I love John Husman's work. I love the charts he includes, well maybe not the 3D ones quite so much, but I definitely love the charts overall.
But I have a problem with his analysis..
We have had an incredible bull run in interest rates since 1983.
Using relatively simple maths and some round numbers:
1983 to 2023 is a nice round 40 years.
Long term interest rates in 1983 were say 15%. The inverse of that is 6.67. That can be thought of as the PE ratio of the bond in a loose sort of way.
In 2023 long term rates are say 3.5%. The inverse of that is 28.6. That can be thought of as the PE ratio of a bond issued today.
So if the long term bond rate is an imperfect but big picture reliable guide to the relative returns of other assets, the running yields soughtt for various assets will have followed a similar path.
And if the yields required have followed a simple path, then so have the PE ratios.
You know I don't mean that they are the same numbers for all assets, and you all know that some assets will be doing materially better and others materially worse at any individual point in time. You know I am talking big picture here.
So this poses the question as to whether that undermines the use of historical returns as any guide to the medium term future. Why should we think that a benchmark based on periods when rates were aberrantly high is relevant? It is only relevant if we think there is a real possibility, even a likelihood, that there will be another aberrantly high period of rates in our investing timeframe.
Will governments and their institutions allow a period of inflation such that interest rates will need to go anywhere near the aberrant peaks of the past, or can we rely on governments and central banks to react to stop inflation?
The current experience says that Yes, central banks will raise rates so high that the economy will slow sufficiently that inflation will be beaten back to the target range of 2 to 3%.
In that case surely the expected range of long term interest rates is 0% to 5% and so the range of PE's of bonds is likely near infinity to say 20.
And other assets will as always in the past generally reflect that in their earnings and PE ratios unless there is some unusual event eg ChatGPT being released with profound global potential.
And how long will rates remain at an elevated 5%? Maybe 1 year in 7? Or 10? Or 15?
So when the market is confident that inflation will be beaten, how should it price returns and assets?
A mix of discount rates would seem appropriate. Maybe we thing 4% will be the cap and that next year rates will be 3%, then 2% the following 5 years and then a recession will take them to 0 for 3 years.
This sort of approach is used in valuing commercial buildings with a significant vacancy. A period and expense of getting the building to an average vacancy rate and rental income is allowed and is in effect deducted from what the value of the building would otherwise be. (#Covid, #SanFranVacancy and #WorkFromHome has thrown this methodology into some doubt, but not so much in the share market).
So if bonds are goiing to have an interest rate of 2% and a PE of 50 in 3 years time what is the value of shares and residential property now? Will shares go from PE of say 28 to PE of say 50 over 3 years? Does that mean a $100 investment today will be worth (100/28)*50 = 178? And if the costs of getting there is 5% pa for 3 years or say $20 when compunding is considered. is $58 a good return on $100 over 3 years?
It all depends on your faith in Central Banks and governments to hold the line against inflation. The stock market has bifurcated. Huge growth in revenue and profits possible for those best able to implement and commercialise AI (very high PEG ratio on those stocks now, maybe little growth likely from here), and the rest of the market as a whole excluding the AI stocks barely able to hold there value, and worse for regional banks under pressure.
Place your bets on governments running a tighter fiscal policy for the masses to reduce demand on the one hand and the central banks holding the line on rates until inflation is dead. The former is much less certain than the latter.
John has loooked carefully at the effects of demographics and their impact on economy wide growth, but there are a whole lot of other changes that have taken place. A few of the big ones going back are:
1 Leaving the gold standard
2 A fixed exchange rate regime
3. A floating exchange rate regime
4 The second world war
5 The Cold war.
6 The invasion of the eastern bloc countries by Russia
7 The Korean war.
8 The Vietnam War
9 The Oil shocks
9 The Rise and Collapse of Japan real estate and stock market
10 The collapse of the USSR
11 The "independence" of central banks (they are actually interdependent with government fiscal policy).
12 The 1992/3 recession
13 The formation of the European Union
15 The dot com bubble
16 The 2002 recession
17 The rise of China and 600 million lifted out of poverty, competing way more for food and other resources
18 Falling population growth and the commencement of decline in population in developed countries leading to projected increasing dependency rates and requiring dramatic but gradual redeployment of labor and capital.
19 The Global Financial Crisis
20 Covid disruptions to global economies and societites
21 The war in Ukraine and the current inflation breakout.
Why on earth do we think that anything before about 1970 is really relevant, other than to show the rhythms of fear and greed?
The average return over the last 100 years is not a meaningful guide to the next 50 years but John seems fixated on that as his guide. It certainly was not an appropriate guide over period of the greatest bull market in interest rates from the Mid 1980's to 2020. John has updated his menthodology to focus more on "market internals" which is more about market momentum now rather than the returns over the last 100 years.
Many stock markets have not had the returns of the US market, I suspect mainly because the USA, for all it's faults, is an English speaking country that embraced a culture that has permitted it to dominate intellectual property in the forms of movies and software, for 50 to 100 years. This enabled it to ride the wave of English becoming the main language of commerce. Other languages like Spanish and Mandarin Chinese and Hind just did not come to dominate commerce. The US experience is thus very different to that of most other markets.
No comments:
Post a Comment