Your C Fund is a tech bet: diversifying with the S and I Funds

1. What you actually own in the C Fund

The C Fund holds the 500 or so companies in the Standard & Poor’s 500 index, weighted by float-adjusted market capitalization. That last phrase is doing all the work. A company’s share of your C Fund dollar is its market value divided by the market value of all 500. When a handful of companies become enormous, they become a handful of your portfolio, automatically, with no decision on your part.

Here is where that stood on December 31, 2025, from the TSP’s own fund information:

• Information technology was 34.4% of the index. The next largest sectors were financials at 13.4%, communication services at 10.6%, and consumer discretionary at 10.4%. Health care was 9.6%, industrials 8.2%. Energy, materials, utilities, and real estate together were under 9%.
• The largest 100 companies were about 75% of the index’s value. The other 400 shared the remaining quarter.
• The largest company was worth about $4.5 trillion; the smallest about $5.5 billion. The biggest holding is roughly 800 times the size of the smallest.
• The top ten holdings were Nvidia, Apple, Microsoft, Amazon, Alphabet (two share classes), Broadcom, Meta, Tesla, and Berkshire Hathaway. By S&P Dow Jones Indices’ figures, those ten were roughly 40% of the index at year-end, up from about 19% a decade earlier, and Nvidia alone was near 8%.

Communication services includes Alphabet and Meta, and consumer discretionary includes Amazon and Tesla. Add those to information technology and the companies most investors would call “tech” are well over 45% of the index. Eight of the top ten are, in one way or another, bets on the same thing: that spending on computing, cloud, chips, and artificial intelligence keeps growing. If you hold only the C Fund, that is your bet too.

2. Sector by sector: C, S, and I side by side

The clearest way to see what the other two stock funds add is to line up their sector weights against the C Fund’s. All figures are the TSP’s, as of December 31, 2025.

3. Why concentration matters

Diversification works because things that are not the same do not fall at the same time. A portfolio of 500 companies in 125 industries is diversified in the sense that matters only if its returns actually come from many places. When 40% of the value and most of the recent return come from ten companies with a shared driver, the effective number of bets you hold is far smaller than 500. Three consequences follow.

Single-company shocks reach you. With one company near 8% of the index, a bad quarter at that company moves your whole C Fund balance. In 1990, when the ten largest S&P 500 companies were about 19% of the index and spanned oil, tobacco, conglomerates, and computers, no single earnings miss could do that.

Correlated leadership means correlated drawdowns. The last time the top ten approached this weight was 2000, at roughly 23% to 27%, led by Cisco, Microsoft, Intel, and other technology names. When that theme broke, the S&P 500 fell for three consecutive years and did not regain its March 2000 level, on a price basis, until 2007. An investor who bought the S&P 500 at the start of 2000 had a negative total return for the following decade. Over that same decade, small-cap and international stocks were positive. The point is not that this will repeat; it is that when it happened, the funds the TSP now offers as S and I were the ones that cushioned it.

Passive flows reinforce the pattern. Every dollar that goes into an S&P 500 fund sends about 40 cents to the same ten companies, regardless of price. That is a feedback loop on the way up and, potentially, on the way down.

For a federal employee, one more factor: sequence risk. A 35-year-old can ride out a lost decade in the C Fund. A 62-year-old about to start installment payments cannot. The nearer you are to drawing on the account, the more the shape of the stock allocation matters, not just the size of it. That is the subject of section 10.

4. Why it is not automatically a problem

Honesty requires the other side of the argument, and it is a strong one.

Cap weighting is not a bug. Market-cap weighting owns companies in proportion to how much the market thinks they are worth. It lets winners run, it never has to guess which companies will lead, and it turns over very little. Decades of academic work find that most attempts to improve on it, including equal weighting, fundamental weighting, and sector caps, do not beat it after costs over long periods. Concentration in a cap-weighted index is what a period of extraordinary success at a few companies looks like. It has also tended to be self-correcting: leaders change, weights normalize, and the index carries on.

The C Fund has earned its position. Over the ten years to December 2025 it returned 14.79% a year against 11.04% for the S Fund and 8.70% for the I Fund. A participant who diversified away from it in 2016 has less money today than one who did not. Any argument for diversifying now is an argument about risk, not a claim that the C Fund is about to lose.

