Aug 252015

There are three major types of Perth Mint Depository investor:

  1. Those that only buy gold
  2. Those that only buy silver
  3. Those that buy 50% gold and 50% silver

There are others who include platinum, or have different percentages, but the above three types are a significant majority of our clients. I find it interesting that most investors who weren’t strong goldbugs or silverbugs and couldn’t decide between them went with a simple 50/50 strategy. This begs the question: is this a good strategy and what is the ideal percentage allocation one should make between gold and silver?

To answer this I have assumed an investor saving regularly for retirement, for simplicity $100 a month, including rebalancing each month to bring the value of gold and silver held back to the target percentages. I also assume an investing period of 25 years, on the basis that one does not start saving serious money until 40 (see this post for the investor lifecycle logic behind this) and retires at 65.

I then ran through every combination of gold and silver percentages to come up with a total value at the end of the 25 year investing period (which is 300 months, or $30,000 in total cash invested). As an example, the chart below shows the portfolio value for an investor who started in 1985 and retired in 2010, across various percentage mixes.


If this person only invested in gold then 25 years later their portfolio would be worth just under $89,000. A silver only investor would have just shy of $93,000. However if this 1985 investor had perfect forecasting ability they would know that investing 37% into gold and 63% into silver would maximise their return at $94,932, over three times the total cash they put in.

However, given the varying performance of gold and silver over time, the year a person starts and finishes investing affects their return. To compare different years I divided the total portfolio value by the dollars invested, to give a simple “times” figure (eg 3 means you’ve tripled your money). The chart below shows this calculation for three selected years, each demonstrating that a different strategy maximised the return.


The year 1985 is shown here again, but as a “times invested money” figure. While a gold/silver mix was the best return when starting in 1985, if you started in 1988 you would have been better off just investing in silver as you would have increased your $30,000 by over 5 times, compared to a gold only strategy of quadrupling your money. Starting in 2002 (which only covers 13 years) a gold only would have been the best strategy, increasing your $16,300 investment by 1.67 times to $27,181.

Of course the above is all very handy if you have hindsight, but what if you are starting to invest in precious metals in 2015? In 25 years will starting in 2015 end up like 1985, 1988 or 2002 in terms of what is the best strategy given the relative performances of gold and silver? One way of answering this question is to look at the strategy that gave the best performance for each year and see if there is a strategy that performs best across all different time periods.

The chart below looks at each year starting from 1975 up to today and shows the gold/silver percentages that gave the best return for a person starting to invest in precious metals in that year over the following 25 years. From 1990 onwards it is not for a full 25 year investment period, but I’ve included it for comparison purposes. Note, the chart is not showing the best strategy for that year, but the best strategy for an investor starting in that year).


The chart shows that there isn’t any consistency across time in which gold/silver proportions are best. A skew towards silver seems to be best during the 1980s with 100% gold being best at other times. This is reflecting the relative outperformance of gold or silver over the various 25 year timeframes. I didn’t find this very helpful in terms of identifying an all-weather strategy.

I then noticed that some years, like 1985, don’t have a big difference between the best and worst strategy – in the case of 1985, between $89,000 and $95,000, a gap of less than 7% on an increase of around 200%. To get some sense of the variability of returns, I’ve charted the “times increase” of the 100% gold, 100% silver and 50%/50% strategies below.


This chart is similar to the one above, showing that a silver only strategy performed best in the 1980s and gold better than silver for most of the other investment time periods. The reason for the high performance in the late 1980s is because an investor starting then is averaging in during gold’s bear market, lowering their cost of purchase, and then exiting at gold’s relative highs in 2011-13.

What is interesting is how close the 50/50 strategy is to gold or silver when they are the best strategy and indeed there are times when the 50/50 strategy outperforms a gold or silver only strategy. The key for me is that over time the 50/50 is closer to the best strategy than the worst – for example, if you picked a gold only strategy and the next 25 years turned out like those for an investor starting in 1985 and retiring in 2010, then you would see an increase of 4 times when a 50/50 would have given you a 4.7 times increase, not far off 5 times for a silver only. A silver only approach however, would have underperformed a lot if the future turned out like the 1990s, when a 50/50 was very close to a gold only strategy anyway.

My conclusion is that a 50/50 strategy is a very good all-weather approach to take. You will not give up too much return if gold only or silver only turns out to be the best strategy but a 50/50 will protect you from underperformance if your choice of metal does not turn out to be the best.

Tomorrow I’ll have a look at the situation for Australian investors to see if the advice is any different, and also look at the effect of not rebalancing (just buying on a 50/50 basis each month).