Housing weakens CPI conclusions
Time to look at inflation numbers again, what with the latest consumer price index report showing that deflation accelerated slightly last month.
The twist today is how the housing component affects the overall CPI, given that it is the biggest single force pulling prices down on the index.
It is an unusual component for a price index as it does not in fact measure prices at all. It measures only rents plus a few incidental service charges, and almost all of these rents are private sector rents as public sector rents do not move much and in any case are so ridiculously low that they are given a low weighting.
So if you live in public rental housing or own the flat in which you live, and those two categories represent by far the largest bulk of our population, this CPI housing index has very little to do with your living costs.
Some other countries have tried to refine the housing CPI by taking account of changes in monthly mortgage payments to get a better picture of living costs for home-owners, but we are nowhere near that yet. Few countries take much note of public rental housing as few have anywhere near as much of it as we do.
Your effective inflation rate is therefore more likely to be the one represented in the first chart by the blue line, which takes out the housing CPI and shows a deflation rate of only 2.2 per cent year over year, than the red one, which represents all items in the index and shows a deflation rate of 4.6 per cent.
But now let's take two other things into account. The first is that private sector rents do not change on a yearly basis. Almost all Hong Kong residential lease contracts are written for two-year terms.
If you therefore measure changes on a year-over-year basis what you get is roughly half the rents not moving at all and the other half rising or, in the present circumstances of the property market, falling more steeply than they would if we had one-year lease contracts.
Average the two out and you are probably near the truth, but you always have to remember that you have introduced a possibility of error that you do not have in other CPI items in which you can get an exact year-over-year measure.
And to give you an indication of how great that scope for error may be at present, look at the second chart. The red line measures change in the Government's domestic property rental index over prices the previous year. The blue line (the way rents actually move) does it over prices two years previous.
The other thing to note is that rents for small to medium-sized flats are rising in the New Territories while still falling on Hong Kong Island. Unless the gradual population shift to the New Territories is accurately represented in the CPI, and this is not always easy to do, we could have underestimated average rent increases.
So take your CPI with a pinch of salt. The housing index is uncertain.