My post yesterday on federal worker pay generated a large and aggressive response from federal workers, both in my inbox and on websites such as Fed​smith​.com. (See also Federal Times and Govexec). Here are four points raised in criticism:


First, people accuse me of producing distorted data somehow. Actually, it’s essentially just raw Bureau of Economic Analysis data, but the data is usually overlooked by the media because I don’t think the BEA puts out a press release on it. Anyway, the average wage data is from BEA Table 6.6D. The average compensation data is simply total compensation (Table 6.2D) divided by the number of workers (Table 6.5D).


Second, people argue that reporting overall averages for wages and compensation is somehow illegitimate. People email me comments like “my federal salary is only $50,000, yet you claim that federal workers make $79,000.” All I can say to folks like this is that there must be a federal worker out there making $108,000 who balances you off.


Third, people argue that a better analysis would be to compare similar jobs in the private and public sectors, rather than looking at overall averages. I agree that that would be very useful. Unfortunately, the BEA data is not broken down that way. At the same time, the BEA data provides the most comprehensive accounting for the value of employee benefits of any data source. Benefits are a very important part of federal compensation, and so that’s why I look to the BEA data.


Fourth, many people argue that the federal government has an elite workforce with many highly educated people. Certainly, that’s an important factor to consider. However, that is the reason why I focused on the pay trend over the last eight years. The federal worker compensation advantage rose from 66 percent in 2000 to 100 percent in 2008. Has the composition of the federal workforce really changed that much in just eight years to justify such a big relative gain? I doubt it.


A final consideration is to look at a “market test” of the adequacy of compensation in the public sector–the quit rate. The voluntary quit rate in the federal government is just one‐​third or less the quit rate in the private sector (Table 16 near the bottom here).


That is strongly suggestive of “golden handcuffs” in federal employment. While many federal workers probably grumble about their jobs (as many private sector workers do), they know that the overall package of wages, benefits, and extreme job security (Table 18 here) is very hard to match in the competitive private market, and so they stay put.