A Reform Roadmap for the Fed’s Task Forces

Reopening the Fed’s framework review was the first recommendation the Cato Institute’s Center for Monetary and Financial Alternatives made at the start of Warsh’s tenure.² The case for it was straightforward: The Fed adopted its flexible average inflation targeting regime in 2020, promising to let inflation run above target to make up for past shortfalls, and then presided over the worst inflation in four decades while insisting the surge was transitory. When the Fed conducted its scheduled framework review in 2025, it treated the exercise narrowly, revisiting the language of its consensus statement without seriously reexamining how it sets policy, reads the economy, or communicates its decisions. The five task forces are, in effect, the review that should have taken place then.

For each task force, in the sections below I identify the underlying problem with current Fed practice and the reform that best addresses it. The recommendations share a common principle: The Fed serves the public best when it operates with more transparency, predictability, and restraint, and when it interferes less in private markets.

After 2008, the Fed came to rely on forward guidance, telling markets in advance what to expect from policy, and it built an elaborate communications apparatus around that practice, including the Summary of Economic Projections and its widely watched dot plot. The intention was to make policy more predictable, but the practice carries a deeper problem. Financial markets and incoming data often provide a better real-time read of economic conditions than the Fed’s own forecasts, and forward guidance can ignore or obscure those signals.

Two recent episodes illustrate the problem. In September 2025, inflation data pointed to holding the rate target steady or raising it slightly. Yet the Fed had so firmly signaled a cut that reversing course would have shocked markets; forward guidance forced the Federal Open Market Committee (FOMC) to cut the rate despite prevailing data.³ The Summary of Economic Projections offers another example. In March 2026, not one policymaker projected a rate increase for the remainder of the year even as inflation ran well above target, while private markets had already begun to price in at least one rate hike. By June the committee’s projections had swung toward a hike, catching up to what markets had signaled months earlier.⁴ In each case, the better read of conditions came from outside the Fed’s projections, and the Fed’s own guidance had pointed the wrong way.

Scaling back forward guidance and the projection apparatus moves in the right direction, and central bankers abroad have voiced similar regrets about being bound by their own guidance.⁵ The communications task force can resolve the underlying problem with a single reform. The Fed should pick and then publish a reaction function—an arithmetic formula connecting the FOMC’s rate target decisions to macroeconomic indicators—and let that stand as its communication. If policy follows a transparent rule of this kind, the public can see how the Fed will respond to changing conditions without the Fed having to forecast or precommit to a path. Predictability would come from the rule itself, and forward guidance in its current form would become unnecessary. The communications problem is a symptom of Fed discretion, and it is best solved by adopting an objective policy rule.

Before the 2008 financial crisis, the Fed’s balance sheet was small and consisted almost entirely of short-term Treasury securities. It has since grown to many times its historical size, peaking near $9 trillion in 2022 after successive rounds of quantitative easing.⁶ What began as an emergency response to two crises became a permanent feature of monetary policy, and the composition of the balance sheet changed along with its size. The Fed now holds large quantities of long-term Treasury securities and mortgage-backed securities (MBS), and it operates in an ample-reserves regime that requires it to pay interest on the reserves it created.⁷ The balance sheet task force is charged with examining that footprint and figuring out how to reduce it.

There are several problems with a large and broadly composed balance sheet. The Fed’s MBS holdings channel credit toward housing, favoring one sector of the economy over others in a form of credit allocation the central bank was never meant to perform.⁸ The ample-reserves framework requires the Fed to pay interest on reserves to affect its policy rate—a program that has cost the Fed billions, transferred risk-free income to large banks, and complicated the fight against inflation—because every rate increase raises the Fed’s interest expense.⁹ And this post-2008 operating framework makes it easier for the Fed to absorb federal debt, raising the risk of fiscal dominance, in which the Fed’s price stability mandate gives way to the government’s financing needs.¹⁰

The task force should endorse and accelerate a return to a smaller, simpler balance sheet. The Fed should continue reducing its holdings and prioritize running off its MBS portfolio, moving toward a balance sheet composed of Treasury securities.¹¹ It should exit the ample-reserves regime and return to a framework in which it influences its policy rate through the open market for reserves rather than through interest payments to banks, which would allow the interest on reserves program to be wound down.¹² To ensure that these solutions are permanent, Congress should place statutory limits on the size of the balance sheet and restrict the Fed’s purchases to short-term Treasury securities, thus closing the door to future rounds of open-ended asset purchases and to the credit allocation that MBS holdings represent. Obviously, where Congress places the limit on balance sheet size will be crucial; one option is to cap the Fed’s assets at 10 percent of total US commercial bank assets, which is roughly the ratio that existed pre-2008.

