A Different Analytical Approach

With Thanks to Charles Gave of GaveKal Research who kindly gave me permission to share his latest thoughts on his analytical framework to valuing the stock market.

I have read with great interest the arguments put forward here by Anatole that equities are in a “structural” bull market. Having listened closely to his presentation at Gavekal’s London seminar, I now understand where our main point of difference lies. Anatole argues that we are in a bull market that began in 2013 when US stocks broke above their long-established trading range and which continues to this day. 

In a nutshell, he looks at the red line in the chart below and concludes that the consolidation which started in 2000 ended in 2013. Ever since, the market has been in a structural bull trend similar to those it saw between 1900 and 1929, and from the mid-1940's to 2000. This approach allowed Anatole to make a great call in 2013-14.

My approach is slightly different. To me a “structural” bull market takes place not when the stock market goes up in nominal terms, but when the total return of the stock market is greater than the total return of the long-dated local government bond maintained at a constant duration. When this happens, it means that the growth rate of corporate earnings is higher than the cost of borrowing. In turn this implies that repaying the debt incurred to generate growth will be painless.

The ratio of the total returns between the stock market and the bond market tells me what is happening. If the stock market does better than the bond market, then we are in a structural bull market, if the stock market does worse, we are not. If the returns are similar, then the jury is out. This is illustrated by the blue line in the chart below, which shows the ratio of the total return of the S&P 500 to the total return of a 16-year zero coupon treasury bond, maintained at a constant duration.

SPX vs 0 Coupon Bond.jpg

Why 16 years? Because over the long term, the volatility of a 16-year zero is roughly the same as the volatility of equities, which implies that “risk” of the two assets is the same (if you accept that volatility equates to risk, which I do not, but let’s leave that aside for the purpose of calculating the ratio). Naturally, I am aware that 16-year zeros did not exist before 1980, but it is possible to calculate theoretical prices for a synthetic bond. In any case, the
economic reality does not change.

Looking at the chart, the first thing I see is that the two methodologies—Anatole’s based on the price return index, and mine which seeks to identify when the world moves from a phase in which the growth rate of earnings is lower than the cost of borrowing to a phase in which it is higher—both identified the same upside break-out points in 1900 and in 1943. However, the results since the early 1980's have been completely different. Since 1981, my ratio of the total returns of stocks and bonds has moved sideways, swinging backwards and forwards between 25 and 100. This gives a very different view of the world.

1) The peak of the stock market relative to bonds did not occur in 2000, but in 1981.

2) The first two break-outs—in 1900 and 1943—occurred at the same times according to both measures. But as far as I am concerned, the third upside break-out has not yet occurred. Although the market is approaching the upper boundary of its range since the early 1980's, unlike Anatole I cannot yet say that we are back in a structural bull market in equities.

3) However, I can say that very low short rates have fueled a big increase in debt, and a big rise in the prices of the assets financed by that debt. So on a balance sheet basis, the debt is not a problem, as long as asset prices do not fall. But on a cash flow basis, we can have no certainty that borrowers will be able to service their debts easily, especially if short rates go up. If short rates do rise substantially, the only way borrowers will be able to service their debts will be by selling assets. This would precipitate an abrupt decline in the stock market, in a classic “debt deflation”.

This suggests to me that equity prices have risen since 1981 not because the growth rate of earnings has exceeded the cost of borrowing, but because short rates have been pushed far below returns on invested capital. This propelled a massive increase in leverage, which led to higher asset prices. This is potentially a very dangerous state of affairs, because servicing and repaying the debt could become very difficult if asset prices start to go down and we find ourselves in the final phase of Hyman Minsky’s eponymous cycle.

This puts me in a quandary. I keep oscillating. I still do not know whether the most likely outcome is a structural bull market in equities or debt deflation. On the one hand, my total return ratio is approaching the upper bound of its long term range. So an upside break-out into a structural bull market in equities is possible. Indeed, if the economy moves into the upper, inflationary, half of my Four Quadrants grid, we will certainly move into a period when
stocks outperform bonds.

On the other hand, the US stock market is looking richly valued and there is no shortage of risk—both macroeconomic and political—in the world. This tells me I cannot be certain that asset prices will continue to rise, and that my total return ratio could fairly quickly revert from its present level of 80 to its long term average of around 50. If that happens, US long bonds will once again become a good hedge for equity investors.

At this point, I simply don’t have enough information to determine which is more likely. Perhaps it is time to invest in some straddles.

From Charles Gave, GaveKal Research, April 13th 2018 (www.gavekal.com)