Over Christmas, I read a blog on an online newspaper about ‘top strategists’ forecasts of the stock market. Quite rightly, the blogger wondered about what the point is of asking so-called experts what will happen in the market over 2011 when they all typically get it wrong. I wasn’t cited in the article upon which the blogger commentedbut I still felt hurt. Was it my pride? The person certainly wasn’t being nasty or even impolite. He or she was simply bewildered.
I have no idea why other people make forecasts or what they do with them. They are old enough (and paid well enough) to look after themselves. I’m not a Catholic but I feel like going into confessional. Not for my forecasts which have been wrong (plenty!) but for not having explained what it is that I do for a living. If you bear with me, I might help you understand what a professional in this industry does for a living. Of course, not everyone is a professional and maybe I am pushing it to call myself one. You can decide!
I now (as of Christmas) perceive that there is a huge gulf between me and the people I would like to engage. Mea culpa. I also need to differentiate my way of forecasting from some others. I have a very systematic (I like to call it scientific) approach to forecasting. Unless I have a systematic approach to forecasting, I have no idea why I am wrong when I am (and I expect to be on occasions) and how to correct or improve my forecasting procedures. What I do does not necessarily make me as good as others and certainly not necessarily better, but I do not understand how someone who mainly relies on gut feeling survives in this forecasting world.
My plan for this research note is simple – explain how I think; how I use forecasts; how I rationalise my forecasts; and how I take aim at celebrity forecasters who make a career out of getting it wrong, upsetting people, and profiting!
I expect my market forecast to be wrong. I even hope they are (at times). And I profit by their error! It is absolutely necessary for anyone who wants to consider my forecasts to understand my processes (without going through the full deal, just some simple statistics).
I believe that there is some process (equation) that generates data such as the S&P/ASX 200 (the market). In my world, there are no Fibonacci retracement levels, resistance levels, head and shoulders and the like. That is not to say that I think that people who follow those methods are wrong – it’s just not my thing.
I believe in the movie Sliding Doors. As a propeller head, or nerd, I am not sure who the actress was but she lived two parallel lives in the movie. In one, the sliding doors on the underground train closed and she missed it. In the other, she just made it. If she caught the train, she caught her partner in bed with another. If she didn’t, a different life ensued.
In my world of applied finance, the stock market faces sliding doors every day, if not much more often! Real events with real people but the outcomes can vary so much on such small trigger points. If she faced that sliding door every day, on average, she would catch out her man. And that’s how forecasters like me think. There are many paths to the end of the year in the stock market. None are predictable, but the average is!
Let’s take my current forecast for capital gains on the market for 2011 – at 14 per cent. That is the thick black line in Chart 1. The other ten coloured spaghetti lines in that chart are 10 possible sliding doors situations – assuming that my 14 per cent forecast is correct, as is my market volatility forecast of 12.5 per cent. With 10 ‘scientifically drawn’ sliding doors outcomes for 2011, there is a wide variety of outcomes. The lowest has the market falling over much of the year and the higher one doing quite well. All are equally likely outcomes assuming that 14 per cent growth and 12.5 per cent volatility are correct.
Chart 1: Sliding doors – 10 alternative forecast scenarios
: Woodhall Investment Research
Of course, there are many more than 10 possible outcomes. In my discipline, we would consider hundreds of thousands, if not more. In Chart 2, I show the first 100 outcomes from an Excel-based experiment. Unless a forecaster does not believe in volatility, a reasonable range for forecasts has to be about 4000 to 7000 for 2011. Of course, higher or lower trend growth changes the outcome but the width wouldn’t be much different.
So I am trying to forecast the black line – the average – but people naturally want to compare me to the actual. Am I wasting my time? I think not. I cannot live in my world of investing without my forecasts. Indeed, I quite like it when they are wrong!
There are three uses of my forecasts that are essential for my way of investing. First, without a credible forecast I cannot work out whether the market is under- or overpriced. It is not quite as simple as comparing the lines in Chart 2 with the average because the forecasts evolve over time as new information comes to hand. Nevertheless, my forecasts, which are based on broker consensus forecasts of dividends and earnings, are the centrepiece of my measure of market exuberance.
Chart 2: Sliding doors – 100 alternative forecast scenarios
: Woodhall Investment Research
Second, by having a credible, systematic forecasting system, my daily updates of 12-month-ahead forecasts give me vital information on whether or not to adjust my investment decisions. Thirdly, I forecast each of the 11 major sectors of the markets. While each sectoral forecast suffers from the sort of variation I show in Chart 2, some of this variation is common to each sector. That is, if my market ‘average’ forecast is too high, all sectors will have a high-side bias. Therefore, I have a strong degree of reliance on my sector forecasts in terms of rankings. It is being in the right sectors that gives outperformance.
I show my sectoral forecasts for 2011 in Chart 3. The usual suspects of energy, materials and industrials (particularly mining services) stand out as the likely key performers. As it turns out, my view of 2011 is much the same as my view of 2010 was 12 months ago. The market was flat last year. Both years started at about 4800.
Chart 3: Sector forecasts for 2011
: Woodhall Investment Research
In February 2010, exuberance was negative – the market was underpriced by about five per cent. Importantly, industrials were also underpriced and had a strong forecast. A buying opportunity! In June/July 2010, the market was even more underpriced (10 per cent to 15 per cent) – as were materials. Materials had a strong forecast and so another buying opportunity. These opportunities, together with some others, and a good sectoral allocation and stock selection is how I achieved substantial outperformance over my benchmark S&P/ASX 200 index in the flat market that was 2010. None of that could be done without some (substantial) faith in my forecasting methods. My forecasts of the average turned out to be well above the sliding door that was 2010 – but I made money and that is the main reason for me to forecast the market!
