This study is from the Federal Reserve Bank of San Francisco:
Chart 1 - The middle-age cohort (aged 40-49) divided by the old-age cohort (60-69) in Blue. The S&P 500 P/E Ratio in Red. As the Blue line drops, the old-age cohort(divisor) is getting larger. We know this old-age cohort gets larger for the next 30+ years. That won't change. And as they get older, they will continue to need "true yield" as offered by Fixed Income investments that are backed by an indenture (a legal obligation to provide, among other protections for investors, a steady coupon every 6 months). This demand for Fixed Income will come at the expense of equities.
Chart 2 - The Blue line if the Actual S&P 500 P/E multiple. The Red line is the Model generated S&P 500 P/E multiple. Note that it shows the P/E declining persistently from about 15.8x today to about 8.4x in 2025 before recovering all the way up to 9.14x in 2031!
So watch out when talking heads tell you that the only reason "bonds" have as low a yield as they do is because the FOMC has artificially kept them this low. True, that is part of the reason, but also true is that the weak economy (still only 2.5% growth after 4 years of massive monetary and fiscal stimulus) requires it. Also true, is that inflation is at a 50yr low. And finally true, is that the aging demographic requires "true yield". I think 10yr treasuries at 2.25% will draw in demand.
I am not so sure stocks have a ton of value at new all-time highs with GDP at barely 2.5%. One would think the GDP growth rate would be at 4% today and on it's way to 5-6% in 1yrs time. So I am not so sure the P/E multiple at 15x makes a lot of sense right now given the current and future demographic.