• There will be a three month wating period after someone starts a job before their employer has to auto-enrol them into a pension.
• Instead of being auto-enrolled as soon as someone’s pay exceeds the bottom of the earnings band on which contributions have to be paid (£5,564), a new auto-enrolment trigger was introduced. This was set at the level at which people start paying income tax (£7,745). In other words as soon as your pay exceeds £7,745 you are auto-enrolled in a pension, and you and your employer both have to pay contributions on your earnings in excess of £5,564.

The government’s consultation on uprating these thresholds has just closed. Our big concern is that the government keeps the link for the auto-enrolment trigger to the personal income tax allowance as the coalition agreement, on Lib Dem urging, says that this should rise to £10,000 a year.

Women would be the main losers from a new higher earnings trigger as the vast majority of workers with pay between the lower limit of the earnings band and the income tax threshold are women working part-time. The auto-enrolment trigger should therefore be frozen.

A £10,000 earnings trigger could eventually stop around two million women from being auto-enrolled into pensions. More than one in seven female workers (15.5 per cent or 1.9 million) currently earn more than the current lower earnings band (£5,564) but under £10,000. One in three female part-time workers (1.7 million) earn between the current lower earnings band and £10,000.

Men are less likely to be affected by this change as just half a million men earn between the current lower earnings band and £10,000. Nearly one in five workers (4.4 million) currently earn less than £10,000, although this includes two million workers who earn less than the current lower earnings band and would never have been auto-enrolled into pensions. A table at the end of the post shows the figures in more detail. (We asked ONS to do a special run of these figures for us from ASHE, but used the NI primary threshold figure rather than the personal tax allowance – but they are close enough to show the same picture).

What we need instead is a bit of fiscal drag. Those who know their economics jargon will know that this is the term used when government freezes tax thresholds. As pay generally goes up, it means that more people pay more tax without any rate actually changing in cash terms.

The government could freeze the lower limits (ie the bottom of the earnings band and the auto-enrolment trigger) but increase the the upper limit of the earnings band to £42,475. This would increase the band of earnings on which contributions have to be paid. This  new upper limit would keep the link with the National Insurance Contributions upper earnings limit, as recommended by Lord Turner’s Pensions Commission.

The government’s decision to further extend the timetable for auto-enrolment will leave those in the first wave starting later this year on a one per cent employer contribution for a full five years. This is disappointing but if the lower band is frozen, then at least there’s a chance that the amount they put into a pension will increase as their pay goes up (if it does!).

If however the bottom of the earnings band goes up in line with average earnings, this is likely to reduce the amount lower-paid workers put in their pension as low pay tends to increase at a rate less than average pay.

This is not a very deep measure of this, but the table shows that cash pay for the biottom decile increased 24.1 per cent over the last ten years, while it went up 29.2 per cent for the median and 33.9 per cent for the top decile.

 Employees earning between lower earnings band (£5,564), national insurance primary threshold (£7,605) and £10,000

Source: Annual Survey of Hours and Earnings 2011

- Figures are all workers, including those outside the age range for auto-enrolment. This is likely to overstate low paid workers.

My strong impression is that this delay was forced on the DWP, and could even be seen as a victory over a strong push for the permanent exemption of small firms from pensions auto-enrolment. But it is still bad news.

Pensions auto-enrolment compels employers to put their staff into a pension scheme. Unless the worker opts out, both employer and worker have to make contributions. These are 4% for the worker and 3% for the employee on a band of earnings, with an extra 1% tax relief from the government. The government is currently consulting on the figures for the band of earnings, but in round numbers it will be about £5,500 to £35,000.

Auto-enrolment is undoubtedly a cost for business as they will for the first time have to make pension contributions. Auto-enrolment has therefore been both staged and phased.

Staging brings different employers in at different times. It starts with the biggest employers later this year, and finishes with the smallest. The announcement today changes the staging timetable.

Big employers (ie those with more than 250 staff) face the same timetable and will all need to auto-enrol by 2014. Medium sized employers  will still start auto-enrolment in 2014, but the last batch will now be delayed until April 2015 – a relatively minor change.

But small employers (ie less than 50) will now not start auto-enrolling until 2015 and finish in 2017. This is a significant delay.

The minister rightly points out that most people work in big or medium firms. This means 70% of workers will be auto-enrolled by 2015. So does this really matter?

Yes it does. This is because changing the staging timetable also delays phasing – and that hits everyone who is auto-enrolled.

Phasing is the second way that auto-enrolment is being eased in. It means employers start off paying just one per cent of the earnings band. In the second phase, they pay 2% and in the final phase the full 3%.

The problem comes with the interaction of staging and phasing. The original plan was for a much shorter staging timetable. But both Labour and now the coalition have extended this. Phasing was designed so that early employers stayed on the first rung for longer and the second and third stages only kicked in once everyone had auto-enrolled for a year.

But the extended staging timetable means that people enrolled this year will now have to wait five full years until 2017 before the two per cent contribution kicks in and the whole system won’t be up and running with the full – but still modest – contributions until 2018. (Don’t forget the effect of the earnings band – the most that anyone will get is 6.8% of their pay, and anyone earning less than the top of the earnings band will get less.)

Delay also has a knock on effect for pension schemes. Most of the cost of running auto-enrolment schemes will depend on the number of members they have. If each member is only making tiny contributions then it will be hard to keep charges down even in those schemes that want to.

