Toxic Assets and Recapitalisation

Last night’s RTE news was full of breathless commentary from Brussels and Montrose to the effect that governments needed to do more than just re-capitalise the banks.  From Brussels, Sean Whelan excitedly talked about “toxic assets” and how it was going to be necessary to “buy the toxic assets at a steep discount, leaving the banks to continue with the viable parts of their business.  Simply re-capitalising the banks isn’t seen as enough.”  Back in Montrose, David Murphy authoritatively informed us that “In terms of doing something else in addition to re-capitalisation, they have an option of setting up a bad bank or, for instance, insuring the banks against the some of the bad loans that are coming down the track.  But it’s pretty clear that the re-capitalisation on its own won’t be enough.   And if they do something that isn’t enough, the markets won’t like it one little bit and Ireland will be punished for that.”

Despite the confident tones in which this expert analysis was delivered, as far as I can see it doesn’t make any sense. 

There is no logical distinction between the issue of undercapitalisation and the problems created by so-called toxic assets.    Banks could get rid of all their toxic assets in an instant if they just wrote them down to zero.  But then they would be undercapitalised—and that’s the problem with toxic assets.  What appears to be going on is that the various re-capitalisation programs around the world have not managed to offset the likely losses from bad loans.  However, this is a problem with the size of the re-capitalisation programs, not their nature.  There is no logical argument for now augmenting these programs with a scheme to systematically overpay for impaired assets.  (Sean Whelan may think the proposal is for buying assets “at a steep discount” but that discount is relative to what they were originally worth, not what they are worth now.)

What appears to be going on is that neither banks or governments want to inject more government equity capital because going any further than we are now requires effectively admitting that the banks need to be nationalised.  Government concerns about nationalising banks are well-founded but it seems that, in many cases, we are well beyond that point.  If the banks can’t find private equity to re-capitalise them and government is willing to do so, then that’s where we should end up.

As a final point, it’s worth noting that despite the constant commentary about toxic assets and all the problems created by the evils of structured finance (CDO-squared and all that), this stuff is essentially irrelevant to the Irish banks.  Their toxic assets are largely plain old loans to bankrupt developers and rocket science isn’t required to come up with a guess at the size of the loan losses.  As Colm McCarthy has joked “We didn’t import any toxic assets, we grew our own.”

As usual, I have the sneaking feeling that I’m missing something.  Perhaps our team of commenting pundits can explain to me what I’ve got wrong.

19 replies on “Toxic Assets and Recapitalisation”

Perhaps what the RTE correspondents meant was that the recaps “won’t be enough” to position the banks to relax lending standards and “get credit flowing again”, which seems to be one of the Government’s objectives. If that is what they meant, then they surely have a point.

Setting up private “bad banks” is still an option for AIB and BOI. They are simply mechanisms to maximize value of the banks’ assets. Because each would have its own private bad bank, the pricing of transferred impaired assets is not as tricky an issue as in the case of sales to a state-owned AMC. Of course, whether they have enough capital to do meaningful transfers is another question. As you say, Karl, large injections of ordinary equity may be what’s needed, but the Government seems keen to avoid nationalization of the big two.

Yes, I noticed and particularly liked that comment about how governments would “buy the toxic assets at a steep discount”. For discount, read premium.

Presumably if nationalisation is going to happen anyway, then it would make sense to do so before flinging money at the banks rather than afterwards?

TARP coverage via

Prof. WARREN: Well, they’re trying to push money into banks and the question the oversight panel was asking is, `are you getting an equivalent amount back?’ And so that’s what this was about. Now there could be lots of policy reasons that Treasury might decide that it wanted this money to be in the banks. But our question is the one we put to Secretary Paulson, and that is, `are you putting it in and getting back assets that are worth equivalent value?’ He told us yes; our independent investigation said no.

CHEN: So are you saying he was lying?

Prof. WARREN: Well, I’m telling you he told us yes and our independent investigation said no.

I think one of the key issues in this debate — and the reason for a disintinction between bad banks and insurance schemes on the one hand and plain recap on the other — is the need to in some way remove fear in the markets that some dreadful future horrors are hidden on bank balance sheets.

Here the crux is that the banks don’t want to write down all their bad debts at once ( and claim anyway that they are not as bad as everyone thinks). But not doing so still leaves the uncertainty in the market. Some kind of bad bank or insurance arrangement might remove this uncertainty, though Karl is of course correct that either way the State ends up picking up a large part of the tab.

The other part of the issue, then, is the split in the cost between the shareholders and the state. Have to admit with some confusion about the range of proposals emerging now both here and internationally to deal with this !

The way I see it, there is a conceptual difference between recapitalisation and buying toxic assets.

If you recapitalise, you still have the uncertainty regarding the value of the banks’ balance sheets. You have a toxic asset valued at 50c, but it may be worth 100 or zero, depending on the state of the world. Therefore, counterparties are still reluctant to provide term funding, investors are reluctant to invest in equity, and the bank will also tend to hoard capital as a precaution against the most adverse scenarios.

