1. Skip to navigation
  2. Skip to content
  3. Skip to sidebar
Source link: http://archive.mises.org/8298/austrian-realists/

Austrian Realists

July 17, 2008 by


Anyone who peruses the top mainstream economics journals will quickly realize that economic theory has been crowded out by mathematical formalism. The neoclassical economists’ uncompromising quest for precision in their models has been achieved at the expense of the accuracy of their predictions. The old joke of the drunk searching for his keys near the streetlight is an appropriate metaphor. FULL ARTICLE

{ 22 comments }

Stan Warford July 17, 2008 at 10:34 am

Great article. But, what’s this inside joke about a drunk looking for his keys?

RKArmour July 17, 2008 at 10:49 am

It’s really old. I remember seeing it on a black and white television show, but it may go much further back.

The straight man asks why, if the drunk lost his keys by the dark door stoop, is he looking under the streetlamp.

The drunk responds that it’s too dark to find them where he dropped them, so he’s looking in the light.

Book 'em Danno July 17, 2008 at 10:52 am

This topic, Dr. Murphy, is timely. I recently encountered scientists in education policy who simultaneously claim to reject the most glaring aspects of logical positivism but then engage in apologetics for its modern offshoots- including some aspects of Augustus Comte thought.

William Shadish, for example, teaches “Experimental and Quasi-Experimental Designs for Generalized Causal Inference”. In this work he lists and reacts to the many problems of theoretical and human error in lab experimentation. “Gone, too, is the belief that the power experimental methods often displayed in the laboratory would transfer easily to applications in field settings,” Shadish writes.

Fair admission.

Shadish says too that “…all scientists are epistemological constructivists and relativists…ontological realists but weak epistemological relativists.”

To be clear, it is key to know that the balance of Shadish’s critique of experimental science is referring to lab and physics. But what is his book about? Conducting social science. I believe, at this point in my reading, that Shadish fails to differentiate between the search for physical cause and that of human origin.

Is it merely a clever obfuscation by Shadish and other scientists or an honest attempt at resolving what you have clearly explained as an unbridgeable gap between Realism and fictional crutches? What role should modeling play in social science?

Thanks again, Dr. Murphy!

Jim Lippard July 17, 2008 at 11:01 am

Nassim Nicholas Taleb’s book, _The Black Swan_, is an extended response to Caplan’s argument. You can’t use mathematics that presupposes Gaussian distributions on phenomena that doesn’t follow Gaussian distributions. In Taleb’s terms, Caplan is using mathematical formalism for “Mediocristan” in an “Extremistan” environment, which leads to failures like Long Term Capital Management occurring far more frequently than expected.

Landsburg’s argument makes no sense to me at all. If I have three children and love one of them slightly more than the others, would he suggest that the best way to demonstrate that is to only give birthday presents to the one?

I give to charitable organizations that I support because I want to see them make a difference. My information about how effective they are and, more importantly, how effective they will be in the future with my contributions, is imperfect. I also support more than one cause, and I apportion my donations among them for at least two reasons. First, to increase the probability that at least one of the supported groups will make progress in its cause, and second, because I want to see progress made in all of the causes I support. What reason is there to take a winner-takes-all approach to charitable donations rather than diversify, regardless of whether my motivation is altruistic or selfish? (And I don’t think the fact that I share the aims of a charitable organization and want to see it succeed makes it “selfish” in the ordinary sense of the term. I am happy in response to its successes because I share the goals, but I’m happy because of the success, not simply contributing in order to generate the subsequent feeling of happiness.)

Fephisto July 17, 2008 at 11:30 am

The fact that Austrian economics is based on a fundamentally sound axiomatic basis makes it, IMHO, a lot more mathematical than any neoclassical theory.

Stranger July 17, 2008 at 12:32 pm

One of the Mises Institute epistemologists needs to urgently look into physicist Stephen Wolfram’s work. He wrote a book called A New Kind of Science in which he arrives at the following conclusion:

1 – That most physical processes in the universe are maximally complex, and prediction is not possible.
2 – That they cannot be modeled through mathematics but only through explicit computational simulation (i.e. deduction).

