RPICPI User Group

last person joined: 3 months ago 

To foster co-operation and the exchange of information between the ONS, its advisory bodies and other users. Please see 'Aims of the RPI/CPI User Group' in the library for further details.

A fishy analogy from Simon Briscoe

  • 1.  A fishy analogy from Simon Briscoe

    Posted 18-10-2018 13:38
    Simon Briscoe has asked for his recent blog to be posted here.  It is also available at -


    The king and his fish: the RPI fairytale

    A fishy story about a blind king

    Beware: this story contains upsetting scenes.

    Once upon a time there was a king who fished the fish that were consumed in his kingdom. He lived for his fish. All the people ate his fish. One day the king was eating a piece of cod, the most popular of all the fish, and inadvertently took a mouthful that contained a small bone. It got stuck momentarily in his throat and caused him a bit of discomfort. He felt humiliated at the coughing fit and turning red in front of his people – on account of eating his own fish. He hunted for excuses to cover his embarrassment. His courtiers and servants provided many excuses: one said that no one realised that cod had bones, or that it was made of something bad, another that people can't be expected to know what they put on their fork, and one claimed that cod was just horrible and no one in their right mind could like it. The king was loud and clear in the message he gave as he toured the land. "No one would ever fish for or sell cod again", he said. The people were surprised at his ever more desperate delivery as everyone had happily eaten cod for as long as anyone could recall. The king started a campaign to try to stop everyone eating cod. Strangely, he was happy to eat all other fish despite warnings from the proletariat that they contained bones and if he did not up his vigilance they too could cause distress. There were worries that the king was going mad. Finally he threatened to ban cod from sale and force the peasants to eat coley. In village halls, factories and inns revolution was stirring: people loved eating their cod and were very upset at being told that the age-old tradition was to be banned. Dissenters protested that there was no good reason to ban cod. The problem is not the filet itself. And while cod bones are an inconvenience and can be dangerous, they said, bones are a problem with other fish and those fishing, cooking and eating any fish just have to take care. But no, the king carried on and tried to ban cod, the nation's favourite fish. The king persuaded his knights and princes (for it is mainly men who know about cod) to travel the land with the message about cod. Anyone who said they wanted cod not coley was sniggered at and told that they had no right to prefer something that tasted nicer – and was so dangerous. The proclamations from the state's agents become more vehement and the discontent among the people grew. Eventually a good and lawful man at the moot court said "wait a minute, lets look at this". Then the Lords began to look at, what had by then become, the cod wars and …………..

    to be continued ……….

    ………… Will the Lords convince the king that his overreaction can be forgiven and that cod may be eaten, and the people need not be unreasonably deprived for the rest of their lives?

    (Footnote: A couple of readers have suggested that I have been unfair to coley. They wondered why I'd not found a way to squeeze in a carp and a grunter or any crappies, pollocks or cobblers. With thanks.) 


    Like many fairy tales the truth behind this one is far worse than the sanitised version above.

    The Office for National Statistics produces the nation's key figures. That's all it does and the numbers are widely used. Mostly this is not controversial but in 2010 there was a major cock-up when some untested new raw data were carelessly and unprofessionally used to calculate the inflation figures. This error caused the gap between the CPI and RPI to rise by ½ a percentage point. A small figure to many but with billions riding on the RPI it matters to many people's finances.

    It caused much consternation and embarrassment in the ONS. Civil servants "changed" jobs, new systems were brought in and blind eyes were turned but no one bothered to do the simple thing: correct, or even improve a bit, the statistics by using more sensibly chosen raw data. After a period of denial, the ONS realised that the problem was not going to go away and, as criticism mounted, various experts (some of questionable independence) were called in. They wrote reports (of varying quality) explaining why everything would be fine if the RPI was abolished. (Though no one had any idea how to abolish the nation's most widely used statistic without getting the government bogged down in decades of legal battles.)

    Meanwhile the government decided to take advantage of the gap between the two measures and changed the link used to uprate the payments made to people away from the RPI to the lower rate, CPI. This robbed millions of benefit recipients, workers and pensioners of their legitimate and long-standing promised rise in income. The ONS, the Bank of England and Treasury, who have regular secret meetings about the RPI, produced many reasons why the RPI was suddenly, despite its long history and widespread use, complete rubbish. This is despite the fact that all that had changed was the collection method for raw prices for some clothing items – perhaps just 3-4% of the basket of goods in the RPI.

