Too Big To Fail Isn't The Problem... It's The Hidden Risk That's The Problem
from the as-if-that's-possible dept
Duncan Watts has a thought provoking writeup in the Boston Globe talking about the problems of systematic risk, and why no one could successfully see exactly how the various dominoes would fall, leading to our current (and still ongoing) economic financial crisis. Basically, his argument is that the system has become too intertwined and complex, such that no one can really manage the risk. This is hardly a new idea. Watts' suggestion (which, again, is not necessarily new, and has been discussed by many, including Treasury Secretary Tim Geithner) is that perhaps we need a "systematic risk manager" within the government, whose job (like anti-trust folks) is to look at various companies and determine if they're too big to fail -- and then see how to change things such that they're no longer too big to fail.It's a nice idea... in theory. In practice, it's a lot harder. The very reason systematic risk is such a problem is that it's so hard to even imagine the scenarios taking place. The idea that Lehman Bros. failing would have so much impact elsewhere is simply beyond the scope of what most people could have even imagined -- and that would almost certainly include any "systematic risk manager." While I agree that it's a problem that we end up with companies that are "too big to fail," I tend to think, in the long run, it's futile to try to predict ahead of time who's really "too big to fail," but that such an issue should only come up in the event of a gov't bailout. Thus, if you need to take gov't money to stay alive because you are deemed "too big to fail," then it should be required that as a part of the terms of the deal, you need to work out a plan that makes you small enough to fail.
Otherwise, you end up in a situation where companies who are successful get penalized for it. The only time "too big to fail" is a problem is when such a company fails. We shouldn't necessarily be penalizing a company that's too big to fail if it's not going to fail.
Separately, Watts notes that this idea of trying to prevent "too big to fail" is a way of avoiding systematic risk. I'd argue he has the equation a bit twisted. Too big to fail isn't the problem. It's the hidden risk that leads a company that is "too big to fail" to fail that's the problem. The answer to that is not breaking up successful companies -- it's increasing transparency into actual risk. That means increasing openness and data sharing, rather than the status quo of quarterly reports with the real details hidden and buried beneath complexities, combined with Wall Street putting together packages whose sole purpose is designed to hide the actual risk. Make the real data transparent (and real-time) and let anyone access and mess around with the data, and you get a much more accurate view of the risk, and you avoid situations where "healthy" investments suddenly turn sour.
Watts has the right idea that systematic risk is a problem, but the wrong solution. Companies that are too big to fail failing is a symptom of a lack of transparency over the actual risk. The answer isn't to stop companies from getting so big. It's to provide more transparency into the actual risk.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: radical transparency, risk, systematic risk, too big to fail, transparency
Reader Comments
Subscribe: RSS
View by: Time | Thread
nobody saw it coming?
[ link to this | view in thread ]
Re: nobody saw it coming?
[ link to this | view in thread ]
[ link to this | view in thread ]
[ link to this | view in thread ]
No. Wrong. The problem isn't that we are unable to see what the consequences might be. The problem is that the consequences can't be determined, and could not be determined, even if we had perfect transparency.
You are talking as though what we have is a complicated system - there are many moving parts, but if we look at it closely enough, we can figure out what's happening. Watts is saying that it's a complex system - the parts depend on each other and influence each other recursively, which is essentially a chaotic system, which cannot be predicted.
When you have a chaotic system, you cannot prevent or mitigate conseuqneces. Your only defense is to minimize the influence of each part, and in addition, to lower the probablility that one part will directly impact another part (i.e., to regulate the market).
[ link to this | view in thread ]
Hmm...wasn't there a boat...
[ link to this | view in thread ]
Transparency needs to be combined with managing consequence of failure. One part of managing the consequence is to set capital requirements for firms based on impact of their failure. So larger firms would have greater capital requirements as the consequence of their failure is greater.
But managing consequence is more than setting capital requirements for firms. More broadly it is about shaping how the system (economy) as a whole responds to an event. Managing the impact includes how redundancies are managed, how bankruptcies are managed etc.
The area that I see having a large amount of impact is how redundancies are managed. In effect you want hiring and firing to be as easy as possible. The problem is that a redundancy causes demand destruction which on large scales intensives impact. Problem with statutory redundancy payments is it hits a company's cashlfow at a time when they can least afford it. Unemployment benefits can go away towards addressing the problem but do result in large taxes.
A solution to the problem is to provide government back loans for redundancy payments that are recouped through the businesses taxes. It avoids the cashflow hit while also addressing demand destruction caused by redundancies.
[ link to this | view in thread ]
how bout we re-enact
[ link to this | view in thread ]
[ link to this | view in thread ]
When you have that mentality, transparency does nothing, because they will just ignore the risk.
[ link to this | view in thread ]
How do you protect a bank that is financially fine one week, when faced with a run on the bank caused by rumors, liquidates in two weeks. People are not rational and neither are our markets. Bubbles go to high and the bottom goes too low.
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Too Big To Fail
[ link to this | view in thread ]
Today we have privatized the profits while making public the loss. You can't do that on the down side. The down side is what keeps most people honest. Take away the down side? Look out.
