Tactical Gold Allocation Within a Multi-Asset Portfolio Charles Morris Head of Global Asset Management, HSBC Introduction Thank you, John, for that kind introduction. Ladies and gentlemen, my name is Charlie Morris. I am a portfolio manager based in HSBC in London; I have been there for 15 years. Thank you very much for having me to speak to you here today. It is always a great pleasure to come back to Hong Kong. Like Marc, I first came to Hong Kong in 1974 when I was just three years old, so I did not have a great economic impact at that time. I am going to talk a little bit about my multi-asset portfolio and how we see whether there is a gold bull market and how we size our position within it. High Quality Assets Have Performed Well First things first: what has happened in recent years? We have had a great acceleration in high quality assets. I think if you talk to a lot of people they will tell you there has been a dash for yield. People think that there is a need to earn income, but I would disagree with that point. I would say there has been a dash for quality and, quite simply, there are plenty of high yield investments out there that are not being bought, but anything that is of high quality has been bid up to very, very high levels indeed. That is true in every single asset class: In equities, the reliable stocks with a known outcome have done very well. Cyclical stocks with an unknown outcome have done less well. For example, the consumer goods companies have low volatility in price because we know what their profit will be in the future. They have low regulatory risk, they have low competitive risk, they have low government risk, etc, and they are globally diversified businesses. Technology companies, by and large, cyclical companies and regulated companies have fared much worse. High quality bonds from countries that, so far, markets have believed in have done well. Bonds from countries that have been questionable, for example, southern Europe, have not done very well. It has been nothing to do with yield. In foreign exchange, the high quality currencies such as the Yen and the Swiss Franc have performed well. In commodities, the best performing, by and large, has been gold and I will come to that point in a second. In property, prime has outperformed all other sectors in most countries. Hence the concept of quality is key to what I am going to tell you over the next 10 minutes. Quality in Equities To illustrate that point, I am going to look at the stock market over the last 20 years. The top line on this graph shows you the high volatility equities during the 1990s. Essentially, this is the tech stocks and so forth. They performed extremely well during that period. The low volatility, high quality equities performed very badly during that time and look how interesting it is: at the top of the bubble the high quality equities were sold in order to fund the purchase of the more risky stuff. Over the next 10 years it was exactly the other way around: we have the high quality equities doing very, very well, up three times since 2000 and the low quality equities down 40%. The middle line is the index. Putting it all together, the tortoise has beaten the hare over this 22-year period. There has been a very volatile ride in high beta assets and low volatility has won the day. This gap is now quite considerable and, in my opinion, as long as this high quality asset trade continues that gap will continue to widen.
Quality in Commodities How can we measure quality in commodities? If we take the idea of price volatility that we did in equities, the same is true in commodities. Incidentally, there is another factor we can reckon there as well, but first and foremost, it is more or less true that the lowest volatility commodities have performed better than the higher ones in recent years, particularly since the credit crisis. What is more, the order of return is more or less inversely proportional to volatility. The second point is above ground supply. It is impossible to get correct data and so this data is the best that I could do. It is publicly available data of what is above ground supply and what annual consumption is. The commodity market of course does not have accurate data, which is the fun of it. Here, we have 42 days above ground supply for natural gas. The result is that it is incredibly volatile. As we go up the list we have more and more and when we get to gold it is 14,000 days, but that does include investment demand. If you strip out last year s investment demand that number is very much higher. Quite simply, the fact that we have a high above ground supply means that the volatility of the price is lower and what I am effectively saying is that supply and demand for gold does not really determine price, which is quite an extraordinary point. Most people would disagree with that. I think that supply and demand in the short term massively impacts natural gas, but there is a much lower impact on gold. What really matters in the gold market is investment demand. The Gold Market (a) Technical Analysis I am now going to talk a little bit about a few techniques for how we are going to measure whether gold is good or bad and how we should position it in our portfolio. Something we are going to do is trend following. Technical analysis is something that is frowned upon in most financial circles, but I think the longer term the data series the more credible it becomes. Quite simply, it works very well on securities where the volatility is low and it works very badly on securities where the volatility is high. Gold has low volatility, as we have already established, so therefore we should follow the trend and what we are going to do is try to, firstly, identify whether or not we have a bull market in gold. Secondly, we are going to decide how big our position should be and thereafter we are going to see how they can have an overlay to increase or decrease that position as we go through time. (b) Audience Question The first question I have for you is: how many times has the price of gold in dollars crossed its 200 day moving average since 1988? 