Safeguards At Risk:
John Graham and Corporate America's Back Door
to the Bush White House

March 2001

This report was researched and written by Laura MacCleery, with research assistance from Paul Schmitt, Debbie Quint, Daniel Becker, and Ed Ricci. Editorial assistance and direction were provided by Joan Claybrook, Booth Gunter, Angela Bradbery, Melissa Luttrell and David Vladeck. Reproduction of the report was managed by Johnette Washington.

Public Citizen is a non-profit, 150,000-member organization based in Washington, D.C. that represents consumer interests through lobbying, litigation, regulatory oversight, research and public education. Since its founding by Ralph Nader in 1971, Public Citizen has fought for consumer rights in the marketplace, safe and secure health care, health and safety standards, fair trade, clean and safe energy sources, and corporate and government accountability. Public Citizen is active in every public forum: Congress, the courts, governmental agencies and the media.

Public Citizen
1600 20 St. N.W. th
Washington, D.C. 20009
(202) 588-1000
March 2001

©2001 Public Citizen. All rights reserved.
ISBN No. 1-58231-022-X

WHO IS JOHN GRAHAM?

Introduction

Patricia Pena didn't know what hit her. Just three days after she left her job to become a full-time mother, her car was broadsided at an intersection and her only child, a 3-year-old girl, was killed. The police investigation revealed that the driver who hit her car was dialing a number on his cellular phone when the crash occurred. Pena has since become a national advocate for banning the use of cellular phones while driving, traveling to Washington, D.C., to speak with legislators and writing about her story in USA Today.

Opposing her are the wealthy cellular communications industry and a well-coordinated network of corporate lobbyists, conservative public policy think tanks and industry-funded academics. One of these academics is John D. Graham, whose Harvard Center for Risk Analysis (HCRA) accepted $300,000 from AT&T Wireless Communications to do research on driver distraction and cell phones. Graham's study concluded that a ban on cell phones while driving was too "costly" to be worthwhile. Graham's cell phone cost-benefit analysis misapplied basic statistical methods and weighed projections of the public's injuries and fatalities against lost industry profits and other consequences of a ban. This approach to matters of public health and safety is typical of his work, this report shows. 30 Now, Graham has been nominated to be the regulatory gatekeeper of the Bush administration in the Office of Management and Budget (OMB) regulatory affairs office, with the power to undermine standards set by federal agencies under congressional mandates.

Graham's appointment to the Office of Regulatory Affairs within OMB represents a serious threat to public health and environmental protections. Graham worships at the altar of economic analysis, proposing that all government regulatory actions conform to a methodologically suspect system of decision-making. Although it is marketed as "sound science," in fact his approach is perfectly aligned with the financial interests of regulated industries, such as chemicals, oil and gas, agribusiness and industrial metals. More than a hundred major companies have provided direct, unrestricted funding for his Center over the past decade.

Acting as a political strategist for these companies, Graham has advocated that the government give economic factors (like the cost to industry of complying with new regulations) the virtual trump card in any decision involving the enactment of new safeguards. His paradigm for regulatory 31 decision-making is anti-consumer, anti-environment, and hostile to preventive measures in public health. As a practical matter, Graham's regulatory proposals will also make it far more difficult for the federal agencies to act to protect public health and safety.

Graham is currently the director of the Harvard Center for Risk Analysis (HCRA), which is connected to Harvard University as a separate sub-unit of the Harvard School of Public Health. The Center offers some twelve courses in a field called "decision analysis" to Harvard students but does not itself grant degrees. In addition to the student courses, the Center also offers, at a fee, short courses and multi-day seminars that grant "continuing executive education" credits to corporate officers. Graham teaches several of the courses, which train business executives to, among other things, "identify subgroups of lay people who are likely to be particularly outraged or tolerant about a potential risk."

The Center also publishes a monthly newsletter called Risk in Perspective. One recent issue suggested that "speculative [and] minor" risks to children include pesticides and Bisphenol-A and phthalates (toxic chemicals found in plastics and some children's toys), without acknowledging that funders of HCRA make these products. Another issue of the newsletter from April 1999 was titled, "Toxic Pollution from Power Plants: Large Emissions, Little Risk."

Although Graham himself has sometimes been called a "scientist," and he often uses the honorific "Dr." to the press while invoking the name of Harvard's School of Public Health, he has earned no degrees in medicine, public health or the hard sciences. The subjects of his degrees are omitted from his resume on the HCRA Web site, but Graham has previously told Congress that both his undergraduate and master's degrees are in public policy. We called Wake Forest University, where Graham earned his bachelor's degree, and were told by staff in the alumni office that there is no undergraduate program in public policy and that, according to their records, Graham earned his B.A. in economics. Graham's doctorate from Carnegie Mellon is in a subset of public policy called "decision analysis."

For the uninitiated, the name of Graham's Center for Risk Analysis might be a bit misleading. "Risk Analysis" is a broad, umbrella term that includes at least two sub-disciplines: "risk assessment" and "risk management." Risk assessments survey hazards from human exposure to a substance (or an activity) and the effects upon human health or the environment, and often involve original research in an area of the hard sciences, such as biology, chemistry, toxicology, etc. Risk assessment also uses estimates and theoretical models, because the scientific data are often lacking.

Risk assessment asks an objective question: "How risky is this situation?" However, the results of risk assessments are hardly "objective." As Mark Shere, a conservative lawyer who represents industrial clients in pollution cases, noted, "A typical risk assessment consists of about 50 separate assumptions and extrapolations, each of which may skew the analysis by a factor of 10 or more... these assumptions and extrapolations can alter the final numeric estimate of risk by a multiple of billions, and this result is again, unverifiable."

Risk management, on the other hand, is a field of public policy that uses the results of risk assessments, as well as information about social and political values and data and statistical modeling tools, to come up with recommendations about risk policy. Risk management is a normative inquiry, asking "What should we do about the risk?" Graham also frequently invokes concepts that rightly belong under the rubric of "cost-benefit analysis" or "risk-benefit analysis." These phrases refer to two closely related sets of public policy tools that weigh either projected costs or risks against the projected benefits of some activity or rule.

Most Ñ if not all Ñ of Graham's work falls into the risk management category, meaning that it is informed by public policy as well as by data from the underlying hard science discipline. Many risk managers do have degrees in biology, or chemistry, etc., and the federal regulatory agencies employ them to make policy recommendations based on the science they know. Crafting regulations that protect health, safety and the environment is normally viewed as a highly specialized endeavor.

Driver Concentration Is the Problem...
In a recent Dallas Morning News article about the causes of recent traffic deaths in Texas, Graham emphasized his faith in the driving ability of ordinary citizens, indicating that "virtually everyone, except people who have significant health problems, is capable of being a safe, competent driver." Graham implied that unnecessary traffic deaths could be prevented by increasing the level of driver concentration Ñ saying that "the problem we have is maintaining people's level of attention."

...But Driver Distraction Is Not the Problem.
Despite this statement, Graham has come down against eliminating one possible source of driver distraction: the use of a cellular phone while driving.

One week after the National Highway Traffic Safety Administration (NHTSA) held a public hearing on driver distraction and recommended that drivers pull over before using cell phones, the Harvard Center for Risk Analysis, using Graham's name, self-published a report funded with $300,000 from AT&T Wireless Communications which assessed the risks of using a cell phone while driving.

The report, released in July 2000, was very timely. Communities and states all over the country were in the midst of considering whether to enact bans on the use of cell phones while driving. Just prior to the study, the township of Marlboro, New Jersey, banned drivers from using cell phones, and Brooklyn, Ohio, had passed the country's first such law in March of 1999.

While twenty-two states had considered bans, none had been passed at the time of the report's release. As in the battle over environmental tobacco smoke described above, industries that anticipate that new laws and public attitudes may discourage the use of their products take these threats very seriously indeed.

The HCRA report surveyed existing data and concluded that because the level of risk to drivers was not clearly indicated, further regulation in this area was unwarranted. HCRA's report stated that "although there is evidence that using a cellular phone while driving poses risks to both the driver and others, it may be premature to enact substantial restrictions at this time. We simply do not have enough reliable information on which to base reasonable policy."

The report also made use of a "collateral benefits" theory of risk tradeoffs. According to the report, the benefits of cell phone use while driving may actually outweigh the potential dangers of using them. The report attributed gains to users such as "peace of mind," "expanding productive time," and "strengthening social networking." The obvious problem is that these benefits are difficult to quantify so as to weigh them against the risk to human life from driver distraction. And to the extent that the benefit of "peace of mind" derives from having a phone in the car for emergencies, a ban on using a phone while driving could be crafted to preserve this benefit entirely.

The HCRA study drew the main power of its conclusion that the risks of driving and cell phones had not been adequately demonstrated using general data that showed that while "cellular phone use has grown 17-fold between 1990-1998, U.S. traffic fatalities have continued a steady decline." These data did not include any specific information on cellular phone use while driving, nor did the study analyze crash records to ascertain whether cellular phone use was a cause of the crash, as NHTSA has done.

HCRA's comparison of gross figures on cell phone use with gross crash data does not account for recent innovations in air bags and increased use of seat belts that were likely major contributors to recent declines in fatalities. Yet these data form the basis of the study's position against banning the use of cell phones while driving.

Comparing Apples and Oranges
The central problem in Graham's study actually relates to the evidence presented in support of his conclusion that the risks are undemonstrated. In a media interview about the study, Graham applied a misleading risk tradeoff: "Based on the information to date, the risks [of driving while talking on a cell phone] look fairly small compared to risks people face in daily life." But the central question is whether the activity creates an additional or unnecessary risk, not whether it comprises just yet another risk that we should learn to live with.

This simple problem plagues each part of his analysis. According to Graham's study, driving while using a cell phone causes fatalities of 6.4 deaths per million drivers annually. But instead of comparing this figure with drivers under similar conditions who do not use cell phones, the study spuriously compares this data to cases involving extreme risk factors, such as driving with a blood alcohol concentration of .10, which causes annual fatalities of 30.9 per million drivers, and driving without wearing a lap and shoulder belt, which causes 49.3 annual fatalities per million drivers. This choice of methodology shows a basic lack of commitment to elementary standards of research. As Charles Osgood said on The Osgood File when reporting on the study, these categories of data are "apples and oranges.

It is true that, when compared with these few circumstances, driving while talking on a cellular phone may appear relatively safe Ñ but the study fails to compare driving with a cell phone to the risks of driving without a cell phone under normal conditions. In addition, these comparisons are wrong because the risks could be cumulative Ñ for example, drivers may be at risk from both talking on a cell phone and being intoxicated while driving.

Therefore, regulation banning the use of cell phones while driving might help to eliminate an additional risk on the road, as many communities appear to believe. Fundamentally, comparisons like the ones in the Graham study are flawed because comparing the riskiness of some activity to some other risk makes little sense if we can choose to live with no additional risk at all.

My Study vs. Your Study
As the media framed it upon its release, the HCRA study "contradicted the finding by another study done in 1997." That other research, published after full peer review in the New England Journal of Medicine, avoided such ridiculous comparisons, and had concluded that the risk of car crashes is four times greater when a driver uses a cell phone.

Dr. Donald Redelmeier was one of the authors of the New England Journal of Medicine study. In an attempt to approximate something like peer review, which is the accepted practice in the sciences to assure the validity of a study, Graham asked 12 independent specialists to "peer review" Graham's study, Redelmeier among them. After reviewing it, Redelmeier publicly disapproved. But the study was published and amply advertised regardless, thus demonstrating the toothlessness of Graham's "independent peer review" process.

Redelmeier publicly criticized Graham's assessment of cell phone risk, suggesting that the HCRA report lacked rigor because it "provides no new data, gives no new expertise and provides no new analysis." Redelmeier also told reporters that the "Harvard researchers left the report open to conflict-of-interest questions because they didn't publish it in a scientific journal or take other steps to demonstrate the study's fairness."

Other Problems With the Data
The policy conclusions in Graham's study appear to assume that cell phone use while driving will not increase in the future. This factor is taken account into a recent NHTSA study of the issue, which noted that "with a growth rate of about 40 percent per year, it is estimated that by the year 2000 there will likely be about 80 million cellular telephone users in the United States." For comparison, that NHTSA study concluded that using a cell phone while driving does increase the risk of a crash.

From a Dubious Risk "Assessment" To Even More Suspect Public Policy
The report's policy evaluation of bans on cell phone use while driving is even more questionable than the risk "assessment" piece. The report suggests that such restrictions are "inefficient," because the net cost per life-year of the regulation is alleged to be significantly greater than the use of lap/shoulder belts, daytime running lights, and even greater than the use of air bags.

The HCRA study asserts that it would cost approximately $700,000 per life year - saved to restrict cell phone usage while driving, compared to $24,000 for front-crash air bags for drivers, and less than $0 for lap and shoulder belts.

These conclusions are highly suspect, as a relatively expensive but life-saving safety system such as an air bag requires a financial outlay for development and implementation of new technology. Passing a ban on cell phone use while driving requires no such initial resources.

Moreover, HCRA assumes that regulation must be cost-effective in order to be warranted. But the proposition that using a cellular phones while driving is distracting and potentially dangerous is common sense. It relates to a common experience Ñ driving Ñ and requires little technical expertise to evaluate as a hazard in comparison to, say, varying levels of toxic chemicals.

Without saying so explicitly, the HCRA study assumes that regulators alone must carry the burden of proof on the cost-effectiveness of their recommendations: It requires that a ban be justified on cost efficiency grounds (based on the limited information that we have available) before we can act. In so doing, it leaves common sense, the rights of other drivers and pedestrians not to be injured or killed by a distracted driver, and the precautionary logic and experience of individual communities, far behind. As the eminently logical Tom and Ray Magliozzi, hosts of Car Talk from National Public Radio, put it in response to a similar study by Robert Hahn, Graham's ally at the American Enterprise Institute-Brookings Joint Center for Regulatory Studies, "This seems to us to be a clear case of cost/benefit analysis run amok."

The timing of the study in relation to the NHTSA recommendation about limiting cell phone use while driving was surely no accident Ñ and the study's conclusion concerned the political wisdom of enacting a ban. This assumption that risk "experts" should quash good sense by finding regulation is unjustified unless the harm of an activity has already been demonstrated blatantly favors business interests and the status quo, given the political context of the study.

Often, incomplete information on risks means that the benefits of preventive regulation can appear very small, when in fact they could actually be enormous. Unless we are very careful about adjusting cost-benefit equations to account for our uncertainties, it will always turn out that regulation which prevents a future risk is not cost-efficient. Until there are demonstrated risks, such as bodies on the highway, cost-benefit analysis can tell us that protecting safety (or health, or the environment) is just not worth a dime. And because up-front prevention often must occur in the absence of complete information, using a cost-benefit approach to evaluate health policy at the outset will doom us to forever locking the barn door after the horses are gone. The recklessness of these conclusions Ñ and the study's lack of a protective attitude towards human life Ñ is stunning.

*********************
LETTER FROM TOM AND RAY MAGLIOZZI,
HOSTS OF CAR TALK FROM NATIONALPUBLIC RADIO,
TO THE NEW YORK TIMES ABOUT AEI-BROOKINGS' ROBERTHAHN'S STUDY
ON COST BENEFIT ANALYSIS AND CELLULAR PHONES

18 November 1999

Letters to the Editor, The New York Times
229 West 43d Street
New York, NY 10036

Re: "Driving and Talking Do Mix" (New York Times, November 12, 1999)

To the Editor:

As proponents of "Drive Now, Talk Later," we must take issue with the op-ed piece by Robert Hahn. This seems to us to be a clear case of cost/benefit analysis run amok.Just what is this guy thinking? First, he refers to accident reports Ñ which for the most part do not even record the use or nonuse of cellular phones. Then he cites his own estimates of the mere "10,000 people (who will be) in serious accidents and the minuscule number of "100 people who will die" due to cell phone use. And the benefits associated with these accidents and deaths? First is the ability to summon help on a lonely highway. Are we a little short of logic here? Just how would a ban on cell phone use while driving prevent one from using the phone from the breakdown lane? The almost invaluable benefit is the convenience of reminding your spouse of your daughter's school play.

But not if you happen to be Patricia Pena. Her two-and-a-half-year-old daughter, Morgan Lee, was killed by one of Robert Hahn's new breed of highly productive citizens. What price has Mr. Hahn plugged into his nice, clean economic model to account for the misery and tears that such outright selfishness has wrought?

Mr. Hahn, you're just not that important that you need to talk and drive, and endanger thelives of innocent people on the roads. And next time you consider writing an op-ed piece, please remember the admonishment of Ted Williams to one of his not-too-bright teammates: "If you don'tthink too good, then don't think too much."

Tom and Ray Magliozzi, Co-hosts
Car Talk, on National Public Radio

Update from the Web Site of Tom and Ray Magliozzi
Co-hosts, Car Talk, on National Public Radio
About the Cell Phone Issue

A couple of interesting developments that you might be interested in.

1. That Great Newspaper and arbiter of what's "fit to print" Ñ the "New York Times" Ñ opted to print a guest op-ed column on November 12 ("Driving and Talking Do Mix") by some guy named Robert Hahn from the American Enterprise Institute.

2. We wanted to reprint it so you could read it yourself, but the NYT wants $250 for permission to reprint. Instead, we'll paraphrase. Mr. Hahn says that it's silly to ban the use of cell phones while driving. In fact, according to his personal estimates, cell phones will cause only 10,000 serious accidents this year, leading to (a mere) 100 fatalities.

Then this moron (in our opinion) goes on to explain to us that he has done a cost/benefit analysis, and in his opinion 100 deaths and 10,000 serious accidents is a small price to pay for the enormous benefits we derive from using the cell phone.

Here are a couple of his benefits: #1. The ability to call for help when your car breaks down. (Right away one has to wonder if this guy is playing with a full deck. If your car breaks down, you'd hardly be using the cell phone AND driving. Duh, Mr. Hahn.)

And this one is a beauty. The 10,000 serious accidents and 100 fatalities are a small price to pay (says Mr. Hahn) for "the ability to remind your spouse that your daughter's school play starts in 20 minutes." At this point you know that he is DEFINITELY not playing with a full deck.

Naturally we immediately penned a letter to the editor, who apparently did not consider it to be "fit to print." (Admittedly, the letter was a little short on tact and diplomacy. We didn't think such moronic thinking deserved tact and diplomacy.)

Fortunately the NYT did print a couple of letters that did address the issues. One said, "Does he really consider a cell phone call from a car to ''remind your spouse that your daughter's school play starts in 20 minutes'' more important than even one life? Since when do small conveniences rank above human life?"

#2. A few weeks ago, we at Car Talk Plaza were approached by NBC News. They had heard our ranting and seen the feature here on the site, so they wanted to do an interview with Tom and Ray on the cell phone issue. We generally eschew the media whenever possible. But this seemed like an opportunity to help a good cause, so we agreed to do it. Everything was arranged; a date, time and location for the taping were set up.

But...a few days before the event, we were informed that "the producer in New York had decided not to do it."

#2 part 2. We were approached by CBS with a similar request. Again, mysteriously, the "producer in New York decided not to do it."

Now, it's certainly possible that there's a simple explanation for these two incidents. The most likely--even we admit it--is that the producer had never seen or heard of us before. I mean, that would certainly explain it. If you were a producer of a major network show and...well, you get the idea.

On the other hand, there's a distinct possibility that the news that's fit to print and the news that's fit to put on the air are very much influenced by big bucks. Let's face it: there's a lot of money involved in the cell phone industry--and lots of it goes to advertising.

Just a thought.

[ Drive Now, Talk Later Index ]