In contrast to the analysis of Lehman skeptics such as John Taylor (2008, 2009) and John Cochrane and Luigi Zingales (2009), the evidence we present supports the view of many practitioners that the decision not to rescue Lehman represented an immediate and massive shock to the financial system that was larger by an order of magnitude than anything seen over nearly two decades.
The evidence Sterling presents focuses on an index of daily financial conditions created by Michael Rosenberg for Bloomberg. The index drifted down in in the summer of 2008, then fell sharply from Lehman weekend until October 10, and then began a recovery. It returned to pre-Lehman levels earlier this year. The biggest three-day drop in the index occurred post-Lehman, which includes Reserve Primary "breaking the buck."
Sterling also produces evidence that the index of financial conditions is predictive of economic activity, which would show a connection between Wall Street and Main Street.
If you look at Sterling's analysis of how the market reacts to policy, it seems to me that general policy moves, such as providing protection for money market fund deposits, have more positive effect than bailouts. It would be interesting to go back and look at the Bear, Stearns bailout or the AIG bailout and see if they produced any positive result before arguing that a bailout of Lehman would have been a good idea.