Last Friday, Rod Nordland published an article in the New York Times alleging that Afghan security forces (Afghan National Army (ANA) and police) had suffered an average of 57 killed in action (KIA) per day during the previous week, up from 22 killed per day in 2016. If true, such reports would indicate a dramatic increase in loss rates over the previous years.
These reported figures should be regarded critically, however. It is not clear how Nordland arrived at the total of 22 KIA per day for 2016. His article cited another article by Thomas Gibbons-Neff, published in the Times on 30 October 2017. This reported Afghan security forces casualties for 2016 at 6,700 KIA and 12,000 wounded in action (WIA), which works out to an average of 18.36 KIA per day (6,700/365), not 22. The total number of KIA + WIA works out to an average of 51.23 per day (18,700/365).
The lede of Gibbons-Neff’s 2017 article was that the U.S. and Afghan governments had stopped providing official strength and loss figures for the Afghan security forces. Citing the last report of the U.S. Special Inspector General for Afghanistan Reconstruction (SIGAR), Gibbons-Neff reported Afghan security forces losses from 1 January-8 May 2017 (126 days) as 2,531 KIA and 4,238 WIA. This works out to an average of 20.08 KIA (2,531/126) and 53.72 KIA + WIA (6,679/126) per day.
Nordland arrived at the figure of 57 KIA per day based on a report of 400 Afghan security forces killed in the week leading up the publication of his article on 21 September 2018. He calculated it by averaging the total over the previous seven days (400/7). Casualty rates in combat have been highly variable, historically. Brief spikes in rates are common. In the same paragraph reporting the 400 KIA total, Nordland quoted senior Afghan government officials stating that the daily average for recent months had been 30 to 40 KIA per day.
It is possible to use the figures cited by Nordland and Gibbons-Neff to make ballpark estimates for Afghan security forces casualties in 2017 and 2018. Even if the weekly loss of 400 KIA for 14-20 September 2018 represents an abnormally high total, it is reasonable to conclude that the Afghan security forces are very likely incurring sharply higher combat losses in 2018 than the previous two years. These figures do not include counts of missing or captured and thus underestimate the actual numbers of battle casualties being suffered by the Afghan forces. It is also possible that the estimates of 30-40 KIA per day apply only to the peak spring-to-autumn fighting season, which would somewhat reduce the overall 2018 KIA and WIA totals.
As Nordland reported, these losses are resulting in an increasing strain on the Afghan forces. His article stated that the strength of the ANA and police in April 2018 was 314,000, 38,000 below the authorized total of 352,000, and that the actual total was likely even lower due to fraudulent reporting and unreported desertions. The ANA suffered a monthly attrition rate of 2.9 percent in early 2017 from combat casualties, desertion, and failed reenlistments, requiring one-third of the overall force to be replaced by new recruits annually. That attrition rate is undoubtedly far higher now and almost certainly not sustainable for long.
In comparison, the Afghan government reported in August that its security forces had killed 42 Taliban fighters per day, or 1,300 per month. For the year ending in March 2018, it claimed to have killed 13,600 insurgent fighters. There has been no independent confirmation of these claims and they should be treated skeptically.
John Conger recently reported in Defense One
