Too Complicated to Work

Yves Smith has a long excerpt from testimony by Robert Johnson before the House Financial Services Committee on regulation of OTC derivatives. (Johnson’s testimony is not up at the committee site.) Johnson brings together the issues of too big to fail and derivatives regulation: “Absent a drastic simplification of derivative exposures and a transparent and comprehensive improvement in the monitoring of those positions when imbedded in large firms, complex derivatives render these behemoth institutions Too Difficult to Resolve (TDTR).”

In short, he argues that even if you give regulators the ability to “resolve” a Tier 1 financial institution in the event of a crisis, regulators will be afraid to pull the trigger as long as there is still this complicated web of non-standardized derivatives linking it to the rest of the financial system. In addition, this creates a bizarre incentive: if you think that you can escape being shut down by having an intimidatingly complex derivatives portfolio, then you will go out and create such a portfolio.

I think there is a real risk in financial regulation that it becomes too technical and technocratic. We already know that the people who are into finance, economics, and economic policy are suckers for complicated models — the new flavors are contingent capital and size-based capital requirements. But we have to weigh their theoretical elegance against the chances that they will not work, or will be overridden by other considerations, in a crisis. For example, here’s Mervyn King on contingent capital:

“[E]xperience has shown that it is difficult to assess risks of infrequent but high-impact events, and so it is dangerous to allow activities characterised by such risks to contaminate the essential – or utility – services that the banking sector provides to the wider economy. Both of these drawbacks mean that it is almost impossible to calculate how much contingent capital would be appropriate.”

In other words, since crises are by definition tail events, there is no good way to estimate the amount of contingent capital that will be needed beforehand, so there is a high risk that financial institutions won’t have enough and we’ll be left where we are today. Assuming perfect regulation is no different from assuming any other can opener.

By James Kwak

17 responses to “Too Complicated to Work

  1. If you allow firms the luxury of dealing in the technical and technocratic, don’t you think the same courtesy should be extended to their regulators?
    And I challenge the idea that there’s no way to predict how much contingent capital will be needed. The insurance industry does this all the time, as well as the reinsurance industry. One doesn’t need perfect regulation; any regulation at all would be a decent start.

  2. Also, I want to emphasize that 99% of the events we are talking about are man-made. They can happen as often or infrequently as we allow them to. It’s not like a “financial tsunami” is some kind of act of god.

  3. the people who are into finance, economics, and economic policy are suckers for complicated models

    IMO the real problem is that top financial policy makers are suckers for “the market knows best” belief. To appreciate how messed up this situation is, just imagine if FCC or FDA of FAA suffered from the same reverence to the industries they regulate! Yet, this is exactly what the Fed and Treasury have been preaching for years and under varying administrations.

    It is not the inability to regulate that is a challenge, but the lack of desire to regulate on the part of those who are in charge of regulation.

  4. reality isn’t simple. intermediaries can only do their jobs well if they reflect that.

  5. > And I challenge the idea that there’s no way
    > to predict how much contingent capital will be needed

    I agree. At the very least, 100% is probably a number that can be shown to be valid.

    If CDS contracts could only be issued if the counterparty had a valid interest in the underlying asset, then these arrangements would be much less complex.

  6. “The market knows best, except when it prices things cheaper then we think they should be. In such cases, our abstract models know best!”

  7. Yes, regulation cannot fix problems like these.

    1. Regulation by its very nature has to assume basically decent, social actors who just get stupid or carried away and make mistakes.

    But that’s not what we have with today’s bankers. They are, on the contrary, nothing but antisocial gangsters. There is zero common ground between them and the public interest. The null set.

    2. Regulation would therefore be a permanent war of attrition, which the regulators, having far less resources and unstable professional/ideological will (subject to the vagaries of politics, change on administrations etc.).

    3. Even if (1) and (2) weren’t operative, the problem is still too complex. As we’ve seen with this Smith post and her previous one she was following up on, even given her desire to “regulate” she has to throw up her hands and say it can’t be done.

    Put these three together and the thing is clear: we can’t regulate any of this, we simply have to get rid of it.

  8. I agree with you 100%. Derivitive are a way to get out of ownership and usually because they already made their killing and need to destroy the evidence. Lets face it, the only money being made in the US now is from gambling with risk and the only people pulling ahead are the crooks fueling this crap. They need taxed at he 90+% rate and trades need recovery fees added to them. Bet that would slow them down a little and hey, if the want to leave Wall Street. bye bye. We’ll be back later with the warrants for your arrest.

