Arnold Kling

The Case for Killing Textbook Macro

PRINT
From the Preface of My Next Bo... Ben Bernanke Explains Systemic...

Seeing two posts today on macro, I am ready to share Nassim Taleb's despair.

Tyler Cowen writes,


I agree with your point that fiscal policy can work through V and that is the correct way to think about it. In fact Alex and I present this in our forthcoming Principles text.

I'm sorry to hear that. I think that the main lesson of the past few years is that monetarism is not even close to right when the inflation rate is reasonably low (say, less than 5 percent). The notion that "fiscal policy can work through V" (that is, it can increase the velocity of money) is embedded within a faux-physics paradigm that has no value, in my view. It s my view that financial institutions, bubbles, and crashes matter. Monetary policy does not.

Some Austrians will insist that monetary policy is responsible for bubbles and crashes. I will suggest instead that institutional factors, including perverse regulations, are far more important.

Next, we have Greg Mankiw.


What is the distinction between a shock to potential output (aka the natural level of output) and a shock to the Phillips curve?

Back when I still followed textbook macro, I thought that potential output was the long-run aggregate supply curve and the Phillips Curve was the short-run aggregate supply curve. I am not sure why Greg did not answer the question along these lines.

But my main point is that I think this whole framework is not helpful. It is important to remember that for the economy as a whole, "supply and demand" is only a metaphor.

Instead, I prefer to describe the economy as having labor markets that are close to equilibrium or far away from equilibrium. When labor markets are close to equilibrium, there are lots of people taking new jobs and lots of people leaving old jobs (not necessarily voluntarily), but there is an overall balance. When markets are far away from equilibrium, there is an excess of people leaving jobs relative to people finding jobs, and this excess persists.

What should we call this labor market imbalance? In the textbooks, we might call it a demand shock if we think that most of the unemployed could go back to working at their old jobs. Otherwise, we might call it a supply shock.

The working assumption in textbooks is that there is something that policy can do to solve the labor market imbalance, particularly if it is a demand shock. I think that we rarely, if ever, observe a pure demand shock. I think that monetary policy is ineffective, for reasons stated above. I think that fiscal policy is likely to be ineffective, unless policymakers can outguess the market about the sorts of jobs that need to be created in the long term.

In my view, the policymaker who thinks like a textbook macroeconomist, and therefore believes that he sits at a control center with buttons to press and levers to push that will create jobs, is deceived. The buttons and levers are there all right, but they do not work as the textbooks propose.

If I were a policymaker in the current environment, I would try Bryan Caplan's idea of cutting the employer portion of the payroll tax. My thinking is that with more profits and a lower marginal cost of labor, we might get back to labor market balance faster. I would not be confident that this policy would produce great results. But I am persuaded that it stands a better chance than other stimulus policies, both enacted and proposed.

The key point is that the problem is not to create jobs. The problem is to hasten the process of restoring labor market balance.


Comments and Sharing


CATEGORIES: Macroeconomics



COMMENTS (9 to date)
Danny writes:

I consider myself an Austrian but agree with you on monetary policy not being the prime culprit. It was a factor, but not the major one.

I think that institutional and regulatory factors were also important, but I attribute the lion's share of blame on the so-called 'Greenspan Put'. By that I mean the underpricing of counterparty risk, and therefore risk in general, which caused the derivative markets to grow significantly faster and larger than they otherwise would have. There was no discipline.

That is one of the main reasons why I disagree with you on Credit Default Swaps. I don't see them as bad, nor do I see a lack of natural sellers in that market. What there was was a general underpricing of counterparty risk. I do appreciate that you don't call for an outright ban, and I think, in the end, you and I hold similar views on what the ultimate size of that market would be, ie, significantly smaller than it is today.

Don Lloyd writes:

Arnold,

In my view, the policymaker who thinks like a textbook macroeconomist, and therefore believes that he sits at a control center with buttons to press and levers to push that will create jobs, is deceived. The buttons and levers are there all right, but they do not work as the textbooks propose.

