Wikipedia: "Kondratiev waves (also called supercycles, great surges, long waves, K-waves or the long economic cycle) are described as sinusoidal-like cycles in the modern capitalist world economy.[1] Averaging fifty and ranging from approximately forty to sixty years in length, the cycles consist of alternating periods between high sectoral growth and periods of relatively slow growth. Unlike the short-term business cycle, the long wave of this theory is not accepted by current mainstream economics.
...
"Long wave theory is not accepted by most academic economists, but it is one of the bases of innovation-based, development, and evolutionary economics, i.e. the main[citation needed] heterodox stream in economics. Among economists who accept it, there has been no universal agreement about the start and the end years of particular waves. This points to another criticism of the theory: that it amounts to seeing patterns in a mass of statistics that aren't really there."
Good lookup. There's simply not enough data for something like K-waves. If the average cycle is 50 years, then only 6 or so could have occured since the Industrial Revolution. Way too small sample size to draw any conclusions, especially since it's likely the economy behaves more according to 'rough' fractal and power laws than clean, repeating sin wave cycles.
I've never been remotely convinced by cycle models.
Fourier Analysis shows you can model any curve by summing enough sine waves of different amplitudes and frequencies; so given any shape of any market over any time period, I can show a selection of long, medium and short cycles that will approximately match the market. But this will have zero predictive power.
This is the chart (I may not agree on the dates) that I had in mind... the issue for me is that the best asset class to invest in, will change over time.