  9. We don’t need regulator as much was we need cops with chops.

  10. Unless you are willing to argue against derivatives as a whole (this is a separate discussion), 100% is not practical. I tend to agree with the author, here. The insurance industry manages to do this via risk sharing and the relatively well understood correlations between various types of events they are insuring. With many financial instruments, especially credit linked ones, correlation varies widely with time and is extremely difficult to predict. Also, the insurance industry doesn’t really deal with arbitrage (since the events can’t really be changed by human intervention). In this context, expectation pricing works!

    Re: your CDS point– although this is tempting, what constitutes a valid interest? Suppose you are trying to hedge systematic credit risk using an index (CDX). Does this constitute valid interest in the underlying? If I hold debt, paper or equity, does buying CDS to hedge constitute valid interest? It seems it would, but at this point there isn’t really any direct tie to the firm from a non-investment perspective.

    Yes, some people buy CDS to speculate, but a lot of CDS is used for hedging purposes. Also, CDS makes a lot more sense if it’s relatively liquid. A lot of liquidity is driven by buy side index arb, which transfers risk to the right places (i.e. away from the tax-payer) but has many benefits to anyone using the underlying for hedging.

    It’s just not so cut and dried.

  11. nice can opener reference.

  12. There is one way to call a halt to the derivative issue. Simply outlaw them, and give all firms who own them, or have rights (e.g. CDS’s), a year to clean them off their books. You might suddenly see a scurrying to make treaties with counterparties and actually own a transparent risk. Of course this won’t happen, but it might be possible to reach a situation where each derivative must be clearly defined, as well as its risk profile, and that there must be sufficient capital (real assets) assigned to protect from potential losses. You would need to put teeth in this (if auditors determine that the risks are substatially underestimated there would be criminal penalties attached to all involved in the valuations (can you say rating agencies?).

    The way things are, this problem, which is the real issue in re-regulation, will not go away, and will lurk until it breaks the economy further. They say that 2010 will be the year of the foreclosure, including lots of commercial loans, so the assets become more toxic by the day.

    Time for the Fed to remove all guarantee support from the TBTF’s.

  13. as it is from the LRB http://www.lrb.co.uk/v31/n20/brom01_.html
    I guess the description of the anti-Obama-istas might be a bit exaggerated but even if so it eerily resembles statements on this blog.
    Is the circle of “right” and “left” closing? are the twain about to meet?

  14. OK, help me understand if I have this right.

    1. Derivatives are necessary because they provide liquidity, risk sharing, blah blah blah.

    2. We can’t monitor (hence regulate) modern derivatives because they are too complex and no one can figure out the impact of failure scenarios.

    How can these statements both be true at the same time?

  15. So, the committee has created new incentives for banks to take greater risks funded by the taxpayer. Nice.

    How do I become a big bank?

  16. Next Step?
    It may be helpful if we were to step back and define what are the outcomes we believe are required. Then define the issues blocking achievment of those outcomes.

    I suggest we need to be prescriptive in generating a starter set document else apologists seeking to maintain an actor set advantage will seek to derail any fix effort. This is not to imply that other folk have not prepared similar documents so plagiarising (with permission) should be OK. alternatively access such completed document and critique same?

    This document is then followed (Part 2) by a linked solution set and action plan for consideration. OK is seems so simple but we appear to be nationally spinning wheels while such stoppers as ‘TBTF, can’t be done, will destroy the economy’ are continually blocking advance. Note that such negatives seem to be unsubstantiated opinions!

    I suggest comments to date are helpful but don’t constitute a tool set for the President to consider. If he rejects this suggestion at least it can be placed in the public arena and discussion encouraged to add pressure for acceptance and action?

    We have to have a strategy with the tactics to achieve our goals. A finite document can be campaigned against doomsayers, contrarians and crooks.

    May I suggest a distillation of blog contributions to date will probably provide needed data.

    My required outcomes list thoughts are:
    1. Recovery of public funds and reasonable rent.
    2. Cease further public funds deployment.
    3. True disclosure of all matters such as asset valuation (MtM?), dark pool trades in real time view of all investors.
    4. No deals without real interest (derivatives gambling a no-no)
    5. Derivatives to be subject to regulation to extent other assets are traded.

    This list is not prioritised and is an offered as a suggestion only.

    Any comments?

  17. These banks couldn’t get anywhere with all this without congress. Check this link out:

    http://americaspeaksink.com/2009/10/charlie-crist-and-the-film-outrage/