very near danger of such a liquidity position as to requiring their closing. We would do the same thing if we found that practice to exist in any bank, whether it is the C. & S. or First National City of New York or whatever it is.

Mr. WELTNER. You have no regulations pertaining to this at the present time?

Mr. SAXON. Not specifically; no, sir.

Mr. WELTNER. Do you anticipate that if this bill passes that the situation would possibly require the adoption of regulations to that end?

Mr. SAXON. It could be, and certainly if we found any instances of this, we would not only take action, remedial action with respect to that, but perhaps generally as we did here recently on instructions to examiners on CD's obtained through money brokers. We just foreclosed them.

Mr. WELTNER. Do you know of any widespread instances since the passage of the Glass-Steagall Act of abuse by larger banks in selling underwritten general obligation bonds to correspondent banks?

Mr. SAXON. No, sir; I do not know of a single instance.

Mr. WELTNER. Well then, you say the danger here that has been raised is not a substantial danger, is that correct?

Mr. SAXON. I think it is unsubstantial; yes, sir.

Mr. WELTNER. If there is no substantial danger in the bill, I am wondering what benefits might be derived? In your statement, of course, you say that "any measure which would lower the cost of such financing would be of significant benefit to State and local governments." That is a worthy aim. Two years ago we had testimony from, I think, four witnesses from the State of Georgia, probably generated by the school certificates decision from your office. Can you supply us with any documentation as to the reduction in "spread" by having banks in the position to underwrite revenue certificates?

Mr. SAXON. Yes, sir; and from the Governor of the State and other authorities in the State. We will be delighted; yes, sir. Governor Vandiver himself on an issue last year.

Mr. WELTNER. I am familiar with the testimony of our State auditor on that matter. But other than that one instance, is there any other documentation you can give to the committee?

Mr. SAXON. If I may, sir, we will look.

Mr. WELTNER. Mr. Chairman, with the Chair's permission, that might be helpful information.

The CHAIRMAN. Without objection it is so ordered. (The information requested follows:)

REPLY OF HON. JAMES J. SAXON TO QUESTION OF MR. WELTNER

Question. Are there any other cases (besides Georgia State authorities) in which commercial bank participation has resulted in savings to the issuer? Answer. One example that has been cited is the experience in financing the Chicago Civic Center. During the last week in June 1963, $87 million of bonds were sold by competitive bidding to pay for construction. The civic center is to be leased from the public building commission by the city of Chicago with rent payable from tax revenues. In line with the ruling in the case of the Georgia authorities we determined that these bonds are, in fact, general obligations and may be underwritten by commercial banks.

The winning bid was submitted by a syndicate managed by the Continental Illinois National Bank and which included many large commercial banks. The winning bid was a 3.33-percent interest cost to the city of Chicago while the next

best bid was 3.4 percent. It is, of course, impossible to determine accurately what the winning bid would have been without bank participation in the action, but if the second bid had been low, the additional cost to the city would have been approximately $60,000 over the life of the bonds.

Another example is the sale by the Port of New York Authority of $25 million of bonds in November 1964. The highest bid submitted by the investment banker groups competing for this issue was a cost to the authority of 3.51 percent. The Chase Manhattan Bank submitted a bid at 3.43 percent thus saving the authority 8 basic points. (Chase bought the entire issue for its own account).

In March of this year the North Kern (California) Water Storage District sold $2.25 million of bonds to Security-First National Bank of Los Angeles at a net interest cost of 3.82 percent. The next lowest bidder offered 3.84 percent.

Mr. WELTNER. I have no further questions, Mr. Chairman.
The CHAIRMAN. We might limit the questioning to 2 minutes each.
Mr. WIDNALL. I suggest asking for permission to sit during the
session this afternoon.

The CHAIRMAN. If we get more time we will extend the time.
Mr. Gettys is next, I believe.

Mr. GETTYS. Mr. Chairman, in view of the time limitation, I would like to ask only one question.

Mr. Saxon, the commercial banks actually now by administrative order are doing what is requested under this bill, isn't that correct? Mr. SAXON. Yes, sir; in many instances.

Mr. GETTYS. For 30 years the interpretation was different, is that correct?

Mr. SAXON. Yes, sir.

Mr. GETTYS. The merits or demerits of the bill are not what bothers me so much, Mr. Chairman, but that the legislative intent of the Congress as of 1933 by administrative action has been voided. That bothers me. Could you explain why?

Mr. SAXON. That is a question of what was the legislative intent of the Congress at that time, and with respect to what securities, whether or not, as a matter of fact, whether the revenue bond was known. It was practically unknown at that time.

