The January 2013 Special Inspector General for Afghanistan Reconstruction report has been out for some time now, but @SIGARHQ has still been tweeting about it regularly. One of their tweets yesterday brought my attention to the section of their report (pdf) where they discuss force size for Afghan National Security Forces. Since the interruption in training brought about by decreased interactions between US and Afghan forces during the  massive outbreak of green on blue attacks, I have maintained that the claim of 352,000 for ANSF force size was no longer credible. It appears that my skepticism is well-founded, as the pertinent section of the SIGAR report bears this heading:

ANSF NUMBERS NOT VALIDATED

The section begins:

Determining ANSF strength is fraught with challenges. U.S. and coalition forces rely on the Afghan forces to report their own personnel strength numbers. Moreover, the Combined Security Transition Command-Afghanistan (CSTC-A) noted that, in the case of the Afghan National Army, there is “no viable method of validating [their] personnel numbers.” SIGAR will continue to follow this issue to determine whether U.S. financial support to the ANSF is based on accurately reported personnel numbers.

There are several important bits to unpack in that paragraph. First, note that even though the US (well, officially, NATO) is training the Afghan forces, it is the Afghans themselves who report on their force size. It appears that our training of the Afghans, however, has not trained them on how to count personnel in a way that can be validated. But the end of the paragraph is the kicker, because it appears that our financial support of the Afghans is based on their own reporting of the force size. Since we are paying them for the force size they report, why wouldn’t they inflate the numbers to get paid as much as possible? The Afghans know that the bulk of US policy is built around the 352,000 force size myth, so they know that there will be absolutely no push-back (aside from an obscure SIGAR report that only DFH’s will read) for inflating the number to get the result the US desires. For further enticement, recall that NATO has proposed extending the time over which a force size of 352,000 will be supported, in a move that I saw as a blatant attempt to dangle an additional $22 billion ready for embezzling in front of Afghan administrators.

It comes as no small surprise, then, that SIGAR has found that the Afghan-reported numbers somehow manage to include over 11,000 civilians in the reports for security force size that is specifically meant to exclude civilian personnel.

A related area in which SIGAR has found a disgusting level of dishonesty is in how the US goes about evaluating Afghan forces in terms of readiness. Because it became clear to the trainers in 2010 that they had no hope of achieving the trained and independent force size numbers that NATO planners wanted (and because SIGAR found that the tool they were using at the time was useless), they decided that the only way to demonstrate sufficient progress was to redefine the criteria for evaluating progress. From the report:

In 2010, SIGAR audited the previous assessment tool—the Capability Milestone (CM) rating system which had been in use since 2005—and found that it did not provide reliable or consistent assessments of ANSF capabilities. During the course of that audit, DoD and NATO began using a new system, the CUAT [Commander’s Unit Assessment Tool], to rate the ANSF. In May 2010, the ISAF Joint Command (IJC) issued an order to implement the new system which would “provide users the specific rating criteria for each [ANSF] element to be reported by the CUAT including leader/commander considerations, operations conducted, intelligence gathering capability, logistics and sustainment, equipping, partnering, personnel readiness, maintenance, communications, unit training and individual education, as well as the partner unit or advisor team’s overall assessment.”

Since the implementation of the CUAT, the titles of the various rating levels have changed, as shown in Table 3.3. In July 2012, the Government Accountability Office (GAO) raised concerns that the change of the title of the highest rating level from “independent” to “independent with advisors” was, in part, responsible for an increase in the number of ANSF units rated at the highest level. GAO also noted that “the change lowered the standard for unit personnel and equipment levels from ‘not less than 85’ to ‘not less than 75’ percent of authorized levels.” In a response to SIGAR last quarter, the IJC disagreed with GAO’s assessment, saying a change in title does not “equal a change in definition.” Since last quarter, the IJC has initiated a CUAT Refinement Working Group to standardize inputs and outputs in the areas covered by the assessments.

SIGAR noted the evolution of assessment levels and provided this handy table for comparing the various ratings levels that were used:

But, when SIGAR drilled down to the definitions used for the current classification levels, we get to the bit that they tweeted yesterday and caught my eye:

Established: The unit is beginning to organize but is barely capable of planning, synchronizing, directing, or reporting operations and status, even with the presence and assistance of a partner unit. The unit is barely able to coordinate and communicate with other units. Leadership and staff may not adhere to a code of conduct or may not be loyal to the Afghan government. Most of the unit’s enablers are not present or are barely effective. Those enablers provide little or no support to the unit. Coalition forces provide most of the support.

Just wow. “Leadership and staff my not adhere to a code of conduct or may not be loyal to the Afghan government.” And how many units are in the category where their loyalty is openly questioned? As of December, 2012, the report indicated only one Afghan National Army unit and three Afghan National Police units at the “established” level where they may not be loyal to the Afghan government. However, there also were 81 ANA and 301 ANP units that were not assessed, so we are left to wonder if many of those units were not assessed primarily because they would fall into this most embarrassing level.