History Says Buying The Nasdaq-100 At The Dot-com Peak St…

An 83% loss, then a 15-year wait

The Nasdaq-100’s top came late in the bubble: its closing high on March 27 arrived more than two weeks after the broader Nasdaq Composite (^IXIC +0.39%) had already peaked on March 10, 2000.

From there, the fund fell in stages for about two and a half years. It bottomed on Oct. 9, 2002, closing at $20.06 — an 83% decline from the peak. At the low, the $10,000 stake was worth about $1,700.

The climb back took much longer than the drop.

With dividends reinvested, the investment first returned to even in February 2015, nearly 15 years after the purchase. And even that didn’t last. The fund slipped back below the starting value repeatedly over the ing 16 months before moving above it for good in mid-2016.

However, the price itself took longer still. The fund didn’t close above $117.75 again until September 2016. Reinvested dividends, small as they were, brought the recovery forward by about a year and a half.

The fund recovered before its biggest stocks did

What eventually rescued the March 2000 buyer was not the era’s leadership coming back.

Cisco Systems (CSCO -0.66%) closed at $80.06 the same day the index peaked, and it didn’t close above that price again until December 2025 — more than 25 years later. Intel (INTC -0.18%) set its 2000 closing high that August, and the level stood until this past April. Microsoft recovered faster, and it still took until 2016.

Put another way, the fund’s price was back above its previous peak almost a decade before Cisco and Intel got back to theirs.

The reason, I’d argue, is the index’s design. The Nasdaq-100 holds the 100 largest non-financial companies listed on the Nasdaq, and the biggest companies have the biggest weights. As leadership shifted, so did the fund. The investor who bought at the top of one era eventually got paid by the winners of the next one.

Zoom out, and even from the worst entry, the long-term pace was respectable. The total return from that March 2000 close works out to about 7.8% annualized over 26 and a half years. That pace is arguably solid, considering the starting point. But it’s nothing the returns investors associate with the Nasdaq-100 today.

The bigger risk was time

It’s worth noting that a broad S&P 500 (^GSPC +0.17%) index fund did a little better from the same starting point. The SPDR S&P 500 ETF Trust (SPY +0.13%), bought at that same March 27 close with dividends reinvested, is worth about 8 times the original stake today, compared to 7.2 times for the growth index.

Paying a bubble-era valuation for a concentrated fund meant a quarter century in which the broader, cheaper market did just as well.

As of this writing, the fund trades around $717, within about 4% of its 52-week high, and nearly half of its assets are in its 10 biggest holdings, led by Nvidia, Apple, and Microsoft.

NASDAQ: QQQ

Invesco QQQ Trust

Today’s Change

(0.63%) $4.53

Current Price

$721.45

Key Data Points

AUM

$485B

Dividend Yield

0.42%

Expense Ratio

0.18%

Top Holdings

NVDA

8.47%

AAPL

7.90%

MSFT

5.90%

That does not mean another 2000 is coming. But the shape is familiar — a concentrated growth index near a high, with artificial intelligence (AI) in the role the internet played then.

Judging by the worst case on record, an entry that didn’t threaten ruin. The buyer who caught the exact top, reinvested every dividend, and never sold still ended up with a multiple of the original money. The risk was time. It took about 15 years just to get back to even.

In the end, I’d only put money into a fund this when it can remain there for a decade or more, and I’d set expectations closer to that 7.8% number than to the recent past. History treated even the worst-timed buyer reasonably well, so long as the money could wait. It had nothing to offer anyone who needed it back earlier.