# HLW vs Lubik-Matthes Neutral Rate (r*) Models (2026)

> Compare NY Fed Holston-Laubach-Williams (HLW) semi-structural r* at +0.82% against Richmond Fed Lubik-Matthes TVP-VAR r* at +2.15% and market TIPS forwards.

Canonical: https://alphapulsefin.com/analysis/holston-laubach-williams-lubik-matthes-neutral-rate-r-star
Last updated: 2026-11-14T19:27:41.424Z

**Publication:** AlphaPulse Finance
**Category:** Fomc Treasury Term Premium
**Published:** 2026-11-14T14:35:25.915Z
**Author:** Dr. Marcus Vance, CFA (CFA, PhD in Financial Economics (Wharton))
**Reviewer:** Not recorded

## Answer-first summary

Divergence of 133 bps between NY Fed Holston-Laubach-Williams (+0.82%) and Richmond Fed Lubik-Matthes (+2.15%) real neutral rate estimates frames the macro debate over FOMC policy restrictiveness.

## Key takeaways

- Official Fed real r* trackers span a 133 bps range from +0.82% (NY Fed HLW) to +2.15% (Richmond Fed Lubik-Matthes TVP-VAR).
- Adjusting 5y5y forward TIPS real yields (2.12%) for a ~30 bps real term premium implies a market-priced r* near +1.82% (3.82% nominal).
- At a +1.48% ex-ante real EFFR, policy appears +66 bps restrictive under HLW yet -34 bps accommodative against market-implied TIPS forwards.
- One-standard-deviation Kalman filter standard errors around HLW exceed +/-115 bps, reinforcing why FOMC officials navigate by realized labor and inflation data.

## Related financial entities

- **10-Year U.S. TIPS Real Yield & Breakeven Curve** (TIPS-10Y-REAL) — product: 10-Year U.S. TIPS Real Yield & Breakeven Curve (TIPS-10Y-REAL) refers to an institutional financial benchmark and regulatory standard monitored by FRED / U.S. Treasury (US), currently calibrated at 1.98% (Real Yield) as of October 2026.
- **Federal Reserve System** (FED) — company: The central banking system of the United States responsible for FOMC monetary policy, reserve management, and systemic financial stability.
- **FOMC 3.75%–4.00% Target Rate Corridor** (FOMC-CORRIDOR) — product: FOMC 3.75%–4.00% Target Rate Corridor (FOMC-CORRIDOR) refers to an institutional financial benchmark and regulatory standard monitored by Federal Reserve Board (US), currently calibrated at 3.88% (EFFR Median) as of October 2026.

## Analysis

Whether an [FOMC 3.75%-4.00% Target Corridor](/entities/fomc-corridor-fomc-3-75-4-00-target-rate-corridor) actively slows US aggregate demand or merely sits near equilibrium depends entirely on the unobserved real neutral interest rate, `r*`. Across the [Federal Reserve System](/entities/federal-reserve-system) research network in late 2026, official econometric trackers disagree by 133 bps: the New York Fed Holston-Laubach-Williams (HLW) semi-structural model places real `r*` near +0.82%, whereas the Richmond Fed Lubik-Matthes (LM) time-varying parameter vector autoregression (TVP-VAR) estimates +2.15%—closely matching the 1.98% spot real yield on the [10-Year US TIPS Curve](/entities/tips-10y-real-10-year-u-s-tips-real-yield-breakeven-curve).

## IS-Curve Slope Fragility in Holston-Laubach-Williams vs Agnostic TVP-VAR

Holston, Laubach, and Williams (2017, updated 2023 with COVID-19 outlier variance scaling) specify a New Keynesian state-space system linking the output gap to real short-rate gaps (IS curve) and core inflation to the output gap (Phillips curve), decomposing `r* = c * g + z`, where `g` is trend potential GDP growth and `z` captures non-growth savings/investment shifts. When the US economy expanded at 2.7%-3.0% real GDP during 2023-2025 despite 5.25%-5.50% nominal policy rates, HLW attributed resilience mostly to positive transitory demand shocks and supply healing rather than raising `z`, keeping `r*` below 1.0%.

Lubik and Matthes (2015) avoid imposing theoretical IS or Phillips curve elasticities altogether. Their three-variable Bayesian TVP-VAR over real GDP growth, PCE inflation, and real short rates defines `r*` as the five-year-ahead conditional forecast of the real federal funds rate. Because real rates stayed near +1.8% to +2.3% without triggering recessionary output contraction, the Bayesian posterior rapidly updated `r*` above +2.10%.

