Rehearing and Rehearing En Banc Denied October 16, 1990.
POSNER, Circuit Judge.
Central National Bank is one of two national banks in Mattoon, a town of 20,000 in downstate Illinois. After an investigation of the bank's trust department, the Comptroller of the Currency, who regulates national banks, issued in 1987 a notice of intent to revoke the bank's permission to provide trust services to its customers. The notice was issued pursuant to 12 U.S.C. § 92a(k), which authorizes the Comptroller, upon notice and hearing, to revoke trust powers that the bank has "unlawfully or unsoundly exercised." After a hearing before an administrative law judge, the judge found that the bank had committed many violations of the Comptroller's regulations and engaged in many imprudent practices, but decided that it would be enough to order the bank to cease
At oral argument the lawyer for the bank surprised us by asking us to expel a reporter for a Mattoon newspaper whom the lawyer had spotted upon entering the courtroom. After giving counsel for both sides an opportunity to address the motion, we huddled briefly and denied it. The bank had earlier moved to seal the record, and we had granted that motion, but the motion had said nothing about conducting the oral argument in secrecy. A party who wants such extraordinary (albeit not completely unprecedented, Application of United States, 427 F.2d 639, 641 (9th Cir.1970)) relief must ask for it in advance of argument, not only to give the other party fair warning and the bench an opportunity for due deliberation but also to give the press — which may be the only adversary of the request for secrecy — a chance to be heard. Globe Newspaper Co. v. Superior Court, 457 U.S. 596, 609 n. 25, 102 S.Ct. 2613, 2621 n. 25, 73 L.Ed.2d 248 (1982); Gannett Co. v. DePasquale, 443 U.S. 368, 401, 99 S.Ct. 2898, 2916, 61 L.Ed.2d 608 (1979) (concurring opinion). Perhaps the lawyer was surprised that a reporter would want to attend the argument, but he should not have been.
Even if the motion to close oral argument had been timely, it would not have been granted. The bank's interest in secrecy would have had to be weighed against the interest of the press, as the representative (albeit self-appointed) of the public, in access to judicial proceedings. The proceeding before the Comptroller, the decisions of the administrative law judge and of the Comptroller, and the briefs in this court have all been secret, so that if the oral argument were also secret the public would have no inkling of the Comptroller's findings or the contents of its order until we issued our opinion — and not even then, if the bank asked us to seal it and we acceded to that request too. The bank is concerned that public knowledge of these things may impair its standing with its customers. Indeed it may, but perhaps it should, for it is information material to the decision whether to do business with the bank. The private and the social interest in secrecy must not be confused. An individual, or, as here, a firm or other institution, may be injured by disclosure of a disreputable fact about it, such as a record of crime or incompetence. But the very revelation that injures it may be useful to others, the potential transactors with it; so the net social value of the information, when the interests of all who are affected by it are summed, may be positive. Evaluated socially in this manner, the bank's interest in keeping the bad news about its management secret is meager in relation to the claims of a free press for access to governmental proceedings. Joy v. North, 692 F.2d 880, 894 (2d Cir.1982); Brown & Williamson Tobacco Corp. v. FTC, 710 F.2d 1165, 1180 (6th Cir.1983); cf. In re Knoxville News-Sentinel Co., 723 F.2d 470, 477 (6th Cir.1983). The sealing of the record strikes the balance between the competing interests as favorably to the bank as could be thought proper, if not more so.
The case might stand differently if the Comptroller, or the Federal Deposit Insurance Corporation, which insures deposits in national (as in other insured) banks, was asking for secrecy. The history of bank "runs" could be thought to argue for controlling public access to information
We turn to the merits of the petition for review. The bank does not deny that it has engaged in improper practices in its trust department, but pleads a variety of mitigating circumstances. The improper practices not only are numerous but also reach back many years and were only partly corrected in previous remedial proceedings that resulted in the entry and subsequent dissolution of a cease and desist order in the early 1980s. The improprieties range in gravity from the technical (failure to comply with certain reporting requirements) through the mild (minor violations of the terms of trust instruments) to the grave (deliberately overvaluing securities in one of the bank's common trust funds, the consequence being to overcharge customers for units of the fund). The bank has very little to say in rebuttal to these charges, and nothing worth discussing. It trains all its guns on the alleged impropriety that precipitated the Comptroller's decision to order the bank to shut down its trust department — the purchase in 1985 of 20,000 shares of Eagle Bancorporation for a common trust fund administered by the bank's trust department.
