Thursday, December 11, 2014

The allure of design

This post from Simon Wren-Lewis, titled "The Imaginary World of Small State People" is an interesting corollary to Don Boudreaux's complaints from the other side.

Wren-Lewis' basic complaint is:
My first major problem with small state people is that they are not prepared to look at these items on their merits. Instead they have a blanket ideological distaste for all things to do with government.

Boudreaux frequently bemoans the tendency for supporters of state expansion to assume the pure execution of good intentions in public policy.  There is some imperfection in free society.  We have an idea of how we would like to fix that problem through the sausage-making apparatus of the state.  Then a miracle occurs.  Problem fixed.

Clearly we are all self-serving or tendentious to some degree - which I'm sure describes some more than others.  I don't want to spend any time here describing the worst cases.  Instead, I think it's worth thinking about what causes these differences in approach.  Both Wren-Lewis and Boudreaux are asking for some cost-benefit analysis regarding state actions, but both see the other side as being derelict in this regard.

Boudreaux notes that those seeking state solutions to asymmetrical information, negative externalities, etc. in markets don't seem to address those problems in their proposed public solutions, even though principal-agency problems and many other issues present difficulties especially in the public realm.  Wren-Lewis expresses the complaint from the other end of the spectrum that many participants in public debates simply dismiss state action, out of hand, seeming to assume the costs exceed the benefits.

I think the problem here stems from the fallacy of good intentions, or the bias toward design, and the difficulty we have of analyzing the value of emergent phenomena, ex ante.  A state-solution is a solution from design.  Visible costs and benefits are limited to the imaginations of the designers and observers.  But, the unseen effects will tend to tilt negative.  This is because the context in which the action will take place is already the product of some emergent order.  This bias toward negative unintended consequences is something that, today, we universally notice in a context like ecology.  It's funny how two hundred years ago, we didn't tend to notice this problem in ecology, but we did seem to accept it in human affairs, but these biases switched in the early 20th century.

The problem is the opposite for actions outside the state.  Free societies and markets lead to complex interrelations that we almost entirely don't understand.  We tend to only notice exceptions.  An extreme example of this is the frustration we feel when, say, our smartphone isn't doing something quite the way we would like for it to.  We should, objectively, feel awe beyond our capacity for emotion every time we look at our phones.  The only reasonable thing to expect of our smartphones would be for them to cease to function.  In contrast to Boudreaux's presumed miracles, they really are a miracle - a miracle a billion times more impressive than a pencil, which is itself a miracle beyond anyone's understanding.

The only way to understand an emergent system ex ante is through broad principles.  So, while the benefits of state action will generally be contained within our imagination of those benefits when we propose them, the most important benefits of private actions will always be the benefits that exceed our imaginations.  In public debates, both sides are simply failing to come to grips with the emergent portion of the potential outcomes.  This is understandable.  It is an impossible task.

For someone proposing private, emergent responses to a given constraint or problem, the honest description of the benefits is, "I couldn't begin to tell you what the benefits will be.  Millions of people more motivated and skilled than I am will be working through voluntary cooperation on millions of solutions and improvements, and the most long-lasting versions of those improvements will exceed any appreciation or understanding I could possibly provide you now.  And when they are implemented, they will be implemented with such ease that we won't notice there was even a problem - like when we walk into a modern grocery and it seems unremarkable that tens of thousands of items were grown around the world, packaged, and delivered, waiting there for just the time when we would want them."  On the other hand, our fears about emergent phenomena are easy to quantify.  The life we currently have is palpable, and the extent  to which some part of it could be displaced or damaged is something we can understand and tally today.

That's a tough sell.  It's basically what Wren-Lewis is describing - a bias against design and for emergence because of a principle.  And, frankly, I don't know of a convincing way to distinguish a thoughtful version of this from a vulgar version.  Wren-Lewis would agree with me that we should have a bias against bulldozing through the rainforest in the name of managing ecology.  How can we come to some similar principle in human affairs that satisfies both the skeptic and the supporter of the state?  It seems obvious to me that, as a start, we would move toward public regulation that is transparent and simple, minimizing the potential for unintended consequences or regulatory sclerosis.  I doubt if there is much disagreement about that.  In practice, though, this probably tends to look like deregulation, even when it would be more accurately described as safer, more effective regulation.