Buying toxic assets removes this risk (more correctly transfers it to the state, which is assumed to be in a better position to deal with the volatility / uncertainty) and swaps big question marks for cold, hard cash.

This is the frustration that I had with the whole TARP debate – I feel the focus on overpaying somewhat misses this point. The aim is to remove uncertainty from the banks’ balance sheets, and whether Treasury pays 50c or 70c is somewhat less important than this removal of uncertainty. In my view, it’s the potential volatility of the balance sheet and hence the heightened tail risk, not the snapshot best guess expected value, that is the most important driver of behaviour.

As things stand, we’re trapped in a reflexive situation, whereby banks are holding these toxic assets, hoping for the best, but fearing that these assets are worth very little. Hence they hoard capital, choke off lending, and hence generate the very tail scenario that they are praying won’t come to pass. We have to break this cycle. If we recapitalise, the fear is that we will need to recap further in the months and years ahead, and the debt-deflationary spiral continues.

I think both AIB and BOI should be nationalised, and then all bad assets send off to a bad bank. Thereafter, the goods parts should be sold off to private investors. Recapitalising banks simply wont work. Time after time, we’ve seen money going into banks and then they come back to the trough for more. No investor will rationally believe a quick fix recapitalisation will sort out a balance sheet. Esp today, when levels of risk aversion are sky high. Recap didnt work in Japan. And TARP didnt work. Despite TARP, look what Merrill did to BOA. (The idea that government recapitalisation will induce private investors in is also a joke; with the gov already investing *on their behalf*, why should they go in for a second time?)

The distinction between undercapitalisation and the problems of toxic assets is that the former does not explicitly recognise the root cause of the problem, and the latter must.

The root problem with toxic assets in Ireland, it seems to me, is that loans were made on underlying property assets which were grossly overvalued. The more overvalued these securities, the less the loan asset is worth, for any given risk of default. Worse, as the property asset values adjust downwards, the risk of default rises, because the debtor just stands less to lose. And worse still, as the risk of default rises, the property asset values fall due to tighter credit. Thus, toxicity is exponential in falling house prices and rising default.

This suggests to me that the toxic asset problem is more widespread than has been acknowledged – it relates to an overvaluation of all property-related lending, including performing loans.

The only way to really take account of this is to nationalise all the banks and conduct a root-and-branch revaluation of their entire portfolios of secured loans, based on new, underlying security valuations that reflect fundamental value (ie based on a conservative, CAPM style approach).

I agree that this recapitalisation of the banks is a very questionable thing to do at the moment. Let’s face it that conditions are undoubtedly going to deteriorate over the next 12 months as unemployment rises, growth falls faster and house prices continue to plummet.
With this in mind I can’t see any other outcome than the banks needing more capital in 12 months or perhaps even sooner. There is also the question of will the funds actually be used to lend to the various businesses that need it, I think not.
I think that nationalising the big two banks may be the most sensible course of action or perhaps the creation of a new clean bank. At least if we nationalise the banks we as taxpayers have risks but there are more potential rewards, it seems like an investment albeit a forced one. A simple recapitalisation has its rewards too if it works, perhaps the banks will start lending again raise additional funds. But this seems less likely especially when the TARP programme in the US has not been the success, especially when you see what BOA have used the Funds for.
I am also interested to learn why nationalisation is such a bad thing for our banks given the situation we are in. One of the main arguments against nationalisation is that states are not experts in lending, but surely this is a small price to pay for lending to resume in some capacity to the SME’s that need it most.
What are the other main reasons against nationalising the big two banks at the moment and do the pros outweigh the cons?

“…the focus on overpaying [for “toxic assets”] somewhat misses this point. The aim is to remove uncertainty from the banks’ balance sheets, and whether Treasury pays 50c or 70c is somewhat less important than this removal of uncertainty.”

I think the question of overpaying doesn’t so much miss the point as it relates to a different point – that restoring the financial system need not necessarily entail a massive gift from the taxpayer to existing management and shareholders. Or, if it does, this should be done with some transparency rather than pretending that the state is doing something else.

James, I agree; perhaps I should have written ‘the exclusive focus on overpaying somewhat misses the point’. I didn’t mean to dismiss the issue of asset valuations. (Although in this regard, another issue that seldom seems to get an airing in the debate about ‘getting a good deal for the taxpayer’ is that if one strikes too good a deal for the taxpayer, it becomes something of a pyrrhic victory, as the banks go down the tubes.)

I guess my problem is that so many actions have already been taken which, without much detailed calculation, entail a massive gift from the prudent to the imprudent. That’s the nature of the beast. Hence it seems a bit misguided for us to get stuck on whether we are getting exactly the right price in a TARP-like program, when simultaneously, money is being thrown at the problem by way of rate cuts, guarantees, capital injections, regulatory forebearance without any sort of similar calculus.

