Kaleidoscope Thematic Equity Model — June and First Half 2026 Attribution Report

June Review

The Kaleidoscope Thematic Equity Model had a difficult June, declining -3.57% versus SPY down -1.03%, producing -2.40% of active underperformance through June 30. The month was a reversal from May’s strong AI- and momentum-led rally, when the model had materially outperformed SPY as AI, cybersecurity, cloud, semiconductors, blockchain, and clean-energy themes led the market. The May attribution report used a similar structure, emphasizing model performance, ETF-level attribution, sleeve attribution, return/flow trends, market backdrop, model interpretation, and bottom line.

June’s weakness was not caused by the entire technology sleeve. SMH and XBI were strong positive contributors, and ITB also helped. The problem was broader thematic dispersion. Blockchain, Bitcoin-linked exposure, solar, fintech, gold miners, natural resources, autonomous technology, cloud, and China internet all detracted. That combination produced a month where the model still owned some of the right AI infrastructure exposure, but the rest of the thematic complex failed to keep pace with the benchmark.

The broader market backdrop was mixed. Reuters noted that the S&P 500 and Nasdaq posted their strongest quarterly gains since 2020, but both indexes still recorded losses for June as investors worried about high technology valuations and the scale of AI spending. Technology led gains on June 30, and the semiconductor index rallied sharply that day, but the month was more complicated beneath the surface.

Model Performance and Attribution

The month’s attribution was dominated by a split inside the model’s growth exposure. SMH remained the strongest single contributor, confirming that semiconductor-linked AI infrastructure exposure was still being rewarded. XBI and ITB also helped, suggesting investors were willing to rotate into biotech and housing-sensitive exposure even as some speculative technology themes weakened.

The main detractors were BKCH, GBTC, TAN, ARKF, GDX, NANR, ARKQ, SKYY, KWEB, COPX, and FDN. This was a broad thematic unwind: crypto equities, Bitcoin-linked exposure, clean energy, fintech, gold miners, natural resources, cloud, China internet, copper miners, and internet platforms all weighed on relative return.

June Sleeve Attribution

The sleeve attribution shows that June’s underperformance was not simply a “Technology bad” month. Technology Themes were only modestly negative because SMH and CIBR helped offset weakness in AIQ, ARKQ, SKYY, KWEB, and FDN.

The larger drag came from Consumer Themes, where BKCH, GBTC, and ARKF overwhelmed positive attribution from XBI, ITB, ESPO, and PEJ. Commodities were also weak as GDX, NANR, COPX, and URA detracted, while MLPX was the only positive contributor in the sleeve.

First Half 2026 Review

The first half still tells a constructive story. The model returned +16.45% YTD versus SPY up +10.09%, producing +6.43% of active outperformance. Despite June’s setback, the model remains meaningfully ahead of the benchmark for the first half of the year.

The first-half attribution profile shows that the model’s structural growth allocation worked. SMH, AIQ, BKCH, XBI, and CIBR were the biggest positive contributors. Industrial and commodity exposures also helped, with GRID, NANR, PAVE, ARKX, TAN, COPX, URA, and MLPX contributing positively.

The detractors were concentrated in a different part of the thematic universe: GBTC, KWEB, ARKF, FDN, ESPO, GDX, PHO, and SKYY. That mix highlights the central first-half tension: investors rewarded AI infrastructure, semiconductors, cybersecurity, biotechnology, blockchain equities, grid infrastructure, and electrification, but remained skeptical of China internet, fintech, Bitcoin-linked exposure, internet platforms, gaming, gold miners, water, and cloud software.

First Half ETF Attribution

First Half Sleeve Attribution

The first-half result confirms that the model’s technology allocation remained the primary source of alpha. Technology Themes contributed +4.52%, led by semiconductors, AI, cybersecurity, and autonomous technology. That strength was partially offset by KWEB, FDN, and SKYY, which shows that investors were differentiating between AI infrastructure winners and broader internet/cloud exposures.

Industrial Themes contributed +1.48%, helped by GRID, PAVE, ARKX, TAN, and IGF. The sleeve’s positive attribution supports the idea that AI is broadening into physical infrastructure, electrification, grid investment, and industrial capacity.

