Opinion, Berkeley Blogs

Privatization: Maybe good, maybe bad, definitely ugly

By Steve Tadelis

New York’s deputy mayor, Stephen Goldsmith, a champion of privatization, announced plans to “insource” tasks and services that the city previously privatized and assigned to contractors and consultants, citing opportunities to save money for the city. Isn’t saving money the motivation behind recent plans by Florida’s Republican leaders to privatize prisons and probation services? Governor Tom Corbett of Pennsylvania, likewise, announced earlier this month that he will study the benefits of privatizing state owned liquor stores. But he said that it “isn't about the money. It's about the principle. Government should no more run the liquor stores than it should run the pharmacies and gas stations.”  What, then, should governments be providing, and does privatization actually save money or inflate expenses? The answer is, it depends.

There are services that have been proven easy to privatize, such as garbage collection. The task is simple and well defined: drive the truck, pick up the cans, dump the trash and don’t spill it on the street.  How do we ensure high quality service? Citizens will complain if it isn’t.

But not everything that seems straightforward is easy to privatize.  Take information-technology (IT) services. There are dozens, if not hundreds of private companies that provide a variety of IT services, making it easy to find qualified private providers who can bid for, and supply, IT services. But for governments to use the market mechanism, they have to specify in advance what services are needed, how they will be delivered, how success will be confirmed, and how payments will be made. What makes IT services challenging is the rapid change in the ways we use IT and in the technologies that are used to deliver it.

Not long ago I heard a telling story from a city manager in California. Some city council members suggested that there is no reason to have an in-house IT department when there are so many excellent private companies that do the same thing. Shortly thereafter, they hired a private firm at a price that was significantly less than their costs of having an in-house team. Things went great until the first time they asked the company to produce a new report. The company’s response was “that’s not in the contract and we will have to charge an extra fee.” This made sense. The problem was that every few weeks a new report or a new analysis was needed for some ongoing activity, leading to a succession of change order requests with extra payments.

Since the company owned the database, the city could not get competitive bids for new changes. Within months, the city had to hire a part-time employee to deal with change orders.  After a couple of years it was clear that the city was paying more than double for the same quality of IT services they previously received from their former in-house staff. When the two year contract was up for renewal, they terminated the contract and brought back an in-house team.

This kind of problem is not exclusive to local governments. Big corporations can fall into the same trap. In 2004, Sears, Roebuck and Co. outsourced its IT services to Computer Sciences Corp (CSC). After less than a year, Sears terminated the $1.6 billion 10-year agreement with CSC and brought the services back in-house. In 2002 Diebold outsourced its IT services to Deloitte with a 7-year contract in place. In 2006 Diebold backed out of the contract 3 years early and brought back 80 workers to provide the services in-house. More recently, in July, 2009 Boeing acquired one of its most important contactors, Vought Industries, to achieve better control over the much delayed production process of Boeing’s 787 Dreamliner.

At the heart of these problems lies a conflict of interest that is present in any outsourcing or privatization relationship. The buyer, or city, seeks to provide a service at lower costs compared to in-house provision. On the other hand, the contractor wants to make a profit. This tension means that the relationship must be designed carefully in order to ensure a successful outcome for both local government and private contractor. If the service that is outsourced or privatized is difficult to scope, define, bound or monitor, then what the government may seek and what the contractor delivers are neither aligned nor easily described. This leads to an imperfect contract for which neither party’s expectations are adequately accounted for. Costs can skyrocket due to change orders. In a public procurement setting where there are no immediate market pressures to control costs, this can lead to long lasting waste. Furthermore, if the service requires adaptation as the relationship evolves, then the flexibility a government would have with its own workforce is lost.

Politicians like simple messages.  The trouble is that political agendas seldom align with the cost-benefit analysis required for good privatization policy decisions. The tough part is strategically choosing the right projects and services for privatization that have a good chance of avoiding outsourcing’s pitfalls.

Politicians like simple messages. Conservatives like to say that “privatization provides good services at low costs,” while many liberals will claim that “privatization reduces quality and costs jobs.” Both can be right or wrong, depending on the particulars of the service involved. The ugly part is that political agendas seldom align with the cost-benefit analysis required for good privatization decisions. An operational framework for strategically choosing whether or not to privatize a given service requires an ability to identify both the opportunities from privatization and its potential pitfalls. We can only hope that Mr. Goldsmith has some people who can do it for the benefit of New York’s citizens.