In this entry I discuss two reforms of banking regulation proposed in Financial Regulatory Reform: A New Foundation: Rebuilding Financial Supervision and Regulation, which the Treasury Department issued on June 17 and which I'm calling the "Report." The first proposal is to vest the Federal Reserve with responsibility for regulating "systemic risk." (As usual, I define "banking" to include all financial intermediation--and this turns out to be particularly important in the present context.) The second proposal is to empower the Fed to regulate the compensation practices of firms that it classifies as potential sources of systemic risk; such firms are to be known as "Tier 1 Financial Holding Companies."


The Federal Reserve and the Federal Deposit Insurance Corporation between them exercise comprehensive authority over commercial banks, particularly (in the case of the Federal Reserve) banks that belong to the Federal Reserve System (but I'll ignore that detail). Runs on insured banks are rare, because depositors are insured; and while banks have uninsured creditors, the usual sequel to a bank failure is for the bank's liabilities as well as assets to be assumed by another bank. Member banks of the Federal Reserve System can moreover protect themselves from insolvency caused by lack of liquidity (which might occur because the bank could not liquidate assets fast enough to meet withdrawal demands) by borrowing from the Federal Reserve itself, which can increase the cash balance in a bank's account with the local federal reserve bank by a computer stroke. (This is called "borrowing at the discount window," an archaic phrase that confuses people about how the Federal Reserve System operates. There is no window and "discount" just means loan.) There is little concern that a string of commercial-bank failures would precipitate a recession or depression.


The crisis that engulfed the financial system last September primarily involved what is called the "shadow banking" system, which is to say the financial intermediaries that provide bank-like services (such as lending borrowed capital) but are not regulated as commercial banks. The shadow banking system provides in the aggregate more credit than commercial banks do.


An illustration of the perils of nonbank banking is provided by the commercial-paper market. Checkable money-market accounts are close substitutes for demand-deposit accounts in commercial banks, and they pay interest, which until 1986 commercial banks were not permitted to pay on demand deposits. To be able to pay interest, the money-market funds have to be able to earn interest with their depositors' money, which they do by using those deposits to buy debt, such as commercial paper. The term refers to short-term unsecured promissory notes issued by companies that have sterling credit records, such as Proctor & Gamble, to finance their day-to-day operations. Sometimes these notes are issued directly to money-market funds, but more commonly they are issued to broker-dealers (firms that either broker, or deal in--that is, buy and sell--financial instruments), such as the ill-fated Lehman Brothers. The broker-dealers issue their own commercial paper to the money-market funds, and the cash they receive in return (that is, the money they borrow from the funds, the commercial paper being their promise to repay) is what they use to buy commercial paper from, which is to say lend to, nonfinancial companies such as Proctor & Gamble. Lehman thus was a dealer in commercial paper--an intermediary between nonfinancial issuers of commercial paper and money-market funds. Because it had a lot of risky investments in other parts of its business, it defaulted on the commercial paper that it had issued to money-market funds, that is, failed to repay the money it had borrowed from them. This caused a run on those funds--because they were not insured--until the government stepped in and agreed to insure them. Since Lehman, broke, could no longer buy commercial paper, the nonfinancial issuers drew on the standby lines of credit that they had with banks, as a result of which the banks had less money to lend to the many firms that were clamoring for bank credit in the crisis atmosphere of last fall.


Lehman and the money-market funds were, for all practical purposes, acting as banks, which is to say lending borrowed capital, but their borrowed capital was not insured and they were not regulated by the banking authorities. The Report proposes that the Federal Reserve be authorized to designate a nonbank as a Tier 1 Financial Holding Company (FHC) and, having done so, to place restrictions on the bank's capital structure, management, and operations (including its compensation practices, which I discuss below) designed to prevent a repetition of the failure of Lehman Brothers and the near failure of other broker-dealers, such as Bear Stearns and Merrill Lynch. For example, the Fed could require a Tier 1 FHC to increase the amount of its capital or reduce its debt-equity ratio (leverage).


The basis for classifying a firm as a Tier 1 FHC would be that it poses a "systemic risk," meaning that its failure, like that of Lehman Brothers, could endanger the financial system, or the larger economy. Usually this would be because of the firm's web of direct and indirect relations with other participants in financial markets, such as, in Lehman's case, the money-market funds, the nonfinancial issuers of commercial paper, and the banks that had issued standby letters of credit to those issuers.


Another example of the problem of systemic risk involves what is called "prime brokerage." Some broker-dealers provide extensive financial services to other financial firms, such as hedge funds, notably by holding the hedge fund's money, much as a broker-dealer holds an individual's investment in a customer account, while the hedge fund is between investments and needs somewhere to park its money. When the broker-dealer, normally because of proprietary trading (that is, speculating, often with borrowed capital) or other risky activity, gets into financial trouble, the hedge fund will quickly withdraw its money from the broker-dealer. For like a person with a money-market account (before those accounts were insured), a hedge fund is merely an unsecured, uninsured creditor of the prime broker. A run by hedge funds seeking to withdraw their money before the prime broker declares bankruptcy can bring down the broke. And if because of the broker's insolvency the hedge fund in turn gets into financial trouble, this will precipitate demands for redemption by the hedge fund's investors, and thus another run, with the result that the hedge fund may collapse along with the prime broker.