1. Up to 25 2. 26 to 50 3. 51 to 75 4. 76 to 100 5. Above 100 I am not going to ask you to raise your hands, there is no need, but have a little think about it. The answer is it has crossed in both directions 106 times, so 212. I think a lot of people do not realise it is as frequent as that, so I suppose what I am saying is the 200 day moving average is pointless. Gold has crossed its 200 day moving average since 1988 in 106 trades, 32 of them made money, 74 lost money and most of that time it was in a very strong bull market. The return from crossing the moving average is 143% before costs, so after you have paid the brokers, many who are here today, you are probably under water. Had you done nothing and just held the asset, the return would have been 266%. (c) The Principles of Trading Moving Averages That is our starting point, so where does that take us? If we zoom in on a very volatile time, we can see how incredibly frequent these crosses of the moving average are and so it is a licence to lose
money very, very quickly. A better idea is to slow down, have fewer data points and slow down your time series. In the top left of this chart we have 200 day moving average, 40 week and 10 month. They are all the same time horizon. The fewer data points we use, very quickly our return increases and the number of trades falls of a cliff very rapidly. Hence it is a better idea to trade less frequently with less data. If we use the gradient rather than the crossover, again the returns pick up because we have fewer trades. Better yet, if we use longer term date of 700 day, 140 week, 35 month, they are all the same time period and again the returns go up, the trades come down. The best thing of all is this in the bottom right: not much of that and the returns go up quite significantly and the number of trades. I used 1988 data because I think the gold market is relatively normal post-1988; that is another conversation, but just take my word for it. There are only 13 trades in all that time, which you can hardly call over-trading. (d) The Tactical and Strategic Approach The first thing we are going to do is say that if gold is not in a bull market we do not want to own it, because we are portfolio managers and we do not believe in diversification for the sake of it. We own it when gold is relevant to the times and obviously believe those times are now. Once we have a bull market we are going to have 40% of our position. When we think the times are good we are going to increase it to 70% and when it is fantastic we will increase it to 100%. What do those things mean? Core Gold Models to Identify a Bull Market in Gold (e) Real Interest Rates You have already seen this slide of US real interest rates twice today. Very simply, when US real interest rates are below 2% or 1.7% to be precise, historically that is going to work very well for the gold market and so that is one of our conditions: not necessarily negative real interest rates, but low, below plus 2%, so the carry cost of holding gold is not too bad. That has happened 13 times since 1988 and the return from doing that has been 347% versus 239%, so an attractive, simple and occasional strategy. (f) Gold and the Long-Term Trend The second one and one I am very fond of is gold against a multi-currency basket. We look at the gold price in dollars, which is very interesting, but not that much if you do not live in America. The experience of owning gold is very different for different people around the world and the starting point of the gold bull market to most people s minds would be 1999. However, what I think is interesting in the multi-currency world is that it only really started in 2005. Therefore, I would argue that this gold bull market is not nearly as old as people think it is. It only really began seven years ago. Since that time it has done very well indeed, but the point is that here we have a 35-month moving average and this simple strategy has only had five gradient changes in the last 24 years. Thus again it is a very occasional strategy and it has significantly outperformed buy and hold. (g) Gold versus the S&P 500 The final one is the relationship between gold and the stock market. You have all heard of the Dow Gold Ratio; it is a brilliant and simple concept. Again, if we do the moving average gradient on that and only own gold when it is positive as one of our conditions, we outperform the market. There have been 18 changes since 1970, three since 1988, so again a very occasional signal. (h) Core Model Summary Putting all those three together gives us some degree of confidence that there is a gold bull market. All are true today. The interest rate signal was given in 2009, of course, when interest rates went positive in the credit crisis because inflation collapsed. The gold traded in multi-currency is just off its all-time high as we speak and gold is beating equities, although it has had a bad year. Hence gold