Oh, and he is also the multimillionaire inventor of the Mathematica software package, which makes him an authority on math as well as physics.

http://www.wolframscience.com/

Ken Zahringer July 17, 2008 at 12:40 pm

Great article, Dr Woods! We cannot overemphasize this issue. I had the same argument when I got my MA in Sociology thirty years ago. Data about race, gender, income level, and education may be easy to gather and easy to manipulate statistically, but that doesn’t mean they actually explain anything! Praxeology is the only approach to social science that makes any sense to me.

fundamentalist July 17, 2008 at 12:44 pm

Nice rebuttals! Thanks! I think Arnold Kling had a good response, too.

If Caplan knew anything about econometrics, he would know that all models have error terms. Most modelers look at the error to make sure it’s normally distributed and their model isn’t biases somehow, but what the error tells you is that you haven’t included some important variables. If you had all of the variables, you would have no error term. That’s why the r-squared value is often around 0.6 in economic models. That means that your model explains 60% of the variation in the variable you’re trying to predict. What explains the other 40%? Well if we knew that we would include it in our model! Caplan wants us to model that unexplained 40%. What a knuckle-head!

You’re right that “… there are all sorts of motivations that could acquit philanthropists of Lansburg’s charge…” Guessing people’s motives is one of Caplan’s favorite magic tricks, too. But economists should not try that without the expert supervision of the masters of motivation diving—fiction writers. Fiction writers understand that it’s possible to attribute any motive to any individual if you’re creative enough. Economists just don’t have the imagination to do that well.

rtr July 17, 2008 at 3:50 pm

Answer to the Trojan Horse.

It’s called *subjective valuation*. There is no mathematical probability distribution for calculating changing subjective valuation. QED. Any assumption to the contrary is fundamentally, fatally, flawed. There is no mathematical omniscience which can model subjective valuation.

Mathematical model assumptions are a regression to the pre-scientific era before marginal utility was established. Mathematical probability models are utterly perplexed as to how a diamond can be worth more than a glass of water. Mathematical probability assumptions are nothing more than hidden fixed objective “use-value” assumptions.

This is precisely why mainstream monetary theory was built on a huge error suggesting a mathematical *equation* of value between things traded for money. Trade is an action of preference, not an inaction of indifference. But we don’t see mainstream economic formulas with “greater than” and “lesser than” mathematical signs, especially when the exact same good traded is valued differently, both “greater than” *and* “lesser than”, by two different individuals.

In economics, A is greater than B, and B is greater than A. Mathematics is thus killed.

averros July 17, 2008 at 6:40 pm

> The fact that Austrian economics is based on a
> fundamentally sound axiomatic basis makes it,
> IMHO, a lot more mathematical than any
> neoclassical theory.

As a mathematician, I’ll second that statement.

Mark D. Hughes July 17, 2008 at 6:52 pm

I simply do not get Landsburg’s reasoning:

{If you care about the recipients, you’ll pick the worthiest and “bullet” (concentrate) your efforts. But if you care about your own sense of satisfaction, you’ll enjoy pointing to 10 different charities and saying, “I gave to all those!”}

This may be true if you are relying on “your own sense of satisfaction” or self aggrandizement. But so too could a bunch of other options – self satisfaction is the ultimate in subjectivity.

{If your charitable contributions are small relative to the size of the charities, and if you care only about the recipients (as opposed to caring, say, about how many accolades you receive), then you will bullet all your contributions on a single charity.}

This is simply nonsense. If what you crave is “accolades” then making small donations to multiple charities is going to be a slow (and perhaps futile) process. As anyone who follows the world of philanthropy knows the fastest way to get noticed is by making one big splashy donation.

I venture to guess that I’d get notices allot sooner and with greater “fanfare” if I gave LvM Institute $40,000 rather than ten free enterprise oriented institutions $4000 each.

rtr July 17, 2008 at 7:49 pm

Mark D. Hughes shocked the academic world earlier today, with his record $50 billion dollar subjectively valued post to an economics thread on mises.org, with a multitude of media and cameras present. The event included plenty of ticker tape, and lots of cheering fans, as a circus procession to the administrative headquarters made its way for the presentation of the post. Said Hughes, as he presented his charitable gift, “As anyone who follows the world of philanthropy knows, the fastest way to get noticed is by making one big splashy donation.”

