Arnold Kling  

A Pat on the Back for Macro

PRINT
Work-Safe Readings for Macro... The Call for a Commission...

It comes from Olivier Blanchard.


a largely shared vision both of fluctuations and of methodology has emerged. Not everything is fine. Like all revolutions, this one has come with the destruction of some knowledge, and suffers from extremism, herding, and fashion. But none of this is deadly. The state of macro is good.

Thanks to Paul Walker for the pointer. I'll excerpt more below, and try not to get personal about an economist who for thirty years has embodied everything I despise about macro. Back then, he was insufferably smug and I was childishly rebellious. Not much has changed.

Blanchard depicts mainstream macro thusly:


a specfic model, the so-called new-Keynesian (or NK) model, has emerged and become a workhorse for policy and welfare analysis ...The model starts from the RBC (real business cycle) model without capital, and, in its basic incarnation, adds only two imperfections. It introduces monopolistic competition in the goods market. ..It then introduces discrete nominal price setting, using a formulation introduced by Calvo, and which turns out to be the most analytically convenient.

Blanchard lavishes praise on this model. Of course, without capital or investment, it has no financial sector to speak of.

The current financial crisis makes it clear that the arbitrage approach to the determination of the term structure of interest rates and asset prices implicit in the basic NK model falls short of the mark: Financial institutions matter, and shocks to their capital or liquidity position appear to have potentially large macroeconomic effects.

The main imperfection around which thinking about credit markets is built is asymmetric information.

This has two direct implications for macro fluctuations. First, these constraints are likely to amplify the effects of other shocks on activity. To the extent that adverse shocks decrease profits, and thus reduce the funds available to the entrepreneurs as well as the value of the collateral they can put up, they are likely to lead to a sharper drop in investment than would happen under competitive markets. Second, shifts in the constraints can themselves be sources of shocks. For example, changes in perceived uncertainty which lead outside investors to ask for more guarantees may lead entrepreneurs to reduce their investment plans, leading to lower demand in the short run, and lower supply in the medium run

to the extent that investment projects have horizons longer than those of the ultimate investors, financial intermediaries may hold assets of a longer maturity than their liabilities. Because financial intermediaries are likely to have specific expertise about the loans they have made and the assets they hold, they may find it difficult or even impossible to sell these assets to third parties. This in turn opens the scope for liquidity problems: A desire by the ultimate investors to receive funds before the assets mature may force the intermediaries to sell assets at depressed prices, to cut lending, or even to go bankrupt|all possibilities the current financial crisis has made vivid. (The standard non-macro reference here is Diamond and Dybvig (1983), which has triggered a large literature.)

Discussing empirical factor analysis undertaken by Stock and Watson, Blanchard writes,

There is a tempting but slightly worrisome interpretation for these results: That shocks to aggregate demand, which indeed move most quantities in the same direction, have little effect on prices, and thus on inflation. That shocks to prices or wages, and thus to inflation, explain most of the movements in inflation, with little relation to or effect on output. And that asset prices largely have a life of their own, with limited effects on activity.

Blanchard proceeds to discuss methodology.

A macroeconomic article today often follows strict, haiku-like, rules: It starts from a general equilibrium structure, in which individuals maximize the expected present value of utility, ¯rms maximize their value, and markets clear. Then, it introduces a twist, be it an imperfection or the closing of a particular set of markets, and works out the general equilibrium implications. It then performs a numerical simulation, based on calibration, showing that the model performs well. It ends with a welfare assessment.

So, the state of macro is this:

1. We have a workhorse model, with no capital or financial markets.
2. We have some work on asymmetric information and financial markets that is not really integrated into the workhorse model, but which suggests that when "shocks" occur, their effects may be amplified relative to the workhorse model.
3. Real-world data have interesting patterns that either are unexplained by or contradict the most widely-used models.
4. Papers follow a "haiku-like" ritual in order to be published.

And this is "good." I agree with all four propositions, but I disagree with the conclusion.

For example, Washington is currently discussing an economic stimulus package. This sounds like an issue on which macroeconomists ought to have an informed opinion. How large should it be? Should it be enacted at all? In all of the haikus that have been published over the past thirty years, is there one that offers a clue to the answer?

With regard to the bailout, I don't see macro doing much better. We have theoretical models that tell us that financial sector losses can "amplify shocks." On the other hand, we have an empirical study that says that asset price movements don't have big effects on employment and output. If we believe the empirical work, then the likelihood of another Great Depression resulting from recent financial turmoil is actually pretty low.

