Judgement Day – RPI Damp Squib

Today has seen the release of the decision by the National Statistician about what to do with the Retail Prices Index. We were told of the consultation in September last year, and were presented with 4 options, ranging from 1) to do nothing, to 4) to make RPI as much like CPI as possible.

Our view was always that the consultation arose as a result of the desire to correct an error made in the clothing component of RPI in January 2010 see blog. This change had seen the ‘wedge’ between RPI and CPI anomalously and erroneously increase by close to 1% following its implementation. We therefore believed that it was perfectly appropriate for the National Statistician to correct this error, and so we were expecting to see Option 2 materialise, which most closely targeted correcting this source of the wedge.

UK linkers had noticeably underperformed other markets since the announcement of the consultation. The market had initially started to price in a 30 to 50 basis point reduction in the wedge of RPI over CPI in expectation of Option 2’s intention to rectify the error.  However, as Judgement Day approached nervousness increased in the linker market as people started to worry that the more severe options could be implemented.

Were Option 4 to have been recommended today, the wedge of RPI over CPI would have been reduced by approximately 100 basis points. This would have been a severe and brutal change for the index linked bond market. All else remaining equal, this change would have seen breakevens on index-linked bonds fall by approximately 70 basis points (allowing for 30 basis points of underperformance already priced in).  To put it another way, this would have see the price of the longest index-linked gilt, the UKTi 0.375% 2062s, fall from 107.7 to about 85, a fall of 21%. Today, things really could have got nasty!

But the decision today has been Option 1. No change. Whilst highlighting that “the RPI does not meet international standards” and recommending that a new index be published, Jil Matheson “also noted 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 in accordance with user expectations”. For the release, go to this link.

All the lobbying that we – and some others – have been doing behind the scenes has been worth it. In the Financial Times today, Chris Giles (who was on the Consumer Prices Advisory Committee) stated that the ONS rejected the committee’s advice in the face of  ‘overwhelming opposition to changes in the calculation of the RPI’.  The market has recently opened, and is removing the expected reduction of 30 basis points or so from Option 2. Breakeven inflation rates at the moment are up by 37 basis points at the 10 year part of the curve and by 22 basis points at the long end. The 2062 index-linked gilt is up by 12 points in price terms, and the whole linker market is rallying in the relief that no change is being made…

…for now! We will soon see the creation of a new RPI index, called RPIJ. This effectively makes RPI equal to CPI through making the older RPI index more modern by removing arithmetic mean and replacing it with geometric mean. This will be run in parallel with the old, untouched index. But it suggests that this debate is not over forever. We could again see recommendations to move from RPI to RPIJ, but more likely, we will soon start to debate moving the index-linked corporate bond market from RPI linkage to CPI linkage.  The creation of RPIJ does seem a little irrelevant, where a new index has been created that few people will care about given that inflation linked bonds will continue to be linked to RPI and the government is clearly dedicated to linking other forms of government compensation to CPI.

Ultimately, though, even if we had seen a brutal reduction in RPI today, I still think that the strong case could be made to want to own UK index-linked bonds over the medium and long term. And changing the calculation to option 4 could have saved the Treasury a whopping £3bn per year, so while the decision to make no change has been great for inflation linked bond holders, it’s not so great for the UK’s coffers.   Finally, the strong opposition to the RPI changes gives you a good idea of how hard it will be to implement austerity measures, and if we aren’t going to get out of this debt crisis through austerity, then the likelihood of us getting out of it with the help of inflation has just increased a bit!

Discuss Article

  1. Justin Pugsley says:

    Looks like they didn’t want to upset investors, not a good idea when the UK is probably set for credit downgrades & still has a very sluggish economy which is a growing worry to the investment community.

    It’s best to do things like this during the good times.

    And yes I agree it does show how difficult it is to implement austerity. Reduce income to any particular group and you get howls of protest and the government gets bad press. I think the most absurd example of this was the ‘pasty tax episode’, the plan of charging VAT on hot food sold in shops & the government backed down after storms of protest. It made them look spineless.

    The government is hostage to a headline driven instant reaction media and well organised pressure groups and better organised they are and the more noise they make, the more effective they are at getting their way.

    I suspect in the end they’ll settle for trying to create inflation and that might come down to the next Labour government. I strongly expect them to win the next election and they are supported by all kinds of constituencies relying on the public sector who will expect to be rewarded.

    Arguing that the current government’s austerity measures have been a disaster Labour may well go for an all out ‘growth’ strategy of spend, borrow and tax. Welcome back Gordon Brown!

    Posted on: 10/01/13 | 11:24 am
  2. Chrisr says:

    Surely new issues will be CPI or RPIJ so the problem is time limited (til 2060 !!! anyway).

    I understand why geometric mean is correct for averaging between years as it deals with the compound interest effect.

    I still don’t get why geometric mean is correct for averaging a basket of stuff. Weighted arithmetic mean would achieve this. There is a flaw in thinking where the focus is on the annual % change (which is supersensitive to small errors in basket cost) rather than the basket cost, then calculating the % change between the years. The former accumulates any errors for each year whereas the latter cancels out the intermediate years.

    I can’t find basket costs for CPI although I can for RPI.

    Posted on: 11/01/13 | 12:42 pm

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.