The US Federal Reserve’s job has become a lot less enviable

Are interest rates too high or too low? A simple enough question, you might think. If only. The problem is that the system connecting the US Federal Reserve’s main policy tool, the setting of interest rates, to what it ultimately wants to affect—inflation and employment—has many moving parts. One can’t answer the question without a grasp of the context.

The Fed sets a course for its policy rate, which (along with things it can’t control) affects the cost of credit, asset prices and other financial conditions. Changing conditions (and other uncontrollables) then affect aggregate demand. Changing aggregate demand (with other beyond-control factors) then shapes employment and inflation. The Fed’s goal is vague: It has to balance (over an unspecified timeframe) maximum employment and price stability, none of which is clearly defined. It’s hard enough to look back and say whether a given policy was right or wrong. To say what’s good policy in real time is infinitely harder, involving many tricky assumptions.

Yardsticks that simplify the Fed’s task, and thus constrain its choices, do have their place—so long as they’re handled wisely. The classic rule proposed by John Taylor deduces the correct policy rate by weighing the gap between actual inflation and the Fed’s target for it against the gap between actual output and that at full employment. Taylor argued that a formula yields better results than the Fed’s discretion.

The Fed follows a Taylor-type rule most of the time. But what works under normal circumstances breaks down if the economy is struck by big supply-side shocks. Taylor-like rules say today’s policy rate is much too low—on the order of 7 percentage points. Yet, raising rates this much would deliver a huge financial shock, driving up the output gap and rendering the rule’s prescription irrelevant. The economy would change so fast that a policy rate deemed too low one week would be judged too high the next.

With Taylor-type rules sidelined, other simplifying concepts are vying for attention. A favourite contender lately is the so-called neutral rate of interest—the rate consistent with full employment and the inflation target. In effect, this gestures at an appropriate Taylor-type interest rate once the economy is roughly back where the Fed wants it to be.

But the point is that the economy is not where the Fed wants it to be. Even if we knew what the neutral rate was in this other world—0.5% in real terms is a popular choice—this wouldn’t say anything useful about where things stand right now. Still, a simple answer is ever in demand. One can highlight any number of other metrics and say, “Use this to judge the policy rate.” Together with the ghostly neutral rate, labour-market measures are in vogue—understandably, because the Fed’s challenge is to get inflation down without job losses. So the Fed is advised to focus on quit rates, or the ratio of vacancies to joblessness (the Beveridge curve), or separations, or ‘discouraged workers’ or whatever.

All these indicators and more convey information about where the economy stands and the trade-offs the Fed has to weigh. None can say whether policy is too tight or too loose.

Aside from each indicator’s specific defects, it’s also worth keeping in mind Goodhart’s Law: “Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.”

The best approach is for the Fed to direct its own and others’ attention to the broadest aggregate it hopes to influence while emphasizing the limits to its control. It should say what it believes its projected policy rate implies for the path of aggregate demand, and why this path is appropriate. Strangely, despite all the comment and information the Fed provides, these intentions aren’t directly stated: They have to be inferred. According to the Fed and the average of private forecasts, the indicated course of interest rates implies gently declining growth in demand, consistent with a gradual easing of inflation, alongside little or no increase in unemployment.

The Fed should keep reminding us that things are likely to turn out differently because of all the factors it can’t control, and, if so, a new path of interest rates will be needed. But with that understood, one can reasonably ask: Is the Fed’s intended path of demand well-judged? My guess—which is all anybody can do here—is that a slightly brisker reduction of demand would help reduce inflation a little faster without materially worsening the outlook for jobs. All things considered, the Fed should push interest rates up a bit faster.

Those qualifiers are infuriating, I know. But that’s also the point. They’re needed to acknowledge the doubts, uncertainties and space for legitimate disagreement. Weak commitments are never satisfying, but, when it comes to economic forecasting, they are always wise.

Clive Crook is a Bloomberg Opinion columnist and member of the editorial board covering economics, finance and politics