In microeconomics, humans are assumed to be "rational" or "self intersted". To this end, they will seek to maximize their pleasure or their "utility" against certain constraints. Where people draw their utility depends on their preferences. One could even derive utility from helping others, thus even good samaritans can be totally self interested.

Is this assumption valid? Well that's very hard to say since it's impossible to test. Neoclassical decision theory is unfalsifiable. Let's say I try to shoot myself, then they couldn't be self interested could it? Wouldn't I be mad? Not according to the resiliant microeconomist. I simply have a preference for not being alive and I draw utility from acting in accordance with that preference.

This holds for any activity you can think of. Anytime you see someone acting outside of what you think would be "self interest", a resourceful microeconomist could wave his hands and come up with a story of how different preferences could lead a self interested individual to doing the seemingly self destructive. This is the basic model of economics, and many other important models flow from the conclusions of this model.

So that leaves us with one question:

How is economics scientific? Surely any good scientific theory must be falsifiable. We must be able to test it and that test have the possibility of failure. The theory of gravity predicts an apple will fall to the ground, but it is possible that, if the theory were false, the apple might hang in midair. Neoclassical decision theory doesn't have this characteristic.

Not matter how many tests you do, an economist can always say the person was acting rationally according to his preferences. Thus, no test could ever show Neoclassical decision theory to be false. kwsmith2, an economics grad student, did a good job of illustrating this line of argument in the thread on selfishness and capitalism.

So how can economics (or at least this particular theory of economics) be considered scientific? How could one ever test a theory that, by its very nature, can't be tested?

What a sad reality for economists. Or is it?

[Edited on July 5, 2005 at 5:30 PM. Reason : ``]

7/5/2005 5:25:03 PM

duh

7/5/2005 5:33:07 PM

It may fall into the same hole as relativism, what if someone gains pleasure out of getting rid of pleasure?

7/5/2005 5:33:16 PM

What, did you read the introduction to Freakonomics? Good job.

7/5/2005 5:36:47 PM

haha

7/5/2005 5:39:07 PM

I doubt Steve Levitt would come up with this shit. It might affect his book sales. Sell Out.

7/5/2005 5:44:41 PM

So this isn't known by everyone?

hmmm.

7/5/2005 5:47:36 PM

His book is still interesting.

7/5/2005 5:47:51 PM

economics is a social science

7/5/2005 7:03:41 PM

Economics is just a tool for social exploitation.

I'll make a very good quote on this once I get a little crunk

Thanks to my wallet's economical conditions.

7/5/2005 7:31:52 PM

Quote :

"What, did you read the introduction to Freakonomics? Good job."

everyone should read this book

7/5/2005 8:00:36 PM

Economics is not scientific in the manner that physics or other hard sciences are, but that doesn't mean it is useless. You say any good scientific theory must be falsifiable. This is true when you are studying subjects like gravity or light particles where the assumption is that, assuming full understanding of all other forces, every experiment and set of actions is completely reproducible and falsifiable. That is the objective nature of the hard sciences. Economics on the other hand tries to study what I think of as the function of the human mind, which is not objective. Human minds take in sets of inputs and through some complicated process end up with an output. One person may take a situation and react one way, another person may take the same situation and react another way, and even the original person on a different day under different circumstances may even act differently. The study of economics is based on generalization and summarization, not the pursuit of set in stone scientific laws. Thats why you will graduate in the CoM ceremony and not the PAMS one yo!

[Edited on July 5, 2005 at 9:38 PM. Reason : .]

7/5/2005 9:36:39 PM

But economics doesn't seek to prove that people are rational and will always seek to maximize their utility. Rather economics assumes that people are rational in order to speculate as to how people will operate under a specific set of conditions. Given that the goal is not to prove that people are rational, such an assumption need not be falsifiable, especially given that the assumption is merely made to simplify the equation and there is an understanding and acknowledgement that not everyone will act rationally with respect to a given frame of reference.

