Housing, Jobs and Bernanke Revisited

Good post CR... I think that is the biggest mistake people make when looking at stats... confusing cause with effect and/or coincidence and ASSUMING a correlation proves a causation.

I look at process data as part of my engineering past… and looking at Hamilton’s graph you can see a ‘line’… But if that was data predicting the strength of a material and I was being asked if I was comfortable with the relationship enough to use that material in a design of say a bridge. I would probably be comfortable using the relationship from the data points in the range x = -4 to x =4. Whether or not there is proof of causation... the data points are so tight as to suggest a 'useful' relationship.

But it looks like the relationship blows up from about x>4 and up... something else is going on in that region.

So maybe if you teased the numbers you could make the case that job growth causes small housing price increases… but there is another undetermined variable in there too that isn't well'behaved'. My guess is that variable is ‘speculation’… how you measure and model that is anyone’s guess.

Interesting post CR...

I must take exception to your insistance that Dr. Hamilton is confusing correlation with causation ... It simply is not in his piece ... Both he and Bernanke suggest that tmultiple factors appear to be at work ... No where is it asserted that job growth is the single cause of higher home prices ... a simple model was offered to point out just one of the factors that Bernanke called out in his comments ...

You appear to be the one who is insisting that we have a housing bubble and attacking all who explore alternative views as to what is happening in the economy.

The economy is almost always driven by a sector ...that is one sector is growing more rapdily than others ... this occurs because we aggregate multiple sectors ... please note that I am not asserting a singular factor causal argument ... I am just making an empiracle observation ...

Hey Simon,

Umm, the 'correlation vs. causation' thing has nothing to do with a 'single factor model.'

CR has, correctly in my view, pointed out that the direction of causation might flow from RE-related employment to prices. At the very least, it can be argued they are jointly determined. In such a case, the estimated coefficient will be biased up.

Prof. Hamilton knows this to be sure; he is an econometrician of the highest reputation. However, perhaps he overlooked mentioning this possibility in this case.

Hamilton uses statistical regression analysis to demonstrate a “highly statistically significant” “relation” between housing prices and employment and offers this interpretation of the result:

“While speculative behavior appears to be surfacing in some local markets, strong economic fundamentals are contributing importantly to the housing boom.”

In other words, employment growth causes increases in housing prices.

Verdict: Guilty as charged.

Job growth was higher in the 1995-2000 period in most areas and housing prices (with the exception of the Bay area) were nowhere near as high.

Since 2000, several areas that have had weak employment growth (if at all) like Boston, NYC metro, Bay Area in Cal have seen large gains in prices.

So my conclusion is that job growth in some areas explains some price rises, but not all or even most.

Also, does Dr. Hamilton's model take a possible feedback loop into effect, especially in areas like So Cal where a great deal of new jobs are real estate driven ?

touche, who is guilty as charged? you and/or CR or ...?

“While speculative behavior appears to be surfacing in some local markets, strong economic fundamentals are contributing importantly to the housing boom.”

First,it is acknowledged that there is speculation ...
Second, strong economic fundamentals are contributing strongly ...

What is so strongly stated to imply causation?

Your assertion on regression is yet another clear example that nothing truly causal has been asserted unless you believe that regression is a causal construct ... if so, then the problem lies with you not with the Hamilton or Bernanke

Again, it looks to me that you are trying far too hard to push the points further than the either Bernanke or Hamilton have pushed their respective points.

The relation developed was qualified by the magnitude of the R^2.

I guess what bothers me is that I do not see the blindness that CR is alleging ...

Simon, if you check the comments for Dr. Hamilton's post I acknowledged my possible bias: "I am probably biased about housing - I am as confident as I ever get that there is a housing bubble".

Bernanke clearly implied causation - that jobs were one of several factors in higher house prices. I agree with Dr. Thronberg that causation is probably running the other way - that the house price bubble is creating jobs - the evidence for this assertion is in the large percentage of RE related jobs created in the last few years. Also it appears RE related jobs are primarily being created in the bubble states.

I respect Dr. Hamilton's opinion highly, but I am willing to challenge anyone's view - including my own.

Best to all!

Simon, after rereading your comments ... instead of suggesting that Dr. Hamilton "confused" correlation and causation, perhaps it would have been better to suggest that Dr. Hamilton's correlation post was insufficient to defend Bernanke's argued causation.

