By Felix Salmon
July 20, 2011

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Utterly depressing column by Nathan Myhrvold, patent troll — Bloomberg

How to differentiate an economist from almost anyone else in society — Gelman

Paying taxes your employer keeps, by David Cay Johnston — Reuters

Wendi Deng’s Five Best Enraged Expressions — Awl

SNL’s “Ronald Reagan mastermind” sketch — YouTube

US Treasury Awards $142.3 Million to Benefit Organizations Serving Economically Distressed Communities Nationwide — CDFI

“Hackgate” – the movie trailer — YouTube

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
4 comments so far

One of the easist ways to differentiate economists from almost everyone else in society is to simply look for the people who are never wrong regardless of whether or not the data prove their theories.

I am an engineer. It would be very useful to me to simply ignore data that appears to disprove my working hypothesis – it would make solving problems so much easier. It would also be useful for when something I design fails, because you can simply state that the data demonstrating that the failure occurred is incorrect and the theory shows that the structure is still standing, regardless of what the eye-witnesses say. In fact, because the theory says that the structure is still standing, we should replicate the design and increase the rewards to the designer!

Posted by ErnieD | Report as abusive

@ErnieD: One thing we engineers could contribute would be to get the economists to understand that there is always a tradeoff between efficiency and reliability (some call it “robustness”). It isn’t efficient for the Space Shuttle to carry all those redundant systems into orbit each trip, most of which never get used. But engineers understand the value of reliability and its associated costs.

Similarly, it isn’t necessarily efficient for banks to carry lots of reserve capital. It also isn’t necessarily efficient for an economy to maintain production capacity for a vital product (food, for example) when specialization and trade are viable options. But redundancy and other inefficiencies can often buffer a system from unexpected shocks.

Posted by engineer27 | Report as abusive

@engineer27: I think one of the most important things that is going to come out of the 2000-2020 timeframe is an understanding that economists do not have the tools yet to predict failure or poor performance of their systems to anything like the precision and accuracy that they believe they currently have.

Modern “quantitative” engineering is a couple of hundred years old now. During that period, engineers have learned that models and calculations only take you so far – there are too many variables to nail everything precisely. as a result, the safety factors and redundancy you mention are essential to having a functioning system with longevity.

The 1930s Glass-Steagal provided the equivalent of an engineered financial system with safety factors and redundancies built in. It worked very well for over a half-century. During that period, the economists went mathematical and believed that they could accurately predict real responses to policy changes. They then decided they could become Icarus, repeal Glass-Steagal and fly close to the sun. We have now rediscovered that those engineered factors of safety and redundancies in Glass-Steagal were the primary factors preventing financial system collapse, although very few economists and politicians want to admit that.

The RMBS, debt-ceiling, government pension, European bonds, and Wall Street derivatives crises are likely to drive home the point over the next decade that a new Glass-Steagal Act is essential for democratic capitalism to survive. Unfortunately, the financial and economic carnage to achieve that end game winning position is likely to resemble the ’30s.

Posted by ErnieD | Report as abusive

As a matter of curiosity, how do you explain all the bank failures in the 80s and the relative robustness of non-US, non-UK banks in the crash if Glass Steagal was a casual reason for stability? What was that about data and theories?

Posted by Danny_Black | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see