These cumulative delays means that it will be 13 full years from the Pensions Commission report that recommended auto-enrolment that it is fully implemented. Ironically the law requires a review of how auto-enrolment is working in 2017, even though it won’t be fully operational until 2018.

No wonder Brendan Barber in the TUC’s response said that this echoes St Augustine: Make me good, but not yet.

Good question I thought. So I’ve knocked up a very quick spreadsheet that worked out whether you would be better off with a pension (or anything else for that matter) uprated by earnings, RPI inflation or CPI inflation.

This is it.

It shows that something uprated by earnings from 1970 would now be 79.8 per cent higher if uoprated by earnings rather than RPI inflation. (The CPI measure only goes back to 1988, so only more recent figures allow a CPI inflation).

What the figures show is that you do better with indexation to earnings starting in all the years shown other than 2005.

It is of course possible that this recent trend will continue into the future, but I think few would agree with that – especially since the government has shifted to using the generally lower CPI inflation measure, which the OBR think will be 1.4 per cent lower than RPI in future.

Of course the shorter the period studied the higher the risk will be that the result will not be typical. Inflation has been atypically high over the last couple of years so while earnings growth has slowed over time, we cannot assume that prices will continue to be high over the decades long period over which pensions accrue.

For those intererested in the technical details, I’ve used the September figure for change over the last twelve months for RPI and CPI. I’ve alse used the September figure for earnings, but used the Average earnings index up to 2000, but the avereage weekly earnings since 2000 as that is the figure the government now uses to measure earnings increases, but as it was not published before 2000 I’ve used the AEI figure.

I’m aware that this is probably not the best way to measure the difference between these indices as it would probably be better to produce an average annual figure for the difference between earnings, CPI and RPI over different periods. Comparing 1970 with 2000 in the chart above is not very revealing as there is 30 years more uprating in the 1970 figure. I suspect someone who knows more stats than me would use a geometric mean for this.

But another way of putting this is that if we compare earnings with CPI inflation, earnings have been higher than CPI for every year bar five since CPI started in 1989. But four of those years were the last four. (The other was 1993.)

Pensions auto-enrolment for the first time will force employers to contribute to the pensions of their staff. This is absolutely right, but unlike some employer complaints it is undoubtedly a burden on business as pensions contributions have a price tag. No doubt for staff higher up the income scale employers will attempt to recoup some of the costs from their wages, but as you cannot cut the minimum wage this is harder for those who on the legal minimum (though there may be some pressure on hours).

But closer examination of the figures suggests that the cost of auto-enrolment, particularly for the next few years, will be tiny for minimum wage workers.

How auto-enrolment works

First, let’s remind ourselves of the detail of auto-enrolment. Auto-enrolment starts in September. Employers have to choose a suitable pension scheme for their staff. They must auto-enrol all staff over 22 who earn above a threshold into this scheme within three months of starting a job. (This House of Commons Library note is a great detailed resource).

Unless the worker opts out, the employer and employee must both pay contributions into the scheme. The minimum is set as a proportion of a band of earnings. The employer pays 3% and the employee pays 4% of this band. (In addition the state contributes one per cent tax relief.)

The earnings band figures were set in 2006 at £5,035 and £33,540, but need to be uprated. The government is currently consulting on what the band should be when auto-enrolment starts later this year.

Originally the lower earnings band was also the threshold for auto-enrolment. As soon as someone’s pay rose above £5,035 contributions would need to be paid on their  wages above this. figure.

The new government set up a review into auto-enrolment. Ministers accepted its recommendation that an auto-enrolment threshold should be set at £7,475.  This means that people will only be auto-enrolled once their pay goes above this (the income tax threshold).

There is still an earnings band, and employers and employees still have to make contributions on pay within this band. But below the earnings threshold workers have to opt-in. Above the threshold they are auto-enrolled and stay as scheme members even if their pay subsequently falls below the earnings threshold.

We were not very keen on this. While it does stop people earning just above the lower earnings band making tiny contributions, it introduces a ‘cliff-edge’. When  someone’s pay rises above the threshold, they become liable for contributions on what is likely to be around £2,100 of earnings once the bands have been uprated.

Staging and phasing

Introduction of auto-enrolment is being both staged and phased.

Staging brings different employers into the regime at different times. The biggest have to start auto-enrolment in October 2012. The smallest do not have to begin until after the 2015 election. (Employer size is measured by PAYE scheme size.) The government delayed the final phase for  small businesses for a year in the Autumn Statement. The precise timetable is about to be published.

Phasing means that employer contributions start at one per cent and are then increased in two phases to 3%.

Employers pay just 1% until September 2016. Employers then pay 2% until September 2017. They only start to pay the maximum in October 2017. While the delay was sold as a help to small firms, it has also delayed the second and third phases for all workers. This means that those auto-enrolled this year will be stuck on 1% minimum contributions for four years (although many employers are intending to be more generous than the minimum.)

