Commentary: Global Growth Strategy

• Global equities rose modestly, but a sharp AI-led growth selloff and rising rates drove a pronounced rotation into value.
• The Strategy underperformed as an overweight to AI infrastructure holdings built over the previous quarter reversed sharply, while underweights to rebounding U.S. hyperscalers also detracted.
• We view the AI-led momentum selloff as sentiment driven and remain constructive on the technology’s long-term potential. At the same time, we continue to diversify our exposure, managing risk and positioning for broader enterprise adoption.

Global equities delivered mixed results in the third quarter, with growth stocks pressured by a sharp selloff in momentum-driven AI companies while rising rates and inflationary effects supported value names. The benchmark MSCI All Country World Index advanced 1.6% for the quarter. Investment style dispersion was pronounced, with the MSCI ACWI Growth Index falling 0.5% to trail the MSCI ACWI Value Index by 420 basis points.

In July alone, the ACWI Growth Index declined 2.8% compared to a gain of 3.1% for ACWI Value. Year to date, value has outperformed growth by over 600 bps while over the trailing one-year period the differential is 710 bps.

Exhibit 1: MSCI Growth vs. Value Performance

Growth stocks have thus far this year been driven by AI buildouts across sectors including consumer staples, consumer discretionary, industrials and, of course, technology. Behind this were the industry concentrations, leveraged ETFs and hedge funds playing the global AI trade, most notably in emerging Asia. This was followed by a sharp reversal in July and into August as investors questioned the return on investment from historically high AI capex and unwound highly leveraged bets on the AI buildout.

Value stock performance, on the other hand, has been driven more by geopolitics and the implications of rising inflation and interest rates. These three factors have seen extremes in the last few years, in large part due to the geopolitical conflicts across Europe and the Middle East. Interest rates have risen due to the return of inflation fears while the macro backdrop remains uncertain. We believe this environment has been conducive to value companies.

The July growth selloff accounted for the bulk of the ClearBridge Global Growth Strategy’s quarterly underperformance versus the benchmark, with the heaviest losses occurring among the Strategy’s IT and industrials holdings. At the start of the third quarter, we believed the portfolio was well-positioned but underestimated the degree of overexposure to AI-indexed stocks.

We have historically focused on high-quality growth companies with durable competitive advantages, strong returns on capital and attractive long-term growth prospects. This has included substantial exposure to structural beneficiaries of AI investment, such as NVIDIA and ASML. During 2026, however, market leadership has broadened well beyond these established technology leaders. Some of the strongest returns have come from companies operating in more commoditized and historically cyclical industries, including memory, optical and electronic components as well as batteries.

The magnitude of the rerating in some of these companies has been extraordinary. Japanese memory chip maker Kioxia provides perhaps the clearest example. Its share price increased from approximately ¥10,000 at the beginning of the year to more than ¥100,000 by mid-June (up 10x), before subsequently correcting sharply (-50%) in July. While Kioxia represents an extreme case, similarly elevated returns and volatility were evident across a number of companies exposed to supply-constrained areas of the AI capital-expenditure cycle.

Many of these businesses had historically fallen outside our growth investment universe. They operated in industries characterized by commoditized products, significant cyclicality, periods of excess capacity and, in many cases, relatively weak long-term shareholder returns. The scale and speed of the current AI infrastructure buildout, however, created acute supply-demand imbalances across several of these markets, resulting in significant earnings upgrades and a rapid repricing of the companies exposed to them.

We recognized relatively early that AI-related demand would extend beyond semiconductors and into a much broader range of infrastructure components. What we underestimated was the speed and magnitude of the resulting earnings revisions and share-price appreciation in some of the more cyclical beneficiaries. As the strength of the underlying demand became clearer, we selectively added exposure to several of these areas. However, some of those investments were initiated after substantial appreciation had already occurred, leaving the Strategy more exposed to the sharp technical correction in July.

Importantly, our underperformance has therefore not reflected an absence of exposure to the AI investment cycle. Rather, it has reflected the particular character of market leadership in 2026, where a significant portion of benchmark returns was generated by higher-beta companies whose historical business quality and return profiles had generally not met our investment criteria.

Notably, our active repositioning left the portfolio overweight in stocks that fell 20% to 30% in the third quarter. From an individual security standpoint, several of the largest relative detractors were AI infrastructure companies, many of which were among the Strategy’s leading contributors in the previous quarter. These included semiconductor names Infineon, ASM International and Samsung Electronics, Chinese battery maker CATL, optical networking maker Nokia, data center cooling and power provider Vertiv and high-voltage cable supplier Prysmian. This cohort was at the center of the July AI unwind sparked by concerns around AI compute oversupply, execution of the latest high-bandwidth memory chips, cheaper Chinese AI models, doubts over AI capex returns and forced hedge fund deleveraging. Importantly, this was not an earnings problem: nearly every AI hardware company we own reported record results or saw their earnings continue to be upgraded.

Relative performance was also impacted by an underweights to hyperscaler Microsoft and not holding Meta Platforms. Microsoft, our largest purchase during the quarter, was supported by continued Azure cloud momentum, encouraging Copilot usage data and signs of recovery in enterprise software. Meta advanced following a strong reception to the launch of its Muse personal AI agent. Early downloads surpassed those of the ChatGPT and Gemini debuts, reinforcing our view that agentic commerce is becoming a broader opportunity.

