UK Inflation and the indices by which it is measured take centre stage.

Firstly thank you to those who sent me good wishes for my holiday. Lake Como is a lovely part of the world but was extremely hot even for an Englishmen whose climate has this year given him something of an acclimatisation programme. As I did play some golf I did my best to live up to Noel Coward’s lyric only “Mad dogs and Englishmen go out in the midday sun”, however to be fair to them mad dogs were noticeable by their absence! During my time away I notice that the mood in financial markets has become noticeably more optimistic with equity markets undergoing strong rallies with the American S&P 500 index up by more than 3% on Wednesday. One factor which has caught my eye is the fact that the Bank of England maintained its official interest rate at 0.5% for the 16th month and the European Central Bank kept its equivalent at 1% for the 14th month. If you think about it such moves (as in lack of them) is perhaps something of an indictment of the state of economic recovery prospects in Europe and the UK. If you look back to previous recoveries from output setbacks of the size we had in 2008/9 they have usually been much stronger than what we have now. A small increase in the IMF’s forecast for world growth simply does not cut it when looked at in that way and interestingly included a trimming of its UK forecasts.

Changes in the UK inflation index

I wrote an article on this subject on the day of the UK Emergency Budget as I feel very strongly about this issue. I did not realise that I was going to be proved prescient quite this quickly however.

Furthermore there is another corollary to this. These moves tend to spread so that anybody who has a private sector pension that depends on a promised benefit level rather than a fund value is likely to be affected as time goes by from a type of inflation index creep.

Yesterday the Pensions Minister Steve Webb announced plans to extend the (new) use of the Consumer Prices Index to legislation surrounding personal defined benefit pensions in the UK.

“We believe, therefore, it is right to use the same index in determining increases for all occupational pensions and payments made by the Pension Protection Fund and Financial Assistance Scheme.”

In some ways the lowest blow is the impact of the changes on the Pension Protection Fund which helps those who are members of defined benefit schemes which have become insolvent. For the protection they receive was relatively limited anyway. To properly discuss this area I intend to explain the differences between the indices and the implications for the members of such schemes.

The Differences between the Consumer Price Inflation index and the Retail Price Index

The beginning of the CPI as a method of inflation measurement came from the European Union where it was originally called the Harmonised Index of Consumer Prices (HICP). This index was developed to assess whether prospective members of European Monetary Union would pass the required inflation convergence criterion and then of acting as the measure of inflation used by the European Central Bank to assess price stability in the euro area. So its introduction as CPI in the UK was a type of convergence with the European Union.

However,there are significant differences between the CPI and the RPI. The CPI excludes a number of items that are included in RPI-X the previous target inflation measure and these mainly relate to housing and housing costs. These differences include council tax and a range of owner-occupier housing costs such as mortgage interest payments, house depreciation, buildings insurance, as well as estate agents and conveyancing fees. It also excludes Trade union subscriptions and vehicle excise duty. Conversely the things included in CPI but not in RPIX are of more minor influence ( Unit trust and stockbroker fees, University accommodation fees, Foreign students tuition fees and foreign exchange commission for purchases of sterling by overseas visitors).

The two indices cover different population bases as the base for the RPI and its variants excludes the highest income households and pensioners who depend on state benefits. Whereas CPI covers all households including foreign visitors to the UK. There are also some specific price measurement differences for example with second-hand cars.

The last major difference is that individual prices are combined in the two indices in different ways according to different formulae in each expenditure category.The CPI uses the geometric mean, which allows for the substitution of cheaper goods for more expensive goods when relative prices change. The RPI and its variants use arithmetic means which do not allow for substitution.

How much does this matter?

The original documentation for the change was produced in the period of late 2003 and it has an interesting statement on this subject.

Since January 1989, RPIX inflation has exceeded CPI inflation by an average of 0.7 percentage points and, at 1.3 percentage points in October 2003, the difference is currently quite wide.

One might think that such a time “currently quite wide” might be not the time to make such a change but the then Chancellor did not have such doubts. I demonstrated the differences since 2002 in the two inflation indices in my article on Budget Day (22nd June) but today I have added RPIX to the comparison as it excludes mortgage costs. This is because some suggest that for pensioner households this is not appropriate. The annual rates of change from 2002 are. 