Even in the theoretical models, we have no idea what sort of financial sector distress could lead to a quantitatively large effect on the real economy. We have no idea whether, taking opportunity cost into account, injecting capital into banks is a net plus or a net minus for the overall economy. Remember that Ken Rogoff, correctly in my opinion, describes the financial sector as bloated. That means that the financial sector might be the last place where we should be injecting capital.

There may very well be adverse macro effects to follow from the loss of confidence in the way that financial institutions handle risk. But isn't that loss of confidence justified? Didn't financial institutions in fact do a lousy job of managing risk?

Common sense would suggest that injecting capital into banks will not restore confidence if nobody believes that the banks know what they are doing. In that case, the capital injection could turn out to be a waste of resources, keeping financial institutions going when the market is screaming at them to shut down.

Neither I nor anyone else knows what the macroeconomic consequences will be of the various rescue plans and stimulus proposals. At best, these moves can be justified as sheer panic. At worst, they reflect a long-term agenda of the Left, which is using the financial crisis as cover for looting corporate America, just as thieves use an urban riot as a cover for looting televisions.

Macroeconomics can tell us nothing useful about the current policy environment. All we know for sure about what is taking place is that there has been a massive shift of power to Washington, with much more likely ahead.


Comments and Sharing





COMMENTS (17 to date)
Frejus writes:

"try not to get personal about an economist who for thirty years has embodied everything I despise about macro. Back then, he was insufferably smug and I was childishly rebellious. Not much has changed."

Very odd way to not get personal. By getting personal. Particularly the way you describe him as "smug"--all negative--and then try to use self-deprecating humor by describing yourself as a childish rebel--which can be viewed positively.

The way to not get personal is to focus on the facts. And not get personal at all.

John Jenkins writes:

@ Frejus: But AK does not get personal in his critique. The statement you focus on is simply a warning about AK's biases that helps you in evaluating the balance of the post.

To challenge your last claim: where in the actual discussion did AK get personal?

fundamentalist writes:

Kling: “Neither I nor anyone else knows what the macroeconomic consequences will be of the various rescue plans and stimulus proposals.”

At the risk of irritating non-Austrians, I think the Austrian perspective might shed some light on the matter. And those undecided about whether to learn Austrian econ might want to watch for empirical evidence of the Austrian thesis. From the Austrian perspective, the problem with the economy (not the financial sector necessarily, but it is related) is a shortage of both capital goods (material and equipment to make consumer goods) and consumer goods. It was fairly clear that all the last stimulus package did was to raise prices because more money was chasing the same shortage of goods. The same thing should happen with any further stimulus.

As for the rescue plans, the whole purpose was to keep the money supply at the high level of growth that it had experienced during the period of low interest rates. Rescuing banks will allow them to loan more and thereby pump more money into the economy. So the result of the bank rescue plan will be the same as the stimulus, but with a lag. With more money chasing a limited supply of consumer goods, prices of consumer goods will rise and profits for the sellers will rise.

Rising profits also means lower relative wages, that is, wages relative to the prices of output. Such lower wages causes the Ricardo Effect to kick in. Ricardo said that when wages fall relative to the price of output (consumer goods), businesses will shift to more labor intensive methods. In other words, they will hire more labor and buy less equipment. In standard micro econ textbooks the Ricardo Effect is described in the chapter on optimizing the combinations of labor and capital. The combination of higher profits in consumer goods industries and lower demand for equipment means that investors will shift investment from capital goods producing industries to consumer goods producing industries. Sales and profits will fall in the capital goods industries. The net result will be more business failures in the capital goods industries and more lay-offs.

Mencius writes:

Note that Blanchard makes the same mistake I described in the last thread - he assumes that "liquidity" in the sense of a maturity crisis must have something to do with asymmetrical information. Ie, he mixes up liquidity(a) and liquidity(b).

Also, it's pretty hilarious to start from the "real business cycle" model, which as Blanchard says (I'll take his word for it) is the heart of "New Keynesianism," when - as he in fact explains quite clearly - you have this enormous shock amplifier, systematic maturity transformation, in your financial system.

Which is easier: maintaining a vast staff of diligent central planners whose task it is to smooth out every last ripple in prices, wages and interest rates? Or removing the gigantic Marshall amp that turns the ripples into tsunamis?