[ link to this | view in thread ]
[ link to this | view in thread ]
The problem with the Lehman bankruptcy was not the collateral effects of the collapse but the fact that the government acted in a manner that was so arbitrary, unpredictable and unrelated to past behavior that no one had any idea what the ground rules would be. That is not "too big to fail." That is regulatory failure. That is what left everyone scrambling to figure out how to manage their issues and too nervous to do anything until there was some clarity about what the government would permit or prohibit and the situation stabilized.
No amount of "transparency" or regulation is going to address these issues. These companies, public and private, are subject to voluminous reporting requirements to various agencies and government bodies already. There is no lack of "transparency" (a word that has lost all meaning in the last several years anyway).
The only thing that will discipline them is the notion that their failures will actually cause them to fail rather than fall into a governmental safety net.
There is no person or group of persons in government who are smart or capable enough to perform this function, particularly if when you consider the thousands that are already charged with performing these functions. Throwing more money into the regulatory black hole and tying down companies with more bureaucratic red tape and reporting requirements will do absolutely nothing to address any issue in the current environment.
How about Sarbanes-Oxley? Billions of dollars and the beginning of the transfer of capital markets from New York to London in the name of "transparency," increased reporting, enhanced criminal sanctions for things that were already illegal, bowing to the latest trends in corporate governance and enhanced accounting standards and internal controls to address issues of risk. It only took about 5 years until this altar to regulation, transparency, and risk management crumbled into the meltdown.
[ link to this | view in thread ]
Re:
[ link to this | view in thread ]
Dumb joke
We're too big to fail.
[ link to this | view in thread ]
Re:
Auto makers can't use 3rd party manufacturers? Or 3rd parties can't become bigger than 10% of the main manufacturer? Or big manufacturer can't buy 3rd parties? Or 3rd parties can't buy each other? Or 3rd parties can't further subcontract? Or...
Recognize of course, that "too big" doesn't happen at a specific point. There is no particular number (revenue? expenses? debt? employees? function of each?) that indicates that a company is too big.
The only thing I could possibly think of would be a function of cash flow, debt and risk. But even that would have to take into account market fluctuations, company size (smaller companies often need to risk more), etc.
[ link to this | view in thread ]
enforcement of transparency ...
In software we call a software system that has been patched and added to like that "spaghetti code" because changes start affecting unintended parts of the system. You often write about the unintended consequences of new legislation. It is the same thing.
To fix the situation you have to essentially start over with a fresh new version. The problem is that there are stakeholders that do not want this. So we keep limping along until some catastrophe forces change. We may be at that point but I do not see the kind of changes needed occurring yet.
[ link to this | view in thread ]
And once you answer that why do people continue to use the phase?
Call me crazy but I don't see a difference if a local I use goes down or a national . either way there is someone else to pick up the slack.
[ link to this | view in thread ]
Re: nobody saw it coming?
I think that some people saw bad stuff was on the way, and knew it would be really bad (even worse than it has been). But no one connected all the dots to figure out how the collapse would occur.
[ link to this | view in thread ]
Re:
When you have that mentality, transparency does nothing, because they will just ignore the risk.
I disagree. A big part of the problem was people believed certain assets were less risky than they really were, and thus they couldn't price them properly. These assets got passed to ever more unsophisticated investors who relied on others to rate the risk... but the ratings were a guessing game, because the details were all hidden.
Open up the details, let there be *real competition* in terms of rating the risk, and you'll find a very different, and much more efficient market place.
[ link to this | view in thread ]
shouldn't the risk of failure matter?
[ link to this | view in thread ]
Too Big To Fail
Actually one of the problems is letting companies get too big, not just from a "To Big to Fail", but being "to big for the market".
When you have five or more companies that have 80% or more of the market, they they will inevitably collude -- passively or actively -- and the customer suffers. There wouldn't be much talk of charging users outrageous fees for text messages or tethering phones if they could choose from ten cell phone companies; there wouldn't be a discussion over net neutrality if there were ten ISPs for users to choose from.
Allowing companies to merge -- such as the mergers of the cell phone carriers over the last decade are anti-competitive, and shouldn't be allowed.
[ link to this | view in thread ]
Re:
The reason for the crisis is ultimately the Fed, who manipulate the interest rates. If I have capital and you want to borrow it, why should a 3rd party say what rate I can charge you? There is a reason (but it is the man behind the curtain): fiat currency. The reason the Fed has to control the interest rates is my capital I want to lend to you is actually worthless. It isn't backed by any physical object, just millions of US taxpayers.
[ link to this | view in thread ]
Nice to see more serious discussion of theory here...
But where can you look to find those correlations? Surprisingly, some of them are observable in cross-correlations of short-time behavior over long-time periods:
http://arxiv.org/abs/0903.2099
Those signals may be arbitraged away eventually without any change in the correlations that they signal. The role of a risk regulator should be to gather and process information about the types of risk and time-horizons for potential or actual liquidity needs with an eye toward preventing too much flow from the finance industry into one particular industry over a particular window of time (dot com, real estate, &c.)
There is a physical limit to the IRR that can be achieved by legitimate (non-fraudulent, sustainable) growth. We don't need a whole new set of federal authorities to measure how fast money is flowing into particular things, with an eye toward slowing and spreading the flow more evenly if it looks like it's going in too fast.
The reason we have the problems we do is because the mental models we all tend to rely upon in understanding markets -- the comparative statics of shifting supply and demand curves -- simply fail to help us in understanding market dynamics far from equilibrium.
[ link to this | view in thread ]