What this covers

George Selgin of the Cato Institute sits down with host Russ Roberts to challenge the prevailing narrative of the 2008 financial crisis. Rather than defending the Federal Reserve's emergency response, Selgin argues that Ben Bernanke and the Fed made catastrophic errors both before and after the collapse—errors that worsened the recession instead of alleviating it. The conversation moves through two distinct phases of Fed action: the decision to bail out Bear Stearns' creditors in March 2008, and the subsequent decision to pay interest on reserves once the crisis deepened. Selgin contends that each choice violated sound monetary principle and deepened the very problems it claimed to solve.

The core dispute turns on what the Fed should have done during a financial panic. Selgin grounds his critique in Bagehot's Dictum—that a central bank should lend freely but only on good collateral and at high rates—and argues the Fed abandoned this standard almost immediately. The Bear Stearns rescue was not merely a tactical mistake but a watershed moment: by accepting toxic assets and protecting creditors against standard banking principle, the Fed triggered moral hazard that emboldened larger institutions, particularly Lehman Brothers. More provocatively, Selgin asserts that the Fed's later decision to pay interest on reserves was deliberately designed to prevent newly created liquidity from reaching the collapsing economy. This choice—maintaining a high Federal Funds rate while spending imploded—represented a fundamental inversion of the Fed's core responsibility to maintain stable total spending. The conversation distinguishes sharply between solvency and liquidity, questioning whether many rescued firms were truly solvent and challenging the Fed's claim that its balance sheet ultimately broke even when the recovered value of mortgage-backed securities depended entirely on the Fed's own price-support efforts.

Sharpest takeaway

Selgin argues the Fed mishandled the 2008 crisis both before the collapse (by bailing out Bear Stearns' creditors against Bagehot's rules, deepening moral hazard) and after (by paying interest on reserves to deliberately prevent its emergency liquidity from being lent into a collapsing economy), making the recession worse rather than better.

  • The Bear Stearns creditor bailout, not Lehman's failure, was the critical turning point because it emboldened Lehman's creditors and reinforced Too Big to Fail.
  • The Fed accepted toxic, non-Bagehotian collateral, so 'not losing money' is a red herring given the Fed itself propped up those asset values.
  • Interest on reserves was expressly designed to keep newly created liquidity from being lent, preserving control of the Fed Funds rate while spending collapsed.

The argument · threads5 threads · 25 claims
0.85

Bailouts of failing firms generate cascading moral hazard by signaling that large firms will be rescued, breeding expectation and escalating systemic risk.

6 pointscentrality 5/5
  • What matters in moral hazard is not the rescued firm itself but its creditors, who do not care whether they are repaid out of the firm's assets or out of a bailout because the dollars are equally good; thus after Bear, money rushed to Lehman not because it was better run but because its perceived standing among potentially-too-big-to-fail firms had improved, illustrated by money market funds lending to a firm with an unhealthy balance sheet.

    “It doesn't have to be out of the assets of the firm you are putting money in. It can be out of a bailout. And that's just--those dollars are just as good”

  • Rescuing failing financial firms avoids immediate cascading failures but creates moral hazard because other firms then expect to be rescued, and each successive rescue reinforces this expectation, escalating the size of firms eventually bailed out until you reach firms that genuinely are too big to fail.

    “if you rescue them, you avoid the immediate problem of triggering further failures. But you also create a moral hazard that consists of the expectation on the part of other firms that they will be rescued, should they get in trouble.”

  • The six-month respite Bernanke credits to the Bear bailout was harmful because larger firms like Lehman believed they were off the hook and behaved accordingly; absent the Bear rescue, Lehman would have had an incentive to prepare for bankruptcy and retrench, but instead it took bigger gambles, borrowing money it was unlikely to repay to compensate owners.

    “were it not for the Bear bailout, Lehman would have had an incentive to essentially prepare for bankruptcy... When it has no prospect of getting a bailout, it's going to start planning for the worst. It's going to retrench.”

  • The moral hazard from bailouts is not Bernanke's personal failing but a problem inherent to any human central banker with discretion, since it would take extraordinary courage to let a firm like Bear fail; the solution is therefore rules that prevent such emergency lending rather than reliance on discretion to avoid moral hazard.

