Wednesday, December 21, 2011

The Interaction of Cycles

But there is no ground to believe that there should be just one wave-like movement pervading economic life.
-- Joseph A. Schumpeter

Wandering through Google results for unemployment historical data the other day turned up The Natural Rate of Unemployment, a PDF by David Brauer of the Congressional Budget Office. (16 pages, lots of pictures.)

The paper is from April, 2007 -- shortly before the crisis. Brauer's Figure 2 shows a prediction based on pre-crisis thinking:

Brauer's Figure 2: Demographics and the Natural Rate of Unemployment

David Brauer writes:

In its summer 2006 forecast, CBO reduced its estimate of the current natural rate from 5.2 percent to 5.0 percent, reflecting a decline since the mid-to-late 1990s in the share of the labor force that is below the age of 25. That decline is attributable not to a change in the age mix of the general population, but to an unexpected sharp and sustained drop in the labor force participation rate of teens and young adults, most likely because of higher rates of school enrollment.

Weird science:

1. People can't find work.
2. People go to school to improve their chances of finding work.
3. The "labor force participation rate" falls because people are in school.
4. The "natural rate of unemployment" falls.

It's sort of like: People can't afford so much steak, so they switch to pasta. And then steak is taken out of the "basket of goods" used to figure inflation, and pasta is put in. Because of this change, the inflation rate number is less. This is sick-puppy arithmetic, all of it. And bad economics.

But that's not why I called this meeting.

I think it's a little funny that the projected rate is five point oh! percent. The number wants to convey more accuracy than it can hold, I think. And once that trend line hits 5.0, it is rock-solid. A straight-line projection out to 2015 and beyond. Laughable.

I also think it's a little funny that in 2006 when CBO reduced its projection to 5.0 percent, apparently they reduced it all the way back to the year 2000.

We have always been at war with Eastasia. But that's not why I called this meeting.

Here is the opening sentence from the Abstract of Brauer's paper:

This paper assesses the natural rate of unemployment—the unemployment rate that arises from all sources other than fluctuations in demand associated with business cycles.

A simple definition. Unemployment arising from all sources other than business cycles. And that brings us to today's agenda: the fluctuations that are associated with business cycles.

In his Kondratieff rejection piece, Murray Rothbard wrote:

... the market is a seamless web. All facets of the market are interconnected through the price system, and the profit-and-loss motive. Booms and busts spread throughout the system; that is precisely why they are important. It is absurd to think that one part of the economy can peg along on a nine-year cycle, another on a three-year cycle, and still another on a 25-year cycle, with each of these cycles barreling along on a hermetically sealed track, not influencing and modifying each other. In fact, there can only be one real cycle going on in the economy at anyone time.

Rothbard is right:  It *IS* absurd to imagine completely separate cycles existing with absolutely no influence on each other. On the other hand, it is not absurd to imagine that more than one cycle might arise from economic forces generated by concurrent economic activity of billions of people spread over the face of a planet. Nor is it absurd to imagine such cycles themselves in continuous interaction.

It is, however, absurd to invent the notion of the non-interaction of cycles, pretend that others invented the notion, and then reject the work of others by saying that the non-interaction of cycles is absurd.

In Business Cycles (PDF, 385 pages apparently), on page 209 Schumpeter attempts to present an extremely simplified or "clean" version of the interaction of business cycles. He limits his presentation to three cycles only, "for the sake of simplicity". He eliminates not only other cycles but also "external disturbances" and also "Seasonals and Growth". And he makes his three cycles as regular and unvarying as possible by using "three sine curves the amplitudes of which are proportional to their duration".

Schumpeter warns that this is a great simplification: "Of course, we do not, as a matter of principle, postulate either internal regularity or sine form...

...we may look upon the charts as an illustration of all the boldest assumptions which it is possible, and to some extent permissible, to make in order to simplify description and to construct an ideal schema with which to compare observations. In particular, all cycles have four phases of equal length, amplitudes of plus and minus excursions are equal and constant, periods are also constant, and each of the two higher cycles consists of an integral and constant number of units of the next lower movement."

And then Schumpeter provides Chart 1, showing the interaction of cycles:

Take three extremely regular wave-patterns and add them together, and you get a surprisingly complex shape.


For the stranger to statistical technique the fact alone that extreme regularity of but three components may result in so very irregular-looking a composite should be instructive.

Schumpeter argues that the cycles *DO* interact.

I didn't find the reference, but Schumpeter also argues that if time and chance should bring the lows of various cycles together, the combined low would be deeper than any of the individual lows alone. You can see this behavior on his chart.

Schumpeter said this concurrence could explain the severity of the Great Depression. I don't necessarily buy that explanation of the Great Depression. But I certainly buy that such a thing could happen. So I have to reject Rothbard's assertion that "there can only be one" cycle.


Jazzbumpa said...

I think you would do yourself a favor if you just forgot about Rothbard. He's a hack.

