Best Infrastructure ETFs

Infrastructure investing has drawn attention because of three overlapping tailwinds. Reshoring is pulling manufacturing back to the United States, which requires new factories, roads, ports, and logistics. The electric grid is aging and needs large-scale upgrades to handle electrification, renewables, and rising demand. And the data centers behind the AI buildout consume enormous amounts of power and construction, straining both the grid and the companies that build it. Together these support demand for the firms that pour concrete and lay cable, and for the utilities and transport assets that deliver power and goods.

That is a long-term thesis, not a certainty. Infrastructure spending is cyclical and sensitive to interest rates, government budgets, and policy, so the buildout funds in particular can fall hard in a slowdown even when the multi-year story is intact. The global-income funds are steadier but carry their own interest-rate sensitivity, since utilities and toll-road operators often use debt. Anyone holding infrastructure should size it with that in mind. None of this is a prediction about where these funds go next.

The most direct way to bet on the US construction cycle is through the buildout funds, which hold companies that make and supply infrastructure rather than companies that own it. These funds lean heavily into industrials and materials: engineering and construction firms, machinery and equipment makers, and producers of steel, aggregates, and cement. Because those businesses are cyclical, the funds behave like a growth-oriented play on the economy and reshoring, and they can move sharply in both directions.

PAVE (Global X US Infrastructure Development) is the largest and most popular US infrastructure ETF, holding a basket of industrials and materials companies tied to domestic construction at an expense ratio of around 0.47%. It is a fairly concentrated buildout tilt. IFRA (iShares US Infrastructure) is broader and cheaper at roughly 0.30%: it spreads across companies that could benefit from US infrastructure activity, including a meaningful slice of utilities alongside equipment and materials names, and it is often equal-weighted in style. Both pay relatively small dividends, because the appeal is capital appreciation from the construction cycle rather than income.

The other kind of infrastructure fund holds the assets themselves. Instead of the companies that build, these ETFs hold global owners and operators of infrastructure: regulated utilities, toll roads, airports, ports, pipelines, and communication towers. Those businesses tend to generate steady, often regulated-style cash flows, so these funds usually pay more income and behave more defensively than the buildout funds. They are frequently used as an income-oriented, diversifying slice rather than a cyclical bet.

IGF (iShares Global Infrastructure) is the largest of the group, holding developed-market utilities, transport, and energy infrastructure at an expense ratio of around 0.41%. NFRA (FlexShares Global Infrastructure) casts a somewhat wider net across developed-market infrastructure at around 0.47%, and GII (SPDR Global Infrastructure) tracks a global infrastructure index of utilities, transport, and energy names at around 0.40%. All three lean toward utilities and toll-road-style assets, so they carry more dividend income and less cyclicality than PAVE or IFRA, but they can be sensitive to interest rates because these asset owners often carry debt.

The single most important thing to understand about infrastructure ETFs is that the label covers two very different exposures. A US-buildout fund like PAVE or IFRA is essentially an industrials-and-materials bet on who constructs infrastructure, so it is more cyclical, more growth-oriented, and pays less income. A global-infrastructure fund like IGF, NFRA, or GII is a utilities-and-toll-roads bet on who owns the finished assets, so it is more defensive, more income-oriented, and less tied to the construction cycle.

That means they can perform quite differently in the same market. When the reshoring and data-center buildout is running hot, the industrials-heavy funds tend to lead; when rates fall or investors want defensive income, the utilities-heavy funds tend to hold up better. Neither is better in the abstract; they express different views. Reading what a fund actually holds matters more than the word infrastructure in its name. This is descriptive, not a recommendation.

Costs are approximate expense ratios as of mid-2026; verify the current figure on each issuer's site. The first two hold US industrials and materials companies and behave like cyclical buildout plays, with IFRA the broader and cheaper option and PAVE the more concentrated one. The last three hold global infrastructure assets like utilities, toll roads, and pipelines, so they tend to pay more income and move less than the buildout funds. For the broader angle, you can also explore the infrastructure theme, or read our companion guide on the best infrastructure stocks.

The hard part of infrastructure is not picking the fund; among the US-buildout funds, PAVE and IFRA express a similar theme with different breadth, and among the global funds, IGF, NFRA, and GII overlap heavily, so cost, holdings, and yield are reasonable tie-breakers. The harder question is which version fits your portfolio at all: do you want cyclical buildout growth or defensive infrastructure income, how large a slice makes sense, and how much it overlaps with what you already own. That depends on your goals and existing holdings, which is where an AI assistant that can reason over your real portfolio helps.

That is where Walnut fits. It connects your existing brokerage so you can ask, in plain language through Claude, ChatGPT, or a built-in assistant, how an infrastructure ETF would fit what you already hold, how much a position like PAVE or IGF moves with the rest of your portfolio, and how the buildout funds versus the income funds are doing against the market. Walnut keeps your accounts read-only, so an infrastructure position is only ever added when you place that order. As something that informs rather than advises, it sizes the question of an infrastructure sleeve against your real holdings instead of recommending one, because Walnut is not an investment adviser.

Infrastructure ETFs split into two jobs. For the US construction cycle, the buildout funds PAVE (most popular, around 0.47%) and IFRA (broader and cheaper, around 0.30%) hold industrials and materials companies that move with the economy and pay little income. For global infrastructure income, IGF (around 0.41%), NFRA (around 0.47%), and GII (around 0.40%) hold utilities, toll roads, and pipelines that generate steadier cash flows and pay more dividends.

Whichever route, the honest framing is the same: reshoring, the grid, and data centers are a real multi-year tailwind, but infrastructure spending is cyclical and rate-sensitive, and the two fund types can behave very differently, so read what each one holds. Holdings, fees, and yields change; treat the specifics here as a starting point and confirm on each provider's site before deciding. For the full category map, see our best ETF in every category guide.