Reading the Economic Calendar
CPI, NFP, PCE, ISM — how to read the surprise, not the number
① Economic data moves markets on the surprise (actual vs. expected), not the level. Jobs can grow and stocks still rise if growth was slower than expected; inflation can fall and stocks still drop if it didn't fall as much as expected.
② The four big prints — jobs report (NFP), CPI, PCE, ISM — measure different things and come from different agencies. When the Fed talks about its 2% target, the yardstick is core PCE, not CPI.
③ Know the transmission chain — data → rate-cut/hike expectations → Treasury yields → stocks and the dollar — and the day-after move makes sense. For investors holding U.S. stocks from abroad, there's one more layer: the exchange rate.
📅 What the economic calendar is — the monthly rhythm
The major U.S. data releases arrive in roughly the same order every month. Knowing that order tells you what to watch this week.
- Early month — ISM Manufacturing PMI (business day 1), ISM Services PMI (business day 3), ADP private payrolls (Wed), and the jobs report (NFP) on the first Friday. The month's first read on the economy is bunched into this week.
- Mid-month — consumer prices (CPI), producer prices (PPI), retail sales. Around the 10th.
- Month-end — PCE (the personal-consumption price index) with personal income, plus GDP each quarter. PCE, the Fed's preferred gauge, is the last big print of the month.
Most releases land at 8:30 a.m. ET (jobs, prices) or 10:00 a.m. ET (ISM, retail sales). marketbrief's daily brief lays out "released today" and "upcoming" in exactly this order.
| Indicator | Agency | Covers | Why watch it |
|---|---|---|---|
| Jobs report (NFP) | BLS (Labor Dept.) | Prior month | The labor market is the root of growth and inflation. Highest volatility. |
| CPI | BLS (Labor Dept.) | Prior month | The most widely used inflation gauge. Shelter is a large weight. |
| PCE | BEA (Commerce Dept.) | Prior month | The gauge the Fed's official 2% target is defined against. |
| ISM PMI | ISM (trade body) | Prior month | The earliest sentiment/activity read. 50 is the expansion/contraction line. |
💼 The jobs report (NFP) — the most volatile hour
Released the first Friday of each month at 8:30 a.m. ET, the jobs report moves markets more than any other release on the day. One report actually contains two separate surveys.
- Establishment survey → the change in nonfarm payrolls (the "+XXK" number) and average hourly earnings.
- Household survey → the unemployment rate and the labor-force participation rate.
Because the two use different samples, they can point in opposite directions — payrolls up while unemployment also rises (which can happen when participation increases). Read only the headline and you'll misread the report.
Nonfarm payrolls are revised twice over the following two months, and the revisions can be large. The number on release day is closer to a preliminary estimate. In recent years U.S. payrolls have often been revised sharply lower after the fact, occasionally turning "the number the market reacted to" into something that didn't happen. That's where "don't set your direction on a single print" comes from.
Which direction counts as "bad news" flips with the regime. When the Fed was weighing cuts, cooling jobs gave it a reason to cut, so weak jobs = good for stocks. In a regime like now, with hikes on the table, a too-strong jobs number signals "tighten more" and reads as bad news.
A concrete case: the July 2026 jobs report (released in early August) showed nonfarm payrolls at −23K, well below the roughly +85K expected. September rate-hike odds dropped immediately and Treasury yields fell. The August ADP private-payrolls figure then came in at +38K (vs. +47K expected), reinforcing the "labor market is clearly cooling" read. Both numbers moved markets because they undershot expectations, not because the level was low.
📈 CPI vs. PCE — why two inflation gauges
Both measure inflation, but the math differs and so does what the market watches.
| Feature | CPI | PCE |
|---|---|---|
| Agency | BLS (Labor Dept.) | BEA (Commerce Dept.) |
| Basket | Nearly fixed — what consumers actually bought | Substitution-adjusted — assumes people switch when prices rise |
| Shelter weight | About one-third — large | Smaller than CPI |
| Healthcare | Only out-of-pocket spending | Includes insurer/government outlays (some items use PPI) |
| Fed target | — | Core PCE 2% is the official benchmark |
CPI usually prints a bit higher than PCE (the shelter-weight gap is the main reason), so "CPI 3%" and "PCE 2.5%" can describe the same environment. To gauge the Fed's stance, watch PCE — core PCE in particular.
"Core" strips out food and energy, which swing with weather and oil and obscure the trend. The Fed treats core as the trend signal. But consumers feel the headline, so on release day the market looks at both.
Producer prices (PPI) come out the day after CPI, and PCE draws on PPI rather than CPI for some categories — healthcare, airfares, portfolio-management fees. So even when CPI is quiet, a jump in core PPI signals "this PCE could run hotter than expected." July 2026 was exactly that case (see July CPI & PPI, fully analyzed).
🏭 ISM PMI — 50 is the line
The ISM Purchasing Managers' Index is the first activity read each month. It asks purchasing managers at hundreds of firms whether conditions improved or worsened versus the prior month.
- Above 50 = expansion, below 50 = contraction. 55 is strong expansion; 45 is clear contraction.
