The problem with public consultations

The case of the RPI and the Carli Index

One of Britain's two main measures of inflation, the Retail Prices Index (RPI), is statistically flawed.  It uses a method known as the Carli Index, thought to overstate inflation by around one percentage point per year.  As Chris Giles notes in the FT, this method was ditched by Canada in 1978, Australia in 1998, and the US in 1999.

The Office of National Statistics (ONS) has been deciding whether or not to change our index, too. Today, it has announced that there will be no change, despite the widely acknowledged innaccuracy.  Instead, it will produce a more accurate index - the RPIJ - for those of us for whom the purpose of a price index is to tell us what is happening to prices.

The main justification given for their decision was that "there is significant value to users in maintaining the continuity of the existing RPI's long time series without major change, so that it may continue to be used for long-term indexation and for index-linked gilts and bonds".  This followed a public consultation which received 406 replies.  332 of these were against the change, but only 64 objections provided a statistical argument, according to the FT.

Who would gain and who would lose if the RPI was corrected? Aside from the obvious benefit in having accurate information on prices disseminated every month, the main gain would be to the British taxpayer, who would pay out less on index-linked gilts (government debt for which the interest payments depend on the RPI).  The main loser would be investors in that debt: those with large savings, and pensioners.

The latter group is obviously more concentrated than the former.  The costs of the innacurate index are spread across all taxpayers, while the benefits accrue to smaller numbers of people.  Given the time-and-effort cost of participating in a public consultation, it seems obvious that most of the people who bother to reply to such a technical consultation will be those with a vested interest in maintaining things as they are.  Public consultations are of little value in these types of scenarios, and should not be treated as a good way of measuring the costs and benefits to the public.  We don't have the details of the replies yet, but it is probably the case that a mobilised minority has stopped an improvement of policy that would benefit the (justifiably inactive, yet large) majority.

Incidentally, this is another case of the bias towards the old when it comes to public policy and media narratives, which will be the subject of a future blog post.  Further losers from today's announcement are students, whose debt repayments are also linked to RPI.  But how many students, this blogger excepted, are likely to have responded to an obscure consultation on the Carli index?