A method to chose your workplace pension scheme.


I’d value your feedback on a scoring system we are developing which aims to provide employers with a method of rating one pension proposition against another.

We want it used by employers looking to establish a new workplace scheme, and those who have an existing scheme and are wondering if it needs attention (a second opinion).

Clearly this will need some clarity from Government about what makes for good (the Quality Test) .

Here are three questions to you, trusted readers

1.         Is this a fair method to assess workplace pension schemes?

2.        Can we expect employers to engage in rating a series of propositions like this?

3.         Would providers be comfortable to directly offer pensions to companies chosing in this way?

Answers on a postcard (or better still in “comments”).

Case Study

Here is how an employer rated the Providers at a recent beauty parade we ran.

Provider A

Attribute  Weighting  Out of  Score 
Charge  30  25 
Investment  25  20 
Payroll/HR support  20  10 
At Retirement  12 
Member engagement 
Security of proposition 
Overall  n/a  100  68 

Provider B

Attribute  Weighting  Out of  Score 
Charge  30  20 
Investment  25  15 
Payroll/HR support  20  15 
At Retirement  12 
Member engagement 
Security of proposition 
Overall  n/a  100  66 


As they say in boxing, Provider A won on points, the judge marked it 68/66. This seems a simple and elegant basis for taking and documenting the decision.

Charge -overall impact of charge TER (including an assessment of impact of nominal per capita (NOW) and contribution charge (NEST)

Investment -subjective view of default plus organization of other fund options

Payroll/HR support– mainly AE related but also at implementation and ongoing (Note we assume all providers will have excellent record keeping – this is now a hygiene factor)

At Retirement –treating customers fairly – member support provided to all participants

Member engagement –effort put in to provide staff with comms –including FE

Security of proposition– how likely is the provider to be still in the game in 20 years’ time.

This is a slight development on the thinking on the six DC outcomes established in the Pension Regulator’s paper (Nov 2011).

We’ve moved on from “security of assets” to security “of proposition”, “getting higher contributions” maps onto member engagement while “administration” maps onto “Payroll/HR support”.

The essential difference between this and PQM is that this is about the scheme chosen and not about the sponsor and member covenant (the contribution structure).



This system of rating is based on my personal view of the importance of each subject to my decision making; someone else might place Payroll/HR support at the top and charges less important . Others would argue that Member Engagement should be higher

Any fiduciary should be able to change the weighting order to suit their preferences but the default order should be set by the expert with conviction (in my case First Actuarial).

I don’t think that the attributes and the “out of” scores should be variables, they are hard coded into the process. Bespoking attributes and the scoring system would be an operative disaster and smacks of our having no conviction. It’s doing away with the concept of guidance.


During the process of choosing, a small number of employers will become enthused and want to “go further into it”. This might mean them wanting to go to a consultancy “after all” and pay fees for a second opinion or for detailed help on investments, engagement or on a full on wrap proposition .(for instance).

Exceptional companies should  be given easy links to further assistance, something we think hard about at www.pensionplaypen.com.

Similarly , a system like this must point companies both to mainstream providers but also to  industry specific workplace schemes such as SHPS , the Pension Trust and the Printing Industry scheme . To know what makes for good is one thing but to find and impliment “good”  even more important.

In the months to come, we will build a machine that will help companies work out what makes for good and assess either an existing scheme (or workplace schemes available to them).


The chief output is the overall % as this gives the personal assessment of the person managing the staging process. If a company wants to get this rating done by a number of people (a committee) then we should let them run this a number of times and save each result for them.

What if the employer can’t or won’t score?

This is a big conviction question . Should we have default positions on all providers? . The way I’d like it to work is that we ask the employer to make their own decision on “weightings” and “scores” but give each answer a “can’t choose? ” option which leads them to a default position.

(One snag with default positions is with providers (insurers) with variable responses. Our default view may be  that xyz are generally the cheapest insurer but what if abc comes in with a superquote when they are normally very expensive ?)

If any “expert” can be sophisticated enough to give a bespoke rating on attributes based on the response received – well and good, but I think this is expensive and risky.

Apart from the dangers of assuming a standard charge from those with variable choices, we should not force companies to adopt a single “provider view”

Employers using this methodology should be encouraged to think for themselves and this means either requiring them to fill in certain fields or giving them the strongest of warnings that adopting the default position is not going to be as accurate a way of assessing as personal engagement.