CPI: 1.3%;1.4%;1.3%;2.1%;2.3%;2.3%;3.6%;2.2% 


RPIX: 2.2%;2.8%;2.2%;2.3%;2.9%;3.2%;4.3%;2.0%

The differences are plain to see where CPI is consistently lower and if we aggregate my numbers with those from the original research they have been consistently so since at least 1989. If you are not persuaded by these numbers then perhaps a statement by a government agency(its own savings body) may help as on the National Savings website one can find.

The RPI includes housing costs and council tax, and is calculated in a different way, so it tends to be higher than the CPI. NS&I has always used the RPI to calculate increases in the value of Index-linked Savings Certificates, and we continue to do so.

There is a clear implication that it is a better index in that statement.

The impact of this on private sector defined benefit pensions

This is a complicated area and I will try to explain it as simply as possible.There are two main types of pension in the private sector. One which is called defined contribution where you build up a fund value is unaffected by these changes although there may be future implications. The other type is where the member of a scheme is promised a level of benefits in the future hence the name defined benefits and these are also called final salary pensions as the pension relates to ones “final salary”. These will be affected.

The impact comes as follows. The government sets legislation which provides for minimum annual increases when a pension is being paid. These matter because with rising life expectancy pensioners these days can reasonably expect to live for around 20/30 years. Over this period even a low rate of inflation will eat substantially into their pension. The current minimum level of inflation protection is called limited price inflation where benefits accrued before the 6/4/2005 rise at the RPI capped at 5% per annum and those accrued after this date saw the rise capped at 2.5% per annum. Please go back to the figures I have quoted above to see the impact of this as I believe example figures help.

There is a group who are likely to be affected more severely and it is those who leave such a scheme early and they are called early leavers or deferred pensioners. Their benefits are protected by legislation which gives them a minimum level of inflation protection. This is to cover the gap between them leaving their employer and the retirement age to help protect them against the impact of inflation. For them benefits accrued before 6/04/09 are protected by LPI of up to 5% and benefits accrued after this date are protected by LPI of up to 2.5%. So for them a change in the rules has a double whammy effect as the period up to retirement will then be followed by the effect in the previous paragraph. So the joint impact could be somewhat severe in real terms.

I have deliberately excluded some elements in my explanation above as there are differences in the treatment of benefits when they relate to private pension schemes covering what are in effect government set benefits such as a Guaranteed Minimum Pension. They add enormously to the complexity and add little to the explanation so I have glossed over them to try to help readers eyes from glazing over…


There are some clear implications from my analysis.

1. This change in inflation index is a retrograde step which will affect many pensioners in the future as it is being extended from public sector to private sector schemes.

2. It could in the end affect others as the government already intends to extend the use of CPI to such matters as the indexation of tax allowances. This move is even more inappropriate than the move for pensions.

3. Pension legislation during the last government did in effect break the unwritten rule in the UK that governments do not act retrospectively. This government is now continuing and adding to this trend. This is very serious as it brings into question the nature of long-term contracts and this must lead to individuals wondering if pension saving is worthwhile. Changing the rules always leads to fears that they are being changed unfavourably and undermines the credibility of any contract or guarantee. In this instance the change is clearly unfavourable.

Now one needs to be careful with words here as I have seen a lot of misguided writing on this subject as the indices which the government is changing set minimum levels for private sector pensions so it is not forcing them to change as they have the option of exceeding them. However as these schemes in general have large deficits they are increasingly likely to offer only the minimum level of inflation protection. Should this happen then as I have suggested above if you believe RPI will in general exceed CPI, then this change means that this government is effectively proposing to allow pension schemes to reduce the value of members’ accrued benefits and not just future ones. This has very serious implications, as such changes have never been permitted before.

