There's been a lot written about Merrill Lynch International (MLI), but it seems no one has bothered to actually analyse the accounts that include the $16 billion dollar loss. I have. For your convenience they are here, and those for 2006 here.
I'm not for one minute pretending what follows is a full analysis by the way. And this is emphatically not market information: because this is one company in a group the analysis here is looking at that company in isolation and gives no indication of the future potential returns for the group as a whole. Take this as an exercise in tax and accounting analysis, at best, and a quick one at that.
But let's knock a couple of things on the head. First it's been said that MLI may have been making strong profits in the UK in the past and so claiming losses now is fair quid pro quo. No, that doesn't look right. In 2006 (a good financial year) they reported a loss before tax of $63 million (see 2006 accounts here, the comparative in 2007 is restated). OK, because of tax timing differences tax was due, but cumulatively there was a loss to date. In 2005 there was a profit: $409 million, and tax was certainly paid. But nothing like enough to justify the losses declared now as being fair quid pro quo.
Second, let's be clear, profit variation seems largely dependent upon the volatility of a group service charge, $1,169 million in 2005, £1,909 in 2006 (subsequently restated to $2,360, which is baffling in itself) and then $1,307 in 2007. That should take someone quite a lot of time to justify. The fall might be convenient though, it paves the possibility for future profits to mop up those losses.
Most interesting though is the change in the balance sheet over the three years. Just look at two numbers in each year, the long inventory position (hedge assets) compared to the short inventory position (hedge liabilities) (and yes, I know I am using shorthand). In 2005 the comparison was long $111 billion to short $85 billion. In 2006 it was long $182 billion to short $153 billion. In 2007 the position changed radically. Long was $467 billion (amusingly, not far short of 17% of UK GDP to give this some context) compared to $463 billion short.
It's obvious the margin has gone (and I suspect this is where the provision has come from) but the scale of change is staggering. The 2007 position is more than 2.5 times the 2006 position. Of course this was an exceptional market where liquidity was hard to find. That may be the reason for the change, but I'm curious to note that MLI have not booked all the benefit of the tax advantage they might get on their staggering loss. $3.7 billion of deferred tax asset (almost certainly the future value of the loss if tax credit were to be given) has not been claimed in the accounts. I think they're realistic.
Losses in the current year are probably going to be allowed for group relief (if any such option is available) and maybe carry back to the limited extent allowed but I think MLI are going to have fun and games proving that they haven't changed the nature of their trade as a result of the changed appearance of their balance sheet. And let's be clear, this is very important: losses don't carry forward ad infinitum against any trade. They are only carried forward against future profits from the same trade. The losses are said in the 2007 accounts to have arisen on trades in Super Senior United States Asset Backed Securities Collateralised Debt Obligations. Now I suspect the market in these is not looking too good right now: I wouldn't even be surprised if after the credit crunch the whole trade in anything vaguely like these CDOs might look very different indeed, or non existent.
What's the significance of that? Well, simply that it takes very little change in what you do for loss carry forwards to be disallowed. For example, a company who made losses on making medical equipment using rubber who then switched to using plastic was deemed to have started a new trade and was denied loss carried forward. Tescos way back in the 1970s (for those with long memories) stopped giving trading discount stamps and instead cut the price of products. That was deemed to be the end of one trade and the start of a new one. Losses did not carry forward.
What's the chance MLI will ever use all these losses in that case given the simple volatility of the volume and type of this trade? Very little if there's anything like a good tax inspector on the case in my opinion. Which is pretty good news, I think. And makes MLI's accounting look pretty realistic. I wouldn't be banking on seeing that $3.7 billion is I was them.
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:
Richard,
These claims (from TR-UK) seem highly dubious (at best). What if a firm changes from pencils to pens, XP to Vista, Apple to PC? These are akin to TR-UK’s vague example of the medical equipment company and their rubber rings.
Do you have a link to that particular instance so the reader can ascertain the true facts?
JG Ingram & Sons Ltd v Callaghan (1968) 45 TC 151
Credit (no pun intended) where it is due. I am not often in agreement with RM’s approach, but the point about the denial of losses on a change of trade is a well made observation. I have no first hand experience of investment banking taxation but, applying general principles, it is hard to envisage how a sufficient change in trade will not now have taken place in the nature of the instruments dealt with etc. And, if not, does it mean that not one at MLI has seen fit to question whether it might be a good idea to change tack.
What presumably follows is whether the preservation (or otherwise) of the tax losses then becomes a bargaining tool in the UK, with MLI using any threat of their denial as a reason to reduce their UK activities (with the usual arguments about loss of UK investment, jobs, spending etc.)
Readers may be interested to know that, for a time at the end of the 80s, I was senior tax manager for a Big 4 firm with responsibility for tax advice to Merrill Lynch’s UK operations. It consumed the lion’s share of my time. I imagine the ML business and the entities have changed a lot since then.
Client confidentiality would prevent me from commenting in detail on current matters, but I will just say that the UK and worldwide tax affairs of ML are complex to say the least, as you might expect. The issue of where trades, intra group items, assets and liabilities etc should most appropriately be booked is a perennial problem for all worldwide financial businesses.
Richard,
Putting aside the citation is from a 1968 Australian ruling….
Are you saying that ML fails the “same business test”? “new business test”? “new transaction test”? “changed their “essential character”?
I think I rather agree with Lord Donovan who observed in J G Ingram & Son Ltd v. Callaghan:
“I doubt if one can as a rule segregate the various activities involved in carrying on a trade, select one of them as being of the essence, and then designate the one selected as the real trade. There is, I think, an organic unity about a trade which invalidates this sort of dissection, and I think Rowlatt J was saying much the same thing, though more incisively, when he remarked in Graham v. Greene that a trade differs from the individual acts which go to make it up, just as a bundle differs from odd sticks.”
Wouldn’t you?
Georges
I presume you will argue with lamp posts
I gave Andrew Brooks comments rather more weight: he is a fellow tax professional
If you think tax professionals waste their time looking at this issue day in, day out, please carry on
The reality is that the body of case law on this issue is large, and stacked against the taxpayer
You’re at liberty to disagree, as any client would be, but I make my case with good reason: my professional judgment is that MLI have a problem.
Richard
Richard, that seems a fair summary.
As I explained in reply to a query over at mine, it is a highly risky strategy (possible attack re change in nature of trade, also, the IRS won’t take kindly to ML transferring future US profits over to the UK to mop up the loss, so ML might end up paying tax on future profits twice!).
What is all the more chucklesome is that the humugous write down of $16 bn is only 3% of the book value of ‘hedge assets’, so they’ve a long way to go…