Calculated Risk · Housing & Cities
TIER 4 Mon, 20 Apr 2026 14:20:09 +0000
Update on Neutral Interest Rate "Models" ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ ͏ | | ---|---|--- | | | Forwarded this email? Subscribe here for more --- # Lawler: Update on MBS Yields and Spreads ### Update on Neutral Interest Rate "Models" | | CalculatedRisk by Bill McBride --- | Apr 20 --- | --- --- | | | --- | | --- | | --- | | --- | | READ IN APP --- From housing economist Tom Lawler: After rising significantly during the first four weeks of March, current-coupon MBS (CCMBS) yields have reversed over half of that increase. CCMBS yields hit a 2026 low of 4.81% on February 27th and a 2026 high of 5.53% on March 27th, and settled at 5.11% on April 17th. CCMBS/Treasury spreads widened noticeably from February 27th to March 27th, but have since retraced virtually all of that widening. The major catalyst for both the March widening and the April narrowing in CCMBS/Treasury spreads was a change in market-implied interest-rate volatility. Rate volatility surged in March but has plummeted so far in April back to late February levels, as shown in the MOVE index chart below. | | ---|---|--- | | ---|---|--- | | ---|---|--- As a side note, CCMBS yields as of last Friday were 5 bp higher than the day before the President's post in January that the GSEs would buy $200 billion of MBS, and CCMBS/Treasury spreads are about 2 bp lower. ### Update on Neutral Interest Rate "Models" Below is a table showing various models' estimates of the so-called "neutral" US real interest rate for the fourth quarter of 2025. Note that the institutions in parentheses are sources to find these estimates, but the estimates are not those of the institutions shown. | | ---|---|--- The L-W, H-L-W, and Z models are semi-structural New Keynesian models; the L-M model is a time-varying parameter vector autoregression model; and the F-L-P model uses estimates of convenience yields, safe asset supply/demand, productivity, and demographics to estimate the neutral rate. Note that the L-W and H-L-W model are similar, though the H-L-W models was adjusted both to estimate other countries' neutral rates and to allow for time-varying volatility. Note also that the H-L-W model has for over the last year produced estimates of the so-called "output gap" that are not just lower than the L-W (or other) models but also that in my view do not pass the sniff test. E.g., the H-L-W estimate of the output gap was negative over the last 5 quarters, and for last year was about the same as that for the second half of 2010, when the unemployment rate was 9.5%. As a reminder, the concept of the "neutral" or "natural" interest rate is a "real,' or inflation-adjusted one. Thus, if inflation and expected inflation were at the Fed's target of 2%, then the average of these models would suggest that the neutral nominal interest rate were at 3.36% to 3.47%. Of course, inflation and expected inflation are NOT as low as 2%, implying that the current neutral rate would be higher than that 3.36% to 3.47% range. CalculatedRisk Newsletter is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Upgrade to paid Share _You 're currently a free subscriber to CalculatedRisk Newsletter. For the full experience, upgrade your subscription. _Copyright CR4RE,LLC 2026 Upgrade to paid --- | | | Like --- | | Comment --- | | Restack --- (C) 2026 CalculatedRisk Email: dive_hike@yahoo.com, 548 Market Street PMB 72296 San Francisco, CA 94104 Unsubscribe