$1,000 in 2006: 20 years of market in one ETF race
Educational content. A historical experiment on past data, not a recommendation or a forecast. Past performance is not a guarantee of future results.
The experiment, in one line
January 2006: $1,000 on the same day into 13 ETFs — 8 US sectors (Tech, Healthcare, Energy, Industrials, Consumer, Real Estate, Telecom, Utilities), Europe, Emerging Markets, Gold, 20+ year Treasury, TIPS. Dividends reinvested, no rebalancing, no timing. Just monthly closing prices — 245 data points — turned into an 80-second animated race.
The podium (in nominal terms)
| ETF | Final value | CAGR | |
|---|---|---|---|
| 🥇 | Technology (US) | $21,544 | 16.27% |
| 🥈 | Gold | $7,856 | 10.65% |
| 🥉 | Industrials (US) | $6,935 | 9.98% |
| 14th | TIPS | $1,960 | 3.36% |
| 15th | 20y+ Treasury (TLT) | $1,741 | 2.76% |
Three things the race shows better than any table:
- Tech's lead isn't linear: it stabilizes around 2010 and becomes uncatchable. Staying anchored to classical "defensive" sectors meant giving up a 3-4× multiple.
- Gold beats 11 of 12 equity ETFs. The "dead asset" label collapses against $7,856 from $1,000.
- The "safe haven" is the slowest boat: long Treasuries and TIPS barely more than doubled in 20 years — in nominal terms.
The twist: the same numbers, adjusted for inflation
US inflation from 2006 to 2026 was +66.8% (a 2006 dollar is worth about 60 cents today). Deflating, every ETF loses about 40% of its apparent value. The real podium:
| ETF | Real value | Real CAGR | |
|---|---|---|---|
| 🥇 | Technology (US) | $12,668 | 13.30% |
| 🥈 | Gold | $4,413 | 7.58% |
| 🥉 | Industrials (US) | $4,131 | 7.23% |
| … | 20y+ Treasury (TLT) | ~$1,057 | +0.27% |
| … | TIPS | ~$1,177 | +0.81% |
The figure that sticks: TLT and TIPS, the "safe havens", are essentially flat over 20 real years. They didn't protect, they merely kept value. Inflation is the tide that moves all boats — and it must always be measured.
What to take away
- Always measure in real terms. A double-digit nominal return can hide a modest real one: only a positive real CAGR builds wealth.
- "Safe haven" is a nominal label. In real terms, the safe harbor can be nearly flat.
- Dividends matter: this is a total return experiment. Without them, the numbers change.
- Concentration ≠ guarantee: Tech won this 20-year window. Past performance is not a guarantee.
Methodology
Total return (dividends reinvested), no rebalancing, monthly prices, declared 2006→2026 period, nominal and real comparison (US CPI deflator). The same discipline we apply to models: dated data, explicit assumptions, no start-date cherry-picking.
Continue
- Rerun the experiment on your parameters with the ETF Portfolio Simulator.
- Why "real" beats "nominal": how to measure an investment.
- All data races: Data Races playlist.
Educational experiment on historical data. Not advice or a recommendation. Past performance does not guarantee future results.