What this covers
Eugene White of Rutgers University argues with Russ Roberts that bank regulation has pursued the wrong strategy: instead of constraining what bankers can do through rules, regulators should reshape the incentives facing them so prudence emerges naturally. White traces this insight through nineteenth-century practice, showing how mechanisms like double liability (which exposed shareholders to additional losses if their bank failed) and performance bonds for cashiers created genuine skin in the game. When these devices aligned stakeholder interests with risk exposure, banks closed before collapse rather than gambling for survival. The conversation then moves to how twentieth-century changes—deposit insurance, implicit government bailouts, and the proliferation of regulators—removed those natural monitors and enabled moral hazard and excess leverage.
White examines why the U.S. banking system became uniquely fragile compared to Canada's. The culprit was not nervous depositors but the National Banking Act's prohibition on branching, which created thousands of small, undiversified unit banks vulnerable to local shocks. The Federal Reserve itself, designed in 1914 with twelve regional banks rather than a single central authority, reflects this same legacy of political pressure from those small banks. White argues that regulatory arbitrage—where banks shop for the weakest regulator—thrives because authority is split across competing agencies, and that Dodd-Frank largely failed to address too-big-to-fail. Throughout, he illustrates how leaders like Richard Fuld and Jimmy Cayne escaped meaningful downside despite appearing to lose heavily, and how the Ratchet Effect has expanded FDIC coverage upward and made it nearly impossible to reverse.
White argues bank regulation should change bankers' incentives so prudence emerges naturally rather than restricting their choices through top-down rules, drawing on 19th-century mechanisms like double liability and performance bonds that aligned stakeholder interests with risk.
- 19th-century devices (double liability, cashier performance bonds) gave managers and shareholders genuine skin in the game, encouraging banks to close before collapse
- Deposit insurance and implicit bailouts removed creditor and depositor monitoring, enabling excess leverage and moral hazard
- U.S. branching prohibitions, not nervous depositors, made the U.S. system more panic-prone than Canada's nationwide-branching system
Implicit government guarantees and moral hazard remove market discipline and encourage excessive leverage.
- When Wall Street investment banks converted from partnerships to publicly traded companies about 15-20 years ago, partners could diversify capital out of the firm, and the public firms could become more leveraged, larger, and pay higher salaries to attract talent; this created a competitive dynamic that pushed the whole industry toward going public and taking more risk.
“when one of these firms went public, the partners got stock... they were able to move some of their capital out of the firm... But then what happens? The public firm can become more leveraged; it can borrow more and take more risks.”
- FDIC insurance began as a limited program but underwent a slow creep upward between 1934 and 1980, as depositors learned to split and move accounts, and competitive pressure spread insurance from commercial banks to savings banks and other intermediaries, making the expansion of liability insurance very hard to reverse.
“if you look between 1934 and even up to 1980, the levels of deposit insurance creep up. People get better at splitting their accounts and moving them bank to bank.”
- In the S&L crisis, Congress responded to insolvent thrifts not by closing them but by raising deposit insurance and granting new, riskier lending powers, which, like letting a losing gambler keep playing at greater odds, encouraged insolvent institutions to take more risk since managers captured gains if bets paid off but bore no further loss if they failed.
“the reaction in Congress... was to not close them all down, and absorb the large losses, but to raise the level of deposit insurance, keep people in, and then give them new lending powers. To take more risk.”
- In the aftermath of the Savings and Loan crisis, large depositors who held funds above the FDIC limit were nonetheless reimbursed, and the limit was later raised from $100,000 to $250,000 and extended to Money Market Mutual Funds.
“And in fact, we raised the limit from $100,000 to $250,000 during the crisis, and we extended it to Money Market Mutual Funds.”
- Implicit government insurance means even depositors above the formal limit and large creditors believe the government will not let institutions fail, so they don't seek safer banks or demand higher returns, which removes natural restraints on risk and allows firms to take on more leverage.
“many people who had $200,000 or $300,000 in the bank would never even think of running on the bank because there's an implicit insurance. They believe the government won't allow the institution to fail, so why should they seek out a safer institution? That allows the firm to take more leverage.”
- Double liability was abolished because the Great Depression's systemic shock made it nearly impossible to collect funds from shareholders to pay depositors; at the same time that FDIC insurance was introduced, double liability was cancelled, and deposit insurance began as a limited mutual-pool program funded by banks rather than government.
