US headache over Afghan deserters
By Gareth Porter
WASHINGTON - One in every four combat soldiers quit the Afghan National Army
(ANA) during the year ending in September, published data by the US Defense
Department and the Inspector General for Reconstruction in Afghanistan reveals.
That high rate of turnover in the ANA, driven by extremely high rates of
desertion, spells trouble for the strategy that US President Barack Obama has
reportedly decided on, which is said to include the dispatch of thousands of
additional US military trainers to rapidly increase the size of the ANA.
US officials have for years touted the ANA as a success story. General Stanley
A McChrystal, the top US soldier in Afghanistan, called in his August 2009
strategy paper for increasing the ANA
to 134,000 troops by October 2010 and eventually to 240,000.
But an administration source, who insisted on speaking without attribution
because of the sensitivity of the subject, confirmed to Inter Press Service
(IPS) that 25% has been used as the turnover rate for the ANA in internal
discussions, and that it is regarded by some officials as a serious problem.
The 35,000 troops recruited in the year ending September 1 is the highest by
the ANA in any year thus far, but the net increase of 19,000 troops for the
year is 33% less than the 26,000 net increases during both of the previous two
Those figures indicate that the rate of turnover in the ANA is accelerating
rather than slowing down. That acceleration could increase further, as the
number of troops whose three-year enlistment contracts end rises rapidly in the
next couple of years.
Meanwhile, the Defense Department (DoD) sought to obscure the problem of the
high ANA turnover rate in its reports to the US Congress on Afghanistan in
January and June 2009, which avoided the issues of attrition and desertion
Instead, they referred to what the DoD calls the "AWOL" [absent without leave]
rate in the ANA, which measures those unavailable for duty but still in the
army. It claimed in June that the AWOL rate was 9% through May 2009, compared
with 7% in 2008.
The reports also confused the question of turnover in the ANA by using
questionable accounting methods in the DoD's reporting on monthly changes in
personnel. It provided figures for total ANA personnel in 2009 showing an
increase from 66,000 in September 2008 to 94,000 in September 2009.
Those figures made it appear that ANA manpower increased by 28,000 during the
year. But nearly half the increase turns out to be accounted for by a decision
on the part of the US command responsible for tracking ANA manpower to change
what was being measured.
Previously, the total had included only those who had been trained and assigned
to a military unit. But in late September 2008, the Combined Security
Transition Command-Afghanistan (CSTC-A) started counting 12,000 men who had not
previously been considered as part of the ANA.
In response to a query from IPS, Sergeant Grady L Epperly, chief of media
relations for CSTC-A, acknowledged that the US command had abruptly changed
what it included in its overall strength figures for the Afghan army in late
"The way numbers were reported was switched from reporting only operational
forces to including all soldiers, officers and civilians, regardless of
training status and command," Epperly wrote in an e-mail.
The graphs in the DoD reports of January and June 2009 are still identified as
"Afghan National Army Trained and Assigned". But the text of the report reveals
that the personnel totals shown on the graph were no longer for the ANA but for
the Ministry of Defense.
That meant that the totals included for the first time those still in training,
including even high school cadets, and others not assigned to any unit.
That deceptive accounting change obscured the fact that the total number of
personnel assigned to ANA units in September 2009 was actually 82,000 rather
than the 94,000 shown, and that the increase in ANA personnel over the year was
only 16,000 rather than 28,000.
Using the corrected totals for changes in personnel during the year, the 25%
turnover rate for ANA combat troops can be calculated from the available data
on recruitment and the breakdown between combat and non-combat troops. 
ANA turnover as a proportion of ANA combat troops is a more significant
indicator of instability than turnover as a proportion of all personnel,
because there is little or no desertion and far higher reenlistment rates in
non-combat jobs. ANA non-combat personnel totals also include thousands of
The impact of the 25% combat troop turnover rate on the ANA is actually more
acute than it would appear, because of the high absenteeism rate in the ANA. A
Government Accountability Office report revealed that, as of February 2008, out
of 32,000 combat troops on the rolls, only 26,000 were available for duty - a
19% absenteeism rate.
Assuming that same rate of absenteeism remained during the past year, the
number of ANA combat troops actually available for duty increased only by about
9,000 from 37,000 to 46,000.
As serious as the turnover rate was in 2009-2009, turnover in the first two or
three years of the ANA was much worse. ANA recruitment and reenlistment figures
show that 18,000 of the first 25,000 troops recruited from 2003 to 2005
That desertion rate prompted analysts at the US Army Center for Lessons Learned
at Fort Leavenworth, Kansas to conclude that the ANA would not be able to grow
beyond 100,000, according to an article in the current issue of Military
Review, published at the same army base.
The authors, Chris Mason and Thomas Johnson, both of whom have had extensive
experience in Afghanistan, write that that the analysts at the Army Center
concluded that by the time the ANA got to 100,000 troops, its annual losses
from desertions and attrition would roughly equal its gains from recruitment.
The Center for Lessons Learned refused to confirm or deny those assertions.
When asked about the assertion in the Military Review article, an official of
the Center for Lessons Learned, operations officer Randy Cole, refused to
comment except to refer IPS to the authors of the article.
1. The turnover rate in any organization in a given time period is the total
number of personnel who quit the organization divided by the total number who
belonged to the organization during that period.
The ANA recruited 35,000 men from September 2008 through August 2009, according
to quarterly reports issued by the Inspector General for Reconstruction in
Afghanistan and semi-annual DoD reports. With 66,000 as the personnel base for
the year beginning September 2008, the total number of personnel in the
organisation for the year was 101,000.
The difference of 19,000 between the 35,000 recruited and the 16,000 net
increase in personnel during the period represents total turnover from a
combination of attrition - soldiers who do not reenlist after their three-year
contracts have expired - and desertion.
The 19,000 turnover is 19% of the total of 101,000 men who belonged to the ANA
during the year ending September 2009.
However, the more meaningful measure of turnover is the percentage of combat
troops who left the ANA.
The total number of combat troops increased only from 46,000 to 58,000 during
the year ending in September for an increase of 12,000, according to the
official published data.
Four thousand of the new 35,000 new recruits either went into non-combat units
or were not assigned, leaving 31,000 recruits who were assigned to combat
The difference between the 31,000 recruits assigned to combat units and the
12,000 increase in combat troops, representing the turnover of ANA combat
troops, is 19,000. That 19,000-man total turnover was 25% of the 77,000 total
ANA troops assigned to combat units during the year (46,000 plus 31,000).
Gareth Porter is an investigative historian and journalist specializing
in US national security policy. The paperback edition of his latest book,
Perils of Dominance: Imbalance of Power and the Road to War in Vietnam, was
published in 2006.