It’s understandable that seeing your portfolio value drop seems like it must be a bad thing, surely that means you have less money?  Well actually no.  Ultimately what you have is “units” of things and each of those units entitles you to income which is ultimately linked to the profits of the underlying companies.  If you have the same number of units, you haven’t made or lost money when the price moves; you make and lose money when the profits change. (focus on the right measures)

The problem with selling to later re-buy “when things are better” is it doesn’t work.  The reason is quite simple – there is no method of accurately getting out and back in which ensures success.  Because you are buying “units of income” the price of these need to be lower when you re-buy at least 80% of the time to break even.  Furthermore because markets generally rise, they regularly achieve levels to which they will never return and therefore prices will always be higher.

Let’s look at this in practice.

Eg You own 1000 shares priced at $1, each paying 5 cents per year in dividends.  The market drops 25% and now they are worth $0.75 each or $750.  What’s been lost?  You are still entitled to 1000 5 cent dividends.  Even if you sell, you still haven’t lost as you still have the opportunity to buy 1000 dividend entitlements with your $750 (ignoring transaction costs).

If you sell

You now have $750 cash and provided you buy back in before the price rises above $0.75 you could actually make money.

Eg prices drop to 60 cents and you buy 1250 units.  You’ve now made 250 units!  That’s 250 more dividends forever! However if prices don’t drop further, or recover to say $0.80 before you re-buy, then now you can only get 937 units.  You’ve now lost 63 dividends that you can never get back.

So what! That’s only $3.15 per year, you still get $46.85 right?  What happens over time though?

The example below follows an investment in the US stockmarket from 1988 to 2011.  We compare a “Hold” vs “sell out to protect capital” approach.


I’ve used the S&P 500 index to measure prices and dividends.  Portfolio first purchased in March 1988 (great time to buy, just 5 months after Black October!).

I’ve assumed full sale of the portfolio in June 2002 (index at 989 after the Tech crash and September 11) but well before the bottom (index at 848 March 2003) with proceeds invested in cash earning 2% pa interest (roughly rates available at the time).  I’ve assumed re-entry to the market in June 2004.

I’ve assumed another full sale of the portfolio in September 2008 (index at 1166 after the GFC) but well before the bottom (index at 797 March 2003) with proceeds invested in cash earning 1% pa interest (roughly rates available at the time).  I’ve assumed re-entry to the market in September 2010.


In this example, both times whilst there was some downturn, the portfolio avoided further significant drops in capital value.  Missing the full rebound however ended up costing a little bit, by June 2011 “Hold” portfolio was worth $1.53M whilst “Protect” was worth slightly (11.3%) less, $1.36M.

Well isn’t that worth a few less sleepless nights?  After all you missed a lot of the worst of it?

But what about income?

Well, that’s the same.  The portfolio is worth less because you own less “units” (11.3% less in fact) and therefore you are entitled to 11.3% less dividends.  This will continue forever as the number of units won’t change unless you do this again and pick up on the downside.  By “protecting” your capital you have consigned yourself to a permanent 11.3% pay cut.

Maybe you lose in this example but if you’d bought in lower you would have won

Sure, that’s true. Whilst I’ve tried to use timings that are at least reasonable, you can always be accused of “cherry picking” the data when you reach your preferred conclusion[1].  Given my articles are general having a crack at someone else for doing this, I’d hate to be accused of committing the same offence.

To do this then, we’d need to measure the returns of real investors based on returns actually achieved (not what they say they got).  Fortunately someone already measures this!

It’s called the Dalbar Quantitative Analysis of Investor Returns and by looking at actual investor returns over periods up to 20 years, it measures:

the effects of investor decisions to buy, sell and switch into and out of mutual funds over both short- and long-term time frames. The results consistently show that the average investor earns less – in many cases, much less – than mutual fund performance reports would suggest “

So let’s look at what this study has found over the years.

The chart below shows the 2005 results of the Dalbar study for share fund investors returns.


The chart below shows the 2005 results of the Dalbar study for share fund investors returns.

“In its 15th annual study of mutual fund investor behavior, Dalbar discovered that equity, fixed income and asset allocation fund investors experienced average annual losses for all time periods examined except the longest (20-year) time frame. And even those positive returns did not keep pace with the average inflation rate.

This year’s report examines how investor behaviour actually exacerbated the devastating losses of 2008”

By 2010 Dalbar found the average investor had recovered to a small positive return,


Just by staying invested, you would have 2½ times as much as the average investor!

I appreciate this is a lot however hopefully I’ve provided enough real evidence that the golden rules of investing and wealth creation are still true,

that’s precisely why they are the golden rules.


[1] Actually I’d argue I’ve reached this conclusion because of the data, but supporting evidence is always good!
Matt Battye

Matt Battye

CEO, Financial Adviser

Analysing what can seem to be like complex issues, Matt is effective in using analogies to better explain scenarios and truths to the rest of us. This is what Matt enjoys – educating clients on the truths and debunking the commonly held (wrong) view.

Share This