Column: Fed's 'R-star' becomes black hole

LONDON, March 10 (Reuters) - After six weeks of frantic forecast changes, world markets seem to have lost the plot on where interest rates will go in this cycle - mainly because the central banks are in the same boat and both groups seem to have lost confidence in a lodestar.

Difficulty figuring out just where the Federal Reserve's year-old tightening cycle is going to end - and where it pans out after that - is replicated far and wide, but the U.S. central bank puzzle sits at centre stage as usual.

The problem is simply that long-term visibility has disappeared in the face of distortions related to the pandemic and war-related energy shock, and forward policy guidance has been effectively ripped up - with everyone now flying blind from economic data update to update.

Fed Chair Jerome Powell basically admitted as much this week - suggesting just two February reports on jobs and inflation over the next week could dictate whether the central bank would return to 50-basis-point rate hikes from the slower 25-basis-point pace to which it had only last month confidently agreed to gear down.

Such jumpy 'data dependency' speaks poorly of the guiding light of policy models or faith in quasi-scientific assessments of 'natural' long-term real rates of interest - the so-called 'r*' from algebraic models that denotes the sustainable interest rate that neither stimulates nor reins in the wider economy.

A chronic inability to gauge economic slack and potential in real time means no one's quite sure what that new equilibrium is or should be - and, by extension, whether the Fed's policy rate path is overkill in fighting inflation or still way short.

The upshot has been wastebaskets full of crumpled forecasts.

Since the start of February - when economists and policymakers alike were confounded by a new year boomlet in retailing, hiring and a resurgence of core inflation - peak interest rate bets seem to be changing almost by the week.

Bond volatility has returned to the highs of late last year (.MOVE), with 2-year Treasury yields briefly back above 5% for the first time in 15 years, 10-year yields back above 4% and the 2-10 year yield curve inverting to its deepest since 1981.

By Feb. 17, Goldman Sachs and Bank of America upped so-called 'terminal rate' calls by an additional quarter of a percentage point to 5.25%-5.50%. Just three weeks later, Goldman raised it yet again and BofA and BlackRock were talking about a chance of a 6% peak all of sudden - some 120 basis points above what markets were pricing only in January.

Others followed and the rethinks were not just about the Fed, with JPMorgan lifting its peak European Central Bank policy rate forecast twice in the space of just two weeks to 3.75% - and still a quarter of a percentage point below where markets are now indicating.

So is everyone now just second guessing a Fed that is itself less than sure anymore?

'DARK SKY'

For its part, Fed policymakers' quarterly projections of long-term neutral policy rates have remained remarkably stable during such a turbulent and unpredictable four-year period.

Barring a one-quarter downtick early last year, the median long-term Fed projection has stayed at 2.5% since April 2019.

Any upward revision in that projection at the March 21-22 policy meeting would signal the Fed thinks the economy has indeed moved to a different sustainable plane - resilient to more than 400 basis points of rate hikes in a year - and would suggest it needs to do even more today to cause sufficient drag on activity to sap inflation.

But the debate is all over the shop.

Based on traditional and long-abandoned fixed policy models, Cleveland Fed researchers reckon policy is already more aggressive than any of those rules suggest. And estimates from San Francisco Fed economists suggest the real impact of policy, modelled by the current 6.3% 'proxy' rate, which combines Fed rates and broad financial conditions, is harsher than it looks.

What all that means for R-star is another question. If the macro volatility hinges on temporary supply distortions, then not much may have changed longer term - even if that's hard to assert. The wild economic swings of the pandemic forced New York Fed staffers to simply abandon regular updates on their estimates of the R* natural rate in late 2020.

In a deep-dive review into a variety of R-star models this week, JPMorgan's Joseph Lupton and Dan Weitzenfeld concluded they were all basically flawed as policy-setting tools - "uncertain at best and useless or misleading at worst."

In a report called "Dark sky: On the ill-fated search for R-star," Lupton and Weitzenfeld said lack of clarity points to greater data dependence for gauging the policy stance.

"This is not ideal as it points to central banks feeling around in the dark," they wrote. "It also diminishes the power of forward guidance as reaction functions can shift with incoming data."

What are investors to do? Work it out themselves.

Ashok Bhatia, deputy chief investment officer for fixed income at Neuberger Berman, reckons there is a structural "multi-year" change going on in the bond markets that at least means the era of zero or negative real, inflation-adjusted, yields is over.

The political and policy appetite for zero interest rates or quantitative easing - which seemed to chase estimates of R-star ever lower over the past decade - is gone. At the same time, real yields above 4% have proven unsustainable historically.

Bhatia feels real yields somewhere in the middle is where markets will settle. Given the economy-wide accumulation of debt over recent years, real 10-year yields in a 1.5%-2.0% range probably works. And if inflation returns to 2.5%, then 10-year nominal yields of 4.0%-4.5% may be around for some time to come.

But he too knows that's a shaky assumption on R-star.

"We'd all admit there's a lot of uncertainty about where that number should be."

The opinions expressed here are those of the author, a columnist for Reuters.

Our Standards: The Thomson Reuters Trust Principles.