Look: I am eager to learn stuff I don't know--which requires actively courting and posting smart disagreement.

But as you will understand, I don't like to post things that mischaracterize and are aimed to mislead.

-- Brad Delong

Copyright Notice

Everything that appears on this blog is the copyrighted property of somebody. Often, but not always, that somebody is me. For things that are not mine, I either have obtained permission, or claim fair use. Feel free to quote me, but attribute, please. My photos and poetry are dear to my heart, and may not be used without permission. Ditto, my other intellectual property, such as charts and graphs. I'm probably willing to share. Let's talk. Violators will be damned for all eternity to the circle of hell populated by Rosanne Barr, Lady Gaga, and trombonists who are unable play in tune. You cannot possibly imagine the agony. If you have a question, email me: jazzbumpa@gmail.com. I'll answer when I feel like it. Cheers!

Wednesday, September 14, 2011

Correlation and Causation

In comments late last night, Stone Glasgow pointed out that correlation is not causation, which is, of course, a fact.  It falls into the category of true, but irrelevant

The way I look at it, there are meaningful correlations and non-meaningful correlations.

In the meaningful case, there might be common cause, or there might actually be cause and effect.  If there is common cause, at least one phenomenon serves as a barometer for the other.  Help me out here, if I'm missing any other possibilities.

In the non-meaningful case, an apparent correlation may simply be a data artifact, or there can be pure coincidence, or what I call "coincident correlation" where two unrelated phenomena happen to follow a similar course for a while, before veering off on their own trajectories.

To sort out if a correlation is  meaningful, you have to put together a coherent narrative.  In the case of this graph, I've done that.  Here is some elaboration.  The indicator value of finance sector peaks in identifying recessions since WW II is ten for ten - 100%.  On the contrary, there is one false positive in 1986.  This gives us 10 for 11, overall, or 90.9%.  Few indicators achieve that kind of reliability.

As additional confirmation, consider the times when the percentage of finance sector capture is below trend for several years in a row.  These times are specifically the longest recession-free periods in the time frame.  Politically, these are the Kennedy-Johnson and Clinton administrations.

To call this correlation non-meaningful is asking an awful lot of coincidence.  Occum's razor firmly in hand, I prefer to think that policy matters.  Recessions are a characteristic of Republican administrations.   This relates to tax enforcement policy, deregulation that enables capital misallocation, fiscal and monetary policies, and the wealth disparity that has grown continuously since about 1980.

My coherent narrative lends credence to the meaningful correlation between high finance sector profits and poor economic performance.

High finance sector profits contribute to a concentration of wealth at the top of society.  This is ultimately destabilizing.  It happened in the 30's and it's happening again now.


.

3 comments:

The Arthurian said...

That was an interesting exchange, I thought.

One thumb down: "The indicator value of finance sector peaks in identifying recessions since WW II is ten for ten - 100%."

There are probably a lot of datasets of which this is true. (This is just a first reaction, not a well-considered reply. But my initial reaction is that your claim is ...probably not meaningful.)

One thumb up: "As additional confirmation, consider the times when the percentage of finance sector capture is below trend for several years in a row. These times are specifically the longest recession-free periods in the time frame."

This is a brilliant insight. I like it. I have this.

And one simple observation: "High finance sector profits contribute to a concentration of wealth at the top of society."

Yes, and -- apart from the relative fall of wealth and income for everyone below the top -- the increasing income of the financial sector is an increasing cost to the non-financial sector.

Jazzbumpa said...

There are probably a lot of datasets of which this is true.

I very much doubt it.

"High finance sector profits contribute to a concentration of wealth at the top of society."

Exactly.

the increasing income of the financial sector is an increasing cost to the non-financial sector.

Right. It also involves the wrong use of debt.

Cheers!
JzB

The Arthurian said...

me: the increasing income of the financial sector is an increasing cost to the non-financial sector.

you: Right. It also involves the wrong use of debt.

me: As financial costs increase for the non-financial sector, financial profits increase relative to non-financial profits. Because of that, the financial sector is more appealing to investors than the non-financial sector.

The thing that drives the "wrong use of debt" is that profits are better in the financial sector than elsewhere.