Ask the diligent central planners, of course, and you know what they'll say. The basic problem is that the entire discipline of macroeconomics emerged in order to explain these tsunamis. If you identify the actual cause of the "business cycle" (a better name would be "banking cycle") and fix it, the customer need not keep coming back to the shop.

If the answer is merely what Keynes ridiculed as "orthodox economics" and his discredited Victorian opponents called "sound money," 20th-century macroeconomics belongs in the same file as, say, psychoanalysis. Ie: it is pathological science.

Keynes and Freud certainly do make good bookends. Once you read Rothbard's essay on the former, you'll realize that any school which names him as its founding father is in need of serious cranial examination.

Mencius writes:

fundamentalist,

I'm not so sure I believe the conventional Austrian analysis about rising prices of consumer goods. I think this is a spurious result of Mises' tendency to try to explain financial fluctuations in "real" terms - an approach very typical of his era.

So, for example, Austrians will say that lenders at short maturity lend short because they have a "desire for present satisfaction." This is certainly accurate in a praxeological sense. But this need not be confused with their motivation to stop lending short, which is what causes the crisis.

In theory, lenders/savers could withdraw (cease rolling over) because they all wanted to buy Doritos and Pepsi at the same time. In which case the prices of Doritos and Pepsi would indeed go up. But they could withdraw for any other reason, and - more to the point - once they start withdrawing, the crisis generates its own momentum.

The motivation becomes the insolvency of the financial institutions. Lenders withdraw because they want their money back, sauve qui peut, devil take the hindmost. And, maturity crises being maturity crises, the devil will always take someone.

The result of the panic is just a contraction in the cash price of financial assets, which makes people feel poorer. And praxeologically, it is hard to get from people feeling poorer to people spending more.

fundamentalist writes:

Mencius: “I'm not so sure I believe the conventional Austrian analysis about rising prices of consumer goods.”

If the money supply increases without an increase in the production of consumer goods, you don’t think consumer prices will increase? Maybe I don’t understand what you’re saying.

Mencius: “But they could withdraw for any other reason, and - more to the point - once they start withdrawing, the crisis generates its own momentum.”

Lenders would quit lending short because they want the cash, but based on the Minneapolis Fed’s myths of the crisis, it doesn’t appear that lenders have quit lending short term. What happened was that some businesses in the capital goods industries, mainly housing manufacturers, started going broke and couldn’t pay back their loans. That caused the highly leveraged credit structure to unwind as everyone sought cash to pay liabilities.

Mencius writes:

fundamentalist,

Some Austrians argue that one epiphenomenon of a crisis should be a rise in the demand for consumer goods and a recovery of consumer industries, at the same time as longer-term malinvestments are liquidated. Reading your original comment more carefully, you were making a slightly different point. So you're right - my response is not pertinent.

I have a different response to the point you were actually making, though, which is that even though money supply in the base M0 sense is rising, MZM is flat and personal net worth is declining sharply. Mortgage-equity withdrawals, especially, have ceased for obvious reasons. There is no statistic which tells us how much of peoples' portfolios they expect to use as near-term spending money, but as their assets fall we should certainly see an increase in saving and a decrease in spending.

As for the Minneapolis Fed, frankly, I have my doubts.

fundamentalist writes:

Mencius: "...MZM is flat and personal net worth is declining sharply."

Yes, that's because the Feds expand the money supply mainly through credit expansion and fewer businesses are borrowing. The volume of businesses paying off loans or going broke, which caused money to contract, is balanced by about the same volume of new loans, which causes the expansion of money.

Mencius: "...we should certainly see an increase in saving and a decrease in spending."

That will give Keynesians a heart attack. But Austrians think that is exactly what is needed. With increased savings and lower spending, profits fall in the consumer goods industries. That kicks the Ricardo Effect in reverse because lower profits means higher wages relative to the price of output. Producers will again start buying labor-saving equipment in order to reduce labor costs, causing the producers of that equipment to hire more. The producers of equipment will get the funds to hire more and produce more from the increased savings. Then we will be back on the road to recovery and the next boom caused by interest rates being too low.

Mencius writes:

Hm - the Ricardo reasoning seems a bit tenuous to me. The effect of deflation on both wages and prices should be, well, deflationary. The change in the wage/price ratio is not obvious to me.

