Malcolm Kerr: Has Bill Bengen’s ‘4% rule’ been rendered obsolete?


The “four per cent rule” has been around for more than 20 years. Actually, it is a theory, not a rule, but it is highly relevant today nonetheless. The idea was the brainchild of Bill Bengen, a US adviser who in 1994 asked the question: how much of my savings can I spend each year without outliving my money or losing buying power to inflation? He set about running hundreds of scenarios of how asset classes correlated from the mid-1920s to the mid-1990s.

Bengen made some assumptions pretty much in line with many drawdown strategies in the UK today: the money would be in a tax-free fund and last for 30 years, and withdrawals would be linked to inflation (remember that?). His initial conclusion? Four per cent initial retirement income and a 50/50 stocks and bonds portfolio using direct equities and limited mutual funds, rebalanced regularly.

The idea gained swift traction in the US 401K rollover market but it had its critics, not least that the assumptions were too conservative and the model too simplistic. In 2006 he raised the rate to 4.5 per cent. Of course, by 2009 it was considered by many to be too risky (and still too simplistic).

UK advisers now tailor retirement income to individual objectives, risk appetite and so forth, using sophisticated tools and hard-earned expertise. Is 4.5 per cent still a good starting point? Possibly not. The asset growth needed to generate 4.5 per cent income would be very challenging to achieve via what seems to be the current “sinking fund” model. And if the client seeks to maintain capital value it needs some fairly heroic assumptions. The problem is costs.

First, the vast majority of portfolios still comprise actively managed mutual funds and I gather the actual cost of these is more than double the annual management charge when you factor in transaction costs and so on. Say 200 basis points per annum? Second, most advisers use a platform of one sort or another for which the client pays perhaps another 30bps per annum. Third, the average adviser fee is now around 70bps pa and, finally, many advisers outsource to a discretionary fund manager, which can add between 10bps and 100bps pa.

This all adds up to a potential performance drag of over 3 per cent – at least 2 per cent more than Bengen’s scenarios contained. And that is before we factor in the need to hold, say, two or three years’ income in cash to mitigate sequencing risk, as well as facilitate the advice fee.

Income scenarios

There is a further potential problem. Some advisers will be running their lifetime income scenarios on the basis of the average life expectancy of their client. Let’s say age 86 for a male currently age 65. However, over 90 per cent will live longer than this. What is more, for a couple both aged 65 there is a 50 per cent chance of one of them reaching age 90, a 25 per cent chance of one of them reaching 97 and a 17 per cent chance of one of them reaching age 100.

I was reminded of these numbers when I re-read Ned Cazalet’s excellent When I’m Sixty-Four publication from 2014. In it, he suggested: “Providers are busying themselves preparing a tsunami of new accumulation and deculmulation propositions, including offerings without guarantees, as well as contracts in with profit, variable annuity and constant proportion portfolio insurance formats.”

He goes on to say: “It is crucial advisers, providers and regulators move away from naïve and potentially misleading and harmful reliance on simple metrics such as reduction in yield and annuity rates to embrace money-weighted evaluation techniques and the use of stochastic modelling to generate realistic projections of client outcomes that allow for downside as well as upside scenarios to help devise sturdy, enduring client propositions that, in many cases, will be built from a blend of solutions.”

This is exactly what many advisers are doing now. But is it the majority? I moderate many workshops for advisers and so far I have found very few familiar with CPPI, structured products and variable annuities. I am told they are too expensive and too complicated. But there is seldom any precision around the actual cost or how they work.

Returning to costs: could a consumer create a cheaper do-it-yourself retirement strategy and avoid the portfolio drag that impacts Bengen’s rule? The answer is yes.

A consumer could ask a DFM to construct a portfolio of direct investments or passive funds into bonds and equities just like Bengen. The underlying assumptions could be a yield of, say, 2 per cent and capital growth of, say, 3 per cent pa. Inside a Sipp with a fixed fee the annual cost would be around 1 per cent plus transaction charges.

But the non-advised process is almost certain to be sub-optimal because a retirement strategy is not just about income and investments. It is holistic.

It can include income tax, capital gains tax and inheritance tax planning, real estate and residency planning and countless other factors. The range of potential solutions is huge, ranging from offshore discounted gift schemes to equity release, with a myriad of propositions along the way.

Given this complexity it seems likely professional advisers will dominate retirement strategy for wealthier consumers for the foreseeable future. But some of the Financial Advice Market Review proposals may open up opportunities for clients to be more actively involved in the planning process. This may improve their experience, lower the cost of providing the advice and open up even more opportunities for advisers.

Malcolm Kerr is senior adviser at EY