7/5/2005 10:38:07 PM

^hence, hardly different from any soft science such as psychology. The point in ^^ was to show that the real difference between a real science and a soft science was pursuit of a natural law. Falsifiability is neither here nor there.

It's simply a social science heavily cloaked in mathematics.

7/5/2005 10:58:07 PM

Quote :

"Is this assumption valid? Well that's very hard to say since it's impossible to test. Neoclassical decision theory is unfalsifiable. "

This isn't true.

The axioms of consumer choice are testable.

Consider the Axiom of Transitivity: For any three elements a,b, and c, if a is preferable to b, and b is preferable to c, then a is preferable to c.

Many, if not most, microeconomic books will acknowledge this is a "controversial" axiom: more specifically, there have been experiments, which have shown consumers' choices are not always transitive. Nonetheless, it's not a particularly horrible assumption. First, it jives with intuition. Secondly, in spite of the behavioral experiments showing otherwise, there are still quite a few experiments showing it's a reasonable assumption.

Moreover, the rest of the axioms of consumer choice (at least those used to derive basic microeconomic results) are for the most part, falsifiable. Those that aren't are simply used for mathematical regularity.

[Edited on July 5, 2005 at 11:11 PM. Reason : edit]

7/5/2005 11:04:41 PM

^^ I was actually responding to socks, not arguing with scott

7/5/2005 11:22:03 PM

you should go with what scott said

7/5/2005 11:31:44 PM

Keynes, you answered a complaint I didn't even make.

The axioms of preference (like transitivity or completeness) dictate the shape that a "rational" utility function can take. That isn't what I'm talking about.

I'm talking about the assumption that people seek to maximize thier utility (that's the part of the math problem where you maximize the utility function subject to the budget constraint). In other words, that people will always follow their own "self interest"--broadly defined. No matter what a person does, from buying coke to shooting themselves in the head, you can always say that they were maximizing their utility (like kwsmith2 did in the thread on capitalism and selfishness). IOW: it is unfalsifiable.

I think I made that pretty clear in my post. It's a shame you wasted all that hot air.

[Edited on July 6, 2005 at 12:59 AM. Reason : ``]

7/6/2005 12:56:02 AM

A good way to tell whether an action maximizes a person's utility is to examine his purpose for committing that action. If the action itself contradicts the reason for committing that action, then it is generally considered irrational economic behavior.

An well-informed investor could say, “I want to have a low risk stock portfolio.” In an effort to achieve this, he goes out and purchases a high-risk stock portfolio. Meanwhile, there is another stock portfolio, with a much lower risk and same rate of return, which is left untouched by the investor. This would generally be considered irrational economic behavior.

Or consider the case of suicide. A person could shoot himself for the sole reason that he thinks space aliens told him he would join Johnny Cash in heaven. On the other hand, a person could commit suicide because he has a debilitating, interminable disease. The first case typically does not fall into the bounds of rational behavior, as it is very unlikely space aliens were egging him on. The second case could probably be considered rational, I don’t know.

[Edited on July 6, 2005 at 1:44 AM. Reason : edit]

7/6/2005 1:37:18 AM

^

1) don't confuse rational and reasonable. It isn't reasonable to believe aliens are sending messages to you, but you would still be acting "rationally" if you were optimizing your utility by killing yourself at their bequest.

2)

Quote :

"An well-informed investor could say, “I want to have a low risk stock portfolio.” In an effort to achieve this, he goes out and purchases a high-risk stock portfolio. Meanwhile, there is another stock portfolio, with a much lower risk and same rate of return, which is left untouched by the investor. This would generally be considered irrational economic behavior."

Maybe he didn't know about the other investment, maybe he didn't like the guy who was selling it to him, maybe he was just lazy and got more utility from buying the first portfolio he sees, maybe maybe maybe maybe. There are so many maybe that it's very easy for a resourceful microeconommist to tell a story that fits the "self interest" assumption--even easier because we cannot directly observe the process. I hope you see the difficulty in trying to second guess human behavior.

