Shopping at Amazon?
Wait! If you click on the Amazon banner link at the top of the page first, HCN will get a percentage of anything you purchase for 24 hours!
Open predictions
Our local book authors
User login
Who's online
There are currently 3 members and 135 guests online.
Currently online
Recent Recommendations
Another word of caution: many economists, including former Fed Chairman Greenspan, have little confidence in the Conference Board leading indicators.
Oh yeah?
Well Mr. Kasriel has some rather harsh words for those economists:
(pdf) When The Facts Change, I Change My Model - What Do You Do?
ac, thanks for linking to Kasriel's article!
Best Wishes.
Update: And Northern Trust's Paul Kasriel (hat tip ac) argues that the LEI has value:
Note, however, that he doesn't make a similar statement about the economists.
Put this in you LEI pipe and smoke it.
INO Futures and Commodities - Energy - RBOB GASOLINE Jan 2010 (NYMEX:RB.F10) Price Chart and Quote
As per the Kasriel article, the Wright Model "B" yield-curve also isn't subject to revision.
It inverted in Q3 2000, in advance of the 2001 recession.
It inverted in Q1 1989, in advance of the 1990 recession.
It remained inverted almost continually from Q4 1978-Q3 1982, with 2 recession occurring during that stretch.
It inverted in Q2 1973, in anticipation of recession in 1974.
It has not inverted in this cycle.
Sebastia
CR,
I found some interesting information on the MBA purchase index "conundrum." (Ignore the first sentence as it talks about manufacturing.)
Link
--
Any indicator that predicts a recession is to be discredited, or down-played. We simply can't handle a recession with the level of Debt and Leverage that exist.
Jas
Note that the March LEI was revised up to 0.6% from 0.1%. That revision just about offsets the forecasting miss for April. That is to say, if we average things a bit, nothing has changed much.
Same with the big starts number for April. It followed downward revisions to prior months. The 3-month average for starts is just about where it was generally expected to be, prior to the release of the April data and prior revisions. No real sign of a greater slowdown (LEI was already signaling below average growth), and no real sign of a pick-up in starts, seen over a three month period. Going nowhere fast.
Is it elitist to say that bloggers without unique data can't resolve points of contention that the most educated, experienced and connected disagree over?
Beating the mainstream media is easy work. I give the FRB more credit.
ECRI's criticism of one of the Conference Board leading indicators:
ECRI | Home
One of the strong points about our economy is that it is flexible enough to change quite drastically in a decade or two. It's not surprising that economic measurements and indices would have to change with it.
ECRI doesn't like the yield curve as a leading indicator in general. While it's true that is has proved to be a successful predictor of recessions in the United States, it has not provided the same clues in other economies (as Macro Man points out). Because of this Moore never considered it a fundamental driver of the business cycle, despite its success in the US.
Hopefully Sebastian will see this...
Sebastian, the Wright B fits for the period indicated. There are two problems, both noted by Mr. Wright in his paper that presented this model, which go a long way toward explaining why the model cannot be completely trusted.
Problem one: The model doesn't work in most other nations. Problem two: The model doesn't work for recessions prior to the period discussed. Mr. Wright posits an explanation for both, which I'll rephrase as: Assuming the fund rate is set using the principles used in the US since 1970, and assuming the bond rates are not influenced so as to misrepresent market forces, Model B works very very well.
If any of those assumptions are voided, the model's effectiveness is also voided.
I'm also going to point out Mr. Wright's reason for not giving whole reliance upon the model. There haven't been enough recessions to test it. It looks good, it might be good, but relying upon it utterly would be like... buying a house with as large an ARM as you can grab because, for the past decade and a half, by the time the reset comes around you can sell it for a serious profit. Never mind the history of the decades before, or the experiences of people in other places.
I'll supplement with a link to this Minyanville article. To summarize, John Succo asked an unnamed large broker firm to explain why CDOs are stable when the collateral values of the market are deteriorating. The sum of the answer is "conflict of interest" - the agency(ies) trusted to provide an impartial pricing mechanism have a vested interest in the pricing remaining high.
Since there is a relationship between CDOs and 3-month (among other) notes, this means at least one of the assumptions necessary for Wright B to work (see preceding post) is questionable. Not necessarily void, but questionable.
Kirk Spencer,
If I have read correctly, Mr Succo also states that the "mark to market" price (based on the rating) may be higher than the actual price at which trades are being executed.
Whoa!!
Kirk Spencer said: "I'm also going to point out Mr. Wright's reason for not giving whole reliance upon the model. There haven't been enough recessions to test it. It looks good, it might be good, but relying upon it utterly would be like... buying a house with as large an ARM as you can grab because, for the past decade and a half, by the time the reset comes around you can sell it for a serious profit..."
Thanks for the response.
A couple of points:
I've never, ever suggested that anyone rely on any single indicator, I've always said to use multiple indicators and cross-check them for confirmations/non-confirmations.
The last time the Model "B" yield curve "missed" a recession was 1960, and it's captured every one since. So even as a "stand-alone" indicator it beats every indicator (measured over the same timeframe, and shorter ones) that I've ever seen the bears use to justify their position.
Finally, I never rely on an indicator just because of its "track record," which could just be a statistical coincidence. I always make sure I understand how the indicator works and why before I use it.
Sebastia
Sebastian, I agree you've never said you don't use multiple indicators. However, you have said we cannot possibly have a recession because this (and the unemployment) triggers have not presented themselves. That's a different argument, and it's the one I'm refuting.
I'll point out that another thread in this forum shows why the unemployment dog may not have been barking. I also - as implied by the Succo article to which I linked - demonstrated that there's reason to believe the Wright B display may fail to show. These are done to show your absolute belief that a recession is not possible in the near future is, if not unsubstantiated, placing great trust on less than trusworthy evidence.
Thanks for the link to the article. Only just found your blog, but very interesting reading. Thanks.