And as Bob has said, there isn't an empirical study by the Board that demonstrates, at a statistically significant confidence interval, that increased direct investment leads to increased failures. There is such a study on the losses to the FSLIC fund. You know, almost nothing in life is decided on the basis of statistically significant empirical evidence.

If you wait for statistcially significant empirical evidence on an insurance fund, you've gone out of business -- broke -- three years earlier on loss experience, before you learn that indeed cigarette smoking, in that first sample of thirty-seven institutions, is statistically significantly correlated, once you reach higher sample sizes - four years later and

allow the progression of lung cancer. That type of thing.

In terms of what people normally make decisions on, this is, to my mind, one of the most monumental cases for adopting a rule that exists. Beyond that, there is statistical information that, at an absolute minimum, if you were an insurance company, would cause you to zap them.

MR. GRAY: Well, let's talk about that. If you were an insurance company, a private insurance company, and you had this kind of information, and you knew that you were very low in your reserves, protect against claims, what would you likely do?

MR. BLACK: Well, first thing you would look to whether it was perfectly rational or not, if you're simply an insurance company, is your loss experience. And on the basis of your loss experience, you'd either have such a prohibitive premium, or an exclusion from coverage.

MR. GRAY: Which is something that you cannot do.

MR. BLACK: Which is something we cannot do --- an exclusion from coverage already just from the loss experience. But here you've got more than that. You do have the studies that you've talked about. You haven't done the statistical significants tests yet to know whether the difference in the means is statistically significant. Given the very small number of institutions, thirty-seven, you may find - I don't know what the answer is going to be. Certainly we will do it and should do it. But the point is, as I said, you wouldn't wait around. If you had that difference in two groups, as an insurance company, you'd do something about it.

If you looked at Professor Benston's own studies, and you looked at his studies of the relationships between direct investment to total assets, an analysis of failed savings and loan institutions -- where he finds a statistically significant correlation between increased direct investment and failed institutions, at the 95 percent confidence interval for certain of its year's, and, though he does not state it, above the 90 percent confidence interval for other years. Particularly considering that he is a consultant for folks posing the rules. I'm referring to Specification 15 of his '85 study, that the Board did the review on and found that I think it was the '85 data, that was statistically significant, at approximately the ninety slightly above the 90 percent confidence interval.

MR. WHITE: A change in the institution.

MR. BLACK: Yeah, that's the time -- that's all I'm saying.

MR. GRAY: Okay. Let me just get back to this. Just get back to this. We have $2.2 billion today. We have other cases coming up that are going to reduce that. We all know that. And substantially, in the absence of recapitalization.

MR. BLACK: Well, you can't afford the additional losses.

MR. GRAY: Well, this is the point. If you're an insurance company, you call a halt to this until you can figure out a way to deal with it.

But it always seems very strange to me, Bill -- and Bob -- that we have a good deal of evidence here, that institutions that were heavy into direct investments -- I mean even 5 percent, or in excess of 5 percent, who have, or are on the way to reaching their demise, and we have such a small margin for error here, where the stakes are so high for this fund, that I guess I have difficulty understanding, I guess, from a prudent man's point of view, how it is that whatever lack of statistics there may be, we know we have a problem here, and we know that it involves high levels of direct investments.

Now, we also know and recognize that direct investments can, indeed, provide a higher rate of return, and they are therefore inherently correct me if I'm wrong -- a higher rate of return generally means that there may be higher risk involved. And if there is higher risk involved, there is higher risk of loss, losses that the FSLIC simply cannot continue to sustain without going out of business, or without some other kind of action being taken by other parties.

And I suppose, then, from a prudent person's point of view, as the operators of an insurance company, that we operate not on behalf of the thrift industry--we operate this on behalf of the American people, on behalf of the American people. This is not the property of the thrift industry, it's the property of the American people. And here we are, it seems to me, discussing whether or not, to anyone's particular satisfaction, the riskiness of these kinds of investments is so risky that -- or maybe not as risky as it might be -- when we have almost no money to take care of claims, anyway. It seems rather strange.

MR. SAHADI: Well, I think a key point here is, we're not prohibiting this type of investment. All we're doing is establishing a supervisory threshold, that once people bump up to that 10 percent or twice their net worth, they come to a supervisory agent and demonstrate a case that what they're doing is -- does present some higher return to the institution, does have some portfolio diversification, is something that they have the management capability to pull off.

So, in effect, it's saying, even though we're in this very difficult situation of having a $2.2 billion fund in a and obviously with problems much higher than that; we're saying you can still sky-dive but we just want to check your parachute to see if it's got any holes in it.

MR. GRAY: All right. And by the way, it seems to me, from an underwriter - insurance underwriter's point of view, that one might liken this supervisory threshold, where supervisory agents become involved, to an underwriting device. It's a kind of underwriting device. I gather that 60 percent of all those institutions which have applied to exceed the regulation, or have been allowed to do that those, I assume, that have not been allowed to exceed the threshold, have financial problems, and other problems that relate to the criteria that supervisory agents use.

MR. SAHADI: Well, many of the problems were withdrawals. When asked for further information, the institution just withdrew.

MR. GRAY: In other words, it's likely, then, when "push came to shove," they didn't want to go through with this because it might raise concerns about the very criteria that's involved here. And it sounds to me like it's reasonable criteria, such as the quality of management, the track record of the institution, the financial strength of the institution, the potential impact on the FSLIC. Those seem to be rather reasonable steps of criteria.