Mr. GETTYS. It appears that an interpretation standing for 30 years would be pretty good evidence that the intent of Congress has been rightfully construed.

Mr. SAXON. Not necessarily.

Mr. GETTYS. Why not come to the Congress with this bill before action is taken to implement an administrative interpretation? Other Comptrollers before have not interpreted in this manner.

Mr. SAXON. The issue has been up before, as our records show, and this is the first instance

Mr. GETTYS. I mean when does the law of Congress stand?

Mr. SAXON. It is a question, Mr. Gettys, of what was the legislative intent of the Congress and what was the disposition of it

Mr. GETTYS. Wouldn't that be a court matter rather than an administrative executive matter?

Mr. SAXON. It could be. It is within the entitlement of any party, investment banker, or other, to raise it as a matter of judicial proceeding, if they were so disposed, but none have raised.

Mr. GETTYS. Suppose this bill is adopted. What is to prevent the Federal Reserve Board from interpreting it after we adopt it--adopt the bill, that it should not be this way?

Mr. SAXON. I would think it would be a question that here the legislative intent is clear. It is by no means clear with respect to the Glass

Steagall provision in question, by no means. In fact, very seriously
doubtful.

Mr. GETTYS. My time has expired, Mr. Chairman. Thank you.
The CHAIRMAN. Yes, sir. Mr. Stanton.

Mr. STANTON. Thank you, Mr. Chairman. Mr. Chairman, I am of the view of Mr. Harvey of the Banking and Currency Committee, that while this dispute arises between the Federal Reserve and the Comptroller I hope we can solve the point one way or the other. Mr. Saxon, it seems to me that your testimony is primarily based upon the fact that this will increase competition which will be of great aid to the cities and States and communities involved. It has been my experience in the last 8 years of my life to be in county government where we have dealt with 50 or 60 revenue bond issues. At no time can I recall where we had an issue that was not highly competitive in the bidding. Anywhere from 3 to 4, and at one time in my mind it was 8 to 10, and it seemed to me to be a very healthy situation. My question to you, Are you inferring that the investment bankers are not doing a good job in this particular field at the present time?

Mr. SAXON. I am suggesting that there would be benefit from the introduction of competition of additional banking capital, of a broadened market through the participation by commercial banks. Whereas in your instance there has been good competitive bidding, this has been healthy. There has been benefit to the community. But this has not, certainly not been the case nationally, Mr. Stanton. Mr. STANTON. Have you got statistics?

Mr. SAXON. We will do our best.

Mr. STANTON. Whereby there has only been one bid?

Mr. SAXON. We will do our best to provide this.

Mr. STANTON. I think that would be good because the crux of the matter lies in this, as far as the point is concerned. You have got to prove that the investment bankers are not being competitive. It has been my experience in Ohio that they have been competitive. (The information requested follows:)

REPLY OF HON. JAMES J. SAXON TO QUESTION OF MR. STANTON

Question. Present data on the number of cases in which there was only one bidder for an issue of revenue or general obligation bonds.

Answer. It is much more common for revenue bonds than for general obligation bonds to be sold through negotiated sales. That is, sales handled by negotiation with one underwriter rather than through competitive bidding. One reason for this, of course, is that commercial banks have not been allowed to bid for the revenue bond issues. The Investment Bankers Association reported that in 1961, 38 percent of the revenue bonds issued were handled through negotiated sales whereas only 3 percent of the general obligation bonds were placed in this way.

Even when issues are sold through competitive bidding, there are fewer bidders for revenue bonds than for general obligation bonds. Professor Fox presented data on this matter on behalf of a committee of investment bankers to the House Banking and Currency Committee in 1963. Even on the basis of the investment bankers' figures it seems clear that competition is greater in bidding for general obligations (in which commercial banks participate) than for revenue bonds. The figures show that in 1962, 82 percent of all general obligation issues of over $1 million offered for public bidding attracted four or more bidders. Only 72 percent of the revenue bond offerings of this size had four or more bids. Put another way, in 16 percent of the revenue bond issues there were only one or two bids received, while this was true in only 8 percent of the general obligation issues.

Number of bids submitted for general obligation and revenue bonds issued $1,000,000 and larger, 1962

Source: Adapted from table in "Increased Flexibility for Financial Institutions." p. 733.

In total, about 40 percent of revenue bonds are sold without benefit of competitive bidding by two or more underwriters.

Mr. ST GERMAIN. Will the gentleman yield?

