The following is in response to Adam Kanzer’s letter that appeared last week. That was in response to several letters to the editor that appeared June 11. Those letters were in response to Adam Kanzer’s article, Exposing False Claims about Socially Responsible Investing, which appeared June 4. Kanzer’s article was in response to Adam Apt’s article, Measuring the Cost of Socially Responsible Investing, which appeared May 21.
Dear Editor,
In reading Adam Kanzer’s letter, I had the thought that I should have reworded Mark Kritzman’s and Timothy Adler’s proposition:
Successful active management – that is, successful stock picking – produces an excess return. Unsuccessful stock picking does not. The cost of SRI of which Kritzman and Adler speak comes out of that excess return. No excess return from stock picking, no cost. Small excess return, small cost (as a percentage of the entire portfolio).
And over what benchmark would this putative excess return be earned? Plainly, for Kritzman and Adler, it must be a benchmark that encompasses more than just socially responsible companies, or they couldn’t even begin to frame their argument. But that makes sense, because the analytical skills required for successful stock selection draw upon economic, accounting, price, and strategic information that is in no way limited by corporate social responsibility. Kritzman and Alder explicitly, in the part of their paper where they attempt to estimate the cost, use the S&P 500 and three broad international and global indices. They could as well have used the Russell 3000.
As to why the cost is incurred (if it is), Kritzman’s explanation in his letter (June 11) cannot be improved upon.
Kanzer, in his letter, continues to draw a distinction between active benchmarks and passive benchmarks.
There is no such distinction. All benchmarks are passive reference measures that characterize the performance of groups of investments, for example, stocks in general, the stocks of large companies, the stocks of small companies, value stocks, growth stocks, or the stocks of socially responsible companies. All require careful design and management. Benchmarks don’t occur naturally, and they aren’t produced by simple mathematical formulas. Once, the S&P 500 contained Royal Dutch Shell. It no longer does, because the managers of that index decided to exclude foreign companies. Once, it contained the stocks of small companies. It no longer does, because the managers decided to make it an index of the stocks of large companies. Its managers have also decided to exclude MLPs.
Kanzer expresses a partiality to the Russell indices. The Russell 3000 was once rebalanced semi-annually. Then its managers decided to rebalance it annually, on the last trading day of June, unless that day is the 29th or 30th. It does not simply average the returns of the 3000 largest companies, because the designers chose to adjust it for float. The Russell Growth and Value indices require judgment as to what criteria differentiate a growth stock from a value stock. Two years ago, Russell made these criteria more complex. Others may disagree with their judgment.
Kritzman’s and Adler’s argument applies also to growth investing and value investing, with one important distinction from SRI: A portfolio manager conceivably may have special expertise in the evaluation of economic, accounting, price, and strategic information that is limited by the corporate characteristics that make for a growth stock or a value stock, and so contributes to success in stock picking. Some growth and value managers may not appreciate this.
Unless I’m much mistaken – and Kanzer himself says he agrees with me – the KLD 400 does not consist of stocks that have been deliberately picked to have greater returns than the S&P 500 or the Russell 3000. It is not an actively managed portfolio. It is a benchmark, passive, but constructed by means of screens that are rather more complex than those used for other indices. Kritzman’s and Adler‘s argument, as they have stated more than once, is unable to say anything about the performance of that index relative to a larger universe.
None of us has yet pointed out that were the KLD 400, for whatever reason, to have a long-run return that exceeded that of the S&P 500 (or the Russell 3000) by a sufficient amount, then Kritzman’s and Adler’s argument concerning the origin of the cost, while valid, might be rendered moot. (In such cases, though, active management, while adding value to the S&P 500, might actually subtract value from the KLD 400.) But this seems an unlikely eventuality for a number of reasons, and Kanzer himself states that he doesn’t argue that the KLD 400 will necessarily outperform a larger universe.
Still, as I said in my previous letter, regardless of the logic of the arguments, I believe that real numbers, if and when we learn them, will show that this is much ado about very little.
Adam Apt