After the comment quoted above, Wren-Lewis continues:
The evidence that government is ‘always the problem’ is just not there. The idea that private sector activity is always welfare enhancing and is best left alone was blown out of the water by the financial crisis.
This is the bias for design and for good intentions (battling a strawman, no less).  This comment, taken literally, is preposterous.  Left alone?  (There was a run on securities facilitated by quasi-public firms, rated by a legal oligopoly, so that they could be highly leveraged according to an arbitrary standard set by regulation and justified by mandatory public insurance?  Just goes to show what happens when you try to leave 'em alone, I guess.  Anything short of the NIRA is practically anarchy.  We just need to find and fix those remaining areas of discretion.)  In any situation, when things go wrong, it is natural to think that, if only someone had taken control, if only we had regulated this a little more closely, we could have stopped it.  Is there any context where, when something goes wrong, there is a consensus for giving less power to the state?  When BP creates a spill in the Gulf, we all know that the obvious response is to strengthen and fund state regulatory agencies overseeing their work.  When the VA commits egregious lapses of care, we all know that the obvious response is to strengthen and fund the VA, so they can better do their work.  I don't believe I heard anyone propose subsidizing the oil industry after the BP spill so they could better fund their safety protocols.  The double standard here is extreme.

Or, consider the incredible outcome that the U.S. was the only major power not to sign on to the Kyoto Protocol, and yet, is the only country to have actually met the 2012 global goal for CO2 reduction.  And, how did we meet it?  Because of private fracking for natural gas.  This clearly is an unintended positive externality resulting from decentralized economic activity.  But, we naturally experience this as an "accident", because there was no intention, and in fact, there was no way to seriously propose this as an alternative to other CO2 mitigating plans, ex ante.  So, we have the fact that the legal and cultural incentives for aggressive wildcatting in the U.S. are the most powerful force for CO2 reduction in the world right now, and yet there is no way for market advocates to claim this as a victory for emergent order.  Imagine the shame we would feel if every nation that signed on to the Kyoto Protocol met the target, and only we had not.  "We could have done something, and we didn't."  But, now, we're just left with an accident.  Luck.  And it is just that.

We are conditioned by the scientific movement to look for falsifiability.  In the search for things we can learn, falsifiability is very useful and important.  Where we do our work, this is central.  We find topics to concentrate our efforts on, and we find little details to test - something falsifiable that moves our understanding to a higher plane.  But, our self-guided attention gives us a false sense of the pervasiveness of falsifiability.  The fact is, there is no promise that the truth is falsifiable, and, in fact, in the range of true facts with the complexity of broad human endeavors, falsifiability is unlikely.

I think this is why libertarians tend to be fatalistic about direct political activity.  Opponents demand falsifiability.  Either our principles are misplaced anyway and we are wrong, or we have a strong point to make and it's simply not possible to make it convincingly.  I know how it feels to be on the other side.  I feel that way myself about much of the Austrian Business Cycle stuff - it's not convincing to me, even though I am drawn to much of the Austrian work on decentralized activity.

So, if a common understanding is not available with which to check ourselves, how do we keep ourselves from falling into a vulgar form of our principles?  I'll look into that on the next post.


  1. Your smartphone reference struck a chord with me. My boss was complaining just the other day that Prime had failed to deliver an order within the promised two days (and, indeed, it had not been delivered even several days later.)
    After commiserating with him a bit, I commented to him that rather than be frustrated at the one time out of his hundreds of Prime orders placed annually, he should be amazed at how few of them fail to meet the promised delivery standard.
    Myself? I've never had one fail. That's awesome.

    1. Thanks for the comment, Murph. I think there is an interesting idea here about desert and free lunches. I am as happy to talk about the bad incentive effects of some redistributionist policies, and they are clearly real. But, even without public redistribution, we are all beneficiaries of free lunches from markets, to the extreme, as your and my examples above demonstrate. Sure, I work to pay $100 a month or so to have a magic box in my pocket that connects me instantly to the world, and we work to pay a few dollars a month to have a global array of products available to us at the click of a button. But, these are services that would have cost a billion dollars a month or more in our palpable memories. So, my work is contributing $100 a month to the remuneration of these services, and the other billion dollars worth was a freebie provided by markets. We are all inundated with unearned abundance, so our public expectations of minimum quality of life rise, naturally. And, it probably leaves all of us weaker, compared to our ancestors. This is an unavoidable trade off. We have a subconscious puritanical reaction against this feeling of unearned abundance. I think this feeds a lot of misapplied intuitions about the economy. For instance, people widely attribute great disruptions like the recent recession or the Great Depression to the booms that preceded them. We have a deep distrust of abundance.

      Contra Milton Friedman's frequent retort that "There ain't no such thing as a free lunch.", markets are indeed a sort of free lunch, which is a reason we are often uncomfortable with them. Here's an old post I did on the idea (as if I haven't already gone on too long):

    2. Oops, edit. "I am as happy to talk about the bad incentive effects of some redistributionist policies as anybody...." My error in phrasing wrongly appears to attribute this opinion to Murph.

  2. Nothing to add other than that this was a brilliant post. I too struggle with how to explain the problem solving capabilities of a complex adaptive system. To those not getting the paradigm and seeing the role of decentralized division of labor coordinated by prices, it seems like wishful thinking.

    Special thumbs up to the comment on "effective regulation" as opposed to deregulation.