The FT has noted this morning that:
New accounting rules could force Europe's top banks to recognise an extra €61.5bn in loan losses, new analysis shows, as a major UK pensions group has warned that bank accounting is not fit for purpose.
Analysts at Barclays have calculated how 27 of Europe's biggest banks would fare under new global rules governing how much lenders should set aside for potential bad loans.
In a report published on Tuesday, they found that the rules would trigger an increase of about 34 per cent in loan loss provisions across the group, as well as lower bank valuations and more volatile earnings.
This is deeply significant. The new rules are what are called International Financial Reporting Standard 9. What they demand is a change to the way that banks provide for bad debts in their accounts.
Until the adoption of IFRS in 2005 UK banks were required to anticipate the losses that they might incur on their loan books. That meant if they had a reason to think a loan might go bad they had to anticipate that and make a provision for the loss when it could be anticipated. The approach was prudent but wholly relevant: it meant that managers had to focus on the future consequences of current loans: that is exactly as it should be as all loans exist over time, and not in a moment.
IFRS changed that: what it said was that loan losses should only be booked as they were incurred. So, unless and until a loan actually went bad nothing should be done about it in accounting terms even if it was entirely anticipatable that it would, at least in part, fail. This measure also had the net impact of reducing general loan provisions that recognise that some loans will fail, but you just don't know which ones as yet.
The 2005 changes had a massive impact on bank profits: they rose enormously. This in turn had a massive impact on the economy. Bank loans could be offered to people who may not be able to repay without any impact on current reported profit: in fact, the upside of fees and charges could be taken immediately and the losses could be deferred. So, not only did profits increase, and with that bonuses, the share price, stock option valuations and so much more that all favoured bankers in the short term, but risk for everyone else rose at the same time. Banks did, effectively, dump their risk on the rest of us.
As a result people were recklessly offered loans.
The economy was over-geared.
Bank shares were over-priced.
Bank balance sheets were over-stated.
A myth that all was well when the opposite was true.
The 2008 crash followed.
That was not all down to bad accounting, but that accounting had a massive impact on it: the fact that the US did the bad accounting first did not help.
It will take until 2019 to put this right. Itself this is a scandal. Tim Bush at PIRC, a former chartered accountant who I have known for many years and have had the pleasure of co-operating with on a number of occasions, is the hero for exposing this risk. He did so most emphatically in 2010 in evidence to parliament. The case he made then was unambiguously right: bank accounts were not true and fair and the consequences for the UK had been catastrophic.
It is clear that they still are. If banks have €61.5 billion of excess assets on their balance sheets (not all in the UK, admittedly) three consequences follow.
First, bank shares, and so stock markets as a whole, are overpriced. There are losses to come.
Second, the banks face major crises in filling these holes in their balance sheets that will test their ability to perform in the economy in the next few years.
Third, there is considerably more stress in the economy than current forecasts suggest likely.
These facts (for that seems to be what they are) then suggest that George Osborne's expectation that he can rebalance the economy on the basis of increased lending to companies and consumers utterly implausible.
This is yet another case where an elite - the accounting and auditing profession acting on behalf of their banking clients - took control of a process that should have been subject to control by the state and abused the power they were given with disastrous consequences. The lessons need to be learned. As I argued yesterday the first lesson is that the state needs to take back control of accounting standard setting. The second is to realise that over0-geaing is not the solution to boosting private sector activity. The third is to anticipate this crisis with an alternative economic plan. You already know what I think that that is, and I am certain it will be needed.
Thanks for reading this post.
You can share this post on social media of your choice by clicking these icons:
You can subscribe to this blog's daily email here.
And if you would like to support this blog you can, here:
The obvious question: if the bread-and-butter business of bank lending is mis-stated in the banks’ accounts, what of their trading and derivatives exposures?
See yesterday’s blog on that
‘True and fair’……………..hmmmm?
If we allow the State to take control of accounting standards then accounts from different countries will no longer be comparable. Different Governments will be elected on different mandates and will have conflicting views and priorities when it comes to accounting?
They are far from compatible now
Prem Sikka and I have done research on this issue
Perhaps – but the intention is there. Giving the power to local governments means you are giving up on that objective.
