By this we mean the rate at which an investor can erode their capital (for income in retirement, for example) at a ‘safe’ pace.
That is to say, the funds should not run out before the longevity of the individual does.
In 1994, US financial adviser William P. Bengen first articulated the idea of using 4 per cent of starting capital as the ‘safe withdrawal rate’.
However, a lot has changed since 1994 so we will review to see whether this general rule of thumb still applies in today’s world.
What affects the withdrawal rate?
There are several factors that affect the withdrawal rate that advisers and clients may wish to consider, as follows:
- The age at which someone retires (or starts drawing on their funds)
- Inflation and the subsequent need to increase withdrawals (in notional £ terms) over time
- Volatility of the investments in the portfolio
- Tax – that is, how much of the fund has to be eroded (gross) to give the client their (net) retirement income?
The safe withdrawal rate is also affected by the level of natural income a portfolio produces.
For example, if a portfolio has a dividend yield of 1 per cent and your client requires 4 per cent per year, there is 3 per cent that must be met by capital erosion.
In a rising market this is generally not a problem as the portfolio value is increasing, however in times of falling markets we have a negative compounding effect.
This is sometimes referred to as ‘pound cost ravaging’.
On the other hand, a portfolio with a natural yield of 3 per cent only requires 1 per cent of capital erosion. This reduces the negative compounding effect significantly.
Does the safe withdrawal rate still work?
In his 2015 study, US adviser and blogger Michael Kitces showed that for investors taking 4 per cent withdrawal and adjusted for inflation, most of the time the 4 per cent rule works in that when the retiree has been drawing down on the pension for 30 years, there is still money in the pension pot.
In around two thirds of cases, the client finishes the 30 year retirement time horizon with more than double their starting capital.
In fact, the study shows that less than 10 per cent of the time the retiree finishes with less than the starting principal amount (this is assuming a 60/40 equity:bond portfolio).
Investing for Income versus Total Return
From the above analysis you may be tempted to look at portfolio options that carry an attractive level of natural (dividend) income.
This is a very valid choice for some clients however one must appreciate that investing for income does require some skill to avoid ‘dividend traps’ and unsustainable levels of income.
In times when growth focused investments look attractive, there may be some relative underperformance from an income orientated strategy.
However, the inverse is also true - when economic and profits growth is scarce, a portfolio of income generating assets can outperform the wider market as these investments tend to look attractive for their defensive qualities.
As with many decisions in investing, there is no obvious right or wrong answer to which strategy is best.
However, a measured approach and a good understanding of the merits of both approaches is a very good start for the conversation.
The focus on either total return or investing for natural yield also has an impact on potential sequencing risk, as noted below.
Sequencing Risk
One of the issues with taking the average long term market returns and aiming for withdrawals less than this long term average is that when there are the inevitable negative years, the withdrawals actually compound the erosion of the portfolio.
If these negative years are clustered around the early years of drawing down on a portfolio, this can erode the capital value to such a point that the fund will be exhausted. This is known as ‘Sequencing Risk’.
In a great blog post, the US Financial Advisor and blogger Michael Kitces analysed how the 4% safe withdrawal rate had held up since the tech bubble of 2000 and the financial crisis of 2008. You can see the full blog post here.
The findings are rather startling and should serve as a useful guide to the effect that returns in the early years of retirement can have on the longevity of a retirement portfolio. In a further post, Kitces notes the correlations between the safe withdrawal rate and various timescales.
The key to this is that the first decade of retirement seems to have the most significant effect on whether the conventional 4 per cent safe withdrawal rate is appropriate.
When we think logically about this, it makes sense.
A bad market in the first or second year of a retirement portfolio may not be too damaging as the portfolio is still relatively large for when the market begins to rise again and can therefore benefit from the subsequent rise in asset prices.
Similarly, a bad market when the portfolio has been paying out income for a long time (20 years let’s say) is potentially less damaging because the portfolio has paid out a significant sum of money already and the income time horizon required is reducing.
The real driver of sequencing risk is the real (that is, inflation adjusted) returns achieved over the first decade of retirement.
There is also the question of where the withdrawals are funded from. In good markets it may make sense to reduce the equity content and in bad markets it may be most prudent to realise funds from fixed income or alternative assets.
At this point the relationship and lines of communication between the financial adviser, discretionary fund manager (if used) and client are key.
We find that this discussion can offer real added value to the client and can make a meaningful impact on the longevity of their pension fund.
Conclusion
In conclusion, we have established a few main points for consideration:
- The 4 per cent rule still seems to work when looking at historical scenarios, although it is very dependent on the inputs (time horizon, tax, risk)
- Sequencing risk in the first 10 years of retirement is arguably the most important driver of establishing whether a fund will last for the duration of retirement
- There is no ‘formulaic’ response to managing safe withdrawals and sequencing risk - this should form part of the adviser’s ongoing dialogue with the client and any other providers (that is, discretionary fund manager)
Whilst we are primarily looking at pensions in retirement in this article, the same logic could be applied to other cases.
For example, life interest trusts seeking to produce a return for a life tenant and a degree of long term capital growth for the remainder.
Louis Coke is senior investment manager at Charles Stanley