What is clear is that profits are going to be negative and a lot of capital is going to be liquidated. Unless, of course, the CBs find a way to subsidize it.

fundamentalist writes:

Mencius: “….the Ricardo reasoning seems a bit tenuous to me. The effect of deflation on both wages and prices should be, well, deflationary. The change in the wage/price ratio is not obvious to me.”

No, it’s not obvious. It takes a while to get your head around it. But the Ricardo Effect is nothing but the macro results of the micro optimization of inputs taught in every micro econ class.

Empirically, wages and prices don’t move in lock step. Wages lag behind prices by quite a bit. If they moved in lock step, profits would almost always be a constant percentage of revenue. The fact that profits rise and fall indicates that wages and prices change at different rates. Profits are high when wages are low relative to prices, and profits are low when wages are high relative to prices.

Mencius writes:

I understand the Ricardo effect - what I'm not quite sure of is your price-wage logic which invokes it. Labor costs are not the only non-capital factor in production. The chain of reasoning from decreased consumer spending to increased capital investment is quite a long one, and seems to depend on holding a lot of things constant that I'm not sure are.

The Snob writes:

Prof. Kling,

As someone who got his BA in the dismal science, I was curious how early in the educational process you think the macro rot sets in.

I focused my studies on micro and industrial organization, but was required to take intermediate macro as well. Compared to micro, which seemed based in logic and psychology as much as anything, even relatively basic macro felt to me like a sort of faith system where you simply had to accept that you could manipulate these great aggregations as undifferentiated lumps, or the whole edifice crumbled. Needless to say, I wrote my own baby-haiku for the sake of my GPA and got out as fast as I could.

Later, I dreamed about running a simulation of an entire economy based on some millions of individual actors, each acting according to micro principles, using a computational approach. Just as chemistry can be thought of as a subset of physics, is not macro a shorthand abstraction of a massively-parallel micro system? Has anyone done this type of simulation?

PT writes:

Of course Blanchard's paper is nonsense. What did you expect from a French theorist anyway?

fundamentalist writes:

Mencius: "Labor costs are not the only non-capital factor in production."

On the micro level, that's true, but not at the macro level. Inputs other than labor are just stored up labor. At the macro level they all conflate to labor and capital.

Mencius writes:

fundie,

I understand your thinking, but "stored up labor" is an awfully ethereal concept for me. I guess I prefer my abstractions to be a little more concrete.

Basically, the bottom line of the crash is that a lot of capital goods are repriced downward, a lot of investments are no longer viable at the new, higher interest rates and must be liquidated, a lot of people are less wealthy, etc. This creates a completely new equilibrium. I am not sure it is necessary to try to deduce specific details of this shock, such as new wage-price relationships - the event is painful enough as it is. For me it suffices to know that (a) it's bad, and (b) it is caused by unsustainable credit expansion.

L. Burke Files writes:

The consequences are clear - at least to anyone with a wee bit of history under their money-belt. The diaspora of losses will be concentrated so that those who have money will be able to snap them up at 30 to 50 cents on the dollars - hold them for a period of time and meter the assets back out into the market.

Farm property in the 1930's, Commercial property in the late 60's early 70's, property Limited Partnership in the 80's along with the RTC, and now residential property.

But it is going to get worse if government has its way as government tries to fix what government has caused.

We face continued market distortions as long as the systemic risk posed by bad, yet nearly identical world wide, accounting requirement and regulations persist. Further, it, this systemic risk, is getting worse as all of the financial regulations are slowly being harmonized by and or through the initiatives proffered by the Bank for International Settlements. (www.bis.org for more info.) Mind you they are not bad, they just do not understand the fragility of a worldwide financial systems with all of the same rules. If this too leaves you a bit confused - imagine if all Sheep had the exact same DNA the population becomes fragile, in the same was as the markets, and one virus could wipeout the entire population.

This is what we are seeing now - waves of selling and buying that are reinforced around the globe - thus swings are larger and wilder than ever before. It is a symptom of this systemic risk.

All I can say is I hope bi-partisinship fails and the government is deadlocked for the next 20 years, it would be safer for all of us.

Steve Roth writes:

Paul Romer points out that the economists who opposed the bailout were primarily the theory/model-driven "fundamentalists."

http://www.growthcommissionblog.org/content/fundamentalists-versus-realists

You object to both the bailout and the modelistas. Worth discussing?

Comments for this entry have been closed
Return to top