right now is in a bull market and because of that we can switch on our tactical models. The tactical models do not exist unless gold is in a bull market. Tactical Models to Increase the Position Size (i) Gold Volatility The first model is gold volatility. The level of volatility is shown on this slide and I hope you can see that is a chaotic picture. Between 1988 and 2000, gold was not in a bull market, so therefore why would you want to trade it? You would not. However, look at the difference, when gold is in a bull market, how the behaviour is completely different and suddenly it makes perfect sense. All I am saying and to reiterate a point I made earlier, is when the volatility is high you should mean revert; you should think the opposite of what is happening. When the volatility is low you should follow the trend. It is a very simple idea and I could give a presentation for an hour on the volatility of gold. I will not bore you with an hour on the volatility of gold, but it is a big subject and I am just touching on it here. However, I think that principle holds true and of course you can measure volatility over multiple time horizons. (j) The Gold-Silver Ratio The second tactical idea I have is the gold-silver ratio. I think it is a very interesting ratio because it is something that goes back in economic history for many hundreds of years and longer. Since it is a ratio money does not matter, it is just the relationship between these two metals. Pre-2000, there was an average gold-silver ratio of 72. In the bull market, the average is now 60. I think it is fair to say that over the next few years that average will come down yet again. However, the gold-silver ratio does not just tell us about precious metals. It tells us about general risk. High readings here tend to coincide with major buying opportunities in risky assets and low readings tend to coincide with excessive speculation in the market. Therefore, we should always be cautious when the gold-silver ratio is too low and we should always be more optimistic when it is higher. I could give a long talk about how exactly we read the signals here, but very simply, we should be encouraged to own gold when silver is, firstly, relatively cheap compared to gold and, secondly, shows the opportunity to outperform. (k) Sentiment The third one is crowd sentiment. There are many ways of measuring sentiment. We all know the COT data, looking at the futures market and the hedge funds are taking all the gold from one corner of the vault, selling it to the other guy and putting it in the other corner of the vault; they create the volatility in the market. Of course, the COT data always coincides with the medium term highs and lows, so it is useful, but it totally ignores the secular trend, which is driven by the real money. The real money of course comes from wealth preservers, whether they are central banks or wealthy individuals. There are other indicators as well. The Hulbert Sentiment looks at all the newsletters in America and is very effective, but my favourite of all is internet hits. This is a graph from the kitco.com website measured by Alexa, showing the number of hits on their website. The general public is getting very excited about the gold market when there are highs and they could not care when it is low. Every single one of these highs exactly coincides with a high in the gold price, so I think sentiment is a very important factor to include in the tactical allocation. The Tactical and Strategic Approach To put it all together, we want to identify that we are in a bull market and then how to position thereafter. When we know that real interest rates are low, we know that the trend is working well in a multi-currency basket and we know that the gold price is beating the stock market, it gives us some confidence that we are in a bull market. Forget the noise, just look at those three core factors, so we have a core position and we buy some gold. Thereafter, if we think the volatility is low or we think that silver is cheap relative to gold or we think the sentiment is weak that gives us some confidence to increase that position to the second level.
If that is an and statement and not an or statement so volatility is low and silver is cheap and sentiment is weak we can have a fully-fledged max position, but be careful, because at this point it is getting too good to be true and when that happens it normally is. At that point, volatility will be high, the price is over-bought, the crowd love it and we have to go back to our core position. Where are we today? Volatility is low, as has already been said twice today by various speakers and the volatility of the price in the last 12 months has been dropping significantly. The silver ratio has come right the way back to the high 50s and is now advancing again following silver s decline from the peak last year. Sentiment, as you saw on the internet hits, is incredibly low. I would be cautious on the COT data because it misses the secular theme. Summary In summary, high quality assets are beating the market. This is a major theme, but it will reverse one day and I think that the highest point in high quality assets will be in the lowest point of the world and so when they peak that will be the time to buy risky assets and we look forward to that day. Sell your house in Mayfair and buy cheap industrial council houses in the suburbs. Trading rules work very, very well on low volatility assets and not so well on high volatility assets, which I think is a point that is lost on many people. The core models are very, very important. These are the ones that I would consider long-term being a portfolio manager. A central banker might think they are day trading, but to me they are the big idea: are real interest rates driving this trade? Is the price strong in many people s eyes around the world? Is gold beating the king of risky assets, which is the stock market? Thereafter, once we have established we have a bull market, we switch on the tactical models. We never switch them on unless we are in a bull market. We go with volatility when it is low, we buy gold when silver is leading and we like the crowd to be adverse. I think the most important point is you do not say, I am bullish on gold or I am bearish on gold or This is my price target. None of that matters. It is how much of it you own in your portfolio that matters and that is a personal decision, but at least I hope to have given you some sort of framework to help you make it. Thank you very much.