention to the massive NPL balances the Irish Banks carry whereby something will eventually be done about them.

  2. The assumptions may be related to each country’s exposure to senior hurling Ie pure neoliberalism. Ireland has a less resilient economy than Belgium, for example cos it has a weak SME sector and a penchant for property bubbles. So blame it on the assumptions but look at the bigger picture.
    The government has decided to prop up property prices for the benefit of the twins but no asset has an intrinsic value as they know at the EBA.

  3. Excellent analysis and spot on, and a bit too kind as to reasons why Ireland was brought down a few pegs more than the rest.
    All the more surprising when you consider that currently, based on its published balance sheet at June 2016, AIB has an real capital ratio of 12.7% whereas Duetsche bank has a real rapital ratio of 3.7%.
    Put simply, ‘Ireland’ was getting too big for its boots, again. There is of course more than a little truth in that. Still, the EU thought process, that the pixie heads need to kept in their place, is well entrenched in EU mandarin circles.
    It also helpfully keeps the foces on those pesky peripherals and southerners.

  4. There’s a whiff of ‘original sin’ off this approach.

    Ireland’s integrated economy means per se greater effect of any global shock.

    Ergo Irish banks must hold more capital essentially forever to guard against this.

    1. Ruadhán MacCormaic, in a superb article about Brexit back in January , noted that Irish Governments have been of a very neoliberal bent during the recent past.

      http://www.irishtimes.com/news/ireland/irish-news/brexit-a-referendum-on-ireland-s-future-1.2476474

      American military stopovers in Shannon, not building social housing despite over 100,000 on the waiting list (because public spending on infrastructure is WRONG) , giving that concession off Mayo to Shell, not investing in rail (but paying how much in climate change induced flood compo?) , paying off bondholders and loading the sovereign balance sheet with the cost (because public spending on mispriced financial insanity is a no brainer) , Public Private Partnership (because public spending on infrastructure is WRONG but public spending on mispriced private traffic projections is a no brainer ) , giving bookies tax breaks, selling NAMA property to vulture funds, killing the National Pensions Reserve, 12.5% corporation tax, the double Irish, other blatant tax scams like Apple paying whatever token amount it is , the IFSC and light touch regulation (which blew up the native financial industry, fantastic job that. ) are all rather neoliberal, not to mention stupid, in addition to being ultimately self defeating.

      Insightful analysis is mostly only to be heard from people RTE and most of the media classify as oddballs, primarily serving the mostly unheard people who have been at the sharp end of a lot of this tomfoolery endorsed by the soberest suits. Joe Higgins was on the ball about the Lisbon treaty and look at what happened to the ultra middle of the road and “respectable” comme il faut Fianna Fail when the bailout sheriffs rolled up.

      Things have moved on since the bailout. The ECB is unable even to generate its own inflation target of 2%. Central Banking has been reduced to optics. “Whatever it takes” except the bleeding obvious. You couldn’t make it up.

      Whining about unfair treatment is deluded. Ireland runs an extremely risky macro model.

      1. I would be every bit as critical of Ireland’s model as your good self, but these EBA numbers simply don’t stack up, unless Ireland is to have another horrendous crash that Europe is largely insulated from.
        [Europe could, of course, reinstall Trichet and send it bills to Ireland, like the last time, which is great method of insulating oneself]

        For the adverse scenario, AIB would have to lose approx €5 billion between now and the end of 2018, in addition to ‘losing’ approx €3 billion of its deferred tax asset (as per new rules on deferred tax assets).
        In order for this to happen, AIB’s reserve for bad loans would have to move from 9% (June 2016) to about 18%!
        Could it happen?
        Yes, it is possible, but is it possible in a scenario where other EU countries are not nearly as badly affected?

        In any case, to my mind the news is somewhat helpful, in that it will postpone hopefully for a lengthy period, AIB being handed over to some vulture fund, masquerading as a charity, to make a tax free killing.

        1. As discussed on twitter it’s a model not a forecast. We’re more open so we get hit more (and rebound faster) in a crisis. Our banks have lots of NPL so these mean we are bottom of the heap.
          As we can’t solve the oneness issue …..