Commodities Themes contributed +0.83% despite a weak June. NANR, COPX, MLPX, and URA contributed positively YTD, while GDX detracted. Commodities were important earlier in the year, but their June weakness shows that the inflation/geopolitical hedge trade is not currently leading the model.

Consumer Themes were roughly flat, with XBI, BKCH, DRIV, ITB, and PEJ offset by GBTC, ARKF, and ESPO. This sleeve contains some of the most divergent holdings in the model, ranging from biotech and blockchain to fintech, Bitcoin-linked exposure, housing, travel, gaming, and electric vehicles.

Thematic ETF Performance and Flow Trends

According to the ETFThemes.com Thematic Return and Flow Database, using FactSet Research Systems Inc. data, June flows were strongest in semiconductor, smart grid, infrastructure, and copper-mining exposure, while the weakest flows were concentrated in China internet, housing, gold miners, solar, cloud, AI, and internet exposure.

Flow totals reflect available ETF flow fields in the return and flow file. Some model holdings had incomplete or unavailable flow fields in the supplied file. The copper-miners row in the source file was treated as the model’s copper-miners exposure for flow-context purposes.

The flow data were more constructive than the return data. The strongest June flows went into SMH, GRID, PAVE, copper miners, and IGF, showing continued investor interest in semiconductors, electrification, infrastructure, and real-asset supply chains. That is an important distinction: even as the model declined in June, the flow data still showed investors adding to the infrastructure around AI and electrification.

The biggest outflow was KWEB, followed by ITB, GDX, TAN, SKYY, AIQ, and FDN. This suggests investors were reducing exposure to China internet, housing, gold miners, solar, cloud, AI basket exposure, and internet platforms after a strong spring rally.

Top June Flow Winners and Losers

The June flow data were highly informative. SMH was both the strongest flow winner and the strongest model contributor, confirming that investors continued to fund semiconductor exposure even after a very strong first half. GRID and PAVE also attracted meaningful capital, reinforcing the idea that AI infrastructure is broadening into electrification, power systems, grid modernization, and physical infrastructure.

At the same time, KWEB outflows were large and performance was weak, suggesting investors continued to reduce China internet exposure. ITB was an interesting disconnect: the ETF performed well but still saw meaningful outflows, likely reflecting profit-taking or skepticism around the durability of rate-sensitive housing strength.

First Half Flow Winners and Losers

The first-half flow picture confirms three durable investor preferences. First, investors are still allocating aggressively to semiconductors. Second, they are funding the adjacent power, grid, and infrastructure beneficiaries of the AI buildout. Third, they are pulling capital from internet, China internet, cloud, fintech, and some consumer/digital themes that have not produced consistent leadership.

Market Backdrop

June was a tension-filled month. The quarter ended strongly, with the S&P 500 up about 14.9% and the Nasdaq up 21.4% for Q2, but June itself was weaker for the major indexes. Reuters noted that investors remained upbeat on economic and earnings growth, but also flagged concerns around technology valuations and massive AI spending.

The macro backdrop was also less supportive for speculative themes. The Federal Reserve held the federal funds target range at 3.50%–3.75% on June 17 and said inflation remained elevated relative to its 2% goal, partly due to supply shocks and energy-sector price increases. The Fed also characterized economic activity as expanding at a solid pace despite elevated uncertainty tied partly to the Middle East conflict.

That mix matters for the model. Higher-for-longer rates and inflation uncertainty tend to pressure long-duration thematic equities, especially fintech, cloud, speculative growth, crypto-linked equities, and unprofitable innovation themes. At the same time, strong capital investment and AI infrastructure demand continue to support semiconductors, grid equipment, electrification, infrastructure, and power-related themes.

AI power demand remains one of the most important tailwinds for the thematic universe. Reuters reported that EIA expects U.S. electricity demand to rise to record levels in 2026 and 2027, with demand growth driven in large part by data centers tied to AI and cryptocurrency. FERC also ordered major U.S. regional grid operators to consider new protocols for quickly connecting large energy users such as data centers, citing record electricity demand and grid constraints.

The investment implications are visible in the flow data. GRID, PAVE, IGF, SMH, and copper-miner exposure all attracted capital. In addition, Bloom Energy and Brookfield expanded an AI infrastructure power-financing partnership to $25 billion, underscoring how quickly AI demand is turning power availability into a strategic bottleneck.

