The US Surge Revisited: What's the evidence for decreased casualties after one year of troop deployment?

In September last year we published a critique of the available data on the impact of the US 'surge' in Iraq. Comparing the figures for US coalition casualties with the data on attacks presented by General Petraeus to the US congress it was evident that things just did not add up. At that stage, in mid September 2007, US casualties had not declined and there was no evidence of a decrease in civilian casualties either. But now, after more than a full year of data has become available on the 'surge' (Feb 2007 - Mar 2008) what is the evidence for an improvement in the casualty burden?

Our updated graph of US Coalition fatalities (non-Iraqi troops only) shows that from mid September 2007 onwards daily fatalities have, in fact, markedly declined. In February 2008 an average of 1.03 deaths occurred per day compared to 3.04 per day in Feb 2007. For March, these figures were 1.26 and 2.65 respectively. This does suggest that conflict levels have fallen substantially. However, it should be noted that some of the decrease in foreign troop deaths may have been due to an increase in the number of active Iraqi units fighting within the US Coalition.

Figures for civilian fatalities are much harder to assess due to the substantial challenges faced in collecting reliable data. We are planning to publish a detailed review of this soon. However, the available trend data does indicate a significant fall. Good news it appears. A major remaining policy issue of course is to what extent the 'US surge' can account for these apparent improvements. The role of the Mahdi Army cease fire and the 'Anbar Awakening' are widely acknowledged to have been critical factors. Future political developments may reveal to what extent these have been responsible.

Data updated 21.04.08