    Ironically, the change had to be made following EU pressure to correct obvious stupidities that the old price collection caused in the CPI. The ONS replaced something that was failing with something that also failed. As arguments used to rubbish the RPI were undermined or dismissed, others, Hydra-like, were invented.

    As the likelihood rose of the wider gap remaining there was an even greater incentive for government to switch between measures. Inspired by the deception of the Treasury, the private sector – notably pension providers – followed suit, choosing to improve corporate P&Ls while hitting the spending power of those dependent on such incomes.

    UKSA, the supposedly independent regulator, then declared it wanted to "abolish" the RPI. This view was adopted despite producing no report itself. The basic truths – that the formulae used have been around and unchanged for centuries, that index creation is an art requiring common sense to overcome minor issues that arise over time, and that bad raw data has and will continue to mess up any statistical index if not chosen with care – were ignored.

    Groups of pensioners, fearful of further cuts to their income, rose up and brought the case to the courts in 2017, and the judges ruled in their favour. The House of Commons in the form of the Treasury Committee chose not to investigate how the Treasury was using statistical errors and civil service professional weakness to hit personal incomes in the name of austerity. At least the House of Lords, with its ability to smell a rat (and perhaps a keener sense of justice than the Treasury) now has a committee looking into it. The incomes of millions will ultimately depend on the conclusions of their report and that is where the nightmare stops. For now.

    Many hope that the rigid, blinkered bluster of the establishment will be seen through and seen off. Innocent if stupid statistical errors should not be used to cut personal incomes. ONS must improve the data. UKSA should not encourage practices that undermine trust in data. The RPI's problems are not fundamental: all its key elements were endorsed over decades by independent committees reporting to parliament. The only statistical lesson from this episode is one everyone knew already: if you put rubbish raw data in, you'll get a less good statistic out. This was the case with the clothing component of CPI pre-2010 and is the case for the clothing component of RPI now.

    Various components of both indices need urgent review to see what is lurking within. A refresh of data and methods needs to be made regularly in a transparent and open way as they were before the politicisation of inflation became the norm under Gordon Brown's chancellorship. Trying to abolish an historic popular statistic because some sloppy statisticians put some silly numbers into the formula is not a sufficient reason to edge more people towards poverty and launch decades of legal battles.

    The dozens of men – and it is mostly men – in UKSA and its official allies who have studied the price of strappy tops know in their hearts that the answer is far simpler than the official line. All that's required are some minor and easy changes to the RPI's data collection – starting with better data for strappy tops. If only the ONS was strong enough to do this, supported by a rigorous and independent UKSA, everyone could live happily ever after with the much touted three measures of inflation. We'll discover soon what the House Of Lords thinks. In the worst case, we could witness the greatest abandonment of principles of professionalism and independence of the UK's official statistics since the CSO was created in the 1940s.

    If anyone knows of any other case that has hit the public's trust through their pockets to the same extent, please let Simon know.

    Tony Cox
    chair, RPI CPI User Group

  • 2.  RE: A fishy analogy from Simon Briscoe

    Posted 19-10-2018 09:43
    This helpfully Semitic explanation of what has happened looks right and I hope Paul Johnson, who was frankly duped by those advising him, reads it.

  • 3.  RE: A fishy analogy from Simon Briscoe

    Posted 20-10-2018 09:08
    My phone's predictive text appears to have changed demotic to Semitic. Apologies.

  • 4.  RE: A fishy analogy from Simon Briscoe

    Posted 13-11-2018 17:36
    Hi All

    I wanted to use today's labour market statistics release to show one important economic impact of which inflation measure is used.

    The official line is shown below.

    "Latest estimates show that average weekly earnings for employees in Great Britain in real terms (that is, adjusted for price inflation) increased by 0.9% excluding bonuses, and by 0.8% including bonuses, compared with a year earlier."

    The problem with that calculation is that it requires the use of the imputed rent driven CPIH to get there as the imputed rents drag inflation lower and thus inflate real wages. If we take the imputed rents out and go back to CPI then we need to take a bit more than 0.2% off the real wages figure. So it reduces to ~0.7% excluding bonuses and to ~0.6% including them.

    However the nub of the issue arrives if we use RPI because we know that in annual terms it has been consistently above 3%. Thus any real wage growth disappears and may even be negative.

    Now if you were the government which number would you want the population to use?


    Shaun Richards