Fed officials have cautioned that shrinking the balance sheet is difficult and that a rapid sale of assets could strain financial stability.¹³ That caution is warranted, but it merely argues for a deliberate, well-communicated runoff rather than abrupt sales. It does not argue for leaving the balance sheet where it is. The Fed maintained a small, Treasury-only balance sheet for most of its history without incident, and a gradual return to that posture is well within its capacity. The goal is a balance sheet that supports monetary policy without further entangling the Fed in credit markets or in the financing of federal deficits.

The Fed relies heavily on economic statistics produced by government agencies, and the data task force is charged with asking whether those sources serve it well. It is a question worth asking. Government data arrive slowly and are subject to substantial revision, sometimes long after the policy decisions that relied on them have been made. In September 2025, the Bureau of Labor Statistics released a preliminary benchmark revision lowering its estimate of total nonfarm employment as of March 2025 by 911,000, one of the largest such markdowns on record and enough to change the labor-market picture that policymakers had been acting on.¹⁴ Revisions of this kind have grown more severe since the pandemic, as falling survey response rates and other collection difficulties have degraded the timeliness and reliability of official statistics.¹⁵ A central bank that describes its policy as data dependent is only as good as the data, and the data have been getting less reliable.

The contrast with private markets is stark. Private stock exchanges price thousands of assets by the second, aggregating information continuously and revising in real time, while the government’s marquee economic statistics arrive weeks after the fact and are then revised for months. There is no technological reason that macroeconomic data must be slow and stale. The reason is largely institutional. Because the government produces headline statistics at no direct charge to users, it crowds out the private data providers who might otherwise compete to collect and deliver economic information faster and more accurately.¹⁶ Free public provision is not costless despite its zero-price veneer. It carries a high opportunity cost, measured in the better private data ecosystem that never develops because it cannot compete with a subsidized government product.

The task force should treat openness to private and higher-frequency data sources as a starting point and go further by recognizing the crowding-out problem directly. The existence of robust private data, and the demonstrated ability of markets to meet nearly every other kind of consumer demand, suggests that the case for exclusive government production of economic statistics is far weaker than commonly assumed.¹⁷ At a minimum, the Fed should incorporate private and market-based data into its assessment of the economy rather than treating official releases as the sole authority. More ambitiously, policymakers should reconsider whether the government needs to crowd the field at all, or whether stepping back would allow a faster and more accurate private data ecosystem to emerge.

This task force examines the interaction of productivity and employment. Warsh has suggested that gains in productivity, particularly from artificial intelligence, could create room to cut interest rates, based on the reasoning that faster productivity growth lowers costs and eases inflationary pressure.¹⁸ Faster productivity growth can, in principle, allow stronger real growth without added inflation. The difficulty is that no one knows how large the productivity gains from artificial intelligence might be, or when they might arrive.

The range of credible estimates is enormous. At the low end, one prominent estimate puts the total factor productivity gain from artificial intelligence at no more than 0.66 percent over an entire decade, while others project gains well above a full percentage point per year.¹⁹ A range that wide cannot anchor a rate decision. Setting policy today based on a productivity forecast means selecting one point from a distribution so broad that almost any policy could be justified by choosing the convenient estimate. This form of monetary policy is just more Fed discretion, and it is the kind of policymaking the inflation framework task force should be moving the Fed away from.

Understanding productivity and employment is a legitimate object of study, and productivity does matter for the economy’s capacity to grow. But of the five task forces, this is the one whose payoff to actual policy is least certain, because the central quantity is unknowable with the precision that a rate decision would require. Anchoring policy to a productivity bet risks drawing the Fed’s attention and credibility toward the question it is least equipped to answer, and away from the reforms where the task forces can deliver concrete improvements, namely the balance sheet and the policy framework. A Fed governed by a transparent rule would not need to place a bet on productivity. The rule would respond to realized conditions as they appear in the data, incorporating stronger growth if it materializes, without requiring the Fed to forecast it in advance. The productivity question is best resolved not by improving the forecast, but by adopting a framework that does not depend on one.

