Peter Williams on Interest Rates, Term Premium, and the Importance of Inflation Expectations

Episode Summary

Executive Summary: Peter Williams argues that the post-pandemic economy avoided recession because nominal income growth stayed strong, fiscal support lingered, and the labor market held up despite aggressive Fed hikes. He explains how markets think about rates, term premium, R-star, and inflation expectations, stressing that models need real-time macro discipline and that the last mile back to 2% inflation will likely be slow and uneven.

Main Topics: Why the U.S. avoided recession in 2023 (Priority: 5/5): Williams says recessionary signals were present—weak housing, soft manufacturing, financial stress, and tighter credit—but the economy ‘skated through’ because nominal growth and household/fiscal support buffered the shock. How markets think about interest rates (Priority: 5/5): He distinguishes the Fed funds rate from longer-term market rates, arguing that the 30-year rate is often the most useful single rate for assessing economic impact, while structural anchors like trend inflation and R-star shape the curve. Treasury yields, term premium, and model limitations (Priority: 5/5): Williams cautions that term premium estimates are highly model-dependent and difficult to interpret in real time because expectations and term premium move together; he favors macro- and survey-disciplined approaches over purely statistical decompositions. R-star and monetary policy navigation (Priority: 5/5): He explains that R-star matters because the Fed uses it and because it informs policy space relative to the zero lower bound; his work suggests R-star fell sharply after the GFC and may have risen again after COVID and fiscal expansion. Inflation expectations and their role (Priority: 4/5): Williams argues inflation expectations matter as part of wage-setting, pricing, and confidence channels, but are noisy and hard to model; he recommends using multiple surveys and market measures in a Bayesian, signal-extraction framework. The remaining inflation problem (Priority: 5/5): He expects the final descent to 2% inflation to be harder because disinflation from core goods is fading, housing is sticky, and services/wages remain elevated, implying a slower and more cautious Fed easing cycle.

Key Arguments: The economy avoided recession not because there was no slowdown, but because the slowdown was partially offset by strong nominal income growth, which let households and firms smooth shocks. Fiscal policy was unusually powerful after the pandemic, especially through direct transfers and accumulated savings, and likely dominated monetary policy in supporting demand in that environment. Long-term rates cannot be understood from the Fed funds rate alone; market participants should focus on the broader curve, especially medium- and long-term yields that affect households and borrowing conditions. Term premium models are useful but fragile in real time because they embed assumptions about the future path of rates; their outputs should be checked against plausible macro and market narratives. A macro-based term premium framework improves discipline by linking yields to yield-curve factors and macro variables, but post-COVID nonlinearities and regime changes make all such models harder to trust mechanically. R-star should be taken seriously because the Fed uses it and because it helps define policy space; the post-GFC world likely had a much lower R-star, but recent cycles may have pushed it higher. Inflation expectations matter most as a medium-run support for pricing and wage-setting, yet surveys are noisy, response rates have worsened, and short-run movements are often dominated by food, energy, and commodities. The last mile of disinflation is difficult because core goods deflation will fade, housing is sticky, and services inflation—especially core services ex-housing—remains more persistent than markets would like.

Data Points: R-star after the GFC: around 0% - Williams says his IMF model found U.S. R-star fell more than most other models and was basically zero for much of the post-financial-crisis period. Nominal top-line growth: 5% to 7% - He argues firms and households had enough income growth to absorb shocks without tipping into recession. Market rate used as benchmark: 30-year rate - Williams says he often thinks about the 30-year market rate as the single most important interest rate for economic impact. Typical end-of-cycle spread between long rate and Fed funds: within 50 bps - He says long-term Treasury yields often end a cycle within about 50 basis points of the Fed funds rate. Fed funds rate at cycle peak: 5.5% - Used in his discussion of why duration risk looked unattractive when long yields were much lower than short rates. Long Treasury yield level discussed: near 5% then below 4% - He describes the 10-year Treasury rising close to 5% and then falling to around 4%, briefly below 4%. Treasury yield level current/mentioned: just above 4% - He notes the 10-year Treasury was a little above 4% at the time of the conversation. Model time horizon for market research: 3, 6, 12, occasionally 24 months - Williams says private-sector research is more reactive and focused on shorter horizons than public-sector policy work. Conference/fiscal context: late last year - Refers to the Bennett McCallum Memorial Conference as a chance to reflect on big-picture macro issues. Inflation target: 2% - He says his R-star framework is tightly anchored to the Fed’s 2% inflation objective. Survey sample issue: a few hundred people - He notes that a small subset of the University of Michigan survey respondents can materially affect inflation expectation readings. Fiscal shock comparison: World War II-esque - Williams characterizes pandemic-era fiscal expansion as historically large in scale and duration. Inflation expectations horizon: 5-year measures are less volatile than 1-year measures - He says longer-horizon expectations better capture underlying inflation beliefs than short-term readings. Rate hike reference: 75 basis points - He mentions the Fed’s aggressive response after a hot consumer inflation-expectations reading.

Pivotal Quotes: "we were coming out of a recovery in a very asynchronized way across the whole of the economy" — Peter Williams: Explaining why the post-pandemic cycle was unusual and hard to forecast. "if your top line is going at five or 7%, you have a lot more buffer than a world where your top line is going at one or two" — Peter Williams: Describing why nominal income growth helped households and firms absorb recessionary shocks. "the last mile hardest" — Peter Williams: Summarizing why getting inflation from roughly target-adjacent levels back to 2% may be more difficult than the earlier disinflation phase.

Implications: Markets and policymakers should expect slower, noisier disinflation and avoid overreading point estimates of term premium or R-star. The Fed may cut cautiously, while investors should focus on macro regime shifts, nominal growth, and sticky services inflation.