What’s normal?

workplace pension dispersion

normal’s that line in the middle

When we allow money to be taken out of our salaries , we don’t really know what to expect.

That’s why the FCA want the independent governance committees that oversee workplace pension providers, to tell us what value for money is and whether we got it. The DWP are doing the same thing for workplace pensions run by trustees.

The idea is that we will be able to log onto a website and see if we’re getting what we should from our investments – and the administration and communication of our pensions savings plan. In future we may  be able to see whether we are likely to get value for money from our pensions spending plan – better known as “drawdown”.

So far we haven’t done too well and that’s because analysis of “value for money” has focussed on abstract ideas like investment performance, service level agreements and “member engagement”. All of these things are measurable but they don’t mean much to ordinary people who secretly want to know how they did and whether they got value for their money. Some of the things that IGCs focus on like “engagement” depend on whether their’s anything to engage with.

I can define value for money simply.

I have been thinking about little else than this question. Here is the answer I have come up with. See if you think it holds water.

I start at looking at all the contributions that you’ve made into your pension over the years and compare it with what you’ve got in your pot (known as the net asset value or NAV). I can tell you the interest you’ve got on your money – this is known as the internal rate of return (IRR)

Then I can look at the interest you would have got if you’d invested the same money at the same time, into an average fund.

If you have got more than you’d have got from the average fund, you’ve got value for your money and if you’ve got less, than you have not got value for money.

It really is as simple as that. The AgeWage score I talk about simply tells you how much you’ve beaten or lost to the normal score. The normal score is 50 and you can get up to 100 or as little as 1.

I’m pleased to say that this idea has proved very popular on our crowdfunding platform and we are now going to go into production, starting with a period of testing our normal fund to make sure it really is normal

Screenshot 2019-04-24 at 06.43.22

So what is normal?

Surprisingly, there is very little academic research into how your money has been invested since pension savings plans (DC pensions) started. We are taking the start point of pension savings plans as 1980, that’s because most of the people who started saving before then will now be spending their savings!

Our normal fund has a price at which your money is invested for every day of the last forty years. The price is decided by looking at how a basket of actual funds grew or fell day by day. As time went by, most of our money became invested “passively”. Our normal fund is increasingly priced by looking at how the indices rather than the basket of funds have done.

If we get the price of our normal fund approximately right, we will be call it normal.

How we test whether the prices of our normal fund are right is by using “big data”. What we mean by that is that we are taking tens of thousands of our contribution histories and seeing whether the outcomes (the NAVs) beat the outcomes from the normal fund.

When we get it right , we’ll be able to see the same number of your AgeWage scores beating normal as losing to normal. We’ll even be able to see whether the losses and gains are of equal measure.

When we can show that 50,000 contribution histories cluster around the normal score symmetrically, we’ll have cracked value for money. We will know what normal looks like and be able to tell you whether you’ve done better or worse than normal.


workplace pension dispersion

normal’s that line in the middle

About henry tapper

Founder of the Pension PlayPen,, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in advice gap, pensions and tagged , , , . Bookmark the permalink.

4 Responses to What’s normal?

  1. George Kirrin says:

    Two comments:

    1. You say you can calculate the IRR, but that is a money-weighted rate of return, whereas most fund reporting uses time-weighted rates of return which tend to inflate the averaging impact of earlier years when capital employed is lower and understate the impact of lower returns in later years when capital employed is highest.

    2. Your “normal” distribution assumes the scattered mix is balanced on either side. My own subjective experience is that a minority (20% or less) tend to do well over the longest periods while the rest tend to do worse, but use short term performance periods and new funds to bolster confidence. The shape of such a graph may be more like a right-facing beaver than a Bell curve?

    • henry tapper says:

      On 1, I don’t think that time weighted reporting is helpful to people; money weighted is better because it tells them how they’ve done , not how the fund managers have done.

      On 2. I think your comment begs the question “than what?” . If the benchmark is how other people have done, you should be seeing 50% of people doing better than the other 50%, that’s what I mean by normal

  2. Richard Chilton says:

    The idea seems fine as far as it goes. However, I am worried about the behaviours this may encourage. People will want to see their funds doing better than normal. After all, that is the way that active investment funds have been sold. Pension providers could then be tempted to go for investment strategies that they think will provide immediate benefits to push themselves higher up the league tables. Those strategies may not be sustainable.

    • henry tapper says:

      I agree that if poor scores encourage people to chase returns then they’ll be doing people no favours. When a pension expectation is underfunded then dialling up the risk is not a good idea. I don’t suppose that most pension providers will be looking at AgeWage scores to promote themselves but they can be useful to IGCs in working out where value for money is being achieved. The AgeWage scores are hard to manipulate as they represent someone’s entire contribution history

Leave a Reply