Brin: The Oligarchy vs the Free Market

In a fascinating and compelling essay, David Brin argues that the existing oligarchy defends its privileges at the expense of the marketplace, not in its defense.

Comments:

  1. I'm beginning to wonder why you find David Brin so fascinating on these topics. If you want something more down-to-earth (there's gotta be a pun in there) on this issue, you might check out any of several recent books by David Cay Johnston.

    • I actually think about many things in a manner very similar to Brin's.

      It's a shortcut - sometimes he says things I've been meaning to say and I can just point you at them. He usually manages some angles that I haven't thought of, which is even better. If you disagree with him in this case, I'd like to know why. Then we could have a useful discussion.

      Similarly, who is Johnston and why should we care? (Book reviews are welcome.)

      • You can find out plenty about Johnston with an internet search, but to quote someone else:

        Johnston is the author of best-selling books on tax and economic policy. Free Lunch: How the Wealthiest Americans Enrich Themselves at Government Expense and Stick You With The Bill, is about hidden subsidies, rigged markets, and corporate socialism. It follows his earlier book Perfectly Legal: The Covert Campaign to Rig Our Tax System to Benefit the Super Rich—and Cheat Everybody Else, a New York Times bestseller on the U.S. tax system that won the Investigative Reporters and Editors 2003 Book of the Year award. The Fine Print: How Big companies Use "Plain English" to Rob You Blind was released in September 2012.

        Johnston has made it very clear that businesses do not like actual markets as competition gets in the way of profit. I don't think I disagree with Brin in this case, at least as to his main points, but his delivery tends to strike me as portentous bordering on pompous, as if no one ever thought of any of this before. (And just to clarify a bit, I think I still have my copies of Startide Rising, The Uplift War, and The Postman around here somewhere - I'm presuming you know who wrote those.)

        I have no problem at all with shortcuts, but relying on a science-fiction writer/physics prof for commentary on economics and power politics may turn out OK in a particular instance, but I doubt it's the best way to do this as a general rule. Kind of like talking to a weather forecaster about climate, I just think there are better sources for this sort of thing.

  2. "The Oligoracle"
    -- by Horatio Algeranon

    I am an Oligoracle,
    The wisest of the wise.
    I'm really technoloracle
    My brain is oversize.

    I earn more in an hour
    Than you do in a year
    No reason to be dour
    I'm worth it, I've no peer.

  3. I think Brin's logic is mostly broken. the bit about CEO pay is silly, and the bit about HFT makes no sense at all to me.

    However, if you're saying that oligarchy is bad and a free market is better then I'm happy to agree. But I'd also say that excessive regulation is part of oligarchy, and we've disagreed in the past about the virtues of regulation.

    • The bit about CEO pay is entirely compelling to me. To summarize: many people could do those jobs as well or better, given a chance. That ought therefore to drive the pay down. But it doesn't. Therefore the game is fixed; marketplace rules do not apply.

      A few genuinely high-value executives like Steve Jobs notwithstanding, one doesn't get the impression that high paid executives have any remarkable talents. Like chess players, it is as important that they have few exploitable weaknesses, but looking at some of the individuals involved it is hard to believe even that.

      The bit about a transaction tax? Well, we have a destabilizing financial marketplace. Let me quote from something that happened across my inbox this morning:

      ===
      Do you want to join a team of developers who are trying to solve hard problems in the financial industry? We provide ***** with next-generation trading technology that enables traders, quants, research analysts and other developers to trade in the ultra-competitive high-frequency trading space.
      What are some examples of things that we work on?

      - We developed a low-latency execution platform that allows algorithms to trade futures, options and equities across all U.S. exchanges
      - We built market data libraries and services that disseminate millions of updates per second across the firm
      - We are building our own big-data analytics platform that gives quants and traders powerful insights into how their algorithms interact with the market
      - We have helped developers, traders and quants across the firm to design ultra-low latency trading frameworks. We’ve improved their execution speed by more than two orders of magnitude.
      - We built monitoring tools with HTML5 and websockets to provide historical and real-time statistics about our trading platform

      Skills and qualifications

      We are looking for someone with the following skillset:
      • A strong background in data structures, algorithms, and object-oriented programming in C++
      • Familiarity with the new C++11 standard
      • Experience with high-performance, lock-free containers and data structures
      • Strong understanding of the network layer, in particular TCP and UDP
      • Solid grasp of concurrency and parallel processing
      • Fair knowledge of python and or other scripting languages