- Manufacturing PMI lands on business day 1, Services PMI on business day 3. Services is far larger in the U.S. economy, so it matters more for the index.
- The sub-indexes carry more information than the headline — new orders leads next month, employment hints at NFP, and prices paid is an inflation preview.
August 2026 ISM Manufacturing PMI was 54.6, holding in expansion. With PMI, the signal is crossing or losing the 50 line and the direction of travel, not the absolute level.
🔗 How a data point reaches stocks and FX
When a release hits, it flows to asset prices in roughly this order.
① The surprise (actual − expected) → ② a shift in rate-cut/hike expectations (visible in CME FedWatch odds) → ③ Treasury yields move (the policy-sensitive 2-year first, the 10-year following) → ④ stocks and the dollar.
Within equities, high-valuation growth stocks are the most rate-sensitive, because the rate is the denominator when future earnings are discounted to today. When rates rise, growth wobbles first and dividend/value names hold up better (covered with data on all 11 sectors in the sector rotation analysis).
Strong data → higher yields → a stronger dollar is the base case. For an unhedged investor holding U.S. stocks from abroad, that cuts both ways — a rising dollar delivers an FX gain, but the same cause (higher rates) pressures the share price. Return in your home currency = index return ± FX P&L, so on a data night the two can offset or compound. Check staggered-conversion timing with the live FX calculator.
The same release often sees its move reverse within a day. After Fed Chair Warsh's remarks at Jackson Hole in late August 2026, September hike odds jumped from the 30s into the high 50s (%), then eased back a few days later when New York Fed President Williams signaled there was "no case" for a September hike (see the Jackson Hole / Warsh analysis). Reposition heavily off the release-day spike and you get caught in that reversal.
🛡️ A release-day checklist
What to check heading into a release, and right after it.
| When | What to check |
|---|---|
| Before | Note the consensus and the prior print, and decide in advance which direction of surprise hurts your positioning. Avoid large trades or FX conversions right before a release. |
| Right after | Look past the headline to core, revisions, and sub-indexes. For NFP, wages and the revision size; for CPI, core and shelter; for ISM, new orders and prices paid. |
| 1–3 sessions | The first-day move can reverse. Wait about three sessions before judging. This is the stretch where Fed-official remarks move things more than the data did. |
① "More jobs is always good" — regime-dependent. In a hiking regime, strong jobs is bad news.
② "Just watch headline CPI" — the Fed watches core PCE. Headline is for what you feel; core is for the trend.
③ "In line with consensus means no reaction" — if positioning was lopsided going in, an in-line print alone can trigger a reversal.
④ "One print sets the direction" — revisions can flip it. Read it as a 2–3 month trend.
Having an AI summarize a data release (PDF) is convenient, but verify the numbers against the source release and the agency's own tables, not the summary. Why models confuse digits, month-over-month, and year-over-year is covered in why AI gets numbers wrong; handling long documents is in reading documents with AI.
📆 What to watch on the calendar now (September 2026)
The debate right now is not "cut or hold" but "hold or hike" — which is why the market is unusually sensitive to the three releases below.
| Date | Event | Why it matters |
|---|---|---|
| 2026-09-04 | August jobs report (NFP) | A third straight soft print after July's −23K and weak August ADP would push the hike case well back |
| 2026-09-11 | August CPI | The last CPI before the September FOMC — core trend and tariff pass-through |
| 2026-09-16 | September FOMC + dot plot (SEP) | Hold vs. hike, plus hints for the 2027 path |
The lens for reading these three is in the complete Fed rate guide and the higher-for-longer scenario; the recent path of the inflation gauges continues in July CPI & PPI, fully analyzed. Individual terms (NFP, PCE, PMI, and so on) are laid out in 50 essential US stock terms and core concepts explained.
As it turned out, both the 9/4 jobs report and the 9/11 CPI mentioned above beat expectations, and the 9/16 FOMC did deliver a hike — see Inside the September FOMC for that outcome and what the dot plot says next.
※ This article is a guide to reading economic data and is not investment advice in any particular direction. The figures cited (July 2026 payrolls −23K, August ADP +38K, August ISM Manufacturing 54.6, September hike odds, etc.) are the releases and market prices available as of early September 2026 and can change with revisions or later releases. Verify release values and consensus against the source agencies (BLS, BEA, ISM) and CME FedWatch directly. Investment decisions and their outcomes are the investor's own responsibility.
- U.S. Bureau of Labor Statistics — Schedule of Selected Releases (jobs report & CPI dates)
- U.S. Bureau of Economic Analysis — PCE Price Index
- U.S. Bureau of Economic Analysis — Core PCE Price Index (ex food & energy)
- Institute for Supply Management — ISM Report On Business (Manufacturing & Services PMI)
- CME Group — Understanding the CME FedWatch Tool Methodology
- CNBC — Odds the Fed hikes in September tumble following big July jobs miss
- Yahoo Finance — The Odds of a September Rate Hike Have Plunged
※ This report is provided for informational and educational purposes only and does not constitute a recommendation to buy or sell any security.
New research, when it lands
Subscribe and the next deep dive comes to you, along with the daily market brief — free, unsubscribe anytime.