Risk sharing in pensions for USS employers

This paper, by LSE academics Nicholas Barr and Mike Otsuka is welcome at any time, but it is particularly welcome now, with the Work and Pensions Select Committee inquiring into CDC and USS looking for a resolution to what seems an intractable dispute between employers and members of USS.


  • Risk-sharing in pension plans (like risk sharing more generally) improves welfare;
  • Designs for risk-sharing pension plans exist that are compliant with current UK legislation.

A defined-contribution (DC) plan places all risk on individual members, a defined-benefit (DB) plan all risk on the plan sponsor. Each of these designs is what economists call a ‘corner solution’ and, like most corner solutions, generally suboptimal.

As an additional problem, the rules for measuring the health of a DB plan (a) are badly designed and, in consequence, (b) hamper an employer’s ability to offer a good retention package. This note treats those rules as a binding constraint.

The right essay question to ask is: How should risks be shared?

A better design is a plan that shares risk through an explicit set of rules. The example of the plan in New Brunswick. [1] A UK analogy is the proposed WinRS scheme.

The risk-sharing design is based round three central elements:

  1. A division of benefits into base benefits and ancillary benefits;
  2. Risk management through a robust stress test of the plan’s financial health; and
  3. A fully-specified set of rules that establish pre-determined actions in the face of a projected deficit, including changes to contributions and accrual rates and – crucially – a catch-up provision.

The design is very flexible, allowing risk to be shared in different proportions among employers (higher contributions), workers (higher contributions), future pensioners (lower accruals) and current pensioners (less-than-full indexation of benefits in payment). Thus, for example,

  • Less reliance on increased employer contributions in the face of a deficit reduces the risk sponsors face, correspondingly helping to free the plan from unhelpfully onerous funding rules;
  • Greater reliance on lower accrual rates places more risk on future pensioners, though with the possibility of subsequent catch up.[2]

The wider picture

The basic DC design has three strategic deficiencies:

  1. The individual plan member faces all risk;
  2. For that reason a typical plan moves members from equities towards bonds as he/she approaches retirement (often referred to as ‘lifestyle’ de-risking), thus forgoing the higher returns from growth-seeking assets;
  3. Individual plans are administratively costly.


Lowering costs and charges. The DC plan proposed by university employers (UUK) for USS gets round problem (c), with low management and investment charges because of economies of scale and USS’s ability to negotiate low investment fees with fund managers. There is also flexibility over additional voluntary contributions. If an employer were concerned only with (c), there would be no need to move away from UUK’s proposal for USS. But (a) and (b) remain serious problems, which can be addressed via different scheme design such as the following:

Sharing risk among plan members. This approach shares risk among plan members rather than its being borne individually. This type of collective defined-contribution (CDC) plan is DC in that the employer contribution is fixed, but savings are pooled among workers. Such pooling shares investment and longevity risks (addressing (a)) and allows saving to remain invested in growth assets (addressing (b)).

Sharing risk between employers and plan members. A design like New Brunswick shares risk not only among plan members but also between plan members and employers, and thus avoids the corner solutions of individual defined-contribution and defined-benefit plans. WinRS, similarly, shares risk between employers and plan members, and can be implemented within current UK pension regulations. While regulations are not yet in place for a pure version of CDC, a version of WinRS that closely approximates CDC could be adopted today.



[1] See Munnell and Sass (2013); UK Pensions Policy Institute (2014e). Official detail can be found here.

[2] Accruals would need to provide at least the minimum indexation specified by current UK pension legislation.




About henry tapper

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

2 Responses to Risk sharing in pensions for USS employers

  1. Pingback: The case for reimagining the UK's pension models | Great Today News

  2. Asen says:

    Could you provide a link to the paper by Nicholas Barr and Mike Otsuka? I couldn’t find it either here or on the authors’ websites (http://econ.lse.ac.uk/staff/nb/ and http://personal.lse.ac.uk/OTSUKAM/). Also, “risk-sharing” sounds nice, but your post does not explain how it actually benefits people. For example, in the case of DB pensions, there is an element of rolling off risk on the next generation that isn’t there with regular insurance. In the case of regular insurance, I only pay you in a scenario in which I am well off (my house is fine) and you are not (your house burned down). In the case of DB pension plans, there are scenarios in which future generations have to pick up the tab for paying our pensions even if they themselves aren’t particularly well off. In fact, the scenarios in which our investments don’t do well (because stock markets don’t do well) are exactly the scenarios in which future generations are unlikely to be doing much better than us (if stock markets didn’t do well, then it’s likely the economy as a whole didn’t do well either). Also, “risk sharing with employers” sounds nice. But, universities are nonprofit organisations. So, we’re really talking about future employees and students.

Leave a Reply