“if you say, why don't we have double liability any more, it turns out the answer is really the Great Depression. Where, again, you have this systemic, broad, wide shock, and it's very difficult to collect from anybody money to pay out depositors.”
Senior executives escape personal liability for risks they drive, misaligning incentives across the industry.
- In the MF Global case, the CEO who pushed the firm into proprietary trading to make money claims not to know what happened to customer funds, yet does not appear to bear any liability for the customer losses, which illustrates a misalignment where senior executives escape responsibility for risks they drove.
“it doesn't appear that the man in charge bears any liability for this. He's the one--this is Mr. Corzine--who took over the firm and said: All right; we are going to make lots of money... And yet he looks now and says: I don't know what happened.”
- Although crisis-era CEOs like Fuld and Cayne lost roughly a billion dollars on paper, they were left with around $500 million each, which is not a meaningful penalty because it leaves them rich beyond most people's dreams and in the top tenth of one percent, so they did not bear genuine downside risk despite appearing to have skin in the game.
“As one of my friends would say: That's already rich beyond your wildest dreams.”
- Historically prudent investment bankers avoided lavish buildings to signal they were not misspending clients' money, so the erection of a luxurious new skyscraper should prompt the question of where the money is going; Northern Rock built a grand new headquarters just before its collapse.
“when you see a firm erecting a new giant skyscraper, a luxurious one, you should ask the question of: What's going on?”
Layered regulations produce unpredictable interactions that address some problems while creating new ones.
- Rather than improving the financial system step by step as Woodrow Wilson hoped, the U.S. has tended to layer one regulation atop another, producing not a simple model with one imperfection but thousands of interacting constraints whose combined outcome is nearly impossible to predict, so current legislation may address a few problems while adding new ones.
“most of the time what we've done is we've tended to layer, add one layer of regulation on top of another.”
- Dodd-Frank did little about too-big-to-fail and has no teeth because its roughly 1,500 pages sit largely in outline and delegate implementation to the agencies, so the ultimate outcome is unknown; meanwhile the people with the incentive to monitor implementation are special interests rather than the informed but inattentive general public.
“the whatever, 1500 pages of Dodd-Frank, sits in outline and then delegates to the agencies to implement the laws.”
- Because financial-stability authority is split across many agencies with different funding and incentives that fight for position, financial institutions engage in regulatory arbitrage by seeking the weakest regulator, creating destructive competition among regulators; consolidating these agencies into one would reduce this problem.
“you get often to have what we call regulatory arbitrage, which is where financial institutions seek the weakest regulator. And so you have competition among the regulators. And this is very destructive and very weakening.”
- Having many regulators does not improve oversight because the incentive to look out diminishes when responsibility is diffused; each regulator can assume someone else is supposed to do the job, and agencies like the SEC are sometimes starved of funds before crises and then blamed for not catching wrongdoers.
“the incentives to look out are not very big when there are lots of you, because you could say someone else is supposed to do it.”
Well-designed incentive alignment through skin-in-the-game mechanisms historically constrained risk better than rules.
- From after the Civil War to WWI, bank shareholders were subject to double liability: if a bank failed, the receiver could demand from each shareholder an additional contribution equal to the par value of their stock, which made shareholders monitor management closely and prompted them to close failing banks before they went over the edge rather than gamble for resurrection.
“anyone who held a share of stock, if that bank failed, the receivers... would have the right to go back to every shareholder and ask for a contribution to pay out all the creditors, equal to the par value of the stock.”
- In 19th-century banking, the cashier (the second-most important bank officer) was frequently required by the market to post a bond equal to one to three times their salary, which would be forfeited if the bank failed due to mismanagement, creating a strong incentive against excessive risk.
“frequently they would be asked to post a bond equal to one, two, or three times their salary. And the idea there was that if the bank got into trouble or if failed as a result of some mismanagement, that bond would be forfeit. Strong incentive to make sure they don't take too much risk.”
- The dominant approach to bank regulation for the past century has been to restrict what banks can and cannot do (regulating choices), but a better approach is to give bankers the proper incentives to make prudent choices on their own rather than dictating those choices.