In my articles we have looked at may issues including the nature of wealth and money. Now we come to the value and credibility of long-term contracts. It is not a one way street as in the past governments have had good intentions when proposing legislation for these schemes as the minimum level of inflation protection was well intended. However this new move is going in the other direction and if we find ourselves doing a version of the “hokey-cokey” I hope that questions will be asked about the affordability of some of the changes that have been made in the past. Our political elite finds it easy these days to escape the implications of their actions…

For those who hold index-linked government bonds (gilts) 

If you look at the original documentation for the introduction of CPI as an inflation index you first see an interesting statement.

Pensions benefits and index-linked gilts continue to be calculated on exactly the same basis as in the past with reference to the all-items Retail Prices Index RPI or its derivatives.

So we are now seeing what is an example of inflation index creep some 7 or 8 years later and I would suspect that holders of index-linked gilts will now be reading their contracts. Anybody who has an index-linked annuity will probably be wise to do the same.


15 thoughts on “UK Inflation and the indices by which it is measured take centre stage.

  1. This government wants individuals to take personal responsibility for your retirement,which is fair enough. I thought that being a member of an occupational pension scheme was such ‘responsibility’. However I now find that I have been irresponsible and unfair in using an occupational scheme for my own nefarious ends.

    The signal I am receiving is pretty much the same as that which came form Gordon Brown,if you try to save or work (or both) I will punish you.It seems those who jibed me in the past for wasting my money on a pension have had the last laugh.

  2. Welcome back Shaun and thank you for another excellent post.

    The British government is not the only one that is reducing benefits (accrued or otherwise) by underestimating inflation. I take it that this approach does not break any laws. The government badly needs to reduce future liabilies. What other choices — that will not be challenged in court — do they have?


    • Hi John

      Actually I am surprised that these moves have not so far met more of a challenge. I think that a reason is likely to be that we have two concepts which are not well understood. Inflation and its measures I have discussed many times and will no doubt do so again but if anything the pensions environment is even more complicated and has had even more government meddling. So it may well be that there are not enough people who understand the state of play to make the case…

      In a way it all comes down to long-term contracts and their credibility and with our current political rulers ( of all political parties that I can think of) we have a problem.

  3. Welcome back Shaun.

    A quibble I have is that CPI RPI etc are essentially “cost of living” indices, and yet they also seem to get used as inflation benchmarks too. It is my contention that this is not right.

    There are at least three components I can think of that will change the cost of living from one year to another:
    1 – If the price of one or more items in the basket of goods increases relative to the rest of the economy, the measure will rise.
    2 – If the currency of account is debased, the measure will rise.
    3 – If there are technological or other productivity improvements made (“progress” I would label this), the measure will fall.

    Now while it is arguably right and proper for some sort of “cost of living” index to be used by say welfare authorities to determine the level of benefits provided, I contend that it is wrong for monetary authorities charged with maintaining the value of a currency at a stable level to be misled by prices changes that are not related to their core mission.

    So if CPI or RPI is -2% because the price of oil fell because new cheap sources being discovered or perhaps political risk premiums shrink, this is a boon to society and is not a monetary problem that the Central Bank should react to by debasing the currency by +2% in order to wash out this effect.

    And if CPI or RPI is -2% because productivity improves allowing the same goods to be made with 2% less inputs, that too is a boon to society and not a monetary problem that the CB should react to.

    But if the CPI or RPI is 0%, consisting of -2% from each of the above two categories plus 4% of currency debasement, while we may well have “price stability” I would argue that it would be proper to say that we have a 4% monetary inflation problem and the CB had better be working on fixing this…..?

    Someday perhaps I’ll write a serious solid rant along the lines of “why price levels aren’t (exactly the same thing as) inflation”.

    In the meantime thanks Shaun for this useful detail on pension indexation. I had thought that while governments would be able to (relatively) easily get away with fiddling with the terms of government pension, that they did not have easily available direct control over private pensions. So in a way I was not wrong, but effectively they do have some influence through this mechanism of adjusting the rules on the minimum level of indexation that private pensions are required to provide.

    One final detail I get from Shaun’s posting (if I understand it correctly) that seems particularly pernicious is capping of indexation to 5% or 2.5%. What I ask is the point of having inflation protection at all, if there is to be none provided above a very modest figure?