By definition, it is not possible to predict which sliding door will occur this year – or have been predicted last year. However, I can try to rationalise what is happening as we go along in a form of justification. For example, in July of 2010 I was trying to work out why our market was underperforming that in the US. As I show in Chart 4, we almost moved blow for blow with the S&P 500 through the turbulent 2008 and 2009 period. Then we broke loose. We never had a recession during this period and the US economy was really struggling until the end of 2010 but we fell behind.
Chart 4: The S&P 500 (rescaled and the S&P/ASX 200
I put our failure to keep up with the US down to the mining tax that was first hinted at in the press at the end of January 2010. Materials stocks had taken a big hit and the argument was plausible. However, when the dust settled on the leadership, and the subsequent renegotiation of the mining tax, our market didn’t play catch up. End of that idea. Then I thought about our dollar. When Australians invest in overseas equities we usually view their performance as hedged in Australian dollars. So foreigners presumably look at our returns in $US.
I find the $US translation of our market a ‘dead ringer’ for the US throughout 2010 in Chart 5. But the bit in the middle looks well off. Hence, I form my hypothesis. When the world was in turmoil, investors just passed on the overnight change on the S&P 500 to the S&P/ASX 200. There wasn’t time for fancy footwork. However, volatility fell quite sharply at the end of 2009 giving (overseas) investors time to think – and think in $US. I can’t prove this hypothesis and it will take a big change in currency to disprove my hypothesis. But I like it.
In my opinion, it is extremely difficult to predict short-run changes in our currency. However, when the US economy is noticeably stronger and the Federal Reserve talks of interest rate rises from a base of close to zero per cent, our dollar could tumble back to an 80c sort of exchange rate. Many other things are possible but this is my current thinking. I don’t know when the US will look that strong but it seems unlikely to occur in the next six months. It does seem quite possible in the next 18 months so I conjecture that sometime in 2011/12 our dollar could fall 10 to 25 per cent. Couple this with my hypothesis based on Charts 4 and 5 and that translates to big gains on our stock market on top of my 14 per cent that doesn’t take the exchange rate into account.
Source: Thomson Reuters (DataStream)
So, to cut to the chase, I am expecting our market to grow strongly – but not fantastically – at 14 per cent per annum. Then, if we get the dollar ‘kick’, there is the chance of capital growth jumping up to very big numbers – 14 per cent plus any depreciation in the dollar. If the kick doesn’t eventuate, that doesn’t hurt my equity play with a possible 14 per cent growth. If the kick does eventuate, I don’t want to be caught out of the market.
To put all of my hard-core econometrics into perspective, I also like to set my results in an historical perspective. In March 2009, I did an analysis of every bear market there had been to see if there were any common patterns. I found 13 bear markets since 1875 and six of them had very similar characteristics on the way down as the 2008-9 bear market. That then led me to believe that the current market could have bottomed. Not proof – but it wouldn’t be unusual. Since these markets also shared certain common characteristics on the way up, I kept updating my chart.
I show that analysis updated until June 2010 for three markets: the current recovery, the recovery from the Great Depression and the recovery from the 1973-4 recession. I used that Chart 6 on Switzer TV at that time to argue that I wouldn’t be surprised if we didn’t continue to follow the same recovery path. I further update that analysis to the present in Chart 7. Still looks good to me.
Chart 6: Bear market retracement to June 2010
: Woodhall Investment Research
Chart 7: Bear market retracement to January 2011
: Woodhall Investment Research
Such charts do not prove or predict what will happen. But if something keeps happening, I want a reason why this time will be different. We didn’t have a recession this time around so, if anything, I would be less surprised by an upside than a downside break-out. The implication of Chart 7 and my conjecture is that the S&P/ASX 200 might rise 3000 points in three years. In managing risks I always want to manage both the downside and the upside.
As a final straw to break the bear market back, I turn to US GDP in constant prices (inflation adjusted). Nouriel Roubini, the famous – or is it infamous – bear argued at the start of 2009 the GDP chart would look L-shaped implying the economy would fall and stay down for a very long time; in short, a depression. A few months later, when there were signs of life he changed his mind. Now it would be U-shaped because it would be a very slow recovery but not quite a depression. By the end of 2009, he was gunning for a W-shape, or double-dip, because he thought the recovery wouldn’t last.
I show what actually happened in Chart 8 with a letter ‘V’ superimposed for ease of comparison. Within three years, the US has gone through a classic V-shaped recovery. GDP is now higher than it was at the previous peak. Unemployment is still a problem but that should start to move soon as the US consumer seems to be getting back to the stores. The final irony though is that Roubini, a professor at NYU, still comes across as worried about all the markets but it was reported on Bloomberg TV in December 2010 that he had just bought a ‘condo’ in Manhattan for $5.5 million.
Chart 8: US GDP
The questions this purchase raise for me are at the crux of what the big bear is really thinking. Would he buy if he thought that the property market had not bottomed? Did he borrow to buy? If he did, what are his new feelings on carrying debt in this market? If he didn’t, how much of the $5.5 million did he earn from the speakers’ circuit telling everyone the financial world was all but imploding? The final question is when did he ever predict good times?
I am tolerant of other forecasting methods but I prefer mine. But I think any forecaster should be forced to reveal his or her track record – in a meaningful way. Consider a weather forecaster in Manchester (north of England). If that forecaster every day predicted a sunny day, he would get everyone sunny day right. However, he would be laughed at because on very few days does the sun shine in Manchester. It is the same with recessions. Forecasters should point to successful forecasts of both good times and bad times.
Secondly, the forecaster should have skin in the game. If he doesn’t believe his forecasts enough to invest or act on them, why listen to him?
Click here to subscribe to the Switzer TV channel on YouTube and keep up to date with all of our shows.