There’s A Flaw In Your Logic, Good Sir   2 comments

Gerald P.  Jr., over at the CATO institute, offers a missive entitled: “The Gulf Spill, the Financial Crisis and Government Failure”.  My teeth ache.

From the article:

The Gulf oil spill and the global financial crisis both demonstrate the failings of big government.

Already we have a problem.  The Gulf oil spill and global financial crisis illustrate failings of government.  We don’t know for sure that they illustrate the same failing.  We also don’t know that the size of government has anything to do with the problem.  You are arguing in advance of your facts, Mr. O’Driscoll; motion to strike your comment as prejudicial to the jury.

Given that it is your opening line, I don’t expect the rest of this to go well, which is too bad, as you often do write good stuff.

The agency directly responsible for regulating the activity is the Minerals Management Service (MMS) of the Department of the Interior. Government regulation is intended to protect the public interest against bad or irresponsible behavior by private parties. In the case of offshore drilling, the federal government has assumed the role of solving a collective action problem. Potentially all Americans benefit from the drilling, but those living in coastal areas suffer disproportionate harm from mishaps. The government theoretically negotiates on their behalf and establishes rules to protect them.  Obviously, regulation failed. By all accounts, MMS operated as a rubber stamp for BP. It is a striking example of regulatory capture: Agencies tasked with protecting the public interest come to identify with the regulated industry and protect its interests against that of the public.

I agree, MMS is a striking example of regulatory capture.  I disagree with your characterization of regulatory capture.  “Agencies tasked with protecting the public interest come to identify with the regulated industry”.

“Come to identify”?

This is the sort of phrase I use to describe my dog’s behavior in the household.  He has “come to identify” me as the alpha male in the pack.  Has the Materials Management service, as an organizational entity, “come to identify” with the oil industry (i.e., arrived at this identification under its own auspices)… or has it been encouraged to identify with the oil industry (either by actions of the oil industry, or by political entities above it, or some combination thereof)?  How does an organization “come to identify”… an organization does not have a unified cognitive function.  It’s a collection of individuals.  They don’t all decide to identify with an industry independently all at the same time, do they?

The result: Government fails to protect the public. That conclusion is precisely the same for the financial services industry.

The result is not precisely the same.  It is conceptually the same.  You’re misusing “precisely” in literally the same way people misuse “literally”.

Advocates of heavy regulation promise that risky behavior by banks can be controlled and limited by regulators. There are two major reasons such efforts fail. I have already discussed the first: regulatory capture.

Okay, let’s talk for a second about regulatory capture, since I agree regulatory capture is indeedy the sort of thing that represents a real problem, and you haven’t covered it well.

Remember when I talked about audit a couple of days ago?  We are going to rely somewhat on that post here.  Go read it, if you haven’t already.  Okay, now you’re back and ready for a new bunch of genius knowledge.  When you’re talking about regulatory capture, you’re talking about some of the factors I mention at the end of the post… first:  “If we make it harder for people to do one kind of bad stuff, are they going to stop doing bad stuff altogether?  Or are they going to move to a different kind of bad stuff that’s worse?”

Regulatory capture is a case where the organizational process being audited leads the organization being audited to decide “to do a different kind of bad stuff that’s worse”, namely they suborn the audit process either directly (by bribing the auditors) or indirectly (by political maneuvering changing the rules of the audit process), for the purpose of either exempting themselves  from the audit, or for the nefarious and truly reprehensible purpose of entrenching their organization’s place in the market by raising huge barriers to entry.

How you prevent regulatory capture?  This is actually *not* a big mysterious insoluble question.  “Quis custodiet ipsos custodes?”

Let’s look at the particular case of the MMS.  It is a federal agency.  The current acting head, Bob Abbey, replaced Elizabeth (Liz) Birnbaum.  Ms. Birnbaum (who inherited the job in July 2009) seems to have been thrown under a bus, so to speak, but at the same time when you’re the head of a major regulatory agency you ought to have fixed some of the problems that were present when you took the top spot.

However, it’s not terribly counter-intuitive to imagine that stacking a regulatory agency with a bunch of pro-industry appointees for several years might take some time to undo, yes?  Particularly in today’s employment-employer market, where it’s difficult to fire people?  (Yes, even in private industry?)  Stacking was a tactic that President Bush was accused of using in so many different problem domains it’s hard to list them all.  To be fair, there are accusations of the same levied at the current Administration, like any other.  However, when the question is that of regulatory capture, it seems fairly obvious that stacking in the particular case of an oversight agency with industry insiders is a bad idea.