The purchase price of $220,000 constituted more than 20 percent of the assets of the common trust fund. To finance the purchase, Malcolm O'Neill, the bank's trust officer, sold blue-chip stocks and bonds held in the common trust fund. Eagle was a modest (net worth of $14 million), unlisted company owning small, unlisted banks and seeking to acquire more of these; its stock was narrowly held and thinly traded. The president and (indirectly) principal stockholder of the Central National Bank of Mattoon, James Singer, owned stock in Eagle, as did another director of the bank and O'Neill himself. In addition, as O'Neill well knew, Eagle was trying to acquire another bank of which Singer was a major owner.
The Eagle shareholder from whom O'Neill had purchased the 20,000 shares for Central National Bank's common trust fund agreed in writing to repurchase the stock within one year, upon demand, at a price slightly above the bank's purchase price. Within that year an attempt by Eagle to raise money by a new offering of stock failed, and O'Neill belatedly began to worry about the prudence of the bank's investment. But he unaccountably failed to exercise his option to sell the stock back to the seller. He claims to have made an oral demand, which — assuming he did make it — was ignored. He made no effort to follow up with a written demand. So far as appears, the stock remains in the bank's portfolio to this day, having paid during this period a single dividend of $1,500.
The imprudence of the investment does not lie in the fact that Eagle is not a blue chip. It is a myth that prudent investing requires limiting one's portfolio to the stocks or other securities of the strongest companies, the companies least likely to go broke. The prices of the stocks of weak companies are bid down in the market until those stocks yield the same risk-adjusted expected return as the stocks of the strongest companies; otherwise no one would hold stock in a weak company — the prices of such stock would fall to zero. The risk to which we have just referred is the volatility of the stock in relation to the market, or in other words the stock's sensitivity to market swings. This "market risk," what finance theorists call "beta," is different from insolvency risk — that is, how weak the company is. But like insolvency risk, it is compensated risk, since most investors (including most beneficiaries of trusts) are risk averse and therefore demand a premium to hold a security that has an above-average market risk. The prices of stocks with high market risk are bid down accordingly,
But the proposition that the price of a stock that involves above-average insolvency risk or above-average market risk will be bid down to compensate investors for bearing these risks assumes that there is an efficient market for the stock in question, that is, a market in which there are enough well-informed buyers and sellers, actual and potential, to ensure that prices reflect true values as affected by risk preference as well as by expected earnings and other factors. We do not know whether the market in stock of Eagle Bancorporation is efficient, but probably it is not. Its existing shares are thinly traded and its effort to sell additional shares flopped. The fact that in all of 1985 the only price at which Eagle stock changed hands was $11 (and there were several transactions during this period, not just the purchase by the bank's common trust fund), even though the value of a company is rarely constant over an entire year, is further evidence that the price of this stock is not a reliable market price. With O'Neill and his bosses owning stock in Eagle and the big boss owning stock in Eagle's target as well, there is no reason to suppose that O'Neill's decision to buy 20,000 shares in Eagle for the common trust fund reflected an informed and unbiased effort to determine the value of those shares relative to the market value of the stocks that O'Neill sold to pay for the shares. Nor is there evidence that O'Neill in fact made any such effort. The problem, then, is not that Eagle was not a blue chip company, but that, as a prudent investor in O'Neill's shoes would have realized, it may very well not have been worth as much as $11 a share.
The purchase also impaired the diversification of the bank's common trust fund. We pointed out earlier that to investors who are risk averse, unnecessary risk is a cost, which they will willingly bear, therefore, only if compensated by the prospect of an additional return on their investment. We identified two types of risk: insolvency risk, and market risk (beta). Both are compensated risk, in that price will adjust to make securities containing even large amounts of such risk just as good investments as securities with less such risk. There is, however, a third type of risk, which is uncompensated, and that is risk that can be diversified away simply by holding a larger, more balanced portfolio. It is uncompensated because it is not a real burden to the investor, since he can eliminate it at trivial cost by the design of his portfolio. By placing a large fraction of the trust fund's assets in the shares of a small company, itself undiversified (for some companies, of course, are highly diversified — they are like mutual funds), O'Neill significantly reduced the fund's diversification without conferring any offsetting benefit, in the form of a higher expected return, on the fund's beneficiaries. In time, if and when they discovered that the bank was failing to maintain a diversified portfolio, they might withdraw from the fund, but until then they were simply worse off as a consequence of the bank's purchase of Eagle stock.
