This may be one of the most obscure technical points I ever mention in this blog. The Office for National Statistics is consulting on a new version of the Consumer Price Index – the document is here. The CPIH will be intended to take housing inflation into account. That does not have any direct implications for policy, but indirectly it seems more than possible that those elements of the CPI which are used to reflect housing issues will ultimately be geared to CPIH rather than the existing CPI. And that matters because benefits, pensions and other issues are being uprated according to the CPI, which does not currently take housing into account, rather than the Retail Price Index, which does. It was recently announced, for example, that the Local Housing Allowance – the local limit on Housing Benefit – will be uprated according to the CPI. There will be a strong argument for moving this to the new CPIH.
The calculation of the new CPIH is not, however, going to reflect what people actually pay for housing. The consultation proposes to take into account the imputed rent of owner-occupiers, measured in an unrepeatably obscure way by drawing parallels with private rental values. This method is called “rental equivalence”. The concept was used long ago for taxation, and it suffered then from an obvious problem – information about rental values is hard to come by, and what we have is dominated by the social rented sector. I can’t tell as things stand whether this will lead to a higher or lower rating for the CPI. However, it does seem undesirable to produce figures that will influence benefit levels by jiggery-pokery, rather than straightforwardly measuring what people actually spend.