Those ten companies are real businesses. Unlike much of the 2000 leadership, today’s largest companies are extraordinarily profitable. Their weight reflects earnings as well as expectation. That does not make them immune to a reset, but it makes a direct comparison to 2000 less than exact.

So the case is not “sell the C Fund.” It is this: if you are 100% C, you are making a concentrated bet, and you should be making it on purpose. Most people in that position did not decide to be there; they got there by leaving a contribution election alone for twenty years while the index changed underneath them. Deciding is the whole exercise.

5. What the S Fund adds

The S Fund tracks the Dow Jones U.S. Completion Total Stock Market Index: every actively traded U.S. common stock not in the S&P 500. As of December 31, 2025 that was 3,375 companies representing about 12% of U.S. market value. The C and S Funds together are, by design, the whole U.S. market. Because the S Fund holds nothing in the S&P 500, it holds none of the ten companies that dominate the C Fund. That is precisely what makes it a diversifier rather than a duplicate.

Its character is different in three ways. Its largest sector is industrials at 19.1%, followed by information technology at 18.3%, financials at 16.3%, and health care at 14.2%. Its concentration is far lower: the largest 100 holdings are about 29% of the fund, not 75%. And its holdings are mid-sized and small companies whose fortunes are tied more to the domestic economy and less to global technology spending. Historically that has meant higher volatility than the C Fund, sometimes sharply so, and long stretches of both outperformance and underperformance. Its 2025 return was 11.38%; its ten-year return 11.04%; since inception in 2001, 9.42% a year against the C Fund’s 11.34% since 1988.

Cost is slightly higher: investment expenses of 0.017% versus 0.001% for the C Fund, because sampling several thousand small stocks costs more than holding 500 large ones. Both are trivial. The S Fund’s job in a portfolio is not to beat the C Fund. It is to own the 12% of America the C Fund leaves out, and to have a different worst year.

6. What the I Fund adds

The I Fund changed materially in August 2024. Before that it tracked the MSCI EAFE index: about 800 large companies in 21 developed markets, no small caps, no emerging markets, no Canada. Since then it tracks the MSCI ACWI IMI ex USA ex China ex Hong Kong Index: more than 5,100 companies of all sizes across 44 developed and emerging markets, covering roughly 99% of the investable stock market outside the United States, China, and Hong Kong. If your mental model of the I Fund is “large European and Japanese companies,” it is out of date.

What it now owns, as of December 31, 2025:

• Countries: Japan 16.1%, United Kingdom 9.8%, Canada 9.2%, Taiwan 6.8%, France 6.5%, Germany 6.0%, and 38 others making up 41%, including India, South Korea, Australia, Switzerland, and Brazil.
• Sectors: financials 23.9%, industrials 16.3%, information technology 14.9%, consumer discretionary 8.7%, materials 8.1%, health care 8.0%. Technology is less than half its weight in the C Fund.
• Largest holdings: Taiwan Semiconductor, ASML, Samsung, Roche, AstraZeneca, HSBC, Novartis, Nestlé, SAP, and SK Hynix. Several are technology companies, but they are the suppliers and competitors of the U.S. giants, not the same names.

The I Fund also carries currency exposure, which cuts both ways and is part of the diversification. In 2025 the index returned 23.57% in local currencies but 32.01% in U.S. dollars, because the dollar weakened; the I Fund’s 32.45% made it the best TSP fund of the year. In years when the dollar strengthens, the reverse happens. Over the ten years to 2025 the I Fund returned 8.70% a year, well below the C Fund, which is the cost of the insurance. Investment expenses are 0.015%.

What the three funds did, year by year

Diversification shows up as dispersion: in any given year the three funds do very different things, and the leader rotates. The TSP’s trailing figures to December 31, 2025 make the point without commentary.

Over ten years the C Fund won, and by a wide margin. In 2025 it finished a distant second to a fund many participants had written off. Neither fact predicts 2026. What the spread column shows is that owning all three smooths the ride: a portfolio split on the L Fund pattern returned roughly 22% in 2025 and roughly 12% a year over the decade, never the best fund and never the worst, which is what diversification is supposed to feel like. For a retiree, “never the worst” is the property that matters.

