I wouldn't dream of investing in equities over a period of just 4 years

Sarenco

Registered User
Messages
7,876
Copied and moved from another thread - Brendan



I wouldn't dream of investing in shares over an investment horizon as short as 4 years as suggested.[by Brendan]
 
Last edited by a moderator:
I wouldn't dream of investing in shares over an investment horizon as short as 4 years as suggested.

Intend to move in 4-5 years also so not willing to lock savings away for long term

In the old days when investment managers took a huge lump out of the initial investment and probably charged an exit fee as well, one had to invest for around 10 years to recover the charges. Even if you bought shares directly, you were paying 1.5% commission on buying and selling and 1% stamp duty, so you needed time to recover the 4% transaction charges.

There is also the idea that the longer you invest, the more likely you are to get a positive return, so you need to invest for a long time.

These days the costs are much lower so this is no longer a reason why you should not invest for 4 years.

So what about the argument that you might lose money over 4 years? Over 4 years, there are a few possible outcomes ranging from a serious fall in the value of your investments to a serious rise in the value of your investments. However, the chances of a rise are higher than the chances of a fall. The net result is that the return from investing in the stock market should, on average, be higher than putting your money on deposit.





Brendan
 
Last edited:
The net result is that the return from investing in the stock market should, on average, be higher than putting your money on deposit.

Looking at historic US data, there is a roughly 70% probability that US stocks (total market) will outperform 5-year US Treasuries over any given 5-year holding period. However, that ignores investment expenses and, critically, taxes.

There are no investment costs involved with buying 5-year State Savings Certs and the returns are tax free. In contrast, there are costs (commissions, stamp, etc) involved with purchasing equities directly, dividends are taxed at an investor's marginal rate and any gains are subject to CGT.

When you take account of investment expenses and taxes, I would suggest that the risk/reward analysis changes dramatically and the probability of a diversified portfolio of stocks outperforming 5-year Savings Certs falls dramatically.

I would "guesstimate" that the probability of the net, after-tax return on a diversified equity portfolio outperforming the net, after-tax return on State Savings Certs over a 5-year period is materially less than 50%.

In other words, it's a bad bet.
 
Looking at historic US data, there is a roughly 70% probability that US stocks (total market) will outperform 5-year US Treasuries over any given 5-year holding period. However, that ignores investment expenses and, critically, taxes.

There are no investment costs involved with buying 5-year State Savings Certs and the returns are tax free. In contrast, there are costs (commissions, stamp, etc) involved with purchasing equities directly, dividends are taxed at an investor's marginal rate and any gains are subject to CGT.

When you take account of investment expenses and taxes, I would suggest that the risk/reward analysis changes dramatically and the probability of a diversified portfolio of stocks outperforming 5-year Savings Certs falls dramatically.

I would "guesstimate" that the probability of the net, after-tax return on a diversified equity portfolio outperforming the net, after-tax return on State Savings Certs over a 5-year period is materially less than 50%.

In other words, it's a bad bet.

Hi Sarenco

I am confused. You often recommend that people invest in the stock market.

If it's a "bad bet" over 5 years, then why is it not a bad bet over 10 years or over 30 years?

Brendan
 
If it's a "bad bet" over 5 years, then why is it not a bad bet over 10 years or over 30 years?

Hi Brendan

Essentially it's because the probability of stocks beating bonds increases over longer holding periods.

Again looking at historical US data, the probability of stocks (total US market) beating bonds (5-year treasuries) increases to roughly 80% over 10-year holding periods, 90% over 15-year holding periods and (essentially) 100% over holdings periods of 20 years and longer.

It's also relevant that the probability of suffering a loss on equities diminishes over longer holding periods. Remember Mr Buffett's golden rule of investing – never lose money!

The following are the worst real annual returns of the S&P500 over different holding periods from 1871 to 2016 (with dividends reinvested).

