Historical Inflation Rates by Decade: 1920s to 2020s

I was staring at an old receipt and it basically bullied me

I was standing in a dusty antique shop flipping through a box of old papers and I found this grocery receipt from 1937, and the total was 1.42 and I just… froze. My brain did that thing where it goes, “wait, so was everything basically free?” and then you start doing math on your phone and nothing adds up because you’re mixing modern instincts with old-world price tags.

So I went down the inflation rabbit hole again.

And if you’ve ever tried to compare “then” to “now,” you already know the problem: inflation isn’t one smooth escalator. It’s more like a staircase with a few missing steps and a couple of surprise trapdoors.

This post is me walking you through a sane way to think about historical inflation rates by decade (1920s through 2020s), without pretending every year behaves the same. You’ll end up with a method you can reuse for anything: wages, rent, movie tickets, a house your grandpa bought for “about 9,500,” or whatever.

What “inflation by decade” even means (and why it trips people up)

Inflation is just the change in prices over time, usually tracked by an index like CPI (Consumer Price Index). The thing is, when people say “inflation in the 1970s,” they’re usually compressing ten separate years into one vibe. Sometimes that’s fine! But if you’re trying to do timeline math—like comparing a 1924 salary to a 2024 salary—you want a repeatable approach, not vibes.

So here are the two “decade” concepts people accidentally mix:

• Average annual inflation rate for the decade (a typical year in that decade).
• Total inflation over the decade (how much the price level changed from the start to the end).

Those are not the same. Not even close, sometimes.

And there’s another sneaky bit: inflation is usually reported year-over-year, but your historical question might be “what did 1.42 in 1937 feel like?” That’s a purchasing power question, which is basically the inverse of inflation math. I nodded like I understood that at first. I didn’t.

If you want to play with the math quickly, here are a few tools I keep open in other tabs:

A decade-by-decade cheat sheet (not gospel, just a map)

I’m not going to pretend a single table can “explain” a century of inflation. But a map helps. Below is a quick decade guide with the kind of headline events that tend to show up alongside inflation shifts. It’s not a dataset, it’s a memory aid (and honestly, it’s the part I wish more history textbooks did).

So yeah, that table is intentionally squishy. If you need exact rates, you’ll want a specific source series (like CPI for a given country) and then you do the math from the index values. Which brings us to the part that actually matters: how to calculate decade inflation cleanly.

The math that keeps you honest (worked like a timeline, not a textbook)

If you only remember one idea: inflation comparisons are compounding. You don’t add yearly rates like they’re inches on a tape measure. You multiply the growth factors.

And you can do it two ways:

• Use the index values (cleanest if you have them).
• Use yearly inflation rates (fine if that’s what you’ve got), compounding them.

That formula is the “receipt translator.” If you know the index in 1937 and the index in 2024, you can scale the receipt total. If you only know decade endpoints, you can still estimate decade change.

But people also ask for an average annual inflation rate for the decade, because it feels more intuitive (like “about 4% a year”). That’s a different calculation: you’re basically backing out the constant annual rate that would produce the same total change.

Now let me do a worked example with easy numbers, because that’s where it clicks.

Example: Say a price index is 20.0 in 1920 and 24.0 in 1929 (I’m using round numbers on purpose, not claiming these are the actual values).

1. Total decade inflation = (24.0 ÷ 20.0) − 1 = 0.20 → 20% over the decade
2. Average annual rate = (24.0 ÷ 20.0)^(1 ÷ 10) − 1
3. That’s (1.2)^(0.1) − 1 ≈ 0.0183 → about 1.8% per year

See the difference? “20% in the 1920s” sounds big, but “about 1.8% per year” sounds mild. Both can be true. That’s why you have to say which one you mean.

And if you’re comparing something like wages, don’t forget you’re really asking two questions at once:

• Did the number go up?
• Did the buying power go up after inflation?

That second one is where people get mad at each other on the internet.

If you want to sanity-check your math without building a spreadsheet, I’ll often do it like this: compute the total percent change with

But. One more thing.

Decades don’t start and stop neatly in real life. If you’re studying the Great Depression, “1930–1939” might be less meaningful than “1929–1933” or “1933–1937,” depending on what you’re trying to explain. That’s why I like timeline math tools like

That’s a lot of nuance!

How I’d actually use this for 1920s to 2020s (a practical workflow)

Here’s the workflow I use when I’m trying to tell a “numbers + history” story without making stuff up.

1) Pick the exact years.
If you say “1920s,” decide if you mean 1920–1929, 1921–1930, or “the vibe between WWI and the crash.” You’d be shocked how often two people argue while using different endpoints.

2) Grab an index series you trust.
For US readers, that’s often CPI-U. For other countries, it might be a national statistical office series. The key is consistency: don’t mix sources midstream unless you know how they’re constructed.

3) Compute decade totals from index endpoints.
Use the endpoint ratio method. It’s clean and it avoids the “adding rates” mistake.

4) If you need a single decade number, annualize it.
That’s the CAGR formula above. Then, if you’re summarizing multiple decades, you can average those decade annual rates with something like

5) Tie it back to events, gently.
This is where the history part matters. The 1930s can include deflation, and the 1970s are famous for inflation spikes, and the early 2020s had a very specific kind of disruption. You don’t have to claim “Event X caused exactly Y%.” You can just say, “this is the era where the numbers start acting weird.” People get it.

And if your goal is simply to translate an old price into a modern equivalent, skip the decade summaries and just do year-to-year with an

FAQ