Best Regards.

Howdy Calc. New operating systems are a good thing!
Grand blog ya' got here. :•)

Simon sez “What is so strongly stated to imply causation?”

“strong economic fundamentals are contributing importantly to the housing boom.”

That’s a statement of causation.

In San Francisco's Sunset District, non descript houses are selling for $800k up $100k from a year ago. One in ten houses in these neighbor hoods is vacant and has been for a year or more. Some are being used by Asian gangs to grow marijuana. Since much of the new construction is out in commute-distance land and is being built by illegals being paid $10/hr cash; I doubt that housing construction is fueling the bubble.

This distinction between causation and correlation is a real pain in the butt. [Necessary and sufficient conditions are next.]
Who can I throw mud at first?
touche argues:
“strong economic fundamentals are contributing importantly to the housing boom.”

That’s a statement of causation.

I don't know t, not like " the freight train contributed importantly to the bee's sudden loss of headway upon collision."

The train caused the bee to stop. No question.
But things get mushy in economics. Which is why Nobel prizes in this field go to the strangest people.
And why we need non-economists like CR to straighten Boneheads like Bernanke out.
No one believes that high wages are causing the housing boom.
Not even Bernie.
Next.

Yeah, but read this:

Stormy's great post at Brad Setser's web log:

RGE Monitor

I do not see a conundrum at all. Every time the Fed raises short term rates, the long term rates flatten: Not a conundrum.

Industry is not a place for investment; it already is cash rich and labor plentiful. The only place now is in government bills. The reason why investment does not flow elsewhere is that it is simply not needed, thus driving down long term interest rates. Complicating and adding to this problem is that many of the factories of the west simply have move. A solid proportion of industry has not created new, additional factories; they have simply relocated. Their total output has not changed, only their profits have.

Does this not explain why there are interest-only mortgages? There is a cash glut. Bernanke, to some extent, is absolutely right.

There is an enormous flow of capital and money out of the industrialized world. Some of it does return in terms of profits and salaries, but a good percentage of those profits never find their way into public coffers. The rich get very rich. There is no way longer any way of really getting a slice of those profits to pay the national bills. In addition, downward pressure on average wages continues.

Cheap labor has been substituted for investment. Credit is cheap. With profits and cheap labor at these quantities, who needs credit? I do not see this as a conundrum. The pump has been primed from every conceivable direction: tax incentives and cheap labor abroad, tax cuts at home.

Ten million Chinese work for a foreign enterprise. The size of that cheap labor force is staggering. And it is only a tip of the Chinese iceberg. And we have yet to touch India. We are living through a boom economy, except the boom is not really heard here. We are paying for it. The dimensions of that boom cause Chinese entrepreneurs to open fake off-shore businesses, pose as foreign investors, and then be eligible for all the goodies--roundtrip investors, they are called. With credit like this, is there any wonder that capital returns cannot be measured by real interest rates? Hell, I should start such a business. Bet doing so is dirt cheap. No loans required. And, you do not need real investment in automation or equipment when labor is dirt cheap. Look at what the Egyptians did with the pyramids, or the Chinese with the Great Wall.

Normal Keynesian economics was not prepared for anything like globalization of this magnitude or style. There is a glut of capital and credit and labor. And it is coming at the American taxpayer's expense. The average American consumer is going deeper and deeper into debt. Average credit card debt in the U.S. is over $9000. And the interest rates are hefty there! And then place that number against the median wage: credit card debt then becomes one-fourth of the median wage. Some are living on a knife's edge here

Continued:

Some are living on a knife's edge here.

When the bubble does burst, who will pay the bills, personal and national?

The knot that has been tied is Gordian, as Volcker says. There is no easy way out of this one. We cannot wait until China becomes a "consumer society"-- a ten or fifteen year pipe dream. After China, there is India. After India, who knows.

Written by Stormy on 2005-07-24 00:16:41


Yup - I think stormy nailed it. Made me think quite a lot on the subject over the last day or so. Posted a bit on it over on AB.

I really think we are setting ourselves up for an economic equivalent of the 'perfect storm' if we don't get our collective act together.

Login or register to post comments
Syndicate content