7. How the L Funds already split it

You do not have to invent an allocation from scratch. The TSP’s own investment professionals publish one every quarter in the form of the L Funds, and the stock portion is remarkably consistent across them. As of December 31, 2025, the stock sleeve of every L Fund from L 2055 through L 2075 was 51% C, 13% S, 35% I, and the nearer-dated funds held nearly the same proportions with less stock overall. The L Income Fund’s 28% in stocks was split 14 C, 4 S, 10 I: half C, a seventh S, a third I.

In other words, the Board’s view of the right long-run stock mix for a federal employee is: about half in the S&P 500, a third overseas, and the balance in smaller U.S. companies. A participant who is 100% C is not merely more aggressive than the L Funds; she holds a fundamentally different portfolio from the one the plan’s designers consider efficient. That does not make her wrong. It does mean the L Fund split is a reasonable reference point for anyone who wants one.

8. Three allocations and what they do to your exposure

The useful question is not “how much C?” but “how much of my money is in ten companies, and how much is in one sector?” Using the December 2025 sector weights and a 40% top-ten share for the C Fund, here is what four stock sleeves actually deliver. The percentages are of the stock portion only; your G and F allocation sits outside them.

Read across the L Fund row. Moving from all-C to the Board’s split cuts your exposure to the ten largest S&P companies in half, from 40% to 20% of your stock money, and trims information technology from a third to a quarter, while still keeping the S&P 500 as the single largest holding. That is a large change in risk shape for a modest change in expected return, and it is the reason the L split is the default recommendation of this guide for anyone who does not have a considered view of their own.

The equal-thirds sleeve goes further, and it is worth saying plainly why this guide does not recommend it as a default: a 33% S Fund weight is nearly three times the S Fund’s share of the U.S. market, which is a large active bet on small companies, historically the most volatile of the three funds. It is a legitimate choice for someone who wants it. It is not neutral.

Choose the row that matches the bet you are actually willing to make. Then, and this is the part most people skip, write the target down, so that next year’s rebalance is a mechanical act rather than a fresh debate.

9. Rebalancing inside the TSP’s rules

Two TSP mechanics matter here.

The transfer limit. You may make two fund reallocations or fund transfers per calendar month. After the second, for the rest of that month you can only move money into the G Fund. A once-a-year rebalance uses one transaction; the limit is not a constraint for anyone following this guide. It is a constraint for anyone trying to time the market, which is the point of it.

Contribution allocation is separate. Changing where your new contributions go does not count as a transfer and can be done at any time. For an employee still contributing, the gentlest way to move from all-C toward a target is to redirect new money to the S and I Funds while leaving the existing balance alone, then do a single reallocation at year-end to close the remaining gap. There is no tax consequence to any of this inside the TSP; the only cost of rebalancing is the risk that you were right to be concentrated.

A rebalancing rule that works: pick a date (the first business day after your birthday is common), check the actual C/S/I split against the target, and reallocate if any fund is more than five percentage points off. Otherwise do nothing. The L Funds rebalance to target every business day; once a year is plenty for a human. Rebalancing means you will be selling the fund that just did best and buying the one that just did worst, which is uncomfortable every time and is also the entire mechanism by which diversification pays.

10. If you are already retired

For a retiree drawing installments, concentration is not an abstraction. It is the difference between a 20% drawdown in your stock sleeve and a 35% one in the year your withdrawals start. Two adjustments to the framework above.

First, the G Fund bucket comes before the stock mix. Hold the next three to five years of planned withdrawals in G so that no stock fund, however concentrated, is ever a forced sale. Only then decide how the remaining stock money is split.

Second, the case for the S and I Funds is stronger in retirement, not weaker, because the goal shifts from maximizing the expected balance to narrowing the range of outcomes. The L Fund stock split is a sensible default; the L Income Fund itself is an option for anyone who wants the whole thing done for them, though its 72% in G and F is more conservative than most FERS retirees with a pension need. That trade-off is the subject of the L Funds guide for feds with a pension.

The one thing not to do is the thing most people do: stay 100% C because it has always worked, then discover in the first bad year of retirement that “always” had a start date.

11. Frequently asked questions