![Econometric architecture comparing HLW structural Kalman filter against Lubik-Matthes TVP-VAR](/api/figure?type=flow&title=Real%20Neutral%20Rate%20%28r*%29%20Estimation%20Divergence%20Pipeline&subtitle=Why%20post-2022%20economic%20resilience%20under%20high%20rates%20splits%20structural%20vs%20Bayesian%20models&steps=Macro%20Input%20Vector%3A%3AReal%20GDP%2C%20Core%20PCE%20inflation%2C%20and%20nominal%20EFFR%20trajectory%7CHLW%20Structural%20Filter%3A%3AIS%2FPhillips%20slopes%20treat%20resilient%20GDP%20as%20transitory%20shock%20%28%2B0.82%25%29%7CLM%20Bayesian%20TVP-VAR%3A%3ATime-varying%20parameters%20update%205Y%20conditional%20real%20mean%20%28%2B2.15%25%29%7C5y5y%20TIPS%20Market%20Test%3A%3A2.12%25%20forward%20real%20yield%20less%2030%20bps%20term%20premium%20%3D%20%2B1.82%25%20market%20r*&metrics=NY%20Fed%20HLW%20r*%3A%3A%2B0.82%25%3A%3A2.82%25%20nominal%20w%2F%202%25%20PCE%7CRichmond%20LM%20r*%3A%3A%2B2.15%25%3A%3A4.15%25%20nominal%20equivalent%7CMarket-Implied%20r*%3A%3A%2B1.82%25%3A%3ATerm-premium-adj%205y5y%20TIPS%7CModel%20Spread%3A%3A133%20bps%3A%3ARecord%20LM-minus-HLW%20gap)

## Market-Implied Forward Real Rates (5y5y TIPS) vs Econometric Filters

Fixed-income portfolio managers at firms like [Alpine Woods](/institutions/alpine-woods) and primary dealer desks at [Cantor Fitzgerald](/institutions/cantor) compare central bank econometric trackers against 5-year, 5-year forward TIPS real yields (~2.12% in Q4 2026). Even after subtracting an estimated 25-to-35 bps real term premium using the D'Amico-Kim-Wei (DKW) model, market pricing implies an equilibrium real short rate of +1.75% to +1.85%—nearly 100 bps above HLW.

If true `r*` sits at +1.80% rather than +0.82%, then an EFFR median of 3.88% against 2.45% one-year forward inflation delivers an ex-ante real rate of +1.43%—meaning monetary policy would already be mildly accommodative rather than restrictive, explaining robust S&P 500 capex and tight credit spreads monitored by the [SEC](/regulators/sec) and [FINRA](/regulators/finra).

```formula
HLW Decomposition: r*_t = c * g_t + z_t
Ex-Ante Real Short Rate: r_real_t = EFFR_t - E_t[pi_{t,t+4Q}]
Policy_Stance_Gap_bps = (r_real_t - r*_t) * 100

Worked Policy Stance Contrast (EFFR = 3.88%, 1Y Expected Core PCE = 2.40% -> r_real = +1.48%):
- Under HLW (r* = +0.82%): Gap = (1.48% - 0.82%) * 100 = +66 bps (Restrictive -> cut bias)
- Under Market TIPS (r* = +1.82%): Gap = (1.48% - 1.82%) * 100 = -34 bps (Accommodative -> hold bias)
- Valuation Swing on 10Y Fair Yield = +100 bps shift in terminal anchor
```

## Real Policy Stance Gap and Semi-Structural Decomposition Math

Quantifying the real policy gap (`r_actual - r*`) across competing specifications illustrates why FOMC committee hawks and doves interpret identical macro data differently.

![Implied monetary policy restrictiveness gap across HLW, SEP median, market TIPS, and Lubik-Matthes](/api/figure?type=bars&title=Ex-Ante%20Monetary%20Restrictiveness%20%28Real%20EFFR%20%2B1.48%25%20Minus%20r*%29&subtitle=Policy%20stance%20assessment%20assuming%203.88%25%20EFFR%20and%202.40%25%201Y%20expected%20PCE%20inflation&steps=vs%20NY%20Fed%20HLW%20%28%2B0.82%25%29%3A%3A%2B66%20bps%20restrictive%20stance%20-%3E%20argues%20for%20continued%20cuts%7Cvs%20SEP%20Median%20%28%2B1.00%25%29%3A%3A%2B48%20bps%20mildly%20restrictive%20-%3E%20gradual%20calibration%7Cvs%20Market%20TIPS%20%28%2B1.82%25%29%3A%3A-34%20bps%20mildly%20accommodative%20-%3E%20warns%20of%20inflation%20floor%7Cvs%20Richmond%20LM%20%28%2B2.15%25%29%3A%3A-67%20bps%20accommodative%20-%3E%20signals%20neutral%20already%20breached&metrics=Ex-Ante%20Real%20EFFR%3A%3A%2B1.48%25%3A%3A3.88%25%20EFFR%20-%202.40%25%20E%5Bpi%5D%7C10Y%20Spot%20TIPS%3A%3A1.98%25%3A%3ATIPS-10Y-REAL%20benchmark%7CHLW%20Trend%20g%3A%3A%2B1.72%25%3A%3APotential%20GDP%20growth%7CConfidence%20Band%3A%3A%2B%2F-115%20bps%3A%3A1-sd%20HLW%20filter%20error)

