The growth-optimality justification for using logarithmic utility is not new. It goes back at least to the 1950's and was discussed by prominent economists. See

The growth-optimality justification for using logarithmic utility is not new. It goes back at least to the 1950's and was discussed by prominent economists. See for example http://finance.martinsewell.com/money-management/Markowitz1976.pdf http://finance.martinsewell.com/money-management/Markowitz19...