Can the Dodd-Frank Act Defeat Financial Crises?

Can the Dodd-Frank Act defeat the natural cyclicality and periodic crises of financial markets?

The answer is: No. This natural cyclicality is far more powerful than the thousands or tens of thousands of pages of regulations, which will be the principal memorial of the act.

The real name of the Dodd-Frank Act should be the "Faith in Bureaucracy Act." This is a faith I do not share.

This is not because of a lack of intelligence, or diligence, or good intentions, at least much of the time, in the financial regulatory bureaucracies. It is rather because I see no evidence that the human minds operating in regulatory bureaucracies, and driven to political defense and expansion of their own jurisdiction and power, have any superior insight into the unknowable future and its ineradicable uncertainty. I see no evidence that government actors are exempt from the cognitive herding that affects everybody else.

For example, the Federal Deposit Insurance Corporation made this comforting announcement right at the top of the bubble, in 2006: "More than 99 per cent of all insured institutions met or exceeded the requirements of the highest regulatory capital standards." This did not save them from more than 350 bank failures so far this cycle (not counting the banks which were bailed out), with hundreds more on the endangered list, or from a large negative net worth in the FDIC's own deposit insurance fund.

The fundamental problem of financial markets is not risk, but uncertainty. As Frank Knight taught us almost 100 years ago, risk means playing the odds; uncertainty means that you do not even know - and in fact you cannot know - what the odds are. This is what lies behind the sometimes-celebrated "financial instability hypothesis" of Hyman Minsky, which is rediscovered in each financial crisis and then forgotten again.

Treasury Secretary Geithner has written, "We cannot build a system that depends on the wisdom and judgment of future regulators." To be more precise about this, we can build such a system - the Dodd-Frank Act has built one - but it will not avoid Minskian instability.

Let us consider where the uncertainty comes from. Financial markets are not only complex, but wildly recursive - something like infinitely recursive - where very large numbers of financial actors are all mutually influenced by the actions, expectations, strategies, reactions and beliefs of the others. We might say it like this: "What I think about what you think about the probability distribution of future financial outcomes changes the distribution."

The interacting financial actors include government actors. The government actors are not above, but themselves part of the recursive complexities. Once you see that the government actors are not looking down from afar on financial interactions, but are themselves just another set of actors enmeshed in the recursive expectations and strategies of the overall complex system, you can see the difficulty of "regulating" the outcomes of the system as a whole.

Of course, some very important government financial actors are central banks. The U.S. central bank today owns about $1 trillion of mortgages and can fairly be called "the biggest savings and loan in the world." Central banks, while striving for stability, can themselves nonetheless be major generators of what we have come to call "systemic risk."

In 2002, then-Federal Reserve Chairman Greenspan explained to Congress that rising house prices, with their positive wealth effect, were offsetting the negative wealth effect of the collapsed tech stock bubble. So they were, and the Fed, in my judgment, fully intended to stoke a housing boom - I call this the "Greenspan Gamble," and find it ex ante a reasonable gamble. Writing in 2003, economist Richard Koo judged it "a brilliant strategy." Needless to say, it turned out not to be, as the boom got away into the Bubble.

Senator Bunning memorably asked Fed Chairman Bernanke, it is said: "How can you regulate systemic risk when you are the systemic risk?" An excellent question! It captures not the whole truth, but a truth of central importance.

Discussing famous financial managers, Martin Mayer wrote of the "sudden revelation of the truth that becomes obvious only in hindsight, that a lot of them don't know what they're doing." A nice summary of the problem of uncertainty.

Some other exceptionally important government actors in the inflation of our 21st century Bubble were Fannie Mae and Freddie Mac, which were a $5 trillion government intervention in the housing finance sector. As is well known, entirely absent from the Dodd-Frank Act is any attempt to address these entities, which by the time of the act were flat broke.

Fannie and Freddie are best understood as off-balance sheet financing vehicles for the government, which created hyper-leverage in mortgage financing. The key to keeping them off-balance sheet and off-budget was for the government to give them a so-called "implicit guaranty," rather than a formal guaranty. In fact, this status was invented by the Johnson administration in 1968 to take Fannie (Freddie wasn't created until two years later) off budget in a time of rising federal deficits.

So the taxpayers' guaranty was only "implicit." But bond salesmen all over the world said something like this to investors: "This Fannie and Freddie debt is really government debt. You can't go wrong - you get a higher yield, but if they get in trouble, the government will pay!"

Now it is obvious that the bond salesmen were right, and what they told the bond buyers was absolutely true. The result is that ordinary Americans are being taxed so that the bondholders of the hopelessly insolvent Fannie and Freddie can be paid off at par. Yet while the status of Fannie and Freddie were debated for years, nobody - and least of all, their regulator - predicted that they would go broke from bad loans.

Will the regulatory bureaucracies, with vastly expanded presence and cost, somehow see the unknowable future next time, uncertainty, complexity, and their own contributions to the problems notwithstanding? This is the faith of the Dodd-Frank Act, but it hardly makes a convincing faith.