Plenty of research may suggest 4% a year can be ‘safely’ withdrawn from pension pots over the long term but, writes Paul Tinkler, that comes with a number of caveats – in which case, can stochastic modelling help?
Decumulation, should you not be familiar with it, is all about the level of income that can safely be taken from a fund or portfolio, without eroding the capital over time. There is a lot of research around that would suggest around 4% a year is achievable over a reasonably long period of time – say, 25 years – using a portfolio of 50% equities/50% bonds.
Clearly there is no certainty attached to this level when planning ahead, unless some form of guaranteed product is used – and I am not aware of any products offering that level of guaranteed income with a capital guarantee and ongoing access, as of today.
There are some caveats you need to consider here; First, most of the research takes no account of charges – more of which later. Second, scenarios are highly sensitive to timing so, if analysis started as equity markets entered a bear phase, that portion of the portfolio may have been ravaged by falls in that asset class and will not support 4% income levels over the long term.
On the other hand, if the analysis started at the beginning of a bull run, then the portfolio may have been boosted by early portfolio growth, therefore providing support for long-term income withdrawals.
The final caveat is asset allocation. A 50% equity and 50% bond portfolio – let’s call it ‘balanced’ – approach to the risk strategy for a retirement fund feels appropriate, right? Or maybe arguments could be made for a ‘cautious’ approach of, say, 20% equity and 80% bonds. After all, most retirees are cautious by nature. Still, how many would argue for an ‘adventurous’ 80% equity/20% bond strategy?
Opinions will differ but how can those opinions be tested? We know looking backwards can help but analysis will suffer from certain limitations, as mentioned above.
Think about accumulation planning for a moment. Most advisers use tools that provide a stochastic approach to illustrate expected outcomes for clients, in various market conditions. These typically use Monte Carlo simulations to offer a mean/expected/most probable outcome and then ‘good market’ and ‘poor market’ outcomes, which will broaden or narrow depending on the risk the client is prepared, or needs, to take.
In the decumulation planning world, however, most tools are deterministic, with users setting or adopting default levels of expected growth for a portfolio and then setting the level of income the client requires – for example, a growth rate of 6% a year and an income of 4% a year.
The resulting outcomes show the portfolio continues to grow at a net 2% every year, after income, and the client sees a nice linear graph – but how realistic is that? It would be nice if it was, but it is not.
So can we apply the stochastic approach, used in accumulation, to decumulation planning? Well, that is exactly what we are starting to do at Defaqto and we plan to provide advisers with a tool to stochastically model future income strategies, using ‘probabilities’ to provide clients with illustrations similar to those advisers currently use for accumulation.
Although it is still early days, we are starting to see some interesting outcomes, using the stochastic approach.
So, can a 4% level of income be sustained over time, together with capital preservation? The answer is both ‘yes’ and ‘no’.
We have looked at the three portfolios touched on above – cautious, balanced and adventurous – and, based on an initial value of £100,000, the mean residual values, after 25 years, are approximately £23,000, £83,000 and £140,000 respectively.
In other words, only the adventurous approach provides capital preservation – that is £40,000-plus at the mean. Under poor market conditions – essentially, the 95th percentile of outcomes – however, we believe the original capital could become exhausted after as little as 16 years. So, it is starting to look like 4% income levels may be sustainable – but only using certain asset allocation models.
Now for an elephant in the room – which is frequently ignored in most analysis – and that is fees. Remember, all of the above ignores fees.
Add in fees …
So what happens to the above projections if we add a fee level of 1.5% a year – call it maybe 50 basis points each for advice, platform and fund fees?. Remember we are looking over a 25-year period.
The stochastic modelling suggests a cautious approach would exhaust the fund after 22 years at the mean and 16 years under poor market conditions. The balanced approach meanwhile would have £1,000 left at the mean and would be exhausted after 16 years under poor market conditions. For its part, the adventurous approach would have £36,000 left at the mean, but could be exhausted after 15 years under poor market conditions.
Clearly all of the above are market-based projections, take no account of fund outperformance and do not factor in ongoing financial advice, which may be essential to tailor income levels and shifting asset allocation.
Lots to think about there, then but let me just mention the other elephant in the room, which is taxation. Again, much of the industry analysis ignores this, but the impact of poor tax wrapper usage can have a severe impact all around. The effective use of the pension commencement lump sum, personal allowances, capital gains tax nil rate bands and a wide range of tax wrappers will be essential for certain clients, with the goal of maximising tax-free income.
Clearly financial planning for retirees is not getting easier any time soon.
Paul Tinkler is an insight consultant at Defaqto
The post Paul Tinkler: Is there such a thing as a ‘safe’ withdrawal rate? appeared first on Retirement Planner.