Lesson 1 of 8
The facts on the table
Before any account is given, what one has to fit: comovement, the asymmetry, which industries swing hardest, and the link between credit and crises.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- Writing the facts down before the theories is the only way to stop each account being judged against the evidence it was built from.
- Five facts are widely agreed: comovement across sectors, the concentration of the swing in durable goods, the asymmetry between contractions and expansions, the absence of periodicity, and the link between credit and crises.
- Agreement on the list is not agreement on the explanation. Every account in this course fits some of it well and some of it badly.
This lesson is a list, and the list is the marking scheme for the seven lessons after it. Nothing here is controversial in itself; what is controversial is which facts an account is allowed to leave out.
Five things an account has to fit
- 1ComovementOutput, employment and investment move together across most industries
- 2Durables swing hardestCapital goods and durable consumer goods fall much further than the rest
- 3AsymmetryContractions predict the following expansions; expansions do not predict the next contraction
- 4No fixed periodRecurrence without regular length, so statistical tests find no cycle in the strict sense
- 5Credit and crisesCredit growth predicts financial crises across countries and over a century
Why the list comes first
Every school built its account around the facts that were most salient when it was written. Keynes wrote after a collapse in investment; the monetarists wrote after a decade of monetary history; real business cycle theory was built when the measurement of productivity was improving; the financial cycle literature dates from after 2008.
If each theory is judged on the fact it was designed to explain, all of them pass. The list is here so that a reader can ask the harder question: what does this account say about the facts it was not built from.
The five, with their sources
Comovement is the reason the cycle is a subject at all. If industries moved independently there would be nothing aggregate to explain, and any account that explains a downturn in one sector has to say how it spreads.
The concentration in durables is agreed by everyone and explained differently. This tradition explains it through the time structure of production. A critic in the previous course explains it through the ordinary economics of durable goods: a buyer can postpone replacing a stock, so demand for anything durable swings more than income does.
The asymmetry comes from the plucking evidence. In US unemployment data since 1948, the size of a contraction predicts the following expansion closely, while the size of an expansion tells you almost nothing about the next contraction.
The absence of a fixed period is a point both sides use. Statistical tests find a random walk rather than a periodic cycle, which a critic uses against theories that promise a cycle and a defender answers by saying the theory never claimed periodicity.
The credit and crises link is the newest and the strongest, and it has its own lesson at the end of this course.
What the list leaves out
Two things, deliberately.
It leaves out anything about policy, because a theory can be right about a mechanism and wrong about what to do, and the two questions get confused fastest when they are on the same list.
And it leaves out the distinctive claims of any one account, including the claim that the structure of production lengthens and shortens. That claim is not on the list because it is not an agreed fact: it is a theoretical proposition awaiting the measurement described in the previous course.
Check yourself
Why write the facts down before the theories?
- Because the facts are more reliable than the theories
- So each account can be measured against evidence it was not built from
- Because the theories disagree about the facts
Show the answer
Answer: B.
Every account fits the evidence that was salient when it was written. The interesting question is what it says about everything else.What does the asymmetry fact say?
- That expansions are shorter than contractions
- That a contraction predicts the size of the following expansion, while an expansion does not predict the next contraction
- That unemployment never returns to its previous level
Show the answer
Answer: B.
That is the plucking property, measured on US unemployment data from 1948 to 2020 and reported in the previous course.Why is the structure of production not on the list?
- Because it is unimportant
- Because it is a theoretical claim rather than an agreed measured fact
- Because no one has proposed it
Show the answer
Answer: B.
The list holds facts every account has to fit. The stage structure is a proposition this track has argued about and which the evidence does not yet settle.Sources
- Dupraz, Nakamura and Steinsson, 'A Plucking Model of Business Cycles', NBER Working Paper 26351, revised June 2021, section two, for the asymmetry between contractions and expansions in US unemployment.
- Schularick and Taylor, 'Credit Booms Gone Bust', NBER Working Paper 15512, revised July 2011, for credit growth as a predictor of financial crises across fourteen countries from 1870 to 2008.
- Caplan, 'Why I Am Not an Austrian Economist', section 3.4.2, for the durable goods explanation of why producer goods industries swing hardest.
- Tullock, 'Why the Austrians Are Wrong About Depressions', Review of Austrian Economics, volume two, 1987, for the point that statistical tests find a random walk rather than a periodic cycle, and Garrison's article in volume three of the same journal for the reply that periodicity was never claimed.
What this leaves out
- This is a heterodox tradition. The mainstream account of the same phenomenon differs, and nothing here is a forecast or a view on any security.
- This list is not exhaustive and other writers would include other regularities. It is the set of facts this course can source to work it has read.
- Nothing here ranks the accounts that follow. The list is a scoring sheet the reader fills in, not one this course fills in for them.