Of course, employer specific scoring is valuable if it can be collected. It provides a “true view” of propositions (eg what the employer thinks), over a large sample of employers. This is about as good a data as you can get as to what employers really think ( a reasonable proxy for members the closer you get to 2018).

If we can collect this data at a provider level the data becomes even more valuable as it informs not just the general debate about what makes for good, but also the internal development of each provider’s proposition.

Further advantage of a self-service technology led approach.

A lot of this decision making will be imperfect. Even with a beauty parade this happens – (I once spoke with an HR manager who had the casting vote on a provider selection and chose xyz on the colour of the presenter’s tie).

We won’t have those distractions and we want to provide information to decision makers which is clear and easy to compare.


This blog is designed to encourage debate and then action. We can argue all day about this methodology but in the end we need to adopt a way of doing things. Basing our assessment on a tweaked version of tPRs six DC outcomes is a smart move as it ties in with Govt thinking but allows an “expert” to remain a thought leader with a value proposition.

The scoring system that leads to a percentage rating seems about right to me- encouraging a range of engagements from an employer (based on interest and competence) with a relatively small degree of bad outcomes (the beauty parade method produces its own).

Of course the providers will be able to see our hand and we will need to be able to justify our ratings not just to employers but to providers (and the regulator) so we are talking “robust”. That said, “robust” is something we do pretty well!

About henry tapper

Founder of the Pension PlayPen,, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in actuaries, auto-enrolment, Change, customer service, dc pensions, First Actuarial, Payroll, pensions, Personal Accounts, social media and tagged , , , , , , , . Bookmark the permalink.

16 Responses to A method to chose your workplace pension scheme.

  1. handstrat says:

    An important element in the choice for the employer should be an assessment of what the service and costs will be after we are no longer a new client with everyone chasing our business. How sustainable is the offer?
    Does anyone else see that some of the current winning propositions will not be able to provide the same service at the contracted price without allowing that scheme to continually increase losses for the provider. It then only needs an FD or shareholder doing their jobs for red ink to start to flow across the pensions division.

  2. henry tapper says:

    You are absolutely right! Which is why it’s not the price- it’s the sustainable price that matters! Dun and Bradstreet can rate thousands of occupational schemes’ credit worthiness. Let’s hope we can find a way to rate a few qualifying workplace pension schemes!

  3. Pingback: Can social media play a part in pension scheme governance? | The Vision of the Pension Plowman

  4. Thomas Schildhammer says:

    On a similar note (maybe that’s something you intend to capture within the “payroll / HR support” category): I’m a strong believer in rating providers also on support of scheme transferability, i.e. out and off to a new provider (“how fast?” “how painful?”) – both for individual members and their employers.

    While I agree there’s a case for making the beauty parade as simple as possible (otherwise it won’t fly), I’d be a bit worried about the respective weights and the impact these have on the actual decision-making. Clearly, even if the weights were “flawed” – say 90% on “charge” and 2% each on all other dimensions – you’d still get what you call an “elegant basis for taking and documenting the decision”.

    The process is repeatable and documented, OK, but does it generate satisfying results on the “outcomes” front? You will remember several participants at the last playpen luncheon who – beyond all technical details and wrapper preferences – were ultimately interested in just that: How can we make sure the member experience (in the default option) is optimal? Not sure the proposed weights satisfy this requirement.

    Having said that, here’s the answers to your questions:
    (1) Something along the proposed lines could be a fair method to assess workplace pension schemes.
    (2) It is absolutely reasonable to expect employers to engage in rating a series of propositions like this. (However, you want to make sure the liabilities flowing from this form of self-administered “guided” choice are transparent and water-tight.)
    (3) As per my point under (2): As long as providers don’t see a larger part of the pile of liabilities shifted their way, they should feel comfortable to directly offer pensions to companies, no matter what the latter’s selection process looked like. Nevertheless – in my opinion – the world of providers will probably be split in half: Those who feel their wealth of “bespeakable” offerings is unfairly collapsed into some unfairly weighted headlines on the one side, and the ones who feel they could gain market share by going along with the process.