    “it's not a Bernanke problem. It's a problem that we would face with any human central banker.”

  • Firms become genuinely 'too big to fail' precisely because earlier interventions treated smaller firms as too big to fail when they really weren't; studies afterward cast tremendous doubt on the claim that those earlier failures would have had severe systemic consequences, yet the bailouts increased expectations of future bailouts, producing ever-larger firms that eventually really are too big to fail.

    “the studies afterward cast tremendous doubt on the claim that the failures of those smaller Too-Big-to-Fail firms really did or would have had such severe systematic consequences as those who defended the bailouts at the time predicted.”

  • The 1984 rescue of Continental Illinois launched the Too Big to Fail doctrine as an explicit reality, fundamentally changing the game for large financial firms by leading them to perceive part of their risk-taking as subsidized by the Federal Reserve, broadening implicit insurance far beyond prior deposit insurance.

    “the real sea change I think came with the rescue of Continental Illinois. And that really is what launched 'Too Big to Fail.'”

0.75

Interest on reserves was a deliberate device to prevent banks from lending newly created liquidity into the economy.

4 pointscentrality 5/5
  • Paying interest on reserves, implemented in October 2008, was a ludicrous and disastrous policy because its purpose was to make the demand for reserves rise at least as fast as supply—creating an artificial liquidity trap—so the Fed could expand its balance sheet to bail out firms without that liquidity being lent into the economy, ensuring the collapse in spending would not be followed by a quick restoration; all premised on the madness that controlling the Federal Funds rate matters more than total spending.

    “the Fed chose, to see to it, through interest on reserves, that the demand for reserves would go up, at least as much as the supply; and created, as it were, a liquidity trap where none would have been created.”

  • Banks intrinsically need to hold reserves and do not need to be paid to hold them, so paying interest on reserves was unnecessary except as a deliberate device: if the Fed had not paid interest, banks would have lent out the extra reserves the Fed could no longer avoid creating, which is exactly what the Fed wanted to prevent in order to maintain control of interest rates.

    “banks really need to have reserves. They don't need to be paid to hold them... If you don't pay them, they are going to lend them.”

  • Until Lehman's failure, the Fed sterilized all its emergency lending by selling Treasury securities (from a base of about $800 billion) as it acquired other assets, keeping its total balance sheet size constant, because it was determined that emergency lending would not increase the total supply of Federal Funds—a goal predicated on maintaining control of the Federal Funds rate; Quantitative Easing only began once the Fed physically lacked enough Treasuries to keep sterilizing.

    “up to Lehman's they were sterilizing all this lending by selling Treasuries as they acquired other assets.”

  • Modern Fed thinking, influenced by contemporary monetary theories where it is uncool to talk about money or quantities, holds that monetary policy is fundamentally about regulating the Federal Funds rate rather than flows or quantities—an idea that treats a fixed interest rate as equivalent to letting the money supply be demand-determined, which fails because money demand does not map neatly onto any given interest-rate target.

    “They think monetary policy is about interest rates; it's not about flows, it's not about quantities. It's about where the rates are.”

0.56

The Fed's quantitative easing was crippled by policies that ensured monetary expansion would not increase total spending or restore demand.

3 pointscentrality 4/5
  • Quantitative Easing in 2008 was not bad in principle but was bad in this case because it was deliberately paired with policies—especially interest on reserves—calculated to ensure the easing would not revive demand or increase total spending, leaving the country with a Fed four times its prior size and dangerous exit problems but none of the good consequences, analogous to pouring gas into a tank with a leak.

    “I'm against Easing this time because it's accompanied by other policies that are calculated to make sure the Easing has bad effects but none of the good consequences that we really want from it.”

  • The core, non-fancy responsibility of a central bank is to ensure the amount of liquid funds in the economy is sufficient to keep total spending from collapsing; when spending collapses it signals people are trying to accumulate liquid funds that aren't there for everyone, and a properly organized monetary system—gold or central bank—should maintain a stable flow of spending.

    “their job is to see to it the amount of liquid funds in the economy is sufficient to keep spending going. To keep spending from collapsing.”

  • By trying to keep the Federal Funds rate artificially high through interest on reserves and sterilization, the Fed paradoxically all but guaranteed the rate would fall and stay low, because reviving credit and letting banks lend the extra reserves would have boosted spending and nominal GDP, which would eventually have pushed interest rates back up—whereas suppressing lending kept the natural rate collapsing.