Sorry: lost you
My point was that IFRS are flawed, but a key goal is surely international comparability?
If we now put Governments in charge then we have no hope of achieving that (unless we have some kind of global State committee to decide on these matters, but that seems a recipe for rows depending on different priorities leading to inaction).
So it’s better to do the wrong thing many times than do the right thing
I don;t think I agree with that
Why would banks be mis-stating their neoliberal derivatives exposures ? Surely they have to mark these to market daily ? These are then shown on the balance sheet at their market value.
There is no market in many of these products
Simulations are used for valuation
And they can be miles out
If only they did have to mark to market! If this had happened in 2008, we’d have lost most if not all of our current crop of zombie banks as it would have been apparent (and impossible to conceal) that they were all utterly bankrupt.
Government to insure deposits; private sector to pick up the scraps and do with them what they will. At least that’s how it should have worked; but unfortunately Gordon gave the banks loads of free money without strings, and the Tories then refused to regulate properly to the surprise of precisely nobody. So here we are.
It’s re-writing history to state that the Tories refused to regulate. Do you think the banks were adequately regulated before 2010? Do you think regulations are higher now or in 2010?
Mark to market? does anyone remember Enron and Arthur Andersen?
Richard,
an interesting blogpost. I wonder to what extent the problem is magnified by the treatment in commercial accounting of write offs and write downs in the profit and loss account (warning – sloppy terminology probably in use here).
In my field, in National Accounts and other statistics standards, like the IMF Government Finance Statistics manual, there is a separation between a group of things referred to as “transactions”, and a further group of things called “other economic flows”.
The taking out of a loan, or issuing a bond is a transaction, as is the associated payments of interest and principal by the debtor to the creditor. However, were the debtor to go bankrupt, or in the case of a householder, die, the write off of the debt is not treated as a transaction, but as an other economic flow. On the other hand cases where a creditor forgives the debt are regarded as a transaction – an imputed payment from creditor to debtor, that the debtor uses to extinguish the debt.
in National Accounts write downs are not recorded at all.
Consequently, for large financial institutions, in statistical measures, write downs and the probabilistic approach do not feature at all, assets remain in their balance sheet recorded as the full amount owed (as individual debtors almost certainly record their debt in full on their balance sheet. The only things that make it into revenue would be interest receipts, and any fees paid. Expenditure only includes costs of running the business (compensation of employees, consumption of fixed capital, intermediate consumption, taxes etc) and write offs, but not write downs.
Transactions in derivatives provide no revenue at all (other than from commissions for arranging said derivative), they are all regarded as financial transactions that take place “below the line”.
Consequently profit (or net lending / borrowing) in a national accounts world is very different to that in commercial accounts, especially for financial corporations.
This is obviously not to say that there is no place in commercial accounting for estimates of the level of bad debt, provisions, or indeed showing an impaired balance sheet – indeed I think this is very sensible for investors – but national accounting does not take this approach.
I wonder whether this approach could be more appropriate for the banking sector? To essentially reduce banking profits to a more simple measure (it would simplify down to something like net interest receipts and commissions minus running costs and write offs)
What is clear is that the statistical compilers and commercial accountants have arrived at very different places when it comes to accounting for large financial corporations – its interesting to here the views of an accountant on the statistical approach.
I gave kept this back hoping to have time to address he issues you raised
I now have book proofs to read
I can’t see that time happening
Sorry
But bang it in again some time as the point is interesting – although I definitely do not agree with it all
A parallel thing going into 2008:
A pension fund buys stock or other tradable instrument (like sub-prime ABS’s).
Their custodian offers them a deal. The custodian will waive custodial fees if the custodian has the right to lend the shares [to short sellers].
The shares are lent to short sellers who sell them–and never take title.
The short sellers put up collateral to cover the loan of the shares.
As the value of the lent stock falls this collateral is returned to the short seller.
When (if) the short seller covers he books a profit.
Since the pension fund has bought but not yet sold the stock it does not book a loss.
In some particularly egregious cases pension funds were buying the same stocks that were being shorted meaning that they were buying shares that they already owned from their own inventory (via the short sellers).