It's sort of like: People can't afford so much steak, so they switch to pasta. And then steak is taken out of the "basket of goods" used to figure inflation, and pasta is put in. Because of this change, the inflation rate number is less. This is sick-puppy arithmetic, all of it. And bad economics.

Funny you should use this example. I just saw an official takedown in the last couple of days of shadowstats using a steak-hamburger example. Sadly, don't remember where. This isn't it, but it's related.

Aha - found it!

Anyway, your analogy is unsound. Brauer is talking about why the natural rate would legitimately change.

Five percent is a roundish number. Your criticism would make more sense if he had landed on 5.1 or 4.9.

Now when you get to adding cycles, with all their non-sinusoidal intricacies, at trend levels from seconds to millenia, you get Elliott waves.


The Arthurian said...


5.0 has exactly the same accuracy as 5.1 or 4.9... ONE DECIMAL PLACE.

If they just used a one-digit estimate I would not have belittled them. FIVE is far different from FIVE POINT OH!

The Arthurian said...

Change To Inflation Measurement On Table As Part Of Budget Talks -Aides -- NASDAQ
By Corey Boles and Janet Hook

WASHINGTON -(Dow Jones)- Lawmakers are considering changing how the Consumer Price Index is calculated, a move that could save perhaps $220 billion and represent significant progress in the ongoing federal debt ceiling and deficit reduction talks.

According to congressional aides familiar with the discussions, the proposal would shift how the Consumer Price Index is calculated to reflect how people tend to change spending patterns when prices increase.

Jazz, not only have they done it before... They are going to do it again.

From the link you provided, this is exactly right: "Of course, replacing hamburger for steak in the calculations would reduce the inflation rate, but it represented the rate of inflation in terms of maintaining a declining standard of living."

Are you saying these substitutions do not take place?

The Arthurian said...

Jazz, from your link:

And here's the response by Greenlees and McClelland: To begin, it must be stated unequivocally that the BLS does not assume that consumers substitute hamburger for steak...

Yeah, they DO NOT assume that people stop buying expensive meat and start buying cheaper cuts.

Nevertheless, people do. Amd people buy smaller cars. And people go out less often to dinner. And people watch Netflix instead of going out to see a movie. You can call that technological change. I call it cost-cutting.

In addition, when people stop buying steak and start buying cheaper stuff, the reduced demand artificially reduces the inflation rate of steak. So even if the CPI counted steak instead of crapmeat, inflation would be understated.

// Idunno how objective this link is, but the parts I quote are historical:

The government has made significant changes to how it calculates inflation over the last 30 years. Many believe that these changes have led to an understatement in the current inflation data.

The most common inflation measurement talked about is the Consumer Price Index (“CPI”). CPI is the weighted average of prices of a basket of consumer goods and services, such as transportation, food and medical care. The CPI is calculated by taking price changes for each item in the predetermined basket of goods and averaging them.

Starting in 1983, the government significantly changed the way it calculates CPI. It has also made a few major changes since that time that would lead many to wonder if current inflation data is understated.

The three changes are:
1. In 1983, housing prices were removed and “owner’s equivalent rent” was used instead.
2. In 1998, Hedonics was introduced to estimate a product’s value as technology advances.
3. In 1999, the idea of product substitutions was introduced to the index.

Through Hedonics, the government calculates an adjusted price of the original TV ($150) that includes the quality enhancements (perhaps an additional $900), so the total calculated price would be $1,050. Many times, as in the case just illustrated, the price of the improved product as calculated is higher ($1,050) than the retail price you’re paying ($850). Therefore in the government’s eyes, the price is going down, not up, because the government says you’re getting more for your money.

In 1999, the CPI began using product substitutions as a part of the process. This simply means that as prices go up, consumers will substitute the originally desired product with an alternative. If your family loves steak, and the price of steak goes up, the government believes your family will stop purchasing steak and buy hamburger instead. So if the price of steak and hamburger both go up, steak can be removed from the index and replaced with hamburger. Even if hamburger now costs what the price of steak was before the increase, the index would show no change in inflation...

The rebuttal in your DeLong link was incomprehensible.

Jazzbumpa said...

Art -

I am on the verge of exhaustion, and cannot get into a lengthy discussion on all of this.

I'll just suggest that if anyone has shifted from steak to hamburger, or to a smaller car in the last several years, it hasn't been because inflation has eroded their purchasing power.

It's been because lack of money has been reducing their purchasing power.


The Arthurian said...

(some years later)
From Irving Fisher, The Debt-Deflation Theory of Great Depressions

4. The old and apparently still persistent notion of "the" business cycle, as a single, simple, self-generating cycle (analogous to that of a pendulum swinging under influence of the single force of gravity) and as actually realized historically in regularly recurring crises, is a myth. Instead of one force there are many forces. Specifically, instead of one cycle, there are many co-existing cycles, constantly aggravating or neutralizing each other, as well as co-existing with many non-cyclical forces...

The Arthurian said...

And the concluding thought from Irving Fisher's article 4:

"Any one cycle, however perfect and like a sine curve it may tend to be, is sure to be interfered with by other tendencies."

As Schumpeter showed.