Mr. BEALES. The system is maintained by the three credit bureaus and they would be able to tell you how many people have in fact opted out, I mean, how many people are on the list.

But we do not have that information.

Senator STABENOW. Okay. And would you make any changes from your perspective in how that is working, that whole process for consumers?

Mr. BEALES. As I say, the Commission doesn't have legislative recommendations at this time. I do not think we have seen anything that has seemed to us to be a particular problem in that area, that really needs to be fixed.

Senator STABENOW. Okay. Thank you very much.
Chairman SHELBY. Thank you, Senator.

I want to go back to the issue of credit report accuracy.

Is there an acceptable tolerance level for errors, and what is that, if there is? In a risk-based system, there has to be some tolerance, but what is the threshold?

Mr. BEALES. The statutory standard focuses on procedures, reasonable procedures to assure the maximum possible accuracy.

It doesn't set a numerical threshold. Even if the error rate is very low, if it is cheap to fix, you should fix it. But if it is really difficult to fix or really expensive to fix, if there is no reasonable way to correct it, or no reasonable procedure that would prevent it, then that would be acceptable under the statutory standard.

But it is not a numerical threshold. It is a balance of

Chairman SHELBY. How would you excuse me a minute. You said if there is no reasonable way to fix it. But what if it were so prevalent, it called for fixing?

I am not saying it is, but you said if there is no reasonable way to fix it. First of all, assuming that the number of errors are small, we understand that.

Mr. BEALES. Let me back up because I think what I should have said is, there is no reasonable way to prevent it.

Chairman SHELBY. Okay.

Mr. BEALES. Because if there is an error and it is called to your attention, you have to fix it.

Period. End of story.

Chairman SHELBY. That is correct.

Because accuracy is important.

Mr. BEALES. Because accuracy is important. Absolutely.
Chairman SHELBY. Right.

Mr. BEALES. But the reasonable procedures focus on what kinds of steps can you put in place to keep that from happening in the first place?

Chairman SHELBY. Is there any way to gauge what is or should be an acceptable error rate? You said that they do not do it statistically.

In other words, you do not do it numerically. You do not say that the error rate is I am just throwing this out there-3 percent or 5 percent or one-half of 1 percent or one-hundredth of 1 percent. Is that what you were saying a minute ago, that you do not gauge that?

Mr. BEALES. There is no bright-line standard in the statute of what is acceptable. Even if the error rate was a hundredth of a per

cent, if you could avoid that for free, then under the statute, you have to do that.

It is a question of what kinds of costs do you have to incur, what is reasonable to do to avoid that particular error.

Chairman SHELBY. I think Senator Dodd asked you the question of, something to the effect, how many complaints did you have at the Federal Trade Commission a year? And you said, around 2,000, more or less, on a yearly basis.

Mr. BEALES. Per credit bureau.

Chairman SHELBY. Per credit bureau. Six thousand? Three credit bureaus?

Mr. BEALES. I think it is probably a bigger number than that.
Chairman SHELBY. A larger number.

Mr. BEALES. Let us get you the precise number.

Chairman SHELBY. Can you furnish that for the record because we are building on it.

Mr. BEALES. Yes, sir.

Chairman SHELBY. What are the considerations or trade-offs involved in the calculation of an acceptable rate of error?

For example, maintaining as much participation as possible within a voluntary furnisher system versus accurate record of consumer credit history, and ultimately, the appropriate pricing of credit.

Accuracy goes to the very heart of all of this. And it would seem to me that not only would the credit bureau, or whatever, but also the credit-checker, would want their reports to be accurate.

The user of that information-let's say it is a mortgage company or a bank or something-they would certainly want it to be accurate, wouldn't they?

Mr. BEALES. They certainly would. I think everybody in the system, consumers, users, credit bureaus, benefits from accuracy. I think that is absolutely right.

Chairman SHELBY. Benefits from accuracy, starting with the con

sumer on.

