Roman Empire ‘raised HIV threat’ BBC
The failure of two empires Dmitry Shlapentok, Asia Times
Car sales collapse by almost a fifth as loan drought and fuel rises bite The Independent
Price to Rent Ratio Calculated Risk
The Alternative Asset Management Industry: Inexorable Forces for Change Roger Ehrenberg
The Perilous Price of Oil George Soros, New York Review of Books
Shadowstats debunked Econbrowser. Anyone with specific knowledge is encouraged to comment. I have just skimmed the BLS paper referenced, but, for instance, in discussing geometric weighting, it uses examples within the food category. I have been told (I thought it was as a result of geometric weighting, but perhaps came about otherwise) that health care, which as we all know has been rising more rapidly than most costs (except commodities in the last year) is very much under-represented in the CPI. My dim recollection is 6% versus about 16% of GDP. Links and sources very much appreciated.
Antidote du jour:
10 comments
Comments are closed.
I do have a question about Shadowstats to which I’ve never found a good answer. I recently discovered this site and I’m impressed with Yves and the commenters so this seems a good time to ask…
As I understand it the main change beteween Williams’ pre-1983 and the post-1983/pre-Boskin method was to change from tracking home prices (or payments) to owner’s equivalent rent. If that’s the case, then the spread beteen the pre-1983 and post-1983 levels should be falling dramatically (or even going negative) from say 2005, because home prices were rising fast back then and now are falling just as fast.
However, the spread seems to remain fairly constant. Any ideas?
Sorry for the lack of references… it’s way past my bedtime but I can follow up with more tomorrow if this is a topic of interest.
thanks,
rich
The thing I have always found odd about shadowstats is that a lot of his figures figures pretty much differ by a constant spread from official government figures. I have always had difficulty believing the various changes in government methodology only had that simple an effect on reported figures.
The thing I find odd about economists who criticize shadowstats is that they try to debate the merits of the government’s changes in claculating GDP, CPI, etc, by saying the old methodologies were wildly wrong and the new ones are way more accurate. That may be true, but it is completely irrelevant. shadowstats subscribers aren’t paying shadowstats to measure economic reality, just to back out changes in government methodology.
The Last American Homeowner clinging to his perch with a view.
The argument by the BLS seemed more hair-splitting than meaningful.
It is a justification of their method changes, and not a proof that anything that Shadowstats claims is actually incorrect in the net result.
As someone else said above, Shadowstats primary function is to back out the changes. I believe John Williams is able to show attributions, and may even have done so in the past. I cannot recall specifically.
But the point remains that there has not been an explanation and debate over MANY of the changes that the government institutes in its economic statistics. Why get rid of M3 reporting because ‘it costs to much to calculate.’
Blind acceptance of government numbers, any ‘official numbers’ for that matter, is weak, and upholding them for the sake of sounding wise is lame.
Most recently several took it upon themselves to challenge the ADP employment calculations and some bad assumptions were unofficialy acknowledged because of changed environment, and a lack of willingness to expend the effort to change them.
Anyone who accepts official statistics and methods changes without peer review and debate before the fact should expect an enduring loss of credibility.
Based on my own detailed examination of government numbers, I would prescribe a strong skepticism to them all, not for any conspiratorial intent, but rather for the natural course of human error and foibles.
And anyone who thinks that there is no pressure on the numbers people from political sources has never worked in Washington or is simply maive.
Any “inflation” statistic is politically motivated. This is von Mises “index number problem”. All the “technical” objections to shadowstats amount to: the author of shadowstats has no PhD, we are smarter than he, trust us. Buncombe. I have said for decades that US official inflation statistics are driven by a need to reduce “tax bracket expansion” and reduce real social security payments. That’s all you have to know. The rest is economists playing with numbers. I’m A CPA, I see numbers being played with daily. Chains, links, geometric, arithmetic, fuggedabout it. It makes no difference in the long run. The BLS and Fed want you to hold Uncle Sam’s paper. I say: do it at your own risk. And have a nice day.
The thing I have always found odd about shadowstats is that a lot of his figures figures pretty much differ by a constant spread from official government figures.
This is not an accident. This is by design. I noticed this myself, and so I went hunting through his “about” pages to find his methodology.
It seems that John Williams does not actually recompute CPI etc using the old techniques. What he does is take the current measurements, and then offsets them with a correction factor. This correction factor is derived from some internal government document created around the time that the measure is changed (e.g. someone writes a memo claiming that the new CPI will be 2% lower). Hence, his system has the same problems that the new government measurements have, namely that they cannot meaningfully be compared with the original measurements. So all he is doing is providing us with yet another alternate set of metrics, and a dubious set at that.
This is a really shame. We desperately need a site that actually does what Shadow Stats claims (but fails) to do. If I hear another mathematically illiterate economist/pundit numerically compare 2000s CPI with 1980s CPI (or 2000s unemployment measurements with 1930s unemployment measurements), my head is going to explode.
walker: “It seems that John Williams does not actually recompute CPI etc using the old techniques. What he does is take the current measurements, and then offsets them with a correction factor.”
I think Williams / shadowstats basically calculated the spread between the last reported government statisic before and after a change in methodology, and then assumes reports of the statistic post-change are off by that spread. Obviously, this would be simplistic and wrong, but it is a lot easier and cheaper for him to do something like this, than to actually try to calculate the numbers. I haven’t been willing to subscribe to his service because I am suspicious of his methodology.
I have been told … that health care … is very much under-represented in the CPI. My dim recollection is 6% versus about 16% of GDP.
There is no reason why the goods & services basket covered by the CPI should be the same as the goods and services composition of GDP. In fact, they are different by design and there is nothing sinister about this (the consumer basket is mainly PCE, while GDP has several other components).
Prima facie, I am however surprised that healthcare would have a lower weight in the consumer's basket than in GDP – perhaps this is accounted for by government expenditure on healthcare.
While Williams’ methodology is the main issue I believe Elaine Chao’s (wife of KY Senator Mitch McConell-R) character and performance at United Way is an important aspect of this discussion.
Please note I resisted any reference to jelly. Ok, I didn’t.
The real question is why the BLS would see any need to debunk Shadowstats.
Unless of course they knew they were wrong about stuff.