[Edited on July 6, 2005 at 2:44 AM. Reason : http://cepa.newschool.edu/het/essays/get/consumer.htm for the uniniated]

7/6/2005 2:38:58 AM

scottncst8

Quote :

"Economics is not scientific in the manner that physics or other hard sciences are, but that doesn't mean it is useless. You say any good scientific theory must be falsifiable. This is true when you are studying subjects like gravity or light particles where the assumption is that, assuming full understanding of all other forces, every experiment and set of actions is completely reproducible and falsifiable. That is the objective nature of the hard sciences. Economics on the other hand tries to study what I think of as the function of the human mind, which is not objective. Human minds take in sets of inputs and through some complicated process end up with an output. One person may take a situation and react one way, another person may take the same situation and react another way, and even the original person on a different day under different circumstances may even act differently. The study of economics is based on generalization and summarization, not the pursuit of set in stone scientific laws. Thats why you will graduate in the CoM ceremony and not the PAMS one yo!"

huh? do you know what falsifiable means? If so, are you claiming that Economics (and other "soft" sciences) should make claims that are not testable? If so, then how do we know that one description of human behavior is better than another?

And are you sure that the theory of gravity is really falsifiable?

7/6/2005 2:56:06 AM

define economics please

7/6/2005 3:00:42 AM

^ plz to read the first god damn post to know what i'm talking about. I was very specific about what I ment.

7/6/2005 3:02:22 AM

well then you cant call economic unscienitfic, just a certain area. i think the concept of supply and demand is very scienific, the statement that supply decreases prices is testable.

so i think youre really arguing against

Quote :

"Neoclassical decision theory"

which really isnt synonymous with the whole field of economics

7/6/2005 3:27:41 AM

Socks, you're still ignoring the fact that economics does not seek to prove that humans are rational. Rationality is a variable in the equation, not the end result of the equation, and thus it does not matter if rationality is falsifiable because that isn't the object to be proven.

7/6/2005 3:28:01 AM

^^ Josh, "neoclassical consumer choice theory is not scientific" wouldn't fit in the title box. My title also gets more clicks. The post clears up any confusion the title might give. Sorry you're too god damn lazy to read it.

Also, you might want to click that link i gave in my post before last. You'll see that the Demand function you love so much is DERIVED from consumer choice theory. You should brush up on yer econ.

^ 1337, huh?

[Edited on July 6, 2005 at 3:40 AM. Reason : ``]

7/6/2005 3:38:11 AM

Quote :

"Josh, "neoclassical consumer choice theory is not scientific" wouldn't fit in the title box"

i think it would. but i just wanted to know if it was sensationalistic. post a link to a primer on this topic then. ive never taken an econ course. know of any at teh introductory level that actually teach something useful?

7/6/2005 3:45:09 AM

As far as books go, I'd recomend "Understanding Economics" by Lester Thurow, Economics by Thomas Sowell, or Naked Economics.

Good websites are...
New School: http://cepa.newschool.edu/het/essays
David Friedman's online textbook for Microeconomics (best Micro textbook evar): http://www.daviddfriedman.com/Academic/Price_Theory/PThy_ToC.html
and here are some more online textbooks
http://www.oswego.edu/~economic/newbooks.htm

7/6/2005 4:06:56 AM

Quote :

"But economics doesn't seek to prove that people are rational and will always seek to maximize their utility. Rather economics assumes that people are rational in order to speculate as to how people will operate under a specific set of conditions."

-1337

But economics does seek to describe human behavior, and to that extent they assume people are utility maximizers, an assumption which cannot be falsified. If the assumption cannot be falsified, then how can we be justified making it?

The argument you're probably struggling to remember from your EC201 textbook is Milton Friedman's positivism, which states that the assumptions of our models don't have to be realistic or justified--all that matters are the predictions our models make.

