The Illusion of Wealth & How to Beat Inflation: A 100-Year Perspective
We tested 30 ideas about how to beat inflation against 100 years of US money supply data, from gold and housing to savings, the S&P 500 and Bitcoin.
Topic
100-Year Analysis: A Century of Assets vs M2
A century is long enough to see past most cycles, which is why we built The Illusion of Wealth on 100 years of data.
Read more
Over that time the US money supply went from $43.7 billion to $22.8 trillion, a 522-fold increase, and the series overview pulls together what that meant for every major asset.
The short version is that very few assets kept pace on price alone. Gold rose 168x and US houses 83x against 503x for M2 in Part I, and only assets that reinvested their income, held for decades, came out clearly ahead.
- 522x US M2 growth since 1926 From $43.7 billion to $22.8 trillion, as set out in the series overview.
- 168x Gold, 1926 to 2026 An M2-adjusted value of 0.33, so about two thirds of its share was lost.
- 11,570x S&P 500 total return since 1928 With every dividend reinvested, about 24 times what M2 did.
Articles on 100-year Analysis
1 article from our research on this topic.
Plain-English definitions
Key terms for 100-year analysis
- M2 money supply
- The Federal Reserve’s measure of cash, current accounts and easily accessible savings in the US. We use it to see how quickly new money is being created.
- M2-adjusted value 1.0 = break-even
- An asset’s growth divided by M2 growth over the same years. Above 1.0 it gained share of the money pool, below 1.0 it lost share.
- The 6.56% rule 6.56% a year
- The average annual growth of US M2 over the last century. Assets compounding faster gained ground, slower ones lost it. More about The 6.56% rule
- Normalised to 1.0
- Setting every asset to the same starting value so growth can be compared on one chart regardless of the original price.
FAQ
Questions about 100-year analysis
Answers drawn from the figures in our research. Educational only, not financial advice.
Why look at 100 years instead of the last ten?
A short window can be dominated by one unusual spell, and gold is the example we kept running into. In the series overview its five-year growth rate came out at about 20% a year, roughly four times its 100-year average of 5.6%, so a ten-year view would have made it look far stronger than it has been over the long run.
That is why we built the series on the St. Louis Fed’s M2 data and Aswath Damodaran’s return series going back to the 1920s, a span that takes in the end of the gold standard in 1971, the 1970s inflation, the dot-com peak, 2008 and COVID.
Does 100 years of history matter for investors today?
For the 6.56% rule, yes, and for choosing between assets, less so. We set the rule on the full century because it is the only stretch in our data that includes the 1940s and the 1970s, when money printing ran at close to 10% a year.
Part V then makes the case that someone under 50 should pay more attention to the last 10 to 20 years, and between 2016 and 2026 Nvidia and Bitcoin gained around 80x after M2 while gold, the S&P 500 and Berkshire Hathaway managed 1.9x to 2.1x.
Sources