Friday, December 31, 2004
The Island of Lost Posts III: Computer Price Deflation
by Tom Bozzo
Some of this was pure paranoia unworthy of further mention, but another strain of thought focused on quality adjustments to price measurements. The investment component of economic growth, for instance, would have looked worse if computer prices weren't measured to be dropping rapidly, so a dollar spent on IT hardware today constitutes more "real" investment than a dollar spent in the past.
Part of the reality is that there is a lot of sausage making to macroeconomic measurement. For background on price measurement with application to computers, I'd recommend Ariel Pakes' relatively unflinching "A Reconsideration of Hedonic Price Indexes with an Application to PC's" (American Economic Review Vol. 93 No. .5: 1578-1596).
For computers, there is actually a case to be made that the official computer price index doesn't measure enough deflation. The Consumer Price Index (CPI) series for computers (Dec. '97=100) stood at 33.9 when I got the PowerBook in February '01, and the most recent reading, from November, is 14.2. This can be interpreted as saying that I should be able to get a Feb. '01 PC now for 42% (14.2/33.9) of its price then, or about $1050 in the case of the PowerBook. Prices for all consumer goods increased by a bit less than 9% over the same period per CPI.
The anti-quality adjustment crowd would say, whoa, you can't actually hold quality constant and what's actually available is a $1999 Aluminum model, 80% of the old entry-level Titanium price. The CPI for computers is overstating deflation by this standard.
My response is that as much as I love my old PowerBook, it's not marketable now for anything close to the $1050 implied by the computer CPI. (With $1050 one can, for instance, get change back from the purchase of an iBook that is superior in most respects to an old Ti.) Computer deflation is by this standard understated in CPI.
There are several further rounds of objections and responses that could follow, but won't unless someone asks.
Through the election season, the comment sections on economic news releases at higher-traffic blogs that post on such things were rife with discussion about possible manipulation of economic statistics to bolster the Bush campaign.
Some of this was pure paranoia unworthy of further mention, but another strain of thought focused on quality adjustments to price measurements. The investment component of economic growth, for instance, would have looked worse if computer prices weren't measured to be dropping rapidly, so a dollar spent on IT hardware today constitutes more "real" investment than a dollar spent in the past.
Part of the reality is that there is a lot of sausage making to macroeconomic measurement. For background on price measurement with application to computers, I'd recommend Ariel Pakes' relatively unflinching "A Reconsideration of Hedonic Price Indexes with an Application to PC's" (American Economic Review Vol. 93 No. .5: 1578-1596).
For computers, there is actually a case to be made that the official computer price index doesn't measure enough deflation. The Consumer Price Index (CPI) series for computers (Dec. '97=100) stood at 33.9 when I got the PowerBook in February '01, and the most recent reading, from November, is 14.2. This can be interpreted as saying that I should be able to get a Feb. '01 PC now for 42% (14.2/33.9) of its price then, or about $1050 in the case of the PowerBook. Prices for all consumer goods increased by a bit less than 9% over the same period per CPI.
The anti-quality adjustment crowd would say, whoa, you can't actually hold quality constant and what's actually available is a $1999 Aluminum model, 80% of the old entry-level Titanium price. The CPI for computers is overstating deflation by this standard.
My response is that as much as I love my old PowerBook, it's not marketable now for anything close to the $1050 implied by the computer CPI. (With $1050 one can, for instance, get change back from the purchase of an iBook that is superior in most respects to an old Ti.) Computer deflation is by this standard understated in CPI.
There are several further rounds of objections and responses that could follow, but won't unless someone asks.
