The Math Behind VGT's Compounding Promises — And What It Ignores

If you invest $100 a month in the Vanguard Information Technology ETF (VGT) for 30 years, you could end up with over $2 million. That is the claim you will see repeated across financial websites today. The arithmetic is simple: $100 monthly at a 22% average annual return, compounded for three decades, does produce that number.

The question is not whether the math checks out. The question is whether a 22% annual return for 30 years is a realistic assumption for an ETF that currently offers almost no margin of safety above a Treasury bill and has, in its 20-year history, been knocked down by more than a third at least three separate times.

VGT is not 300 diversified companies. It is three companies wearing a diversified costume.

The concentration behind the diversification

VGT tracks the MSCI Investable Market Information Technology Index and holds roughly 320 technology stocks. It charges 0.09% annually and now manages more than $160 billion in assets. It is one of the largest sector ETFs in the world.

But the fund's index methodology is market-cap weighted, which means the largest holdings dominate everything. As of September 2026, Nvidia accounts for 17.16% of the fund, Apple for 16.26%, and Microsoft for 10.97%. Those three names alone make up over 44% of every dollar in VGT. The top five holdings control more than half the fund. The top 10 control 62%.

Compare that to the S&P 500, where the top 10 holdings account for roughly 38%. You are not buying broad technology exposure. You are buying Nvidia, Apple, and Microsoft, with about 317 smaller positions tacked on as noise.

That matters because the valuations of those three are very different. Nvidia trades at about 27 times trailing earnings, Apple at 38, and Microsoft at 28 — each with a market capitalization above $3 trillion. The fund as a whole trades at roughly 29 times non-GAAP trailing earnings and about 21 times forward earnings. Those are not bargain multiples. They are premium multiples on premium companies.

The return assumption that carries the projection

Here is how the compounding projection actually works. Three scenarios for $100 monthly over 30 years:

• At 11% annualized return — roughly the S&P 500's long-term historical average — the account grows to about $239,000 from $36,000 in total contributions.
• At 16% annualized — VGT's approximate return since inception in 2004 — the account reaches roughly $636,000.
• At 22% annualized — VGT's 10-year average — the account grows to approximately $2.1 million.

The headline number requires you to believe that VGT's best decade will repeat itself for three more decades. The 16% assumption requires you to believe the fund will match its historical average despite having grown from a tiny fund in 2004 into a $160 billion behemoth today, with its largest holdings already worth more than the GDP of most countries.

VGT's annualized return since inception is about 15%. That is the number, not the 22%. The 22% comes from the last 10 years — a period that includes both the AI infrastructure boom and the post-pandemic technology super-cycle. These are not conditions expected to hold in their current form for three decades.

The valuation gate that the projection does not test

Projections backward from past performance treat the current price as irrelevant. But your future return depends entirely on what you pay today and how much those earnings grow from here.

VGT's current total yield — dividend yield of about 2% plus an earnings yield near 3% — sits at roughly 5%. The risk-free rate on short-term Treasuries is about 4.8%. The difference, known as the equity risk premium, is about a quarter of one percentage point. That is essentially zero.

A near-zero risk premium means the market is already pricing in very strong future growth from these technology companies. You are not buying VGT at a discount that can cushion you through a downturn. You are buying it at a price that assumes the growth delivers.

When the growth does deliver — as it has over the past two years, with sector earnings expected to grow roughly 24% this fiscal year — the price holds or advances. When it does not, a portfolio priced at near-zero risk premium has very little room to fall before the math turns ugly.

The drawdown record that the compounding chart hides

VGT's long-term chart is a smooth upward line. That line is drawn by connecting the highs. It does not show that this fund has fallen 35% in the 2022 inflation shock, 32% in the 2020 pandemic crash, and 27% during the 2025 tariff disruption.

These are not minor pullbacks. A 35% drop means you need a 54% gain just to get back to even. If you are contributing $100 a month, a sharp decline at the wrong time — near the end of your contribution period, for instance — permanently reduces the number of shares you accumulate. Dollar-cost averaging does soften the blow, but it does not eliminate it.

The reason VGT drops more than the broader market is precisely the concentration described above. When interest rates rise, high-multiple growth stocks are the first and hardest hit. When earnings disappoint, there are only three or four names doing the heavy lifting, and a stumble from one of them rips through the entire fund.

VGT's 10-year annualized volatility is roughly 21%, compared to about 14% for a moderately diversified portfolio. You are accepting substantially more whiplash for a return that may or may not exceed the broader market by much after the downturns.

What a realistic assessment looks like

VGT is not a bad investment. It is a pure-play bet on U.S. technology companies, delivered through a low-cost index wrapper. If you believe that the sector's earnings growth — currently estimated at 24% for the coming fiscal year, with the AI-driven capex cycle still unfolding — will sustain at a meaningful pace for years to come, VGT is one of the cleanest ways to capture that bet.

But the compounding projections deserve a reality check. The $2 million headline number rests on a 22% annualized return assumption that ignores valuation, concentration, drawdown history, and the simple fact that extraordinary performance tends to mean-revert over time. A more honest projection uses the fund's full-history average of about 15%, or better yet, anchors to the broader market's 10-11% long-term rate and asks whether VGT's concentration and premium valuation justify the additional volatility.

The question for any investor is not whether $100 a month can grow into something large. It can. The question is whether you are comfortable betting that three technology companies, priced at a premium that offers almost no safety margin above risk-free rates, will compound at an extraordinary pace for the next three decades — and whether the 30% drawdowns along the way will test your resolve.

If the answer is yes, VGT is a legitimate vehicle for that bet. If the answer is "I'm not sure," the $2 million projection is not encouragement. It is a reminder that the difference between $239,000 and $2.1 million is not the strategy. It is the assumption you make about the future.