This page has been archived and commenting is disabled.
Never Speak Ill of the Dead, But ...
From The Daily Capitalist
Paul Samuelson died yesterday at the age of 94 as I am sure you have heard. He was very successful in his field, having won a Nobel in economics and his economics textbook, Economics: An Introductory Analysis, sold over 4 million copies. I used his textbook in Econ 101.
His contribution to economics was to make it a science, much as physics, by bringing mathematical analysis as the prime means of empirical research and as a means to the truth. He even applied the law of thermodynamics to the study of economic equilibrium. He was very influential and mentored many of today's economic leaders, including his nephew, Larry Summers, and Ben Bernanke.
There is one problem with his work: he was wrong. He did more to undermine the study of economics as a means to truth than anyone else. His textbook penetrated the globe and still is representative of the predominate school of economics. He was a Keynesian, or Neo-Keynesian, who believed that you could accurately predict human behavior and originate economic technocratic policies through "science." It gave rise to econometrics and the influence of schools like MIT which perpetuate these fallacies.
It is a shame that instead of following Mises and Hayek, the world went Samuelson's way. Samuelson never "won" the debate over epistemology with Hayek or Mises because, as far as I know, he never knew of or addressed the significance of these issues, other than to make fun of them. Now we are reaping the rewards of the limits of mathematical analysis as the generations of economists in charge of business and government policy have blindly followed this false science.
As a tribute to Hayek, I urge you to read his Nobel acceptance speech on the limits of so-called science as it applies to the social sciences. It is a devastating rebuke of Samuelson's econometrics.
»
- advertisements -
Every action creates reaction.
Whatever measures adopted by those policy makers, economists, dealers, bankers, etc to combat and manipulate the economic condition will eventually create counter measures and unforseen consequences.
We would be better off to bite the bullet now in stead of kicking the can down the road. What happens when the can drops off the red cliff?
Yes, all empires collapsed at the time when the can finally dropped off the red cliff. BO could be the last president of the US of A.
Well, he had a disadvantage: he never lived under an authoritarian system. In the end, it turns out that ideology matters more than logic. You either believe that the government is a force for good, or you don't, the rest is just fitting the unclear "facts" to your believes. Unlike this idiot I was "lucky", I was born in the USSR so I knew what government could represent. Eventually I learned that it winds up representing it everywhere. Power corrupts and all that.
I can't quite honestly wish for him to burn in hell, mostly because I don't believe in hell, but also because he didn't know what he was doing so it wasn't really a sin. Keynes on the other hand knew. There is only one way to treat socialists, like mad dogs, whatever their true beliefs are. Whenever you hear someone suggesting that the government should solve this or that problem, and it's neither national defense nor road improvements, spit or worse in their general direction. Good bye Samuelson, and thanks for all the damage you've done.
What priviliges national defense and road improvements to exempt them from your frothing? Why can't Blackwater mercenaries protect us instead of the US Army? In point of fact, aren't they doing that right now?
Economics has become nothing but a pseudo-scientific justification for government interventionism, empire building and power grabbing, and a public works employment project for economists.
The most amazing thing is they actually believe the crap they spout. It's not surprising that JP Morgan et al got the Fed created to socialize their losses but it's is simply mind boggling that it's now taken as gospel that it exists for the public good.
Bravo, sir; you speak the truth. Your post is the antidote to Leo's deification of Samuelson. He was a man, not an economic god with infallible tenets & ideas. His ideas have failed the crucial test of time. The true tragedy is, his ideas have spread like a virus worldwide to effect all economists which has resulted in "group think" for the lot of them. Group "thinkers" are not necessarily great problem solvers as we have witnessed, rather they apply & define the definition of insanity, repeating the same thing over & over again expecting a different result.
From the cited speech of Hayek, The Pretence of Knowledge:
And this one:
I ve studied econometrics. It's pretty much application of statistical science for economic purposes. Now as any other tool, it is only as good as the end user of such a tool. Besides, if any of so called "banksters" would actually use it, they would be able to survive without gov bailouts.
At any rate, econometrics is used by all economic "schools", it's a valuable tool to test your ideas. And if someone uses it to make up some data - Climate Debacle, well don't blame the car, blame the driver.
p.s. you might be barking at the wrong tree and as far as I know, Paul S. tended to stay away from politics or keeping any official government position.
I do believe that DINOSAURS must die, however an anger is a bad substitute for a reason. .
Tu ne cede malis sed contra audentior ito
A fantastic rant, Econophile !
This schumpeterian loved it and can only add that Socionomics masterfully fills the void of uniting the social sciences themselves with actual scientific classification ... where Samuelson so desperately tried / failed.
Great piece, thanks for it.
Excellent post, Jeff. In life or in death, we must repudiate false ideas and the men who perpetuate them. I think a fitting afterlife would be for him to spend all of eternity with JMK and a gaggle of central bankers. Let us pray that, if the 1930's depression was the rise of Keynesianism, perhaps this latest depression will usher a return to the Austrian school and fiscal and economic sanity.
Oh yes...
