If you’re not exhausted with the neutral rate discussion by now, something’s wrong with you. You need more friends, probably.
I’m tired of it, but apparently not exhausted, because I keep writing about it and otherwise editorializing around it. What does that say about me? Nothing great. But nothing that hasn’t been said before, I suppose.
Bear with me. This’ll be mildly interesting and, more importantly I’m sure, very short.
On March 22 (so, a few days ago), the New York Fed released its latest DSGE model forecasts for the real natural rate or r-star, the equilibrium rate that balances the economy at full employment and stable inflation.
DSGE stands for dynamic stochastic general equilibrium, which is gibberish to the vast majority of humanity. Suffice to say it’s a model the Fed’s used to forecast economic variables for going on 15 years. It was made available to the public starting a decade ago.
The figure above shows the model-implied neutral funds rate (nominal, using core PCE) dating to the beginning of the Fed’s hiking cycle (light blue bars) and extending out to Q1 2027 (dark blue line, leveraging inflation forecasts from the same New York Fed model). Anyone see a potential problem?
The black line is just Fed funds. It was below the DSGE-implied nominal neutral rate until Q4 of 2023 (it might’ve overshot in Q4 of 2022, but not meaningfully). So, Fed policy wasn’t restrictive on this interpretation until a few months ago.
If you go by the new dot plot (which, you’re reminded, featured upward shifts to the 2025, 2026 and long run median markers), policy will only be restrictive relative to the model-implied equilibrium rate through this year, and not by a lot even in 2024.
Plainly, there’s ambiguity here. With the exception of the historical Fed funds rate (in black), we’re adding up and comparing model outputs and macro forecasts from economists and policymakers. As such, you might plausibly suggest this is a largely meaningless exercise.
And yet, it underscores the ongoing, lingering tension between i) a pervasive suspicion that rates need to be structurally higher in the post-pandemic world, and ii) what’s likely to be a reluctance on the part of policymakers to set rates accordingly. That’s not meaningless. It’s important to the extent it’s an accurate characterization of the current macro-policy zeitgeist.
As for market expectations, traders generally see Fed funds settling around 3.70%, roughly consistent with the DSGE-implied neutral rate but, as JonesTrading’s Mike O’Rourke pointed out, more than 50bps higher than the 2026 median dot.
“If equity investors believed the terminal level of the Fed funds rate was 3.625% instead of 2.5% or 3.125%, would they remain this exuberant?” O’Rourke, whose latest daily inspired this article, wondered, before noting that 18 months ago, markets expected Fed funds to peak at 3.75%.
The FOMC hasn’t been slavishly adhering to its internal models for some time, and maybe never did.
Powell has, IIRC, acknowledged that the Fed’s models have not worked that well during the pandemic/post period. Every time he says “data dependent” he’s implicitly saying “not model-dependent”.
Possible explanations for Powell’s all-but-promise of rate cuts starting this year despite two months of disappointing inflation data – FOMC might
1. consider current rates restrictive, notwithstanding a dynamic stochastic general something or other
2. see increasing financial risk +/o economic weakness
3. have confidence that inflation will be ~2% by end 2026, regardless of visibility on the near-term path
4. be coordinating with Treasury and administration economic team on one or both mandates
5. factor in institutional implications of political outcomes
6. be using market anticipation of cuts as tool
I sometimes think about what part of the economy does the Fed have more information about than investors do, seems that would be financial system plumbing.