Evaluating Reports & Surveys: The 3.8 Billion Dollar Question Mark

The New York State Department of Financial Services issued a report on October 17th regarding the state’s public retirement systems. Specifically, the report reviewed the “Common Retirement Fund” from March 31, 2009 to March 31, 2016. That is, the two state pensions over the last seven years. Unique compared to other state pensions, in New York they are run solely by the State Comptroller. The report found that “under the Comptroller’s watch the State pension system has spent large amounts of pension system funds chasing returns and performance that has fallen far short for years. Specifically, over the past eight years, the System has paid over $1 billion in excess fees to hedge fund managers who underperformed to the tune of $2.8 billion.” The report makes several very valid points, and you can read it in its entirety here. Reserving judgement on the overall situation, one major red flag jumped out at me: the dates of the report. As a general rule of thumb, it is never a good sign when a date range starts or ends right before or after a significant crash. This report starts March 31, 2009. The exact market bottom was March 9, 2009; 22 days earlier. That is like conducting a report on the most successful teams in NBA history and starting the report Michael Jordan’s rookie year. It doesn’t make the facts any less valid; however, it frames the situation in a certain way that may distort the picture. If the New York State Department of Financial Services wanted to analyze the Comptroller’s decision to invest in hedge funds, why not look at the entire time these investment were held? Why cherry pick a date close to the market bottom of one of the greatest market declines in U.S. history?

This is what you need to know - whenever a financial report has a starting or ending point right before or after a significant market event, the authors are almost always doing so, intentionally or otherwise, to frame their argument in support of their message. Regardless of the content of the message, it is something to be cognizant of as affecting the conclusions of the report.