Mr. STANTON. Yes.

Mr. ST GERMAIN. He asked for a specific instance. I have in my own town in 1960, we had a $2 million issue for our water system, and there was no competition. In fact, they ended up having to go to negotiations with one firm. They had no bidders. Now I say you are very fortunate but there are some communities that are not so fortunate. That is one specific instance.

Mr. STANTON. Thank you, very much, Mr. St Germain. But primarily when you get to specific cases like that we become involved too. We have been getting aid and help by investment bankers, commercial bankers, and the rest of them. Before you put up a general revenue bond into a project, it could have been the project itself rather than the lack of competition that caused no bidders. Maybe they did not like the project.

Mr. ST GERMAIN. I would say this. The commercial bankers would try to convince them to go to GO's because they wanted it, so it could not have been that bad and it has paid off very nicely to date.

Mr. STANTON. Did the commercial bank buy the bond after it was issued?

Mr. ST GERMAIN. Yes; they did.

Mr. SAXON. This is so often the case, unfortunately. That is what makes the problem so difficult.

The CHAIRMAN. Mr. Todd.

Mr. TODD. Thank you, Mr. Chairman.

Could I ask a question and ask for a reply later on?

The CHAIRMAN. Yes, sir; for the record.

Mr. TODD. This question is somewhat related to Mr. Harvey's and Mr. Weltner's question. I will read it and then, perhaps, each of you gentlemen could supply the answer. Will you kindly supply me with copies or studies of expert opinions of a technical nature establishing the effects of allowing revenue bond underwriting by the banks on, (a) underwriting spreads; (b) reduction in interest rates on revenue bonds; and (c) effect on interest rates of general obligation bonds. I think these are all interrelated matters, and if you please I would appreciate technical studies rather than the sort of opinion we have had expressed here this morning which convinces me of nothing, I must frankly state.

The CHAIRMAN. Will each of you gentlemen answer that for the record?

(The information requested follows:)

REPLY OF HON. JAMES J. SAXON TO QUESTION OF MR. TODD

Question. Discuss studies of effect of commercial bank underwriting on (a) underwriting spreads; (b) rates on revenue bonds; and (c) rates on general obligations.

Answer. To our knowledge, two such studies have been undertaken, one by the Federal Reserve and one by Prof. Bertrand Fox for the Committee for Study of Revenue Bond Financing. Both these studies used similar techniques and, not surprisingly, reached similar conclusions. We believe that both studies have the same shortcomings.

Both studies examined price data for a sample of bonds. In the Federal Reserve study the sample consisted of offerings made in the first half of 1963 in amounts of $2 million or more and which were rated A by Moody's. The Fox sample consisted of issues of over $1 million offered from 1957 to 1961.

The apparent lower yields on general obligations (when quality is held constant) is explained in part by the longer average maturity of revenue bonds. After taking account of this factor, both studies found a differential of 11 basis points by which yields on revenue bonds exceeded those on comparable general obligations. Mr. Martin testified that this differential "might be narrowed by permitting commercial banks to underwrite revenue bonds."

The sponsors of both studies suggest that this potential saving-one-tenth of 1 percent is too small to be worth bothering about. We think that the studies under discussion underestimate the yield spread between general obligations and revenue bonds, but even on the basis of their results the benefits of wider commercial bank participation in the revenue bond market seem large. There were about $4 billion of revenue bonds sold in 1954 and this amount promises to increase. A saving of one-tenth of 1 percent, multiplied by billions of dollars each year is clearly significant.

The reason for believing that these studies understate the difference in yields lies in the way in which the sample data were drawn. Both studies examined only relatively large, competitively sold, issues. This is not where we would expect the significant effects of commercial bank participation. It is in the smaller issues that broader commercial bank interest would be most helpful. Obviously it is investment in such securities by commercial banks rather than underwriting that is necessary to effect yields, but, as Prof. Roland Robinson has pointed out in his "Postwar Market for State and Local Government Securities," since "commercial banks cannot underwrite these issues, they are less enthusiastic investors in them."

The measurement of the potential effect of revenue bond underwriting by commercial banks on underwriting spreads poses difficult problems. Both studies found that spreads are virtually identical on revenue bond underwritings and general obligation bonds. They concluded that since spreads on revenue bonds are now as small as spreads on GO's, there is nothing to be gained by allowing commercial banks to underwrite revenue bonds.

Both studies suffer from a serious bias in the sample data used. Both studies consider only sales through competitive bidding and exclude sales of bonds through negotiation with one underwriter. It is in the latter case that we would