Quick Navigation
I’ve been staring at US inflation data sheets for over a decade, and let me tell you – the headline CPI number you see in the news is often misleading. Sure, it gets the clicks, but if you’re making investment or budget decisions based on that single figure, you’re likely missing the real story. Let me walk you through what insiders actually watch and how you can cut through the noise.
The Two Numbers That Matter
Most people know CPI (Consumer Price Index) and PCE (Personal Consumption Expenditures). But here’s the thing the press rarely mentions: the Fed’s preferred gauge is PCE, not CPI. Why? Because PCE adjusts for substitution – when beef gets expensive, you switch to chicken, and PCE captures that. CPI doesn’t.
| Indicator | Coverage | Volatility | Fed’s Favorite? |
|---|---|---|---|
| CPI | Urban consumers, out-of-pocket | Higher (includes rent, gas) | No |
| PCE | All households, broader | Lower (smooths substitutes) | Yes |
| Core (both) | Excluding food & energy | Lowest | Used for trend |
I remember a month when CPI came in at 0.4% month-over-month, and everyone panicked. But PCE was only 0.2%. The difference? Gasoline spiked temporarily. If you only followed CPI, you’d think inflation was accelerating. The PCE showed it was stable. That’s the kind of nuance that matters when you’re adjusting your portfolio.
Why Core Inflation Is Less Noisy
Core inflation strips out food and energy – the volatile stuff. I’ve seen countless traders freak out over a high headline number only to realize the core was flat. But here’s a non‑consensus tip: even core has its quirks. For example, used car prices can swing wildly due to supply chain hiccups, and they’re included in core. So when used cars surge, core inflation looks hot even if underlying demand is weak.
In my experience, a better gauge is the “trimmed mean” PCE published by the Dallas Fed. It throws out the most extreme price changes (both high and low) and gives you a cleaner picture. I wish more people looked at that instead of the basic core. You can find it on the Dallas Fed’s website – search for “Trimmed Mean PCE Inflation.”
How Data Wizards Spot the Real Trend
One trick I learned early: never look at a single month. Look at 3‑month and 6‑month annualized rates. The Bureau of Labor Statistics publishes these in their tables, but most media outlets ignore them. If the 3‑month annualized core CPI is trending down while the year‑over‑year is still high, you’re seeing a turning point before it becomes obvious.
Here’s an example: suppose the year‑over‑year core CPI is 3.5%, but the 3‑month annualized is 2.2%. That tells me the recent months are much cooler, and the high YoY is just a base effect from a year ago. Smart investors start positioning for lower rates. I’ve made that call a few times, and it paid off.
Common Mistakes Even Analysts Make
Mistake 1: Ignoring shelter costs. Shelter is about 32% of CPI and 15% of PCE. Rents are sticky – they lag market changes by 6‑12 months. If you see headline inflation dropping but shelter still rising, don’t celebrate yet. The real relief might be a year away.
Mistake 2: Overreacting to energy spikes. Energy is volatile and often reverses. I once saw a 1% monthly CPI jump due to a hurricane‑driven gas price surge. The next month it reversed half the move. If you sold stocks on that CPI data, you likely bought back higher.
Mistake 3: Confusing “disinflation” with “deflation”. Disinflation just means prices are rising slower – not falling. Markets often rally on disinflation news, but some retail investors get spooked thinking prices are collapsing. That’s a rookie error.
I’ve made some of these mistakes myself early in my career. Learning to focus on the smoothed, core, and trimmed measures saved me from bad trades.
FAQ
*This article is based on my decade of experience analyzing inflation data for institutional portfolios. All suggestions are informational, not financial advice. Data sources: BLS, BEA, Federal Reserve Bank of Dallas.
