“Kevin Warsh Says the Fed Is Not a Repair Shop. He Must Draw the Line”

In 2010, Kevin Warsh told a room full of Wall Street executives that the Federal Reserve is not a repair shop. His meaning was plain: step in when markets seize up, step out when they stabilize. A sign on the Fed door, by that reasoning, might read: "Emergency repairs only." A narrow Fed Put.

Now Warsh is headed to the Senate Banking Committee as the nominee for Fed chair. Senator John Kennedy, a persistent critic of Fed overreach and the moral hazard it breeds, is likely to pounce. You can almost hear the drawl: "Mr. Warsh, today's market expects the Fed to fix everything from a bad earnings season to a full-blown crisis. They see a sign on the door that says, 'From bumper to bumper, we've got you covered.' What say you?"

Kennedy's question lands because the gap is real. What Warsh described in 2010 and what markets expect today are not minor variations on the same theme; they are different jobs entirely.

That is where the Fed chair nominee comes in: he should walk into the hearing with a definition. One that puts the emergency back in emergency repairs.

The problem with the Put

The Fed Put is not a policy. It is an expectation. When financial conditions deteriorate, markets believe the central bank will step in.

The result is not a rule. It is a habit.

And the habit has been creeping in scope. You can't hold the line at a boundary that was never drawn. Since the Fed has never defined the Put, markets have supplied the definition themselves, and it is quite generous and growing.

What the Put should mean

The Fed Put should mean only this: the Fed intervenes when financial market dysfunction blocks the transmission of monetary policy to the real economy, threatening employment or price stability.

That is the trigger. Everything else is noise.

The trigger is impairment, not volatility. Those are not the same thing. In practice, that means evidence in credit and funding markets, not just closing prices.

The difference is not the size of the drop. It is what the drop does. A twenty percent selloff that leaves borrowing costs unchanged and unemployment steady is not a trigger. A ten percent drop that freezes corporate debt markets and sends layoffs surging is. The difference is transmission, not tick marks.

When the transmission mechanism breaks down, when credit channels close and systemic contagion threatens to pull the real economy into free fall, the Fed must act to preserve its ability to do its job. Bagehot codified this principle more than a century ago. The 2008 crisis confirmed it under conditions no one would choose to repeat.

Outside of this definition, there is no Put.

Critics argue that the line is easier to draw in hindsight than in real time, and they are right. That is precisely why it must be drawn now, in calm conditions, before the pressure arrives. A definition cobbled together in the middle of a crisis is not a definition at all. It is capitulation dressed up as policy.

A second objection holds that any rule can be gamed -- that officials will always find a reason to intervene. Perhaps. But a rule does not eliminate judgment; it disciplines it. It narrows the space for excuses.

Where the definition holds

The 2008 financial crisis belongs within it. Interbank markets froze. The commercial paper market seized. Credit channels closed. The Fed acted aggressively, and the justification was sound. The transmission mechanism had broken.

The early months of the 2020 COVID response fit as well. The real economy collapsed virtually overnight. Treasury markets and money market funds came under severe stress. The Fed's intervention restored functioning that had been impaired.

One can argue about the scale and composition of the responses, but under this standard they were justified.

Where the definition breaks down

The problem is not that the Fed has a firefighter. It is that the firefighter keeps getting called for false alarms.

The original sin -- 1998

In 1998, Long-Term Capital Management imploded. The Fed coordinated a private rescue and cut rates three times. Defenders argue that LTCM sat at the center of a web of derivatives exposures that could have cascaded through the system. Even granting that risk, credit to households and businesses never meaningfully faltered.

It is better understood as the original sin. The economy was booming, the stock market was rising, and banks were lending. No transmission mechanism was impaired, but the Fed acted anyway to soothe markets after a hedge fund blew up.

Under the definition proposed here, 1998 would not have qualified. Systemic risk has to show up in the real-economy plumbing, not just on Wall Street's balance sheets, before the central bank reaches for the fire hose.

The dot-com hangover -- 2001-2004

After the dot-com bust, the Fed cut rates to one percent and held them there long after the real economy had recovered. Holding rates at one percent once credit was flowing freely is where Put logic took over from macro data. Rates stayed too low for too long. Cheap money flooded into housing and, when it met lax underwriting and securitization incentives, it helped along the conditions for 2008.

The Jackson Hole signal -- 2010

In August 2010, with no recession underway and markets functioning normally, Chairman Bernanke previewed QE2 at Jackson Hole. He was explicit that QE2 was partly filling the gap left by a gridlocked Congress. With that speech, markets learned that the Put covered equity drawdowns, not just systemic dysfunction.