Congress requires the Fed to promote stable prices and maximum employment, but it imposes no binding requirement on how the Fed pursues that mandate. Over time, this open-ended grant of discretion has hardened into the assumption that the Fed can manage the economy, adjusting interest rates to steer inflation and employment toward precise targets. That assumption asks too much of any central bank. Inflation is shaped by fiscal policy, supply conditions, global shocks, and the decisions of millions of private actors, and the Fed influences only a portion of the forces at work.²⁰ The inflation framework task force is charged with examining the drivers of inflation, the Fed’s responsibility for inflation, and how inflation is measured, the last of which overlaps with the data task force. Warsh has said the task force will “weigh the full range of ideas for delivering price stability in a changing economy.”²¹

The current Fed inflation framework rests too heavily on the FOMC’s period-by-period judgment about the sources of inflation. The appropriate response will only ever be as good as that judgment, and the record since the pandemic shows how costly a lapse in judgment can be. The Fed misread the post-2020 inflation as transitory, held policy loose as prices accelerated, and then tightened sharply once the misjudgment became undeniable.²² The problem is not a single bad forecast. The problem is a framework that makes policy hostage to the FOMC’s subjective economic assessments.

Of the ideas the task force might weigh for delivering price stability, the most effective is to require the Fed to set its policy rate according to a rule. Under such a system, the FOMC would tie its interest rate target to observable economic conditions, such as inflation relative to 2 percent and a measure of economic slack, rather than resetting policy from scratch at every meeting.²³ A rule does not promise perfect policy. Instead, it offers policy that is systematic, transparent, and insulated from the pressures of the moment. The Fed would commit to a rule, follow it, and revise it only at scheduled intervals, such as the framework review it already conducts every five years. Any deliberate departure from the rule would require a public written justification, a plan to return to it, and testimony to Congress.

A rules-based framework directly addresses the failures the task force is meant to study. It removes the need for the Fed to correctly diagnose the source of every inflationary impulse in real time, a task at which no central bank has proven to be reliable. It anchors expectations, because households and firms can see how policy will respond to changing conditions. And it lowers the risk that political pressure or institutional overconfidence would drive policy off course, because deviations become visible and costly to justify.

The reform does not require endorsing any single rule. The goal is not to bind the Fed to one economist’s formula but to require that policy follow some transparent, precommitted rule that the Fed itself selects and defends. In any case, Cato conducted a comprehensive comparison of several commonly prescribed rules and found that they have relatively similar macroeconomic effects; what matters most is picking one and sticking to it rather than the precise rule calibration itself.²⁴

One question sits outside this task force’s remit for now. Warsh has said the 2 percent target is the Fed’s long-held objective and that the task force will not revisit it until the Fed has reestablished its ability to deliver on it.²⁵ Setting the target question aside is sensible while inflation remains above the target and while the more pressing reforms are underway. Over a longer timeline, though, the target itself deserves scrutiny. The choice of 2 percent is not the product of a model or a structural estimate, and a fixed positive target forces prices upward even when rising productivity would otherwise let them fall. A target derived from the economy’s long-run productivity growth would allow consumers to capture the gains from cheaper production in the form of lower prices, rather than surrendering them to a predetermined rate of inflation.²⁶ That is a project for a later Fed review, but it belongs on the agenda.

Taken together, the five task forces describe a Fed that could do considerably less than it does today and do its core job better as a result. No one should expect the Fed to ever deliver perfect macroeconomic outcomes—that requires a privatized monetary system and is not possible under our centralized fiat currency regime. But a Fed that binds itself to a transparent rule, holds a smaller and simpler balance sheet, communicates through that rule rather than through forward guidance, draws on better and more varied data, and refrains from staking policy on speculative forecasts would be a more predictable, more accountable, and less intrusive institution. Each reform stands on its own, yet each advances the same principle: A central bank that interferes least with private decisions, and operates by clear and stable rules rather than by discretion, serves the public best.