    While reported inflation north of 2.5% has not been *too* much of a problem so far, I am old enough to remember double digit inflation of the late 70s and early 80s. In fact I remember the “6 and 5” program the Canadian government applied to civil service payrolls, to the great shriek and howls of the public sector unions, which was an attempt (successful) to hold civil service pay rises *down* to 6% one year and 5% the next in order to break the back of inflation.

    • Hi DanielC, of course you are right. In fact if you go back to the classical definition of inflation prior to around 1960 it was not confused with price rises as “price inflation” as it is now. Price rises are not a measure of real inflation, but are a symptom of it. Of course the primary reason for the change was to enable politicians to fool the people and debauch the currency so as to help balance their deficit budgets. They enlisted the help of fraudulent statisticians to help them use smoke and mirrors to fool the people. The earlier exit from the Gold standard was the foundation for the whole rip-off, since whilst fiat money had to be backed by gold there was a severe limit to what debauchment tricks the politicians could play with the currency. Have a look at which has been posted on here previously.

    • Hi Daniel

      Just to comment on limited price indexation and its cap. It is an example of government meddling which was probably well intended but is going wrong. There was a time when there were no statutory minimums for inflation rises to pensions. So the government put in rules but also meddled in other areas which made the schemes more expensive. Now they are facing the problem of being very expensive and one of the easiest ways is to cap the very same inflation protection rules. So we saw a cap in 2005 for inflation increases in payment and one in 2009 for deferred pensions ( in their gap to retirement). Now we see CPI replacing RPI.

      It probably would have been better if they had never interfered in the first place but of course politicians love to do things particularly if it looks like they are free and any complications wont be on their watch…

      To someone like me who is concerned by the UK’s propensity for inflation the very concept of a cap has problems as whilst it has not been a big problem in the last decade or so there are scenarios where it might be in the next decade. However even in sophisticated financial markets it is very difficult to get past the human mind which when faced with something of lowish probability invariably replaces low with zero!

      As to measures of inflation then I agree that there are issues. However before there was the meddliong we have seen then in the UK we had RPI which was known and understood. In that sense it served various purposes and in many ways was one of our better economic measures. Examples of poorer measures are trade figures and retail sales. One of economics essential problems is transferring theory to reality. One of the UK Monetary Policy Committees problems is that it seems to prefer theoretical results to reality… I hope you are better served in Canada.

  4. One important effect of what the government is doing is to alter the balance of incentives for itself and future governments in favour of keeping inflation above 2.5%. Within “government”, I include the discredited MPC. With indexation of government liabilities limited to 2.5%, the MPC/government will want to keep inflation above that for as long as it can. Always of course, promising the little people that it will return to 2% in the medium-term. (If Keynes were alive, he would no doubt say that in the MPC’s version of the medium-term we will all be dead).

    • Hi Ian, well of course Keynes wrote: “”Lenin was certainly right, there is no more positive, or subtle or surer means of destroying the existing basis of society than to debauch the currency. By a continuing process of inflation, governments can confiscate, secretly and unobserved, an important part of the wealth of the citizens. By this method they not only confiscate, but they confiscate arbitrarily, and while the process impoverishes many, it actually enriches some. The process engages all of the hidden forces of economics on the side of destruction, and does it in a manner that not one man in a million can diagnose!” (John Maynard Keynes Economic Consequences of the Peace, Unseen Hand, Page 57)”. So as you point out if he were still alive it is clear what he would opine concerning the MPC’s total failure to do the job they are paid to do!

      The only way to prevent this it seems is to re-establish the Gold standard to back fiat currencies?

