News: At $72.8 Trillion, Presenting The Bank With The Biggest Derivative Exposure In The World (Hint: Not JPMorgan) | Zero Hedge

30 April 2013 | Bill Hodgson

Yet another news article scaring the pants off ordinary people by suggesting that Notional = Value, and that Collateral = Zero. It's a shame, the comments show how taken-in people are when the underlying facts aren't presented properly. See the other side of the story on this site here (How Big is OTC Really). At $72.8 Trillion, Presenting The Bank With The Biggest Derivative Exposure In The World (Hint: Not JPMorgan) | Zero Hedge.


Thanks BIll. I'm getting tired of this. Especially in this case because:1. The more meaningful figure (i.e. positive market value of EUR 776bn) was 1 cm away on the page from the one they used.2. A reasonable reaction to the bar graph is "are these two things comparable?" and the answer is no!I've registered for Zero Hedge and you should too. We should comment.

It's like killing zombies, the commenters will never be happy. I tried it on the Naked Capitalism post of a similar nature and had to engage with comments which seemed to be from a parallel universe.

In US amusement arcades they have a game called whack-a-mole - see video.... http://www.youtube.com/watch?v=9Lvv7TX2GycIs there a better way to tackle this? (I doubt congress would pass a law...)

I just read through your naked capitalism comment chain. Sorry but I found it quite funny. It reminded me of this scene from Monty Python and The Holy Grail:http://www.youtube.com/watch?v=ekO3Z3XWa0QMaybe we need another simple post that explains current and future exposure with some numbers both gross and after netting and collateral (all from published statements).

Does N = NP? Why do monkeys open bananas at the other end? Why won't people believe notional isn't the problem? ;-)

This point about notional being used for scaremongering is a good one - it might just be an annoyance, except that the European FTT is supposed to be calculated based on notional. My analogy is with insurance - your travel insurance might have a premium of 50 / year, and offer maximum cover of 3m if you get, e.g., charged with a crime in the US. They propose to tax the 3m.

That's a nice start to a blog post on FTT, fancy writing it?

yup - will do... getting quite interested in FTT actually. give me a few days!

The insurance analogy is somewhat useful. Except that at least in the insurance case the 3m is the worst case loss whereas even that is not he case with swaps. An FTT for derivatives could only make sense if it were a tax on revenue or risk rather than notionals (but of course how to calculate revenue and risk is non-trivial).Judging by what I've read derivatives are the least of FTT's problems though. The most straightforward article I've read is from the Economist: http://www.economist.com/news/leaders/21572205-plans-transactions-tax-ou.... Aside from derivatives, the 0.1% tax on repos per day equates to about 22-25% per year. As applied to the repo notional amount this would mean it would be cheaper for banks to issue equity capital to fund their bonds. In other words banks would withdraw from onshore EU repo and therefore government bond trading overnight - withdrawing nearly all the market making capacity from both primary and secondary government bond markets. That would make 2008 and the current EU sovereign debt crisis look like a picnic lunch.It would be useful to confirm the Economist's interpretation of 0.1% per day. Though even if they changed the 0.1% to be an annualized rate the repo market would be seriously impacted leading to withdrawal by some banks.

The insurance example relates to CDS not IRS, but I think that it does work for CDS - if you sell $100m of protection on the iTraxx main index, then you could be required to settle $100m, if every name in the index defaults and recovers at zero. (i.e. if the world ends)I agree that taxing IRS based on notional is very hard to make sense ofthe point impact of FTT on repo markets is covered in some detail in a report produced by ICMA and the European Repo Council, here: http://www.icmagroup.org/Regulatory-Policy-and-Market-Practice/short-ter... makes fairly scary reading

If the purpose of FTT is to have banks pay on an accrual basis for the historic or future costs to the taxpayer incurred in crises, it would be a good start to create an estimate of this cost. Anyone have any numbers?The reason FTT is a flawed premise for derivatives and repo is that taxing transactions is based on the key premise that transactions are an exchange of an asset for cash. Derivatives and repos do not fit that model! Repos in particular appear to fit but in fact are a cash loan secured by a pledge of securities. I presume for political reasons unsecured loans are excluded from FTT to avoid disincentivizing lending to the real economy. In truth repos also contribute liquidity to the real economy - perhaps they should be exempted for the same reason. Derivatives don't fit the model at all e.g. a spot starting on market IRS involves negligible exchange of income or variation margin on trade date + 1 though initial margin will be put up on T+1 once clearing and bilateral margin regulations are in place. What is true though is that repo, derivatives along with cash securities and FX transactions influence the liabilities (both overnight, short term, term debt and equity issuance) of a bank. This suggests that a better place to start than taxing transactions would be to tax liabilities in the manner of the UK bank levy (which by the way already exists).