One Last Volley on the Keynesian Cross
Somebody emailed me the relevant section from Krugman/Wells on this stuff. Unfortunately it’s not a smoking gun; both sides will see what they want to see. Krugman walks through the original Keynesian logic of using the MPC to derive “the” multiplier (with math, not geometry). So someone who wants to bust him, would just look at that section.
However, Krugman does work in various points about crowding out, supply constraints, etc. in his broader framework of Aggregate Supply / Aggregate Demand, so someone who wants to defend him could point to those things.
I only taught intro to macro very rarely, and we were using a “free market” textbook by Gwartney et al. So in my teaching of this stuff, obviously the fallacy that Hazlitt/Rothbard/Landsburg identified, wasn’t there. But I can’t say whether modern students still get it, in the typical classroom. Nick Rowe kinda sorta thinks so.
Anyway, here’s my parting contribution to this discussion. Here’s Daniel responding to Nick Rowe:
All of [Nick Rowe’s] points are actually well taken, the problem is he makes those points as if it’s an either/or choice between AD/AS and the Keynesian cross. Of course it’s not. He also makes those points as if you don’t have a semester to clean up the Keynesian cross with other intuitions.
…[T]he Keynesian cross was nice to teach too because:
1. They could actually solve it – the math is trivial so you can take non-econ majors that are required to take it and they can work through it.
2. It reinforced the income-equals-expenditure point which is one of the critical differences between macro and micro and for that matter between macro and how people think generally, and
…
4. Contrary to what some people have been saying in this discussion, the Keynesian cross is the perfect time to teach about crowding out. Why? Because when you give them plausible numbers to plug in and have them solve you get stupidly high multipliers (and I’m talking about multipliers of four or five, not Landsburg’s multipliers). Then you show them a chart that Mark Zandi put together and ask “so why don’t we actually get multipliers that are that high? where can we see daylight between what the model says and what happens in the real world?”. My (limited) experience is that that tends to be a very fruitful discussion. [Bold added.]
Does everyone see what just happened here?
==> Landsburg said that the Keynesian Cross, as typically taught in Econ 101, rests on a silly fallacy. It uses simple math to give students the impression that there is a causal relationship between more government spending and more national income, when that doesn’t at all follow from the accounting identity.
==> Daniel comes back and says he thinks the Keynesian Cross is still good to use in class. He admits that it’s based on simple math, it reinforces the idea that spending equals income, and because using it literally as a model gives you totally wrong outcomes, it’s an excellent pedagogical device to show why the Keynesian Cross by itself must be supplemented with other things (like crowding out). Indeed, you need to spend future lectures during the semester to “clean up” (Daniel’s words) the false intuitions one might have gotten just from studying the Keynesian Cross.
In light of the above, one might have expected Daniel to originally write a post, “Landsburg makes some good points on the Keynesian Cross, but I regret nothing!” Instead he titled his original response, “As misleading as comparing penetration by a rapist to penetration by photons but (mercifully) less offensive.”
I know I do it too, but it’s really funny when you step back and look at how much the economics bloggers actually agree with each other, but since we’re on different teams, we flip out and accuse the other side of perfidy and stupidity. (And yes, as someone in the comments mentioned, to see Krugman complaining that people are saying he’s stupid, and not giving him the respect he deserves, is breathtaking, given his routine treatment of even other Nobel laureates who think the government deficit is a problem.)
I would say the misuse of the Keynesian model is a fallacy. The model itself is not.
And I don’t think he made good points about the Keynesian cross.
If Landsburg had said “a plausible multiplier from a raw Keynesian cross is bigger than we get and you can’t get everything from those couple of equations” there would be nothing to disagree with.
As we all know, he wrote nothing of the sort.
He dropped MPC from the model and used income share instead. That’s wrong and it has implications. He dropped expenditure from the model. That’s wrong and it has implications. Those two things were mostly what my post was about – all the crowding out/supply stuff really came in a substantial way after I answered your first comment.
So I repeat, if the only disagreement was over how to deal with (theoretically) plausible multipliers that are too high and whether we should judge the model on the basis of two equations or on the basis of the actual theory behind it, I would have written a very different post.
But that was not the only disagreement.
“The model itself is not.”
Yeah, it is. Modelling human action is a fallacy in and of itself, much less the ridiculous parameters of the Kenyesian models…
So modelling human action during an Austrian business cycle theory (e.g., human agents moving prices and making alleged capital malinvestments) is “a fallacy in and of itself”!!
Way to disprove Austrian economics.
You build beautiful strawmen.
You’re familiar with the terms praxeology and catallactics, right?
Of course he isn’t.
That isn’t modelling human action.
Daniel,
You are right that Landsburg dropped expenditure and MPC but why does this matter? He can get the same result without those omissions. Instead of L = Landsburg’s income and E = everyone else’s income, he could have let L = Landsburg’s expenditures and E = all expenditures other than his own. So your objection about leaving out expenditures goes away. Take derivatives on Y = L*10^8 and we get his marginal propensity to consume so your second objection goes away. So now I’ve patched up the two things you have identified as problems and just like before we can get to the same result by the very same algebra employed by Landsburg.
Now I’ve shown that a reductio can be constructed satisfying your added requirements. Seeing that your first two objections have exactly zero implications for Landsburg’s argument, do you now grant the validity of the argument? Or do you want to assert still more requirements for a valid reductio against the Keynesian cross?
I did pretty much the same thing on my post (patching up my concerns about Landsburg’s post for him).
If you’re curious about my thoughts after already doing that, see the post.
You’ve had a number of posts on the topic but I can’t figure out which post you are referring to here. In any case, since I’ve shown that Landsburg’s argument can be modified in the most trivial way to address your income/expenditure objections, I guess now you would agree that Landsburg’s argument is a valid one, right?