Minimum wage workers and auto-enrolment  

There are a number of reasons why the impact on minimum wage workers and their employers is modest:

1. DWP research suggests around one in three of low paid workers will opt-out. The earnings band studies goes above the minimum wage, and it is possible that opt-out rates will be higher for those on the lowest pay.
2. Many minimum wage occupations are high churn. The three month waiting period (introduced by this government) will mean that employers do not have to pay contributions for the first three months of employment.
3. The earnings band is very important in minimising the cost to employers. If the government uprates the lower earnings band by maintaining its link to the Natinal Insurance Lower Earnings Limit it will be set at £5,564.  At the current NMW of £6.08, this means that staff will only pay and get contributions for hours they work above 17.6 a week. Many NMW workers are part-time and may not get above the lower earnings band.
4. The new auto-enrolment threshold means that they will need to work more than 23.6 hours before they are auto-enrolled. 70% of all part-time workers work less than 23 hours according to the ONS ASHE survey. Part-time minimum wage workers may have a different hours profile, but it’s not likely to vary a huge amount from that. (If anyone knows a source for these figures – let me know in the comments.)
5. A full time worker (40 hours a week) on minimum wage earns £12,650 (with some rounding). Their employer will contribute one per cent of the difference between this and the earnings band. Until 2015 this will be one per cent of £7,086 or £71 a year or £1.36 a week. This represents an increase in the wage bill of 0.56%.
6. Only employers employing more than 1,250 staff will start auto-enrolment before October 2013. Those employing fewer than 50 will not have to pay pension contributions until at least 2015.

It makes sense to think of auto-enrolment as the pensions analogue of the minimum wage. But it is not going to do a great deal to help minimum wage staff, particularly part-timers and particularly before the staging and phasing have finished.

This means the burden on minimum wage employers is minimal as much of the earnings of low-paid staff are not eligible for pension contributions. Many fall below the earnings band; many work for small companies that will not face auto-enrolment until after the next election in 2015; AND, contributions on earnings above the earnings band are restricted to 1% until 2016.

There is therefore no case for holding down the minimum wage due to auto-enrolment in discussions about setting the minimum wage for next year.

Indeed I would hope that most people reading this would see it as an argument for increasing pension contributions.

I’ve not published any graphs for some time as public opinion was pretty constant for most of last year. People obsess over short term changes in voting intention, but these are often due to the natural variability in any survey or represent a short-term response to whatever is in the news. The truth is that not much happened on the opinion front for most of last year.

But there are some signs of a slight move. Unfortunately it goes in the wrong direction. But it’s not huge, and the government are still losing important parts of their argument, while still ahead on the need for cuts.People are still directly affected by the cuts, but slightly less than a few months ago.

This charts “Thinking about the way the government is cutting spending to reduce the government’s deficit, do you think this is having an impact on your own life, or not having an impact on your own life?”

It shows the net result – that is those saying “having an impact” less those who say “not having an impact”. (All the following charts plot net measures unless I say otherwise.) It goes back to the general election and includes one 2012 result.

It does not show a huge movement, but is consistent with a trend that we will see more strongly in other charts – a shift away from the government at the end of 2010 with a slight recovery since.

However every result shows a substantial majority who say that the cuts are having an impact on them.

The next chart plots a net fairness measure – those who think the cuts are being done fairly less those who think they are being done unfairly.

This shows a much stronger shift in sentiment away from the government into 2011, and the same small recovery.

This charts whether the cuts are good or bad for the economy.

This chart needs some caution. Many people will agree that the cuts are damaging in the short-term but will still say they are needed for long term economic success. This chart is very similar to the previous one.

This is probably the most significant chart. It is a net measure of whether people think the cuts are necessary. YouGov only started to ask this question at the start of 2011, so we do not know whether this measure showed a shift after the election (as the charts above all do.)

The government have a consistent lead on this. Economic news got worse in the closing months of the year, but this does not seem to have led to people moving away from the government. Rather it looks as if people think that bad economic news makes nasty medicine more necessary.

Looking at the detailed figures, the percentage thinking cuts were necessary has risen from 55 per cent to 60 per cent between last February and today, while those who think they are unnecessary has fallen from 33 per cent to 21 per cent.

Of course thinking some cuts are necessary is not the same as thinking all the cuts are necessary now. The next two charts provide some insight into this.

This chart – which also only dates from the start of 2011 – shows the proportion of the electorate who think the cuts are too deep.

and this one shows those who think the cuts are too quick.

These two graphs are not net measures. As respondents were given three choices, “too quick”, “about right” or “too slow” for the latter one, I’ve only plotted those choosing “too quick” or “too deep”.

While the trend is the same in both graphs, there is slightly more opposition to the speed of the cuts than their depth. Again we see some movement helpful to the government in both graphs.

I have a mental map of attitudes to cuts that tries to make sense of this data. On one side of the argument are those think the cuts are unnecessary and are opposed to them. On the other side are those fully supportive of rapid cuts and probably don’t care very much whether they are fair or not.

In the middle are a range of views including a lot of people who agree that the cuts are necessary but don’t like them very much. They worry that they are unfair and may well think they are too fast and/or too deep, but do not see an alternative. People do not always have consistent attitudes, and – hard as this may be for some I know – often do not think much about such issues.

Polls are simply one bit of evidence that must be considered in thinking through campaign strategies, though an important one. They show you the your starting point and your challenge, but not what you need to do or what you should aim to do.

I’ve also been round long enough to know that most people read polls to look for the evidence that confirms what they already think. So those who think that you can’t fight the worst cuts without supporting others will take comfort from these charts. But so will those who say that what is needed is clear leadership and political campaigning for an alternative.