But surely this can't be right either. If it were, then Chinese Accupuncture would be scientific because it makes some good predictions about pain relief based on the theory of the Chi in the body. Is there really such a thing as Chi of the body? Should we believe there is just because the theory makes good predictions?

Surely explainatory power counts for something in scientific inquiry. So why do so many economists follow friedman's half-baked philosophy? They would rather explore minimialist models than scientific truth.

7/6/2005 1:10:31 PM

Quote :

"They would rather explore minimialist models than scientific truth."

That pretty much sums it up. But supposedly the rational choice model is close enough. Like in a physic problem where they tell you to ignore friction or whatever. But I guess you don't by that...

Anyway, does this mean you're going to join the revolution now, Socksie?

Kris-style conditioning is quite scientific...

7/6/2005 2:35:35 PM

It don't mean butt if it ain't got that jutt

Quote :

"But supposedly the rational choice model is close enough. Like in a physic problem where they tell you to ignore friction or whatever."

The difference is that it is only simplifing the physics problem, we could still solve it if we didn't ignore friction. There is no way economics could solve it's problems without having to make it's assumptions.

7/6/2005 2:43:47 PM

There is no way physics could solve problems without its assumptions either.

Physics does not have a grand unified theory that explains everything and is correct all the time yet.

Hell Newtonian equations still work pretty well and theyre based on assumptions that are false.

7/6/2005 2:56:08 PM

I don't really know enough about physics to say one way or the other.

The real question is, when does the revolution start?

7/6/2005 3:08:31 PM

Quote :

"There is no way physics could solve problems without its assumptions either."

But science has a way of testing it's assumptions to see if they hold true.

7/6/2005 3:41:05 PM

i'm not reading this whole thread, but the whole point of economics is that it isn't perfect, and that it operates under a few basic assumptions to make approximate macro or micro analyses and projections.

7/6/2005 4:10:55 PM

Am I mistaken, or didn't a guy recently win a Nobel price for proving that this same underlying assumption is invalid? He basically proved that people are not inherently rational, and do not inherently act in their own best interests...

7/6/2005 7:26:43 PM

^must be getting easier to win nobel prizes, nowadays

7/6/2005 10:22:12 PM

Clear5, I'm not arguing that one shouldn't make assumptions when modeling human behavior (or anything else). I'm arguing that a particular assumption is unjustifiable because it's unfalsifiable.

and Lowjack struggles to contribute.

[Edited on July 7, 2005 at 3:16 AM. Reason : ``]

7/7/2005 3:10:48 AM

But it is falsifiable for a given set of paramaters in any particular instance. For example, if we establish that person A views having money as being the best possible utility he can have and give hima set of choices for which he is informed and he does not make the choices which maximise the ammount of money he can have, then you have shown that he does not act rationaly where rational is defined as working to maximise utility.

7/7/2005 3:33:04 AM

GoldenViper,
VIVE LA REVOLUTION!!11


actually i'm not much of an iconoclast. i don't actually think consumer choice theory is unfalsifiable, but everything else i said is true. I think Friedman's Positivism is wrong headed, minimalist models are useful but limited, and I think economists should focus more on developing models with more explainatory power. Oh and Steve Levitt is a sell out.

Of course none of these complaints are original, and not very outside the mainstream (unless you come from the Chicago School).

[Edited on July 7, 2005 at 3:49 AM. Reason : ``]

7/7/2005 3:39:17 AM

Quote :

"But it is falsifiable for a given set of paramaters in any particular instance. For example, if we establish that person A views having money as being the best possible utility he can have and give hima set of choices for which he is informed and he does not make the choices which maximise the ammount of money he can have, then you have shown that he does not act rationaly where rational is defined as working to maximise utility."

-1337

are you suggesting that it would be falsifiable if we were able to know what maximizes a person's utility (i.e. look inside their brain) and to see if they act accordingly??? please fill me in on how you propose to do this. Ask them? How do you know they are not lying or that one is fully aware of what will maximize his/her utility? People do say thing they don't mean afterall. HEY maybe you can look inside their head to make sure they're not lying AND to know what maximizes their utility!!!

