This note is structured as follows: first, I cover Michael Somers' very revealing and honest testimony to the PAC, then I review Friday Davy note on NAMA, lastly, I provide complete estimates of expected losses for NAMA.
Michael Somers - some fresh air on NAMA
On Thursday NTMA ceo Michael Somers told the Public Accounts Committee that putting valuations of the bad loans will present “an enormous dilemma”. Of course in logic, ‘dilemma’ always leads to two undesirable alternatives – in other words, it cannot be resolved within the same logical reasoning chain that leads to it. In layman’s terms, this means that the only way to resolve NAMA problems is to dump the idea alltogether.
So what is this ‘dilemma’ that the country has learned about only this week?
It turns out to that "there will be arguments down the courts if we don’t get it right. The implications of this thing are enormous and the legislation will be very complex,” Mr Somers said. Hmmm… this is hardly new. This blog and many well publicised articles, some written or co-written by me plus a massive wave of media reports that certain big developers are preparing to challenge NAMA - all were well ahead of Mr Somers. But Mr Somers’ testimony is so news worthy now because it is no longer the dirty scoundrels in media and academia who are beating up Lenihan’s dream baby, but one of the Golden Public Circle’s own.
Mr Somers also said that he believed up to 5,000 officials in the main banks “were currently examining bad loans. …At the moment, we really have no feel for how Nama will operate. But my preference would be a core group of between 30-40 people.” Now, wait a second. 5,000 banks officials cannot get the loans right, but 30-40 NTMA/NAMA folks will? This is after Mr Somers admitted that NTMA has not experience in managing distressed assets whatsoever. Of course, Mr Somers was saying 30-40 NAMA officials on top of 5,000 banks officers already in place will manage NAMA, but in such a scecnario, any final cost of NAMA will have to include the cost of those 5,000 bankers as well...
Mr Somers said the NTMA paid its 170 staff a total of €19.4 million in 2008 – “an average of almost €90,000 per person” as Irish Times puts it. Well: 170 staff at €19.4mln is €114,118 per person in pay. Including secretaries and other auxiliary staff, NTMA is now the best paid state entity on the record – ahead of ESB. But hold on, further €8.9 million was paid out in expenses. So total compensation (inclusive of expenses) came to €166,471 per head.
The 30-40 people that Mr Somers would like to have for NAMA is about 20-times smaller than normally is required to run a €90bn fund. Now, given that NAMA will be a distressed assets fund with less active management, say 600-700 people specialising in asset management, portfolio structuring, risk pricing etc would suffice. If the NTMA wage bill were to apply (and there is absolutely no reason as to why it won’t) – we, the taxpayers, are looking at paying something to the tune of €100mln in wages and expenses for these ‘servants’ of the state annually.
Another dilemma, clearly of unresolvable variety, is that NAMA “is expected to pay for the bad loans through the issue of Government bonds to the lenders.” (Quote per Irish Times). Apart from not being new (first disclosed back in April) and being banal (the state has to borrow cash and issue bonds to pay for its day-to-day spending, let alone NAMA), this claim is rather sterile.
In reality, the Government hopes that it will be able to borrow directly and at reasonable terms, but of course, it might run into some tight spots asking the markets to lend money
- to a half-backed bad bank to be run by 30-40 inexperienced staff;
- mired up to its chin in the mud of court challenges in our notoriously costly and slow legal system;
- operating under the umbrella of our Guinness-serving bond issuing authority;
- in a country whose Government cannot balance its own books;
- with economy managed so poorly, that we are now presented internationally as the worst case scenario possible; and
- the Government so grossly clientelist that it can't even manage its own employees without having to run crying to the 'mommy' of the Social Partnership;
- add to it that NAMA has no popular or political support; and
- that the same Government is doing everything possible to make certain Irish private economy will not come out of this recession with any strength left
Irish Times also told us that “Profits from the eventual sale of the loans will be given to the state which may be used to service the €54.2 billion national debt.” Ok, what national debt do they have in mind? €90-110bn or more debt we will have once Mr Lenihan ends his current deficit financing of the public sector employees lifestyles? In fact, the same Irish Times reported as a comment from NTMA chief that he expects Irish debt to top 100% of our GDP in 2010 (see below). So, does anyone in the Times editorial have a calculator at hand?
It's all down to the cost of NAMA
“However if Nama makes a loss, the Government said it will apply a levy on banks to recoup the shortfall.” Ireland’s stockbrokers decidedly focused on this statement much more than on anything else that Mr Somers said. Why? Because this is the real unknown unknown for banks shareholders. This is, of course, ultimately the question of how big the loss will be.
So let’s do some counting of the beans… shall we?
First what the financial markets analysts say: per Davy note assumption, “costs for NAMA will be covered by interest income from the performing loans. Taking a conservative view that only 50% of the €30bn investment loans to be transferred are performing, this could generate initial income of c.€1.2bn as yields are currently at 8%. This would cover the coupon (assumed to be in line with the borrowing rate on ECB liquidity facility) of the bonds issued to the banks in return for their €80-90bn of loans transferred to NAMA. Income will likely be higher because the investment loans to be transferred to NAMA are not bad loans in themselves but are selected due to cross-collateralisation with development loans. An investigation of staff costs as a percentage of properties under management for large real estate trusts suggests that staff costs will be covered, especially given the greater operational leverage due to NAMA's size. Legal costs are less certain as the legality of the agency is one of the greatest obstacles to its performance.”
