Sometimes the honesty is just too much, POGO 211

I was reviewing the 2011 study Bad Business: Billions of Taxpayer Dollars Wasted on Hiring Contractors from the Project On Government Oversight (POGO) and I came across this little gem of a 'review' AFL-CIO Struggles With Reading Comprehension by Geoffrey Lawrence.

It quotes sections of the study and gives them a 'proper' critical review, recasting the results in the light of being (1) completely unreliable because the study itself says so and (2) showing exactly the opposite, that government workers are overpaid compared to their counterparts.

Wow, if you just read that page and didn't actually bother to read the study you would be strongly inclined to side with Geoffrey. What a bunch of dishonest folks over at the AFL-CIO!

Just look at his evidence:

The authors explicitly admit:

POGO is aware that its methodology does not incorporate some governmental cost factors: i.e., non-directly associated overhead (e.g., executive management and administration, information technology, and legal support), material and supplies (e.g., insurance and maintenance), or facilities (e.g., depreciation, rent, insurance, maintenance and repair, utilities, capital improvements).
In other words, POGO admits that it is impossible, based on available data, to compare federal employee compensation to contract employees’ compensation and so they instead compare only federal labor costs to the total expenses faced by contractors.

That’s as apples-to-oranges as it gets.

What a shockingly POOR study! Bring out the torches and pitchforks!

Before we burn anyone... let's just look that up in the study:

POGO is aware that its methodology does not incorporate some governmental cost factors: i.e., non-directly associated overhead (e.g., executive management and administration, information technology, and legal support), material and supplies (e.g., insurance and maintenance), or facilities (e.g., depreciation, rent, insurance, maintenance and repair, utilities, capital improvements).[70] However, given the fact that POGO relied exclusively on GSA’s listed contractor billing rates for performance at government sites, many of those cost factors would essentially be canceled out.[71] In fact, when contractors perform work at contractor sites, POGO found that contractor billing rates were, on average, 15 percent higher than rates for work performed at government sites. In addition, POGO did not include in its comparative analysis additional costs that the government incurs as a result of outsourcing services to contractors. Those costs would only add to the costs associated with outsourcing documented in this report

Oh... my... it seems Geoffrey did a little quote mining there. I'm 'sure' that was 'just' an accident.

Now, I have no doubt that the folks at the AFL-CIO were willing to overlook these limitations when they decided to trumpet POGO’s report, given that their mission is far more political than academic. However, they might have thought otherwise had they paid more careful attention to the reports’ findings. Specifically, the authors acknowledge in a passing statement:

POGO confirms the results of studies that compared the public with the private sector, finding that federal employees generally make approximately 20 percent more in salary and full compensation than do their counterparts in the private sector.

Oh…so the report actually finds the exact opposite of what the AFL-CIO is claiming! Federal employees are overcompensated relative to their peers performing similar jobs in the private sector…which is also to say that there is a labor cost savings from federal outsourcing. Huh.

Damn, Geoffrey is on fire! I'll sharpen my pitchfork immediately! Wait... let's just double check this...

POGO confirms the results of studies that compared the public with the private sector, finding that federal employees generally make approximately 20 percent more in salary and full compensation than do their counterparts in the private sector. While POGO’s investigation adds some credibility to the government and private sector comparison studies, POGO has some concerns about those other studies.

For example, USA Today analyzed a sample of 40 occupations using BLS data for 2008 and found that the typical federal employee is paid 20 percent more than a private sector employee in the same occupation.[177] The study found that “federal employees earn higher average salaries than private-sector workers in more than eight out of 10 occupations.” USA Today used median salaries and did not adjust for any other cost factors, including health care and pension benefits. That study was refuted by National Treasury Employees Union President Colleen Kelley, who stated it “compares apples and oranges” because certain government work “has more complexity and requires more skill than…work in the private sector.”[178]

A similar opinion has been stated by former OMB director Peter Orszag:

…the truth is that a comparison of federal and private-sector pay, even by occupation, is misleading because the employees hired by the federal government often have higher levels of education than their counterparts in the private sector—even within the same occupations. When you factor in the education and experience of the federal workforce, there is no statistically significant difference in average pay levels.[179]

Now I'm just disappointed.

What I cannot figure out is, this gross mischaracterization by quote mining done out of ignorance or malice.

Surely a champion of the Truth and Reading Comprehension such as Geoffrey could have managed to read the very next sentences following his selected quotes?

Conclusion
Contrary to popular belief, many government services are not performed by federal employees, but by contractors. The government spends hundreds of billions of dollars annually on services—in fact, approximately one-quarter of all discretionary spending now goes to service contractors[209]—and POGO’s analysis found these contracts may be costing taxpayers, on average, 1.83 times more than if federal employees had done the work. In order to reduce those excessive costs, a government-wide system to conduct federal employee versus service contract cost analyses needs to be created. Instead of directly hiring service contractors without considering hiring federal employees—perhaps from a newly created pool of part-time or temporary federal employees—conducting cost reviews at the start of the process would allow the government to save billions of dollars annually.