    “by trying to keep interest rates artificially high, they all but guaranteed that they would fall and stay low.”

0.44

Bernanke's housing-collapse explanation as temporary liquidity-driven was unconvincing because fundamental asset values fell due to real economic deterioration.

6 pointscentrality 2/5
  • A solvent bank is one whose assets are at least equal to its liabilities (traditionally valued at maturity rather than marked to market), while liquidity concerns the ability to service liabilities in the short run; a bank can be solvent yet illiquid when it holds good long-term assets but lacks cash for immediate obligations, which is the precise condition under which a lender of last resort should help.

    “A solvent bank is a bank where the value of its assets is at least equal to the value of its liabilities... Now, liquidity is different. Liquidity has to do with the ability of the bank to service its liabilities in the short run.”

  • Bernanke's hypothetical defense that the housing decline was a temporary liquidity-driven run that would bounce back is unconvincing, because mortgage-backed securities collapsed in value not from fire-sale panic but because housing prices fell and default rates rose substantially, changing the perceived riskiness of the assets; moreover, due to the securities' complexity, no one—including the Fed—really knew where their value would settle.

    “They were collapsing because housing prices fell... it was this change in the perceived riskiness of the assets due to the incidence of defaults that was the proximate cause of the collapse in value of those mortgage-backed securities. Not liquidity.”

  • Walter Bagehot's principle for central banks facing a crisis is to lend freely, at high rates, on good banking collateral—the high rates serving both to combat an external drain of gold by attracting foreign capital and to discourage banks from borrowing merely to arbitrage cheap credit—so that the central bank provides liquidity to illiquid but solvent firms rather than cheap credit to relend.

    “faced with a crisis, the Bank of England... should lend freely, at high rates, on good banking collateral.”

  • Some commentators predict a financial crisis every year and then make a great deal of it when one finally arrives, which is not a meaningful prediction; merely predicting that there will be another crisis someday is virtually risk-free.

    “some people predict one every year and then make a great deal about it when it finally comes about. I'm not involved in that racket. I'm merely predicting that there will be another one some day.”

  • Letting Lehman fail might have been a good thing in itself, but in context it was harmful because it was contrary to all market expectations after Bear, generating a tremendous shock and uncertainty about what the Fed's actual policy was, after firms had concluded that investment banks were too big to fail.

    “I actually think that letting Lehman fail in itself might have been a good thing. But in the context it was quite harmful because it was contrary to all of the expectations at the time.”

  • It is never in anyone's interest to keep an insolvent bank alive because you can only delay an inevitable failure, the firm wastes resources as an inefficient enterprise, and once capital is exhausted the bank's managers lose incentives to manage money well and are tempted to 'shoot the moon' with hail-Mary gambles; with plenty of solvent banks available, the goal should be to protect them from falling dominoes, not to keep insolvent institutions on artificial support.

    “once a bank becomes insolvent, lacks the incentives to manage money as well, because once people are off the hook, once the capital has been exhausted, well, the temptation is to use any money to shoot the moon.”

0.38

The Fed's vast balance sheet of long-term securities carries severe interest-rate risk and was artificially supported by the Fed's own policies.

2 pointscentrality 3/5
  • The value of the Fed's roughly $4 trillion-plus balance sheet of mortgage-backed and long-term Treasury securities depends crucially on future interest rates—the 'elephant in the room'—so whether those assets are actually worth their face value is deeply uncertain and carries substantial interest-rate risk, which is precisely why such long-term assets are not normally considered good banking collateral.

    “the elephant in the room is interest rates... The values of all of these securities depends crucially on what interest rates are.”

  • The argument that the bailouts were fine because the Fed ultimately lost no money is a red herring, because the toxic securities (e.g. Maiden Lane assets worth 39-44 cents on the dollar in 2010) only recovered value as a result of the Fed's own concerted efforts to pump up their prices by continuing to buy them and keeping interest rates artificially low—not because the assets were genuinely sound.

    “they did so mainly as a result of the Fed's concerted efforts to pump up the value of those securities by continuing to buy them and by otherwise keeping interest rates artificially low.”