Wednesday, July 15, 2015

Some Economics of the Freedom of Information Act

The Freedom of Information Act passed back in 1967 allows people to request access to records held by federal government agencies. A number of journalists and other researchers often seek to mine these records, while government agencies often push back--either passively by responding very slowly to such requests or more actively by redacting and selecting among records before releasing them. The US government has just announced a pilot program in which any records released under the FOIA would be released pretty much simultaneously both to the party that made the request and also to the general public.

Teachers of intro econ will recognize the problem. As a society, we want people and firms to innovate, because over time that process drives improvements in the standard of living. But we fear that if anyone can just copy successful new innovations, then the rewards of innovation will be too low. Thus, we have laws about intellectual property--about patents, copyrights, trademarks, and trade secrets--that give innovators some partial insulation from competition for a time and thus help them to earn a reward.

Similarly, as a society we want people to delve into government records because it is part of the checks and balances that keep a democratic government accountable. However, the new FOIA policy raises the danger that if the results of such requests are immediately available to the public, then the incentive for journalists and other researchers to spend the time and energy to file and follow-up on such requests will be reduced. A paradox seems to arise here: Is this a case where making any government records from FOIA requests publicly available could lead to a situation where there is less incentive to request such records, thus leading to a situation where fewer government records actually become publicly available?

This paradox in markets for information is well-known among economists, with a classic 1980 paper on the subject by Sanford J. Grossman and Joseph E. Stiglitz titled "On the Impossibility of Informationally Efficient Markets" (American Economic Review, June 1980, pp. 393-408). The main argument of the paper is made in math, rather than in English, but let me suggest how it applies here.

The key insight of Grossman and Stiglitz is that if our society wants people to seek out information, then there need to be incentives for seeking out such information--basically, those who seek it out need to get paid. This basic notion holds true for information related to news coverage, but also for information related to, say, investing in the stock market. However, if all the available information is spread quickly and easily, then it will be very hard for those who seek out the information to get paid. As they write: "There is a fundamental conflict between the efficiency with which markets spread information and the incentives to acquire information." In the case of the Freedom of Information Act, greater efficiency in spreading information--by making the results of all FOIA requests public--is in conflict with the incentive to acquire information.

The solution here seems fairly straightforward. The protection from conventional intellectual property law, like patents, is only supposed to be temporary. The innovation then passes into the public domain. Similarly, it would seem easy enough to give those who make an FOIA request access to that information for, say, three or six months, and then release it into the public realm. Sure, some journalists and researchers will say that they need exclusive access to the information for much longer. Innovators would like their intellectual property rights to be stronger and last longer, too. But a balance needs to be struck.