Read on Twitter
1/ Market cap weighting is the approach to use if you don't have an edge or a hypothesis (and its OK not to have one).
However, if you have an edge or a hypothesis, then you should deviate from market cap in proportion to your edge and be humble about what that edge might be.
However, if you have an edge or a hypothesis, then you should deviate from market cap in proportion to your edge and be humble about what that edge might be.
2/ An edge can be as simple as the belief that p/b is a broken value signal.
https://www.osam.com/pdfs/research/_19_Commentary_Price-to-BooksGrowingBlind%20Spot_Nov-2016.pdf
https://www.osam.com/Commentary/negative-equity-veiled-value-and-the-erosion-of-price-to-book
https://www.twosigma.com/insights/article/diagnosing-the-recent-decade-of-drawdown-in-value/
https://www.osam.com/pdfs/research/_19_Commentary_Price-to-BooksGrowingBlind%20Spot_Nov-2016.pdf
https://www.osam.com/Commentary/negative-equity-veiled-value-and-the-erosion-of-price-to-book
https://www.twosigma.com/insights/article/diagnosing-the-recent-decade-of-drawdown-in-value/
3/ if you believe that, then look for a fund that looks cheap on P/E, P/CF or other value metrics but looks expensive on P/B.
Then look to size this holding according to how active it is versus the benchmark and your conviction against P/B.
Then look to size this holding according to how active it is versus the benchmark and your conviction against P/B.
4/ An active view on the efficacy of a signal is much better than an active view on a particular stocks as it has more breadth.
Alpha = edge * sqrt(breadth)
A small edge over a large breadth can be a much more powerful source of alpha then one might think.
Alpha = edge * sqrt(breadth)
A small edge over a large breadth can be a much more powerful source of alpha then one might think.
5/ Another active view might be the ability to access investments outside of a benchmark. This is harder for US large cap stocks (hi Tesla fans!) but might be easier in fixed income (hi fans of non-agency mortgages).
6/ If you have a hypothesis and are deviating from market cap, then you also have something that should be observable and testable if it is true or not.
This should serve as a guide to help determine if you are right or wrong and/or stick with the deviation.
This should serve as a guide to help determine if you are right or wrong and/or stick with the deviation.
7/ For example, you can show that P/B has theoretical problems (observable) and does better when value does poorly (also an indication it captures value poorly).
You also are able to observe who is willing to lose money on the view - anyone who benchmarks value as P/B
You also are able to observe who is willing to lose money on the view - anyone who benchmarks value as P/B
8/ This gives you the knowledge of who the willing loser in the trade is and when they may be gone (when not much money is benchmarked to P/B anymore).
9/ For holdings that aren't included in benchmarks, it might be that you measure the potential return and diversification and simply view the investment through the perspective of information ratio and tracking risk. This might be easy for fixed income and harder for stocks.
fin/ Hopefully a helpful perspective to take when tilting away from market cap weights, what to watch and when to close down the tilt.
