Two Scenarios for the Banks

On last night’s Prime Time, Brendan Keenan argued that it didn’t matter much how the government dealt with the problem of bad loans at the Irish banks, as long as they got on with doing it, though he noted he would be very reluctant to nationalise.  Similarly, David McWilliams said that the key thing was to do something to deal with the bad assets and that it doesn’t matter whether we nationalise or not, i.e. that we needed to produce cleaned-up banks and it didn’t matter who owned them.

Let me explain why I think it does matter how we go about this and who owns the cleaned banks.  Start with a few figures. The combined capital of AIB and BOI is currently estimated at about €20 billion, but they are only beginning the process of writing down their bad loans. The fact that their combined market capitalisations are below €1 billion suggests that the market believes these losses will amount to at least €20 billion. Let’s say, just for illustrative purposes, that €30 billion is the figure for losses.

Consider now two scenarios for creating cleaned-up versions of these banks, both of which involve the removal of the bad loans from their balance sheet. I’m going to assume, for illustration, that after the bad loans are replaced with other funds, the banks will need €15 billion in capital to carry on as healthy operations (I’m thinking here that they should pay off some debts and operate with a lower loan-to-deposit ratio and a smaller balance sheet, thus requiring less capital.)

Scenario 1: The Irish government announces that it is not going to renew the liability guarantee beyond September 2010 or provide further supports. Without taxpayer support, the banks will not be viable and will thus need to be nationalised. The loans are taken over by a government asset management company and replaced on the bank balance sheets with government bonds worth €5 billion less than their book value. The government now owns two types of entities: An asset management company whose job is to sell off the bad property assets over time in order to minimise the losses for the state (but which, on current valuations is worth €25 billion less than the government paid for it) and a couple of cleaned-up banks with no bad assets and equity capital of €15 billion. As soon as the operation is complete, the government can look to privatise the cleaned-up banks, thereby recouping its €15 billion it has provided in capital.

Scenario 2: The Irish government sets up an asset management company and it buys the bad loans from AIB and BOI for €5 billion less than their book value. It sets up an asset management company to get as much as possible for these assets over time but the current valuation of this company is minus €25 billion. Again, we now have two cleaned-up banks with no bad assets and equity capital of €15 billion. 

As Patrick Honohan pointed out in comments yesterday, it is now widely accepted that the Irish banks need to be cleaned up, so the real issue is how to allocate the necessary losses between the existing shareholders and the taxpayer. Under Scenario 1, the €30 billion in necessary losses are shared so that shareholders lose their €20 billion in equity capital and the Irish taxpayers lose €10 billion (the €25 billion loss on the bad property assets minus the €15 billion gained from spinning off the new banks.) Under Scenario 2, the losses are split so that the current shareholders lose €5 billion and the Irish taxpayer loses €25 billion.

Under both scenarios, the outcome is a cleaned-up banking system ready to perform its crucial financial intermediation functions.  Perhaps Keenan and McWilliams would be indifferent between these two scenarios.  However, I prefer Scenario 1 because

  1. It costs the Irish taxpayer €15 billion less.
  2. By nationalising, cleaning-up, and re-privatising, it leaves the banks in new private ownership, so that the management and shareholders of the banks that caused this trouble are not rewarded with the gift of a valuable asset courtesy of the taxpayer.

As Patrick pointed out in his comment, there are still a lot of details to come out about Bacon’s bad bank proposals. However, they appear to look a lot more like Scenario 2 than Scenario 1, particularly when you factor in that the Independent article describing the leaked details of the report cited “staving off nationalisation” as a key achievement of Bacon’s proposal.

One final issue is worth flagging. Both of these scenarios have an asset management company, a “bad bank” if you must.  So, to say that the issue is that “we need a bad bank” is simply beside the point.  The real question is how the losses are allocated.  And previous international examples of bad banks have operated along the lines of Scenario 1 above, not Scenario 2.  On Prime Time, Keenan noted his reluctance to nationalise but came out in favour of the bad bank idea (so presumably endorsing something like Scenario 2) and said that the IMF had noted that such schemes had worked well in the past.  Here is a link that brings you to yesterday’s IMF report and here’s what it says on page 11:

Insolvent institutions (with insufficient cash flows) should be closed, merged, or temporarily placed in public ownership until private sector solutions can be developed. While permanent public ownership of core banking institutions would be undesirable from a number of perspectives, there have been numerous instances (for example, Japan, Sweden and the United States), where a period of public ownership has been used to cleanse balance sheets and pave the way to sales back to the private sector.

PS. The most confusing aspect of last night’s program was McWilliams’s references to “Noddyland”.  I’m pretty sure Noddy lived in Toyland.




11 replies on “Two Scenarios for the Banks”

Two silly questions.
First question refers to the following:
“it buys the bad loans from AIB and BOI for €5 billion less than their book value”.

Why 5bn? Why not 7bn less than book value? Or 12bn? Or 17bn?

