Wiki Wiki Web

ValuFocus Investing

Rawley Thomas, 2010

Book is an introduction to a now defunct service selling lifecycle valuation to retail investors.

Refers to Bart Madden's concept of Shareholder Value Review. Explains that casflow matter more than earnings that are subject to distortions and manipulations.

Main tenets are that a valuation model should have 4 properties

  • robustness
  • accuracy
  • non-bias
  • predictive power

Review of other methods:

  • analysts cannot predict cashflows far into the future
  • dividends cannot be used to value most of the stock universe
  • 8x ebitda model is simple, but does not consider capex and interest charge
  • earnings cannot be predicted beyond 2 years
  • price level from regression analysis like Feltham Ohlson 1995 are not causal enough
  • FCF models can be used but require a terminal value assumption
  • EVA or CFROI are lauded by the author, though he believes his approach to be fit for the purpose.

Author refers to more advanced articles in the Valuation Handbook he edited and co-authored.

He also refers to Mandelbrot research on fat tail risk, noting that

  • the 20% most undervalued have low risk and tend to regress to the mean
  • the 20% most overvalued are low risk shorts, which tend to regress to the mean, making 130-30 long/short approach possible
  • the 60% fair valued stocks have very high risk with a stable distribution with an alpha below 2, pointing at infinite variance

The infinite L2 norm results in choosing the L1 norm (max abs) of the log price as a metric of choice.

The discount rate used is the same for all companies and does not depend on market beta (using beta is taught to CFA students but does not seem to improve statistical estimation).

The growth evolution is modelled as having a single exponential fade parameter and cashflows are summed up to 50 years. Fade pattern is different for small, medium and large firm.

The book explicits what the Cashflow Economic Return is not, but does not define what it is. It seems that the author wishes this information to remain proprietary.