Perhaps (and I’m just brainstorming, just thinking outside the box, here), regulatory capture is just one symptom, here?  Perhaps the root problem is that our political system is very nearly engineered so that the party in power feels that it must embed itself in every little facet of government operations in order to further its long term goals, which are most often aligned with the good of the party, not the good of the country?

Is this less an issue with large organizations in general, than it is with large organizations that perforce must trade leadership (between diametrically opposed philosophies) every 4-12 years?

Maybe we could compensate for this problem in this particular domain, without throwing our hands up in the air and claiming that we can’t build anything bigger than a breadbox without dooming it to failure?  The CDC seems to operate fairly well.  The NSF seems to get along okay.  Maybe we ought to try having these other regulatory agencies exist without having the capability of having an Administration (of any stripe) suborn the process they’re supposed to oversee?

The second source of regulatory failure is the knowledge problem identified by Nobel Laureate Friedrich Hayek. The knowledge required by regulators is dispersed throughout the industry and broader economy. For regulation to work, that dispersed knowledge must be centralized in the regulatory agency. To successfully accomplish this requires central planning of the industry, if not the economy.

Now, this is actually germane.  For this reason, if no other:

By way of explanation, MMS Lake Charles District Manager Larry Williamson told the Acting Inspector General, “obviously, we’re all oil industry. We’re all from the same part of the country. Almost all of our inspectors have worked for oil companies out on these same platforms. They grew up in the same towns. Some of these people, they’ve been friends with all their life. They’ve been with these people since they were kids. They’ve hunted together. They fish together. They skeet shoot together … They do this all the time.”

It’s certainly the case that if you want to have experts in oil drilling, you’re probably going to be hiring experts from the oil drilling industry to provide some of the functionality of your oversight mechanism, and thus they’re going to have friendship ties to the industry (this can often work out to your advantage, by the way, since people will tell friends about fiascoes that they won’t discuss with a nameless pencil-pusher).  But it’s also not anywhere near the case that nobody can be knowledgeable about audit if they don’t work in the industry.  It takes anyone with a decent knowledge of security maybe 10 seconds to identify this sort of a problem.  “Gee, Bob has worked for BP before.  Maybe we ought to read Bob’s audit reports with a certain amount of scrutiny if Bob is auditing BP instead of Shell.  Or maybe we can have Bob audit Shell in the first place, and get Louie to audit Shell.  That seems less outright stupid, doesn’t it?”

But the local knowledge of specific circumstances of time and place cannot be aggregated in one mind or agency. We know that is impossible, and that impossibility was the reason for the collapse of the Soviet Empire and the transformation of the Chinese economy.

(sigh) No, Mr. O’Driscoll, we know no such thing.  Your statement is flatly ridiculous.  By this line of reasoning, any sort of aggregation of knowledge is useless, and any attempt at structure is equivalent to a totalitarian, planned economy.  You don’t have any sense of measure or trade-off, here.

Yes, it is true, as you aggregate knowledge you have a tendency to lose detail.  This is a problem with abstraction layers and information hiding, something that people have known about in the Computer Science field… forever, relatively speaking.  At least, in technology terms.  It’s relatively new in organizational science, but people are starting to take note.

An organizational structure is a lot like a computer program.  As the structure gets more complex, the areas in which large functions of the organization interact with other large functions need to be normalized, or you can’t get any work done.  Typically, when you do this, you’re going to lose some efficiencies in return for the others that you gain.  It’s a classic example of the trade-offs of top-down vs. bottom-up structure.  However, something the small-government people have failed to recognize is that if you don’t do this, your structure likely cannot scale.  You can’t build the Great Pyramid of Cheops with one guy.  You can’t even build it with a million guys if they don’t coordinate somehow… and if they’re coordinated, there are going to be times when a bunch of them are sitting around, doing nothing, waiting for somebody else to get something else done.