Actually these charts ‘prove’ neither approach. That is not how anyone should approach this kind of polling.

Argument between those advocating different approaches can be interesting, though also energy absorbing. In practice though the task is to shift opinion away from the government’s current approach. Different tactics, issues and techniques will work with different audiences.Getting more people to think the cuts are too deep and too fast is progress just as much as getting more people to say they are unnecessary.

But whatever strategy people endorse this polling (and others) suggest to me a number of starting points:

1. There is very wide acceptance that the deficit is a big problem. Unless you acknowledge this, people will not listen.
2. It is hard to get people to see the deficit as anything other than a function of how much government spends.
3. The important difference between the cyclical and structural deficits is not widely appreciated.
4. Those (like me) who said opinion would move once people saw the effect of the cuts in their own lives do not have evidence for this in the last twelve months.

And of course good campaigns never accept your opponents’ framing of the issues, and work at the emotional and psychological levels, not through rarefied rationalist argument (though people quickly spot if you can’t win those too).

The increase in the normal retirement age from 60 to 67 for public sector workers has not led to significant savings in the cost of public-sector pensions.

The story is odd in a number of ways. We can wonder why the BBC’s Business Editor is reporting on a non-business story in which he has not been much  involved before – unlike say John Moylan, their industry correspondent who has covered the story in depth. It is also strange that the BBC is covering research that has not – as far as we can tell – been published. It is certainly not on Mr Ralfe’s website.

Mr Ralfe has been a leading critic of public sector pensions for many years. If a journalist wanted an “unsustainable” quote they knew where to go. His consistent argument has been that the discount rate used to value the cost in today’s money of public sector pension payments in the future is wrong. His preferred  figure produces significantly higher costs. They are a great source for those who like big scary numbers to attack future public sector pension costs.

It’s a point of view, but not one that is widely shared beyond the predictable right-wing critics of public sector pensions.

The National Audit Office and the Hutton Report both say that expressing future costs in today’s money is not a helpful way of valuing the cost of pensions.  They say the better measure of measuring future commitments is to look at the share of future GDP they will take. The Government’s Actuary Department predicted this to be steady after the deal negotiated with Alan Johnson, and now says it will fall, mainly because of the impact of the change in indexation from RPI to CPI. This approach is also endorsed by the OBR. Government ministers came unstuck on the Today programme on consecutive days when they tried to argue the “unsustainable” line against this analysis.

There has also been a Treasury led review of the discount rate used in pensions. This resulted in a not insignificant change from a member perspective, but Mr Ralfe was disappointed by the review as it rejected his arguments. 

As Mr Ralfe has not published  his work, we cannot say how much of the argument in Robert Peston’s blog are his and how much Peston’s gloss, but whatever the balance they are misleading and confusing.

They focus on just one change in public sector pensions – and ignore the other two. If Robert Peston had put his claim to the TUC that contribution increases and the switch to CPI indexation:

“were not seen as the major source of acrimony between ministers and trade unions”

we would have laughed (or perhaps more appropriately cried).

Government changes need to be seen as a package. The union complaint is that people are being asked to pay more (ie higher contributions), work longer (ie increase in the normal pension age) and get less (ie the change in indexation to CPI).

In addition there are scheme design issues, such as the move to career average pensions. These flow from the Hutton report. Most of these can be done in ways that are cost neutral. Some may lose and some may gain but the change can be made in ways that do  not change the overall cost of pensions.

Mr Ralfe’s figures do not chime with our analysis, nor with Channel Four’s factcheck that  found againt the government’s claim that pensions would be just as generous under their proposals. But even if you accept them, it is absurd to ignore all the extra income that will be raised not just from significantly higher contributions – Peston is right about this – but also from the extra years of contributions that will be paid.

Nor is it right to say that the government’s proposals are a straightforward increase in pension age from 60 to 65.  Local Government already has a pension age of 65 for all. In other schemes the Johnson changes have meant that new starters have had a pension age of 65 for some years.

But the most misleading argument is the section about accrual rates. As Jon Rogers shows, the statement about existing accrual rates is wrong.

What is worse is that accrual rates in a final salary scheme are compared with those in a career average scheme. A career average scheme needs a significantly higher accrual rate to provide as good a pension as a final salary scheme for the vast majority.

This is because people’s earnings tend to go up over time. If you take a calculation based on a final salary it will therefore be higher than if you use the same factors but start with someone’s (almost inevitably) lower average salary.

It is made even more complicated by how you work out average salary. In any career average calculation you need to uprate past service by inflation to express past earnings in today’s money. This is not straightforward as there are different ways of doing this. The civil service NUVOS scheme, the one existing career average scheme, uses price inflation (which of course has changed from RPI to CPI to the detriment of NUVOS members).

Hutton argued for the increase in average earnings to be used for indexation, rather than prices. As this tends to be higher than prices over time, this produces a higher figure for average salary. To maintain a cost neutral pension package a career average scheme with earnings related indexation would therefore have a somewhat lower accrual rate than one with price indexation.

You do not need to follow all the technicalities of this to see that the generosity of a career average pension depends on three factors: the accrual rate,the normal pension age and the indexation factor, while a final salary pension only depends on the accrual rate and the pension age.

