Saturday, November 24, 2012

Rational Expectations and Low Probability Events

Self interested people have an incentive to form their best guesses about the probabilities and impacts of different future scenarios.   Climate change poses some trouble here because it creates a "non-stationarity" in the sense that the random variables (the impacts of climate change) have means and standard deviations that change over time.   Intuitively, how do you plan to hit a moving target?    If the bullseye target never moves, where your darts land is still a random variable but you know how to practice.  How do we adapt as the target moves?   Professor Rumsfeld has taught us that when you "know that you don't know" what to expect, the prudent person builds some slack into their decision so that they don't regret their choice some time in the future.

With this background, let's consider two pieces in today's NY Times.  First, we hear from the President of Tulane University.  Universities are place based and his University is in New Orleans. He has strong incentives to argue that his University (which was horribly injured by Katrina) is back and is robust in the face of the next storm.


To the Editor:
Orrin H. Pilkey (“We Need to Retreat From the Beach,” Op-Ed, Nov. 15) makes what appears to be a reasonable argument against rebuilding shorelines or homes near the beach destroyed by Hurricane Sandy. Unfortunately, this is reminiscent of what New Orleanians heard after Hurricane Katrina: Why rebuild New Orleans, because it will always be prone to flooding?
Since when did our country develop a standard that we abandon places prone to repeat disasters? People live in danger zones knowing that danger may strike again: San Francisco sits on a fault line, much of New Orleans is built below sea level, and the Eastern seaboard is a flood zone.
In every case, community feeling and the attachment to home has trumped scientific facts and urban planning.
The “resilient development” Mr. Pilkey refers to as a less good alternative to “retreat” can work. New Orleans and neighborhoods like the Lower Ninth Ward exist today because of significant improvements in their flood protection system, proving that the art of the possible can work.
President, Tulane University
New Orleans, Nov. 19, 2012

In my Climatopolis, I argue that it is fine for New Orleans to rebuild if the investors there use their own $.  President Cowen forgets that billions of federal tax payer dollars were used to rebuild his University's city.  That's bad incentives.

Note that at the end of his letter he engages in some public relations to signal to outsiders that New Orleans is now safe. I hope he is right.

A more salient and optimistic example of rational expectations is provided in this piece .  The boss of this firm was prudent enough to build his key floors 4 feet higher to reduce flood risk.  Here is a quote that highlights how we will adapt to climate change:

"But the real storm preparations had been accomplished six years earlier, when Sims Metal Management approved a design for a state-of-the-art city recycling plant that is rising at the South Brooklyn Marine Terminal.
Reviewing projections for local sea-level rise, the company and its architects decided to elevate portions of the site to heights exceeding city requirements by four feet. Using recycled glass and crushed rock discarded from projects like the Second Avenue subway line, they raised the foundation for the plant’s four buildings and a dock.
The fill added $550,000 to the plant’s costs of around $100 million, said Thomas Outerbridge, Sims Metal’s general manager.
But it proved more than worth it. When a 12-foot storm surge swept through nearby streets and parking lots on Oct. 29, the plant’s dock and partly completed buildings did not flood.
“It paid for itself long before we expected it,” Mr. Outerbridge said. “It was built with the idea that, over the next 40 years, this would prove a prudent thing — and the proof came during construction.”

This case study highlights the key adaptation recipe.   Note that if we are "behavioral agents" who do not engage in forward planning then we have a problem.  Such individuals will not have taken the precautions that Sims Metal Management took.  This point has not been discussed by academic economists but in my Climatopolis I argued that climate change planning poses the ultimate test for distinguishing whether neo-classical economists or behavioral economists have the right model of predicting human behavior.   The "doom and gloomers" embrace both a behavioral economics view of individual rationality and a benevolent paternalistic view of government.  That's quite a statement.

UPDATE:  Contrast the company I discussed above, with the New Jersey Transit Authority.   Here is my source and here is a quote:

"Less than a month after Hurricane Sandy damaged nearly a quarter of its rail cars and locomotives, New Jersey Transit is facing withering criticism this week for keeping much of its equipment in low-lying yards during the storm, despite forecasts of potential flooding."

Do you have to be a genius to think about choosing another site for the rail cars?  A cynic might ask if the NJT would have been more likely to have figured this out ex-ante if it had been using its own $ rather than tax payer $.   Moral hazard lurks and the absence of moral hazard (i.e using your own $) tends to focus the mind!!  

You don't have to be Darwin to see that forward looking companies and decision makers will be rewarded by Mother Nature for being smart.  This is evolutionary capitalism at work.   Sink or swim sounds rough but such "tough love" creates strong incentives to change your game and adapt.

No comments :