Strategy

How many ETFs do you actually need?

Somewhere between "just buy VWRL" and the guy on Reddit with fourteen ETFs lies your portfolio. The math mostly sides with the first guy.

This is one of those questions where the honest answer disappoints both camps. The minimalists are right that most multi-ETF portfolios are elaborate ways of owning the same thing twice. The collectors are right that there are defensible reasons to hold more than one fund. The trick is knowing which reasons are yours, and the uncomfortable part is that most people have never checked.

What one ETF already does

A single world ETF holds 3,000+ companies across every developed market and sector. That's the diversification job, done. Everything you add after that is either a tilt (you think tech/small-cap/Europe will outperform) or a duplicate (you own the same Apple through three wrappers).

It is worth sitting with how strange this is. A century ago, owning a sliver of thousands of businesses across continents would have required a family office and a staff. Today it is one ticker and one order, at a fee that rounds toward nothing. The problem the ETF industry originally solved, access to broad, cheap diversification, is now so thoroughly solved that the industry has moved on to manufacturing new problems: thematic funds, factor funds, leveraged funds, funds for whatever appeared in the news last quarter. None of that changes the baseline. The diversification job is done at fund number one. Every fund after that needs its own justification.

How portfolios accumulate anyway

Almost nobody designs a fourteen-ETF portfolio. They accrete one. A world ETF to start, sensibly. Then an S&P 500 fund, because US returns looked better. Then a Nasdaq fund, because tech looked better still. A dividend ETF after reading about passive income. A clean-energy fund from 2021 that shall not be discussed. Each purchase made local sense; the sum has no design at all. The tell is that removals never happen: additions have a story, but nobody revisits fund seven to ask whether it still has one. Portfolios, like garages, fill up because things go in and nothing comes out.

The overlap problem

The classic accidental portfolio: a world ETF + an S&P 500 ETF + a Nasdaq ETF. Feels diversified; is actually one bet on US mega-tech, held three times with three expense ratios. Check the top-10 holdings of each fund, if you see the same names, you own concentration wearing a diversification costume.

The mechanics of the trap: a world index is already heavily weighted toward US large caps, because that is where global market value sits. An S&P 500 fund adds more of exactly those companies. A Nasdaq fund adds the biggest of them a third time. The result is that your effective exposure to a handful of mega-cap names is far larger than any single line in your portfolio suggests, and your three "different" funds will rise and fall nearly in lockstep. In good years for US tech this feels like genius. The point of diversification is what happens in the other years, and in those, the costume comes off. Whether the S&P 500 or the Nasdaq is even the right yardstick for that bet is its own question.

Legitimate reasons for two to four

A core-satellite structure has real logic: a world ETF as the boring core, plus one or two deliberate tilts you can name out loud ("I believe semiconductors outgrow the market this decade"). If you can't finish the sentence "I hold this because…" with something falsifiable, it's clutter.

Income investors sometimes split accumulating and distributing funds, fine, that's plumbing, not strategy.

Other honorable exceptions exist. Some investors combine a developed-markets fund with a separate emerging-markets fund to control that ratio themselves rather than accept the index's. Some add a bond ETF as ballast, which is an asset-allocation decision, not an equity-diversification one. Some hold near-duplicate funds across accounts for tax or platform reasons. What unites the legitimate cases is that each fund answers a question no other fund in the portfolio answers. What unites the illegitimate ones is that the honest completion of "I hold this because…" is "it was going up when I bought it."

The costs nobody itemizes

Extra funds are not free even when their expense ratios are low. Every additional position is another thing to rebalance, another decision about where new money goes each month, another line to check at tax time, and another temptation to tinker. Complexity also has a stealth cost: it makes your portfolio harder to evaluate, which makes underperformance easier to not notice. A two-fund portfolio that lags its benchmark gets caught quickly. A fourteen-fund portfolio can hide a laggard for a decade, because no single line looks bad enough to investigate and the blended result is too muddled to attribute. There is a reason most investors trail the market, and unexamined complexity is one of its quieter mechanisms.

The test

Group your holdings by actual role, core, tilt, income, bets, and check the performance of each group separately. Most people discover one group does all the work and another has quietly trailed for years.

Run it honestly and the results tend to be clarifying. The boring core has usually done the heavy lifting. The tilts are a mixed bag, some earned their thesis, some are three years past the point where the thesis failed. The "bets" group is educational in the way tuition is. None of this is a reason for self-flagellation; it is simply information you cannot act on until you have it. The follow-up question, whether each group beat what a plain index fund would have done with the same money, is the whole discipline of benchmarking in miniature.

Doing the test requires nothing exotic, just your transaction history and a willingness to be graded. Assign every holding to exactly one role, no fund gets to be "sort of core, sort of tilt", and be strict about it: if two funds claim the same role, one of them is probably redundant. Then look at each group's return over a period long enough to mean something, a few years rather than a few months, and compare it against the benchmark it was supposed to beat. Repeat annually. The point is not to churn the portfolio every year but to make sure every fund keeps facing the question it was bought to answer.

That grouping is exactly what sub-portfolios in BullBenchmark are for: label each holding, see per-group returns against the benchmark, and find out which of your fourteen ETFs earned their spot.

So, how many?

For most people, one world ETF is a complete answer, and two to four funds covers everyone with a deliberate reason: a core plus a named tilt or two, or an accumulation/distribution split, or a self-managed developed/emerging ratio. Beyond that, the burden of proof flips. The question stops being "why would I sell this fund" and becomes "what does this fund do that the others don't", and any fund without an answer is a candidate for consolidation. Your future self, the one doing the year-end review and the tax return, will thank you for every line that isn't there.

The first week of BullBenchmark is free, no card, upload the export you just downloaded and see where you stand.