The simplest solution would be to forbid broker-dealers to trade on their own account or engage in any other speculative or highly risky financial activities; and perhaps that is what the Federal Reserve will do to broker-dealers that it classifies as Tier 1 FHCs; nor need it stop with broker-dealers, since other financial intermediaries that operate as critical nodes in the global finance network can also if they collapse carry much of the financial structure down with them.


The Report rejects the idea of specifying criteria for classifying a firm as a Tier 1 FHC, because it does not want to tie the Federal Reserve's hands, or enable a firm to skirt classification by keeping just under whatever threshold in terms of assets or interconnectedness with other financial intermediaries that Congress, or the Fed by regulation, might set. But by granting the Fed uncanalized discretion to subject firms to draconian restrictions, the proposal if adopted would endanger the Fed's prized political independence. As long as it just manages the money supply and regulates commercial banks, which are the instruments by which it manages the money supply, for it adjusts short-term interest rates and thus the money supply by altering banks' cash balances, it is engaged in a limited, technical activity that does not involve picking and choosing among individual firms outside the commercial-banking industry. But once it has a roaming jurisdiction to place the mark of Cain on whatever firm it deems a potential source of systemic risk, it is bound to be accused of playing favorites and invite political interference by the Administration and Congress.


The Report does not consider how the banking (including "shadow banking") industry will try to game the Federal Reserve's systemic-risk authority, but try it will. On the one hand, some banks may try to become Tier 1 FHCs on the ground that, since the Fed will not allow such a firm to fail, lest the potential systemic risk that by definition it is believed to pose becomes actual, it will be immune from bankruptcy and therefore able to borrow money at a lower interest rate than its competitors. But this seems unlikely, since recent experience teaches that when the government bails out a failing firm, it may impose conditions that wipe out not only shareholders and managers but also creditors, even secured creditors. Moreover, the Report recommends giving the Federal Reserve the power to "resolve" a failing Tier 1 FHC, a term that refers to the streamlined administrative bankruptcy procedure that bank regulatory authorities employ when a commercial bank or a thrift goes broke; as in conventional bankruptcy, the usual  consequence is that the shareholders are wiped out and the creditors recover only a small fraction of their claims. And recalling creditors' unhappy experience in the Chrysler and (in all likelihood, though it is still under way) General Motors bankruptcies, a firm classified as a Tier 1 FHC may find itself unable to borrow at attractive rates because lenders may fear that if it gets into trouble and has to be "resolved" it will be dealt with mercilessly by the bank regulators, as a macroeconomic culprit, and its creditors wiped out.


Not that it is certain that "resolution" or any other form of bankruptcy would be the fate of a Tier 1 FHC that got itself into deep financial trouble, as firms like Citigroup, Goldman Sachs, and Merrill Lynch did last September. The Fed might decide that the shock value of bankruptcy would be too unsettling for the economy, or that just the mechanics of taking over and running a giant financial institution would be too much for the Fed or other "resolver" and so the institution should be bailed out with minimum harm to creditors, as was done last fall (except with respect to Lehman Brothers). Government has problems with precommitment; it cannot tie its hands the way a private firm, by signing a contract, can do; and it does not want to. So creditors can always hope that, when the chips are down, the government will balk at allowing Tier 1 FHCs to fail and be "resolved."


Nevertheless, although Tier 1 FHCs might turn out in a crisis to benefit from their status, it seems equally and perhaps even more likely that firms that are candidates to be classified as Tier 1 FHCs will decide they'd be better off by spinning off enough of their operations to avoid the classification and so the restrictions that come with it. For example, a broker-dealer that like Lehman was both a dealer in commercial paper and a trader on its own account might be better off spinning off its trading operations, so that its shareholders would have shares in two companies, than continuing in its dual role and be subjected to restrictions that might make its trading unprofitable by preventing it from making attractive deals that were highly risky.


And the restrictions that the Fed could place on Tier 1 FHCs might make them noncompetitive with foreign banks regulated under looser standards. It is a terrible fate to be a regulated company forced to compete with a nonregulated, or less regulated, company. Candidates for Tier 1 FHC classification might do a lot to avoid being so classified.


Yet even if all firms that create systemic risk decided to shrink, or to reduce their interactions with other financial firms, systemic risk would not be eliminated. For such risk is a property of the financial system rather than of individual firms. That is, systemic risk is correlated risk. If the entire banking industry were heavily invested in home mortgages, and a housing bubble caused a drastic fall in the value of those mortgages, it wouldn't matter if the industry consisted of 10,000 banks of equal (and therefore equally small) size that had no dealings with other financial firms. The whole industry would be brought down.


