1/n This is an interesting question we've tried to calibrate. The challenge is the term "valuation". Makes sense when money resides with discretionary investors where cash is base asset and securities are purchased based on forward expected return. Presumably, a scenario
2/n exists where a discretionary manager finds nothing priced for positive expected return and sits 100% in cash. Obviously restrictions on portfolio construction reduce the proportion of cash allowed. Rodriguez at FPA was last I remember with 50% cash and he caught huge flack
3/n This ability to “flex” cash levels is largely what determines valuations. Ultimately if I buy and you sell, “cash levels” don’t change. The urgency which I get rid of my securities relative to your urgency to sell cash determines price which changes cash % of market
3a/n Note inverse relationship between cash in money market funds and S&P price. This roughly approximates cash % of market (scale is irrelevant here to be clear) 
4/n These prices are then compared to some denominator – Book Value, Sales, DCF, etc – to determine “valuation”. But unless market participants respond to the valuation signal, it has no impact on prices. The only thing that matters is cash levels.
5/n And this is where passive changes the game by agreeing in advance they will NEVER hold cash. As passive gains share, the proportion of cash falls mechanically driving “valuations” higher. As discretionary managers evolve to keep up, they in turn hold less cash as well,
6/n creating conditions where the only source of liquidity is cash external to the market. Needless to say, getting this cash into the market is challenging under most conditions. The easiest way the Fed can create buying power is by pumping up collateral, ie bonds
7/7 Systematic rebalancing means bonds must be sold and equities bought. This leads to the mistaken impression that cutting rates or buying bonds provides “stimulus” that then gets discounted into “valuations.” This is not the mechanism at work. It’s a collateral play.