Austrian school econ is sanity.
Disagree.
In the hands of a good analyst who understands the limits of what implications are reasonable to take from the analysis it is a valuable tool that helps society and limits the foolishness of those who spout "theories" (and a bad theory is just as bad as a bad analysis )
In the hands of an arrogant fool bad analysis and bad use of analysis does not help society.
For all his brilliance, name a society that can implement Hayek for any length of time and have enough stability to provide social justice and political stability?
An econometric view uninformed by political economy (Polyani, Rawls,etc) is only good for studying the impact of say farm subsidies...not runing a country
Lastly, since when have banks, unharnessed by politicians needed econometrics to screw up their generation of society? Panic of 1837 require econometrics? Dutch need econometrics to screw up? Did we have black box models and quants between 1865 and 1910? In 1929-1941?
Samuelson believed in a tool kit and rigor. That is a contribution in the hands of those who are rigorous. Polemicists have used whatever data was available to argue and advocate in the service of power or greed since 2500bc
Steve Keen, who's developing computer and mathematical models of his post-Keynesian ideas, predicts that unless the post-Keynesians come up with sophisticated models of their own then the neo-classicals will once again win by default. (I'm sure the Austrians are in the same position.)
http://www.debtdeflation.com/blogs/2009/12/16/circuit-theory-and-the-sta...
Probably a nice introductory textbook incorporating the models would be a very smart move too. It all shows how effective Samuelson's approach was.
There is nothing wrong with the use of mathematics in economic analysis. The problem is when you use it as a substitute for theory. The issue is a rather large and important one. If you believe we humans live in a mechanistic world where we are subject to certain laws of behavior as in the physical sciences, then it should be easy to use math to figure out what to do. But that's not the real world. We have free will; whatever you predict that you think I will do, I guarantee that I'll do something different. Also, you can never know if your models work. If the economy consists of millions of decisions made every minute of the day, how can you know if you've got the right data much less complete data? You can't and that's the flaw. Yet modern economics is built on this mechanistic foundation. Mises and Hayek and other Austrians have demolished this theory rather convincingly. Assuming that you have honest and benign technocrats making economic decisions based on mathematics rather than on human behavior, it will never come out right. You don't need arrogant fools to prove that. It's not a question of good people, its problem is that it doesn't work and never will.
We agree...you cannot make decisions based on mathmatics. Not the way humans work
However, mathmatics can help you make more well examined decisions and weed out bad ideas upon rigorous examination faster than humans can and less prone to the emotions of the moment than humans will
We are talking about probability and optimization applied to essentially political decisions about resource allocations by societies...not mathmatical proofs or perfect predictions. (we agree that is not possible)
Absent rigor the profession is half polemic and half political science. If you follow the logic that you can never know if you have the right data and complete data...you essentially are paralyzed from making any decision.
In life you do not know if you will make it crossing the street....so you make a decision based on probability. Same difference. Same goal. To have more inputs from which to make "somewhat" better decisions granted that pefect ones are not possible.
btw, your solution is not the real world. Decisions without perfect information have to be made all the time. Not using all available tools, (including logic and gut instinct) is not real world either.
Over reliance on models is foolish. So is thinking that there are enough Hayeks around to run a complex world. So is thinking you are not predictable. A bus is speeding down the middle of the street and you are stepping off the curb....tell me about your unpredictability?
Could you cross the street? Yes you have free will. Now lets examine the need to put a stop light on the corner where city busses often speed down the street. That decision is based on the probability of the city bus speeding, the predictability of how many city busses speed and the predictability of how many men of free will ...will do unpredictable things X the payout the city subsequently has to pay to the widow. Hence a model ...despite not having complete or even the right data about what you, as a man of free will...might do when seeing a speeding bus.
Thanks for playing
Great post, but I have to take issue with one thing. It seems in your back and forth with econophile, you're treating the human predictibility problem like a light switch. We're either predictable or we aren't. Obviously there is a degree of predictibility for everyone, and it's probably different for every person. On a scale of 1-10 with one being "I have no way to know if you're going jump out of the way of the bus or let it hit you", and 10 being "I know the exact minute that you will scratch your butt", how predictiable do you think we are in general?
The problem with your side is that there aren't enough variables to account for the flapping of butterfly wings in china.
Simply put, not everybody goes to lunch at exactly 12:00, every day of the week, nor do they eat the same thing. Models do not account for these types of variables, and make far too many assumptions to be of any use.
Climate change models have exactly the same problem.
A) Don't have a side
B) Talked about rigor...not models. But anyone who is a good analyst understands what models can and cannot do....indicate during stable times and help one narrow the odds of an outcome...not eliminate, not predict during times of uncertainty...they are just another indicator an analyst considers
C) Any models in biology, transportation, purchasing, supply chain optimization, epidemiology, crime control, google search engines and production line efffectiveness and quality control much less oil exploration are of great use.
D) We agree there are a legion of examples of models badly used...and the same is true of any other tool invented by or used by humans
Well said: Thank you!
'Twas a fatal conceit.
- Heretic