As you may imagine, library privatization, or the outsourcing of the operations of a library to a private company, is not a prospect to warm the cockles of many librarians’ hearts. The American Library Association (ALA)’s official policy manual flatly “opposes the shifting of policy making and management oversight of library services from the public to the private for-profit sector”. California Governor Jerry Brown recently signed into law a bill to “restrict the privatization of public libraries”. In contrast (and unsurprisingly), private companies that contract with cities to provide library services have vigorously defended themselves. For instance, LSSI, the most prominent for-profit company in the category, asserts that it “fixes broken libraries” by instituting much needed efficiency-enhancing and cost-saving measures.
This is a complicated issue. To properly answer it, one needs both carefully collected data and a theoretical framework for making sense of that data. (The theoretical model is needed to provide guidance on questions like “what are the preferences of library users?”, “how homogeneous are such users?”, and “how do their preferences aggregate?”.) Only with both elements thoughtfully designed and executed can one attempt to calculate changes in social welfare due to the privatization of a library system, i.e. to conclude whether library privatization is good or bad (and for whom).
This is a complicated issue — one on which I’d be suspicious of easy, categorical answers. I haven’t any answer to offer myself, but for anybody attempting to do such a study, I would offer three points that need to be checked off.
- Please make sure that you are comparing the proverbial apples to apples. Specifically, if you’re going to examine a particular execution of a privatization plan, then you want to compare that to a realistic counterfactual, not the idealized abstract of how a library is run by public workers.
- Please don’t confuse inputs with outcomes. An argument against privatizing libraries is that the privatized entities lay off workers or substitute MLIS-certified librarians with para-professionals. Leaving aside the question of whether this is true or not, I would posit that this is something I, as a MLIS-in-training librarian, care a lot about (because I want a job), but not something I expect the public automatically to. The public cares about the service it gets; bad service from a lot of workers is not valued more than good services provided by fewer workers; crummy service from MLIS librarians is not preferred to good service from para-professionals. (In more technical jargon, the arguments to the agent’s utility function are different from the inputs to the production function. Furthermore, when private companies’ arguments rest on the proposition that they have a different production function than the library staff of old, I don’t know how much informational content is added is by examining the inputs.)
- Please be clear about the assumptions behind your arguments. For instance, asserting the LSSI does well on measurables such as number of books purchased, hours opened, and circulation counts but that it does not measure everything (e.g. the quality of the service offered during the hours opened) is a valuable argument insofar as you assume that one cannot infer the unmeasured from that measured. For example, to cast doubt on the quality of service even though the hours of operation are expanded, you must be assuming that there is a negative correlation between those two things. Whether this is true or not, I have no idea, but I certainly need to know that for your argument to carry weight beyond sounding ominous.
I’m not in favor of privatized libraries; I simply don’t have enough information to be. I would, however, very much like to have cogent, convincing argument from either side.