In the past couple of months, the flaws in the Federal Reserve’s management of the economy have become hard to ignore. The central bank’s current way of doing things is increasingly self-defeating.

To its credit, the Fed seems to realize this and is conducting a broad review of its goals and methods. It would be a pity if this process ended with an inconspicuous mid-course correction — the default outcome of most such exercises. What the Fed needs is an avowedly bold new approach, leaving no room for doubt that something big has happened.

Such a change will require a drastic shift from Fed Chairman Jerome Powell’s customary posture of cautious pragmatism. In explaining the bank’s approach to setting interest rates, he tends to emphasize patience and data-dependence. The economy is performing pretty well, he said until recently, with extremely low unemployment, growth at or above its long-term trend, and no sign of inflationary pressure.

The Fed stood ready to respond if or when the data pointed to a deviation from that satisfactory course. Among other things, he seemed to be telling investors to calm down and not to over-interpret every fleeting particle of information, including comments from him or his colleagues.

That was pretty good advice. With so many uncertainties about the domestic and global economies, a promise of patient open-mindedness made sense — as a holding action, at least. But such a policy would be counterproductive if applied inconsistently or if it descended into opaque ad hockery. And that’s exactly what’s happened.

In his Senate testimony last month, Powell as good as promised there’d be a cut in interest rates at the next meeting of the Federal Open Market Committee. The reasoning, however, was obscure. The economy was still in “a very good place,” Powell said. Few observers thought that the most recent data demanded a policy change.

What had changed, apparently, was Powell’s understanding of the data. The relationship between unemployment and inflation has broken down, he noted during his testimony, and the so-called neutral rate of interest seems to be lower than previously estimated. The somewhat puzzling implication: The economy might be in a very good place, but interest rates were too high.

Powell’s post-meeting press conference made things worse. He described the just-announced quarter-point cut in the benchmark interest rate as “a bit of insurance” against the risk of weak global growth and ongoing trade frictions. Sounds sensible. But wait: What does insurance have to do with the supposedly broken trade-off between inflation and unemployment, or the sinking neutral rate of interest? On the face of it, nothing. (And by the way, using interest rates to insure against things that might or might not happen is not the data-dependence that Powell used to talk about.)

A fair question to ask at this point is, so what? The Fed just executed a negligible change in interest rates that will, in itself, have almost no effect on the economy. What does it matter if this non-event hasn’t been well thought through or convincingly explained?

Unfortunately, it matters a lot. In monetary policy, expectations rule. The Fed nudges the economy chiefly through changing expectations about monetary conditions and the level of demand. When interest rates are only a percentage point or two above zero, leaving monetary policy tightly hemmed in, guiding expectations isn’t just the main thing — it’s about the only thing. If the Fed can’t do that, it’s virtually helpless.

That’s why the need for change is so urgent. A range of choices presents itself.

The simplest step forward would be to inject some intellectual discipline by adopting a policy rule — not to set the interest rate mechanically but as an expository aid. It could be some kind of Taylor rule, which ties the policy rate by formula to the neutral interest rate, the inflation rate and the gap between actual and potential output. The Fed needn’t and shouldn’t follow such a rule blindly — the world is too complicated — but its deliberations could be more explicitly guided by it, and officials would have to explain deviations from what the rule suggests.

If nothing else, this would bring some badly needed order to the Fed’s all-important communications with financial markets.

There’s much to be said, though, for bigger changes. The simplest Taylor rule essentially looks just at now (so far as now can be measured); it doesn’t look at the future or the past. In principle, a monetary-policy rule should do all three: take in information about where the economy currently stands, how it’s likely to evolve, and where it has been.

This last point is subtle but essential, especially when interest rates are very low. If inflation persistently undershoots its target, effective policy should aim to make up the difference with a period of overshooting. In effect, it wouldn’t let bygones be bygones. When interest rates are at or close to zero, ignoring past failures to reach a suitable level of inflation imparts an additional strong disinflationary bias to the system.

Other considerations also need to be weighed. For instance, there’s a strong case for casting the target for monetary policy in terms of total demand (nominal GDP) rather than in separately specified numbers for its components (inflation and real output). A target based on forecast growth in nominal GDP (averaged over time), or on the forecast level of nominal GDP, would push all the buttons.

In particular, it would be easy to explain, be agnostic about things nobody knows (such as the neutral rate of interest and the economy’s potential output), and have the right backward- and forward-looking properties.

Such a far-reaching change would be asking a lot, of course, testing the Fed’s political skills and requiring other changes in the way it organizes itself and reports its deliberations to the public. But if very low interest rates are going to remain a persistent feature of the U.S. economy, which seems likely, some pretty radical thinking is exactly what’s required.

Clive Crook is a Bloomberg columnist who writes about economics, finance and politics. He was previously chief Washington commentator for the Financial Times, a correspondent and editor for the Economist and a senior editor at the Atlantic.