packer16 wrote:The one thing about a belief, like value vs. growth, is that you have to determine if the parameter you are talking about (value in this case) has a positive excess return or is more cyclical. With long and not regular cycles, the apparent excess return could be nothing more than the upside of a cycle. I really did believe that value had excess return but am now coming to the realization than Bogle is probably correct & the growth/value difference is largely cyclical and with increasing efficiency over time the excess return we have seen in the past is just the market becoming more efficient, therefore, waiting for a value premium to show up is based on hope more than anything else.
The alternatives I like are based upon cash flows from firms whose economics are tied to contractual or recurring revenues so they can provide bond like diversification versus trading strategies where the underlying securities have exposure to stock risk & are dependent removing that risk by some means whose correlation to stock returns can change over time. Whenever, I think of these strategies my head explodes at the complexity & how hard managing it must be. I put it in the too hard pile very quickly in comparison to my contractual or recurring revenue alternatives.
1) When one looks for example at long-short momentum for US stocks, the track record is just not there.
ten year........... 1997-2016: CAGR = -12%
twenty year.......1987-2016: CAGR = -2.2%
thirty year.........1977-2016: CAGR = +1.7%
forty year..........1967-2016: CAGR = 4.5%
since the standalone returns are so meagre and the tails so fat, if one wants to persist in belief in momentum one needs to fall back on "correlation benefit" type arguments. the trouble with this is that "investing is not physics"- the correlations are not stable as you point out above.
I'd rather invest in something that has strong evidence for a positive expected return on its own rather than hope that future correlations turn out as they did in the past.
2) Fama and French don't believe in the value premium any more either.
in their latest modelhttps://www.forbes.com/sites/phildemuth ... 561ece1138
its been replaced by profitability and "low investment". i.e. companies with high gross profitability = (revenue - "cost of goods sold")/assets have higher expected stock returns as do companies with low investment (i.e. they are not upgrading their factories all the time i guess). when those two factors are added in the value factor is no longer statistically significant.
their new factors sound related to your notion of contractual or recurring revenues. what morningstar calls "wide moats" i guess. e.g. companies that are in the position to achieve high gross profits and don't have to invest a lot in their businesses (cause they are insulated from their competition by those wide moats i guess)
"...people always live for ever when there is any annuity to be paid them"- Jane Austen