  5. Somewhere deep in the Treasury I speculate that there is a small team whose brief it is to find a way to use inflation to ‘help out’ the government with its debt problem, and also by extension, those with personal debts. And they do need helping. The whole structure of our country’s finances is built around the assumption that you must have a ‘modest’ level of inflation in order for the economy to prosper. The team would be working with a higher rate than that. My hypothetical team would have its Excel sheets filled with assumptions about the economy, but the key one would be the inflation rate (measured by whatever criteria they use). Tweak that and you can see debt declining in the future. Or not, if the inflation rate is too low. The key is that incremental inflation is gradual, a slow process not very noticeable on a day to day basis. That’s very convenient for politicians, and makes it the weapon of choice for wealth shifting! So much easier than those obvious tax increases.
    Unfortunately for us, the team’s brief does not include modelling what could happen in the rest of the economy and what the popular reaction would be to yet another round of overspending followed by a chunky inflationary adjustment. You can only do this so many times before the pulbic get wise and start to take evasive action en masse.
    I predict that this will be one time too many, and that people will do what they can to avoid keeping funds in Sterling. That’s what happens, for example, in Argentina, where the middle class save in USD, not Pesos, because the local economy has large inflationary tendencies.

  6. Dealing with pensions all day in my work I tend to ignore it outside. Is there a carry though to the increases funded by the Pension Protection Scheme?

    Don’t forget increases in deferment vary depending upon the years the DB service occurred and when the member left the scheme.

    There has been a two way trend: the withdrawal of Govt subsidy of increases in DB with the ending of GMP and a requirement for DB schemes to apply increases to all of the benefits a member qualifies for.

    The replacement of GMP with the reference test could probably be another article. But a dull one.

    It was probably cheaper to provide these schemes when it was easier for employers to rob them blind. Murdoch being the outstanding example.

    As DB schemes mature and move more to fixed interest securities as part of their portfolio I can’t help thinking that increases will be more costly to provide. Compare funding a level annuity with an increasing annuity from a fixed fund. Check out how much lower the increasing annuity starts at, how long it takes to reach the level annuity’s level and then overtake the total amount paid.

    • Hi Sean
      I tried to dip into pensions knowledge and present the relevant bits. I had noticed that many areas including the BBC were misrepresenting the changes and their impact on private occupational schemes and wanted to present it as it is likely to turn out. Perhaps the likeliest impact is that we wil see a differential between stronger schemes which will exceed legislation and weaker ones which probably cant wait for CPI to come in and buy some cheaper annuities….

      I decided early on that discussing GMP was in a way a red herring as whilst it is important it would add a lot of complexity and as it has its own rules isnt especially relevant here. As to previous pension rules and regimes they add to the complexity without helping with the themes. Within those two sections are plenty of articles! Although perhaps the area I like the least is the use of the AA Corporate Bond rate so I will get around to that at some point.

      I know you have read many of my articles from your comments on here so you are probably aware of what I am likely to think of the trend towards fixed interest securities for DB schemes. They are buying at what I think is the tailend of a 20/25 year bull market for bonds in the UK. It looks ok for now as gilts etc. have continued to rally but if you look at the UK’s past economic history then current yield levels require a large fundamental change to be good value. Or sustained recession. There are many other scenarios where they will be a poor investment.

      As to the Pension Protection Scheme/Fund one could write on it all day. A well intended idea but it is plainly stuggling and of course its funding affects the remaining live schemes… As to the changes then it is my understanding the CPI will apply to escalation in it.

    • …the trend towards fixed interest securities for DB schemes.

      Sure sounds to me like a desire for certainty overriding any thoughts about the performance of the investment.

  7. I’m sure that your comment about this not being understood is correct – it doesn’t help that most politicians and journalists are (like me) neither economists nor mathematicians. You are, I think, not much interested in the public sector, but the implications there are actually much more serious because of the nature of the reward deal: relatively lower salaries have for many years been made up for by more relatively generous pensions, conveniently for politicians because payment is deferred, and also (I would argue) for public sector workers because in accordance with the character of their jobs they tend to be more risk-averse than people in the private sector. Thus pensions – a form of deferred income – are a much bigger component of overall reward, and hence the inhibition of pension uprating is much more damaging.

    In fact most of the media coverage seems to have been founded not only on the assumption that the only difference between the CPI and the RPI is housing costs, but on the even more erroneous assumption that this somehow means that, using the CPI, the housing element of people’s living costs does not get uprated at all (when in fact all that happens is that the basket is differently constituted).

    What I have not been able to understand is why using a geometric mean is thought to allow for substitution, while using an arithmetic mean does not. Can you point me to an explanation of this?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s