Some shares went to zero which meant that there was no need to cover the short.
“IFRS changed that: what it said was that loan losses should only be booked as they were incurred.”
Well, that’s just not correct. Adoption of IFRS in 2005 allowed use of FRS 102 on impairment which specifically allows that:
“A company is required to assess at each balance sheet date whether there is any objective evidence that a financial asset or group of assets may be impaired.”
FRS 102 set out indicators that the asset holder would need to consider, such as whether the issuer is experiencing significant financial difficulties, whether a default or breach of contract has already occurred, or whether an active market for that asset has disappeared because of financial difficulties.
So impairment was specifically allowed for in anticipation of a loss. An actual loss was, quite simply, not required.
indeed FRS 102 requires:
“where there is objective evidence of impairment, the entity recognises the impairment loss in the profit or loss immediately.”
Not only did a company not have to wait until a loss was actually incurred, it was obliged to recognise an impairment when there was objective evidence.
Quite frankly I’m surprised that someone claiming to be an expert of accounting can get something so wrong.
Respectfully, if that was right IFRS 9 would not be needed
I am entirely confident of my analysis
Debts are provided at present on an incurred and not an anticipated basis
Then your confidence is entirely misplaced.
Paragraph 59 of IAS 39 says (using words which mean what they say and not what you wish they might mean)
“a financial asset or a group of financial assets is impaired and impairment losses are incurred if, and only if, there is objective evidence of impairment as a result of one or more events that occurred after the initial
recognition of the asset…
Objective evidence that a financial asset or group of assets is impaired includes observable data that comes to the attention of the holder of the asset about the following loss events:
(a) significant financial difficulty of the issuer or obligor;
(b) a breach of contract, such as a default or delinquency in interest or principal payments;
(c) the lender, for economic or legal reasons relating to the borrower’s financial difficulty, granting to the borrower a concession that the lender would not otherwise consider;
(d) it becoming probable that the borrower will enter bankruptcy or other financial reorganisation;…”
So (inter alia) significant financial difficulty, default in payment, the granting of concessions or the probability that the borrower will enter bankruptcy would be objective evidence requiring a recognition of impairment. Not actual bankruptcy. Not an incurred loss but an anticipated one.
It’s not even a matter of opinion, it’s there in black and white in the wording of IAS 39.
One has to wonder why, if your interpretation that IAS 39 is so woeful is correct, that IFRS 9 isn’t even compulsory until 2018.
Because banks lobbied so hard against it
Respectfully, you are wasting my time
I suggest you write to the FT and Barclays and tell them they have the issue wrong too
I’m new to your blog and wondered whether you’ve addressed the issue of the accounting treatment of intangible capital? I work for a media company with assets that are almost wholly intangible (brand, goodwill etc). These are comparatively novel concepts developed to address the challenge service companies with no tangible outputs and few tangible assets have in developing balance sheet assets. It seems they have been given increasing scope to count as assets (and consequently collateral for borrowing) the estimated present value of projected future income streams. In fact, most large corporations are based on intangible assets/capital. This has two consequences. The first is that they can grow by manipulating accounting codes/auditing codes to count the non-existent as assets. The second, and more pressing, is that intangible assets, being intangible, can be deemed to exist anywhere, in contrast with tangible assets like factories which have to be in a single place. This explains the rise of tax haven companies why there are now so many. If this argument is sound (I am an economist and struggle to accept intangible property/capital as valid from first principles), then perhaps the starting point for addressing unacceptable corporate tax avoidance needs to be containing the growth of intangible capital. This is a legislative issue and (like your point about incurred income about which I have some professional experience) something that the accounting profession should be pressed about.
There is a fuller critique of the ideas supporting intangible capital from first principles here: http://edmundosullivan.com/economics2030/the-intangible-capital-myth/
Yours Aye
Eddie O’Sullivan
This is an issue I have addressed frequently as part of my (currently fondly recalled) work on tax havens and corporate profit shifting
Much of this might be politely called micky mouse accounting for rogues
Can you point me where I can read it.
There is misses of it – sorry!