Model Interpretation

June was a setback, but not a thesis-breaker. The model underperformed because thematic breadth deteriorated, not because AI infrastructure failed. SMH remained the largest contributor, and first-half attribution is still dominated by SMH, AIQ, CIBR, and GRID. That means the model continues to be aligned with the strongest secular theme in the market: AI moving from software and chips into compute, power, grid, infrastructure, and electrification.

The more important message is that the market is becoming more selective. In May, the model benefited from a broad thematic growth rally. In June, investors separated the winners from the laggards. Semiconductors, biotech, home construction, infrastructure, and cybersecurity held up better. Blockchain, Bitcoin-linked exposure, solar, fintech, China internet, cloud, gold miners, and natural resources detracted.

For the first half, the model still delivered a strong result: +16.45% versus SPY up +10.09%. That outcome validates the diversified thematic structure. Technology provided most of the alpha, but Industrial and Commodities Themes also contributed. The biggest challenge is that Consumer Themes were flat on a sleeve basis because strong XBI and BKCH attribution was offset by GBTC and ARKF weakness.

The key setup for the second half is whether AI leadership can broaden again. The most constructive signal would be continued strength in SMH, AIQ, CIBR, GRID, PAVE, IGF, and related electrification/power infrastructure exposure. The biggest risk would be a renewed valuation reset in AI and speculative growth if rates rise, inflation pressures persist, or investors become more skeptical of AI capex monetization.

Bottom Line

The Kaleidoscope Thematic Equity Model declined -3.57% in June versus SPY down -1.03%, producing -2.40% of active underperformance. The weakness came from blockchain, Bitcoin-linked exposure, solar, fintech, natural resources, gold miners, China internet, cloud, and internet exposure. SMH, XBI, ITB, PAVE, CIBR, MLPX, and PEJ helped offset some of the drag.

For the first half of 2026, the model remains ahead of benchmark, gaining +16.45% versus SPY up +10.09%, producing +6.43% of active return. First-half leadership was anchored by semiconductors, AI, cybersecurity, blockchain equities, biotech, grid infrastructure, infrastructure development, and select natural-resource exposure.

The June attribution report shows a thematic market that is still rewarding AI infrastructure, but no longer rewarding all thematic beta equally. The strongest forward setup is in areas where flows, returns, and macro catalysts overlap: semiconductors, AI infrastructure, electrification, smart grid, cybersecurity, and physical infrastructure. The weakest setup remains in China internet, fintech, Bitcoin-linked exposure, and other long-duration growth themes that remain vulnerable to rates, valuation pressure, and investor skepticism around monetization.

Rolling Performance Attribution | Securities Level

Sources

1. Kaleidoscope Thematic Equity Model Rolling Returns & Attribution, June 30, 2026; data from FactSet Research Systems Inc.
2. ETFThemes.com Thematic Return and Flow Database, July 1, 2026; data from FactSet Research Systems Inc.
3. Reuters — S&P 500, Nasdaq register best quarter since 2020 despite Iran war; used for June/Q2 market backdrop, technology/semiconductor context, and valuation concerns.
4. Federal Reserve — June 17, 2026 FOMC Statement; used for rate, inflation, and macro-policy backdrop.
5. Reuters — U.S. power use to beat record highs in 2026 and 2027 as AI use surges, EIA says; used for AI power-demand and electrification backdrop.
6. Reuters — Top U.S. energy regulator pushes grids to overhaul data-center power rules; used for grid-capacity and data-center interconnection context.
7. Reuters — Bloom Energy, Brookfield expand AI infrastructure power partnership to $25 billion; used for AI power-infrastructure capex context.

Disclaimer:  This report is provided for informational purposes only and does not constitute investment advice, a recommendation, or an offer to buy or sell any ETF, security, strategy, or investment product. The information presented should not be relied upon as the sole basis for any investment decision.

Performance, return, flow, and attribution data are based on model portfolio assumptions, ETF market data, and internal calculations using third-party data sources believed to be reliable, including FactSet Research Systems Inc. Past performance is not indicative of future results. Actual investor results may differ due to fees, expenses, taxes, timing, trading costs, implementation differences, and market conditions.

The attribution analysis included in this report is not GIPS compliant and should not be interpreted as a performance presentation prepared in accordance with the Global Investment Performance Standards.