The taper tantrum -- 2013

In May 2013, Bernanke testified before Congress and suggested the Fed might begin tapering bond purchases. He did not announce tapering; he floated a possibility. The 10-year Treasury yield rose from roughly two percent to nearly three percent over the following months. Equity markets sold off. The reaction was swift and sharp.

The Fed backed down, delaying tapering for months. A floating possibility had been enough to rattle markets, and that was enough to rattle the Fed.

The Fed retreats -- 2018

Coming out of a decade of near-zero rates, the Fed was in the middle of a long-overdue normalization cycle. It had raised rates to two and a quarter percent and signaled further increases. Equity markets fell roughly twenty percent in the fourth quarter. Powell and his colleagues cited global growth risks and tightening financial conditions as the rationale.

Those factors did matter for the domestic outlook in principle. But the unemployment rate was under four percent and domestic inflation was contained.

The Fed folds -- 2019

By January 2019, Powell had reversed course. He introduced the word "patient" and ended the hiking cycle. By mid-year, the Fed was cutting.

The Fed cut when markets fell, but It did not tighten when they recovered.That asymmetry is the Put in its purest form.

For 2018 and 2019, the question is simple: is it the economy or is it the market? When that question is left unanswered, markets assume the one that benefits them most. A defined Put removes that ambiguity. If the Fed pauses or pivots in the future, it will have to point to specific data. Not global risks. Not sentiment. Data.

Five times. One result: markets pushed; the Fed complied.

None of these episodes would have qualified for a Fed Put under the definition proposed here: the transmission mechanism was not impaired, and employment and price stability were not threatened. By 2019, the Put had been stretched so far that its original shape was barely visible.

What the drift has cost

The costs fall into three categories, and none of them is theoretical.

The first cost is moral hazard. When investors believe losses will be cushioned regardless of the quality of their decisions, they take more risk than fundamentals warrant. Capital flows toward speculation, leverage builds quietly, and asset prices inflate beyond sustainable levels.

Each Put episode compressed rather than eliminated risk. The next shock therefore hit a more fragile system and required a larger intervention.

SVB, Signature, and First Republic were the most recent bill. Yes, bank-specific mismanagement and supervisory lapses played a role. But those failures interacted with interest rate and liquidity risks built up during years of suppressed rates and QE-influenced valuations. Those valuations did not emerge from thin air. They were in part the byproduct of a decade of Put-driven accommodation.

The second cost is policy distortion. When the Fed responds to market reactions rather than economic fundamentals, rate decisions get evaluated by how markets respond on the day, not by whether they serve price stability and employment over time. When capital gets mispriced this way, the pain doesn't stay on trading floors. It shows up later as lost jobs, stalled investment, and louder calls for bailouts.

The third cost, and deepest, is the corruption of price signals. Markets allocate capital through those signals, and among the most important is loss: the information that a bet was wrong or a risk mispriced. When policy repeatedly cushions that loss, it risks weakening the mechanism that makes markets work in the first place.

Congress never authorized the Fed to mute market signals or soften the discipline of loss. Yet over time, the Fed Put has moved in precisely that direction.

There are costs of non-action too. For example, failing to act when transmission really is breaking has brutal consequences for workers and businesses. The point is not to make the Fed indifferent; it is to make the Fed discriminating.

Why Warsh can draw the line

Most incoming Fed chairs have inherited the Put and passed it on slightly larger than before. Warsh is different for three reasons. He was inside the institution during 2008 and knows what a genuine emergency looks like.

He stated his conviction publicly in 2010, before he had any platform to act on it. And he is arriving during calm conditions, the best possible moment to define the line before the next test arrives.

Now he has the platform. The confirmation hearing is that moment.

What he should say at the hearing

Kennedy will ask where the line is. So will others on the committee who have watched the Put creep in scope for decades. Warsh should have an answer.

The definition is already on the table: the Fed intervenes when market dysfunction blocks transmission to the real economy, threatening employment or price stability. Everything else falls outside it.

We backstop the plumbing, not the price charts. The Fed will act when financial market dysfunction threatens the real economy through the transmission mechanism. It will not act to prevent portfolio losses.

A defined Put gives everyone the same thing: investors, businesses, workers all gain certainty about what the Fed will and will not do. Clarity before the storm is cheaper than improvisation during it.

Holding the line

One hearing answer won't reset decades of habit, but it starts the clock. Every time Warsh holds the line after that, the Put will shrink a little more.

He said the Fed is not a repair shop.

The hearings are his chance to mean it.