Therefore it is highly misleading to say that:

Before the changes, which were finally agreed before Christmas, public-sector workers accrued pension entitlements at the rate of 1/80 of salary per annum, plus a cash lump sum on retirement of 3/80 of salary, which was equivalent to an accrual rate of 1/70. For teachers the new accrual rate is 1/57 of salary per annum, for healthworkers it is 1/54 and for civil servants it is 1/44 – which is significantly more generous than the old arrangements. (my emphasis.)

Curiously everything else I can find on public sector pensions written by Robert Peston also seems to come from John Ralfe. This is not to say that Mr Ralfe is not a legitimate figure in the pensions debate, but it is odd that such a distinguished journalist  should appear to rely on a single source.

I’ve always liked and admired Robert Peston’s work. I’ve not always agreed, but have always respected him as someone who does his homework and knows his subject. His somewhat stumbling piece on the Today programme this morning did not meet this expectation. ‘More in sorrow than in anger’ is probably a bit of a cliche, but it captures my mood rather well in writing this.

Today workplace pension provision is very different. More people are paying into defined contribution workplace pensions than defined benefit pensions. Some defined contribution schemes are run as occupational schemes with trust based governance, but most are contract-based. These are provided by a commercial pensions company and cannot make short service refunds as this provision is only available to trust-based schemes.

Next year auto-enrolment starts. Over the next five years millions of people will start to save in a pension for the first time. Every employer has to choose a scheme. While there are guidelines and suggestions about how they should do that, the selling of pensions to employers of pensions is unregulated, unlike the sale of pensions to individuals where successive scandals have led to a reasonably tough regulatory regime.

Some pensions companies have suddenly seen the attraction of trust-based DC pensions because they can go to employers and say if you use one you can get your contributions back when members leave if they have been contributing for less than two years. In effect if you are an employer who has lots of workers likely to have short tenures – think retail, hospitality and other low-paid sectors – there is a ‘get out of auto-enrolment contributions free card’ if you choose a trust-based scheme.

Insurance companies can provide what is known as a master-trust arrangement, where a commercial arrangement is given a trust governance wrapper. In the past this has been used for multi-employer schemes. But pensions companies are now seeing them as a way of selling their products to employers, with short-service refunds as a prime selling point.

These not only undermine the auto-enrolment regime, but come pretty close to any common-sense definition of theft. Contributions paid by the employer into a worker’s pension retirement pot should belong to the worker, not be subject to a grab back by their boss if they happen to leave.

This is why Pensions Minister Steve Webb’s announcement today that he will end short-service refunds is very welcome. If there is any complaint it is that the timetable has been too relaxed.

A small pot in a defined benefit scheme does present genuine difficulties. Whether the employer contribution should go back to the employer or the member has always been controversial, but there has been an understanding that looking after small DB commitments was an unreasonable burden for employers.

This cannot be argued in DC schemes. A DC pot is just one lump of savings and investments among others. Small pots are proportionately more expensive to administer than large ones, and are thus not very attractive to providers. Yet they are easy to swap between schemes, in a way that DB contributions are not.

Auto-enrolment will produce many more small-pots, especially if the short-service refund loophole is closed. That is why the minister is right to say that we need to deal with small pots alongside ending short-service refunds. This provides a partial excuse for the timescale, although the TUC suggested that an interim measure would be to make the employer contribution go to the member, thus removing the employer incentive to exploit these refunds.

It makes sense for small pots to be consolidated. It serves neither member nor schemes to have lots of small pots distributed around many pension providers.

We also know that we will need a default mechanism. Of course members should be given a choice about where they move their pot when they change job, but we know that many will not exercise that choice. With pensions companies developing what they call active member discounts and the rest of us should see as deferred member penalties – ie higher charges for members no longer making contributions – it is not a good idea to leave a pot behind with an old scheme when you change job.

The government is now consulting about what should happen to small pots. The starting point for this should be the interests of the member. There has to be a default mechanism that guarantees small pots are transferred into a scheme that has the lowest possible charges and a default investment fund likely to serve the interests of lower income savers with small pots.

One obvious option is a default transfer to NEST, as that meets both these criteria (disclosure: I am a trustee member of NEST). But what is important is that scheme member’s interests are paramount, not those of employers or of pensions providers. Creating a default mechanism for dealing with small pots is in itself a boon for them. The precise details must serve the worker.

One of the government’s attacks on public sector pensions is to change the way that pensions in payment are linked to prices. In the past they have gone up in line with the retail price inflation (RPI) inflation measure. In future they are linking pensions to the consumer price inflation (CPI) measure.

CPI differs from RPI in two ways. First, it excludes a number of items, mostly those linked to housing such as rents, mortgages and council tax. Second, there’s a vital if very technical difference in the way that it is calculated. But you don’t need to understand the nerdy difference between the CPI’s geometric mean and the RPI’s arithmetic mean to understand that it will make the CPI lower.

Historically the difference between the two has been between about 0.7 per cent and 0.9 per cent, but the OBR say

Based on the decomposition of the differences between these measures, further analysis in this paper suggests that a plausible range for the long-run difference between RPI and CPI inflation is around 1.3 to 1.5 percentage points (Table 3.1). For the basis of our November 2011 EFO, we assume that the difference between RPI and CPI inflation is around 1.4 percentage points in the long run.

Over time this means pensions in payment will go up a little less each year.

After 15 years this would lead to a CPI indexed pension being 17.4 per cent lower than an RPI indexed pension.

The table show s the difference over 20 years.

