The semiconductor industry is renowned for its pronounced cyclical volatility. Our stock selection strategy is specifically designed to address this inherent characteristic. Rather than simply chasing industry hotspots, the approach combines independent assessments of the current cyclical position with rigorous analysis of financial leverage. The objective is to precisely capture the maximum profit elasticity and valuation recovery potential as the industry moves from trough toward recovery. We focus on identifying financially sound companies with high operating leverage and robust technological capabilities near cyclical lows, positioning the portfolio to deliver substantial returns during the subsequent upturn.
We aim to acquire high-quality assets at compelling valuations when industry sentiment is depressed and prosperity is low. We deliberately avoid pursuing momentum at cyclical peaks where bubbles often form. Instead, we seek to identify industry bottoms through objective, multi-dimensional indicators, enabling us to build positions ahead of the recovery and capture the primary phase of value re-rating. Cyclical positioning is determined by tracking supply-chain inventory levels (including turnover days from fabless designers through to distributors), and by monitoring global capital expenditure trends among major players—contraction in capex frequently serves as a leading signal of an impending cyclical trough. Concurrently, we compare company-specific valuation metrics (P/E and P/B) against their own historical ranges and global peer sets, placing particular emphasis on opportunities where multiples reside in the lower historical percentiles.
We target companies capable of converting incremental revenue into outsized profit expansion. The semiconductor manufacturing business model is characterized by high fixed costs; consequently, as demand recovers and capacity utilization rises, firms with elevated operating leverage experience gross margin and net profit expansion that significantly outpaces revenue growth, generating substantial profit elasticity. We conduct detailed examinations of cost structures, quantifying the proportion of fixed versus variable costs. We also model potential changes in gross and operating margins under varying revenue growth scenarios to assess profit sensitivity to sales. Special attention is given to companies that suffered severe margin compression during the prior downcycle yet preserved their market position and production capacity intact, as these firms typically exhibit the strongest earnings rebound momentum during recovery.
This criterion ensures that selected investments benefit not only from near-term cyclical recovery but also maintain leadership through successive waves of technological evolution. The aim is to exclude ephemeral cyclical names and instead focus on durable industry leaders capable of navigating multiple cycles through sustained innovation-driven growth. We place significant weight on the intensity and effectiveness of R&D spending, with particular scrutiny of accumulated expertise and breakthroughs in advanced nodes, specialty processes, chip architecture, advanced packaging, and core intellectual property. We further evaluate client concentration, product-line competitiveness, and strategic positioning within high-growth end markets. Priority is assigned to companies that have established defensible technological moats in specialized segments and are actively shaping industry roadmaps, thereby preserving strong competitive positioning well beyond the immediate cyclical upswing.
The question “Is it too late?” is often misleading. The more relevant issue for semiconductor investing is determining the current position within the industry sub-cycle and the specific segment of the value chain.
Even after the broader industry has entered an upturn, significant valuation and earnings expectation mismatches can persist across sub-sectors and business models.
Historical patterns indicate that cyclical peaks are frequently accompanied by excessively optimistic capital expenditure plans, while the greatest risk arises from holding maximum leverage exposure at profit highs.
Thus, the suitability of semiconductor exposure in 2026 hinges on whether attractive asymmetric risk-reward profiles still exist—whether driven by supply-demand dynamics, technological differentiation, or structural business model advantages—rather than on the overall direction of the industry cycle.
The distinction between fabless and foundry companies extends beyond business models to encompass fundamentally different risk and return profiles.
Fabless firms typically operate with lighter capital structures, higher gross margins, and greater exposure to end-market demand fluctuations. Foundries, by contrast, are highly capital-intensive with elevated barriers to entry, but their profitability is tightly linked to capacity utilization rates and the capital expenditure cycle.
From an investment perspective, fabless companies function more like “demand amplifiers,” delivering outsized returns during periods of strong end-market growth, whereas foundries act as “cyclical levers,” exhibiting pronounced profit elasticity during capacity-constrained recovery phases.
In the early stages of economic and demand recovery, foundries often display the most significant profit upside; over longer-term structural growth periods, fabless companies tend to command richer technological and margin premiums.
The pronounced cyclicality of the semiconductor industry is not random but arises from the interaction of three reinforcing factors: fluctuations in end-market demand, inventory adjustment cycles across the supply chain, and the lagged response of capital expenditures.
During demand downturns, inventory destocking tends to amplify the downside in both revenues and earnings. In recovery phases, the delayed supply response—due to long lead times for new capacity—magnifies the profit rebound.
This dynamic produces the characteristic “overreaction” in earnings performance: downturns are deeper and recoveries sharper than underlying demand changes alone would suggest.
Cyclicality per se is not inherently a risk; the true risk lies in assuming elevated exposure at the wrong point in the cycle.