AI headwinds were partially offset by diversified contributions across sectors outside of technology. These included U.S. life sciences tools maker Thermo Fisher Scientific, which continued to benefit from its second-quarter earnings beat, with resilient biopharma consumables and pharmaceutical product development end markets and healthy third-quarter guidance supporting the prospect of an accelerating business into 2027. Meanwhile, Intercontinental Exchange rose on solid second-quarter results, strong trading activity and record open interest, with elevated market volatility supported trading volumes across asset classes during the summer. Japanese financial conglomerate Mitsubishi UFJ benefited from higher interest rates in Japan.

Expanding our structural growth exposure and taking advantage of oversold conditions to repurchase secular growers we like were the themes of our quarterly trading activity. We believe these moves balance our higher-beta AI exposure and will enhance our ability to participate effectively across a range of market and macro conditions.

We increased our financials exposure with the additions of JPMorgan Chase and Morgan Stanley. JPMorgan operates what we view as the highest-quality U.S. bank, which consistently wins market share. We repurchased shares as still-rising rates positively impact net interest income and due to an increase in corporate lending. Morgan Stanley is a quality compounder driven by wealth management with an impeccable track record of gathering assets. Market appreciation and operational leverage have led to increasing margins. On the investment banking side, the firm’s prime brokerage, equity fixed income trading and M&A all support superior earnings growth.

After selling out of German enterprise software maker SAP in April amid concerns that agentic AI could structurally disrupt the software-as-a-service (SaaS) industry, our view has since evolved, prompting a re-entry. The evidence increasingly suggests SAP is better positioned to benefit from AI than most SaaS companies, given the mission-critical nature of its offering, deeply embedded customer relationships and ownership of uniquely valuable enterprise data and business context. Other repurchases included Hong Kong Exchanges & Clearing on improving trading volumes and Thomson Reuters on the potential for AI monetization of its proprietary legal and accounting data.

We exited U.S. streaming provider Netflix due to user engagement data remaining soft with limited new content in development until 2027, which has weighed on guidance. We also sold out of semi cap equipment maker ASM International, Arista Networks and Vertiv to better consolidate our AI infrastructure exposure.

We view the third-quarter selloff in AI-related portfolio holdings as rotational and sentiment driven — mainly the unwinding of the Situational Awareness and related hedge funds in the U.S. — as their fundamentals remain robust. We continue to participate in the long-term AI capital expenditure opportunity while maintaining our focus on companies where we believe growth can translate into durable earnings, attractive returns on capital and sustainable shareholder value.

Maintaining this meaningful and deliberate exposure to the AI theme while managing a prudent level of risk and sector weightings in the portfolio is where we are spending a lot of our time right now. AI remains a powerful trend with intermediate-term tailwinds — areas of the AI infrastructure buildout where scarcity and supply or service bottlenecks are creating pricing power and/or new sources of demand. This is occurring across not only semiconductors, but also advanced packaging, storage, networking and optical infrastructure. In addition to the U.S., the global AI capital spending cycle is benefiting companies across Japan, South Korea, Taiwan and parts of Europe.

As active investors attuned to secular growth trends, we are willing to make substantial changes to improve our participation in such trends. While we saw some early signs of success from these efforts in the second quarter, we acknowledge that not all portfolio holdings will be immediately recognized for their growth attributes. This is where diversification and duration can work to our advantage.

Thematic impacts such as AI have become much more pervasive than in the past and it is leading us to regularly measure the correlation of our portfolio to different stocks in the benchmark and make adjustments accordingly. For example, we have reduced our overweights to IT, consumer discretionary and financials while establishing exposure to energy.

We are also upgrading our investment resources to better navigate a very narrow, trend-driven international equity market. We are using new tools and collaborating with ClearBridge’s data science team to more closely monitor risk factors, evaluate our portfolio activity and more effectively leverage shared research across the firm.

Looking ahead, we expect AI to evolve beyond today’s infrastructure beneficiaries into a much broader enterprise productivity theme. While the portfolio continues to invest across the AI supply chain — including semiconductor equipment, electrical infrastructure, utilities, cable manufacturers and other “picks-and-shovels” businesses — we believe the next phase of AI will be driven by adoption across financial services, health care, manufacturing and other industries, creating significant productivity gains and new business models.

The ClearBridge Global Growth Strategy underperformed its MSCI All Country World benchmark during the third quarter. On an absolute basis, the Strategy had positive contributions from two of the nine sectors in which it was invested (out of 11 sectors total). The leading contributor was the consumer discretionary sector, while the industrials and utilities sectors detracted the most.

On a relative basis, stock selection in consumer discretionary contributed to performance. Conversely, stock selection in IT, industrials, health care, financials, communication services and materials detracted from performance. A lack of exposure to energy also weighed on results.

On an individual stock basis, the largest contributors to relative returns were Thermo Fisher Scientific, Intercontinental Exchange and not owning SK Hynix, Tesla and Applied Materials. The largest detractors were Contemporary Amperex Technology, Life360, ASM International, an underweight to Microsoft and not owning Meta Platforms.

In addition to the transactions mentioned above, we initiated positions in Salesforce, Intel, KLA and Kioxia Holdings in IT, Prudential, UniCredit, Tokio Marine Holdings, Robinhood Markets in financials, Linde in materials, Sunbelt Rentals and ABB in industrials, Abbott Laboratories, Argenx, Rhythm Pharmaceuticals, Scholar Rock Holding and Guardant Health in health care and Keurig Dr Pepper in consumer staples. We exited positions in SK Square, Siemens Energy and Brambles in industrials, Lasertec and Halma in IT, SK Telecom in communication services, HSBC Holdings, KBC Group, HDFC Bank and Orix in financials, Pan Pacific International Holdings and Laopu Gold in consumer discretionary and CG Oncology in health care.