Let’s start with the relatively good news. The state retirement pension will be uprated by £5.30 in line with September’s peak 5.2 per cent CPI inflation. There were rumours that it wouldn’t.

While the triple lock of indexing by the higher of wages, CPI inflation or 2.5 per cent will lead to better pensions than the old RPI link over time, pensioners may feel hard done by this year as RPI inflation in September was 5.6 per cent. The old system would have given them a better rise next year.

Pensions credit which is normally indexed to earnings has also been given a one-off higher CPI linked lift of £5.35. However this is being paid for by scaling back savings credit. This means taking cash from the not very well off to give to the poor. One can see why the DWP would want to do this given Treasury constraints, but it would have been better to find money somewhere else to pay for it.

A further rise in the state pension age seems to be one of George Osborne’s favourite ways of looking responsible by saving money in the future. It will rise to 67 on a phased basis between 2026 and 2028. Bad news if you are under 51.And of course any rise in the state pension age, hits the poor with shorter life expectancies and greater difficulties working longer harder than the better off.

This is not unexpected and I have set out the arguments against here. It will have a particular impact on public sector pensions as the government wants to make the state retirement age the pension age in public schemes.

The delay in the final stages of auto-enrolment for smaller companies is very disappointing. While the strong lobby to completely exclude small firm staff from auto-enrolment has not been succesful, I doubt it will go away.  Ensuring that no small firm is auto-enrolled before the election will encourage a huge lobby to get the Conservatives to commit to this for their manifesto. I suspect the DWP will be rather pleased they have headed this threat off for now, it is still very bad news.

It will not just hit the staff of smaller firms (probably just under half the workforce) but all workers. This is because auto-enrolment is being both staged and phased. Staging brings big employers in first (starting next year) and finishes with the smaller employers. We are talking all employers with fewer than 250 here, I think. Although the details are not yet clear, at present the plan is to treat all employers with fewer than 250 staff as the final batch for auto-enrolment. As there are so many they don’t all have the same date, but are meant to auto-enrol in random groups over two years.

Phasing of higher contributions only starts when staging has been completed. This leaves workers only getting a one per cent contribution from their employer until the final small business has been auto-enrolled. Only then do the contributions start to go up to the three per cent of band earnings ultimately required from the employer. This probably means that the system will only be fuly implemented in 2017.

This was not the original intention. Indeed there is a legal requirement to review various elements iof autro-enrolment in 2017 as the last government assumed it would have been up and running for a few years by then.

Finally comes the dog that did not bark. There were reports that there would be a crack-down on higher rate pensions tax relief – a progressive policy favoured by many Lib Dems – and not just those on the left.

This is the ideal source of funds to improve the state pension and ease Steve Webb’s long held plan for a flat rate state pension.

It has to be said however that if it had been cut today, there was not much chance that it would have gone into better pensions.

But it was the one proposal floated in advance of the Autumn statement that would have asked better off people to pay more in tax.

But to make pension provision in the public sector look like those in the private sector we would need to expel two in three public sector workers from their scheme, as well as giving top public servants much bigger pensions.

But at least something was being done about the appalling state of private sector pensions with the introduction of pensions auto-enrolment next year, which compels employers to contribute to a pension unless the worker opts out.

But there are now reliable reports, particularly this one in yesterday’s Sunday Telegraph, that the Chancellor will either delay or abandon auto-enrolment for small businesses in the Autumn Statement on Tuesday.

(Incidentally this looks like a classic case of leaking a technical story to a journalist that is not an expert in the subject as the report is littered with small errors.)

The switch to CPI indexation for public sector pensions and some benefits is already hitting private sector pensioners too, as many company schemes are now following the government and downgrading the inflation indexing of their pensions. This new proposal represents a direct attack on private sector pensions.

Private sector workers have every right to feel hard done by in comparison to their private sector colleagues when it comes to pensions, but if any were under the illusion that their pension prospects would improve with cuts to public sector pensions, they are going to be disappointed.

Ending auto-enrolment for millions of workers in small businesses would be the end of the pensions consensus established after Lord Turner’s Pensions Commission report. The wide agreement on the basis of a somewhat better state pension, topped up by a workplace pension secured by the nudge technique of auto-enrolment would collapse. Excluding a significant proportion of the workforce would mean that we no longer had a new pensions system.

It also provides a perverse incentive for businesses to remain small. The marginal cost of providing a pension for an extra worker for an employer is three per cent of their salary between about £5,035 and £33,400 (though these figures will be uprated by the time it starts). But the marginal pension cost of taking on an 11th worker (if employers with 10 or fewer staff are excluded) will be the employer contributions on all 11 staff.

No doubt there will be a boon on avoidance schemes such as splitting companies up into small units too.

Staging for employers with fewer than 250 staff  is due to start in March 2014 with the last starting in 2016. That is already a very leisurely timetable, with some people waiting a further five years before they can start to build up a pension.

My best guess is that the policy will be a delay beyond the next election for this final wave of companies in the auto-enrolment, as excluding a million employers from the scheme altogether would be a breach of the coalition agreement. I can’t see Steve Webb being very happy with that – although there is evidence that BIS, where Vince Cable and Ed Davey both have ministerial posts, has been at best an advocate of delay, and at worst an opponent.

But there must be a strong possibility that the Conservatives will go into the next election promising a permanent exception for small business staff from auto-enrolment.

