An up and coming forum, libertyhq, seems to be taking over for the, sadly, soon to be closed mises.org forum. One of the members quoted a few lines from the mainstream economics textbook his class is using, and asked some questions about it. Naturally Smiling Dave was right there, with his insightful insights.

Here’s the Original Post, from Gurimbom:

I’ve been recently learning the Keynesian theory at school. With models being represented like: ‘Y = C + I + G – (X – M)’
I know it’s just a representative model of the economy, but I’m not sure if it can be successfully used to explain economic effects and consequences of certain actions.

For example, a sentence reads: “Due to impressive work of country X’s best econometricians, it’s proven that ‘C = 3/4Y’. Meaning that 75% of the national income is consumed. And due the consumption multiplier being 4, a continuous increase in consumption of 1 billion dollars would increase GDP with 4 billion dollars.

I don’t really understand how econometricians could establish the ‘fact’ that people consume 3/4th of their income. Isn’t it fallacious to try to explain human behaviour and action in mathematical models? How can you even be certain these models are even remotely close to the truth?

A poster called Student, a student of the mainstream, gave the best possible defense of the textbook and wrote:

Actually, Y = C + I + G – (X-M) is an accounting identity and not a model. It is how we define GDP. GDP will always be equal to the sum of consumption, investment, etc **by definition**. As a result, if you had a model that described how each of the variables change after a certain action takes place, this equation would give you a very good idea of how GDP would respond. Of course, that hard part is finding that model. :P
http://en.wikipedia.org/wiki/Gross_dome … e_approach
http://en.wikipedia.org/wiki/Accounting … ic_product

“I don’t really understand how econometricians could establish the ‘fact’ that people consume 3/4th of their income.”

This sounds like a homework question or an example from your book. It is just a simplifying assumption, so don’t read too much into it. Depending on the time period, saying that fraction of consumption was constant may not be a horrible assumption (depending on what you’re trying to do). If you look back at the data you will see that personal consumption was indeed a relatively stable portion of GDP from 1950 to the early 1980s (it looks like an average of 62%). However, no one would actually argue that it is an iron clad law that consumption must always stay at 62%.
http://research.stlouisfed.org/fred2/graph/?g=ezm

To make a long story short, my guess is that they are telling you to assume consumption always equals 75% of Y, not because it is true (it isn’t), but so you can more easily deal with multipliers. IOW: The point of this thought experiment isn’t to accurately describe how an economy works, but to get you start thinking in terms of models (i.e. thinking like an economist).

Keep in mind that the goal of these exercises is (or should be) to get you to start asking questions, not telling you answers. For example, as you read about these models, ask yourself what assumptions have to hold for the simple one sentence model to make sense (what has to happen for the multiplier to stay constant). Once you identify what the underlying assumptions are, ask yourself if they make sense. If you decide that they don’t, ask yourself what you would do differently.

“Isn’t it fallacious to try to explain human behaviour and action in mathematical models?”

I guess it depends on who you ask. Hayek had no problem using mathematical models. Nor do more modern Austrians like Peter Leeson, who routinely uses game theory in his academic work.

I think it doesn’t depend who you ask at all. It’s a big mistake. Here’s Smiling Dave’s reply in full:

1.

Student wrote:
Gurm,
Hayek had no problem using mathematical models. Nor do more modern Austrians like Peter Leeson, who routinely uses game theory in his academic work.

Could you give us a link to Hayek using mathematical models? A quick search turned up this quote from Rothbard:
F.A. Hayek is not only the leading free-market economist; he has also led the way in attacking the mathematical models and the planning pretensions of the would-be “scientists,”

Hayek’s Nobel Lecture was all about the uselessness of exactly such models as that in the texbook. I particularly enjoyed his analogy of the topics economics studies being like a baseball game. You can have all the advanced statistics of every single player and every single team all the way to the beginning of time, and you still don’t know who will win the game. Any wonder he won a Nobel prize?

2. Student is your man when it comes to explaining the mainstream position. He got it all spot on.

3. Note that the textbook you are using is actually a subtle form of brainwashing. Imagine if your homework assignments had questions like:
There is a famine in the land. 3 bullocks are offered in sacrifice to Zeus. If the sacrifice multiplier is 4, by how much will the crops increase? Answer: Twelve fold.
You see where I’m going with this. The book will assume, over and over, that there is a sacrifice multiplier, an assumption that is, to put it mildly, questionable.
As time goes by, the innocent student, from constant repetition, will come to believe in the sacrifice multiple.

Equally questionable is the assumption of a consumption multiplier. Future generations will mock ours for thinking consumption, meaning consuming, meaning eating up and destroying wealth, makes for wealth. They will be even more mocking when they read the classical economists who were so patient explaining the folly of such an idea. “Guys,” they will tell us. “James Mill in Commerce Defended spelled it out so clearly. How did you ever listen to Keynes?”

Keynes wrote about the “fallacy that demand is created by supply”. The mind boggles. Such an obviously true statement he calls a fallacy. Has anybody here ever gotten money [= ability to demand] any other way but by working [= creating supply]?

4. The OP is certainly showing healthy skepticism about using mathematical models to predict human behavior. I’d like to elaborate a bit.

The wildest use of math in economics is confusing correlation with causation. As an example, govt statisticians found that homeowners are superior citizens by many metrics. They are more responsible, their families are more cohesive, they hold down jobs, etc. They decided, based on the incredible correlation the math produced, that it’s a great idea to get as many people as possible into houses. Such a move would turn all Americans into model citizens, they were certain. They encouraged banks to provide cheap loans to everyone, to make sure they got into houses. To their chagrin, they found out that a deadbeat with a house is still a deadbeat. Being responsible was what made one able to have a house; having a house did not make one responsible.

There are a very few mainstream economists whose eyes have opened to this. They have pointed out that no respectable science, such as physics or chemistry, ever starts out with an empty correlation and runs with it. Instead, they carefully develop an hypothesis based on logical reasoning, and then go looking for correlations. Only economists commit this egregious error. Sadly, the ones who recognize the error are in the very small minority.

That’s the wildest error. Slightly less wild, but equally wrong, is the use of the fallacy that past patterns predict the future. An economist will draw up a graph, made up of carefully researched data points, of, say, stock market prices. He will find that some math formula will fit the data perfectly. After all, you can always find a polynomial that fits perfectly through any finite set of points. And the formula works, until it doesn’t. Austrians explain that the flaw was assuming that there is some underlying law of nature that forced the data collected to fit the curve. After all, in physics that’s how it works. The laws of gravity, of conservation of energy, and so on, force the projectile thrown at angle X with velocity Y to travel a distance Z. But it doesn’t work that way with people. Hayek’s baseball analogy is pertinent here.

Finally, there is the lesser mistake of correctly locating a force in economics, but incorrectly assuming the force can be given a magic number for all time. For example, certain economists have correctly deduced that an increase in the money supply will cause prices to rise. There is indeed a force at work, as certain as gravity, that forces prices to rise when the money supply is increased, all other things being equal. But by how much? If we write down a formula, say Q = [1+m]P, where Q is the new price, P the old price, and m the increase in the money supply, can we know what the value of m is? Sure we can, after the fact. But the flaw is in thinking that m is a constant, like Planck’s constant in physics. Austrians point out that m can be .3 today, and .2 tomorrow, and something else the day after.

Note that when I talk about these common mistakes being greater or lesser, I am talking about the theoretical madness that led to making the mistake. You have to be really nuts to confuse correlation with causation, and less so to think the future will always be like the past. But all of these mistakes can and have wreaked catastrophe on the world.

About these ads