XLF vs KRE: Which Bank ETF Wins After the Fed’s Hike

The Fed raised rates on September 16, 2026, and the cheapest bank ETF is not the one that won. XLF charges 0.08% against KRE’s 0.35% — $8 versus $35 a year on $10,000 — yet KRE returned 14.79% year to date through August 31 against XLF’s 6.26%. The gap is composition, not skill: XLF holds just 28.41% banks.

Every bank ETF comparison gets framed as a fee fight. This one is not. The three largest ways to own American banks charge wildly different prices and own wildly different things, and the Federal Reserve’s first hike since 2023 has pulled them apart again.

Here is what each one actually costs, what each one actually holds, and which measurable question each one wins.

What did the Fed actually do on September 16, 2026?

The FOMC raised the federal funds target range by 25 basis points to 3.75%–4.00% on September 16, 2026, in a unanimous 12-0 vote. The statement said inflation “remains elevated” and that the action “will support a timelier return to the Committee’s 2 percent goal.”

It was the first increase since 2023. Fed Chair Kevin Warsh told reporters that “the plain fact is that inflation is too high and has been for too long,” per CNBC’s meeting coverage.

The Summary of Economic Projections released the same day puts the median federal funds rate at 4.1% for end-2026 and 4.1% again for end-2027. Median PCE inflation for 2026 sits at 3.7%, core PCE at 3.4%, unemployment at 4.1%.

Markets split. The S&P 500 rose 0.93% on September 17, the Nasdaq finished flat at −0.01%, and the Russell 2000 fell 0.40%, according to TheStreet’s session recap.

Why a hike matters more to banks than to the index

A higher funds rate pulls four levers at once for a lender, and they do not all point the same way:

  • Asset yields reprice fast. Floating-rate commercial loans and reserve balances reset within weeks of a hike.
  • Deposit costs reprice slower — but they do reprice. The lag is where net interest margin is made or lost.
  • Securities portfolios take mark-to-market damage. Higher yields cut the value of bonds banks already own. This is exactly what broke several regional lenders in 2023.
  • Credit gets harder. Higher rates raise the odds that commercial real estate and small-business borrowers stop paying.

Which of those dominates depends entirely on the balance sheet you own. That is why the fund you pick matters more than the sector call.

Which bank ETF has the lowest fee?

XLF, at a 0.08% gross expense ratio, is the cheapest of the three by a factor of more than four. KBWB and KRE both list 0.35%. On fee alone there is no argument — but XLF is also the only one of the three that is not primarily a bank fund, which is the catch.

MetricXLFKBWBKRE
Full nameFinancial Select Sector SPDRInvesco KBW Bank ETFSPDR S&P Regional Banking ETF
Gross expense ratio0.08%0.35%0.35%
Net assets$54.29B (Sep 16, 2026)$6.69B (Sep 17, 2026)$4.13B (Sep 16, 2026)
Holdings7629158
WeightingMarket cap, S&P 500 financialsModified market capModified equal weight
Bank exposure28.41%Banks onlyRegional banks only
30-day SEC yield1.35% (Sep 16, 2026)2.01% div. yield (Sep 17, 2026)2.19% (Sep 16, 2026)
30-day median spread0.02%n/d0.01%
Avg. daily volume12,206,337 shn/d6,521,683 sh
InceptionDec 16, 1998Nov 1, 2011Jun 19, 2006
Fee and spec data from State Street (XLF, KRE) fund pages and stockanalysis.com (KBWB), as of the dates shown.

What a $10,000 position costs per year

Run the arithmetic on a $10,000 position and the annual gap is $27. That is the whole fee argument, in dollars.

  • XLF: $10,000 × 0.08% = $8.00 per year
  • KBWB: $10,000 × 0.35% = $35.00 per year
  • KRE: $10,000 × 0.35% = $35.00 per year

Compound it. At a 7% gross annual return over ten years, XLF nets 6.92% and turns $10,000 into $19,524.93. KRE and KBWB net 6.65% and reach $19,037.44. The fee costs you $487.49 over the decade.