We have wandered a bit afield, for the Comptroller, perhaps prudently, does not emphasize the effect of the Eagle purchase in reducing the diversification of the common trust fund. So far as the trust services of the Central National Bank of Mattoon, Illinois, are concerned, the propriety of the bank's practices is determined by the Illinois law of trusts, 12 C.F.R. § 9.11(a); and while the duty of a fiduciary to diversify his investment portfolio is generally recognized, Restatement (Second) of Trusts, § 228 (1959); Hamilton v. Nielsen, 678 F.2d 709, 712 (7th Cir.1982), leading us to hazard a guess in Hamilton that the duty was a part of the Illinois prudent man rule, we acknowledged that there were no Illinois cases on point. Id. The Illinois Appellate Court has misread Hamilton as holding that "Illinois law does not impose a duty on a trustee to diversify the trust assets." McCormick v. McCormick, 180 Ill.App.3d 184, 198, 129 Ill.Dec. 579, 588, 536 N.E.2d 419, 428 (1988). This leaves the status of the duty to diversify in Illinois law in a fog that neither the statute itself ("in ... investing ... property for any trust ... the trustee shall exercise the
Yet the wrong was substantial. And while the remedy is severe, we disagree that the Comptroller has decreed the bank's demise. National banks are not trust companies, and only a minority of them (though a large one — 40 percent) have trust departments. But while Central National Bank is wrong to argue that the revocation of its trust powers must endanger its survival, the Comptroller is equally wrong to argue that because the department is shown on the books of the bank as losing money the bank will be better off by the revocation. Unless the trust department is being run for the enrichment of the bank's officers — which the Comptroller does not argue — we can be confident that if the department is really conferring a net detriment on the bank it will be closed down voluntarily. Well, reasonably confident, for it is possible that the department is a loser but that revocation would be such a blow to public confidence in the bank (word of the Comptroller's order, and of the findings underlying it, would spread quickly in a town of 20,000 once the curtain of secrecy was lifted, as it now has been) that the losses in the department are worth bearing. More likely the trust department is a loss leader whose services generate other business for the bank that is profitable enough to more than cover the losses in the trust department. (We note in this connection that 80 percent of the bank's trust customers are also depositors, that they hold 30 percent of the total demand deposits in the bank, and that 10 to 20 percent of the bank's loans are to them.) In either event the trust department's losses would be illusory in an economic sense, being completely offset by the profits on other banking business that the operations of the trust department enable.
So probably the bank will be hurt by the revocation. Part of the hurt, however, will come not from the abolition of the trust department as such — for there should be other ways even in Mattoon to attract banking business besides offering cut-rate trust services (we are not even told whether the other national bank in the town has a trust department) — but from the negative inference concerning the quality of the bank's management that the public may draw from the Comptroller's order. But as should be evident from our earlier discussion of the bank's desire to conceal the nature and basis of the Comptroller's action from that same public, we consider this a private harm rather than a social one. The harm to the bank is offset — quite possibly more than offset — by benefits to others, namely consumers who will use the information to make an informed judgment on whether to take their banking business across the street to the other national bank in Mattoon, or to some other bank. The banking public of Mattoon is entitled to information concerning the prudence and probity of the bankers with whom it deals. Perhaps it will conclude that the mismanagement of the trust department bespeaks only an inability to attain the high fiduciary standards required of trust officers and not an inability to manage the other components of banking service. Then the damage to Central National Bank will be slight. The other possibility, of course, is that the bank's customers, or at least many of them, will decide to take their banking business elsewhere. Either way, the impact
Even if the sanction is not so fell as the bank contends, it might still be disproportionate to the gravity of the wrongdoing. While admitting that O'Neill acted imprudently, the bank points out that there is no evidence that Singer or other members of the board of directors knew about the purchase of the Eagle shares until the repurchase agreement had expired. Suspicion, yes; evidence, no. And the Federal Reserve Board, which has final authority to determine whether O'Neill should be removed from his office or indeed excluded from banking, 12 U.S.C. § 1818(e)(4), has decided in an unpublished decision that he should not be removed or excluded, because he received no personal gain and demonstrated no personal dishonesty and because his action did not harm the bank, 12 U.S.C. § 1818(e)(1)(B) — directly. The harm came through the Comptroller's response to O'Neill's action, and this the Board, giving the statute a surprisingly narrow interpretation, thought too indirect. Meanwhile O'Neill has resigned from the bank, which argues that the problem has thus been solved (no other officers or employees having been implicated in O'Neill's actions) and that revocation of the bank's trust powers is overkill. It adds that no one has lost money because of the purchase of the Eagle shares, though it would be more accurate to say that no one knows whether in the end this imprudent purchase will yield a return equal to that of the securities that the bank sold to finance it. (A company's prospects are impounded in the market price of its shares, but we do not know whether shares in Eagle have a market price.) Crying unfair surprise, the bank points out that the Comptroller, perhaps in oblique response to the savings and loan scandal, has toughened his stance on the appropriate sanctions for imprudent practices. The bank also lays great weight on the refusal of the administrative law judge to order revocation, adding that it is unfair for the "prosecutor," as it describes the Comptroller, to impose on his own initiative the very punishment that the (administrative law) "judge" decided would be too severe.
