In response to my article last week, on the good it would do to one’s portfolio if one could avoid the 10 worst plunges in the last 40 years in the Straits Times Index, a reader wrote: “I thought it is difficult enough to ‘miss’ one worst day, and now you have to ‘miss’ 10 worst days.
“Who would have this kind of clairvoyance or prophetic ability and also guts to act? It is impossible to be able to have such high conviction to sell all of one’s stocks on the eve of the crash based on your gut feelings that the crash is coming.”
Valid questions. So the underlying question he is asking is: Is there a way to know when to buy and when to sell the market as a whole?
One would need, like the reader said, clairvoyance if the stock market operates in a world of its own, detached from the logic we know that operates in the physical world that we live in. But here is a central truth to the stock market: Underneath it all, there is an economic reality. Companies have to make money in excess of their cost of capital to be of value.
In terms of the assets that the listed companies own, there is always arbitrage around the replacement cost. For example, if you can buy a drinks factory in the stock market for half the price of building one, nobody will build one. Those with money will just buy their competitors’ plants via the stock market.
This will drive up the stock price of drinks factories. When the stock prices rise to such a high level that it is way cheaper to build a new plant instead, a new set of entrepreneurs will come in to build new plants. This will mark the end of the “up” cycle in the stock prices of drinks factories.
Conversely, if you can invest to build something – say fibre-optic cables – for $1 and sell it for $10 in the stock market, then you can guarantee many people would want to do that until we drown in fibre-optic cables. This is exactly what happened in 2001 and 2002.
We keep reading it, but it is true: Greed and fear of market participants drive stock market prices to the two ends of the spectrum.
This week, I will show that there is a way to tell when the market is over or undervalued. And it is not based on your gut feelings.
Remember, we talked about using price-earnings (PE) ratio – how many times a stock is trading relative to its earnings – to value stocks. We can use this metric to value the entire market as well.
But there is a problem with using the PE ratio without regard to business cycles. For a cyclical stock say, shipping company NOL, if it is at the peak of its cycle, yet you are still willing to pay 20 times earnings for the stock, then you are set to suffer a substantial capital loss on your investments. Because the earnings will come down, and the PE of the stock will rise sharply, say to 40 and 50 times, and the market will deem it too expensive. Its share price will then drop.
Conversely, at the bottom of the cycle, you could pay 50 or 100 times for NOL and you could conceivably consider it a cheap buy.
So in valuing the market using the PE, we need to remove the effects of business cycles. One way to do it is to take the average earnings of the market in the past 10 years, and compare it to the current price of the market.
That’s what I have done in the accompanying chart. The Straits Times Index (STI) data from Thomson Datastream started in early 1973. So the first point of the PE starts only in 1983 – 10 years later.
That line is plotted against the STI. I used monthly data. As you can see from the chart, the 10-year average PE of the STI traded between a range of 10 and 33.5. The average is about 21 times earnings.
The peak PE of 33.5 times was registered on Sept 1, 1987. In January 2000, it was 32.7 times. On Oct 1, 2007, it was 30 times.
If you used 25 times and above as an indication of market overvaluation – so you get out of the market, and you get in only at 15 times and below, you would have avoided the October 1987 Black Monday crash, the Asian financial crisis, the dot.com bubble and the global financial crisis.
The indicator would give you the go-ahead signal to enter the market in end-1984, but get out in the second quarter of 1987; enter again in the second half of 1998, exit by the third quarter of 1999; enter in March 2003, exit by end-2006; and enter again by November 2008.
Following this strategy, you would have obtained a compounded annual return of 12.5 per cent a year. This assumes a 1.5 per cent transaction cost every time you enter and exit the market. Dividends and interest earned in your cash holding periods are reinvested in the market. In other words, you grow $10,000 into $341,500 in 30 years.
A buy-and-hold strategy with dividends reinvested yielded 7.9 per cent over the same period. It grew $10,000 to $97,900.
Of course, you can modify your strategy. Take 15 times PE as the buying signal. But for every point that the PE goes lower, say to 14, and then to 13, then to 12, you just increase your investments in the market, say by 10 per cent each time. You set your own rules. There is after all a cheap sale going on.
And you start selling when it hits 25 times PE. The higher it goes, the more you sell. By 30 times PE, it is best to be completely out of the market.
This way, your return would be even higher than 12.5 per cent a year.
Now the question: Where is the 10-year average PE now? Well, it’s at 14.8 times – not in extreme undervaluation territory but just about, at our arbitrarily set buy level. But bear in mind that we are in an extraordinarily low-interest rate environment. To the extent that companies’ earnings were boosted by cheap financing, then the value in the market may not be as great.
Also, you have to take a view on whether you think Singapore companies can remain competitive going forward. Based purely on the PE, assuming there are no structural changes in the world compared to the past 35 years, there still appears to be a case for being invested.