TCG believes the client deserves the truth. We do not pretend to be infallible, but we promise always to "tell it like it is," and to share the full extent of our insight and counsel.
TCG has been among the pioneers in the field of "relevant benchmarks." We believe that the most accurate evaluation of an investment manager's "value added" can be achieved by comparing his performance with a "relevant benchmark" appropriate for the style in which the portfolio is being managed.
We believe that the use of monthly return data, rather than quarterly, is vitally important. Quarterly data over a five-year time period, for example, provides only 20 observations, not adequate for accurately determining correlation to the benchmark, beta, standard deviation, etc. With monthly data, there are 60 observations over the same five year period, allowing meaningful analyses based upon running 24, 30 or 36 month time periods.
One standard deviation is a partial measure of short-term volatility; but it excludes the 1/3 of returns that are the most extreme. Those extremes are often the returns of most interest. Standard deviation does not distinguish between upside and downside volatility. For proper analysis of a manager's performance, it is essential to show "upside capture" and "downside participation" separately.
Cumulative, quartile, floating-bar charts allow recent performance to impact long-term results. This tempts fund sponsors to fire low and hire high (that is, to hire the manager with what appears to be the best long-term and short-term experience). Comparisons with peer groups for running n-month time periods are superior. In so doing, the membership of each peer group universe must be carefully built to be sure that each member has had high correlation with the relevant benchmark.
We use
both returns-based style
analysis and holdings-based analysis. Both are excellent tools when used separately but are
especially productive when integrated into a side-by-side approach for manager
evaluation.
The "Information Ratio" and the "Sharpe Ratio", based on a relevant benchmark, become more helpful evaluation tools when used on a running 24, 30, or 36 month basis.