      Things that would get us really excited about you

      Things that would really peak our interest:
      • Experience with kernel bypass
      • Experience working or designing big data storage solutions
      • Experience working with or designing simulation/analytics frameworks
      • Prior experience contributing to the design or implementation of a high-frequency trading platform is a big plus
      • Familiarity with machine-learning concepts
      ===

      Now the person who posted this is one of the strongest programmers of my acquaintance, which includes many very strong programmers. He really is (unlike, most likely, his CEO) something of a genius. But should we be diverting our best talents to "kernel bypass" in the pursuit of millisecond advantages on trade? How does this affect valuations and value transfers to the middle class shareholder, for whom the stock market is the modern replacement for a savings account? How does it affect the entire organization of society when transactions of this type dominate the marketplace for real goods?

      And how do these faster-than-thought algorithmic traders interact to affect the stability of the market?

      I know enough about machine-learning to be terrified about machine-learning being trained on the behavior of other machines. Capital is close enough to fiction without this absurd velocity of trade. I would prefer this whole line of work, which extracts wealth without adding value, taxed out of existence.

      • While we're on that subject: I'd love to know more about the actual nitty gritty of these finance structures. Not necessarily even the algorithms, but more things like: what framework has to be in place to allow a machine to make a "real" transaction? I code Java: how do I set up some code to deal in real (virtual) sponduliks?

        Or, more broadly, what were the legal (and less legal / algorithmic) steps that have had to happen, to get us to the point where the majority of capital is now meshed into high finance machine structures that no-one can possibly hope to understand, let alone control?

        And, yes, by God, it's an utterly insane way to run an economy. It's not, actually, running an economy, in fact, is it? It's a parallel parasitic system - still driven by the basic profit drive, but so much more massive than the thing it's parasitising and so easily able to give the host a heart attack, as it did in the crisis.

  4. > To summarize: many people could do those jobs as well or better, given a chance. That ought therefore to drive the pay down. But it doesn’t. Therefore the game is fixed; marketplace rules do not apply.

    I don't find that convincing. Firstly, "could do those jobs as well or better" is just an assertion. I'm dubious that you, or I, have much idea what CEOs actually do. [FWIW my own opinion is that CEO pay goes up because its only a small fraction of the total expense, but a good/bad CEO difference can lose you enormous amounts off your share price. So there's a tendency to overpay, just in case. The same cheeseparing that might apply to an individual engineer - one out of many- doesn't apply.] If you want to believe this is a market failure you need something much better.

    > we have a destabilizing financial marketplace.

    Do we really? And HFT is part of that, is it? That, too, requires evidence.

    > Let me quote

    I'm baffled. You appear not to have noticed that nothing in your quote supports your assertion.

    > should we be diverting our best talents to

    A fair question. Not relevant to the "destabilising" part, of course. It is directly linked to the efficient marketplace in jobs that I think you were favouring earlier.

    > How does this affect valuations and value transfers to the middle class shareholder, for whom the stock market is the modern replacement for a savings account?

    Oh, I can answer that one: it makes things better for them. All this HFT is looking for tiny anomalies and smoothing them down.

    > transactions of this type dominate the marketplace for real goods?

    They don't (I'm overselling my knowledge at this point, of course. But I think I'm right. And I've offered just as much evidence as you have).

    • http://www.washingtonpost.com/blogs/wonkblog/files/2013/09/US_FNSHR0710.gif

      Exactly how is this "smoothing of tiny anomalies" accounting for more than 25% of all profits, then? And what is behind the recent growth in this proportion?

      • Hold on, I thought we were talking about HFT. Now you've suddenly broadened it to the entire US financials market.

  5. ""

    Slightly out of date but:

    Currently, of the largest companies in America (those in the S&P 100), CEO pay has no correlation with either performance or market capitalisation.

    That doesn't necessarily contradict your point, W: it's possible that a small pool of CEOs are all more qualified than anyone else outside of that pool, but within it no correlation shows up. On the other hand, this does tend to suggest a rather weak relationship (if any) between actual profit-making talent (which is what they're hired for) and pay.

    • Also out of date:

      Modern industry seems to be inefficient to a degree that surpasses one's ordinary powers of imagination. Its inefficiency therefore remains unnoticed.

      -- E.F. Schumacher, Small is Beautiful (1973)

      • And the temptation is to observe this, and think "oh, so a command economy would be more efficient". And we know where that leads. So we say "oh, no, we don't want *that* we just wants a teensy little bit of command to get rid of the worst inefficiencies. But that too is bad.

