PensionsMar 23 2015

Are standard retirement assumptions reasonable?

twitter-iconfacebook-iconlinkedin-iconmail-iconprint-icon
Search supported by
Are standard retirement assumptions reasonable?

The surest way to build a decent retirement pot is to contribute more. There is little doubt this is true. But, unfortunately, it is also extremely off-putting to younger savers.

Take the hypothetical example of a 35-year-old earning £50,000 a year who has yet to start contributing to a pension. Using standard projections, our potential client’s 10 per cent-a-year contribution, growing with inflation, may translate into not much more than £11,000 a year in today’s money as retirement income from age 65 (assuming 5 per cent growth, 2.5 per cent inflation, and annuity rates at the current level of just over 5 per cent).

If he wants to achieve a £20,000 a year income on top of state benefits, he would need to increase his annual contributions to almost £9,000 a year – equivalent to a massive 18 per cent of salary.

Pointlessly unrealistic

Last month, former Labour pensions minister John Hutton called for a national savings target of 15 per cent of salary, in contrast to the current auto-enrolment minimum of 2 per cent, which is set to rise to 8 per cent by 2018. It is not difficult to see how these sort of numbers have been arrived at. But they are pointlessly unrealistic – hardly anyone under the age of 50 would be prepared to take such massive cuts in their income to fund a long-term retirement pot.

While it is true to say that contributions are the only element that savers have real control over – which is why it is contributions that receive the focus of attention – any model predicting 30-year outcomes is prone to error. Just how reasonable are the assumptions being used?

The problem with any long-term financial model is that they are all grounded securely in the present day. And the present day is still suffering from a hangover from the financial crisis, austerity and low growth. Last year, the FCA cut standard retirement projection growth rates from 5 per cent, 7 per cent, and 9 per cent (lower rate, intermediate rate, higher rate) to 2 per cent, 5 per cent and 8 per cent. That puts the lower rate in real terms – allowing for the intermediate inflation rate of 2.5 per cent – at -0.5 per cent.

The Barclays Equity Gilt study, which looks at very long-term data, puts the average return on UK shares from 1960 to 2014 at 5.3 per cent above inflation. That would be approximately 8 per cent assuming 2.5 per cent for inflation – effectively the highest rate allowed by the FCA. Is it reasonable to expect long-term planning to be driven by such conservative measures – or to determine risk levels on the same basis? Given the three growth rate options, clients may be driven to take a very aggressive investment approach in order to give themselves the best chance of achieving the 8 per cent or more that they need. Is that the right outcome?

The rates ratings

The other key assumption that is worth testing is annuity rates – or equivalent income rates. We all know that annuity rates are at all-time lows. The first two months of 2015 saw typical annuity rates fall to 5.14 per cent as over-15-year gilts fell to 1.76 per cent (according to Retirement Intelligence) – both historic lows.

Annuities have become more expensive through a combination of low interest rates (represented by long gilt yields) and increasing life expectancy. So what will they be like in 30 years’ time? Actuaries and other annuity experts often put forward convincing arguments for the view that annuity rates could remain low for many years.

With little sign of rapid economic growth or inflation, the market outlook is that interest rates will rise, but only gradually. It is that view which, effectively, determines the over-15-year gilt yield that is central to annuity pricing. What the market is saying is that short-term Bank of England base rates will rise gradually in the coming years, flattening off after a while. So if you want to lend to the government for 15 or more years, the rate accepted in the market at the end of January was below 2 per cent, taking all this into account.

If Bank of England base rates rise as expected, this may not necessarily push up long-term gilt yields which should have anticipated the hike. But market expectations have been hopelessly erratic in recent years, making long-dated gilts one of the most volatile assets around.

In short, the current market price for long-term gilts is no barometer of future annuity rates. And yet according to the FCA, projections are determined by current market yields.

So, what could be a better way of doing it? There is no particular reason why long-term rates should revert to a long-term average (called ‘mean reversion’) as you might expect share prices to do. But assuming that rates remain at all-time historic lows does not seem fair either. For what it’s worth, the straight monthly average over-15-year gilt yield since 1970 has been a staggering 8 per cent, approximately.

Life expectancy

There is good reason to believe that second element to annuity pricing – life expectancy – may not lead to the downward spiral many might expect. Those familiar with mortality trends will know that we are all living much longer than we used to. The average life expectancy for a 65-year-old man was 77 in 1951, and had increased to 87 by 2014, according to the NAPF and Club Vita. This has put up the cost of annuities significantly in recent years.

But the rate of improvement in life expectancy seems to be slowing. Over four years, broadly between 2004 and 2008, the average increase in life expectancy in Liverpool was 0.5 years, and 1.2 years in Crawley, Surrey. But the most recent data from the Continuous Mortality Investigation (CMI) suggests that expectations for future improvement may have ground to a halt. Various theories have been put forward as to why, including unhealthy diets, lack of exercise and an increase in diabetes.

Other long-term predictions project a further 4 years longer life for a 65-year old in 2056. But there are reasons to believe that, as the rate of change tails off, insurers have now got a handle on mortality which means that the margin between annuity rates and gilt yields may have stabilised, having steadily reduced over the past two decades.

So let us return to our hypothetical client’s predicament. Flexing our assumptions to allow for an 8 per cent growth rate, sticking with the 2.5 per cent inflation, but also making a more generous allowance for a future annuity rate of 8 per cent (based on an 2 per cent margin above 6 per cent gilt yields), our unhappy 35-year old client contributing 10 per cent of his £50,000 a year salary suddenly finds himself with an annual income of nearly £30,000 in today’s terms. Hey presto!

Is the annuity rate assumption too generous? Probably. But by switching into a drawdown projection, our client may only need to achieve 3 per cent a year growth on his savings after retirement in order to pay himself £20,000 a year in today’s money – or 7.5 per cent a year to fund a £30,000-a-year lifestyle – assuming he dies aged 91.

The conclusion can only be that any model is prone to error – and the assumptions make all the difference. While hiking up the contribution rate is the surest way to a bigger retirement income, that does not mean that more affordable contributions will lead to deprivation, as much of the rhetoric from experts suggests.

A 15 per cent contribution rate is simply unaffordable to most; scaremongering about the inadequacy of contribution levels is more likely to put savers off entirely.

Bob Campion is head of institutional business at Charles Stanley Pan Asset Capital Management