-

MR. SAHADI: Well, we also have more of you know if an institution doesn't obviously want to get involved in every deal, coming in and ask the PSA for approval, I could understand how that, you know, particularly a well-run institution would resent that type of you know "Big Brother" intervention. If institutions come in and provide a general business plan for this type of investment, then they can you know don't have to come in on a case-by-case basis.

MR. GRAY: Well, let me this is a long Board meeting. You may have all the time you want.

MR. WHITE: A11?

MR. GRAY: You know, I received a letter on December 16th from the Chairman of the Federal Reserve, Paul Volcker, and that is available, by the way.

It says, "You've asked for my reaction to ..." our proposal. "Recent experience, as well as more general 'structural' concerns about the role and activities of federally-insured depository institutions, strongly suggest to me that such extension is not only appropriate but strongly in the interest of the financial system generally as well as the FSLIC."

And he talks about the types of direct investments covered by the restrictions in our existing rule. He said: "I know some of those risks for insured institutions can be reduced by proper supervision.... But such supervision, in my view, is not a substitute", not a substitute "for limiting total exposure. Your existing rule requires that federally-insured institutions must obtain approval from the Federal Home Loan Bank System to make direct investments in excess of 10 percent of assets. Such a limitation, which can be overriden in specific cases of demonstrable strength and favorable experience, seems to me, if anything, liberal. Certainly, it is far less restrictive", Mr. Volcker says, "than proposals for regulating investments by bank holding companies in real estate properties and real estate development projects which the Federal Reserve has been considering."

And he goes on to say: "Our current proposal," that is to say, the proposal of the Fed "would require that direct investments in real estate properties and real estate development projects be made only in nonbank subsidiaries of bank holding companies and would limit the equity investments of bank holding companies in such subsidiaries to 5 percent of their consolidated primary capital."

Now I would compare that, 5 percent of consolidated primary capital, which is a limit, to 10 percent of assets in this current regulation, which is one of the reasons why I think this regulation is not nearly as strong as it ought to be. I think the Federal Reserve is taking more of the kind of approach that I, personally, would like to see taken. He goes on to say: "The real estate subsidiaries would be permitted to lever their positions up to five times their capital, but the total exposure of the holding company (the direct investments of the real estate subsidiary plus loans and guarantees of the real estate subsidiary or any other subsidiary of the holding company to properties and projects in which the real estate subsidiary has a direct investment) would not be permitted..."

Then the holding company, this 5 percent that would be upstream, 5 percent of capital, "would not be permitted to exceed 25 percent of the holding company's consolidated primary capital, itself. In addition, in the case of individual projects in which an organization has an equity investment, its total exposure, as defined above, would be restricted to 10 percent of the holding company's consolidated primary capital."

He said: "This approach will, of course, be reviewed and perhaps modified in the light of comments received. But I think, "he says "the proposal reflects the sensitivity of the Federal Reserve Board to the risks inherent in this type of direct investment. Those risks seem to me relevant to savings and loan associations and their holding companies, as well, operating on the basis of deposits obtained from the public and protected by federal deposit insurance." In sum, "he says "the existing FSLIC rule restricting direct investment of federally-insured thrifts, seems to me to be an appropriate reflection of legitimate concerns." And again he points out, "It is far more liberal" --- far more liberal "than we have thought appropriate for bank holding companies. Consequently, consideration "he says to us, "in my judgment, should be given to changes having the effect of tightening the limitations."

Now, the head of our sister regulatory agency, Mr. Seidman, wrote a letter on December 4th, and he said: "In accordance with your request for my opinion, "-- the opinion about the Direct Investment Regulation, that the Bank Board has--he said, "I'm writing to indicate my support for your proposal to extend for two years the Bank Board's regulation. The limit appears to be more than adequate for the industry's needs and is far more liberal than what we are considering for banks which we supervise. The rule should enhance your efforts to contain the magnitude of risk in the system and protect the reserves of the FSLIC. Such concerns are of overriding importance in the present circumstances." So, we have letters here from the Chairman of the Federal Reserve, the Chairman of the FDIC, commenting directly on our proposed regulation, the Fed having a very restrictive rule compared to this one, with no threshold to exceed, and the FDIC is considering an approach which, I believe, would be something like 50 percent of capital, which is almost infinitely different from a 10 percent of assets threshold.

So, that's why I've said we have, ladies and gentlemen, an extremely liberal regulation here, a regulation that, in effect, constitutes a form of underwriting for the risk that the FSLIC must bear. And we are discussing the inadequacies of studies relating to the empirical evidence we have about losses that are involved, losses that from these few institutions that very substantially exceed our resources, today or tomorrow, to deal with these

cases.

I think it's instructive that our sister regulatory agencies have looked very carefully and seriously at this problem and they, obviously, are quite concerned, or they would not be as restrictive as they are with respect to direct investments for their own insured institutions. Now.

MR. WHITE: Thank you, Ed. You, somewhere back there, asked about risk and what risk was. Let me reserve comment on that. I want to come back to the issue of risk.

Bill, even farther back, you had made some comments. First, you had talked about the risks from cigarette smoking. You didn't mention about the risks of being downwind from two cigarette smokers. (Laughter)

But I don't get hazardous duty pay for this! However, I am newly come to the insurance business and I'm not sure I know what an insurance company would or wouldn't do under certain circumstances. But I would guess that an insurance company tries to get it right, tries to identify what the source of its possible losses are, and wants to find the least costly way of dealing with that. I think we're at least most of us -- are agreed, that the first test would be to try to price the insurance properly to reflect risk. If that isn't possible, then we have to think about other mechanisms to try to deal with risk.