The suggestion that revocation of the bank's trust powers was too drastic a sanction for an isolated lapse by one man whom the bank no longer employs ignores the fact that the purchase of the Eagle stock was the culmination of what the Comptroller, with much support in the evidence, describes as the bank's "escalating disregard" for the requirements of prudent management of trust accounts. The violations go back many years, with periodic improvements invariably being followed by serious backsliding. The bank has found it impossible to institutionalize prudence in its trust department, and one of its most egregious violations, the Eagle incident, was also the most recent. But even if we were inclined to agree that the Comptroller was being excessively severe (we are not so inclined), we would not be justified in setting aside his order. The general posture of a court reviewing agency decisions is deferential; how deferential depends among other things on the nature of the issue. The more openended it is, and hence judgmental, and the more a sound exercise of judgment requires specialized knowledge possibly possessed by the agency but certainly not by the court, the lighter the judicial hand. Here, as in general (for the Sentencing Guidelines do not apply here!), the choice of sanctions is judgmental — an exercise of administrative discretion entitled to judicial respect, American Power & Light Co. v. SEC, 329 U.S. 90, 112-18, 67 S.Ct. 133, 145-48, 91 L.Ed. 103 (1946); Butz v. Glover Livestock Commission Co., 411 U.S. 182, 188, 93 S.Ct. 1455, 1459, 36 L.Ed.2d 142 (1973); Carlson v. United States, 879 F.2d 261, 263 (7th Cir.1989); Communication Workers of America v. NLRB, 784 F.2d 847, 852 (7th Cir.1986); Panhandle Cooperative Ass'n v. EPA, 771 F.2d 1149, 1152 (8th Cir.1985) — and depends on particulars of the banking industry to which the generalist federal judiciary is not privy. The relevant discretion, moreover,
As for the supposed unfairness of casting the Comptroller in the role of a judge in his own cause, this disregards not only the precedent of the Federal Trade Commission, which both issues the complaint and on appeal from the administrative law judge decides whether the complaint has been sustained and a remedial order should issue, but subtle differences between the Comptroller's and Trade Commission's functions which should dissipate any lingering doubts concerning the propriety of combining prosecutorial and adjudicative functions in the Comptroller's person (if not the Commission's). The Commission — say, in policing under section 5 of the Federal Trade Commission Act the truth of advertising representations — does not assume responsibility for the welfare of the advertising industry. As a result it has a natural tendency to judge its success by the number of remedial orders that it issues, and in this tendency lies the potential bias in favor of upholding the complaint it has issued of which critics of the Commission have long, though so far unavailingly, complained. The Comptroller of the Currency, however, has comprehensive authority over national banks. First National Bank v. Smith, 610 F.2d 1258, 1263-65 (5th Cir.1980). He charters them; he audits them; he enforces the laws relating to them; in short he "is charged by the national banking laws with the execution of all laws of the United States relating to the organization, operation, regulation and supervision of national banks." 12 C.F.R. § 1.1; see also 12 U.S.C. §§ 26, 93, 93a, 161. Those banks are his wards, and his only wards; if they fail in droves, he will be blamed. It is not in his self-interest, bureaucratic or otherwise, to endanger their solvency by amputating their powers to engage in profitable lines of business. It is in his interest, rather, in deciding whether to modify or uphold or throw out a remedial order against a bank recommended by one of his administrative law judges, to weigh the harm the order is likely to cause the bank against the order's deterrent effect, which will enhance the safety of the banking industry for the future. If the evidence presented to the administrative law judge points toward lenity in the exercise of the Comptroller's remedial powers, we can think of no reason why the Comptroller would be deflected by his initial prosecutorial role in the case from exercising lenity.
