Research Guide
How to analyze economic trends with free data
Is growth speeding up or slowing down? Is inflation heating or cooling? A few transparent calculations answer better than any headline.
Economic commentary is full of confident labels: "the economy is slowing", "inflation is sticky", "a soft landing". Most of these describe one release, or one chart, or one person's feeling. You can do better with a few public data series and simple, transparent arithmetic. The goal is not to predict the next number. It is to describe clearly what the data is doing and how sure you can be.
Start with year-over-year growth
For a monthly series such as industrial production, payrolls or a price index, the most common starting measure is the change from the same month a year earlier:
Year-over-year growth = 100 × (this month's value ÷ value 12 months earlier − 1)
Comparing with the same month a year ago removes most seasonal effects and smooths out some monthly noise. If you are new to the sources, our guide to where to find economic data lists the series IDs to download.
Then measure momentum: is growth accelerating?
A level and a growth rate are different findings, and so are a growth rate and its direction. Prices can keep rising while inflation slows. To see direction, compare the most recent three months of year-over-year growth with the three months before them:
Momentum = average of the last 3 months of YoY growth − average of the previous 3 months
Here is an illustrative example with invented numbers:
| Months | YoY growth readings | Average |
|---|---|---|
| Previous three | 2.8%, 2.9%, 3.0% | 2.9% |
| Latest three | 3.1%, 3.2%, 3.3% | 3.2% |
| Momentum | +0.3 percentage points |
Positive momentum means annual growth has strengthened across those windows. A very small change in either direction is better described as flat or inconclusive than forced into a trend.
Combine growth and inflation into four regimes
Many investors organize the economy by whether growth and inflation are each accelerating or decelerating. That gives four combinations, each raising different questions:
| Growth | Inflation | Questions to research |
|---|---|---|
| Up | Down | Is demand strengthening while cost pressure eases? Who keeps the benefit? |
| Up | Up | Is demand outpacing supply? What happens to input costs and borrowing costs? |
| Down | Up | Are margins squeezed by weaker volume and rising costs? |
| Down | Down | Is inflation easing because of supply relief, weaker demand, or both? |
To label a regime, use more than one indicator per side. For growth, compare payrolls and industrial production; for inflation, headline and core CPI. Only assign a direction when the indicators agree. If they disagree, call it mixed and ask which upcoming release could settle it.
Keep the clocks straight
Economic data arrives on different schedules. Daily Treasury yields, weekly jobless claims, monthly CPI and quarterly GDP describe different periods. Put the observation date next to every value. A dashboard that mixes yesterday's yield with a two-month-old production figure is fine, as long as the reader can see the ages.
Also watch for revisions. Payrolls, GDP and many other series are revised after first release, sometimes meaningfully. If a trend depends entirely on one month that later gets revised, it was not much of a trend.
Add rates as the link to markets
Interest rates connect the economy to asset prices and borrowing costs. Track the two-year and ten-year Treasury yields and the gap between them. Measure changes in basis points over a defined period, and ask which end moved. A spread can widen because long-term yields rose or because short-term yields fell, and those describe quite different situations.
Mistakes that make trend analysis misleading
- Comparing one month with the previous month in raw, unadjusted data. Seasonal swings can look like a trend.
- Splicing different series together. CPI and the PCE price index measure inflation differently. Keep them as separate lines.
- Calculating a percentage change on a spread or rate that crosses zero. Report the change in percentage points instead.
- Using annualized and year-over-year rates interchangeably. A quarterly annualized rate exaggerates short-term moves; label it clearly and do not compare it directly with a year-over-year figure.
- Letting one release define the story. Wait for confirmation from a second indicator or a second month.
Turn the dashboard into a story
A good trend analysis ends with a short, careful paragraph: what changed, by how much, over what period, which indicators agree, which disagree, and which release comes next. It avoids inventing precision and it says what would change the description. That is also how you build an investment thesis that can be tested rather than merely believed.
Our ebook How to Research Like a Wall Street Analyst walks through building this growth and inflation dashboard step by step, along with an inflation scenario worksheet and a rates and credit module.
Questions readers ask
What is the difference between a level and a growth rate?
A level is the value itself, such as a price index of 120. A growth rate is the change over a period, such as 3% from a year earlier. Prices can still be rising, a higher level, while inflation, the growth rate, slows.
Why compare with the same month last year?
Year-over-year comparisons remove most seasonal patterns and reduce month-to-month noise, which makes the underlying direction easier to see.
Can this dashboard predict recessions or market moves?
Not on its own. It describes what the data has done. Using it to forecast would require testing the method on historical data and measuring how often it was right.
Educational material from The Company Chronicle, not individualized investment advice. Investing involves risk, including loss of principal. Examples are illustrative.