    Going back to the all-important “outcomes”: Giving a theoretical 30 of 100 scoring points away for charges is probably a bit much, since most schemes beyond a certain size have mostly identical charges on their default options anyway. Without being too academic about this, risk-adjusted returns at the retirement point is what this should be about. You will probably find quite a few hedge funds who appear to have unmentionable TERs, but if you look at their maximum drawdowns and Sortino ratios vis-a-vis balanced or lifestyled 25bps p.a. pension portfolios you’d be pleasantly surprised (as a member). I’m convinced a large number of aspiring retirees in the US during the past 5 years would have preferred a 2&20 hedge-fund charge for a stable portfolio evolution to the anticlimactic experience of their target-date funds. (Most of these guys are still working and have postponed retirement for the time being.)

    To conclude: If we can make better predictions about the future (look this one up under Niels Bohr) – and scoring a provider on its expected fitness for purpose is nothing else – then we should rank “charges” even higher; although I’d prefer to call them risk-adjusted outcomes (and that includes charge-adjustments).

  5. Jon Dean says:

    Henry, I think you’ve picked a sensible set of ratings categories although we could debate whether the weightings are appropriate. At the very least it looks a good way to short-list providers for due diligence. An end result like 68-66 would be too close to use for a decision, an employer might choose the lower scored provider if they are “nicer people to do business with”.

    My main concern would be the extent to which employers would want to, or be qualified to get involved in the scoring. Charges (if calculated on a TER/bps harmonised basis) are fairly obvious to compare, but will a company that does not employ pensions and benefits specialists be able to form a view about the right investment choices? I would prefer to see a default scoring system quantified as much as possible in terms the employer can understand.

    • Charges – top score for your reference scheme (30bps AMC?) – with points deducted on a sliding scale for more expensive schemes. This will vary over time and be a relative measure.
    • Investments – ratings firms like Morningstar produce objective, risk-adjusted ratings that quantify performance over time. Something like this should give a pointer to which are the best funds for member outcomes.
    • Payroll/HR – a blend of implementation costs plus how many man-hours of effort needed from the employer to run AE processes each pay reference period, a proxy for ongoing costs which will tell them if they need to recruit help. A full-service model will probably score down on charges but well on this measure.
    • At retirement – TCF is a hygiene factor, the rest is more subjective and difficult to quantify, but really more expert opinion than the employer’s domain.
    • Member engagement – metrics could include number of ways the member has of interacting with the provider, availability of contact centre staff to help etc. as well as the qualitative measures e.g. workplace presentations, clarity of documentation.
    • Security of proposition – really into predicting the future here. The life industry has seen massive consolidation over the last 20 years and we can probably expect a lot more. With wide-scale outsourcing, this has to cover the whole supply chain, but can only really be an educated guess. Data needed here include profitability (historic and trend) and solvency ratios – employers aren’t best placed to assess this.

    These metrics could form the basis for your default view and be enough to create the short-list, but clearly would need regular updating. An objective and open approach would tell providers how they needed to improve their propositions to be short-listed and hopefully encourage more competition – to the benefit of employers.

  6. Pingback: Helping companies take decisions on pensions | The Vision of the Pension Plowman

  7. Pingback: An encouraging tale of choice | The Vision of the Pension Plowman

  8. Pingback: Improving governance and best practice – Work and pensions committee | The Vision of the Pension Plowman

  9. Pingback: The “advice gap” can and will be filled | The Vision of the Pension Plowman

  10. Pingback: Bridging that advice gap – 43rd London Pension Lunch | The Vision of the Pension Plowman

  11. Pingback: Get up! Stand up! For your pension rights! | The Vision of the Pension Plowman

  12. Pingback: Do consumers benefit from risk-based pricing? | The Vision of the Pension Plowman

  13. Pingback: Who speaks for workplace pensions? | The Vision of the Pension Plowman

  14. Pingback: A democratic way to improve DC investment. | The Vision of the Pension Plowman

  15. Pingback: Play up Primark! | The Vision of the Pension Plowman

  16. I loved as much as you will receive carried out right here.

    The sketch is attractive, your authored subject matter stylish.
    nonetheless, you command get bought an shakiness over
    that you wish be delivering the following.

    unwell unquestionably come further formerly again as exactly the same nearly very
    often inside case you shield this increase.

Leave a Reply