Samuelson fashioned his models, which set the standard, after 19th century physics. Functions were assumed to be smooth and continuous. Economics was reduced to various types of the same calculus problem: finding a constrained extremum. The economist’s job was to state the objective function and the constraints, then grind out the solutions.So true, and yet 90+ percent of economistsThis required considerable mathematical ability and stomach for tedium but little imagination and no familiarity with economic reality.

By the 1960s, if not earlier, academic economists who quarreled with this way of doing the job were, as Roy Weintraub put it, “regarded by mainstream neoclassical economists as defenders of lost causes or as kooks, misguided critics, and anti-scientific oddballs.” By aping 19th-century physicists, neoclassical economists convinced themselves and others that they were doing science, but the effort was basically misguided, not so much scientific as, in F.A. Hayek’s term, “scientistic.”Human beings, purposeful and creative, are not like atoms; nor is a market analogous to a physical or chemical system.In the view of Hayek and his teacher Ludwig von Mises, neoclassical economics is, in critical respects, pseudo-science.

*still*try to put our social interactions into physics equations. And they fail to understand, explain or predict.

Physics gave up on this calculus approach with Feynman in the '60s while biology shows daily that this calculus approach does not apply to biology unless you modify it a lot and use computers.

ReplyDeleteIn microarrays and other parts of biology, there are levels at which calculus applies but in the good problems the number of calculus equations to solve is in the thousands or tens of thousands, so pencils and 19th century approaches do not work.

You should stick to water economics.

ReplyDeleteYes, modern econ has gone too far with mathematics, but I don't think that is true for the era when Samuelson was changing the discipline.

@Jeff -- water economics is screwed up b/c of the way that Samuelson changed the discipline. I am sure you will understand if you compare Samuelson's paper on public goods to Ostrom's work on common pool goods (which is the same in terms of dilemma, but different in terms of result).

ReplyDeleteThanks for helping me clarify why that post is relevant.

Great to see this, David, "autistic" is precisely the right term. Much of the bad reputation of contemporary economics is due to Samuelson's autism. Here is an excellent article that expands on the flaws of Samuelson's approach,

ReplyDeletehttp://www.independent.org/newsroom/article.asp?id=2680

MS,

ReplyDeleteNice article that you referenced.

Ignoring Samuelson's personality, theorists of his time were constrained to thermodynamics and other mathematics because, without computers, they needed closed form solutions.

Only with computers (and some fairly big ones) could the behavior of solutions that were not closed form be pursued. A recent pursual was to model a universe in which the nuclear weak force would not exist. Such a universe is possible but to prove it requires a lot of equations, a lot of double checking, and a lot of computer time.

Fortunately, with careful science, we can now show which of the Samuelson generation's assumptions were over simplifications due to lack of computational power.

Eric: What is "modify it a lot" regarding calculus? I read a lot of biological science every day, and much of it uses calculus to understand biology. This is not to say that I do not agree with about 3/4 of the sentiment of David's argument.

ReplyDelete@Fixed carbon

ReplyDelete'Modify a lot' means that where many biological examples use one or a few equations and these equations are embedded in three dimensional space sometimes with time added, we (and others) use thousands of partial differential equations at a time (each one capturing a well defined physical process).

These equations, mathematically, live in thousand or more dimensional spaces and interact with each other through 'tensor-like' things. In addition, over time the data changes but so do the underlying equations. So the action occurs, carefully, in 'functionals' space, which turns out to be the real space appropriate to biology.

We basically have taken the kinds of calculus that biologists use and have extended it through physics and mathematics to much more complex spaces, spaces in which you can actually get the right answer and not have to make the kinds of assumptions that Samuelson and others made for complex systems.

Does this help?

Whether the writing is clear or not, ask questions. I wrote the above quickly, on my way to a meeting, and can write it better and differently if needed.

Thanks for asking.

When I taught economics classes at the high school level, I did so as a social scientist (motto: we are neither). On the first day of class I would explain just why economics was not a science, and I would do it because every high school text defines economics as something like, "the science of choice!" (middle-aged writers fervently believe that exclamation marks make teenagers more interested in the subject...)

ReplyDeleteI would also explain that a free market does not only require that everybody have an "informed self-interest." But, that's another story.

David: Thanks for clarifying. I was thinking more about Macro.

ReplyDeleteDavid (et al.)

ReplyDeleteI think what's fascinating about the post is that it shows the cross-cutting influence of both economics and physics across disciplines. Trevor Barnes just wrote a piece (2009) in the latest Professional Geographer about how the quantitative revolution (influenced by folks like Samuelson) in geography then had a binary separation when the "critical" folks (aka Marxist geographers) came along and started critiquing some models they didn't understand. The choice, then, became you were EITHER a quantitative ________ (fill in your discipline) or a critical _______ (your discipline), when this binary simply doesn't work. Sorry I'm late to post this... Eric P

This is an excellent discussion. I hope I am not so late in posting that everyone has moved on to newer posts.

ReplyDeleteI think the biological model of economics has merits. (We are all driven by biological processes after all.) The physics, the chemistry, and the biology happens at the micro level, not at the macro level. My guess is that calculus that tries to model aggregate behavior will often fail to capture or predict events. More computer power is helping us do simulations in many areas of science. Take weather forcasting as an example. We are not much better at producing long range forecasts than we were before using detailed models, but we have a much better understanding the range of outcomes of uncertain future events. I assume that economics, like weather, suffers from the problem of sensitive dependence of initial conditions. With so many variable to consider we are never very certain of the initial conditions on the micro level.

Since each of us have non-linear utility functions for many aspects of our daily lives, that are subject continuous change, knowing existing conditions is a practical impossibility. However, working with large numbers of heterogeneous agents, approximating groups of utility functions, and running repeated simulations, I beleive we can gain enough insight to make fewer policy errors, and have less severe unintended consequences.

Bottom Line: The lure of centralized, scientific decision making by enlightend leaders, as economists from John Stuart Mill, to Keyenes, to Samuelson, to all of President Obama's advisors have beleived, has great mass appeal but there is little empirical evidence to support the notion.

@Jay,

ReplyDeleteOne piece that I left out is that while the behavior of thousands of careful equations does not have a useful analytic solution, often, the set of equations yields strong bounds on the solution.

So you can make statements, with real support, such as

'The CPI next year will increase by 3% +/- 1.5%'

The difference in this statement with respect to a Samuelson statement is that this statement comes from the aggregate behavior of thousands of agents not from a linear fit to last year's numbers.

As to drifting off to more recent posts, I am looking for discussions not fads and so I continue to pay attention to whatever posts intrigue me.