Tracker · The debt and the dollar
Dollar purchasing power
A dollar has lost 97.1% of its purchasing power since January 1913, so it now buys what about 2.9 cents bought then, and it takes $34.07 to match one 1913 dollar. That is the claim every gold guide makes and it is arithmetically correct. The same loss stated as a rate is 3.16% a year over 114 years.
1868 monthly readings since January 1871
One 1913 dollar, in today's money
$34.07- Purchasing power lost 97.1%
- Average inflation a year 3.16%
- Worst decade 1970s, 50.7%
Both framings come from one calculation. That they are the same fact is the argument of this page.
A claim that is true and misleading at once
What the 97% figure actually describes
What is true
Prices are 34.1 times higher than in January 1913. A dollar bill kept since then buys about 2.9 cents of what it once did. Nothing in that arithmetic is wrong and we have not softened it.
What it leaves out
It describes cash held under a mattress for 114 years. Nobody does that. For most of the period a saver could hold Treasury bills at a nominal yield that broadly tracked inflation, so the currency's loss and the saver's loss are different numbers and the claim quietly swaps one for the other.
The honest version of the concern is narrower and it survives. Inflation is a real tax on cash and on long nominal bonds, it compounds without anyone deciding to levy it, and it has had regimes where it ran far ahead of what savers were paid. The 1970s is the clearest case in this record, and it is a far better argument than the century-long figure because it describes something a real person actually lived through with real money.
What a dollar of each year's money is worth in today's dollars
Purchasing power of the dollar since 1913
January 1913 dollar
$34.07Source: Shiller CPI column, reconciled against the Bureau of Labor Statistics CPI-U over 954 overlapping months
Download CSVWhere the damage actually happened
What a dollar lost in each decade
This is the table the guides never print, and it makes the stronger case. The 1970s took 50.7% in ten years. 4 closed decades in this record ended with a dollar worth more than it started, which a single falling line cannot show.
Change in purchasing power across each decade. Below the line is a loss
| Decade | Purchasing power | Average inflation a year |
|---|---|---|
| 1870s | +28.4% | -2.77% |
| 1880s | +28.0% | -2.46% |
| 1890s | -3.6% | +0.37% |
| 1900s | -21.0% | +2.40% |
| 1910s | -47.6% | +6.75% |
| 1920s | +12.2% | -1.15% |
| 1930s | +22.1% | -2.00% |
| 1940s | -41.1% | +5.48% |
| 1950s | -20.1% | +2.29% |
| 1960s | -22.3% | +2.57% |
| 1970s | -50.7% | +7.40% |
| 1980s | -38.3% | +4.99% |
| 1990s | -24.3% | +2.85% |
| 2000s | -21.8% | +2.52% |
| 2010s | -15.7% | +1.73% |
| 2020s (running) | -22.7% | +4.00% |
A positive purchasing power figure means a dollar bought more at the end of the decade than at the start
The comparison this site is asked for
Gold against the dollar since 1971
Measured from August 1971, when the dollar's last formal link to gold ended. Both sides are stated in what they command in real goods, because a rising dollar price of gold while the dollar falls is partly the same fact counted twice.
An ounce in August 1971
$40.65Worth $332.40 in today's money
An ounce today
$4,562.75+1272.7% in real terms
A dollar over the same span
-87.8%Purchasing power since August 1971
Gold has kept ahead of inflation across that window and the window is not neutral. It starts with gold at an administratively fixed price and runs straight into the largest inflation of the modern era, which is the most favourable possible starting point. Measured from January 1980 instead, gold took more than forty years to recover its purchasing power, and that record is on the inflation-adjusted gold price page rather than hidden from this one. Both windows are real and choosing between them is where most of the argument about gold as an inflation hedge actually happens.
If the 97% figure is why you are here
What the dollar's record means for a retirement account
The figure is true and it is an argument against holding cash for decades, not an argument for any particular alternative. A retirement account is not a mattress. What it does support is paying attention to real returns rather than nominal ones, which applies just as much to the metal as to anything else: a gold IRA carries a dealer premium, a custodian fee and a storage fee, and all three come out of the real return before inflation is even considered.
Related
More inflation and dollar tools
Run your own numbers, see what gold did in the same real terms, and check the fiscal side of the argument.
Common questions
Dollar purchasing power questions
- Has the dollar really lost 97% of its value since 1913?
- Yes, as arithmetic. Prices measured by the consumer price index are 34.1 times higher than in January 1913, so a dollar today buys what about 2.9 cents bought then, a loss of 97.1%. That figure is correct and this page computes it rather than repeating it. What it describes is cash held under a mattress for 114 years.
- So why do you say the claim is misleading?
- Because almost nobody holds cash for a century, and for most of that period a saver did not have to. The same loss expressed as a rate is 3.16% a year, and Treasury bills broadly tracked inflation over long stretches. The 97% is a fact about currency presented as a fact about savers, and those are different things. The honest concern is narrower and it survives: inflation is a real tax on cash and on long nominal bonds, and it has had regimes where it ran far ahead of what savers were paid.
- Which decade was worst for a dollar holder?
- The 1970s, when a dollar lost 50.7% of its purchasing power in ten years, an average of 7.40% a year. That concentration is the point. A single flat figure for the whole period hides the only thing a saver can act on, which is that the damage arrives in bursts rather than evenly.
- Has a dollar ever gained value?
- Yes, in 4 of the closed decades in this record, and the claim never mentions it. The 1870s, 1880s, 1920s, 1930s all ended with a dollar buying more than it did at the start. Deflation is uncommon and it is in the record, and a page showing only a falling line would be selecting as hard as the guides it corrects.
- How does gold compare?
- Since August 1971, when the dollar's last formal link to gold ended, an ounce has gone from $40.65 to $4,562.75. Adjusted for inflation that is +1272.7% in what an ounce commands in real goods, while a dollar lost 87.8% over the same span. Gold has kept ahead of inflation across that particular window, which is one window and includes the largest inflation of the modern era near its start.
- Is CPI a reliable measure over a century?
- It is the best available and it is not precise. The basket, the treatment of substitution and the quality adjustments have all changed over 114 years, and each change is defensible and also means the index is not measuring quite the same thing at both ends. We publish the figure because it is the standard and flag the limitation rather than presenting a century-long index as an instrument.
How this page is built
Purchasing power is the ratio of the consumer price index between two dates, nothing more. The deflator is Robert Shiller's CPI column, which reaches 1871 on one consistent basis and is the same series that deflates the CAPE page, so the two cannot quietly disagree about what a real dollar is.
It is checked against the Bureau of Labor Statistics CPI-U on the 954 months the two share. The levels sit on different bases and will never match, so the comparison is on the path rather than the level: indexed to their first shared month, the two never diverge by more than 0.82%. That check is what would notice a rebasing, which would otherwise silently rescale every real figure on this site.
CPI is the best available measure of this and it is not a precise instrument over 114 years. The basket, the treatment of substitution and the quality adjustments have all changed, each change is defensible on its own terms, and together they mean the index is not measuring quite the same thing at both ends of the record. The gold comparison uses the BLS series rather than Shiller's, because it pairs with the gold record and matches the deflator the other gold pages use.
Sources
- Robert Shiller, online data, CPI column
- Bureau of Labor Statistics CPI-U, via FRED
- LBMA precious metal prices
Full dataset available as CSV, with both framings and the index level behind them.
Gold IRA Digest is an independent publisher. This page is information, not investment advice.