Now put that next to the performance gap below and the fee stops being the headline.

How much does the bid-ask spread cost you?

Almost nothing, and here the cheap fund loses. State Street’s fund page lists KRE’s 30-day median bid-ask spread at 0.01% and XLF’s at 0.02%. On a $10,000 round trip that is roughly $1 for KRE against $2 for XLF.

It is a counterintuitive result: the $4.13 billion fund trades tighter than the $54.29 billion one. KRE’s shares are cheap in absolute terms and heavily used by hedgers, which keeps the book deep.

For a buy-and-hold investor, spreads are rounding error next to the expense ratio. For anyone trading around the Fed, they are not — and neither fund penalizes you meaningfully. The same logic applied when we broke down XLE versus XOP in energy: the structural exposure decided the outcome, not the trading cost.

What do you actually own in XLF versus KRE?

Very different things. XLF’s State Street page shows just 28.41% of the fund in banks as of September 16, 2026. Its two largest positions are Berkshire Hathaway Class B at 12.08% and JPMorgan Chase at 11.85%, followed by Visa at 7.80% and Mastercard at 5.81%.

XLF is a financials fund, not a bank fund

Add Visa and Mastercard together and you get 13.61% of the portfolio in payment networks — businesses whose economics have almost nothing to do with the funds rate. Berkshire alone is a larger weight than either.

If your thesis is “the Fed is hiking, so lenders earn more,” XLF expresses roughly a quarter of that thesis and dilutes the rest with insurance, payments and asset managers.

KRE is 158 regional banks, near equal-weighted

KRE tracks the S&P Regional Banks Select Industry Index across 158 holdings using a modified equal-weight method, which State Street describes as giving “unconcentrated industry exposure across large, mid and small cap stocks.”

Equal weight sounds like diversification. In a single industry it is the opposite: it pushes money down-cap, toward the smaller lenders with the thinnest deposit franchises and the heaviest commercial real estate books.

KBWB sits between them — 29 large-cap banks, modified market-cap weighted, led by JPMorgan at 8.66%, Wells Fargo at 8.51% and Bank of America at 8.48%, per stockanalysis.com.

Which bank ETF has performed better after the hike cycle turned?

KRE, by a wide margin this year, and it is not close. State Street’s data through August 31, 2026 shows KRE up 14.79% year to date on NAV against XLF’s 6.26%. Over one year it is 15.05% versus 8.52%.

Return (NAV, as of Aug 31, 2026)XLFKRE
Year to date6.26%14.79%
1 year8.52%15.05%
3 years annualized20.69%21.55%
5 years annualized10.36%5.07%
Source: State Street fund pages. Returns net of fees, dividends reinvested.

The five-year row is the one to stare at. XLF compounded at 10.36% a year; KRE managed 5.07% — less than half, for 4.4 times the fee.

And the flattering three-year number carries a survivorship flavor. A window starting in September 2023 begins after the March 2023 regional banking failures had already wiped out the losses. Start six months earlier and the picture is uglier.

That is the honest case against KRE: it is a high-beta bet on the exact balance sheets that a hiking cycle stresses hardest. Rising rates help its net interest margin and hurt its bond book and its borrowers, and the market pays that out in both directions. Note that the Russell 2000’s 0.40% decline on September 17 says small-cap risk appetite did not celebrate the hike.

Which bank ETF pays the most income?

KRE, and the gap survives its higher fee. KRE’s 30-day SEC yield was 2.19% as of September 16, 2026 against XLF’s 1.35%. On $10,000 that is $219 of income versus $135 — an $84 difference before costs, and $57 after.

Net of the expense ratio on a $10,000 position: XLF yields $127, KBWB $166, KRE $184. The fee eats a third of KRE’s yield advantage and still leaves it ahead.

Worth the context: none of these clear a Treasury bill with the funds rate at 3.75%–4.00%. An equity dividend yield of 2.19% is not an income substitute for cash right now, a point that also drove our look at TIPS versus nominal Treasuries.

Which bank ETF fits which investor?