Of course there are many arguments that can be put for a referendum (you can read here for one Labour MP’s take) but the clearest new argument that seems to be coming from the right was put by Bernard Jenkin on the Today programme this morning  . This is that EU rules are preventing us from taking the measures that we need if we want to restore growth.

But there’s a fallacy here.

These are rules that apply across the whole of the EU. That does not stop many EU members having a much stronger economy than that of the UK.

The Eurosceptics must therefore tell us why it is that the UK economy cannot do as well as these other countries under the same rules. They are basically arguing that there is something special about the UK that makes it less able to grow than many other economies with the same basic set of minimum regulations.

Given that there is precious little evidence that so-called “red tape” holds back economies, anyone thinking logically might want to first ask what the UK’s particular problems are and secondly how best to deal with them.

I would suggest starting with asking whether the deepest and quickest spending cuts of any country without a sovereign debt crisis might be just such a problem. Unlike the Eurosceps I rarely think that policy can be reduced to simple yes-no choices and straightforward causations. There are many long-running structural problems in the UK, but the biggest issue is a straightforward lack of demand.

I explain why this is deeply misleading in a post on our new Pensions Justice campaign website.

In our view part of the answer should be establishing new sources of credit for businesses through new institutions and a properly resourced green investment bank.

Britain’s political establishment seem to be embarassed about the state ownership of large parts of the banks that needed to be rescued. The ambition is to privatise them as soon as possible and in the meantime they have been given no instructions by government other than get ready for sell-off and carry on as usual.

But the TUC is not alone in calling for much more radical reform. Samuel Brittan in the FT today (£) develops Adam Posen’s call for the Bank of England to restart QE in a way that boosts the real economy. The headline says it all: “Use the UK’s state bank holdings to speed a recovery.”

And he concludes:

Whether any of this would be necessary in a world with an adult attitude to budget deficits is far from certain. But in the world as it is, it is surely far better than doing nothing or leaving everything to the Bank of England. It is time to remember that banks exist to serve the public and not vice versa.”

Trade unions are suffering from a “fundamental misunderstanding” in arguing that public sector pensions are affordable without reform …

Lord Hutton said the figures “assume we have successfully implemented the reforms” that he recommended. “The fundamental mistake the trades unions are making is that the chart assumes that the reforms have taken place,” he told the Financial Times. “They are the post reform costs. But people are still choosing the facts that most suit them from the report and then torturing the data until it confesses. That chart does not show that public sector pensions are sustainable as they stand. If they were, I would not have made 27 recommendations for fundamental change”.

But this cannot be right, as I will explain.

What is at stake is the use of this graph.

It is important that we understand exactly what it measures. It is the cost of making pension payments (from the pay-as-you-g0 schemes) as a share of GDP (in lay terms GDP is the wealth the nation produces each year).

I am almost certain that it does not take account of contributions – either from employer or employee. Few commentators have grasped this and made much of the commentary elsewhere redundant. I say this because if you include contributions the cost comes down to a much smaller share of GDP. Indeed the net cost of some pension schemes (eg NHS) is negative as contributions are greater than pensions payments.

The pensions paid are assumed to be those agreed through negotiations with the last government, but with CPI indexing for pensions and deferred members’ benefits.

It does not take account of any other changes to pension schemes recommended by Lord Hutton or the government. It cannot have done so because it was included in Lord Hutton’s interim report before he made any recommendations and before the government had announced any further changes. The chart was produced by the Government Actuary’s Department and not his enquiry.

An important part of the agreement with the last government was the introduction of “cap and share”. This was a means of sharing the burden betwen employer and employee if longevity rose faster than expected.

By the time this chart was published GAD would have been able to include some of the costs of changed longevity as it had become clear that in some schemes cap and share would be triggered. This was announced by the previous government.

In addition the chart may have made some allowances for reduced public sector staffing numbers, and will have included the effect of people working longer as the agreement with the last government set new pension ages for new entrants.

As the Hutton report itself says:

Ex.11 This change in the indexation measure may have reduced the value of benefits to scheme members by around 15 per cent on average. When this change is combined with other reforms to date across the major schemes the value to current members of reformed schemes with CPI indexation is, on average, around 25 per cent less than the pre-reform schemes with RPI indexation.

Ex.12 All these past reforms, the current pay freeze and planned workforce reductions will reduce the future cost of pensions. The gross cost of paying unfunded public service pensions is expected to fall from 1.9 per cent of GDP in 2010-11 to 1.4 per cent of GDP by 2060 as the central projection of Chart 1.B shows.

My suspicion has always been that Lord Hutton agreed to do his report before finding out about the shift to CPI indexation. At a stroke that reduced the cost of pensions by 15 per cent as he recognises, and went much further than making pensions affordable as the previous negotiations had done into a pretty deep cut. That has always undermined anything the report had to say about affordability, even if he makes more substantial arguments about scheme design issues such as moving to career average pensions.

Its basic arguments will be familiar to anyone who has read any of the RSA’s work on pensions, so ably led by David Pitt-Watson.

• Defined contribution pensions are not really pensions and get eaten up by high charges.
• There is systemic market failure.
• Behavioural economics shows that people do not respond in the ‘rational’ way that neo-classical economists expect them too.
• There needs to be a bigger role for the state, and tougher regulation of private pension providers.