        2. JR

          The risk of the mother of all meltdowns is elevated. The Fed cannot generate growth in the US. 1.2% for Q2 is atrocious. There is just too much debt.
          The dollar is up 20% over 2 years on expectations of interest rate rises (fed by Fed guidance) that are not going to happen because the US economy is not going to grow.

          http://www.marketwatch.com/investing/index/dxy/charts run it over 3 years

          Unwinding this is not going to be straightforward.

  5. It will be no surprise to anyone whos been watching thats for sure. The problem is that the banks need to
    a) get more capital (but income prospects are poor and the state wont inject, and an IPO seems…ballsy)
    b) get less NPL’s (but the state wont allow these to be written off or recouped in normal banking fashion).

    Concerns in 2007-8-9 on nationalised banks were on politicised lending. There was AFAIR less concern than should in retrospect have been on politicised NPL actioning.

    With a fractally minority government, expect little change.

    1. There may be change; but back to the wrong kind e.g.

      http://www.oireachtas.ie/documents/bills28/bills/2016/2416/b2416d.pdf

      Political parties are adept in seeking contradictory objectives, in this instance profitable banks and forbearance for borrowers. You can have one or the other but not both. The electorate seemingly remains unaware of this. Or, maybe, views depend on whether one is a borrower or a saver.

      As to the EBA, querying the basis for the calculations, however well justified, is to miss the point. The institution has been set up to make the assessments. The Irish Banks are not the centre of attention.

      http://www.cnbc.com/2016/07/29/european-banking-authority-releases-results-of-eu-bank-stress-tests.html

  6. Banking as an industry can’t cope with permanently low interest rates which may do more damage than one shock would. Until money starts to get to ordinary people there isn’t a hope of interest rates rising . The unintended consequences risk of ECB policy is off the scale. .

    Galbraith’s observation was that “”The conventional wisdom” gives way not so much to new ideas as to “the massive onslaught of circumstances with which it cannot contend”.

    Some coherent new ideas would be preferable before the whole thing blows up. It would make far more sense to have a once off NPL writedown in conjunction with a tax harvest from the 1% to pay for it. Without growth at least 20% of EU banking capacity is superfluous . AIB and BoI both comprise vital and useless functions . How to save the vital is the question

    And see how many German banks are on a shaky scraw

  7. Meanwhile the IMF independent evaluation office report released on Thursday and already considered done and dusted by the meeja despite numerous nuggets of info. Ireland covered by a single desk economist during 2007-09 etc.

    1. @ Frank Galton

      Would more desk economists have made a difference given that it’s a rare foreign economist who can separate the fantasy from the facts in the national accounts?

      Besides, it’s easy to forget that the denial of the local boosters did not abate overnight.

      In February 2009, exactly 2 years after the subprime crisis broke out in the US, the Irish unit of PricewaterhouseCoopers (PwC), the biggest of the Big 4 accounting firms, delivered a report on Anglo Irish Bank to the finance minister. It said on its proposed bad debt provisions:

      These annual impairment charges were €2.3bn and €3.0bn respectively per annum under the two scenarios for the years ended 30 September 2009 and 2010. The two PwC impairment loss scenarios exceeded Anglo’s worst case impairment loss scenario.

      Jones Lang LaSalle valued a sample of 160 properties held as security in relation to the top 20 land & development exposures on Anglo’s books.

      In 2010 Anglo Irish Bank reported a loss of €12.7bn for the 15 months to the end of December 2009 — the largest loss in Irish corporate history — after charging €15bn to cover bad debts.

  8. Lots of reasonable comments, but nothing to make me feel that EBA have managed to add to what we knew already. There is an issue: what precisely is the purpose of these stress tests?

    1. “what precisely is the purpose of these stress tests?”
      Among other things, to keep the EBA Stress Test Unit in business. First rule of bureaucracy Colm….
      One advantage to these is the rather remote, distant, homogenised view. Market perceptions of individual situations are by nature somewhat idiosyncratic, so these have a place. Also, see point one.

  9. Excellent post.

    The idea that Ireland’s GDP could fall by 10.4% is bonkers.

    The main reason it fell by this amount in 2007-2010 is that the number of new houses built fell from 85k to 8k. This knocked 7.5% off GDP. Excluding house-building, Ireland’s GDP fell by less than the EU average.

