Interestingly, I wrote a computer program last night to find out what the largest percentage drop the combined funds in my portfolio has ever had was.

There's quite a few assumptions - based on buying into the fund at inception (the same amount in all funds), and reinvesting all distributions. I added up the daily value of each of the funds and then found the largest drop from peak to trough (which assumes you timed it to buy exactly at the peak).

It seems that 2002 was a very bad year for funds!

My portfolio would have been down 19.5% from peak to trough. This is a lot higher than I expected. I thought my portfolio (quite geographically diversified) might have coped better than this.

It would be interesting to see how the NavraInvest funds would have performed had they been in operation during 2002 ... it may well have helped keep the portfolio a bit more bouyant.

This means that I need to ensure a maximum LVR of around 60% to cater for a similar downturn without facing a margin call. Which, ironically, is what I was doing anyway. Of course, this does not deal with a crash - I don't enough data from the funds I'm investing in now to see how they coped in the past.

In reality, this peak-to-trough situation is unlikely - unless very unlucky with your timing, you won't get quite as large an impact on your portfolio since your average buy price won't be quite as high. But it does show the need to be careful.

What I need to do is work out a range of scenarios that consider whether it would be better to:

- run at a lower LVR (ie deposit cash into your margin loan)
- keep a cash buffer (ie hold the cash aside to fund a margin call if required)
- just invest it all and deal with a margin call when IF it happens

There's a lot of assumptions in any scenario - but it would be an interest exercise.