Getting groups of people to align along a particular goal predates history as a human societal problem.  There are some areas I agree with libertarians, it’s probably better to just leave well enough alone.  Some problems don’t require a Great Pyramid solution, so don’t build one.  I’m pretty sure that any industry capable of ruining the global economy or the environment for a geographical region larger than many ancient civilizations probably requires some sort of audit.  Yes, granted, it’s going to be tricky to audit such a beast.  The alternative is to forbid such things to exist, or to ride the crazy boom-bust cycle on a greater frequency than we currently do so.  I find neither particularly attractive.

Regulatory practice represents islands of central planning in otherwise decentralized market economies. If we add back in the problem of regulatory capture, then we get industries coddled and protected by government. When business and politics become intertwined we move from market economies to crony capitalism.

Yes, that’s true.  When we remove regulatory oversight, though, we get a whole new set of problems.  Please don’t pretend that they don’t exist.  You’re comparing regulatory oversight and its drawbacks to itself, instead of the lack thereof.

What is the missed lesson from all this? When President George W. Bush had his Katrina moment, the federal government’s bumbling response was blamed on him, on the Republicans, and on conservatives. Now it is President Obama’s turn. His administration’s faltering response to the disaster in the Gulf is attributed to his personal failings, staff ineptitude, communication failures, etc.

George W. Bush put a demonstrably unqualified person with no training in exception scenarios in charge of the governmental body responsible solely for responding to crisis events.  Obama put a demonstrably unsuitable person with some experience but no industry ties in charge of a governmental body riddled with industry insiders which was responsible for auditing that industry.  One major difference: there was some reason to believe that Ms. Birnbaum would be able to approach the problem of regulatory capture as she was not an insider.  She failed, yes, that’s on her and the Administration.  However, it’s clearly not the same problem as Mr. Brown’s inept handling of FEMA.  In Mr. Brown’s case, there was no reason to suppose that he might be successful.  In Ms. Birnbaum’s case, there was reason to believe she might not be successful, but also reason to believe she may be, and until it had been tested, you couldn’t know for certain.

A big-government conservative administration failed in crisis, as has a big-government liberal administration. The regulatory state did not prevent excessive risk taking whether in financial services, nor perhaps in offshore oil drilling. Government response to crises once they occur is slow and inept.

Funny, while the CATO institute fairly often criticized the Bush Administration (particularly in the areas of spending and civil liberties, it should be noted for fairness), they also supported quite a number of the Bush Administration’s policy decisions, often for “small-government” approaches.  I don’t think the privatization of Social Security supported by CATO would have been a very good idea, considering the market performance of the last 5 years.  That aside…

The “regulatory state” did not prevent excessive risk taking in financial services, because the… wait, what is the “regulatory state” circa 2008 when it comes to finance?  According to your article:

In his book on the financial crisis, “Jimmy Stewart is Dead”, Boston University Professor Laurence Kotlikoff counts over 115 regulatory agencies for financial services.

115 regulatory agencies didn’t do a good job of regulating a dozen super-sized financial institutions?  Could this be due to lots of individual regulatory bodies being out of date?  At least, partially?  After all, they had come into being to audit many small financial entities who were legally prohibited from merging from 1933 until 1999, a period of sixty-three years, not to deal with 12 entities, right?  Hypothetically, could you see an alignment problem here?

It’s not quite kosher to blame these organizations for failing to regulate an industry that looked quite different after the repeal of Glass-Steagall, when they were built to audit organizations of a type that no longer existed, don’t you think?  This is like blaming the guy in charge of watching the security cameras at a local bank branch for not noticing embezzlement at another bank branch entirely.  Of a different institution.  That isn’t even a bank.

The reason for the oil spill isn’t quite so complicated.  A bunch of demonstrably corrupt people weren’t doing a pretty simple job.

All this is not because either Republicans or Democrats are in power, but because big government doesn’t work. It can’t deliver on its promises. Big government overpromises and underdelivers. In reaching to do more, big government accomplishes less. That is not an ideological statement, but an empirical observation.

It may be an empirical observation, but it’s hardly a body of data.  I can think of more than a few examples, just off the top of my head, of “big government” delivering precisely what was promised, in some cases before time or under budget, or both.  If arguing by anecdote is an acceptable method of debate at the CATO institute, I can play!

In the case of financial services, virtually all the proposed regulatory reform offers more of the same.

Yes, that’s true.  And you artfully dodge mentioning that many of the people who support “big” government have made precisely this observation, and have complained rather loudly about it.  Like this guy here.  Not to mention the proponents of small government who have complained that it doesn’t give the government enough authority to break up large institutions.

