Friday, 25 May 2018

Don’t believe the OW results: Spain’s banks are screwed

There’s something I don’t like about Oliver Wyman’s “Asset Quality Review and Bottom-up Stress Test Exercise” of Spain’s banking sector. It’s not that the exercise was almost wholly a technocratic models-based one; there was no getting away from the fact that this was the point of the exercise.

IFR Editor-at-large Keith Mullin

IFR Editor-at-large Keith Mullin

And on that score the report was certainly no more technocratic an exercise as you’d expect from a firm of management consultants conducting forward projections on the basis of a whole series of economic and financial variables that let’s face it couldn’t have been any more than best-efforts guesstimates of scenarios that might at some point turn out to be (hopefully) closely correlated with some dimension of future reality.

A lot of the input data came from third parties. Oliver Wyman took estimates from a host of real estate specialists to get a fix on the valuation of the banks’ masses of dud real estate; I imagine it also ran all sorts of weighty simulations around residential mortgage exposures. It certainly waded through huge swathes of loan data.

But at the same time, Wyman was heavily boxed in by the Bank of Spain (which commissioned the study), the Ministerio de Economía y Competitividad, the European Banking Authority, the European Commission, the European Central Bank and the International Monetary Fund.

In aggregate, these heavyweights formed what I imagine were pretty control-oriented committees (the Expert Coordination Committee and the Strategic Coordination Committee aka the Steering Committee) that I bet in practice offered little to no room for manoeuvre as to process, methodology – or even required outcome.

I don’t want to be negative towards Oliver Wyman. (Others have been pretty brutal.) What don’t I like about the report, then? Well, put simply, I don’t believe the results.

The economic scenario in Oliver Wyman’s adverse context was three standard deviations away from Spain’s long-term average for the three years of the exercise

So under the adverse scenario Ibercaja/Caja 3/Liberbank needs €2.1bn; Banco Mare Nostrum €2.2bn, Popular/Pastor €3.2bn, Banco de Valencia €3.46bn; NovaCaixaGalicia €7.17bn; CatalunyaBank €10.83bn and Bankia €24.74bn. It’s not the individual results that I’m concerned about. Most of the banks with projected deficits have been nationalised anyway so we knew they were duds. In particular, Bankia is still a dog; no change there.

It’s the assumptions I’m suspicious of. I didn’t spend the weekend running my own proprietary models; I’m going on anecdotal evidence of the situation reported by real people in Spain about the real estate market, empirical knowledge of how markets react to deteriorating situations, plus my gut feel that a lot of the assumptions – around the extent of negative deposit growth in base and adverse scenarios, future profit generation, future business generation, access to the corporate bond market to bridge the loan-deposit gap at anything like recent observed spreads, degree of real estate and mortgage delinquencies, supposed mortgage LTVs, depth of recession, levels of unemployment etc – just don’t feel believable.

In a situation reminiscent of former Goldman CFO David Viniar’s 25 standard-deviation moves several days in a row to explain away the global financial crisis, we’re told the economic scenario in Oliver Wyman’s adverse context was three standard deviations away from Spain’s long-term average for the three years of the exercise.

Do I feel any more comfortable about this supposedly conservative stance? Not at all, because frankly it’s meaningless. Do I feel any more comfortable that the adverse scenario is more severe than the EBA’s rather comical stress tests that no-one believed either? Err … even less so.

Excess capital buffer

And, like others, in particular I dislike in the extreme the excess capital buffer, i.e. excess over minimum capital adequacy requirements that’s been included in the tests that in my view and the views of many observers, undermines the outcome of the stress tests.

Excess capital derives from the deleveraging impact of Spanish banks selling RWAs. It’s one of four measures of banks’ loss-absorption. The others – existing provisions, asset-protection schemes, future profit-generation (even exaggerated) – I’ve got no problem with. But excess capital? It’s completely illusory.

In the base scenario, the banks are estimated to have €22bn of excess capital standing ready to absorb losses. In the adverse scenario, that rises to €73bn – of which €36bn will be earmarked to absorb losses. [Incidentally, the rather counter-intuitive higher excess capital buffer under the adverse scenario is down to the 6% CT1 requirement compared with the 9% base-case CT1].

We’re told the excess capital amount is within the upper range of the previous top-down exercise. Err, so what?

It doesn’t make it any less illusory. Even so, one of the many constraints imposed by the Expert Coordination Committee is that the “expected credit deleverage defined by the macroeconomic scenario is achieved”. Oh dear.

To figure out how illusory this source of excess capital is, just take a look at the European bank deleveraging process so far.

Extreme caution

Banks have been extremely cautious about offloading at prices much below currently high mark-to-market less fees in order to prevent the direct hit to capital the process is designed to avoid in the first place. The opportunity funds that have been circling Spain for over a year now have had slim pickings as the stressed/distressed prices they’ve been bidding haven’t been hit in size.

That’s because the assets for sale are predominantly performing and the banks aren’t forced sellers. Buyers have been using the process as a means of business expansion, buying into targeted areas such as project finance and big-ticket leasing and transportation finance, or back-engineering relationships with specific corporates in targeted sectors in targeted jurisdictions by acquiring corporate loan exposures. Most importantly, the process has been orderly.

Transplant that environment to Spain under Oliver Wyman’s adverse scenario. In that case, the assets for sale will in most cases be non-performing or close to non-performing, the sellers will be forced and the economy will be cratering. No buyers will emerge for toxic Spanish assets. Very heavily discounted bids might emerge but the sellers won’t be able to hit them because doing so in size would wipe out their capital.

Beyond that, asset-encumbrance will act as a drag on how much they can sell anyway. Be it cedulas hipotecarias, repos, derivatives or perhaps most importantly assets pledged to the ECB, Spanish banks don’t have the latitude to offload in size even if they could achieve the pricing they need. They’re caught between that proverbial rock and a hard place.

I’m going short.

Have your say

To have your say, you have to be signed in