Are Federal Employees Underpaid?

I think not:

The wages of federal workers are 2 percent higher than similar private-sector workers, on average.

The benefits of federal workers are 48 percent higher than similar private-sector workers, on average.

The total compensation (wages plus benefits) of federal workers is 16 percent higher than similar private-sector workers, on average.

I guess the relevant question is this: are federal services so efficiently provided and in such high demand that federal employees are truly deserving of a 16% wage premium? If not, federal employees are definitely overpaid.
An alternative question is this: would federal employees be paid as well for doing the same job in the private sector? If not, federal employees are overpaid.

Perpetually Parsing Pensions

You know this whole debate over whether to get the pension fund to 50% funding may not matter as much as it may seem.  Or at least the problem may be so much worse than we think and that the numbers we assume are good turn out to be wildly optimistic.

The government accounting folks, GASB, are considering a change in pension accounting rules that would bring public pension accounting more in line with private sector accounting.   The result, if it happens, is that the calculations of almost all public pension liability will be much much higher than they are now.  So Pittsburgh’s billion dollar liability would be something a lot more. What if the number was $2 billion? Any % increase on a billion dollars is real money.  If you want to see more on the accounting issue, see this from the WSJ:  Board at Center of Pension Dispute Note the Pittsburgh connection in the story that has nothing to do with the city btw.

I was reminded of that because I was reading a paper that just came out.  Remember when I compared Pittsburgh’s debt and pension obligations per capita to that in Vallejo, California (which is still plodding through a bankruptcy proceeding) to Pittsburgh. Someone has taken the time to parse those numbers more systematically across a number of cities.

The Crisis in Local Government Pensions in the United States (link)

or if you want crib notes, read the Economist’s coverage of their research.

It does not look like they included Pittsburgh in their research, most likely because the city is so small.  No time to parse all that, but in scanning it I think they calculated some liability valuations that use some more normal assumptions on things like the future discount rate. Public pension plans tend to use a very high discount rate of 8%.  Actuarial valuations of private sector pension plans almost always use a much smaller value for their discount rate.   I suspect that if they did do such a calculation for Pittsburgh would be quite a shock and show a lot higher than the billion dollar liability we are talking about these days.  Just imagine where we would be in that case?

Join the forum discussion on this post - (1) Posts

What Does it Mean When a Million People Apply for a Thousand Jobs?

Several economists have commented on the remarkable and relatively new phenomenon that’s seen in India, where a government agency (or a state owned enterprise) advertises (say) 100 job openings and gets a million applications. This is generally interpreted as a problem, as a reflection of the very high extent of unemployment amongst the educated in India.

At the same time, this is hard to reconcile with the picture one gets from private recruiters, who say that it’s hard to recruit fairly minimal levels of skills when paying the market price.

The metaphor of market efficiency is useful in thinking about this. Suppose there is a liquid market with many buyers and sellers. Suppose supply and demand clear and the price of the widget is Rs.100. Now suppose you step into the market and offer to buy at Rs.101. In an efficient and well functioning market, you should be deluged with a very large number of sellers trying to sell to you at 1% above the fair market price. Conversely, if you step into the market and try to buy the widget at Rs.99. Nobody should be willing to sell to you at this price. A dramatic shift in the number of bids that you get — from zero at Rs.99 to a deluge at Rs.101 — is the hallmark of an efficient market.

I think this is a useful way to think about what is going on with government recruitment. As a thumb rule, researchers like Lant Pritchett and Jeff Hammer believe that in rural India, for junior positions, the government overpays by 3x. Also see Wage differentials between the public and private sectors in India by Elena Glinskaya and Michael Lokshin, in Journal of International Development, 19(3), page 333-355, 2007.

I quit the Ministry of Finance in 2005 and roughly a year later, I bumped into a person who had been my driver while there. He said that he’s set himself up to collect the wage of the driver from the government, but has recruited another driver to go to work to do the actual work of driving. He was pocketing a neat profit out of this because the government’s price of a driver is roughly 2x the price in the private labour market.

Policemen are apparently poorly paid but with ubiquitous corruption and outright shakedowns being run by the police, the true income of a policeman in India is massive. I bumped into a young fellow on the beach in Goa a few weeks ago. He makes a living helping tourists do stretching exercises on the beach. A full 25% of his monthly income is paid to the local policemen as protection money.

Junior clerical staff in PSU banks reap a bonanza because they’re overpaid (when compared with the market price of clerical staff) and get job security for life. The NPV of that job is very high.

There is a risk aversion dimension also. People with high risk aversion might particularly favour these public sector jobs because they are both high wage and low risk.

In this environment, when the government advertises for 50 policemen, what do you think would happen? In an efficient market, a large number of suppliers of labour would see that there’s an opportunity to sell their services at much, much more than market price. There should be an outright deluge of job applicants.

The phenomenon of a million applicants showing up for a hundred positions is a reflection of civil service wages and job security being way out of line with what is found on the private labour market, and not a reflection of large scale unemployment in India. If anything, a very big deluge of applicants is a reflection of a rational information-rich environment where many individuals are able to access information and act on it.