Starry-Eyed

On several occasions over the past six or so months, I’ve delved into the r-star debate with one overarching goal: To underscore the possibility that the world changed three years ago, and that we’re not in Kansas anymore when it comes to the natural rate.

That’s not a novel suggestion. Indeed, this is a hot topic among economists attempting to orient themselves (and their models) in a world of hot inflation.

This week, the New York Fed relaunched their r-star series, which was suspended during COVID. The event probably wouldn’t have garnered any attention even in a total news vacuum, but given debt ceiling headlines in the US, it had no chance of landing above the proverbial fold, even in the financial pages.

Esoteric though it may be, this is a (very) important subject, particularly right now, as policymakers ponder the adequacy of 2022’s rate hikes in the context of the “sufficiently restrictive” debate vis-à-vis inflation.

I won’t bury the lede: The r-star estimates are now slightly lower. “We see no signs of a significant reversal of the decline in r-star estimates evident in prior decades,” John Williams said, in a speech accompanying the relaunch. “There is no evidence that the era of very low natural rates of interest has ended.”

The New York Fed estimated r-star through the end of next year using Blue Chip forecasts for inflation, GDP and rates, and came up with a negative estimate for year-end 2024. “Evidently, the value of r-star implied by private forecasts is, if anything, even lower than today’s estimate,” Williams went on. “Time will tell whether this turns out to be the case.”

Yes, time will indeed tell, and it’s worth noting that during remarks of his own on Friday, Jerome Powell was keen to emphasize that “positive supply shocks related to globalization probably contributed significantly to the period of low inflation that either ended, or was interrupted by, the onset of the pandemic.” Those positive supply shocks “do not seem likely to be repeated,” he said.

That’s what economists probably should be factoring into their estimates, or at least thinking about, when it comes to the future of rates in a world defined by fractured supply chains, “friend”shoring, regionalization, rearmament, wage inflation, populism and so on. Alas, our insistence on the notion that The Great Moderation marked the dawning of a “new normal,” rather than a temporary reprieve from a long history of macro volatility, has left us beholden to recency bias.

Williams released a three-dozen-page paper alongside the relaunched r-star estimates. Kathryn Holston and Thomas Laubach were listed as co-authors, and I’d be remiss not to mention that Laubach unfortunately passed away in September of 2020 after a battle with cancer.

The paper claimed to address the “econometric issues caused by the pandemic” — all two of them. In essence, it was just a math exercise. So, economists pretending that economics is a hard science. They didn’t grapple with issues of political-economy or global conflict or climate change or biological threat preparedness or, just as crucially, intense friction between capital and labor across developed markets.

In other words: The paper didn’t grapple with any of the issues which will ultimately determine our macro fate. That’s not to say the conclusion (i.e., that the era of low rates didn’t come to an end in the 2020s) was wrong. It may very well prove to be absolutely correct. Rather, my point is that the work didn’t attempt to deal with any of the questions which actually matter when it comes to macro forecasting. It was just another effort to mathematize a soft science. So, the opposite of the holistic approach I’ve suggested the world needs in light of recent events.

From a policy perspective, one implication is that, as Goldman put it,”the decline in the r-star estimate since 2019 — and the potential for further declines in the estimate if the economy evolves as expected — would suggest that monetary policy is already tight and will likely remain so without rate cuts, which might suggest some incremental dovishness if FOMC officials factor the estimates into their policy decisions.”

That’s (probably) not a concern in the very near-term (i.e., in 2023), but it could raise the odds of premature easing down the road if it biases enough Fed officials against the notion that the world has well and truly changed.

More broadly, economists should’ve learned a lesson over the past three years. Economics isn’t, and never will be, a hard science. Earnest, well-meaning, brilliant and hard-working as most practitioners surely are, econometrics is mostly farce. Because at the end of the day, economics actually isn’t about numbers. Rather, it’s about the vagaries of human behavior, and not just as they relate to how we interact with one another to maximize our material well-being.

Vladimir Putin can’t change the laws of physics. Unionizing baristas can’t make 2 + 2 = 5, even if they can cajole a coffee baron to turn $12/hour into $17/hour. But Putin can drive up the price of energy overnight and the baristas can contribute to a wage-price spiral, each by simply deciding on a given course of action.

If enough people make decisions which together go against the prevailing macro trends (themselves largely a function of decisions other people made previously), then the models are useless. Throw in acts of God (or Mother Nature) and economic forecasting is mostly an exercise in futility. Put differently, econometrics simply can’t control for the factors which ultimately determine macroeconomic outcomes. Sometimes, that’s not a problem. But as we’ve learned in the 2020s, it can be ruinous when something goes wrong.