'Scandal's Statistical Companion - Quantitative Efficiency in $91 Billion of 10b-18 Corporate Stock Buybacks - Performance vs. Efficiency

Abstract

The typical NASDAQ company in a small sample that executed $91 billion of stock buybacks in the 2004 - 2006 period (about 10% of all repurchases) paid 1.52% in excess costs relative to a strategy of investing a constant dollar amount at VWAP (volume weighted average price) every month. Results company by company did not cluster particularly close to the expected 0.00% efficiency cost: 63% of sampled companies suffered efficiency losses on average of 4.05% apiece relative to the naive strategy, and their excess trading losses totaled $1.031 billion; 37% of companies outperformed the naive strategy on average by 2.70% apiece, and their excess trading gains totaled $565 million. A large proportion of companies at periods of peak activity accounted for in excess of 10% of monthly trading volume. One company's peak trading accounted for in excess of 15% of monthly volume, and that company also reported the best efficiency in its volume and timing decisions by beating the naive strategy by 4.63% over the period. A highly tentative look at performance suggests that, in addition to poor execution efficiency, corporations' buybacks may not be generating reasonable performance returns.