“instead of telling people what the choices are they should be making, what the bankers should be making, they should be giving the proper incentives for them to do so”
- Institutions that cannot fail tend to absorb subsidiary enterprises that then become insured too and may begin doing everything poorly (as in the abysmal handling of foreclosure documents); removing the permanent guarantee would likely lead banks to become simpler, more specialized institutions that contract out the rest, which is preferable to dictating what they can and cannot do.
“if they can't fail, they are going to start absorbing all kinds of subsidiary enterprises, which become insured as well. So, they may begin to do everything poorly.”
Unit banking and branching restrictions fragmented the system and shaped regulatory structure by accident and political pressure.
- The United States had more frequent bank panics than Canada, Britain, or France not because Americans were more nervous but because the U.S. prohibited branch banking, producing thousands of small, undiversified, single-product banks (e.g., a Kansas bank lending only to wheat farmers) that were vulnerable to local shocks and propagated panics through the interconnected clearing system; Canada had no central bank but nationwide branching and avoided frequent panics.
“It's because we impose a particular set of regulations--we prohibited branching. So, we had tons of small, undiversified institutions.”
- The U.S. unit-banking system arose from the Law of Unintended Consequences: the National Banking Act of 1864 required banks to conduct major activities in their office so examiners could find all documents (since there was no telephone or good telegraph), which entrenched single-office banks that then formed a powerful lobby resisting branch banking, delaying nationwide branching until 1997.
“the National Banking Act of 1864 specified that the major activities of the banks be conducted in its office. And it seems to be the purpose behind that was to make sure that when an examiner came, he would have all the documents there.”
- The Federal Reserve's decentralized structure of 12 regional banks resembling clearing houses owes its existence to the political influence of the 15,000+ small unit banks in 1914, who were nervous about a single central bank and were sold on a regional system that would process their checks and provide credit, rather than to a clean design for preventing panics.
“in 1914 there are well over 15,000 small banks--many of them are very nervous about the idea of having one central bank. And they are kind of sold the idea that they are going to have 12--12 regional central banks, which are going to look like clearing houses”
Government must set standards and rules where asymmetric information prevents market discipline from working.
- A pure 'everyone on their own' approach fails because of asymmetric information: with perfect information you wouldn't even need banks since you'd know who to lend to, but since depositors cannot monitor banks' proprietary information, government has a role in setting common rules and standards (such as uniform accounting) so institutions can be distinguished, otherwise comparison becomes impossible or very expensive.
“the problem with just saying everyone's on their own is the problem with asymmetric information. If everyone had perfect information then you really wouldn't even need banks”
- Banks are institutions that we create by granting certain privileges, and the state retains the right to regulate them.
“Banks are institutions which we really create--we give them certain privileges; and we retain the right to regulate them.”
The financial system weathered recent crisis well enough; durable stability requires realigning incentives, not new restrictions.
- For a country starting from scratch, White would design a system with relatively minimal regulation but properly aligned incentives, so that depositors, directors, shareholders, and managers all have interests matched to the risks they take; this would not eliminate losses but would ensure those bearing risk also bear the consequences.
“I would set up a system which had relatively minimal regulation, but set up the right kinds of incentives so that depositors, directors, shareholders, and managers all had their interests properly aligned.”
- Because a financial collapse is messy and occurs industry- and economy-wide, it is very difficult from single examples to distinguish a broad systemic shock from the individual actions of virtuous versus risk-taking managers, making it hard to assess who was prudent and who took inappropriate risks.
“It's hard to assess right out what we call a broad or systemic shock from the individual actions of who was a virtuous manager and who was a risky--we'll say an unvirtuous manager--taking more than appropriate risks.”
- Fears that prudential reforms would shrink the U.S. financial system are misplaced because the U.S. uniquely has easy entry into banking with new small and medium institutions opening every year to serve small enterprises, whereas European countries like Spain face heavy barriers preventing new banks from replacing failed ones; thus the U.S. system would remain vibrant even with the proposed incentive reforms.
“One thing about the United States is that there is easy entry into banking. Any year you look at, there are new banks opening up.”
- On a somewhat positive note, the crisis has swept the weaker institutions out of the system like rotten trees from a forest, which should produce a fairly stable banking system for at least the medium term (at least 18 months), though firms may be hesitant to extend credit; durable stability still requires long-term solutions.
“the crisis has swept out a lot of rotten trees from the forest, so it's going to clear things out; and for a while we'll have a fairly stable system.”