With today’s new high (4,479.71) in the S&P 500, the share of funds model now rates the market as having the highest valuation since 1945. The share of funds in stocks is over 52.3%. Projected returns over the next 10 years are -1.87%/year, not adjusted for inflation.
— David Merkel (@AlephBlog) August 16, 2021
Q2 2021 hedge fund letters, conferences and more
What I posted after the close yesterday…
I don’t have much more to say. The next Z.1 report comes out in late September, but will only bring the data up to June 30th. I use a pair of adjustments to adjust the share of assets invested in risk assets (stocks) between releases. So far, the adjustments seem reliable as the interim estimates have been tracking the results from new releases.
Could the model be wrong? Sure. I don’t run into many professionals who endorse the asset share model. Part of that is the industry gets compensated off of assets under management. No one wants to forecast declining future revenue. No one wants to have clients move out of stocks, as the percentage fees from stocks are higher than bonds or cash.
Aside from that, the asset share model is a balance sheet model. Most analysts prefer income statement models, even though they work as well, because P/E ratios are more intuitive to market participants. The asset share model is not the only balance sheet model — the Q-ratio is also a balance sheet model.
Though interest rates are low, they are not negative. 10-year investment grade bonds are competitive against domestic stocks at this point. Even if you are losing against inflation, you are losing less against inflation than the market as a whole. Same for cash. I don’t think that there is no alternative. Here are the alternatives:
- Investment grade bonds (market duration)
- Cash
- Value stocks
- Cyclical stocks
- Foreign stocks
- Emerging market stocks and bonds
So consider the alternatives, and consider hedging. I can’t nuance this anymore, as we are in uncharted waters. We are touching the sky.
Article by David Merkel, The Aleph Blog
