Wednesday, May 28, 2008
Inflation and the CPI
Inflation will still ease in second half: minister
Mon, May 26, 2008
The Straits Times
SURGING inflation will ease in the second half of the year despite spikes in global food and oil prices, said Trade and Industry Minister Lim Hng Kiang...
When I read this two days ago, I had a thought: what the minister is saying really depends on what he means when he refers to "inflation".
Prices for a wide range of goods and services, many of them essential, have been rising at astonishing rates globally. This is well-known, widely reported on and discussed in the media.
What is much less well-known and understood are measures of inflation. Disclaimer: I am not an economist and have never taken a formal class in economics, so read what I write carefully.
Inflation at the retail level is typically measured in most countries by a Consumer Price Index (CPI). Very briefly, the CPI is the weighted average of the price indices for many classes of goods and services. Each individual price index is constructed to reflect the price levels of its components. The components of the indices, and the relative weightages of the indices from which the CPI is calculated, are chosen to reflect the consumption patterns of an average member of society. Information on how the CPI is computed in Singapore is available from the Department of Statistics. From a general eyeballing of a paper from the Department of Statistics, it would appear that the CPI in Singapore is computed largely along international guidelines (which is reassuring), and is free from some of the more glaring issues that the USA CPI is accused of. Of course, it could just mean that the Singapore Government is more adept at concealing any chicanery to do with the CPI. Not that I'm saying there's any chicanery of course.
For media reporting purposes, and in the minds of the average citizen, CPI is inflation. This isn't surprising, as the CPI is judged to be the best and most accurate measure of inflation at the retail level. And its probably what the minister was refering to when he said that inflation would ease. The problem that arises is that most people forget that the CPI is a construct, and that changes in the CPI may not reflect actual inflationary phenomena on the ground, particularly pertaining to individuals.
Do you drive? If you do, with oil at about USD130 a barrel, and climbing, your inflation experience is probably quite different from mine. I am fortunate enough that I live close enough to my workplace to walk to work. Yes, I walk. So the oil price is a non-issue for me. What about people who rely on driving to make a living (taxi drivers, real estate agents)? Their inflation experience is probably considerably worse than yours. The key point here is that the CPI is constructed to reflect the price experience of an average member of society. The problem is that no one is strictly average. It's like the family with 2.5 kids; it doesn't exist. The man-in-the-street is as mythical as the Sphinx, especially in this age of personalization, self-expression, customized solutions and targeted marketing.
The CPI is a measure of inflation, but it should only be used as a yardstick. It is likely that your experience of inflation varies from changes in the CPI. Think about that before you swallow the line that "inflation" should ease. It may not mean very much to you even if it does.
What about the housing component of the CPI? The concept of imputed rent sounds arcane, but it has real consequences in how the CPI is calculated. Suffice it to say, if you own your own home or you don't pay rent (say, because you still live with the 'rents), then a fall in housing prices and by extension rentals means little to you (unless you're shopping for a place). Yet the CPI will still fall to reflect those changes. And hence "inflation" eases a little. Oh look, I think the steam is finally coming out of the Singapore property market...
Then there's that time-honored place where economic sentiment in Singapore is discussed. That's right, the hawker center. Perhaps you've noticed that the portions are getting less generous, or that there's less meat or vegetables with the rice or noodles. But the prices remain the same. Hawkers lack pricing power, so they make it up by reducing their food costs. So, how is that kind of inflation captured in the CPI? "Hedonic regression" the economist answers, but because that's a huge can of worms, the Department of Statistics doesn't do that with the CPI, and rightly so, I agree with that stance. It's one criticism that has been leveled at the USA CPI. But it still points to the CPI in general being an imperfect measure of inflation.
Still think that "inflation should ease" necessarily means clear skies ahead?