Not exactly the most convicing argument.

7/7/2005 4:05:22 AM

Quote :

"I think economists should focus more on developing models with more explainatory power."

Such as?

7/7/2005 3:07:18 PM

^^ Hey, the rules of science don't say it has to be easy to falsify the theory, just that it has to be possible.

Furthermore, the only concern is if they are lying. It doesn't matter if they know something or not because the other part of the equation (which you are conveniently ignoring) is rationality is defined not only as working to maximise utility but also with the condition of given X knowledge and all other conditions being equal.

7/7/2005 3:19:40 PM

^ your "experiment" isn't testable

7/7/2005 10:29:32 PM

^^

here's a longer answer: there's no way to preform your test, since we can't read people's minds to know what maximizes their utility and asking them isn't an adequate alternative because they could be lying or they may not be selfaware. That's what i ment by "they may not know what maximizes their utility", they may not have enough "self knowledge" to articulate what maximizes their utility.

So your experiment is more than difficult, it is impossible to conduct. Better keep thinking.

7/8/2005 12:24:25 AM

^And during Newton's time we wouldnt expect humans to ever have the technology capable of proving his theorems wrong either.

The point is that we could eventually map the human brain or get inside people's heads or whatever and determine whether its false.

And secondly why does it matter whether it is scientific or not? How does that affect anything?

Mathematics isnt scientific or falsifiable either but we dont really use that to question the use of mathemathical models.

[Edited on July 8, 2005 at 1:38 AM. Reason : major edit at 7min]

7/8/2005 1:30:20 AM

Quote :

"or they may not be selfaware. That's what i ment by "they may not know what maximizes their utility", they may not have enough "self knowledge" to articulate what maximizes their utility. "

Impossible. The entire basis of the rational behavior is maximising your utility based on given information. If you are not self aware enough to know certain things that can maximise your utility and you don't perform those actions, then you are still acting rationaly because you did not know. Rationality does not imply clairvoyance.

7/8/2005 1:37:52 AM

Quote :

"And during Newton's time we wouldnt expect humans to ever have the technology capable of proving his theorems wrong either."

But newton was able to test his theory and it hold true in every case. Economics cannot do this. We aren't talking about proving economics wrong, the burden of proof is on economics to prove itself.

Quote :

"Mathematics isnt scientific or falsifiable either but we dont really use that to question the use of mathemathical models."

Mathematics is both scientific and falsifiable. I can have two apples and take one away and then I have one apple. This test will always hold true, thus proving 2-1=1. This was a very simplistic example, but you can test almost any mathematics in the real world, from geometry to algebra to trig to calculus.

7/8/2005 1:49:00 AM

1337,

You're apparently not understanding what i'm saying. I'm not talking about knowledge of market conditions or whatever you're thinking of, I talking about "self knowledge"--not being able to articulate what will maximize your utility. You're assuming people will be able to articulate their mental processes to you in a survey, and I'm saying they might not.

You're also confused about what part of consumer choice theory i'm attacking, i'm attacking the assumption that people will maximize their utility, which isn't conditional on perfect information or imperfect information or any information at all. It's a behavioral assumption.

Clear5

1) Are you saying that economics, a science seeking to explain observed human behavior, is comparable to mathematics, a science seeking to deduce abstract truths about numbers?

2) What do you mean mathematics isn't falsifiable?

7/8/2005 1:49:40 AM

When talking about falsifiability in science we are talking about Popper and philosophy of science and not the dictionary definition.

Mathematics isnt falsifiable because you can only prove mathematical theorems wrong with math's own terms, you cant observe it to be wrong.

And I only make the comparison because you make it sound like it should really matter to the people studying the subject whether or not it is scientific.

7/8/2005 2:13:25 AM