Daron Acemoglu has written an essay on what the financial crisis means for economics. Highly recommended. I will excerpt and comment extensively. Thanks to Acemoglu’s frequent collaborator Simon Johnson for the pointer, which I picked up from Mark Thoma.

In case you didn’t know, Acemoglu is a recipient of the John Bates Clark medal, an award given every other year (the frequency will now be increased to once a year) to the American economist under 40 with the greatest contributions. Within the profession, the Clark medal is at least a prestigious as the Nobel Prize.Acemoglu writes,

We believed that through astute policy or new technologies, including better methods of communication and inventory control, the business cycles were conquered. Our belief in a more benign economy made us more optimistic about the stock market and the housing market. If any contraction must be soft and short lived, then it becomes easier to believe that financial intermediaries, firms and consumers should not worry about large drops in asset values.

I would add that we also believed that better monetary policy had conquered the business cycle.

better diversification also creates a multitude of counter-party relationships. Such interconnections make the economic system more robust against small shocks because new financial products successfully diversify a wide range of idiosyncratic risks and reduce business failures. But they also make the economy more vulnerable to certain low-probability, “tail” events precisely because the interconnections that are an inevitable precipitate of the greater diversi…cation create potential domino e¤ects among financial institutions, companies and households. In this light, perhaps we should not find it surprising that years of economic calm can be followed by tumultuous times and notable volatility.

He is suggesting that institutions and contracts that make the economy more robust with respect to small shocks can actually increase systemic fragility. Although this may be true, I find it hard to think of a good way of addressing it.

I think that part of the problem is the way that risks are measured and disclosed. When it comes to financial institutions, there are few issues as tricky as the Risk Disclosure Problem. It frustrates me when people keep saying, “the solution is more transparency,” or “we need more transparency,” as if they are offering a useful, constructive suggestion. You might as well say “the solution is for everyone to be rich” or “we need world peace.” Unless the person is making specific recommendations regarding risk metrics or financial reporting, anyone who says that we need more transparency is just a blowhard. In contrast with the blowhards, Acemoglu is doing the right thing and asking people to think about hard problems.

Much of creative destruction takes place at the micro level. But not all of it. Many companies are large and replacement of their core businesses by new firms and new products will have aggregate implications. Moreover, many general-purpose technologies are shared by diverse companies in different lines of businesses, so their failure and potential replacement by new processes will again have aggregate ramifications. Equally importantly, businesses and individuals make decisions under imperfect information and potentially learning from each other and from past practices. This learning process will introduce additional correlation and co-movement in the behavior of economic agents, which will also extend the realm of creative destruction from the micro to the macro.

What he seems to be suggesting here is that there are real business cycles, due to creative destruction, and these real business cycles will not necessarily be low in amplitude.

Sure enough, institutions have received more attention over the past 15 years or so than before, but the thinking was that we had to study the role of institutions to understand why poor nations were poor, not to probe the nature of the institutions that ensured continued prosperity in the advanced nations and how they should change in the face of ever evolving economic relations.

His point is that we should not get complacent about institutions. They may seem to work for a while, but then something changes and they do not.

I have very strong beliefs on this point. It is my view that incentive systems degrade over time, as agents learn to “game” the system. Within a firm, it is up to managers to identify the problem and change compensation systems in a timely fashion. Unless they constantly adjust their internal incentive structures, firms will find that employees learn to achieve more and more personal gains with less and less contribution to corporate profits.

The same holds true for regulation–in theory, regulators need to change their systems as private actors learn to game them. In practice, it is very difficult for regulatory agencies to move quickly enough, particularly when the political process is heavily biased toward the status quo.

Our logic and models suggested that even if we could not trust individuals, particularly when information was imperfect and regulation lackluster, we could trust the long-lived large …firms, companies such as the Enron’s, the Bear Stearn’s, the Merrill Lynch’s, and the Lehman Brothers’s of this world to monitor themselves and their own because they had accumulated sufficient “reputation capital”.

I would say that the real poster children for companies that threw away their “reputation capital” would be the bond rating agencies, like Moody’s. I also think that Freddie Mac and Fannie Mae paid little attention to their “reputation capital.” When I was at Freddie, the CEO understood that we had a valuable reputation for safety and soundness. Several years after I left, he was replaced, and his successor did not seem to care so much about safety and soundness.

reputational monitoring requires that failure should be punished severely. But the scarcity of speci…fic capital and know-how means that such punishments are often non-credible. The intellectual argument for the fi…nancial bailout of the Fall 2008 has been that the organizations that are clearly responsible for the problems we are in today should nonetheless be saved and propped up because they are the only ones that have the “specfii…c capital” to get us out of our current predicament. This is not an invalid argument. Neither is it unique to the current situation. Whenever the incentives to compromise integrity, to sacri…ce the quality, and to take unnecessary risks are there, most companies will do so in tandem. And because the ex post vacuum of speci…fic skills, capital and knowledge that their punishment will create make such a course of action too costly for the society, all kinds of punishments lose their effectiveness and credibility.

One position that I have taken on this blog is that the executives of financial institutions that enjoy government backing should be subject to imprisonment for failure to adhere to the spirit of the law of maintaining safety and soundness. This position draws the ire of commenters, but several of the points Acemoglu has brought up are points that lead me to take this view.

First, there is the fact that regulatory incentives degrade over time, and banks start to game the system. This means that “letter of the law” regulation is bound to fail, because over time banks will naturally find ways to take excessive risks that are within the letter of the law. Second, there is the risk disclosure problem–it is very had to design disclosure mechanisms that are fully protective and yet still allow financial intermediaries to perform their function. Third, there is the fact that the government cannot credibly commit itself to allow large institutional failures. As Acemoglu points out, there is too much “specific capital” in these institutions. Policymakers were afraid to shut down Freddie and Fannie completely, because they had become such a dominant source of funds for mortgages.

I think that financial institutions that enjoy either explicit government backing or a “too big and interconnected to be allowed to fail” status ought to have their executives worry that if they mismanage the institution they could wind up in prison. That would get them to care more about preserving their “reputation capital.”

from a policy and welfare perspective, it should be self-evident that sacrfii…cing economic growth to deal with the current crisis is a bad option.

Acemoglu is saying that an economic downturn is nothing compared to institutions that promote or retard long-term growth. Of course, that is anything but self-evident to people who are ready to throw capitalism under the bus because of the latest unemployment statistics.

Because of the reallocation and creative destruction brought about by economic growth, there will always be parties, often strong parties, opposed to certain aspects of economic growth.

…The United States is not Indonesia under Suharto or the Philippines under Marcos. But we do not need to go to such extremes to imagine that when the …financial industry contributes millions to the campaigns of Senators and Congressmen that it will have an acute influence on policies that influence its livelihood or that investment bankers setting up–or failing to set up as the case may be–the regulations for their former partners and colleagues without oversight will likely lead to political economy problems.

…Market signals suggest that labor and capital should be reallocated away from the Detroit Big Three and highly skilled labor should be reallocated away from the …nancial industry towards more innovative sectors. The latter reallocation is critically important in view of the fact that the Wall Street attracted many of the best (and most ambitious) minds over the past two decades and we now realize that though these bright young minds have contributed to fi…nancial innovation, they also used their talents for devising new methods of taking large risks, the downside of which they would not bear. Halted reallocation will also mean halted innovation.

Here, we get to the “public choice” problems of bailouts and stimulus. Whatever the theoretically optimal policies might be, the impetus in practice is to reward the politically well-connected.