... And then, the pigs are soaring high in the sky…
Suppose that NTMA goes out to the markets with two suitcases worth of bonds – government bonds and NAMA bonds. Do you think they can price NAMA debt:
- at a discount to current public debt issues (a scenario that Davy suggests); or
- at a rate that is equivalent to Government bonds, say ca 4.5%; or
- at a rate that is higher than Government bonds, say a premium of 20% to Government debt – for 5.4%, while the Government debt remains priced at 4.5%; or
- as the markets look at two piles of paper, they tell NTMA: “Ok, we’ll take Government bonds at a premium to previous issues to account for vast number of these things being floated in the market – say 5%, and NAMA bonds at a 20% premium on that – at 6%”?
For (2): at 4.5% pa (a very optimistic scenario), buying €90bn worth of loans at 15% discount (as Davy suggest) on their face value will cost us €3.44bn annually in coupon payments. But wait, let’s also look at the downside scenarios: (3) implies €4.13bn price tag, and under scenario (4) the cost of annual NAMA financing alone rises to €4.59bn, or ca 15.3% of the entire Exchequer tax revenue in 2009...
Hmmm... back to that 'country whose Government cannot balance its own books' bullet point above, anyone?
Per debt financing assumptions, Somers said Ireland would be “lucky” to hold on to its sole remaining top AAA credit rating, as its low debt levels could surge to more than 100 per cent of GDP next year, from about 41 per cent in 2008, after the State completed the transfer of the banks’ bad loans. Hmmmm… remember that Irish Times statement that NAMA can be used to pay down €54.2bn debt quoted above? Apparently, NTMA chief expects the debt in excess of €170bn by the end of 2010 - somewhat higher than Irish Times journos do. And, apparently, he does not expect to price any debt (Government or NAMA-issued) at ECB discount facility rates (as Davy assume).
Of course, it is a trite statement to say that NAMA will be able to cover its staff costs, so no dwelling on this, but as far as legal costs uncertainty goes, despite Davy's rather neutered murmurs on this, there is no downside protection for the taxpayers. In other words, once committed to the transfer of a loan, NAMA implicitly assumes that no matter what the legal costs might be, the loan will be moved.
Somers was lethal on this last week: “I see great potential for arguments down in the courts if we don’t get this right,” saying further that he heard “people down in the courts were delighted” about the setting up of NAMA as they were in line for “a bonanza”. He said there would be “eating and drinking” at the committee for decades over the set-up of NAMA. But Davy folks didn't listen, apparently.
Davy goes on to pour more fuel on the NAMA pyre: “…the fear of the unknown has also led developers to call for consultation with the government over NAMA, with many saying that they can be part of the solution. Engagement with developers would arguably help reduce legal challenges and secure buy-in. The NTMA itself has stated that it lacks sufficient skills for setting up NAMA, but many developers will be both skilled and, more importantly, incentivised to work out their projects/loans under NAMA.”
Indeed, NAMA is at a severe risk of getting into bed with developers. Not because developers are ‘evil’ (I certainly do not believe this), but because NAMA should be independent from developers interests and free from their influences. And yet, Davy does not even see the lunacy of its own suggestion that NAMA should engage developers in the management process. This is really worrisome.
Doing the final sums on NAMA cost
Davy’s “simple NAMA model shows that taking our 15% haircut assumption for the sector and assuming no profits on disposal of NAMA assets across a 15-year work-out period implies a present value loss of €4bn. Apportioning this fee on an annual basis over 20 years (similar to Insurance Corporation of Ireland) implies that this fee would be only c.3-4% of normalised profits. However, given the long-term nature of the agency, there is every chance that it can turn a profit similar to that which we have seen in past banking crises such as in Sweden. A more positive outlook, with greater performance from investment assets under NAMA and a 5% profit in aggregate across asset disposals, would result in a present value profit on wind-up of €3bn to the taxpayer from NAMA.”
I do not have Davy’s model at hand to see what assumptions they force into it to get these numbers. Judging by their assumptions on pricing NAMA bonds (above), I have no confidence in any of their numbers. Their concluding scenario in the quote above is so far out there, that the aforementioned Amsterdam brownies come to mind again.
But my own simple model goes as follows (with my assumptions listed transparently for all of you to see and to challenge, unlike Davy):
So, net impact is a loss of €33-68bn.
Remember, Mr Somers said he was “aghast” at the scale of development loans advanced to a small number of borrowers which emerged after the NTMA reviewed the banks’ books. He called the review “a huge eye opener for us”. This is not a statement from a man who expect loans losses to be in 5% category (as my ‘Near Davy’ scenario above assumes).
So turning back to that original concern that investors should have about NAMA - what share of these losses can be recovered through a 3-4% shave on future profits of the banks?