Match the fund to the question you are actually asking. If you want cheap, diversified financials exposure, XLF wins on fee and breadth. If you want a pure rate bet, XLF is the wrong tool at any price.

If your goal is…Best fit on measurable attributesWhy
Cheapest broad financials exposureXLF0.08% fee, $54.29B AUM, 76 holdings across banks, payments and insurance
A concentrated bet on lender marginsKBWB or KRE100% banks; XLF is only 28.41% banks
Large-cap bank exposure with less small-cap credit riskKBWB29 large-cap banks, market-cap weighted, $6.69B AUM
Maximum sensitivity to regional bank marginsKRE158 names, modified equal weight, pushes weight down-cap
Highest dividend yield of the threeKRE2.19% SEC yield vs XLF’s 1.35% (Sep 16, 2026)
Lowest total holding cost over a decadeXLF$487.49 less on $10,000 over 10 years at 7% gross
Tightest trading spreadKRE0.01% median 30-day spread vs XLF’s 0.02%
Rankings reflect measurable fund attributes only — fee, yield, spread, concentration. Not a recommendation to buy any security.

Frequently asked questions

Is XLF a bank ETF?

Not really. State Street’s fund page shows 28.41% of XLF in banks as of September 16, 2026. The rest is insurance, payment networks, asset managers and Berkshire Hathaway, which alone is 12.08% of the fund.

Why does KRE cost more than XLF?

KRE charges 0.35% against XLF’s 0.08%. It runs a 158-name modified equal-weight index that requires regular rebalancing across small- and mid-cap names, and it manages $4.13 billion against XLF’s $54.29 billion, so there is far less scale to spread costs across.

Do bank stocks always benefit from Fed rate hikes?

No. Hikes lift asset yields but also raise deposit costs, cut the market value of bonds banks already hold, and stress borrowers. The 2023 regional bank failures happened during a hiking cycle, not a cutting one.

What is the difference between KRE and KBWB?

Size and weighting. KBWB holds 29 large-cap banks on a modified market-cap basis, led by JPMorgan at 8.66%. KRE holds 158 regional banks on a modified equal-weight basis. Both charge 0.35%.

How much is the Fed expected to hike from here?

The September 16, 2026 Summary of Economic Projections shows a median federal funds rate of 4.1% at both end-2026 and end-2027, against the current 3.75%–4.00% range. That implies roughly one more quarter-point move and then a hold.

Which of the three is most liquid?

XLF by volume, at 12,206,337 shares on average versus KRE’s 6,521,683 as of September 16, 2026. KRE nonetheless posts the tighter median spread at 0.01%.

Does a 0.27% fee difference really matter?

It is $27 a year on $10,000 and $487.49 over ten years at a 7% gross return. Real, but smaller than the 8.53-point year-to-date performance gap between XLF and KRE through August 31, 2026. Composition dominates fee here.

The bottom line

On cost, XLF wins outright and it is not debatable: 0.08% versus 0.35%, a $487.49 advantage per $10,000 over a decade. On purity of exposure to a hiking Fed, it loses outright — 28.41% banks is not a bank ETF.

So the choice depends on exactly one thing: whether you want financials or you want lenders. Those are different trades with different drivers, and the labels on these funds obscure that.

The skeptical read on KRE is the five-year record. A 5.07% annualized return over five years, against XLF’s 10.36%, is what the equal-weight structure delivered through a full rate cycle — and the buyer paid 4.4 times the fee for it. This year’s 14.79% is the same structure working in the other direction. Both are the fund doing what it is built to do.

With the median projection pointing to 4.1% and Warsh saying inflation is still too high, the deposit-cost lag is the number that decides this, and it shows up in Q3 earnings, not in a fund fact sheet. For the fixed-income side of the same question, see our breakdown of what 25 basis points costs bond ETF holders, and for the broad-market fee math, SPY vs VOO vs IVV.

This article is journalism, not investment advice. Do your own research before investing.

Sources

Comments

Leave a Reply

Discover more from Wealth Engine

Subscribe now to keep reading and get access to the full archive.

Continue reading