Risk-based credit pricing-I think Senator Sarbanes alluded to that earlier. I think that we all recognize the many positive, and there are many, developments associated with technological advancement.

Technology has made our credit markets remarkably responsive to consumer demand, as Senator Bennett would have shown us with a bigger chart, right?

[Laughter.]

Senator BENNETT. Right. Colored chart.

Chairman SHELBY. Yes, a larger chart. That said, just to get a handle on just where technology has taken our credit markets, could you explain or expand a little on the use of risk-based pricing and how much more prevalent its use today versus 1996 and 1971, if you could?

Mr. BEALES. I cannot do it in a quantitative way, but clearly, qualitatively, there has been substantial change. I think particularly in 1970, at the time the statute was first passed, the dominant model and the way most credit decisions were made was you applied for credit that was available on a fixed set of terms and you were either approved or denied on that same fixed set of terms.

I think probably the predominant model today is you apply_for credit. The terms-sometimes you are accepted or rejected. But with growing frequency, the terms you are offered, whether it is the interest rate or the credit limit or some other aspect of the credit arrangement, depend on the risk that that individual borrower presents.

And the higher the risk, the worse the terms.

Chairman SHELBY. Sure.

Mr. BEALES. That is a much more common model today than it was. Certainly in 1970, the information-processing technology and the information-sharing technology simply wasn't in place to support that kind of system on any very large scale.

Now it is. Now it is done. It is much more differentiated pricing of credit and insurance products based on the risks that a particular consumer may pose.

Chairman SHELBY. I think the use of risk-based pricing offers numerous benefits to consumers. You alluded to that. Credit is now offered to many people who were previously deemed unqualified. Hence, his chart a minute ago, I think.

And credit pricing is much more tailored now to each individual, more so than it used to be.

Would that be a fair assessment.

Mr. BEALES. I think that is correct, yes.

Chairman SHELBY. But the use of risk-based pricing also raises issues about the continued effectiveness of some aspects of the Fair Credit Reporting Act.

You mentioned, if you want to refer to your written testimony, how accuracy, and I will quote:"Was, and remains a core goal of the Fair Credit Reporting Act."

You then indicate and I will quote:"Adverse action notices. . .are a key mechanism for maintaining accuracy."

With the use of risk-based scoring, however, a consumer may qualify for credit, but not at the best terms.

By not making an outright rejection, creditors as I understand the system-do not have to send an adverse action notice and credit applicants may then never become aware of the need to examine their credit reports.

Does this cause you any concern about the continuing relevance of the adverse action process?

For example-let me see if I understand it.

Let's say they check my credit and I do not have A number one credit like Senator Bennett's father or like he would have liked to have had. Right? And they come back and instead of telling me that, they say, we will offer you something based on the risk.

Is that the way they do that? The credit risk as they perceive my credit, rather than an outright rejection.

Mr. BEALES. Yes. And that is a counter-offer.

Chairman SHELBY. Yes, it is a counter-offer. Explain how that works. That avoids the necessity of the adverse

Mr. BEALES. It depends on what you do with it at that point.
Chairman SHELBY. Sure.

Mr. BEALES. If you reject the counter-offer, then that is adverse action.

Chairman SHELBY. That is right.

Mr. BEALES. And if it is based, in part, on a credit report, they have to tell you.

Chairman SHELBY. Sure.

Mr. BEALES. If you accept the counter-offer, then, because adverse action is tied to the definition of adverse action under the Equal Credit Opportunity Act, then there is no adverse action because you got the credit that you wanted.

Chairman SHELBY. You got the credit maybe not that you initially wanted, but you got a deal and took it.

Right?

Mr. BEALES. That is right. I think that raises a difficult tradeoff. We are thinking about the issue, but we do not have a recommendation.

Chairman SHELBY. Tell us how you are thinking about it.
Mr. BEALES. Yes, sir.

Chairman SHELBY. Just for the record.

