How high and for how long would the Federal Reserve let prices rise before raising interest rates?

Those are the core questions ahead of Fed Chairman Jerome Powell’s speech scheduled for Thursday morning, in which the world’s most powerful central banker may unveil changes to the way the Fed approaches its 2% target on inflation.

The expectation is for the Fed to unveil a strategy known as flexible average inflation targeting, where policymakers tolerate prices rising above 2% for periods of time to compensate for other periods of time when inflation is running below target.

But in the midst of a global pandemic and with mathematical challenges behind getting an average of 2% inflation, the strategy may open the door for more questions than answers.

“The new framework will be deliberately vague – it will be a long way from a mechanical rule with actual numbers,” wrote Capital Economics chief U.S. economist Paul Ashworth.

So what would flexible average inflation targeting look like? And what does the Fed hope to do with it?

The problem

Since the Fed adopted the 2% target in 2012, actual inflation has drifted below its goal. Core personal consumption expenditures (the central bank’s preferred measure of inflation, which excludes energy and food prices) has averaged only about 1.6%, touching 2% only briefly in 2018.

The Fed, in addition to other central banks around the world, feel 2% is the magic number to support healthy levels of consumption.

But steering inflation is like walking a tightrope; rampant inflation brings back nightmares of the runaway prices in the 1970s but persistently low inflation could signal an economy suffering from below-potential levels of consumption and investment.

The (possible) solution

A flexible average inflation targeting strategy would allow the Fed to overshoot its 2% target to compensate for periods of time when inflation has run below its target.

But basic mathematics shows that achieving an average of 2% depends heavily on the window of time in question and how much undershooting the Fed is trying to compensate for.

For example: if the Fed was trying to “make up” for all inflation misses since November 2008, the Fed would need 42 years of 2.1% year-over-year core PCE inflation to average out to 2%, based on estimates from Bank of America’s fixed income research team.

The Fed may not want to lock themselves into a policy with such a wide range of outcomes. Goldman Sachs wrote in a note Tuesday that the Fed is therefore unlikely to take on any strategy “explicitly tracking and committing to make up for past downside misses.”