When the levee breaks, I’ll have no place to stay.
Mean old levee taught me to weep and moan,
Lord, mean old levee taught me to weep and moan
The new Programme for Government document contains the government’s last approach to convincing the Greens, and to a lesser extent the general public, that NAMA will not be a costly exercise for the tax payer: A commitment to include a post-dated levy in the NAMA legislation.
I’ll try to boil the problems I see with this approach down to four observations.
First, NAMA plus an actuarially fair levy has the same effect as doing nothing. What does this mean? Suppose a bank has a loan originally worth €100 million which the government buys for €70 million. The loan turns out to be worth €50 million and the government imposes a levy of €20 million on the bank. In this case, the bank ends up with the same outcome as if it had originally held the loan. So, a levy of this type would mean that the NAMA policy does not achieve its aim of transferring the risk associated with bad loans away from the banking system.
Of course, this simple analysis doesn’t take into account time lags. For instance, if the government loses the €20 million today but waits ten years to impose the levy, then the banks would still be better off. This is because the banks have been able to keep €20 million which can be invested at an interest rate of r (reflecting both inflation and a real return) so that this money can be worth 20(1+r)^10 when the levy is introduced; similarly the government could have kept that €20 million and invested it. If the government chose to impose a levy that reflected the opportunity cost of money, then it would have to charge 20(1+r)^10.
Similar concerns apply to the idea of the levy being paid back over time. If you lend someone €20 and they pay it back to by starting in ten years time to pay you €1 per year, it is pretty clear that the economic value of the income stream paid back to you is going to be below the original €20. I calculated in my first post discussing the levy in April, that with an interest rate of 5%, an income stream of this sort would be worth less than 40% of the original amount.
Second, since the government doesn’t want to implement a plan that has the same effects as doing nothing, it is certain that any levy to be introduced will not be on an actuarially fair basis. Based on the discussion above you can see various ways that this can be done. One way would be to exclude any operating losses NAMA makes in terms of the gap between its income and interest payments (we will see if revised legislation makes this clear.) Another way would to have the money paid back as a fixed nominal amount over a long period of time.
Now, on the one hand, these tricks could be considered a good thing, since we don’t want a plan that’s the same as doing nothing. On the other hand, it means that claims that a levy will fully protect the taxpayer simply cannot be believed.
Third, the levy idea opens up a logistical can of worms. It seems likely that the levy will be based on NAMA’s total losses because the only thing we know about the subordinated NAMA bonds is that any contingency has to relate to NAMA’s aggregate performance. But this then exposes AIB and BOI to having to pay for the losses associated with Anglo’s loans, which hardly seems fair (though, of course, I’d guess that NAMA is applying far higher haircuts to Anglo loans than to AIB’s or BOI’s—not that they’re willing to tell us.)
The alternative approach is to have levies tied to the losses associated with each specific bank. But Brendan McDonagh told an Oireachtas Committee in May that this could not be done:
If there was a clawback within the NAMA legislation affecting the balance sheets of the banks, they would not be able to reduce the assets transferred to NAMA because effectively there would be an unpriced option in terms of what the clawback would be in the future. One cannot do this because it would not be possible to take risk weighted assets off the balance sheets of the banks if the levy was imposed in the NAMA legislation.
It’s unclear whether this would apply if the levy related to all NAMA-related losses. Still, if I was involved in assessing AIB’s solvency, I’d certainly want to factor in a potential future levy, even if it wasn’t listed in the bank’s official liabilities.
Fourth, I have always argued that the financial institutions would lobby hard to prevent future legislation to implement a levy. The levy being part of existing law will make thislobbying effort a little bit harder but not too much. The lobbying will just change form to become a request for new legislation to scrap the levy.
Finally, I would note that the NAMA plan is now moving farther and farther away from anybody’s ideal banking cleanup plan. The textbook approach to cleaning up a banking sector is to (a) write down bad assets to realistic values either directly or via transfer at low prices to an asset management agency (b) Recapitalise the banking system. This allows the banks to get back to playing their normal role without large contingent risks hanging over them.
The government’s approach of transferring the assets at unrealistically high prices while leaving contingent risks associated with bad loans hanging over the banks is a bad one all round.
I’d conclude by noting that there has been an increasing tendency on this site for people to attack me personally, so before people start writing in saying “you’re never happy, do you want the taxpayer protected or not?”, I’d like to point out that in April, long before any legislation was published, I wrote: “I would strongly recommend getting the process of pricing and recapitalisation done correctly now and abandoning any idea of a post-dated levy.” So I’m just sticking to my original position on this rather than shifting it to oppose the latest changes.