Einstein famously defined insanity as the belief that, if we repeatedly do the same thing, we will eventually get a different result.

Yeah, uh, the source of that quotation is in some dispute.

University of Chicago law professor Richard Epstein has observed that we need simple rules for a complex world. The complexity of rules is self-defeating, because that complexity requires more knowledge than can be acquired. Brazil has a simple rule for directors of failed banks: They are personally liable.

Now, here’s the first bit of this piece that actually has value.  It’s true that sometimes (actually, often) you can devise simple solutions for complex problems.  Simple rules for financial oversight are probably a damn fine idea, but if you’re talking about making directors of any corporation liable, you’re talking about getting rid of the idea of limited liability corporations altogether, right?

Now, I’m busting Mr. O’Driscoll’s chops here, but in many cases I agree with the principles espoused by the Cato Institute.  I do believe that there are some problem domains where a large regulatory structure is unnecessary, or is contraindicated.  I do believe that there are some problem domains where a local regulatory structure will be more effective than a larger entity. I don’t think any of these situations are as cut and dried as Democrats, Republicans, or Libertarians would like to believe, however.

And I really wish all three groups would stop re-writing the history of current events to support their belief systems.  Strictly from the standpoint of complex systems analysis, financial collapse and the oil spill are not representative of the same sort of oversight failure.  Pretending that they are is just baloney, and bad failure analysis.


Posted June 14, 2010 by padraic2112 in politics, rants, Uncategorized

2 responses to “There’s A Flaw In Your Logic, Good Sir

Subscribe to comments with RSS.

  1. The problem with oversight and regulation in the context of large companies like BP is that the resources that can be fielded to control the outcome of the audit process is often one-sided. Case in point: in 2010 the MMS has a $348 million budget. BP’s *profits* for Q1 2010 were over 6 billion dollars. So assuming that BP would need to spend one dollar in order to offset each dollar that the MMS budgets, they could easily outspend any attempt to regulate them in any meaningful way. Heck, you could probably identify a dozen or less key players and pay them a couple of million each every year and *still* avoid anything approaching real oversight.

    It all comes down to the basic fact that no matter how hard everyone tries, you can’t legislate morality. The capitalist system *selects* for sociopathy in senior leadership; why then are we shocked when things like the Gulf spill happen?

    As far as I can tell there is only one way to combat this problem- legislation that makes companies legally equivalent to private citizens for purposes of civil and criminal liability, eligible to be jailed (i.e. run by the government for some period of time with all ‘profits’ seized as tax revenue) and subject to capital punishment (nationalization and breakup). There should also be legislation making Board members and everyone in the C-suite automatically co-defendants in such cases. The actual probability of such legislation ever being adopted if introduced is, of course, mathematically indistinguishable from zero. Hence my interest in guns and rural property. 🙂

  2. > So assuming that BP would need to spend one dollar in order to
    > offset each dollar that the MMS budgets, they could easily
    > outspend any attempt to regulate them in any meaningful way.

    That’s a pretty big assumption. Not *everybody* is susceptible to bribery with money.

    > Heck, you could probably identify a dozen or less key players
    > and pay them a couple of million each every year and *still*
    > avoid anything approaching real oversight.

    Maybe. This certainly has been the case before (during Prohibition). However, even before the repeal, organized crime’s control over the liquor market was starting to come under investigation, no?

    You solve this problem by watching the watchers. The trick is to have the watching of the watchers done by people who aren’t generally susceptible to the same system of rewards that the watchers are; again, the NSF, the CDC, the NIH… these are all systems where it works much more often than it fails. Of course it won’t be perfect, but it just hasn’t been tried here.

    It would help immensely if we had a practical method of blackballing people, or we had a third mechanism of legal remedy. We have “civil liability”, we have “criminal convictions”, we need “guilty of incompetence or malfeasance in office”. Right now, you can boot someone out of a governmental oversight body, but they can turn right around and go to work in the very industry they were supposed to be regulating. That’s hardly cricket (it’s also not terribly hard to fix). You can work in the Senate, get an ethics writeup, and keep on bein’ a Senator. Convict these suckers and boot ’em out. Let ’em go get an honest job.

    Why bother to hold only C-level executives accountable? Why not just get rid of the corporation as a legal entity? Kinda hard to be “too big to fail” if you can only be as big as the richest dude…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: