I'm going to start with a very broad question. Are American drivers served by the auto safety system we have in place today?

The auto safety system that we have today is spotty. In some cases, the American public is served; in other cases, it's not served. And it just simply depends upon the particular event and the time that you're involved in.

Is that the way it should be? What are the fixes? What are the flaws?

The system's a good system. ... I don't often compliment the United States Congress. But they did a very fine job on this, and essentially what they have come up with is an agency called the National Highway Traffic Safety Administration -- everybody calls it NHTSA for short. ... They have a system whereby NHTSA designs federal motor vehicle safety standards. These are good things.

In order to have a standard, first of all, you have to have a need, that is, you look at all the things going on and say, "We need to do something to make a car safer in this area," let's say, windshields, for example. So if you have a need, then you look around and say, "Is it practical to do this? Does it make sense to do this?" And while you're doing that, you have to ask a couple of other questions, such as, "If we come up with a standard, can we come up with a test that's repetitive? A foolproof test, never changes, never varies, always will tell us whether or not the motor vehicle met that particular standard?" And then we have to ask some other questions that go along with that, for example, "This thing that we're doing, is it something that the public can afford to buy?" ...

Regulations are like that -- there are parameters. I mean, you can only give the public how much safety it wants. If it doesn't want to pay for the safety, you can't really give that to the public. And you can't give the public safety that it will not accept for some cultural reason.

I'll give you a very good example. In the 1970s, somebody said, "We will save more lives by increasing seat belt use than by all the regulations put together on one motor vehicle." That's a fact, by the way. You do save more -- just upping seat belt use 10 percent, you'll save more than all the regulations NHTSA, the federal government has. And that made sense. They said, "Here's the way we'll do that. We'll have a regulation which says this: When you get in a car, you cannot start the car until you buckle your seat belt," and so they passed the regulation. The American public went absolutely ape. They jumped on all their congressmen, they screamed, they hollered. Congress came back to the federal government, to NHTSA and said, "Sorry. You will save tens of thousands of lives by doing this. But the American people won't accept it." And they passed a law which said NHTSA can never again come up with a regulation that says you must have your seat belt on before you can start the car.

How do you know what American people will accept in terms of cost? Is there an actual cost/benefit analysis built into the rules? How do you know?

There is a cost/benefit analysis built into the rule. You can tell what it will cost to do that, and then you have to make a judgment as to whether or not the public will accept this additional cost.

So it's somewhat subjective?

Oh yes, it is subjective. ...

Let's talk about the Bronco II. Was the Bronco II a safe car, a safe vehicle, from a rollover standpoint?

I think you could ask me that question about any car at all, especially the SUV, and the answer that I have to give you is that all cars that are offered for sale in America have to meet the federal motor vehicle safety standards. If those standards are valid and if they are real, it means that any car that comes off of the assembly line in America is safe. ... In those areas that are measured, it is safe. Now, if there is some unmeasured area, that's something else. ...

The federal government has been unable to develop a rollover standard. There is no standard measurement that we have that can tell you whether any vehicle is safe, from a rollover standpoint. So since you have no standard, since you don't really know what it takes to make a vehicle roll over and not roll over. ... You know lots of things, you know if it's higher and the center of gravity is off the ground higher, certainly it's going to be less stable than a low center of gravity. But there's so many things that we don't know. You can't make the judgment then that this vehicle, from a rollover standpoint, is safe or unsafe. You can't do that. Hopefully one day that will happen, but we can't do that today.

The implication of what you just said, though, is that because there's no rollover standard, there is no limit to the degree to which a car can be prone to roll over. ... Is that the implication of what you're saying -- that basically, any car, no matter how prone to roll over, would be safe by your definition, because there's no standard?

There is no question that if you had a car just like an envelope, and it goes straight up in the air, I think anybody could figure out that's unstable. If you had one that's just totally flat, of course, obviously that would never roll over. It doesn't come out like that in the real world.

Let me give you an example. The car that is the widest and the lowest center of gravity is the Chevrolet Corvette. And because that thing is so low to the ground, we know that, compared to an SUV, that it should never roll over. [But] it has one of the highest rollover rates of all the cars on the road. Why? Because it's driven by very aggressive young drivers, and they drive it in a way that causes it to roll over, because of the way they handle that particular piece of equipment. So you can't just look at this, divorcing it from all reality.

The reality is, if you say "OK, there's no federal motor vehicle safety standard. You can therefore meet any standards you want to," the answer is, "Wrong." And the reason it's wrong is there is another part of NHTSA called the Defect Investigation System. If you have one car that has fatalities that are totally out of sync with all the other cars of its type, and it sits out there like a sore thumb because it's killing people -- other cars are not killing people -- that car will be declared defective and it will be off the road. ...

Now I want to ask you again, because you said this fairly provocatively when you said that a car "is safe if it meets the federal standard by definition." Specifically, about the SUV rollover problem -- doesn't this mean there should've been a standard [even if it was less than perfect] to apply to SUVs, so that we could somehow measure the stability of these vehicles, so that the public has some idea of which ones are less safe than others?

You cannot compare unless you have a valid device for comparison. The problem is, no one has been able to develop a valid device for comparison. And so, yes, intuitively I have some ideas about what makes a vehicle roll over and not roll over; intuitively I have that. But I really cannot prove that I'm correct. ...

And there are people now who are clamoring, saying, in fact, there's a law out right now which says NHTSA has something like two years to develop a dynamic rollover standard. They've been trying to develop a dynamic rollover standard for 25 years. Not only can't NHTSA technically come to grips with this from an engineering standpoint; the government of Canada can't do it. The governments of all of Europe -- and I have talked to all of them, every one of them -- all of Europe cannot do it. The government of Japan cannot do it. We have not been able to find out a way to do this.

So do we develop some half-baked standard, like the airbag, which was too early, and kill more people just because the advocates are out there screaming? Or do we simply say, "Time out, folks. When we get it right, we will field it. We're not going to play games with the American public just to make you happy and to advance your agenda."

But the American people, then, don't have any information. ... I mean, until last fall, when there was a static stability factor measurement -- and that's really, frankly, buried on the NHTSA website -- the American public had no way of knowing not only a) the difference between SUVs, but b) that SUVs are more prone to roll over in general.

Oh, I don't agree with you. The owner's manual tells you that, if you read your owner's manual. And there's a little thing on your dash, a little sticker up there, and it tells you, "This is not like a normal car." ...

I submit to you, the American public isn't stupid; they aren't dumb. You can get in any vehicle like that and drive it and know that it's different from driving a Cadillac.

Let's talk about the [defect investigation] side of NHTSA. [Can you give] a very short description of what this process is, what it's meant for, and how it works?

The defect investigation system is a very good system. ... It [goes like] this: A car comes off the assembly line. The car is fine, don't have a problem with it. But as you get on the road two or three years, a problem develops, let's say with your brakes or transmission, for example. The defect system is designed to identify that this has happened, and then to take action to see that that is fixed at no cost to the consumer.

And what is the actual procedure within NHTSA? A petition is brought?

Usually, 75 percent of the time, it's not from a petition or from somebody outside the agency. Most of the time the agency is the one that spots the problem, and they do this using statistics. They have an organization in the agency, a very fine organization, called the National Center for Statistical Analysis. This is the federal repository of traffic safety data. Anywhere in the world if you want to know about data in the United States, this is the place you come. These people are working along with the Office of Defect Investigation, and they're looking constantly -- and I mean daily -- for trends. Is there some vehicle that suddenly is having crashes that has never happened before, or that is not in keeping with the percentage of crashes of other peer vehicles? And so that's what happens. That's how usually they identify a problem.

So you determine a defect in part, at least, by comparing one vehicle to the other vehicles in its class. Where do you draw the line between being the worst of the group and being an outlier, I mean, having a defect? How do you understand that?

This is a judgment call. ... You can identify that there was a problem, a safety defect with this vehicle. You can identify that by just watching trends. Is it crashing as much as other vehicles in its peer group? But there's another way you do it also. NHTSA is also looking at all of the traffic accident reports that are filed that have fatalities across the nation.

Better still, an example while I was there, we had one accident while I was there that was so violent and the occupants were killed, and it happened in such an unusual way that we initiated a safety investigation, a defect safety investigation on that one accident, that one crash. We didn't have any statistical data. But we said, "If this happened, something is really out of kilter, and we've got to jump on this right now and find out what it is."

So it wouldn't be enough that you were the worst in the class to trigger an actual finding of defect -- you had to be an outlier, off the charts?

No, perhaps I misled you. When you look at the comparison, all that tells you is you may have a problem. That's all you can get out of that. You can't determine any defect out of a comparison. You may have a problem. Then NHTSA's job at that point is to launch into an investigation of that motor vehicle and to look at it in terms of engineering and in terms of all of the scientists that NHTSA has, and say, "Are we really looking at a vehicle that has some kind of a safety defect ... about a mechanical problem of some kind?" Then after you zeroed in, you make a determination as to whether or not that vehicle is just high in the group, or whether it is also a dangerous vehicle.

A petition was made to you when you were NHTSA administrator to initiate a defect investigation into the Bronco II. ...

The Bronco II was under investigation before I became the administrator. The work that was done on the Bronco II was done before I was the administrator. We announced the result on my watch, but I didn't make the decision, so I'm not the person who said, "Bronco II is fine." I didn't do that. We announced it but the work had been done before I got there.

So there was nothing wrong with the Bronco II as far as you were concerned? Or you just didn't know?

No, the agency went through the process with the Bronco II. ... And the announcement was very clear on all of these vehicles, which is, the Bronco II was high in its class in terms of comparison -- not the high

est, by the way. ... But we could find nothing; the agency could find nothing that would indicate there was a safety problem with this vehicle. ...

The Insurance Institute of Highway Safety [IIHS] said that the Bronco II had a three times greater fatal rollover rate than Isuzu Trooper, which was getting a huge amount of publicity at the time for being dangerous, and which NHTSA also looked at. That sounds like a very striking abnormality. Is that wrong, or did you find that to be wrong? Or is that irrelevant?

I think IIHS has their data wrong. I think they're just flat wrong. NHTSA is not crazy. The employees that are in the agency are very fine people who are dedicated civil servants; some of them have worked at the agency 25 years. And they would not sit there and watch a vehicle that was three times more prone to roll over than peer vehicles and just say, "Oh, that's the way it is." Anyone who says that I think is being downright disingenuous.

Fair enough. Were you personally involved in any way in the decision to shut down the Bronco II investigation?

It was made before. I didn't make the decision, and the answer is, no. Was I briefed, was I aware of what was going on? The answer is, absolutely aware -- but I did not make that decision. ...

You've said the Bronco II was not a defective vehicle because it was within normal parameters -- it was on the high side, but it wasn't the highest. But one of the possible flaws of the defect system is that if an entire class has a problem, it doesn't affect that. Was there a problem with light truck rollover in the period we're talking about, the late 1980s, early 1990s? After all, in your administration, all the fatalities went down except in this one area, which went up almost 30 percent. Was there a problem?

There is a problem with rollover. Ninety percent of all rollovers take place off the highway. Only 10 percent happen on the highway. You know what this means? This means if you can keep your car on the highway, you won't roll over. And the way it happens off the highway is the driver loses control. He goes off the highway, the tire starts, say, bunching up into the sod or it hits a curb or something, and it rolls over. It trips it; that's what we call "tripping" it. The vehicle is going sideways, it trips, and it rolls over. But that happens after you leave the highway. So when you say you're going to come up with a rollover regulation that's going to prevent rollover, what you're saying is you are going to come up with a regulation that will help us with the 10 percent of vehicles that stay on the highway; the 90 percent that go off the highway aren't going to be affected by the rollover regulation. ...

So what I'm saying to you is, first of all, if you do a rollover regulation, you're going to handle 10 percent of the rollover problem -- only 10 percent. Is that important to do? The answer is, you better believe 10 percent is a big part of the problem. Should we work on it? Absolutely. Has NHTSA worked on it for 25 years? Absolutely. Have they found a way to fix it? The answer is no. Should they keep working on it? Yes.

Isn't forcing manufacturers, not to eliminate SUVs, but to build them lower and wider, a partial solution both to the highway and the off-highway problem?

The first thing is, if they did that, I would shoot them personally.

If they did what?

If they built them lower and wider. I have an SUV because I live on a mountain. I have rocks and I have tree stumps, and I go over them. And I don't want an SUV that is lower and that is narrower or wider or however you want to put it. I don't want that. I want one that's pretty much just like it is. I bought this SUV for that reason, and the SUV was designed for that reason: to go off-road. And I like to go off-road where I live, and I think people like me want that kind of vehicle. Should I have the right to buy that vehicle? Absolutely. Should the manufacturers then make that kind of vehicle? Yes. Is it more dangerous than a vehicle that is lower and wider? Yes. I'll take the tradeoff.

But 90 percent -- above 90 percent -- of the people who buy SUVs never go off-road with them. And is it your contention that those 90 percent of the people don't deserve protection from rollover, but that they ought to know, just intuitively or commonsensibly, that these things are more dangerous, and accept that risk?

My contention is that the American people are not stupid. They buy what they want to buy. They know that vehicle is higher. And if [after] all of the TV press over the last 20 years you don't now know that an SUV is not like a car, something is wrong with you.

I believe in the American people. I think they have the right to buy SUVs, and I believe that if they don't want them, they won't buy them. That's the way supply and demand works. ...

The justification for regulation is that the market doesn't have appropriate understanding and can't exert the kind of pressure it needs to, to make the automakers do what's right. Do you contend that people have a sufficient knowledge of the rollover problem to protect themselves? Or should there be some kind of role for regulation in this issue?

If I could figure out a way to make SUVs safe, there would be a regulation today. Congressman Timothy Wirth tried that [in] 1986. (He became a senator later, by the way.) He attempted to do that. He wrote NHTSA a letter and he said, "Let's use this little measurement device that I've come up with, a little formula..." He didn't come up with [it], actually, some plaintiff attorneys came up with [it]; they were using it in court cases. And he said, "Let's use this little formula, and then all cars that fall outside of the formula, we'll declare them as unsafe." We can do that today, by the way, with the SUVs.

Well, it meant that every pickup truck in America was unsafe. Of course, the congressmen and senators from Nebraska and Kansas and Oklahoma, they didn't think much about that idea. In fact, it eliminated just about every kind of car that it was even close to a Jeep. That wouldn't work. Can we do it today? Sure, we could do that today. Make sense? No. American people stand for it? You don't believe that for a minute.

You weren't there at the time, but four or five departments within NHTSA endorsed, in whole or in part, the Wirth petition. They said it is possible to do that.

Wrong. Somebody has misinformed you. If you look at the internal notes -- which are public, by the way -- that were taking place, there was only one associate administrator in entire NHTSA that stood up for the Wirth petition; only one. The others said, "The idea makes sense; could you do it? No, not today. But if we could do it, should we? The answer is yes." Same thing as today. We ought to do that. It's a good idea. We can't do it. And if you can find somebody who can do it, you will become a hero.

Is it possible, in this case, that the perfect is the enemy of the good? I mean, a static stability factor based standard would not have been perfect. No one argues that it's perfect. It misses a lot of things. But it might have been good in the sense that it would've lowered and widened these cars. It would have encouraged the automakers to find ways to make them safer, and people would still be able to buy SUVs, probably, with enough ground clearance to go up your hill.

... There's something you ought to take a look at. NHTSA did an internal evaluation of SSF. They did it around 1987. It is a damning indictment of that formula. ... I don't think that you play around with people's futures, with their lives. I don't think you play around with serious injuries, with human beings, just because you have a cute little formula plaintiff attorneys like.

By the way, plaintiff attorneys paid to have that formula developed. It was developed by two folks working conjointly; they paid the bill. As soon as they developed the formula and physically wrote it out and published it, the plaintiff attorneys used that in all their court cases and still use it today, in all their court cases. So I've got tell you, I think it is a bad way to go. We do not know that that will work. We do not know that it'll save lives. ...

Of course today, right now, static stability factor is the basis of a NHTSA rating system. Do you have a problem with that?

Sure, I have a problem with it. I told NHTSA I had a problem. ... One of the salient reasons that I have for why there's a problem with it is this: I said, if you really know enough to establish a rating system for stability, you know enough to come up with a regulation for stability. Now let's have the regulation and the rating system. And if you can't do the regulation, then let's not have the rating system, because the rating system seems to tell the public we have a solution for you. And the answer is they don't have a solution. This is a game you're playing with the public, and it's wrong to do that. You're giving the public false hope, and you should not do that. And if that system is good, then let's have the regulation, and let's make it happen.

Two NHTSA researchers, Anna Harwin and her boss, Keith Brewer, produced a paper within NHTSA that found a correlation between that stability factor and incidence of rollover. Do you disavow that work?

I don't disavow the work. NHTSA has disavowed the work. When that paper came out and it was published, NHTSA itself wrote another paper, an evaluation of that one. And it says, "Yes, this is very interesting and, yes, it is part of the puzzle; no question about it. But you can't use that by itself."

And by the way, if you read that paper, at the bottom it says another very interesting thing, which is, it says, "We really weren't able to look at the totality of the problem when we made the statements we're making in this paper. And if we had looked at all of the information, maybe we might have come to a different conclusion."...

Let me ask you about CAFE [Corporate Average Fuel Economy] standards, particularly as they apply to light trucks. Do you support them, and if not, why not?

It is not a question of supporting CAFE. The act was passed in 1975. It makes good sense, which is we're going to reduce America's dependence on foreign oil. The way we're going to do this is we're going to increase the mileage per gallon, or the range, that you get on your car. I don't think anybody can be against that.

But what Congress did not do was to say to the American people very, very clearly what the tradeoffs are. The tradeoffs are these: You cannot get that kind of increased mileage except by reducing the size and the weight of the vehicle. You can do some things in making the internal engine, the transmission work a little smoother, take a little less energy. You can do that; that helps. You can make the car more aerodynamic; that helps.

But if you're going to save gas, there's only one way to save it, and that is you're going to have to take a thousand pounds, on the average, out of cars. That's what we did. They took a meat cleaver and cut off the trunks of the cars and that's why we have all these ugly cars running around with the flat trunks in the back. That was done by the federal government and NHTSA and CAFE. That's a CAFE car. ...

The American people put up with that. They did fine. But there was a tradeoff that Congress really never shared with the American people. And the tradeoff is the one I learned in the military, which is this: If you take a tank and you run it into a Jeep, the occupants of the Jeep lose. And so what happens on America's highways through CAFE is suddenly we had cars a thousand pounds lighter running into crashes with cars a thousand pounds heavier, plus pickup trucks plus vans, plus 18-wheelers. This was the tradeoff.

Now, what the National Academy of Sciences says is that on the average, every year since then, we have lost about 2,000 people killed on our highways because we are saving gas. What we have done is we have traded barrels of oil for body bags. That's what's happening on our highways today. ...

So my question is this: Why haven't we debated this nationally? Why hasn't the media? Why hasn't Congress debated this? The American people have not been given an opportunity to weigh in on this thing. Do the American people feel that that's a reasonable public policy position to tradeoff 2,000 -- let's make it 1,000 -- to trade off 1,000 people a year, human beings dead, in order to have less dependence on foreign oil and more miles per gallon on my car? Is that a fair tradeoff? I don't think the American people have ever weighed in on that. And if you take it in another way, you're talking 25-plus years since that has happened. And if you take 1,000 a year, that means we have killed more people with CAFE than ... the terrorists did on September 11.

The original reason for treating light trucks differently than cars was because they were used differently; they were clearly industrial. [To hold them] to the same gas mileage was illogical in 1975. Hasn't that changed utterly? And if it has, if the use of these vehicles is now essentially [the same as] cars, what sense does it make to treat them as something other than passenger cars?

I don't think it makes any sense to treat them different than passenger cars. Even when I was in NHTSA, they were primarily used for such things as farms, for example. They were used for carrying things. A lot of women now just love riding around in their pickup trucks as their personal transportation. I think that since we have made these passenger cars, they should be treated like passenger cars. I don't see any reason why not. My only thing is, though, we need to tell the truth. We need to tell what the cost is going to be with the tradeoff. ...

In your career as NHTSA administrator ... have you tried to make the American people understand that tradeoff?

Oh, boy, did I ever. Whew. And I got beat up something terrible for trying to do that. Yes, when I was the administrator I sure did, and that was offset testing. I went into offset testing and I said, "I want to show the American people on television what it looks like when a big car runs into a little car." And we did that, by the way, everybody ran it, all the networks. It was a great topic for a couple of weeks. And then it died away. I wasn't able to sustain it. I just wanted everybody to understand there is a tradeoff; you are killing people. ...

Tell me the story of how you specifically tried to inform the American people. How did the offset test happen?

CAFE was on the front burner in Congress. We had a number of senators, Metzenbaum, Bryan, and some others who were just pushing this as though it was a baked Alaska. They just loved this idea. Joan Claybrook, Clarence Ditlow, that whole crowd, that whole group, they were all out there beating the drums screaming about CAFE, dependence on foreign oil, increasing fuel economy.

I was sitting in NHTSA, and my job was, I was in charge -- I was Mr. CAFE, I owned it. They were all talking about it; I owned it. And so, since I owned it, I was responsible, and I looked at it, and I said, "OK, what is this doing to the American people?" Because, as I told you before at the beginning, there's only two questions you ask: Are you saving lives? Are you preventing serious injuries? I said, "Does CAFE save lives? Prevent serious injuries?" My engineers all stood up and they said, "Absolutely not." They said, "We're killing people with CAFE," says, "This kills people on the highways."

And I said, "You mean I'm responsible for killing Americans, and the Americans don't even know that I'm doing this because the Congress told me?" And they said, "Yes, that's right." And I said, "The American people are going to know if I can inform them. I'm just going to tell them." I said, "Then if the answer is go to it and kill them, they're going to know."

And so I turned all of my public affairs people and some other folks and I said, "We've got to come up with some testing." And I said, "I want the testing of the average large car before CAFE and then the average small car after CAFE." I didn't pick the worst case -- I secretly I wanted to pick the heaviest car and the very lightest new CAFE car. But I didn't do that; I got the average so they were fair tests.

We ran these things together and we filmed them and we released the results to the network. People just screamed. I mean Capitol Hill, the senators, called me, congressmen called me, the Claybrooks just went bananas; everybody did. They said, "You're trying to kill this wonderful program." I wasn't trying to kill the program; I was trying to inform the American people. Unfortunately, I couldn't sustain the effort, and the decision was made to continue. And that's what Congress does. The president signed it; I salute.

You say you didn't pick the heaviest and lightest, but that is kind of what you did. My understanding is that you picked -- and I want to show you this, because I want you to have a fair chance of understanding what I'm talking about -- this seems to indicate that you picked for this test a Crown Victoria, which was [in the] top 5 percent heaviest cars on the road and two cars, Subaru Justy and Suzuki Swift, which were the lowest 5 percent. Is that a fair understanding of that? ...

I don't know.

Look at that and tell me if that's your handwriting - clearly, it seems to be -- that you were selecting those cars for the test.

I may have selected them.

I'm not saying there's anything wrong with it. I'm just trying to establish...

I'm not saying there's anything wrong with it, either. I don't think there's anything wrong with it at all.

Did you intentionally select very heavy and very light vehicles to increase ... to get the video you needed to get?

The answer is [that] I can't answer that. I think it's a good idea to do that. But I must tell you I felt that I was getting the middle of it. If you're showing me something that says, "No, you didn't get the middle, you got the extreme." The answer is, "OK." ...

Is it fair to say that you were trying to create a piece of video that would dramatically illustrate the point?

I was trying to give a message to the American people. And the message is: "If you crash a big car into a little car, the folks in the little car suffer." That was the message I was trying [to give]. And I say, "And that is because we were cutting off the weight, and through CAFE, we are making cars smaller."

And it's fair in doing so to pick a car, two cars, which represented the extremes of heavy and light, as far as you're concerned?

I thought I was picking something in the middle. If I picked the extremes, so be it. I did it; it's done. I don't have any qualms about that, and I don't feel bad about that. The message... In fact, that's good, that's better, because the American people need to know what the tradeoff is. That's the point, and that's what we ought to be talking about: the tradeoff. We ought to be talking about, right now, about "Is that right? Is it right to kill 2,000 Americans a year to increase fuel economy on motor vehicles? Is that right?" I don't think it's right, by the way. I think it's wrong, and I think that the Congress has done the wrong thing in this. And I don't think the American people even today understand what's happened to them aside from the fact, by the way, that the cars are ugly. ...

There does appear to be documentation that the tape of these offset crash tests were rushed out within two weeks of the testing being done, but the results which showed that the actual occupants of the small car would have survived the crash were not released until six months later. Now is there anything unusual about that? Does that connote intentionally withholding the results? Do you remember?

I don't remember a whole lot about it, but I remember that we were under a deadline, because Congress was passing legislation. And if you didn't get the message out before Congress passed the legislation, there was no sense in doing it. So this was getting the message out, and that's what we did.

There was also a narration put on the test. Do you remember the narration? How unusual was that? You approved it?

I don't remember at all.

Would it have been unusual to have a narration put on a Department of Transportation crash test like this, specifically saying this shows the danger of legislation on CAFE?

I don't know if it... It probably was unusual to do it. I think it was a great idea to do it. I liked it. I thought it was the right thing. I thought the message ought to go out, and I thought that would help the message. So the answer is, I'm in favor of it whether it's the thing normally done or not. It's a good idea.

Did you have dealings with the Coalition for Vehicle Choice [CVC] on this matter?

No.

You don't remember meeting with them on this issue when you were NHTSA administrator?

No, I think we talked. I don't remember it. I have a feeling that we talked after this came out, but I'm not sure.

Because CVC then used the footage that you...

Yes.

... in a TV spot or a radio spot. But obviously the footage was from a TV spot.

If the question is, did we make this for the Coalition for Vehicle Choice, the answer is that's the dumbest thing I've ever heard. No matter how strongly I felt about it, I wouldn't do it. I think it's wrong to do something like that. They have enough money. They can make their own.

So you didn't know about the ad in advance?

No.

You were surprised when it came on. ... Like everyone else, that was the first time you'd seen it?

Oh, yes, I hadn't seen it before. I think somebody... You know, this is very murky now, but you are asking me. I think somebody may have called and told me they were doing it, so I think I knew it was happening. But I didn't know anything about that until after we had put out the video.

Do you agree that it would have been illegal for NHTSA, or against regulation for NHTSA to participate in this CVC ad campaign?

That's a dumb question. You don't want to ask that question. Do you want to ask that question?

Why is it dumb question?

Because it's so obvious. You don't use taxpayer money to push the agendas of special interest groups or private groups; you just don't do that. I mean, it doesn't make any sense to do that. And again, these folks have their own money; they don't need the government's money. They can make up whatever they want to make up.

What if a tape of that test had been sent to CVC? Would that constitute bad practice?

How would I know? Whatever the lawyers say, I'll stick with it.

But releasing it to the networks, convening a roundtable as you did -- that's different than working with CVC? That comes under some... That's not lobbying, in the strict sense of the word?

Oh, I think if I were going to accuse me of lobbying, I would accuse me of lobbying Congress because I sent one of those to every congressman and every senator on the Hill. So I think, yes, you would have accused me of lobbying, because I was lobbying. I was pushing to get the message out to the American people as to what was happening. ...

But, you know, if you look at this from a different standpoint, all of these special interest groups, all of these so-called consumer advocate groups, owned NHTSA for four years under a gal by the name of Claybrook. They -- she -- they owned it.

She worked for... You talk about a revolving door. She worked for Ralph Nader, goes to the federal government for four years, and goes back to work for Ralph Nader. That's a revolving door. So they owned it for four years. And if CAFE and rollover was so important to them, those were the four years they should have gotten the legislation through, could have gotten it through, and didn't get it through.

They didn't even get through side-impact legislation. Side impact had been ten years in the process before I became the administrator. I got it through, but it had never been gotten through before that. These folks had their shot at it. They blew their shot at it. And they ought to shut up and quit carping on the margin. ...

You say that a heightened CAFE would have forced the downsizing of the vehicle fleet, and as you dramatically illustrated, when big cars hit small cars, the small cars wouldn't be able to come through as well. But it's interesting because, in essence, by not heightening CAFE, by not creating the rollover standard and everything, NHTSA and the government and everyone else has enabled manufacturers to build bigger and bigger and bigger vehicles, bigger and bigger SUVs. And hasn't it been borne out that, because of the increase in size of the vehicle fleet, there's more danger on the road for smaller cars? ... Unintentionally, you all have enabled bigger and bigger cars. And that makes smaller cars more vulnerable?

It's sort of almost the philosophical standpoint you come at this. If you think the federal government has the right to dictate to the American people the size of the car they should buy and use, then it makes sense to do this. But if you believe, like I do, that the American government doesn't have that right to dictate to the American people that you can only -- this is not the Soviet Union -- "You can only ride in that size car."

I don't think that's right. I don't think that's the American way. I don't think that's what should be happening. Is it happening? The answer is yes. If you build a larger car, we have more disparity between the size of vehicles in crashes, yes. By building larger cars, can you increase the number of people being killed? The answer is yes. I just think that we need to think this thing through, have an open public debate nationally on it, and however it comes out, then we go and do what that is. ...

What is the role of the safety advocates and plaintiff attorneys? Have they become in some way de facto regulators in our system?

The plaintiff lawyers and the safety advocates attempt to become the de facto regulators. If they had their way, they would in fact be the regulators. And they would regulate it in a way that they can litigate it in court and make an awful lot of money out of it. That's what drives them -- the greed, the money that's involved in the process.

The good news is that there have been some brakes on them -- maybe to a degree through ineptness -- but the Congress has had some brakes on them. NHTSA has resisted in some areas; not as much as it should have, by the way. This is, I think, going to be a major problem for us in the future -- NHTSA's failures to really resist some of the encroachment that is taking place.

So I think if you were to ask me what should the role be of the plaintiff lawyers? The answer is they should not have a role. They do not have a role in safety. They have a role in the legal system. They have a role in trying to redress grievances. They have a role there. But in terms of the safety system, they don't have a role, and they should not be playing a role.

So when they say, "We are the reason that Ford has redesigned 2002 Explorer to be lower and wider, we are the reason that various safety improvements have happened because we've made it so expensive for the automakers," you reject that assertion on their part?

Yes, I reject that. ...

Even NHTSA has said that they often rely on plaintiff lawyers and safety advocates, that those people should bring them -- because these guys are getting cases -- they should bring them notice of a problem when they see it. In this [Ford-Firestone] case, the [plaintiff] lawyers have admitted that they didn't on purpose. But is that the way it should work?

It should not work that way. Lawyers cannot be trusted to bring safety problems to NHTSA. And the reason is they're trying to make money, and they want to hide these safety problems, so they can spring them in court in the litigation process. So these folks aren't interested in saving lives; these folks are interested in making money through winning court cases. And they will tell NHTSA as little or as much as they want them to know. And you can be sure they will safe-side the information and not give them what they need to know.

I've seen many, many cases where lawyers have sat on what they claimed were -- they weren't, but they claimed -- some great safety problem. And they never told NHTSA about it at all, because they didn't want NHTSA to do anything about it.