No. See the comment above and I’m referring to the first post about Landsburg.
Do you have any thoughts at all on what I just wrote? What is unclear?
I can’t find any post where you patch up Landsburg’s argument so that L and E are expenditures so that the relationship of L to Y is a MPC. In any case, if those are your objections to Landsburg’s argument and his argument still works when they are both remedied, why don’t you accept his argument? If you want a model that includes investment, that’s easy too. Y = L + E + I…
“What is unclear?”
SInce Landsburg posted first, right now you are operating with the luxury of hearing Landsburg’s argument and then asserting requirements for a valid reductio.
I claim that if you list out all of the characteristics that you’d require in a valid reductio, either your set of requirements would be so severe that you’d have to reject even the original Keynesian cross or it would still be possible to rebuild Landsburg’s argument to meet those requirements.
Specifically what would it take for you to see an argument as a valid reductio against the Keynesian cross?
Hi Daniel. The only problem I find here is that we can play the same game:
Y = C + I if C = 0.8Y then I = 0.2Y
MPC + MPS = 1
Y – C = I
Y(1-0.8) = I
Y = I/0.2
as I = 0.2Y then
Y = Y
If Landsburg/Rothbard/Hazlitt is a tautology, is not this one too?
Been over this one in the comment section of my post and in a brand new post referencing Malcolm’s comment.
Think about what you are doing by including your fourth equation. Justify why that is there.
Yes, you can always pull an identity back out of a system of equations. So? Who cares? This is a tautology. The question isn’t “can you pull an identity back out”. Of course you can. You put the damned thing in there why wouldn’t you be able to pull it back out the other end? The question is how you interpret the set of equations and whether your ability to pull the identity back out negates anything.
Read those posts and let me know if you still don’t get my reaction there.
Daniel, for the very first time on this blog I am going to write an entire paragraph in caps.
LANDSBURG DID NOT ARGUE THERE WAS NO KEYNESIAN MULTIPLIER. HE WROTE A POST ATTACKING A PARTICULAR ARGUMENT. THE EXISTENCE OF OTHER BETTER ARGUMENTS FOR THE MULTIPLIER DO NOT MATTER TO THE ISSUE AT HAND, AND DO NOT DISPROVE WHAT HE WROTE.
I’m not sure you’re understanding what my problem with the Landsburg post is.
Also – quick clarification because it gets tossed around here a lot – I don’t think Landsburg is stupid.
A good rule of thumb is that the only time you will ever see me waste time blogging about something that I think is stupid is when I blog about my cats.
They are as dumb as a bag of hammers.
But at least your cats don’t subscribe to the Treasury View.
It’s true. They are quite convinced that there is in fact a free lunch.
Proud of the role I played.
You shall not crucify mankind upon a cross of Keynes!
haha
Silas_Barta likes this
^^^ This!
Any thoughts on this? I think it’s a good summary and sides with Landsburg with good reasons http://puntodevistaeconomico.wordpress.com/2013/06/28/the-landsburg-multiplier-and-all-that/
“To look at income or expenditure is to look at two sides of the same coin.”
That’s my main point against DK’s rather poorly written post.
The Keynesian Cross is really one more failing of the empirical economic methods. The other failure is utter hypocrisy on political issues. I just did a post regarding that if anyone is interested.
Do you know why the aphorism “Economics is the dismal science” exists?
The answer, given how often the expression is used, is very discomforting and awkward.
What is the answer, exactly?
The phrase “the dismal science” first occurred in Thomas Carlyle’s “Occasional Discourse on the Negro Question”, in 1849. In it, he argued for the reintroduction of slavery as a means to regulate the labor market in the West Indies. He wrote:
“Not a “gay science,” I should say, like some we have heard of; no, a dreary, desolate and, indeed, quite abject and distressing one; what we might call, by way of eminence, the dismal science.”
Carlye thought economic science was “dismal” because his classical liberal contemporaries at the time spoke of “supply and demand”, and calling for the government to leave men alone to their affairs. Carlye was disgusted by this, because he wanted the “idle Black man in the West Indies” to be “compelled to work as he was fit, and to do the Maker’s will who had constructed him.”
In short, the study of economics to Carlye was “dismal” because it helps lead to the emancipation of black slaves.
Yikes.
Next time you hear some speaker mention it, offhand or whatever, clearly not understanding it’s origination, keep in mind the reason why the phrase exists. I always feel awkward when otherwise oblivious people make a joke with it. I think, oh, if you only knew where THAT came from.
I shudder to think what kinds of aphorisms I use, the origination of which I am oblivious.
Less forcefully, think of every model as being useful in some circumstances. Then the set of circumstances in which any kind of GEQ model is useful is of measure zero, whereas the set of circumstances in which the keynesian cross is useful is of positive measure.
You may have seen the Keynesian cross at A-level. This can be a fiddly one to understand at first. The idea behind this is there will be a level of planned expenditure in the economy which is the amount consumers (households), firms, and government plan to spend on goods and services. This is basically the demand for goods and services. When this planned expenditure is higher than firms producing goods expected, then will start depleting their inventory stocks of goods, and this will stimulate more production, they will start hiring more workers and produce more, so income will rise. When the planned expenditure is lower than the firms producing goods expected, then they start building up their inventory stocks with unsold goods, which will cause them to step back on production, lay workers off, and income will fall.
To set the stage, consider equilibrium in the basic Keynesian cross model, such as that presented in the exhibit to the right. Aggregate expenditures are measured on the vertical axis and aggregate production is measured on the horizontal axis. The aggregate expenditures line is labeled AE and the 45-degree equilibrium guide line is labeled Y = AE. Equilibrium is determined by the intersection of the aggregate expenditures line and the 45-degree line , which in this case is $12 trillion.