      • That temptation to overdo either-or thinking... But there's a middle way (tertium datur!) which is oriented at reality. It is known since Aristotle that markets need some regulation to work well.

      • An absolute minimum 'command' or 'regulation' is an agreement on what constitutes money. One might start with "The History of Money".

  6. A little digging yields a NYSE daily volume of 3 billion shares with an average share value of about $40, or $120 billion dollars per trading day. at 250 trading days, that is on the order of $30 trillion in transactions per annum on that exchange alone.

    The entire GDP for the US in the same year is about $16 trillion, just over half the estimated trades on that one exchange, never mind the entire market.

    The "quant" industry is about extracting value from the noise in stock valuations. But once the transactions exceed the entire size of the economy, the noise is increasingly detatched from the underlying valuation.

    The events of 2008 proved pretty conclusively that propaganda aside, local incentives to individuals do not necessarily stabilize the entire system. Indeed, one could argue (and I among others do) that the system must be carefully designed to avoid perverse destabilizing incentives.

    I find it amazing how poorly some people remember the fall of 2008. To say the financial industry stabilizes the economy is more than a little bit counterhistorical.

    • > The events of 2008 proved pretty conclusively that propaganda aside, local incentives to individuals do not necessarily stabilize the entire system

      I don't say why you tie that to 2008. It was obvious anyway. It doesn't affect what I'm saying.

      > the system must be carefully designed to avoid perverse destabilizing incentives

      I'd write the words differently, but the overall idea isn't unreasonable. And I don't think many would object. Where you're wrong is to jump from there to "...and therefore HFT is bad; and therefore a Tobin tax would be a good idea".

      > amazing how poorly some people remember the fall of 2008

      Me too. It was triggered by the US housing market. Boring old mortgages. Nothing to do with HFT, CFD, or any of the financial wizardry that people are now blaming in retrospect.

      • The bad mortgages could not have existed without deregulated financial trickery. Everybody outsourced the risk to someone else. In the end, the risk was socialized, wasn't it?

        Admittedly that was not HFT, but we do know that HFT was responsible for several severe volatility events.

        HFT now dominates all other transactions - is the underlying system secure when a day's worth of meaningless transactions outweigh a week's worth of actual trade? Does it actually provide benefits? There are arguments that it does, but there is clear evidence that it also causes problems. Regardless, it wastes talent and electricity.

        The financial sector as a whole sucks up too much of the profits. It turns out that HFT apparently doesn't get the lion's share of that. But in any case, these are not real services but abstractions. When games played with other people's productive efforts are worth more than those efforts themselves (and they are, by a wide margin) that seems a weird corruption of the model that what we are doing is exchanging labor for value. Of course, I don't believe that labor for value is important anymore. But why that means I should have any respect for the financial lottery escapes me.

        The tiny tax simply fixes this particular silly game. I suppose that would generally drive down wages for programmers and mathematicians so my self-interest would be to oppose it. But the purpose of the tax is to make shareholding less like a casino and more like an investment. I guess it seems obvious to me that we want to do that. Let me think further about how to justify that claim.

  7. My seeking an academically pedigreed connection between programmed trading and volatility keeps coming across this paper from 1989:

    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=330302

    "One form of portfolio insurance uses a trading strategy in risk free securities ("cash") and index futures to synthesize a European put on the underlying portfolio. In the absence of a real traded put option (of the appropriate striking price and maturity), there will be less information about the future price volatility associated with current dynamic hedging strategies. There will thus be less information transmitted to those people who could make capital available to liquidity providers. It will therefore be more difficult for the market to absorb the trades implied by the dynamic hedging strategies, In effect, the stocks' future price volatility can rise because of a current lack of information about the extent to which dynamic hedging strategies are in place."

  8. W:

    "And the temptation is to observe this, and think “oh, so a command economy would be more efficient”. And we know where that leads. So we say “oh, no, we don’t want *that* we just wants a teensy little bit of command to get rid of the worst inefficiencies. But that too is bad."

    Like economic reality's on a single axis between free market and soviet? It's a bit more nuanced than that, isn't it?

  9. I should acknowledge that I have conflated high frequency trading with the whole system, and I guess Brin has too. Rethinking a bit.

    But I still think high frequency trading is insane and that a transaction tax would help a lot.

    Thanks as always to William for keeping us honest.

    • It is hard to keep in mind how the market should work with greedsters skimming the pot in microseconds. The tax is an obvious solution, and will therefore probably never happen.


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>