The policymaker doesn't care whether or not what the textbooks say is correct. Nor is the goal of the textbooks to be accurate. Policymaking is a means to political ends, with any economic effects only significant to the extent that they advance or retard the achievement of those ends. The textbooks create the virtual buttons and levers that justify the existence of policymakers to press and push them, but changing the labels on them at random isn't likely to produce significantly worse results.

Regards, Don


Gary writes:
...Bryan Caplan's idea of cutting the employer portion of the payroll tax. My thinking is that with more profits and a lower marginal cost of labor...

Stop right there. I thought economists agreed a long time ago that it doesn't matter to employers what portion of employee cost is in the form of tax and what portion is wage. At least that's the argument we make when someone proposes raising the employer portion of the payroll tax. Let's not make ourselves look bad by trying to have it both ways!

(Or perhaps you're making an argument about wage rigidity and the benefits of temporarily tricking employees into taking lower salaries than they would otherwise demand. But I didn't get that impression from what you said.)

Snorri Godhi writes:

When labor markets are close to equilibrium, there are lots of people taking new jobs and lots of people leaving old jobs (not necessarily voluntarily), but there is an overall balance. When markets are far away from equilibrium, there is an excess of people leaving jobs relative to people finding jobs, and this excess persists.

The problem with this way of putting it is that there is never a time when unemployment decreases, because there is never a state in which there are more people taking new jobs than people leaving jobs.

It seems to me that it would be more accurate to say:

At equilibrium, employment is at a maximum. When markets move away from equilibrium, there is an excess of people leaving jobs relative to people finding jobs. While markets are far from equilibrium, high unemployment persists. When markets move back towards equilibrium, there are more people finding jobs than people leaving jobs.

Please correct me if I am wrong.

Mike Rulle writes:

Even though I occupied senior positions at IBs and hedge funds (From CEO to President to division head) I can frankly say I did not know anything about economics and its implications for individual businesses. I certainly think that was bad thing--for me at least---even as I was a reasonably successful grad student in political philosophy and finance. One can be successful, of course, in business without knowing much formal economics at all---but for me it would have been better to be more literate.

I have become "dilettante" literate in the difference, say, between Hayek and Keynes over the last 2 years or so. When I first came across Hayek I said "of course", as it was a frame of reference I generally had but it was not explicitly formulated in my mind. The Keynes Macroeconomic argument, or what is called that, reduces down I think to an abstraction that is difficult to connect to reality. The bottom line is his counter cyclical theories result in Government making spending decisions without regard to "return on investment". He either just assumes it happens or in certain cases it is not even necessary. In fact, I believe his digging holes and refilling them analogy is evidence. But why not just by pass the digging holes stage and just print spending vouchers if that is such a great idea.

Stimulus theory strikes me at bottom as designed to "surprise" or "trick" the recalcitrant public into thinking things are getting better when "work" is happening and dollars spent. The investment is not a direct product of the work itself, but an indirect product of creating the belief that it is safe to consume and spend again.

While I do agree with Arnold's suggestion, I am confused as to why we propose policies (which by the way will not get passed in the next 4-8 years under any circumstance), so limited in scope. Since we are "wishing", are not there more things to suggest?

What does government actually do? The largest economic thing it does is "redistribute". Transfer payments to individuals directly and through states are about 30-50% of the budget. I wonder what the|"transfer tax" is? Social Security and Medicare of course dominate. This is also largely similar to a Ponzi scheme, not insurance or investment activities because the money has already been spent. One would have to assume the previously spent money created greater growth than would have occurred had these funds truly been invested. While Samuelson thought that governments can succeed in Ponzi's, I do not. So let's rid ourselves of Ponzis---yes it is not easy--but we will be breaking promises soon enough. It all should be shifted to private saving's accounts.

