“Open government” is one of those bumper sticker phrases with which nearly everyone agrees. But what does open government really mean? And what difference does it actually make?
These questions have grown increasingly salient as President Obama has made transparency a key focal point of his administration. Across the federal government, agencies are now undertaking numerous projects under the umbrella of the administration’s Open Government Initiative.
The federal government has launched several new, cross-cutting websites, such as USASpending.gov and IT Dashboard, that make information about government spending more accessible to the public. Other projects, such as the website Data.gov, make datasets accessible to citizens who can put them to use for their own purposes. Across the federal government, agencies have developed their own open government plans with projects ranging from the creation of online reading rooms to the posting of product safety recalls.
Are these projects working? How can we find out if they are? At two separate meetings over the last month, government officials have joined together with scholars to launch a research agenda on open government. As a participant in these meetings – one held in Washington, D.C., the other in Albany, N.Y. – I have come to realize how little clarity currently exists over what open government means and how to assess its impact.
Part of the problem is simply the dizzying array of open government projects now underway. Without even a simple taxonomy, it’s hard to talk sensibly about how to study these projects. As I see it, though, what people mean by “open government” actually refers to three distinct things: process transparency, data availability, and informational regulation.
The first of these – process transparency – refers to efforts to make the inner workings of government more visible to the public. I have sometimes called this “fishbowl transparency.” Websites like USASpending.gov or Regulations.gov fall into this category, as they seek to make information about governmental decision making more accessible to the public.
“Open government” can also refer to the release of data that government regularly collects on various conditions of the world – data that may also be put to use by citizens for their own purposes. Richard Thaler called attention to this type of open government when he wrote recently about how private sector firms can reformat government data “in ways that are helpful to consumers, workers and companies.” He noted, for example, that the public release of real-time GPS data collected by the San Francisco public transit agency for its own internal management has led to a smartphone app that informs city residents if the bus they want to catch is running on time.
In his popular book, Nudge, Thaler (along with co-author Cass Sunstein) has also written about a third meaning of open government: the disclosure of information about regulated firms as a means to shape their behavior. In this sense, open government refers to what economist Jay Hamilton has called “regulation through revelation,” or what lawyer Mary Graham has termed “regulation by shaming.” By releasing information about private actors’ behavior, government seeks to exploit social or market pressures to discourage unsavory conduct, just as it would by writing a traditional regulation. For example, in Nudge,Thaler and Sunstein laud the Toxic Release Inventory (TRI) program through which the U.S. Environmental Protection Agency releases information about businesses’ pollution levels. Thaler and Sunstein claim that TRI’s information release, combined with social and market pressures, has led companies to reduce their pollution.
Each of these different types of open government serves different purposes, and each involves the disclosure of different types of information. Open government as process transparency discloses information about government actors and their actions. Data availability discloses information that government officials collect for their own purposes but that private individuals and firms can also use for their own purposes. And informational regulation generates and discloses information about private firms in an effort to shape those very firms’ behavior.
In studying the impact of open government, it will be important not only to keep the different types of projects in mind – but also to keep in mind the different impacts that these projects may have, whether external or internal to government.
External impacts – consequences for citizens and businesses – can be further subdivided into two types:
Internal impacts – those taking place inside government – include all the possible ways that open government might affect the behavior, work patterns, and decisions of government officials. These impacts can also be subdivided into two types.
Not every open government project will be intended to have all of these different types of impacts. To generalize, we might say that data availability and informational regulation projects will typically seek to achieve external, instrumental impacts, while process transparency projects often aim primarily for internal impacts. But the reality is that any project might have a variety of impacts. Releasing public transit data on buses, for example, makes possible a useful service for passengers, but it also may change the behavior of the bus drivers themselves.
No matter the category of open government, thinking carefully about the broad range of possible impacts will help not only in designing open government projects at the outset but also in subsequently evaluating their full range of impacts. Although everyone surely agrees that open government is a good thing, this does not mean that all government initiatives to disclose information will yield net beneficial impacts. The only way to find out will be to study open government projects after the fact, just like other government programs are evaluated, determining empirically what impacts they actually have.