Second question. After the 1980s AIB bailoit, there was some sort of bank levy used to claw back money for taxpayers over a long period. Why not same again?

These are civilian questions, I know – but since we’re paying the bill we do have some interest in these matters.


Karl, your example is helpful. Critical to your example, however, is that the government overpays for the assets (25 for assets with book value of 30 but actually worth 0). When the TARP was first mooted it was indeed seen as a backdoor mechanism for recapitalising the bank. This required the government overpay in expected value terms for the assets. There was also the additional goal of reducing uncertainty about the solvency of the bank.

But doesn’t the situation change when we add in a second recapitalisation instrument (e.g. preference shares). Now we have two goals and two instruments. The goals are reduced uncertainty and a well capitalised bank. The instruments are purchases of toxic assets at expected value (to reduce uncertainty about value) and injections of preference shares (to adequately capitalise the banks. With the appropriate use of instuments, it should be possible to achieve: (i) removal of toxic assets; (ii) adequately capitalised banks; (iii) avoid the inevitable politicisation associated with nationalisation; and (iv) avoid a taxpayer-funded windfall to the banks. Hopefully, this is what Peter Bacon is thinking of.

Regarding nationalisation, does anyone believe that it is important that the nationalised banks be returned to Irish ownership and whether this will be possible? Already, we’re seeing the impact of subtle – and not so subtle – pressure on banks to restrict foreign lending and to focus on their home market.

It would therfore seem to follow that Ireland should seek to ensure that at least part of its banking sector remains owned and domiciled in Ireland rather than becoming the Irish division of a larger European Group. However, I do wonder whether the Irish market would have the capacity to absorb a newly privatised AIB and Bank of Ireland at some point in the future.

This seems to be at least part of the problem that Central and Eastern Europe has where their local banks are dependent on their foreign shareholders for capital and there are concerns that those shareholders will develop their domestic businesses at the expense of the CEE subsidiaries.

Here’s another silly question. Why does Scenario 1 have to involve a total cutoff at 09.2010? Think of it like Iraq – under Obama the US is “leaving” but 50,000 troops are staying.

How about Scenario 1a – announces that from September 2010, guarantees will be reduced x%/month to baseline, tapered so that the guarantees are more sharply reduced for large liabilities in order to preserve voter (erm…) depositor confidence. This gives the banks further breathing room to reorganise under the continuing competitive advantage while providing government with a clearer exit strategy.

There is a lot of zeal in some quarters to clean out shareholders rather than given them a “taxpayer funded windfall” but who are those shareholders? Aren’t some of them those fools who followed government advice and took out pensions rather than rely on State assistance in their dotage? What moral hazard is theirs? I’m not saying it should be entirely on the State’s nickel but this is largely a question of management strategy and State under-regulation, matters a typical mutual fund holder or the retiree who loaded up on so-called blue chips had as much chance of influencing as Father Christmas.

This post is spot-on. The difference between Scenarios 1 and 2 is exactly the allocation of the losses between shareholders and taxpayers, as well as the treatment of the current circle of executives.

In his piece in the Irish Times today, Dan O’Brien argues forcefully that Ireland must work hard to dispel its image of crony capitalism. Giving the sack to the banks execs and the wooden spoon to shareholders would be a forceful message of economic justice to send to the markets.

There is a bigger prize at stake here, which is our credibility among those who buy our bonds. If we lose that, we will really be up the creek without a paddle.

One additional point: In the international debate, one argument for nationalisation is that it allows you to impose losses on subordinated debt holders (e.g. see the link below to a recent Buiter post). As Patrick Honohon has pointed out, however, the majority of the debt of the Irish banks is subject to the guarantee. Moreover — and I hope I’ve got my fact right here — my understanding is that no losses were imposed on debt holders under the Anglo nationalisation. So we can be reasonably sure the Irish model of nationalisation does not impose burden sharing of that kind.

I find the argument made by Raghuran Ragan in this NPR radio interview quite compelling (fast forward to about 2 minutes in; Ragan is yet another former IMF chief economist):

I do not have an objection to the government using what governance power it has to replace top management. But I think we should try to keep politicians as far away as possible from day to day decisions about credit allocation. Ownership does matter.

Excellent post and debate, I just wish it was happening more on TV & Radio as well, it would be great to see some of this blog’s contributors on the likes of Prime Time more often debating these points. I fear if it wasn’t for McWilliams & Co there’d be little if any public debate.

There’s plenty of talk and debate about bad banks, asset insurance, nationalisation, recapitalisation etc. What’s amazing is the complete failure of most such discussion to recognise (as Karl does) the degree to which distributive rather than (or as well as) technical considerations are central.

Part of the difficulty is that for those of is that go on the meeja the time and editorial constraints are so significant that details such as the allocation of ownership etc sometimes get cut or not gotten to.
We have, I think, moved the debate a bit towards the details and now perhaps we can move it more forward.

Comments are closed.