Transparency is a fundamental component of democratic government.

It concerns the right of citizens to know about the activities of

their government. Transparency policies have expanded to protect

against institutional corruption, a notion that underlies Louis Brandeiss famous

quote: Sunlight is said to be the best of disinfectants.

Recently, transparency policy has evolved into what my colleagues

Archon Fung, Mary Graham, and I call in our research targeted transparency.

Targeted transparency is the use of publicly required disclosure of specific

information in a standardized format to achieve a clear public policy

purpose. It therefore goes beyond the disclosure of information as an end in

itself. Targeted transparency represents a new regulatory approach, but it also

provides critical lessons to those seeking to build public trust in government

through information disclosure.

Over the last few decades, policymakers at the federal,

state, and even international levels have addressed

major public crises, not by regulatory responses, but instead

by a decision to require information disclosure. For

example, hundreds of people died annually as a result

of sport utility vehicle (SUV) rollovers in 2000 and beyond.

Historically, the federal government would have

responded to this crisis through standards-based regulation.

Instead, Congress chose to require auto companies

to report the National Highway Traffic Safety Administrations

assessment of the probability of a rollover, which

was then translated into a simple rating for consumers.

Instead of telling Ford or General Motors to change

their SUV designs, the government developed a five-star

rating that shows consumers the probability that a given

SUV would roll over during different kinds of maneuvering

a radically different approach to improving auto

safety. This is just one example: counting conservatively,

we found 135 major federal regulations that used this

type of regulatory transparency in the last ten years.

When Does It Work?

Most people support the general idea of transparency

because, as noted, it is basic to democracy. However,

targeted transparency gives rise to new questions. When

does it work? What factors link disclosure to changes in

behavior that achieve public objectives? After reviewing

eighteen major transparency policies, our conclusion is

that more often than not they do not work. They sometimes

do not work because they are political compromises

that were never intended to work. More often,

they dont work because they havent been designed

terribly well, and the policymakers havent really considered

what they want to accomplish through transparency.

Two examples suggest what makes transparency

work or not work as a means of regulation.

Clarity or Opacity

Many people use the nutritional labels on food and

drink in deciding what to consume. They represent specific

federal transparency policy for nutritional labeling

that is dictated down to font size. These simple labels are

the result of a lengthy, involved policy process that included

legislative debate on such minutiae as that font size.

In contrast, there is federally mandated disclosure

on drinking water in local communities, which not

many realize they have the right to. That legislation was

passed in response to a major public health problem in

Milwaukee, which resulted in the death of a number of

citizens because of drinking water contamination. Most

people dont even know there is such a law. The quarterly

disclosure report from the local water authority is

totally unfathomable. My wife is an environmental risk

assessor, who spends much of her time considering the

impact of exposures to low-grade toxins on health, and

she cant make heads or tails of this thing; most other

people who read it cant either.

These examples illustrate the promise of and problem

with transparency policies. These policies only

work if they give users the information they need, when

they need it, and in the form they need for making

an effective decision. If those conditions arent met, the

transparency policy fails. Lets look at another example,

where transparency policy has had a major impact.

Southern Californias Restaurant

Transparency Law

They say you never want to look at what happens

in a sausage factory; you probably dont want to know

what goes on in a restaurant kitchen either. In Los Angeles

in 1997, a film crew decided to find out, taking a hidden

camera into several restaurants and filming atrocious

things: rats in storage areas, failure to refrigerate food,

waiters reserving food from one customer to another, and

other appalling practices. This led to a big outcry, and the

Los Angeles County Board of Supervisors quickly passed

a transparency law that requires every restaurant in southern

California to post a letter grade (A, B, or C) based on

public health inspections. The size and shape of the sign

are mandated, and it has to be in the front window right

next to the menu: A, B, or C. If a restaurant doesnt get a

C, the county public health department shuts it down.

In short order, this system has had a phenomenal impact

on public health, as documented in a number of careful

studies comparing restaurants with bad grades with

those with good ones. The Cs lost significant revenue,

while the As business improved. More important, hospitalizations

related to food poisonings greatly decreased.

The details of the policy make senses: the best time to

give people information about a restaurant is when they

are walking in the door. If you look at the menu and see a

C staring you in the face, you spin around and go somewhere

else; in fact, this is what happened.

This is in stark contrast to the drinking water standard,

which disseminates highly complex information.

Its a quarterly report that arrives by mail, along with all

the other junk mail. Consequently, people throw it away

most of the time. It contains incredibly detailed, difficult

scientific data. If thats not bad enough, the information

is a year old, so it says nothing about the current quality

of drinking water. Not surprisingly, studies of this policy

show that its had very little effect. It offers the wrong

information, in the wrong form, on something about

which people have little choice.

As shown in these examples, to fashion policies that

work, policymakers need to understand how information

affects the decisions of users and how, in turn, changes

in user behavior affect the actions of disclosers. The Los

Angeles County restaurant grading system works because

consumers changed their behavior, and in response, restaurant

operators perceived or anticipated those changes,

which incentivized them to clean up their act. In contrast,

the drinking water measure failed because the information

was too opaque to get users attention and gave no

one incentive to change the system.

Data Citizens Need

Consequently, in thinking about the challenge of

using transparency policy to reduce corruption and

build public trust, we must understand how citizens

and other parties can assimilate and utilize the information

provided. Dumping tons and tons of data on the

Web doesnt accomplish much. We need to know the

data citizens need, whether they are financial, outcome

based, or other. We can only find this out by reaching

out to the communities, not only the individual citizens,

but also civic groups, the media, and the people whose

behavior might affect the public trust.

We need to go beyond posting downloadable information

on the Web, to taking advantage of cuttingedge

information technologies. This includes tapping

the potential of social networking technologieslike

FaceBook and Twitter, the way the new generation communicates

which allow interaction among citizens and

between them and policymakers. Transparency policy

can enhance citizen engagement, interaction, and ultimately

trust in the democratic process, but we must learn

from the current systems, building policies based on how

people use information in making their decisions.