    In the very unlikely event of a new global recession of similar magnitude, could Ireland experience a similar fall in GDP?

    Very unlikely (about on a par with Leitrim winning a GAA football/hurling double). The number of new houses built in Ireland in 2016 is likely to be about 15k/16k. Even if house-building were to stop completely in the next couple of years (and its far more likely to rise), the effect on GDP would be about one-fifth of that in 2007-2010.

    So, I agree with ColmMcCarthy’s main point.

    However, look on the bright side. Those who react unthinkingly to such reports may now sell off shares in Irish banks. Those who have greater understanding of the statistics behind the report, a category that, thanks to Colm’s post, now includes all readers of this thread, can then pile in and make a profit. The markets invariably penalise stupidity and reward intelligence.

      1. Thanks for link.

        So, under the adverse scenario its not a 10.4% fall in GDP that the EBA is forecasting (which is what I read in an erroneous media article), but a 10.4% deviation from a high rate of growth that would otherwise occur, with the deviation spread over 3 years? That’s somewhat different.

        Just scanning quickly through the tables, it seems that under the baseline scenario Ireland is forecast to have the highest rate of growth in the EU between 2015 and 2018 (about twice that of UK) – and, under the adverse scenario only Ireland, Hungary, Malta and Poland are forecast to record positive growth between 2015 and 2018, with Ireland’s being again the highest.

        1. The banks, the banks! It’s about the banks.

          http://www.wsj.com/articles/most-european-banks-pass-stress-test-1469824257

          What the stress tests have come up with is entirely credible in terms of the relative position of European banks, including those in the UK. The majority policy approach is clearly based on booting the weakest out, insofar as the countries in the EA are concerned, in the event of a crisis, and into the agreed bank resolution mechanism. The pressure on governments to avoid this, Italy, mainly, in this instance, is intentional. But not by the state concerned simply bailing the bank concerned out. In sum, the management of national banking policy remains at the level of the individual states. It is hard to see how the position could be otherwise. Who, for example, is best informed on the bad loans of the Italian banks? Or on how they came about.

  10. Colm McCarthy said “The adverse shock assumed for Ireland is the largest and 3.2% above the average for the others shown. There are some other factors but the EBA release makes it clear that these numbers are the main driver of the projected capital depletion. The basis for the large Irish shock is a calibration against the experience over 2008 to 2011 when the downturn in Ireland was more severe than elsewhere.”

    Given that the the assumed shock for Ireland is the main driver of the poor performance, the issue appears to be whether whether that assumed shock is justified and a sensible basis for assessing risk.

    To my mind, It is a bit like saying that the risk of falling off the tight-rope is grossly affected by the assumed heigth of the drop. The question is not how the height affects the calculations but rather whether the assumption is justified.

    JTO said “So, under the adverse scenario its not a 10.4% fall in GDP that the EBA is forecasting (which is what I read in an erroneous media article), but a 10.4% deviation from a high rate of growth that would otherwise occur, with the deviation spread over 3 years? That’s somewhat different”

    This makes the assumption sound more reasonable bit I still do not quote understand it. Is a drop from 2% growth to 1.8% growth a 10.4% deviation from the growth rate? If so, a 10.4% deviation from a high growth rate does not seem particularly large in the face of Brexit.

  11. Clarification – I meant to say:
    Is a drop from 2% growth to 1.8% growth a 10% deviation from the growth rate? (i.e. 0.2% = 10% of 2%)

    1. Zhou: the assumed drop is in the rate of GDP growth, so the cumulative growth rate over the three years is, say, 5% instead of 15%. All explained in the EBA document.

    2. Zhou, I am reading Dikotter’s book “Mao’s great famine” and your namesake gets mentioned frequently. China was hell in the 20 years before Mao died.

  12. Presumably the EBA has ignored all the debt in the system in coming up with its growth assumptions. HSBC numbers are down 45% this week.

  13. A few weeks after a 26% growth in GDP due mainly to certain tax arrangements and the prospect of a 10% GDP fall over 3 years seems silly to most on here? What if the France and Germany get their way?

    @josephryan

    “I would be every bit as critical of Ireland’s model as your good self, but these EBA numbers simply don’t stack up, unless Ireland is to have another horrendous crash that Europe is largely insulated from.”

    See above.

Comments are closed.