We know from the events of last September that financial intermediaries can create "systemic risk," but it does not follow that we need a new regulatory regime to deal with it. The crisis was the product of regulatory error and inattention. The Federal Reserve pushed interest rates too low earlier in this decade and missed the fact that as a result there was a housing bubble in which the banking (including the shadow-banking) industry was deeply invested. Despite warnings, the Federal Reserve failed to formulate a contingency plan to deal with a possible financial collapse, and failed even to inform itself about the structure and practices of the shadow-banking industry, as it could easily have done. It was no secret that Lehman Brothers occupied a key position (along with other broker-dealers, such as Merrill Lynch) in the commercial-paper market. Bernanke and Paulson (Geithner's predecessor as Secretary of the Treasury) argued that the Federal Reserve could not have saved Lehman Brothers because a loan to it would not have been adequately collateralized. I find this hard to take seriously, as the Fed has been busy making risky loans in its programs to buy credit-card and mortgage-backed and other risky debt. However, statutory clarification of its authority to make any loans necessary to avoid financial calamity would be appropriate, if only to make it impossible for such an excuse for fatal inaction to be advanced in future crises.


There is a danger that the Fed will be distracted from its core function of managing the money supply if given the new responsibilities that the Report recommends it be given. The danger is acute because the Fed's mismanagement of the money supply appears to have been a major cause of the financial crisis. One critic of the proposal to make the Federal Reserve the systemic-risk regulator has compared it to responding to an accident committed by one's teenage son by giving him a more powerful car.


If as I believe the financial collapse is rooted in regulatory mistakes, expanding the responsibilities of the regulatory agency that made the most serious mistakes seems a perverse response. The problem was not lack of authority but lack of foresight and knowledge. And that problem, to the extent solvable, should be dealt with by enlarging the capability of the Federal Reserve, the Treasury Department, and perhaps other entities in the government to conduct financial intelligence gathering and analysis and to do contingency planning for responding to macroeconomic emergencies. I elaborate on this suggestion in the fourth blog entry in this series.


The restriction that the Federal Reserve could impose on Tier 1 FHCs that might most alarm a financial enterprise would be a restriction on the size or form of executive compensation, and let me turn to that issue. We should separate issues of compensation of CEOs and other top management from issues concerning the compensation of traders, loan officers, and other financial executives at the operating rather than the management level. The Federal Reserve would be authorized to regulate compensation at both levels, but at the top level the aim would be to make management a more faithful agent of the shareholders while at the operating level it would be to curb the risk-taking incentives of financial executives by requiring that much of their compensation be deferred. The deferred component might consist of restricted stock that could not be sold for a period of years or cash bonuses that could be recovered by the company if the deals for which the bonuses were a reward later soured.


The two aims--better aligning executives' incentives with those of the shareholders and reducing the riskiness of executives' compensation--are inconsistent. Shareholders are generally less risk averse than executives because they have less stake in the enterprise, as they can diversify away any risk that is peculiar to the enterprise by holding a diversified portfolio of securities. Top executives have much more to lose, in reputation and future earnings prospects, from the collapse of their company.


That means that top executives, provided that they are allowed by the Federal Reserve to remain imperfect agents of the shareholders, have strong incentives to establish procedures that will prevent traders, loan officers, and other subordinate executives from taking excessive risks. But "excessive" from the standpoint of private businessmen means something crucially different from "excessive" as perceived by the Federal Reserve. Risks, including the risk of bankruptcy, that are cost-justified from a corporation's standpoint may be unacceptable from a broader social standpoint because of their potential for bringing down the entire financial system, and in its wake the nonfinancial economy as well. But the measures that have been suggested for reducing risk taking by traders and other financial executives have their own problems. Many things can affect a stock's price besides a trader's deals, as critics of stock options as devices for compensating top executives point out, and retractable cash bonuses (would they have to be placed in escrow?) make it difficult for recipients to manage their finances.


Lucian Bebchuk, the Harvard professor of law and economics who is the leading critic of existing corporate compensation practices, does not advise that the government restrict the compensation of executives at the operating rather than managerial level, even in financial firms that might be eligible for classification as Tier 1 FHCs. Rather, he would have the Fed impose penalties for taking risks that can endanger the entire financial system on just the top executives, on the theory that this would motivate them to restrict the risk-taking activity of their subordinates. For example, the Fed might require the CEO of a Tier 1 FHC to place two-thirds of his salary and bonus in escrow, from which he could withdraw the money only after five years. This would motivate him to establish and enforce procedures that would reduce the likelihood that a deal or deals made anywhere in the company would blow up and destroy it and by doing so perhaps create the kind of chain-reaction effect that I illustrated with the example of Lehman Brothers.


This is an ingenious suggestion, greatly superior to the Report's recommendation to loose the modestly paid civil servants of the Federal Reserve on the entire compensation structure of Tier 1 FHCs. But it has drawbacks, mainly the fact that regulators lack the expertise required to establish compensation procedures that will balance a firm's competitive needs against the macroeconomic risks that rewarding risky financial decisions can create. Errors by the regulators will create openings for non-Tier 1 FHCs, including foreign firms that may be identical to Tier 1 FHCs in all but regulator-imposed constraints, to skim the cream of the Tier 1 FHCs' financial executives. U.S. firms that escape the classification may be able to do the same thing: eat the Tier 1 FHCs' lunch by avoiding the heavy hand of regulation.


We want to hear what you think about this article. Submit a letter to the editor or write to