ETF flows decoded: where retail money actually went in April

The April ETF creation/redemption data, published in dribbles by the issuers over the last two weeks and aggregated by ETF.com and the ICI factbook update, is more interesting than any single number coming out of macro this month. The headline — $42.1bn of net inflows across all US-listed ETFs — is a healthy print but not a remarkable one. The composition is where the story is.

The four buckets

I find it more useful to throw the 3,000-plus US-listed ETFs into four functional buckets than to look at category averages. The four:

1. Beta sleeves: broad index trackers (VOO, IVV, SPY, VTI, QQQ).
2. Factor sleeves: smart-beta, dividend, momentum, low-vol (SCHD, VYM, MTUM, USMV).
3. Thematic sleeves: AI, robotics, defence, uranium, single-country EM (IGV, BOTZ, ITA, URA, EWZ).
4. Income sleeves: bond ETFs, covered-call, MLP funds (AGG, BND, JEPI, JEPQ).

In April:

Bucket	Apr 2026 Net Flow	YTD 2026	12m Trailing
Beta	+$28.4bn	+$112bn	+$418bn
Factor	-$1.2bn	-$4.1bn	-$9.8bn
Thematic	+$6.7bn	+$22bn	+$71bn
Income	+$8.2bn	+$31bn	+$94bn

Three things jump out. Beta sleeves continue to dominate at 67% of net flow — the indexation tide continues to come in, year after year, regardless of the regime. Factor sleeves have now seen net outflows in seven of the last eight months — the smart-beta thesis has gone quiet, and the largest factor product (SCHD) has bled $3.4bn YTD even as dividends as a category outperformed. And income sleeves continue to grow share, driven almost entirely by covered-call income products (JEPI, JEPQ, SPYI) rather than traditional aggregate bond products.

Inside the thematic bucket

The thematic bucket is where the social-media tape and the actual flow data diverge most sharply. The narrative this month was that retail was capitulating on the AI trade. The data says something more nuanced.

Thematic ETF	Apr Flow	Status
IGV (software)	+$1.4bn	Strong inflow
SOXX (semis)	+$1.1bn	Net inflow
SMH (semis)	+$0.6bn	Modest inflow
BOTZ (robotics)	+$0.3bn	Net inflow
ITA (defence)	+$2.2bn	Largest single thematic inflow of the month
URA (uranium)	+$0.8bn	Continued accumulation
ARKK (innovation)	-$0.4bn	Outflow
ARKQ (autonomy)	-$0.2bn	Outflow

The pattern: money continues to flow into the broad-index AI sleeves (IGV, SOXX, SMH). It is flowing out of the long-duration, high-multiple, narrative-driven AI sleeves (ARKK, ARKQ). And the single biggest beneficiary of April was none of the above — it was defence.

The $2.2bn into ITA in a single month is the largest monthly inflow into the fund since 2017 and reflects a fundamental shift in how retail is thinking about the medium-term capex cycle. The European NATO commitment to 2.5% of GDP by 2028, the US continuing resolution that protected the defence top-line, and the ongoing replenishment of munitions stockpiles all show up here. Defence has, quietly, become a thematic sleeve that flows like a tech sleeve.

Inside the income bucket

The most under-covered story in flows for the last 18 months has been the rise of covered-call income products. JEPI (JPMorgan Equity Premium Income) and JEPQ (the Nasdaq variant) between them now manage $76bn, up from $44bn at the start of 2024. SPYI, the S&P 500 variant from NEOS, has gone from $400m to $4.2bn in the same period.

Why this matters: these products sell upside in exchange for monthly income. In a market that grinds higher with low realised vol — exactly the regime we have been in — they underperform on a total-return basis but deliver a steady cash yield (11–12% annualised). They have effectively become a retail-distribution vehicle for a strategy that was previously the preserve of insurance balance sheets.

The flow into these products is a leading indicator of two things: first, that retail is increasingly willing to give up upside for income, which is a sentiment signal worth watching; second, that the supply of S&P 500 calls being sold every month is now structurally elevated, which is a real reason why realised vol has stayed low. The vol-sellers have become a structural feature of the tape, not a tactical one.

What flows say about positioning

When you flatten the data and look at the implied portfolio shifts, three positioning signals emerge:

One: retail is not the marginal seller of the index. Beta inflows have been positive every single month since November 2022. This is the single most consistent signal in the data. The idea that "retail capitulation" is the cleanup trade that resolves any drawdown is not supported by recent history — retail has bought every dip from $4,200 to $5,300 on the SPX.

Two: the factor trade is over for now. Seven months of outflows from the smart-beta complex, including bleeding from the highest-quality products. Either the regime has shifted away from where factor models excel, or the marginal allocator has decided that factor exposure should be priced into the broad-index multiple rather than purchased as a separate sleeve. Either way, the active management value proposition that smart-beta represented is in retreat.

Three: the income trade is the structural growth story. Covered-call income, dividend ETFs ex-SCHD, BDC funds, MLP funds — all growing share. This is a flow consequence of two things: the demographic mix of who actually owns ETFs (boomer dominance is increasing), and the fact that 4.5% Treasury yields make income-producing equity wrappers attractive on a relative basis for the first time in 15 years.

The contrarian read

If you want a contrarian signal from this data, the most actionable one is in the factor outflows. Quality and dividend strategies have outperformed cap-weighted indices over the last 12 months on a risk-adjusted basis, yet are seeing persistent outflows. When the flow trends and the relative-return trends diverge for this long, the flow trend usually reverses first. Watch SCHD specifically — the outflow streak there is the most stretched in the complex.

The other contrarian read is the absence of any meaningful EM allocation. Emerging-market ETFs have seen net outflows of $3.6bn YTD even as the dollar has weakened by 2.4% on the DXY and EM earnings have outpaced developed-market earnings by ~400bps. If the dollar weakness extends, the asymmetric flow setup is in EM, not in any of the crowded developed-market sleeves.

What I'm watching in May

1. JEPI / JEPQ aggregate AUM. If it crosses $80bn, the systematic vol-selling supply is large enough to matter for index-level realised vol.
2. ITA. A second consecutive $2bn+ month confirms the defence trade has moved from thematic to structural.
3. SCHD. First positive flow month would mark the end of the smart-beta cleanout.
4. EM. Any month with positive net inflows into EEM/IEMG combined would be the first since November and a signal worth weighting.

The narrative will keep telling you what's interesting. The flows tell you what's happening.

---

Sources: ETF.com flow data, ICI weekly ETF report, issuer-published AP creation files. All figures USD, US-listed products only. Flow data as of 2026-05-22.