Streetwise Professor

October 10, 2015

Igor Gensler Helps the Wicked Witch of the West Wing Create Son of Frankendodd

Hillary Clinton has announced her program to reform Wall Street. Again.

The actual author of the plan is said to be my old buddy, GiGi: Gary Gensler.

Gensler, if you will recall, was the Igor to Dr. Frankendodd, the loyal assistant who did the hard work to bring the monster to life. Now he is teaming with the Wicked Witch of the West Wing to create Son of Frankendodd.

There are a few reasonable things in the proposal. A risk charge on bigger, more complex institutions makes sense, although the details are devilish.

But for the most part, it is ill-conceived, as one would expect from Gensler.

For instance, it proposes regulating haircuts on repo loans. As I said frequently in the 2009-2010 period, attempting to impose these sorts of requirements on heterogeneous transactions is a form of price control that will lead some risks to be underpriced and some risks to be overpriced. This will create distorted incentives that are likely to increase risks and misallocations, rather than reduce them.

A tax on HFT has received the most attention:

The growth of high-frequency trading (HFT) has unnecessarily burdened our markets and enabled unfair and abusive trading strategies that often capitalize on a “two-tiered” market structure with obsolete rules. That’s why Clinton would impose a tax targeted specifically at harmful HFT. In particular, the tax would hit HFT strategies involving excessive levels of order cancellations, which make our markets less stable and less fair.

This is completely wrongheaded. HFT has not “burdened” our markets. It has been a form of creative destruction that has made traditional intermediaries obsolete, and in so doing has dramatically reduced trading costs. Yes, a baroque market structure in equities has created opportunities for rent seeking by HFT firms, but that was created by regulations, RegNMS in particular. So why not fix the rules (which in Hillary and Gensler acknowledge are problematic) rather than kneecap those who are responding to the incentives the rules create?

Furthermore, the particular remedy proposed here is completely idiotic. “Excessive levels of order cancellations.” Just who is capable of determining what is “excessive”? Furthermore, the ability to cancel orders rapidly is exactly what allows HFT to supply liquidity cheaply, because it limits their vulnerability to adverse selection. High rates of order cancellation are a feature, not a bug, in market making.

It is particularly ironic that Hillary pitches this as a matter of protecting “everyday investors.” FFS, “everyday investors” trading in small quantities are the ones who have gained most from the HFT-caused narrowing of bid-ask spreads.

Hillary also targets dark pools, another target of popular ignorance. Dark pools reduce trading costs for institutional investors, many of whom are investing the money of “everyday” people.

The proposal also gives Gensler an opportunity to ride one of his hobby horses, the Swaps Pushout Rule. This is another inane idea that is completely at odds with its purported purpose. It breaks netting sets and if anything makes the financial system more complex, and certainly makes financial institutions more complex. It also discriminates against commodities and increases the costs of managing commodity price risk.

The most bizarre part of the proposal would require financial institutions to demonstrate to regulators that they can be managed effectively.

Require firms that are too large and too risky to be managed effectively to reorganize, downsize, or break apart. The complexity and scope of many of the largest financial institutions can create risks for our economy by increasing both the likelihood that firms will fail and the economic damage that such failures can cause.[xiv] That’s why, as President, Clinton would pursue legislation that enhances regulators’ authorities under Dodd-Frank to ensure that no financial institution is too large and too risky to manage. Large financial firms would need to demonstrate to regulators that they can be managed effectively, with appropriate accountability across all of their activities. If firms can’t be managed effectively, regulators would have the explicit statutory authorization to require that they reorganize, downsize, or break apart. And Clinton would appoint regulators who would use both these new authorities and the substantial authorities they already have to hold firms accountable.

Just how would you demonstrate this? What would be the criteria? Why should we believe that regulators have the knowledge or expertise to make these judgments?

I have a Modest Proposal of my own. How about a rule that requires legislators and regulators to demonstrate that they have the competence to manage entire sectors of the economy, and in particular, have the competence to understand, let alone manage, an extraordinarily complex emergent order like the financial system? If some firms are too complex to manage, isn’t an ecosystem consisting of many such firms interacting in highly non-linear ways exponentially more complex to control, especially through the cumbersome process of legislation and regulation? Shouldn’t regulators demonstrate they are up to the task?

But of course Gensler and his ilk believe that they are somehow superior to those who manage financial firms. They are oblivious to the Knowledge Problem, and can see the speck in every banker’s eye, but don’t notice the log in their own.