3 years -35.2%
5 years -13.2%
10 years -5.9%
20 years -0.2%
30 years 1.9%
40 years 3.2%

I also try to emphasize that investment costs and taxes can skew the risk/reward analysis quite dramatically so that paying down debt (essentially the same thing as buying a tax-free, cost-free bond) is often the best use of after-tax savings on a risk-adjusted basis.

Hope that makes sense.
 
Let's take a one year period for simplicity.

Let's say I can put €100 on deposit at 2%.

The alternative is to put it in an investment with the following outcomes

50% chance that it will be worth €120and 50% chance that it will be worth €90.

The expected value of my investment is €105 (€120 x .5 + €90 x.5)

So I should invest the €100 rather than put it on deposit.

If I repeat that every year, I will lose half the time, but my gains will exceed my losses.

That will work out something like being ahead 80% of 10 year periods, and 100% over 20 year periods.

If there is a positive expected value, then it should not matter what the holding period is, if I can handle the potential losses. ( Excluding transaction costs.)

Johnny has an income of €100k, owns his home mortgage free and has €100k cash. He should invest in the stock market, because he can handle the losses and expected returns are higher.

Mary has an income of €50k and is planning to buy a house next year. She has €60k saved. She can't take the risk of investing in the stock market as a 50% fall would mean she can't buy her house. She must put it on deposit.

In the original question, from which I extracted this thread, Lobster had €600k in cash and needed to spend €250k in 4 years.

He should invest the lot in the stock market. After 4 years, the most likely outcome is that the €600k will have increased in value by more than any alternative. But it's possible that the €600k will have fallen by 50% over 4 years. So what? He will still have €250k to spend in 4 years.

If he had only €250k to invest, I would recommend a deposit account.

Brendan
 
Ahhh! A recurring theme on AAM. Boss just to be a bit more posh on the numbers. A typical model would posit that shares can expect to deliver a "risk premium" of, say, 3%. But with a risk (technically volatility) of 20%. The math gives the following results for the probability that shares will outperform deposits.
Over 1 year 56%
Over 4 years 62%
Over 10 years 70%
So yes the law of large numbers operates as you describe but not quite as dramatically as you surmised.

Now why should this be so? Not because shares are real assets with real entrepreneurs blah blah blah As Rory Gillen has pointed out, Coca Cola has been a very well managed company by any of the accounting metrics. It has however been a poor investment. Why? Because its prospects were overpriced by the market.

This is all about the psychology and dynamic of the stockmarket as a second hand market.

Over the last century or so the stockmarkets (Japan a notable exception) have fairly consistently over compensated for the fear of short term volatility. Let us say that this is still the case. Then whether the stockmarket is a good investment for a particular individual depends on how her risk/ reward calculus stacks up against the market norm. Two examples have been cited, one where the individual would be less risk averse than the norm and one where the opposite is the case - thus pointing to two different but equally rational investment decisions.

But are current stockmarket prices still compensating for fear of short term volatility? Maybe with interest rates so low, the search for some sort of return is dominating the fear of short term volatility. If interest rates return to anything like normal levels of say 3% to 4% in 4 years' time it would seem to me that stock prices will suffer a major correction. Who knows?
 
Last edited:
The math gives the following results for the probability that shares will outperform deposits.
Over 1 year 56%
Over 4 years 62%
Over 10 years 70%

Duke - that is very helpful thanks.

So someone who invests over 4 years will win 62% of the time and lose 38% of the time.

At first sight, those are great odds which I, for one, would be very happy to take.

But that is only part of the answer. If he wins small but loses big, then it could be a bad bet.

Have you any way of guesstimating the potential losses over 4 years?

Brendan
 
Have you any way of guesstimating the potential losses over 4 years?
I'm not sure if this answers your question but the total return of the S&P500 for the four years ended 31 December 1933 was -70.5%.

To be honest, I think we may be talking at cross-purposes. I wouldn't invest in any asset unless the odds of that asset producing an acceptable return over my investment horizon were overwhelming (albeit there are never any guarantees in that regard).

To me, anything else is gambling, which doesn't really appeal to me.
 
Have you any way of guesstimating the potential losses over 4 years?

