Paul Lewis has produced a brilliant study that shows how the outcomes of investing a capital sum in cash would have been better than investing in shares over the past 21 years.

Paul is right, the numbers do  not lie and I’ve included the whole of his press release published this morning on this blog. Simon Read’s excellent analysis is widely available (BBC version here).

I had the chance to critique his work over the past few weeks and ducked the challenge, choosing to pass the research to my colleague Derek Benstead who is both cleverer and more articulate on investment matters.

Rather than paraphrase Derek’s argument, I have published his comments on Paul’s research which appear in the final part of this blog.

To summarise; our view is that Paul Lewis is right, but that shares are still the best investment when regularly saving for a retirement income.

1. Investing a lump sum into shares and converting that lump sum back into cash IN ONE GO – is risky and usually not worth it.

2. Trickling money into shares and trickling it out again (as income) is usually worth it, if the space between the trickles is long enough.

3. People who invest to disinvest are probably better in cash (unless they are thinking 25 years +) and are ISA bound.

4. People who save regularly for a regular income many years hence are probably better in shares and are Pension bound.

CASH BEAT SHARES FROM 1995

(press release only – full research here)

Paul Lewis speaks

Money in best buy cash savings accounts would have produced a higher return than a FTSE100 shares tracker over a majority of investment periods since 1995.

That is the shock finding of new research using best-buy cash data which has never been available before.

The results challenge the traditional view that putting money in a savings account is the poor relation of investing in shares.

The analysis also found that since 1995 investments in funds that track the FTSE 100 would have lost money up to a third of the time over investment periods from one to eleven years. Cash in a savings account always ends up higher than it started.

The new research compared returns from a simple tracker fund – which follows or ‘tracks’ the FTSE100 index of shares in our biggest hundred companies – with cash that is moved each year into a best buy one year deposit account with a bank or building society – sometimes called a ‘one year bond’. The tracker has dividends reinvested and the cash is reinvested each year with the interest earned.

It found that money put into this ‘active cash’ beat the total returns on the tracker in 57% of the 192 five year periods beginning each month from 1 January 1995 to the present. The tracker won in just 43% of periods. For some longer time periods the result were even more marked. For example, for investments made over the 84 fourteen year periods from 1995 cash beat shares 96% of the time.

The research was done by financial journalist and presenter of Radio 4’s Money Box programme Paul Lewis. He gained access to best-buy cash data back to 1995 from the financial information publisher MoneyFacts. Data back to 1995 has never been made available since it first appeared in the monthly MoneyFacts magazine. He says this data makes the research unique.

“This analysis of the new data shows that people who prefer the safety of cash can make returns that beat those on tracker funds in a majority of time periods.

“It also confirms that the risk of making losses on a shares investment is very real. Over any investment period from one to five years from 1995 to 2015 there was about a 1 in 4 chance or greater that the value of the investment would fall. Even over nine or ten years the chance of losing money was around one in ten. Few advisers know those odds still less inform their clients of them.”

“I have long suspected that the merits of cash were underplayed by traditional research which compares poor cash rates with often exaggerated gains on investments in shares.”

The new analysis produces different results for three reasons.

• It uses a real tracker fund so the gains or losses are after all charges
• It uses new data on best buy cash accounts which has never before been collated and 
• It moves savings once a year into the latest best buy – what Lewis calls ‘active cash’.

Further analysis of the data shows that for many starting dates from 1 January 1995 investing in shares over any possible period from one year upwards would have produced a lower return than using properly managed best buy cash. That is true for example for the whole of the two years from 1 October 1999 to 1 September 2001 and for four months from 1 October 2007 to 1 January 2008. Money invested on the first of any month on those dates and left for any period from 1 year to the maximum possible 15 or 16 years would have done better in cash than shares.

There are fewer comparable times when shares produced a higher return over every possible investment period:- 1 November 2008 to 1 September 2009 is the longest run and two earlier dates are 1 October 2002 and 1 October 2003.

Overall for investment periods of five years or more there are 38 starting dates when cash would always have produced a better return but only 24 starting dates when that was true of shares.

Lewis adds: “Cash is not right for everyone in all circumstances. But for a cautious person investing for periods of up to 20 years this research indicates that well managed active cash beat a FTSE100 tracker more often than not. And unlike a shares investment it can never lose anyone money.”

Money invested in best buy cash over the whole 21 year period from 1 January 1995 to 1 January 2016 would have produced an average annual compound return of 5.0%. Over the same period the tracker would have produced a compound annual return of 6.0%. The 1% difference is far lower than the 3% to 8% typically quoted for the ‘risk premium’ of investing in shares.

Derek Benstead on why Paul is right and why shares are still best for pension investing

Derek Benstead responds

Paul Lewis’s comparison of the historical performance of cash and shares is interesting.  It is interesting precisely because it is a historical comparison.  My preferred starting point is to look at history and see what I can learn from it.  I would rather look at history than at an actuarial or an investment model.  Modelling is very hard to do well.  Financial models are at risk of producing an unrealistic representation of the real world.  A difficulty of financial management illustrated by a model may only be a consequence of unrealism in the model, it might not be highlighting an actual problem in the real world.

Also to be avoided is the giving of an opinion without evidence, perhaps an inherited opinion, learned from others and not checked out.

