UK – Increasing member longevity, the consequences of changing investment decisions and realistic return predictions must form the basis of any defined contribution (DC) pension model, the UK's Pensions Institute has argued.

Academics from London's Cass Business School also said all assumptions underlying pension plans should be "plausible" and internally consistent, and should endeavour to take account of a member's other sources of income upon retirement – such as accrued state pension benefits or property ownership.

According to the Institute's latest paper, 'Good Practice Principles in Modelling
Defined Contribution Pension Plans'
, the proposed plan design would remove the risk of a member outliving his accrued assets.

Lead author Kevin Dowd, professor of economics and finance at Durham Business School, said applying the devised 15 principles could have "uncomfortable implications" for DC members.

"They will often show that they will be making insufficient contributions to their pension plan or are planning to retire too early," he said.

Co-author David Blake, director of the Pensions Institute, added that he thought most DC plans were "currently very badly designed".

"If a DC plan were well designed, it would be a single, integrated financial product that delivers at reasonable cost to the plan member a pension that provides a high degree of retirement income security," he noted.

The paper argued that the incoming Financial Conduct Authority's "deterministic" return projections were "highly problematic – and, indeed, wrong in principle".

"Such projections are highly misleading as they give no indication of the likelihoods or probabilities of achieving such returns," the paper said, referring to the FCA's approach of projecting a product's value based on three fixed return rates.

"Indeed, one can easily get situations in which the probability of achieving an annual return of 8% over an extended investment horizon is essentially zero," it added.

The paper said plan designs needed to address the "quantifiable uncertainty", and suggested that predictions offered to members should state the likelihood of achieving the predicted goal.

It further said that DC plans should take account of each member's occupation and therefore salary, gender – as women currently report lower replacement ratios upon retirement – and any existing assets members possess, such as property.

Additionally, the plan should seek to outline the impact of risk-averse investments on replacement ratio levels, the paper said.

"A good DC model will illustrate the consequences of these decisions in terms of, say, the expected replacement ratio and the 5% [value-at-risk] as a measure of the downside risk," it said.

"Knowledge of these consequences might, in turn, influence other plan decisions that the member makes, such as the contribution rate and the planned age of retirement."

Concerns over increasing longevity were also addressed, with suggestions that increasing longevity should be brought to the attention of DC members, so the increasing expense of annuities can be taken into account when targeting the ultimate size of the DC pot.

The Pensions Institute said it was open to feedback on the paper and would be holding a consultation on the matter ending 31 May.