But at the very least (the bank argues) the Comptroller should have given it notice that from now on serious imprudences would be met with revocation of the powers imprudently exercised. However, the statute under which the Comptroller acted, 12 U.S.C. § 92a(k), lists revocation as the sanction for imprudent exercise of trust powers, though naturally the bank does not question the Comptroller's power to impose a lesser sanction. Although we can find no cases on the issue of power, the Comptroller's practice of imposing lesser sanctions for less grave violations exists, is apparently of long standing, and makes much practical sense — sanctions whose severity is disproportionate to the wrongfulness of the sanctioned conduct either overdeter or are never invoked and become toothless. Far from questioning this power, the bank, involving cases like Motor Vehicle Mfrs. Ass'n v. State Farm Mutual Automobile Ins. Co., 463 U.S. 29, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983), Continental Web Press, Inc. v. NLRB, 742 F.2d 1087, 1093 (7th Cir.1984), and (the case factually closest to this one) Essery v. Department of Transportation, 857 F.2d 1286, 1293 (9th Cir.1988), argues that the Comptroller departed from its unpublished enforcement policy by not giving the bank an extra bite at the apple before revoking its trust powers. It submits no particulars concerning
The Comptroller did not mislead the bank, and the statute gave the bank all the notice to which it was entitled. An expectation of lenient treatment is not an expectation on which the law founds rights. There have been occasional suggestions, most recently in Bork, The Tempting of America: The Political Seduction of the Law 96 (1990), that the sudden revival of a long forgotten law carrying harsh penalties (the theme of Shakespeare's Measure for Measure, where the law in question imposed the death penalty for fornication) might encounter a defense of desuetude. But if there is such a defense it is surely reserved for more extreme cases than this one. The revocation of the bank's trust powers was not the first crackdown on the bank. The bank had been disciplined before. It could not be entirely surprised that escalating disregard for its legal duties would call forth an escalation in sanctions.
Yet this last argument, which we have just seen fail, is closely related to an argument that conceivably if improbably might have succeeded if made in timely fashion. In its opening brief to us the bank for the first time invoked 5 U.S.C. § 558(c)(2), which forbids (with exceptions that, for the sake of argument only, the Comptroller concedes are inapplicable here) the revocation of a license unless "before the institution of agency proceedings therefor, the licensee has been given ... an opportunity to demonstrate or achieve compliance with all lawful requirements" (emphasis added). See Gallagher & Ascher Co. v. Simon, 687 F.2d 1067, 1074 (7th Cir.1982). The Comptroller gave the bank ample opportunity to show that it was in compliance with the requirements for conducting a prudent trust department, and he argues that the statute requires no more, because the duty is disjunctive: the agency must give the licensee either an opportunity to demonstrate compliance or, if but only if no such opportunity has been given, an opportunity to achieve compliance. This is a plausible reading, for otherwise it would be unclear when revocation could take place: achieving compliance could take forever, and "opportunity" is undefined. There is some judicial support for the Comptroller's reading, H.P. Lambert Co. v. Secretary of Treasury, 354 F.2d 819, 821 n. 2 (1st Cir.1965); Parchman v. Department of Agriculture, 852 F.2d 858, 865 (6th Cir.1988), and some nonsupport. Atlantic Richfield Co. v. United States, 774 F.2d 1193, 1200-01 (D.C.Cir.1985); Blackwell College of Business v. Attorney General, 454 F.2d 928, 933-34 (D.C.Cir.1971). It is odd that so seemingly fundamental an issue has never been authoritatively resolved — indeed, never clearly faced up to. We need not try to resolve it in this case. The bank failed to make the argument to the Comptroller. It raised it for the first time in this court. This was too late, United States v. L.A. Tucker Truck Lines, Inc., 344 U.S. 33, 37, 73 S.Ct. 67, 69, 97 L.Ed. 54 (1952); In re Chicago, Milwaukee, St. Paul & Pacific R.R., 799 F.2d 317, 334 (7th Cir.1986); Wallace v. Department of Air Force, 879 F.2d 829, 832 (Fed.Cir.1989), and the argument is therefore forfeit.
The petition to set aside the Comptroller's order revoking the bank's trust authority is