Money: Check those savings calculations

By BRENT SHEATHER and PHILIP MACALISTER

Working out how much money you need to save for your retirement is a problem that falls into one of two categories: too hard or too scary.

For the feckless amongst us, the internet has been a real boon because so many websites have retirement calculators that help us to answer the question.

You go to a website, punch in your numbers and, hey presto, the black box tells you how much to save each week or month to reach your retirement goals.

The trouble is, not all retirement calculators are created equal and some can be quite misleading.

If you use a retirement calculator you must ensure that the critical assumptions it uses are realistic.

Unfortunately, many of the calculators get it wrong because their assumptions on key variables such as asset allocation profile, returns from each asset class, fees and taxation, are too optimistic.

If, for example, returns are overstated or annual fees omitted from the terminal sum calculation, this could see individuals underfunding their retirement plan, with disastrous consequences in the long term.

Websites offering retirement calculators should disclose in detail the key assumptions inside their black box.

Four retirement calculators road-tested for this article showed the wide range of answers that can be produced.

In this test we used the profile of a 30-year old doctor, with no retirement savings, who planned to save $350 a month until age 65.

The calculators used were on the Office of the Retirement Commissioner's website (www.retirement.govt.nz) and on three financial planners' websites.

The result was a 108 per cent difference in the final sums.

According to the Retirement Commissioner, our doctor would have a terminal sum of $176,485, while the most optimistic financial planner put the sum $109,000 higher.

One of the sites offered a more interactive calculator and allowed our doctor to change some of the assumptions.

The main reasons for the wide difference relate to the assumed returns.

These ranged from the 2.5 per cent annually used by the Retirement Commissioner to a high of 6 per cent used by a financial planner.

None of the websites specifically considered asset allocation (which research says determines 90 per cent of returns). It wasn't clear that annual fees must also be deducted.

Also somewhat misleadingly, one website highlighted pre-tax, pre-fee, real returns for US cash, bonds and shares over the last 75 years.

In contrast to that, another site used a post-tax, post-fee, real return of 4 per cent annually for a balanced growth fund, but did not state that the return was after inflation.

It merely labelled the return as after fees and tax. Subsequent discussions with the company revealed that the returns were after inflation.

Because it assumed that the doctor would increase his contributions at the rate of inflation, it calculated a higher terminal sum than other sites, despite using an assumed return rate that was 2 per cent less than some other sites.

This treatment of inflation seems inconsistent with the fact that the return assumption is in real terms. All the data should be in either real or nominal terms, not a mix of both.

To get an idea as to what rate of return is likely over the next 35 years, people naturally refer to the past.

The problem is that some refer to the recent past of, say, the last 10 years, while some refer to the returns averaged over the last 80 years.

This wouldn't be too much of a problem except for the fact that over the last 10 years shares have given fantastic returns - 18.4 per cent annually, versus 11.2 per cent annually since 1925.

Unfortunately, it is likely that neither of these approaches are appropriate for an individual with a 35-year time horizon.

Over the long term and particularly over the last 10 years, a large component of the returns from shares has been due to the willingness of investors to pay higher relative prices for shares, based on the earnings of the underlying companies.

The price-earnings ratio of the US stock market has doubled since 1980 and, at 30 times earnings currently, it is almost three times the level prevailing in 1926.

Similarly, because interest rates have fallen over the last 10 years, bond returns have been artificially bolstered.

In a recent report, the United Kingdom Institute of Actuaries, whose members spend most of their day forecasting long-term returns for pension funds, predicted that returns from shares over the long term [from today] will average 7 per cent annually.

The projected growth for bonds is 5 per cent annually.

This view is supported by leading New Zealand fund managers such as AMP Asset Management and Tower Asset Management.

AMP Asset Management's head of investment strategy, Paul Dyer, estimates the long-term return for a balanced fund will be 6-8 per cent before tax.

The return after tax and inflation is likely to be 2-3 per cent.

"Our perception is that many clients continue to anticipate rather higher returns. In the near-term they may well be right.

"But the arithmetic for longer-term returns, based on today's prices and yields plus likely economic growth rates, is very clear.

The past decade or so has been a lengthy transition phase from the high inflation/high yield phase of the 1970s and early 1980s.

"Capital returns have been very strong through this period.

"As this phase ends single-digit expected returns should become the norm into the future."

Tower Asset Management's Tore Hayward supports Mr Dyer's view that investors have to wind back their return expectations.

While shares have been able to produce returns of up to 20 per cent in recent years, the likelihood is that 7 to 9 per cent is a more likely average.

Investors using retirement calculators need to be aware that many are designed for simplicity, and some substantially overstate the likely returns.

They also need to remember that fees, if they remain at current levels, will eat up a bigger proportion of returns if, as has been predicted, returns revert to long-term historical levels.

* Brent Sheather is a Whakatane stockbroker.

Philip Macalister is the editor of SuperTalk - a website where people can learn, discuss and debate issues surrounding superannuation.