TLWP Sam July 17, 2008 at 8:24 pm

This reminds of an article by a Libertarian who was equally against charity as much as he was against welfare. It’s a zero-sum game – if people wanted to help then give them a hand-up not a hand-out. He likewise despised the notion the rich have to ‘give something back to society’ as implies the rich stole something from society. I’m sure you get the idea.

hayesy July 17, 2008 at 9:36 pm

I simply do not get Landsburg’s reasoning

Because it’s crap. You can’t generalise regarding the most “efficient” distribution of charitable donations, it’s all subjective marginal utility. If I give 500 dollars to a cancer charity and 500 dollars to a heart disease charity, it’s because the marginal cost of the 501st dollar to either charity is greater than the marginal benefit of the 1st dollar to the other. If, on the other hand, my father died of cancer, I may be more inclined to consider my 1000th dollar of greater marginal benefit to the cancer charity than any other. Not all charities are created equal. Not all of them use the 1000th dollar as efficiently as they use the 500th. 500 dollar units have differing marginal utility to the differing charities in question, regarding the expense involved in their particular activities (eg, fighting world poverty vs building a local basketball court). And people don’t have perfect information regarding how close to a “breakthrough” any particular charity is regarding the battle against cancer, heart disease, etc, which would justify the 501st dollar as possessing superior marginal benefit, hence the “scattergun” approach. All this “bullet” pontificating is just more publicly-sponsored mainstream bilge. No doubt he’ll write a bestselling book about it, and his readers will glow with pride over how much they know about economics.

rtr July 17, 2008 at 10:39 pm

This just in from our Morse code Atlantic cable: “PWN :P”.

Fephisto July 18, 2008 at 12:48 am

Averros: I’m going into grad school in mathematics, and I know of at least three other math people in the forums.

Perhaps we should start up an “Austrian Economics for Math Students” sub-movement? :p

wuzacon July 18, 2008 at 4:59 am

Landsburg’s reasoning doesn’t make any sense because he is not really reasoning. He is using his mathematical, “logical,” and econometric model to bring his readers to his intended outcome. It doesn’t matter if the origins are specious, only that the answer is a bigger government.

His intended outcome is not for people to give to the “one” charity of their choice, but rather to pontificate for the benefit of higher taxation by a single government, which will then be able to decide the best use of the money. He probably disfavors the idea of multiple taxing districts as well, and would rather see a single, large, all-inclusive government entity that can direct all “charitable” investment in the proper direction to maximize social benefit.

Thanks, Dr. Murphy and others for cutting this off at the knees.

Paul Marks July 18, 2008 at 8:01 am

Yes.

When Caplan and co demand something be put in “standard probability language” they have ASSUMED that their way of the study of economics (i.e. using such language) is correct.

Inquisitor July 18, 2008 at 9:27 am

Maybe I am stupid, but could someone explain what Caplan’s objection is exactly, and how Murphy has rebutted it? TBH it sounds like more nonsense on Caplan’s part.

Befree July 20, 2008 at 7:58 am

To a man whose only tool is a hammer, every problem looks like a nail.

Murph August 1, 2008 at 2:47 am

Charitable giving is driven by emotion and empathy, not Reason.

Landsburg is trying to pin one of two Rational motives on charitable giving:

(1) Solving a problem (curing cancer, etc)
(2) Feeding one’s ego

But charitable giving is not an act of reason, it is an act of compassion. Most people do not have great sums to give, but give anyway, because they want to help, even if in a small way.

If most charitable givers were to base their decision on Reason alone, as Landsburg would have them do, then, Reason would inform them that their personal total annual giving to all charities combined of a few thousand dollars is meaningless in the big picture, and so: they would give NOTHING to anyone !

Thankfully people are not the soul-less drones that Landsburg would have them be, once he and his like run our lives.

Steven E. Landsburg April 15, 2010 at 10:12 pm

The point here is not to offer a rival theory. Rather, my point is that people do not approach charitable giving the way Landsburg assumes in his model.

But Bob, that’s my point too! The argument is a reductio ad absurdum. I start from the assumption that givers care only about the recipients, and deduce that they would behave in a way that that they don’t behave in. I therefore conclude that they don’t care only about the recipients. In other words, my point is that people do not approach charitable giving in the way the model assumes. I’m glad you agree.

Comments on this entry are closed.

Previous post:

Next post: