The following post is extracted from my October 2015 newsletter to clients.
The coming market crash?
Don’t worry - I just said that to get your attention! The following is written quickly and will be unedited for grammar, repetition, loss of thread, and just plain silliness. But in amongst the randomness there might be one or two useful points to consider when thinking about investment strategy.
We can’t predict the future. That makes for an interesting life, and is a reliable theme in disaster movies. The ones where the geeky science guy protagonist discovers instrument readings showing there is about to be an earthquake/volcano eruption/asteroid collision, or whatever. Of course nobody believes him at first, and ultimately the disaster occurs (else we wouldn’t have a movie). But some people did at least listen to him, and make some preparations. The point being, whilst we can’t know the future in advance, we can plan for a range of contingencies. That’s why, in the investment world, we diversify. And it’s also why caution sometimes rewards us more than boldness. (I hasten to add, I’m not a geeky science guy, more like the slightly bemused engineering side-kick.)
Try Not To Lose Money In The First Place
Let’s recap on something we know. A 20% drop in the value of an asset is not fully countered by a 20% rise. If you start with a $100 investment and it falls to $80, you need a 25% rise just to get back to where you started. A 30% decline needs a 43% rise to recover. The effect is more pronounced the greater the price fall - a 50% crash needs a 100% rise to recover. This mathematical quirk is one of the reasons why even though a market can bob along quite happily with a few percentage points change in direction every day, it’s helpful to watch out for the downside. Stock-market observers would call a decline of 10% off new highs a ‘correction’, like a release of pressure before prices continue with the longer term rising trend. But a 20% decline might be indicative of a ‘bear market’, a sign of a change in direction of the underlying trend.
Here’s something else we should remember. Our period of observation should be appropriate for the timescale of our investment. We are not traders, we are investors. We don’t make decisions based on today’s price change. Or this week’s. Or this month’s. We take a long term view, diversify, and give the strategy time to work. If we find ourselves making ‘tactical’ changes more than once or twice a year at most, we should double-check that the time period of our decision-making process is long enough.
You Can’t Time The Market
You might have heard the expression “It’s time in the market that counts, not timing the market.” I’ve just Googled the phrase and got 220,000 hits. It’s a nice slogan, but doesn’t tell the whole story. Figure 1 below is a 20-year performance chart of the FTSE100 index (UK’s largest 100 companies). In the long run, markets tend to go up. But depending on your entry point, you might have to bear long periods (maybe several years) with your investment in negative territory. Leading investor Warren Buffet stated his favourite holding time is ‘forever’ - but speaking for myself, I can’t wait that long to retire. Hence we use the strategies of dollar cost averaging, and periodic rebalancing of our portfolio, to try to create gains where there might otherwise have been none. In my article “Reduce Risk, Remove Stress and Access Market Returns” (available via my LinkedIn page) I showed that the FTSE100 index (UK’s largest 100 companies) took until April 2011 (40 months) to recover back to its level of January 2008 at the start of the global financial crisis; whereas a simple dollar cost averaging strategy into the FTSE100 would have created a gain of 18% in the same period.
Figure 1 - FTSE100 20-Year Chart
Breaking Out The Spreadsheet
Let’s get back to that ‘time in the market’ slogan. Quite often you’ll see articles that show the effect of ‘missing out on the best days’ to be in the market. About half the results of the first page of Google do this; certainly some very smart people are using this metric to illustrate their point. I decided to duplicate the exercise myself. It’s some basic number-crunching, and historical prices for many major indices are available to download from Yahoo. I decided to look at 10 years’ worth of FTSE100 data up to 1-July-2015, hence missing the current downturn. Here’s the results:
- Closing price on 1-July-2005: 5161
- Closing price on 1-July-2015: 6608.60
- 10-Year Gain: 28%
Now let’s exclude the 10 best days (easily done by identifying the best days in terms of percentage gains, and then removing the actual numerical gain in the index on those days):
- Closing price on 1-July-2015: 3788.80
- 10-Year Gain (ten best days excluded): -27%
Yes, that’s amazing but true. Just by missing out the top ten days in the ten year period, a 28% gain turned into a 27% loss. That’s a pretty good argument for ‘time in the market’. But hold on a second. I wonder what will happen if we excluded the top ten worst days instead? Let’s see:
- Closing price on 1-July-2015: 9431
- 10-Year Gain (ten worst days excluded): 83%
This seems to me to be a significant finding:
By missing out the ten worst days in the ten year period, I would have outperformed the index by almost 43%. Relatively speaking, missing out the worst days is probably more important than capturing the best days.
Of course it would be near-impossible to do this in the real world, but it might be worthwhile going a bit further along this line of thought.
Today Is 4th October 2015
Suppose that right now I have $100,000 to invest, sitting in cash in my bank. I’ve read that markets are highly volatile at the moment, and are quite a bit lower currently than three months ago. They have started to come back up, and are nicely above the low point in August. Is this a rebound? Should I invest? The answer of course lies in my time horizon, my appetite for risk, my ability to sustain a loss, etc. But one piece of information that can go into the mix is an idea of which direction markets are trending. If markets are in a long term bullish trend then I’ll feel much more comfortable buying in to the recent volatility. But if the trend has changed and maybe the recent pullback will turn into a bear market, then my instinct (and the mathematics) tells me to wait just a bit longer to see.
‘Technical Analysis’ is the study of share price and trading volume data - via use of charts - to try to glean information about trends and future price movements. It’s a highly popular field and has grown an enormous following over decades. Some people ‘believe’ in it, and some don’t. Being a telecommunications engineer (a discipline that includes a great deal of time series analysis and associated statistics) over the years I have invested thousands of pounds (in books, courses and software) and thousands of hours to get to grips with technical analysis (or ‘charting’). Ultimately my opinion is this: there are some basic principles and simple techniques that can give us some helpful information, but charting should never be considered in isolation, and (for me) most of the techniques are too obscure and ambiguous to be useful.
Timing the markets is almost impossible (but note there are some who do claim to be able to do so). We should NOT try to guess the market bottom, or guess the market top, and jump in or out of investments accordingly. BUT we can use charts to sanity-check that an investment is in line with the underlying long-term trend.
As we already discussed, both mathematically (a 50% decline needs a 100% rise to counter) and statistically (missing the worst days is probably more important than catching the best days), we should be more sensitive on the downside than the upside.
So, What Exactly Is The Current Trend?
Ah-hah, that’s the question. Just as trends don’t go on forever, they don’t usually change overnight. Keeping one eye on some basic charts might help. The S&P500 is as good a place to start as any - USA represents close to 40% of global market capitalisation (source: Financial Times), and the S&P500 index captures about 80% of the USA (source: Standard & Poors). Hence one index covers roughly 30% of global equity value.
Figure 2 - S&P500 20-Year Chart With Price Momentum Indicators
A simple way to filter out short term fluctuations in price is to use a moving average. Moving averages let the trend emerge, based on a number of recent periods of data. In the top panel of Figure 2 above the blue line is the 20-month moving average; the red line is the 50-month moving average. Moving averages are backward-looking and will always lag recent price movements, nevertheless they do help us see the long term direction.
Technical analysts say that stock prices have ‘momentum’. The concept is that a price is more likely to continue moving in the same direction than to change direction. Hence by implication momentum indicators are supposed to be forward-looking. Momentum is deemed to represent the changing balance of supply and demand for the stock. In the short term, intra-day and day-to-day, there is plenty of volatility and it’s usually difficult to see anything useful in the noise. In the longer term, some price momentum indicators (often called oscillators) will occasionally give nice identifiable signals.
Most chartists don’t rely on a single indicator, but use several. Confirmation of a signal by multiple indicators is generally considered to be more reliable.
Two common indicators are the MACD (moving average convergence divergence) and the stochastic oscillator. There are literally thousands of websites that discuss the interpretation of these measures, and that’s not my objective here. What I want do is point out the yellow-highlighted areas in Figure 2 above.
Clear As Mud?
These are three zones of interest. In each case the MACD signals first and the Stochastic oscillator confirms later.
- In early 2000 the MACD bearish crossover was confirmed in the second half of the year by the stochastic dropping through the 50 level. This was the period of the dotcom bubble bursting.
- In late 2007 the MACD crossover was confirmed very quickly by the stochastic; this was the global financial crisis.
- Earlier this year in 2015 we saw a similarly clear MACD signal, although it has only recently been confirmed by the stochastic.
We can’t predict what comes next. Until it happens we have no way of knowing whether the trend really is turning negative. And we should naturally avoid any temptation to let our eyes trick us into seeing something that isn’t there. Either way, it pays be aware of what the charts might say, and maybe have a plan for when the lava flow hits or the Tyrannosaurus-Rex stomps on the car.
Implications For Current Investment Strategy
For money already in the market, we should not be in a hurry to make changes. If a bear market does develop, we’ll have time to act. Furthermore, portfolio diversification across geographies and asset classes gives us a buffer against the potential downside. For sure, it is an excellent time to be in the early phase of a retirement savings plan, or a dollar cost averaging strategy. Notwithstanding the signals from the charts, it’s way too early to say what will happen in the final quarter of 2015.