So, I am very supportive of the notion of looking at historical data as a first port of call to see what we can learn.

I have assumed that Paul is thinking of individual, not institutional investors.  He has used an equity fund which is subject to “retail” levels of charges rather than “institutional” charges.  So most of my comments relate to an individual making decisions about where to save.

Having looked at Paul’s analysis, what would I conclude from it?  That equity performance is very unreliable.  Equities cannot be relied upon to outperform cash over 5 or 10 year periods.  There is even 1 period of negative return on equities over 11 years.

It is dangerous to be committed to making large transactions in equities at a single date.  A large investment in equities in 1999 when the market was very high would not do well, and performance over any period ending in 2009 will also be poor.  As Paul points out, it can take a long time for equities to recover from a trough (up to 42 months), which is a long time to wait to make a disinvestment.  I dare say it is also possible to explore the opposite point, of how long can it take for equities to come down from a boom so it is more worthwhile making a new investment.

If I received a large sum of money all at once (perhaps from a sale of a small business, or an inheritance) then spent it all at once 5 years later (say on a nice house), I expect I would leave the money in cash over those 5 years, and I imagine many people would do the same.  I don’t think I would want to run the risk of needing to disinvest from equities to buy the house at a time the equity market was low.  I also have the risk of investing at the start of the 5 years at a time the market is high, and the problem of making the judgement of whether the market is high or not.

Paul has pointed out in previous correspondence that professionals do not have a track record of adding value from asset allocation decisions, so the average man in the street is not going to be able to either.

In many cases, people will accumulate money gradually and spend it gradually.  So with no knowledge of what an expensive or cheap equity market looks like, one can save gradually into equities and spend gradually from equities.  At times the market will be high and that month’s investment won’t buy many shares, but this will be offset by times when the market is low and an investment buys more shares.  It requires no skill to buy shares at an average price if the money is trickled in gradually over a long period of time.  And conversely for disinvestment.  But this trickle in – trickle out process takes a long time both to pay money in over a wide range of equity market conditions and likewise to withdraw it over a wide range of conditions.

If I am saving for a deposit on a house, it may take a long time to save the deposit, so I could perhaps trickle money into equities, but I would still want the money all at once at the time of buying a house, and to be in equities at that time would not be wise.  It’s probably wisest to save for a house deposit in cash.

I expect the first resort for most people making savings is to save in cash. It is good to have cash on hand to replace a car, maintain a house, tide over a period of ill health or redundancy. If someone starts to save into equities, then I expect this is because they have enough cash savings and want to seek to earn more on money they don’t expect to touch for a long while.  In which case the criterion that equities need time is met.  And there is unlikely to be a need for a forced disinvestment at a bad time.

I have a mortgage, cash savings, and a DC pension invested in equities and property.  This isn’t as illogical as it sounds.  If I paid all the cash into the mortgage I wouldn’t have emergency money on hand, and I have DC pension savings because my employer contributes and it would be stupid to miss out on the employer’s contribution.  The only equity investments I have are in the DC pension.  They arise from trickled in money that I cannot presently withdraw (I’m not quite 55) and do not expect to withdraw from for years yet.

The best way for an individual to get exposure to equity investment is by membership of a defined benefit pension scheme which is open to new entrants.  A defined benefit scheme of longstanding receives income from its assets and contributions which it spends on benefit payments.  There is broad balance in the cash flows in and out and the DB pension scheme has little need to either buy or sell assets.  The risks of buying or selling assets at bad times are avoided.  Of course, open private sector DB schemes are few these days, and the sorry story of the exaggerated appearances of cost and risk arising from poor advice based on poor valuation and investment models is a tale for another time.

Paul’s analysis uses 21 years of data.  I would not use this quantity of data to make inferences for periods over 10 years.  Once the period being examined exceeds more than half the length of the data, there will be years in the middle of the data in appearing in all periods.  As the period lengthens, there is increasing constraint on the start date getting nearer and nearer to 1995 and the end date nearer to 2016.  I would not use this data to make inferences about cash v shares over periods of more than 10 years.  I wouldn’t make the specific point about the boundary between cash and shares being 18 years.  18 years out of 21 is far too constrained and affected by events in the late 90s and 20teens.

I wouldn’t use this data to make inferences about the difference between the expected returns on equities and cash for the next 10, 20 years.  For the next few years, interest on cash is likely to be lower than for much of the past 21 years.

The comments I have made focus on the unreliability of equity performance.  It suffices to look at the unreliability of equity performance only to make strong points about whether and how to invest in equities.  This unreliability means it is best not to make large transactions in equities.  It is best not to have forced timing of large transactions in equities.  It is good to have cash on hand to draw upon if needs arise.  Whatever the long run performance difference between equities and cash for the next 10, 20 years or more, it’s clear from this data that one can’t rely on equities outperforming cash over a 10 year period.  If we trickle money in to equities over 10+ years and trickle it out again over 10+ years, we’re easily looking at a time span of three decades from start to finish.

Paul Lewis will have as much knowledge as anyone of the dangers of “get rich quick” and “something for nothing” schemes and scams. Equity investment is not a get rich quick scheme.  Equity investment is a get rich slowly scheme from money you can spare.

Derek is an actuarial consultant at First Actuarial

Happy now?