I was of the assumption we lived in a market society, if that is the case what would be the downside of this:

nationalise the banks if you must, that puts the tax payer at risk, so in return you only take the good assets onto the national balance sheet.

Shareholders and bondholders (everybody else) can have all of the non-performing assets and see if they can liquidate/service them for any profit, they might only get a few cent back on the euro but they took the risk and thus take the reward as well as the downside risk

shareholders should accept this because the option behind door 2 is ‘you loose 100%’.

Depositors are protected under the existing scheme and the writedowns pass from the bank balance sheet to the rightful owners of the firm – shareholders

i say this as a shareholder too (in case it seems too callous!)

This questions was dealt with explicitly and cogently by Caballero and Krishnamurthy on the FT’s Economist forum and at RGE Monitor back in November. It echoes what many of the commenters have already said, but probably puts it more succinctly:

“The Knightian uncertainty perspective also sheds light on some of the virtues of the now defunct asset-purchases program of the original TARP.  In practice, financial institutions face a constraint such that value-at-risk must be less than some multiple of equity.  In normal times, this structure speaks to the power of equity injections, since these are “multiplied” many times in relaxing the value-at-risk constraint. In contrast, buying assets reduces value-at-risk by reducing risk directly, which typically does not involve a multiplier.   However, when uncertainty is rampant, some illiquid and complex assets, such as CDOs and CDO-squared, can reverse this calculation. In such cases, removing the uncertainty-creating assets from the balance sheet of the financial institution reduces risk by multiples, and frees capital, more effectively than directly injecting equity capital.”

Given the worsened conditions in banking sectors globally since that article was published – most major banks are now technically insolvent – I think we are converging on whole-sale nationalization and then separation of toxic assets as the best solution. That is what Roubini et al have been calling for months and, at this stage I think it is almost inevitable in the U.S. where I am based. I imagine it also will be in Ireland.

“nationalise the banks if you must, that puts the tax payer at risk…”

People continue to say this, but perhaps someone can explain to me why nationalisation makes a difference compared to the guarentee (except that the guarentee is supposedly going to lapse in 2010)? Surely the taxpayer already owns the liabilities if things go belly-up.

The problem with the guarantee is that you can’t trade with the guarantee alone. It’s like running out of petrol on the way to the airport – the fact that you have AA coverage is really no good to you. You need transport now, not when the guy in the van gets round to showing up. The equivalent for the banks is capital. You need capital to be in business, and you need to inject capital somehow. Nationalisation (the State buying shares) is one of the only ways of doing this. You can start a ‘bad bank’ but that effectively means that the State buys and recapitalises only the ‘bad’ debt and gets none of the benefit of the bad stuff.

Once the guarantee went in place, the State was effectively committed to nationalising the banks if necessary. That is my read anyway.

The main reason for not nationalising banks is to avoid having the government directly involved in lending decisions. Private, not public entities make the major investment decisions. That is the way our system works. If the government ends up making those decisions you end up with a poor effort at a planned economy, with short election-driven time horizons on investment. Bottom-up innovation begins to disappear. Anyway, government-controlled lending policy makes lending sense in a world where capital flows across borders.

Sorry, I should have made clear in my previous comment that I was wondering what difference nationalisation would make in terms of the risk to the taxpayer, given that the guarentee already means the state owns the liabilities if things go wrong.

Regarding Antoin’s last paragraph, two ponts:

1) ownership and management are obviously two different (albeit closely related) things. Modern states have developed a fairly sophisticated and often successful technology of delegation – cf central bank independence or even the ESB’s recent wage increase!

2) countries as different as Germany and China have banking systems characterised by state banks (the landesbanks in the German case) without this being obviously incompatible with a functioning economy. Norway is another example I believe. Perhaps someone can suggest some relevant literature here?

James: but if you separate the management and the ownership what is the benefit of having a nationalized bank (other than because you think it is going to earn the government coffers a good yield)? I accept that I have over-stated the badness of government banking! But since you gave a good example of a ‘good’ national bank, here is a tale of woe from a not-so-good institution.

Well, if we accept that the banks do indeed need a hefty recap, I (humble amateur observer) don’t understand how the preference share route works, since it would seem that under the current plan the state will actually be sucking money out of the banks over the next five years.

It follows that what is needed is either a great big unilateral gift to the banks (in the form of buying the dodgy assets at a sweetheart price or – same thing – insuring them against losses on said assets) or a proper recap through ordinary shares. The problem, of course, is that the scale of recap needed appears to be several times the present market value of the banks. So again the choice is a sweetheart deal on the purchase of ordinary shares or effective nationalisation (purchase of all or most of the shares).

So yes, the benefit of nationalisation would indeed be the yield to the government coffers (or perhaps the smaller hit to the coffers). It would also have the ironic merit of applying some market discipline, provided existing shareholders and management were cleared out.

Comments are closed.