Advertisement

You will be redirected to the page you want to view in  seconds.

CBO fails to resolve federal pay impasse

Feb. 5, 2012 - 06:00AM   |  
By HOWARD RISHER   |   Comments

It would be interesting to know how many people have read and understood the report from the Congressional Budget Office (CBO) that compared federal and private-sector pay. The message federal total compensation exceeds private-sector levels is simple, but the methodology is incomprehensible to all but the few who were forced to endure courses in multivariate statistics.

By any standard, it was carefully considered the product of 17 current or former members of the CBO staff who collaborated with 12 reviewers from academia or other organizations.

But despite all the brain power that went into it, the report failed to produce any practical answers. It is technically sound, but the broad-brush conclusions cannot be translated into logical changes in the federal pay program. Moreover, the analysis does not tell us which jobs are overpaid or underpaid with any certainty.

One thing is clear: Across-the-board pay cuts or freezes may reduce the deficit, but they don't constitute a sound human capital strategy. Federal salaries for highly educated workers, according to the report, are already below market levels. As the economy improves, it will be increasingly difficult to recruit new graduates in scientific and technical fields if those salaries fall further behind.

The CBO's study is best described as academic no corporation would consider doing a similar analysis. In the business world, analyses of market pay levels are based on so-called "benchmark" salary surveys. Participants match their jobs to briefly described jobs (e.g., staff nurse, librarian). There are well over 1,000 surveys of that type conducted annually across the country.

Another flaw in the CBO pay-comparison study is its reliance on the Current Population Survey (CPS) database, which is planned as a sample of the U.S. population. It was clearly never intended to be used in analysis of federal pay levels. For example, the CPS database lacks information needed to confirm its jobs are comparable to federal jobs. To its credit, CBO focused on larger companies where, like federal agencies, jobs tend to be more specialized and where pay and benefits tend to be higher; but in those cases, only the occupation of jobs is designated, so it is unknown whether useful pay comparisons are being made. CBO acknowledges that government is not competing with the millions of mom-and-pop businesses, and it correctly excluded that data from the comparison.

Also, CPS data includes industry data that is of little relevance to federal salary planning: Construction, retail, information (i.e., newspapers, book publishers) and restaurants are examples. There are industries included in the CPS where agencies have a regulatory responsibility but the jobs are very different (e.g., banking, mining and transportation).

The same criticism applies to occupations. CBO used 24 broad occupations that may appear reasonable, but it groups diverse occupations like economists and sociologists. The two are both professionals and social scientists, but they are paid very differently. Economists in the Washington area earn roughly 30 percent more than sociologists. Employers need to know the market pay levels for specific jobs.

The likely reason Sen. Jeff Sessions, R-Ala., requested the study is because the gap analyses published each year by the Federal Salary Council have lost their credibility. Those analyses are also an incomprehensible black box.

It would be difficult to find anyone in the federal community who is convinced of the council's finding that General Schedule (GS) salaries are 26.3 percent below the private sector. Even President Obama was reported by The Washington Post in 2010 as saying that his team examined pay and found that highly skilled workers in government are slightly underpaid, while lower-skilled workers are slightly overpaid, relative to the private sector.

Federal salaries very clearly need to be justified. The analyses need to be logical and understandable. It is possible. The data and the methods are readily available. The GS system needs to be replaced.

Howard Risher is a consultant and writer on federal pay and performance issues. He was the managing consultant for the studies leading to the 1990 Federal Employees Pay Comparability Act and is author of "Planning Wage and Salary Programs."

More In Advice & Opinion

More Headlines