Mr. Paul Johnson:

I thank the Chairman. First, I apologise for the fact that my wife seems to have taken over the computer and I cannot stop it from saying that I am Nicola Wilberforce. I am, in fact, Paul Johnson. I will start with a few thoughts and reflections on the UK experience, which I think will be of interest and relevance, and I will try to draw out the broad lessons from that. I will talk about pensions, welfare benefits, tax thresholds and briefly about excise duties and a couple of other things.

Pension indexation is extraordinarily important. In the UK we had a long period, between about 1980 and 2010, of indexing the basic state pension just to price inflation. Over that period, the basic pension fell quite significantly as a fraction of average earnings, from something like 25% to not much more than 15%. While it maintained its real price index value, it fell further and further behind the general standard of living and was not offering much in the way of replacement rates. There was a big review of that in the early 2000s and a decision was made to move back to earnings indexation from 2010. It remains in the legislation that the pension should rise year on year in line with earnings. In fact, we have had in place a triple-lock system, which means that each year the pension rises according to earnings, prices, or 2.5%, whichever is the biggest. As a result, it has grown relative to both prices and earnings over that period, as this was an unusual period in which there were a number of years where prices outpaced earnings, as well as years in which earnings outpaced prices and, indeed, some years in which the 2.5% increase was brought into play. That has resulted in a relatively rapid increase in the value of the state pension relative both to prices and earnings over that period.

This is a very strange policy in one sense because if you rationally decided you wanted the pension to be at a particular level, either relative to earnings or prices, this would not be a very rational way of getting there. The defence of it is that it is the most politically acceptable way of getting the state pension to a higher level over a period of time. One of the general messages of indexation is that different ways of indexing things can make a very big difference over time, and it is a difference which often goes somewhat unnoticed politically. It makes a very big difference to future costs. Whether we continue with a triple lock or earnings indexation, let alone price indexation, we are talking about a couple of percentage points of national income within 30 years. This is a substantive choice about the rate at which pensions are indexed. It is a choice of whether to keep it at the same real level or to move it in line with other living standards.

I can speak more about pensions if the committee wishes.

Welfare benefits for those under the pension age have been significantly reformed over the past 30 or 40 years and some of those benefits, particularly for families with children, have been made very much more generous. The indexation rules have essentially been price indexation year on year over the past 40 or 50 years. For example, looking at the standard rate of unemployment benefit, it stands today at essentially the same level in price terms as it did in 1970, which is over 50 years ago. At the same time, of course, real earnings have something like tripled, although not quite tripled. The level of unemployment benefit for people without children is much lower. It is less than half the value relative to average earnings compared with where it was 50 years ago, and few people believe it is at a level where people could nowadays reasonably live on it.

It is worth saying that within recent years, specifically 2015 and 2019, these welfare benefits were frozen so they were not even indexed in line with inflation for a four-year period. That was used as a way of saving significant amounts of money on the welfare budget. The biggest cut to welfare benefits was over what is known as the austerity period in the UK through the 2010s, but it was the least remarked upon and, in a sense, the least unpopular, using indexation as a way of changing policy gradually, and this seeps through rather easily. Freezing things is a rather bad idea, and I will say a bit more about that with tax thresholds, as we do not know ex antewhat the impact will be on the real level of the benefit. For the first couple of years of the freeze, inflation was lower than expected but after that it was higher than expected. Towards the end of the period, there was a bigger real cut than what was intended.

Tax thresholds are, on the whole within the income tax system, by default supposed to rise in line with prices. That was an amendment in a finance Bill in the 1970s that put this in as a default and the Chancellor must announce if the default is to be varied. We have seen very different policies on different tax thresholds, and the personal tax threshold below which a person pays no income tax rose very fast as part of a deliberate policy through the 2010s. It was well ahead of inflation then but it has now been frozen for the next four years, so it is a policy reversal. Again, this freezing of a personal tax threshold is a substantial tax rise over a four-year period. It is a bigger tax rise than was initially intended because inflation has turned out much higher than anticipated.

That is a personal tax threshold but there are other thresholds in the income tax system that really matter. The higher threshold is the point at which a person starts to pay 40% income tax as opposed to the standard 20% rate. Broadly speaking, it again has been subject to all sorts of changes, such as freezes, increases and so on, but it has risen much less quickly over the past 30 years than earnings and slightly less quickly than prices. The result, again not announced as a policy, is that we have something like four times the fraction of taxpayers who are higher rate taxpayers today than we had 30 years ago simply as a result of the way in which that threshold has been increased. It could be increased in line with prices, in which case the natural effect over time is to increase the number of higher rate taxpayers. It might be increased in line with earnings to, broadly speaking, keep the number of higher-rate taxpayers constant.

Some elements of the income tax system have been frozen since they were introduced. In particular, we have thresholds at £100,000 and £150,000, at which higher rates of income tax come into play. We have a 60% rate on incomes between £100,000 and £125,000, and that £100,000 threshold has been unchanged in 12 years. The point at which the 45p rate comes in at £150,000 is also unchanged over 12 years. The result has been something like a doubling of the number of people affected by those thresholds over time. I know this is a deliberate policy to take more money from relatively high-income people without announcing changes year on year. Those are big unannounced changes to the income tax system.

One could speak about other direct taxes but I will say something very briefly about excise duties. Again, one would expect excise duties to rise year on year in line with prices to keep their real value constant. Most excise duties in the UK are set in what I was going to call a random way on a year-by-year basis; the difference is they rise in line with prices. Excise duty on petrol and diesel has not even risen in line with prices for more than a decade now. It is simply being held constant in nominal terms. One of the biggest tax cuts of the past decade has been the effective fall in the real value of petrol and diesel duties because it was considered impossible politically to increase these even in line with prices.

There is a lesson here that has become very salient. If such things are increased once every year in a budget, once the increases stop, it becomes difficult to move back to doing so. One of our suggestions is that it may make sense to have monthly indexation so the duty goes up by a very small amount every month, and this makes the process much more stable.

There are all sorts of other aspects of public policy that are affected by indexation and the way in which regulated prices, for example, are indexed in rail and energy markets in the UK. One bit of public policy that has recently gained much attention is the way in which elements of the student loan and maintenance systems are indexed, so the threshold beyond which graduates start to repay student loans was frozen this year. It had increased significantly three or four years ago and it was frozen before that. These are supposed to be loans and the threshold is supposed to go up with earnings each year. In practice, again, policy is made on the hoof each year in a way that makes it even more clear that this is simply a graduate tax. Again, one might use indexation to change the structure of that system quite significantly. It is worth saying the maximum loan level has only changed once in the past decade, while the real value of money going to universities has fallen over that period. Through lack of indexation, there have also been significant cuts in the level of student maintenance loans.

These are just a series of examples from the UK perspective of different ways in which different parts of the tax and welfare system have experienced indexation over the past 40 or 50 years in some cases. It is really a statement of the very great importance that indexation plays. As a commentator, I significantly dislike the lack of transparency often involved in the way the structure of tax and welfare systems changes because of the way indexation is not used. There are clearly very big and deep questions in many cases as to whether appropriate indexation is in line with prices or earnings. There are quite deep political and philosophical issues to explore there. We should go into those decisions with a very clear view of the impact. Freezing thresholds and so on is something one can do to change structures significantly over time, but one really ought to be very transparent about that as a policy objective.