Brendan
From the model, which for the nerds is called the lognormal model and is used by practitioners for pricing options.

In 4 years:

There is a 10% chance of underperforming deposits by more than 32% (i.e. losing more than 32% versus a deposit)
20% chance down 19%
30% chance down 9%
60% chance up 2% or better
50% chance up 13%
40% chance up 25%
30% chance up 39%
20% chance up 58%
10% chance up 88%

The model tilts the numbers towards equities both in probability terms and also in distribution of monetary outcomes.

Sarenco's observation that anything which involves an unacceptable chance of loss is "gambling" is worth comment.
I personally define gambling when someone accepts that on balance the expected outcome of a punt is negative, but still undertakes it in the hope of a positive outcome.
This does not apply in the model I have described for the stockmarket which, whilst allowing for negative outcomes, on balance expects a positive outcome.

However, Sarenco's is an equally valid definition. But whilst my definition is absolute Sarenco's is relative to the punter.
 
I personally define gambling when someone accepts that on balance the expected outcome of a punt is negative, but still undertakes it in the hope of a positive outcome.
Interesting observation Duke – I never thought of it in those terms.

Maybe the better distinction is between investing (where there is always a reasonable expectation of a gain) and speculating (where a significant risk of loss is more than offset by the possibility of a significant gain).

As a matter of curiosity, does your model take account of investment expenses or taxes?
 
Interesting observation Duke – I never thought of it in those terms.

Maybe the better distinction is between investing (where there is always a reasonable expectation of a gain) and speculating (where a significant risk of loss is more than offset by the possibility of a significant gain).

As a matter of curiosity, does your model take account of investment expenses or taxes?
Nope, no expenses or taxes.

Yes, "speculation" fills a gap.

So we go from "gambling" where we know the odds are tipped against us through "speculating" where we take significant risks but with the odds in our favour through to "investing" where whilst some risk is present on balance the expectation is overwhelmingly favourable.

So I would suggest that day trading is gambling (because of the costs), short term stock market positions, say indeed up to 4 years, are a tad speculative but long term equity positions are investments.
 
Last edited:
If something has a positive expected value, then it's worth the investment, if you can handle the risk.
I wouldn't arrive at that conclusion Brendan.

Stocks always have a positive expected return but then so do 5-Year State Savings Certificates. I would estimate that the odds of stocks beating 5-Year State Savings Certificates over a 5-year holding period are actually negative once you take investment expenses and, critically, taxes into account. So, from my perspective, purchasing stocks for such a short period isn't investing at all. It's not even speculating. It's gambling!

I'm not interested in gambling – or speculating – to fund my future financial requirements.

Your ability to take a risk is certainly important. But your need and willingness to do so are also important considerations.
 
Hi Sarenco

If you have worked out that it's right to do it over 20 years, then it's arithmetically right to do it over 5 years.

Unless there is some peculiarity in a particularly high return on State Savings Certs over 5 years.

The key point that there is a positive expected value in each year. We just don't know in advance which the good years will be.

Brendan
 
Boss a riddle for you. Jane has worked hard all her life and is celebrating her retirement. Why not? She has a €1M nest egg accumulated all those years. In comes Donald Trump and offers her 10/1 on the toss of a coin provided she bets the full €1M. Should she accept the bet?
 
Absolutely not.

But if an honourable man like yourself offered her 6/5 odds on a €10,000 toss, she should take it.

Or better still a series of tosses at 6/5 for €1,000.

Brendan
 
But if an honourable man like yourself offered her 6/5 odds on a €10,000 toss, she should take it.

Why? She's already met her financial goals - she has no need to take that risk.

Would you take the same view if the odds were 2/1?

Or what if a condition of the bet was that the coin toss had to be an annual event?
 
Well the point I am making is that risk has its price and that price is purely personal. JP McManus would bite Donald's hand off for that bet but Jane, we all agree, must reject it. Expectation is not everything.

PS you have no grounds whatsoever for castigating me as an honourable man:)
 
Last edited:
Back
Top