Why It Matters
When an analyst says "home sales rose 0.8% month-over-month," they mean this month's number is 0.8% higher than last month's. MoM is the fastest-cadence change measurement for monthly data — you see it on mortgage rates, unemployment, home sales, permits, SAAR series, and pretty much every housing release. The critical rule: MoM only makes sense on seasonally adjusted data. For a seasonally adjusted series, MoM tells you acceleration or deceleration in the underlying trend. For NSA (not seasonally adjusted) data, MoM is mostly calendar noise — November-to-December retail employment always falls because the holiday season is ending, not because anything changed economically. The classic investor trap: comparing metro unemployment (LAUS, which is NSA at metro grain) month-over-month. That number moves with the seasons, not with the labor market.
At a Glance
- What it is: Percent or absolute change from the prior month to current month.
- Formula: `MoM % = (Current month − Prior month) / Prior month × 100`
- Use cases: Acceleration tracking on SA series, short-term direction monitoring, newsletter headlines.
- When it's valid: Only on seasonally adjusted data. SA national unemployment, SA housing starts, SA home sales, mortgage rates, federal funds rate.
- When it's noise: NSA data — metro-level LAUS, raw retail sales, raw employment for industries with strong seasonal patterns.
- Common trap: Pulling metro unemployment MoM and reading it as a labor market signal. Use YoY instead.
How It Works
The math is trivial. MoM is just the percent change between two consecutive monthly values. A mortgage rate at 6.80% this month vs 6.72% last month is +0.08pp (percentage points) MoM, or +1.2% MoM in proportional terms. Home sales at 4.02M SAAR this month vs 3.99M SAAR last month is +0.8% MoM. The arithmetic is the same across every series. What matters is whether the comparison is meaningful — which depends on whether the series is seasonally adjusted.
When MoM works. On SA series, month-to-month changes reflect changes in the underlying trend, not the calendar. SA national unemployment moved from 3.7% to 3.8% — that's a 0.1pp deterioration in the labor market. SA housing starts moved from 1.38M to 1.40M — the construction pipeline is accelerating slightly. SA existing home sales rose 0.8% MoM — the housing market is picking up pace. These MoM readings answer real questions about direction.
When MoM fails. On NSA series, month-to-month swings are dominated by the seasonal pattern. Metro LAUS unemployment from November to December always rises 0.3-0.5 points because December layoffs (retail seasonal workers, construction winter pause) are predictable. Reading that as "the labor market is weakening" is wrong — it's calendar, not economy. Similarly, NSA retail sales, NSA housing permits, and any small-sample local series have strong seasonal components that make MoM uninformative.
MoM vs YoY vs SAAR — three tools, three questions. For short-term direction on SA data, use MoM. For seasonal stripping when SA isn't available, use YoY. For pace level on housing and retail series, read SAAR. The three complement each other. A home-sales release typically gives all three: SAAR (pace), MoM SAAR change (short-term direction), YoY (context vs prior year). Pick the right one for the question you're asking.
Real-World Example
María Fernández reads the monthly NAR home sales release without the MoM trap.
María is evaluating markets ahead of a multifamily acquisition. March NAR drops. The release says "existing home sales rose 0.8% month-over-month in March to a 4.02M SAAR."
She pulls three readings:
- MoM SAAR change: +0.8% (March vs February)
- YoY SAAR change: +4.4% (March vs March last year)
- 3-month MoM trend: +0.3% average over Jan, Feb, Mar
Three different conclusions. The MoM says the market accelerated modestly from February. The YoY says the market is materially stronger than a year ago. The 3-month trend smooths the noise — consistent but slow improvement.
María also pulls the same exercise for Austin unemployment (LAUS) to gut-check the local labor market. Raw MoM for Austin LAUS in November vs December shows +0.4pp. Without thinking, that would read as "Austin labor market deteriorating." But LAUS is NSA at metro grain — the December seasonal uptick is always there. She checks YoY instead: Austin December 2025 vs December 2024, -0.2pp. Labor market actually tightening.
Same tool (change between two months), two different contexts, two different interpretations. The SA-vs-NSA distinction matters more than anything else about MoM.
Pros & Cons
- Simplest possible change measurement — just two values and a subtraction
- Fastest cadence signal — catches trend shifts 12 months before YoY does
- Standard across all federal and market data releases
- Lets you build momentum indicators (3-month, 6-month MoM averages)
- Complements SAAR (level) and YoY (annual context)
- Invalid on NSA data without interpretation adjustment
- Single-month MoM moves are noisy — revisions and random variation can produce 0.5-1% swings that aren't signal
- Doesn't answer "what's the level?" — only direction
- Easy to misread: "MoM up 4%" on NSA data is usually seasonal, not trend
- Short horizon can miss slower-developing changes — a 0.3% MoM gain for 8 straight months is a bigger story than one single +2% month
Watch Out
- Metro LAUS MoM is a trap: BLS publishes metro unemployment NSA. A November-to-December MoM rise is calendar, not economy. Use YoY for metro labor questions.
- Single-month moves are noisy: Any single MoM reading can swing from revision, seasonal-model changes, or random sampling variation. Look at 3-month trends, not single months.
- Housing data is SA by default: NAR existing home sales, Census new home sales, Census permits and housing starts are all SA/SAAR. MoM on those is valid.
- Mortgage rates are daily, not monthly: MoM on the 30-year fixed rate (FRED MORTGAGE30US) uses the month-end values or month averages. Neither is standardized across analysts — check which convention the source is using.
- Revisions move history: An SA series you pulled in March might revise February's number by May, making your old MoM calculation stale. Pull fresh data for backtests.
Ask an Investor
The Takeaway
MoM is the fastest-cadence change measurement in monthly time-series analysis. Use it for short-term direction on seasonally adjusted series — mortgage rates, SA unemployment, SA home sales, SAAR starts. Never use it on NSA data without adjustment — metro LAUS unemployment, raw retail sales, any small-sample local series. The correct workflow: pull the SA version if available, compute MoM, watch the 3-month trend. If only NSA is available (common for metro-level BLS data), pivot to YoY comparisons instead. MoM is a tool — it works cleanly on SA data, fails silently on NSA data, and the difference between them is the difference between reading the economy and reading the calendar. Data sources: FRED, NAR, Census, BLS — all publish MoM directly on SA releases.