People like Gensler and Hillary, who are so hubristic to presume that they can design and regulate the complex financial system, are by far the biggest systemic risk. Frankendodd was bad enough, but Son of Frankendodd looks to be an even worse horror show, and is almost guaranteed to be so if Gensler is the one in charge, as he clearly aims to be.

Print Friendly, PDF & Email


  1. Do you think the fact that HFT firms have zero losing days might indicate that they are predatory? Do you know if the most brilliant trader in the world has half as good a record? Do you think if the HFTs truly provided liquidity you would have frequent days where the market falls out of bed (ETFs trading at 30% discount)? Do you not think the HFTs are opportunistic and front-run orders and information? Are HFT’s one of your customers?
    Please provide full disclosure. Thank you for all the other excellent stuff you write

    Comment by thoreau_devotee — October 11, 2015 @ 7:59 am

  2. @thoreau-No, I don’t. They exploit the law of large numbers. Even a slight edge (e.g., profiting on 50.1 pct of trades) can lead to a microscopic probability of losing money in a day, when they place thousands of such trades during the day.

    HFTs do not “front run” in the legal sense. They do extract information from order flow, and make money on that. That’s what traders do. Old school market makers did that too.

    HFTs are sometimes opportunistic. I’ve written about that extensively: look at my posts from the March-April 2014 time period in particular. My point is, and always has been, that HFT is not a uniform thing. It has good, bad, and ugly. I have also consistently made the point that meat ax approaches like those Hillary/GiGi are advocating will do more harm to the good, than to the bad or ugly.

    And no, I don’t have any HFT firms as clients.

    The ProfessorComment by The Professor — October 11, 2015 @ 11:01 am

  3. so I use the rand() function 10K times and use a countif() “> .501” (per example you had). and with 10K trades a day, assuming that one has a .501 advantage, results in a loss very very often.. of course if I bump it up to 1/2 a million trades, one mostly comes up ahead. personally, I hate putting an order and immediately seeing the NBBO move against me. hate seeing that I am only filled when the market moves against me. hate seeing the HFTs use my order as a bookend and get the fill by paying .0001. you are right that a lot of blame has to be put on the door of regNMS – but one would have to be naïve if one doesn’t see the handprint of the HFTs in getting regNMS passed.. the SEC has been pretty feckless so far. let’s see what they can do to create a fairer marketplace.. I am not surprised to see retail volume declining.. the HFTs greed has killed the golden goose.

    Comment by thoreau_devotee — October 11, 2015 @ 5:15 pm

  4. Prof- you have ruined my day. I thought HRC as prez was bad enough, but GiGi is too much to bear. The cold dark winter of Naria will prevail.

    Comment by Cypriot — October 12, 2015 @ 4:30 am

  5. Professor, any ideas on what a fix of Reg NMS would look like? Isn’t the fundamental problem (and I think there is one, though knowing what can or should be done about it is another thing entirely) that in the contemporary world of trading there are many venues and the NBBO cannot really function the way people envisioned it would when they instituted it back in 2005? That is, there will always be gaps in time between the sending of messages out from the exchanges (or more accurately their computer servers) and when the prices are collated into the NBBO by the SIP? And HFT will exploit those gaps in small but real ways even though they have often less than 100 milliseconds to do so? Were people aware in 2005 that there would be substantial amounts of money to be made in these gaps? Financial sociologist Donald MacKenzie calls Reg NMS a Newtonian structure operating in an Einsteinian world.

    Maybe a bigger question behind all this is, isn’t there a need for just one venue for stock trading? Reg NMS went a few steps in that direction by instituting the order protection rule and the NBBO, but it also wanted to foster competition between trading venues. But with the growth of HFT, arguably abusive practices can thrive in the timing gaps between markets and the SIP.

    Another question: how much of the reduction in spreads is due to HFT and how much is due to decimalization? And also simply the application of IT to trading. HFT defenders like to point out that HFT has improved benefits for most traders, but isn’t it more accurate to say that it is an element in a handful of different causes of reduced trading costs?

    Comment by Steve — October 12, 2015 @ 7:23 am

  6. @Thoreau_Devotee: CNBC’s Bob Pisani had an excellent analysis of Virtu’s S-1 last year that addresses your question. The professor is right that successful firms like Virtu depend on their risk management capability and the law of large numbers to implement a sustainable, scalable, and profitable business model. Here is Pisani’s 2014 post on the topic:

    Comment by Scooter — October 12, 2015 @ 8:48 am

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress