Durable decumulation: improving retirees’ withdrawal symptoms

What, when, and how much? For savers entering income drawdown in retirement, the key questions are not only how they should invest, but also how quickly they should withdraw.

In an earlier article, we considered the 4% rule for withdrawals. We found that it offers simplicity but, unfortunately for investors, life is much more complicated.

So, if the 4% rule isn’t up to scratch, is there a simple approach we can use instead?

Retiring is a risky business

To work out how quickly to withdraw, investors need to understand the risks associated with spending too quickly (and running out of money) or spending too slowly and winding up the richest person in the graveyard. There are various factors at play that could influence how much savers withdraw at a given point in time:

1. Investment strategy. This depends on an investor’s appetite for risk. In our article we assumed a strategy targeting 2.8% over gilts (net of fees) with 8% volatility.

2. Future life expectancy. The longer this is, the smaller the percentage of the remaining funds an investor will be inclined to spend.

3. The level of (real) interest rates. The extent this matters depends on two competing drivers: higher real interest rates make higher withdrawals more affordable but also encourage greater saving. Under plausible assumptions these factors offset in deciding the percentage of the remaining pot to spend each year. This doesn’t mean low real interest rates don’t matter - one expects the pot itself to be smaller in the future - so withdrawals eventually suffer.

4. Defined benefit (DB) income. If there’s a significant source of DB income, such as from an occupational DB scheme, then splurging is less consequential. In this article we ignore DB income to encourage well-paced spending.

The pursuit of welfare

To understand the impact of a change in withdrawal strategy we need to understand how it impacts retirement outcomes. There are essentially two things that matter:

(a) The overall level of retirement income received whilst alive.

(b) How smooth consumption is: one could reduce the risk of dying with unused funds via rapid spending early on, but this would run the risk of living in relative penury once in old age.

Our measure of retirement outcome success, ‘welfare’, takes both of these into account and is defined in the full article. The next step is to find the withdrawal strategy that gives the most attractive distribution of welfare outcomes. This is somewhat technical and there are various features we captured, such as the investment strategy and likely future decisions. Again details can be found in the full article but the results are below:

Intuitively, a strategy that draws down a percentage of the remaining pot, where the percentage depends on age, makes a lot of sense. The withdrawal amount then depends on the current pot size and age of the individual.

As an example of the rule-of-thumb in action, we’d expect a £100,000 pot at age 65 to be worth around £69,000 by age 80. The withdrawal amount for that year is then expected to be around 8.6% of £69,000, i.e. £5,900 for that year. However, given the uncertainties involved, we would only be 80% confident it is in the range of £3,800 to £8,800, as shown in the last column of the table.

Of course caution is advisable – the investor may have other objectives, such as leaving an inheritance, healthcare needs or changing recreational activities. Individual needs and risk appetite matter greatly and may change over time. A one-size-fits-all approach is impossible when investors have specific health issues, tax situations and spending goals. However, we believe that our rule-of-thumb, combined with a suitably diversified strategy, could form a useful starting point for decision making.

Here, we have focused more on how investors should spend rather than on how they might actually spend due to behavioural factors. Stay tuned, though – we plan to explore this in the future.