While we are at it, lets undo the EPA's unlimited authority to act first and be challenged later.Lets permit drilling (although we may find that our comparative advantage--Ricardo style--may end up making this less important--but why outlaw it?)Let's permit our school tax dollars to be used anyway we wish via vouchers as we transition to a competitive private education system---at least this should put some breaks on the unchecked growth of the teachers unions.

Lets introduce a health care system which mimics other parts of the economy. That is, lets make marginal decisions cost money. Right now, we are insured and the incentive is to get as much for our money as possible. There are only upfront costs, no or little marginal cost constraints at the individual level. This, of course, is the opposite direction we are going unless you count the "marginal" decisions that will be made by the government. Also, why do we permit the AMA to have a de facto supply monopoly?

Why not require the Congress to publish explicit cost benefit analysis is economic terms when any spending bill is proposed? At least can all get some entertainment.Money comes from the states, gets reshuffled, and goes back to the states. How much "slippage" is lost in the shuffle? Is this necessary? Maybe states who think they need funds can make their case to the public in bond offerings like companies do. Maybe other state pension funds would even invest.Outlaw all Government Employee Defined benefit programs. Defined contribution is okay.

It is relatively clear---just look at the stimulus bill---that government spending is an adverse selection process designed to take money from the broad public and give it back to the portions of the public with the highest chance of wasting it. Oh--yes---lower marginal income tax rates for all individuals--including Arnold's suggestion.Government has similar economic features as a mob activity--except sometimes the mob engages not merely in redistribution but offers products and services the public wants. Why are we being so timid, as long as we are wishing for things that will never happen?

I am sure I left out much.

Greg Ransom writes:

I love the way that you can pretend that (heterogeneous) productive goods don't exist or don't matter -- this is the real crackpot legacy of John Maynard Keynes. All this stuff you criticize are just particular manifestations of this central pathology (see Roger Garrison on this topic.)

Note here how ALL of your focus is on the labor market -- the central Keynesian pathology (among a whole host of others.)

"I prefer to describe the economy as having labor markets that are close to equilibrium or far away from equilibrium. When labor markets are close to equilibrium, there are lots of people taking new jobs and lots of people leaving old jobs (not necessarily voluntarily), but there is an overall balance. When markets are far away from equilibrium, there is an excess of people leaving jobs relative to people finding jobs, and this excess persists."

fundamentalist writes:

"I will suggest instead that institutional factors, including perverse regulations, are far more important."

Hayek has a good discussion of non-monetary theories of cycles in "Monetary Theory and Trade Cycles." He wrote is over 70 years ago but it seems current. Not much has changed in economic theory since then. He makes an important point that even if proponents of non-monetary theories don't consider money to be the prime driver of cycles, they all admit that it is a co-conspirator. In other words, institutional and regulation changes or defects could not have the effects they have without complementary changes in money policy.

Does anyone really think that the current crisis would have taken place had the Fed kept its rate at 6% or higher after 2001? When the Feds expand credit in order to boost the money supply, the money that people don't want to hold as cash has to go somewhere. If it didn't go into housing, it would have gone into another asset, such as commodities, stocks, bonds, etc. If it didn't go into assets it would go into consumption and we would have enjoyed high inflation. Tighter regulations in the housing market would have done nothing but divert the excess money to another sector.

Robert writes:

Just so you are aware of the probable knee-jerk liberal reaction to your payroll tax cut idea -- wouldn't it be far more fruitful to focus on lowering the wages of the top earners, since they account for most of employee compensation? A payroll tax cut would not appreciably change executive pay. But lowering the compensation of top earners would greatly improve profitability.

I realize there are a lot of problems with that view, but it is admittedly knee-jerk and FYI.

Ed Lopez writes:

Is your nexus of labor markets approach a version of the sectoral shifts model of business cycles? Admittedly the texts don't stroke that model like ISLM, K Cross, or certainly AD/AS. Would it be of any greater use to policymakers, in your view?

Comments for this entry have been closed
Return to top