Is Inflation, as Measured by CPI, Incoherent?
Monday September 1, 2008
Part of a blog entry by Arnold Kling caught my eye:
A PDF of it is available here: Inequality and Prices: Does China Benefit the Poor in America. The authors give three reasons for why inflation is lower for the poor than the rich:
My instincts tell me that the poor are more likely to change their consumption patterns based on changes in relative prices than the rich are. Could that be what is driving the effect that Broda and Romalis find?
Heather Wilhelm profiles economist Christian Broda, citing a paper that he wrote with John Romalis which:I have argued in the past that there is no "true" level of inflation - at least one that can be measured by CPI. But the quote intrigued me, so I decided to track down the paper...
"shows that from 1994 to 2005...Inflation for the richest 10 percent of U.S. households, which tend to spend more on services, was 6 percent higher than inflation for the poorest 10 percent, which tend to spend more on nondurable goods"
This offset some of the apparent increase in income inequality over that period.
A PDF of it is available here: Inequality and Prices: Does China Benefit the Poor in America. The authors give three reasons for why inflation is lower for the poor than the rich:
The lower inflation at low income levels can be explained by three factors: 1) The poor consume a higher share of non-durable goods —whose prices have fallen relative to services over this period; 2) the prices of the set of non-durable goods consumed by the poor has fallen relative to that of the rich; and 3) a higher proportion of the new goods are purchased by the poor.One of my big complaints about a consumer price measure of inflation is that it goes against all economic logic - specifically that a household's purchasing decisions are unaffected by changes in relative prices. Of course, attempts are made to account for this effect, but there is no one 'objective' way of doing so. Then there is the whole issue of a 'represenative basket of goods' that the above paper considers.
My instincts tell me that the poor are more likely to change their consumption patterns based on changes in relative prices than the rich are. Could that be what is driving the effect that Broda and Romalis find?
Comments
“Incoherent” is a little strong.
The CPI _does_ allow for some consumer substitution, both within and across item categories.
The use of geometric-mean weighting in the lowest levels effectively assumes an elasticity of substitution of -1; when a given cereal lowers its price by 1%, people are assumed to consume 1% more of it.
To get the top-level weights, the ones that compare how much you spend on food versus fuel or books, the BLS uses an extensive survey of consumer spending: the Consumer Expenditure Survey. Because it is a large survey, those weights are only updated every two years, but at the same time, they’re far from unresponsive to long-run substitution.
If you’re interested in an index that uses variation in monthly expenditures, look at the C-CPI-U, the “chained CPI.” Again, it is delayed two years in production, unfortunately, but it makes use of the best information we have on consumer expenditure patterns.