Mr. BEALES. On one hand, if you are not getting the best terms, and that is based on the credit report, you need to get notice.

Chairman SHELBY. Maybe I wouldn't deserve the best terms. Mr. BEALES. That is right. You may not get the best terms based on accurate information rather than inaccurate information.

But if it is based on inaccurate information, you need the notice to trigger your right to look at the file.

But in a pure, risk-based system, where the best person out there gets the best price, and everybody else gets somewhat worse terms-if you think of it at that extreme

Chairman SHELBY. But everything's a risk. It should be basedif it is based on risk, somebody's more creditworthy and has worked hard, diligently to pay their bills, as opposed to, say I hadn't, they should be rewarded, should they not?

Mr. BEALES. They should. I agree with that completely.

Chairman SHELBY. Because they are less of a risk, say, than I would be.

Mr. BEALES. I agree with that completely. But if you say, well, giving less than the best terms is in some sense an adverse action, and we have said that about insurance where there is not the linkage if you get insurance on less than the best terms, we have said that is adverse action and you have to give notice.

But in a completely risk-based system, that means everybody gets notice all the time, except the best risk. And that degrades the notice because it no longer serves the function of saying, there may be something unusual here, which it does now, and it does under the present system.

And there is a balance between giving notice when people need it and not overwhelming people with notices that say there might be something in your credit report, because you could say that to everybody all the time.

Chairman SHELBY. Sure. But going back to what I mentioned, and you mentioned in your testimony earlier, risk must be gauged accurately, as best we can.

Mr. BEALES. Certainly.

Chairman SHELBY. That is what the system is about, is not it? Mr. BEALES. Certainly.

Chairman SHELBY. Is gauging the risk as accurately-and you gauge it accurately based on the information that you pull up, do not you?

Mr. BEALES. That is right.

Chairman SHELBY. Or should.

Mr. BEALES. That is why the availability of information is important and it is why the accuracy of the information that is provided is important, because it gives you a better gauge of what the risk really is.

Chairman SHELBY. Are there more or less adverse action notices now than before?

In other words, are there more counter-offers?

Mr. BEALES. there is offsetting influences. I think there is more counter-offers that are accepted and that would push it down. But there is more credit, and that would mean more denials.

Chairman SHELBY. Can you furnish that information for the record to the Committee?

Mr. BEALES. I doubt if we have it. But if we know, we will be happy to furnish it for the record. We will see if we have it. Senator BENNETT. Than before what?

Chairman SHELBY. We are talking about before, let's say, 1971. Let's say 1996, the 1996 Amendments.

The timeframe. Before the 1996.

Senator BENNETT. No. I want to see the benchmark.

Chairman SHELBY. I amended my question.

Before 1996 and after 1996.

Mr. BEALES. Okay. My suspicion is that we do not have any information to answer that question. But we will look, and if we do, we will certainly provide it.

Chairman SHELBY. The use of risk-based pricing is what we are getting at.

Technology has made it much easier to transfer, as we all know, massive amounts of information and data, thereby increasing the capability of credit reporting systems in many ways.

We benefit from that.

Can you comment on whether or not technology has enhanced the overall accuracy of credit reports? And do you have anything at the FTC-have you done a study on that?

Mr. BEALES. We have not done a study on that.

Chairman SHELBY. It should be more accurate, shouldn't it?

Mr. BEALES. I think technology has clearly enhanced the speed of the reinvestigation process. It is made it possible to reinvestigate, I mean, just the automated information exchange.

Chairman SHELBY. It is the reaction to something.

Mr. BEALES. Yes, yes. But the technology has made possible automated information exchange both to get the information, to report back to the furnisher that there is a dispute, and then for the furnisher to report back the truth.

That can all happen much quicker than it used to.

We do not know of any objective measure of how accurate the information is in credit reports that would be available over time to say firmly whether it is more or less accurate, what is actually been the trend in accuracy.

Chairman SHELBY. Senator Sarbanes.