One of the solutions proposed to the “derivatives problem” since the crisis was to move them to exchanges. It is still something that surfaces now and then, but less and less often.
I will admit that I disliked this solution from the start, for a number of reasons. The ones that still remain are:
- some products can change value in jumps – such as the infamous CDS contracts. Margining them becomes very hard;
- it moves the systemic risk to exchanges; which then become the next Too Big To Fail (TBTF) category – surely, we don’t want more of those?
- it makes it much more expensive to the commercial users (by that I mean people hedging real exposures). While it removes one risk (counterparty exposure) – bar the above – it creates other risk, namely liquidity. Liquidity is something that can kill any company, in any industry, startup as well as hundreds of years old company, well run and badly run (although more likely that one). I don’t believe trading say FX risk for liquidity risk makes much sense.
Now, the noises about moving to exchanges seem to be dying, let unreason rule, and let’s suggest something off the wall:
Move everything to a single exchange!
Doesn’t sound off the wall?
Then how about making the central bank the exchange. Simply mandate that the only counterparty for an OTC contract can be the central bank, and make all other derivatives (which are not traded on other recognized exchanges) illegal.
The requirement would be that the CB has to accept the contract when offset by equal-and-opposite with other party, but also that the CB can set the collateral requirements.
That would immediately produce a couple of results. For a starter, it would remove the single-point of failure problem, and make a failure resolution somewhat easier, without the TBTF problem.
It would provide a central repository of all contracts, letting the CB see what’s happening – should it wish to use the knowledge. It would let them see even the funny and complicated contracts – and guess what, if they don’t understand it, they could (if they dare), require a very high collateral from the cpty (I can see BoE doing it, I doubt Fed would as it is at the moment). I can hear the screaming even now.
In effect, the collateral requirements would mean that
- the CB would be the calculation agent (no more different values of the same contract for different parties…)
- they could regulate the markets dynamically, as they wish using more or less collateral – and to a large extent know what’s going to happen.
Now, if the transaction is between two financial (i.e. not commercial ) entities, I would expect that the side with negative value of the contract would be posting collateral to the CB, but the one with positive value would not – simply because the CB can make good on the contract, always – although it may be not in the currency of the contract but in CB’s currency (in an ideal world, all CBs would act as exchanges for the derivatives in their own currency. Still leaves out cross-currency ones, but let’s not go too much into details now).
The problem occurs when the party with negative value is a commercial party (i.e. a real hedger – and I’d be happy to leave the decision on that to the CB). As I said earlier, we don’t want them to post collateral (or at least not for the full value).
Well, that may mean the intermediation in this case would add a clause between the CB and a financial institution. When the contract between FI and the CB would be terminated on a commercial party default, the FI would receive any collateral the CB collected (if it decided to collect any) and rights to the remaining debt in bankruptcy proceedings. This could be done on a netting basis rather than per-deal, which would simplify things further (possibly not ideal from bankruptcy proceedings, but that’s how it often works now anyway). Any suggestions to improve on this would be of course welcome.
One thing I’d also like to see if this ever went live would be that after a contract is executed, each and every one is published (with the identyfying information removed). That’s trivial for vanillas, and would encourage slightly more sane structured contracts than we can see now.
Now, one large objection to this could be that they could regulate as much – or as little as they wish. Yep, that’s a feature, not a bug. It would make it their problem very explicitly, with no out. They would find it very hard to say “but we didn’t know/understand what was going on…”. If the information would be also published as I suggest (remember, normal exchanges quite routinely publish volumes, aggregate position sizes and the like), bloggers and economists would have a field day and could keep pretty close tab on the CB’s doing as well.
It could still ultimately fail ala Greenspan and equity markets, but I will point out that that bust was – comparatively speaking – minor, and no regulation ever will stop human stupidity.
Note: This idea is still in its infancy, and I may edit the post or even discard the idea. But then, it does say “unreasonable” in the title – you have been warned.