## Institutional Neutral Rate Benchmark Comparison

Asset allocators synthesize academic, regional Reserve Bank, and cleared market benchmarks under [Federal Reserve](/regulators/u-s-federal-reserve-system-156) and [CFTC](/regulators/cftc) data frameworks.

| r* Benchmark Framework | Publishing Institution | Real r* Estimate | Nominal Neutral (at 2% PCE) | Implied Policy Stance at 3.88% EFFR |
| --- | --- | --- | --- | --- |
| Holston-Laubach-Williams (HLW) | NY Fed ([Federal Reserve](/regulators/u-s-federal-reserve-system-156)) | +0.82% | 2.82% | +66 bps Restrictive |
| FOMC SEP Longer-Run Median | [Federal Reserve System](/entities/federal-reserve-system) | +1.00% | 3.00% | +48 bps Mildly Restrictive |
| 5y5y TIPS Forward (Adj) | Dealer Spline ([Cantor](/institutions/cantor)) | +1.82% | 3.82% | -34 bps Mildly Accommodative |
| Lubik-Matthes TVP-VAR (LM) | Richmond Fed Research | +2.15% | 4.15% | -67 bps Accommodative |

## Fiscal Crowding Out, AI Investment Demand, and the z-Factor Shift

Macroeconomic theory anchors real `r*` at the intersection of desired global saving and investment. Structural upward shifts since 2020 stem from three simultaneous shocks: persistent primary US federal deficits absorbing domestic private savings, data-center and power-grid capital formation raising corporate investment demand, and waning foreign central bank reserve accumulation. Together, those forces lift the non-growth `z` wedge from -0.65% pre-2020 to +0.35% in market-implied pricing.

Macro allocation frameworks connect this note to [FOMC Policy Path, SOMA Runoff & US Treasury Term Premia](/analysis/fomc-soma-runoff-us-treasury-term-premium-architecture), [SOMA Monthly Redemption Caps, MBS Prepayment](/analysis/soma-treasury-redemption-caps-mbs-principal-reinvestment-mechanics), and [TIC Foreign Official Treasury Custody Flows](/analysis/foreign-official-tic-custody-holdings-fx-hedged-treasury-demand), indexed under [FOMC Policy Path](/topics/fomc-treasury-term-premium) alongside [tracked macro entities](/entities).

## Frequently asked questions

### Why did the NY Fed modify the Holston-Laubach-Williams model after the 2020 pandemic?

Extreme Q2-Q3 2020 GDP swings violated Gaussian error assumptions in the original Kalman filter; the 2023 HLW update introduced time-varying innovation variance scaling and a supply-disruption indicator to prevent permanent filter distortion.

### Why does the Richmond Fed Lubik-Matthes r* estimate sit more than 130 bps above HLW?

Lubik-Matthes uses an unconstrained Bayesian TVP-VAR without imposing downward-sloping IS curve priors; when US output stayed strong during 2023-2025 rate hikes, the model raised its 5-year conditional real rate equilibrium.

### Can 5y5y forward TIPS yields be read directly as market expectations of real r*?

Not directly. Five-year, five-year forward TIPS yields equal expected long-run real short rates plus a forward real term premium (+25 to +35 bps) minus a TIPS illiquidity discount (~8 to 12 bps).

### How do chronic federal budget deficits raise the neutral rate r*?

Persistent primary deficits reduce national saving at full employment, requiring higher real equilibrium interest rates (a positive z-factor shift) to clear private investment and sovereign debt absorption.


## Source URLs

- https://www.newyorkfed.org/research/policy/rstar
- https://www.richmondfed.org/research/national_economy/natural_rate_interest
- https://www.federalreserve.gov/econres/notes/feds-notes/measuring-the-natural-rate-of-interest-20230519.html

## Citation

- APA: Dr. Marcus Vance, CFA. (2026). Holston-Laubach-Williams vs Lubik-Matthes Neutral Rate (r*) Estimation in US Rates. AlphaPulse Finance. https://alphapulsefin.com/analysis/holston-laubach-williams-lubik-matthes-neutral-rate-r-star
- Canonical article: [Holston-Laubach-Williams vs Lubik-Matthes Neutral Rate (r*) Estimation in US Rates](https://alphapulsefin.com/analysis/holston-laubach-williams-lubik-matthes-neutral-rate-r-star)
- Site research index: [AlphaPulse Finance](https://alphapulsefin.com/llms-full.txt)
