Decoding Decumulation

Across the world we are looking for THE post-retirement solution for DC savers. Several solutions have been proposed and discussed, but not much progress has been made. Perhaps that is because we are looking at this problem from a provider perspective instead of a member perspective.

Taking the member perspective, we need to decode the decumulation challenge into a set of basic features or trade-offs that actually do matter. Such a framework will help us better understand and categorise post-retirement solutions and provide a foundation for designing a choice architecture that helps members to make informed choices as they approach retirement.

Key decisions and trade-offs

When defining a problem it often helps to go back to first principles to avoid getting lost in rabbit holes of technical details. Approaching retirement, there are three decisions, or trade-offs, that a member faces. Traditionally, these decisions have been made by industry experts, on behalf of members, as part of the product design. The trade-offs illustrate why it is not possible to find a post-retirement solution that fits everyone.

The first, and most important, choice is to decide on the income profile. In other words, how to spend the pension savings. Members could opt for part-time retirement, or choose to take more income in the early years and less later in life. The choice of income profile depends on the size of the state pension and other potential state benefits, such as healthcare and housing allowances.

The second trade-off is between affordability and stability of the income profile. By taking investment risk, the expected income will be higher but the consequence is that payments will be more variable.

A final trade-off is between future adaptability and earning mortality gains. By pooling individual longevity risk, the member can get an affordable life-long income. But this means giving up the adaptability to change the income profile if the member’s circumstances change. Behaviourally, it has proven to be difficult giving up the ability to preserve and pass on retirement savings – experts call this the annuity puzzle.

Decoding current post-retirement solutions

Applying this framework to current popular solutions provides a lens that helps us understand the trade-offs behind the most common retirement solutions.

A DB pension provides a stable inflation linked income for life. The guaranteed inflation linked income made DB pensions financially too expensive for employers which also bears the systemic risk that members, on average, will live longer than expected. The individual longevity risk is mitigated by pooling.

A fixed annuity provides a flat nominal cashflow profile that lasts until the individual passes away. The focus is set on income stability as the fixed amount is guaranteed by the provider and this drives the investment strategy towards safe matching assets. The focus on an income for life means that the individual longevity risk is pooled, while the provider needs to hold reserves for the systemic risk.

A traditional drawdown gives the member control over the cashflow profile and lets the member choose the trade-off between affordability and stability by choosing how the money is invested. By design, the focus has been set on adaptivity, as the option to pool individual longevity is not included. It can be seen as an investment account from which the member withdraws money until the savings pot has been spent.

The 4% rule is an attempt to determine how much money a member can annually withdraw (inflation adjusted) with minimal risk of running-out of money even if living much longer than expected. This is a relatively expensive way to self-insure longevity risk as it prioritises adaptability before earning mortality gains, but it preserves and passes on retirement savings.

Decoding proposed post-retirement solutions

Collective DC (CDC) could be described as an ambition to deliver a fixed inflation linked income for life. To make the income affordable, it takes investment risk with the consequence that the income will fluctuate. To offer a lifelong income, the individual longevity risk is pooled, while the systemic longevity risk remains with the members. The targeted cashflow profile will vary with realised investment returns and mortality gains.

Self-annuitizing Master Trust. By making the same trade-offs as for CDC, it is straight forward to achieve virtually identical outcomes by pooling longevity risk within a section of a master trust. This can be achieved by investing a member’s pot in a unitised investment portfolio and implementing the individual longevity pooling through an overlay. This approach, for example, has been implemented by the Australian Retirement Trust.

Drawdown and deferred annuity. This is a design idea that has been around for years, which combines a drawdown for, say, the first 20 years with a deferred annuity starting to pay out after 20 years. This approach strikes a balance between adaptivity and earning mortality gains. In a recent approach, Moshe Milevsky suggests to use a tontine for replicating the deferred leg of the deferred annuity for the first 20 years. This approach increases the adaptivity as the individual is not forced to buy an annuity when the tontine matures.

A choice architecture for good

An observation is that most of the current and proposed solutions implicitly assume that members target a fixed nominal or real cashflow profile for life. This is based on a common assumption that members have enough savings to consider pooling longevity risk or not. In many cases, the individual may not have sufficient savings to make a meaningful trade-off between adaptability and mortality gains. This leaves those members focusing on the first two trade-offs, in other words, how to use the pension savings as a top-up of the state pension for a given time period.

To address the needs of different types of members we must acknowledge that there are no silver bullets. The problem that we need to solve is: how to design a choice architecture that helps members find a solution, or a combination of solutions, that match their individual needs? I think that the answer is to be found in an accessible choice architecture where members are guided through these different trade-offs in an interactive way. This way we can help members make an informed choice that meets their specific needs.