Even more interestingly David Green’s foreword argues that everyone should be able to transfer personal pensions into NEST, the new low-cost scheme established to make auto-enrolment possible and provide pensions for those on low to medium incomes. (Disclosure – I am a member trustee of NEST). 

At present NEST savers are forbidden from transferring funds in or out of NEST and there is an annual limit on how much anyone can contribute. This was introduced due to pensions industry lobbying, and in the interests of establishing and maintaining a broad consensus around pensions reform (by no means a bad intention – even if this for many was a concession too far).

The best sub-head in the release is undoubtedly “Free markets – great for vegetables, less so for pensions”. In many ways this is a statement of the blindingly obvious.

Free markets work perfectly only if you can have perfect information about every possible product/service you are buying. Of course they can still work pretty well without absolutely perfect information, but you have almost no information at all about a pension as you have to wait decades before you can know whether it has performed well. It’s too late by then to take it back.

But for Civitas to say this is almost shocking.

However there is still room for disagreement about vegetables. Thanks to Stuart White for drawing my attention to the cobweb model  which explains what can go wrong in agricultural markets.

Steve Webb’s dilemma is that he undoubtedly wants people to have decent pensions when they retire, but the Treasury will not give him any extra money. Reducing the number of people eligible for a state pension allows a higher pension to be paid to them at the same cost to the Treasury.

But increasing the state pension age does not extend anyone’s working life. If it meant everyone had a chance to carry on working in a decent job they liked for another year, it would be an easier message to sell.

This however is not going to happen, and the impact will be unfair.

Any increase in the pension age means that a higher proportion of the money spent on pensions goes to those with longer life spans. To take an extreme example an increase from 66 to 67 will halve the amount of pension paid to someone who dies at 68, but reduce it by only five per cent from someone who dies when they are 88.

If life expectancy was entirely random there would be no problem with this. It’s an example of risk sharing (even if living a long time is an odd thing to describe as a risk.)

But longevity is strongly related to social and economic factors – and getting more so. While people are living longer, the gap between life expectancy of those in better off areas and those in deprived areas is getting bigger.

According to the most recent official statistics, in just four years:

The gap between the health areas with the highest and lowest life expectancies at birth increased over the period from 9.8 to 11.3 years for males and from 8.2 to 10.1 years for females. At age 65, the gap increased from 6.7 to 8.5 years for men and from 6.3 to 8.3 years for women.”

Those in decent jobs may be both able and willing to postpone their retirement. Even if they don’t they are much more likely to have the savings or a non-state pension that will allow them to have a reasonable standard of living before they get their state pension. They will probably be better placed to phase their retirement too, perhaps working part-time or in less stressful jobs in the run-up to collecting their gold-watch.

But most people are not so lucky. The worst effect of an increase in the state pension age will be for those who in their late sixties who are not quite old enough to get a pension, but are not working.

Last time I looked at the stats,  more than half of men are economically inactive at 64, the year before they get a state pension now.

Of course some will be perfectly happy early-retirees with proper pension arrangements. But many will be lower-skilled people who find that they can no longer get or do the physically demanding jobs that they used to do.

Many may not be ill or disabled enough to meet a rigorous ATOS test, but are no longer fit or in the best of health. Our benefit regime is far less generous than it has been, and is based on the idea that every claimant should be treated as a work-shy scrounger. However it is highly likely that many of these 64 year olds have held down jobs for most of their lives. That may be even more true for tomorrow’s 66 year olds. An increase in the state pension age will mean that for those years immediately before they retire their income will be very low, and they will be in a tough regime that is looking for ways to reduce it further.

Yet this simple narrative has worked for the government. A majority still believe cuts are necessary (even if they want them to be temporary and reject the small state ideology that drives a good part of the cuts/public service “reform” agenda).

We can summarise ministers’ arguments as:

Labour spent too much and left us with a huge bill called the deficit. The overwhelming priority must be to close that deficit and we will do that in four years. That is why we have to make painful decisions to cut spending and raise VAT and other taxes. If we don’t the economy will collapse as markets lose confidence.

The trick here is ignoring that tax income depends more on the level of economic activity than tax rates. This might be rather obvious as soon as you point it out but the success of the government narrative depends on most people thinking the deficit is defined by public spending and tax rates alone.

Yet policies that depress the economy will make the deficit worse as the tax take falls. Spending money can pay for itself if it results in greater economic activity that raises more in tax – what economists call the multiplier effect.

But the conditions for this sleight of hand to work are disappearing.

• The IMF and NIESR are warning that the government is unlikely to meet its deficit reduction target as the economy is too depressed.
• Increasing numbers of people – including many Conservatives – are calling for tax reductions to stimulate the economy.

But calling for a tax cut goes against the central simple plank that we have to cut spending or raise taxes as we have ‘maxed out the nation’s credit card bill’.

Conservatives make it even more difficult for the government’s narrative by concentrating on the 50p rate for those earning more than £150,000 thus breaching the already very tattered claim that ‘we are all in this together’. Employing an advocate of ending maternity rights as a key adviser hardly helps here either.

Narrative collapse is perhaps the most serious problem for any modern politician.

And Chris Dillow in his own post points out just how cheap it is for the UK government to borrow.  As he says:

The average yield on index-linked gilts with a maturity of over five years is below 0.4%. When borrowing is so cheap, doesn’t it make sense to do more of it?