The Diversification Illusion

The Great De-Risking

Between 2018 and 2023, corporate America ran a playbook. Move manufacturing out of China. Diversify the supply chain. De-risk.

Crocs shifted footwear production from Chinese factories to Vietnam. YETI moved 80% of its cooler manufacturing to Vietnam, Thailand, and Malaysia, rerouting imports through the Port of Houston (Supply Chain Dive, 2023). Sunrun’s panel suppliers, JinkoSolar and Trina Solar among them, built massive cell and module assembly lines across Southeast Asia. Analyst notes praised the moves. “Supply chain diversified.” “China risk mitigated.” Upgrade.

The strategy had a flaw that should have been obvious: everyone moved to the same handful of countries.

Vietnam, Malaysia, Thailand, and Cambodia now sit at the center of overlapping tariff actions. Antidumping duties. Countervailing duties. Reciprocal tariffs. The layers stack. For some companies, the combined rates exceed 400%.

The market treated the China-to-Southeast-Asia migration as risk reduction. It was risk relocation. And the new address is now in the crosshairs.

A Handful of Countries, One Chokepoint

Start with solar, where the concentration is most extreme.

Ninety-seven percent of the world’s silicon wafer production happens in China (DOE Solar Energy Technologies Office, 2024). Chinese companies don’t just make wafers domestically. They operate cell and module assembly facilities across Southeast Asia through subsidiaries and joint ventures. JinkoSolar runs factories in Malaysia, Vietnam, and Florida. Canadian Solar manufactures in Vietnam, Thailand, and Indonesia. Trina Solar operates in Vietnam, Thailand, Indonesia, and the UAE.

When Sunrun lists “multiple suppliers” in its 10-K, the diversification sounds real. Jinko, Canadian Solar, Trina, LONGi. Four names. But trace the supply chain one level deeper and those four names converge on the same geography: Vietnamese industrial parks, Malaysian free trade zones, Thai manufacturing corridors.

The U.S. Department of Commerce noticed. On May 15, 2024, Commerce initiated antidumping and countervailing duty investigations covering crystalline silicon photovoltaic cells from all four countries. On April 21, 2025, Commerce announced final affirmative determinations with duty rates ranging from 14.64% (Hanwha Q CELLS, Malaysia) to 3,521% (certain Cambodian producers). The U.S. International Trade Commission voted affirmative on May 20, 2025, confirming that the domestic industry was materially injured. AD/CVD orders were published in the Federal Register on June 24, 2025 (Inv. Nos. 701-TA-722-725 and 731-TA-1690-1693).

The tariffs are not upcoming. They are in effect.

And they layer on top of reciprocal tariffs. Section 201 safeguard tariffs, which added roughly 14% on top of everything else, expired on February 6, 2026 (Solar Power World, February 2026). Their removal helps at the margin, but the remaining stack is still punishing. Reciprocal tariffs (which have been adjusted multiple times since their April 2025 announcement) add another layer that varies by country. Vietnam’s reciprocal rate was initially scheduled to escalate from 10% to 46%, but was negotiated down to approximately 20% as of August 2025 (DHL, 2025). The rate has continued to shift with each policy revision. When you stack AD/CVD and reciprocal tariffs together (with Section 201 now gone), the combined rate on Vietnamese solar cells from non-exempt producers still ranges from roughly 60% to well over 400%, depending on the specific producer and the current reciprocal rate.

The AD/CVD rates alone, which are fixed and published, tell the story for Sunrun’s major panel suppliers:

• JinkoSolar (Malaysia): 40.30% AD/CVD
• JinkoSolar (Vietnam): 244.95% AD/CVD
• Trina Solar (Thailand): 375.19% AD/CVD
• Trina Solar (Vietnam): 201.69% AD/CVD

(Commerce Department Final Determinations, April 21, 2025. Reciprocal tariffs stack on top. Section 201 safeguard tariffs expired February 6, 2026.)

“Multiple suppliers” is not diversification when every supplier manufactures in the tariff zone.

Sunrun’s $14.7 Billion Problem

A note on entities. Sunrun Inc. (NASDAQ: RUN) is a residential solar installer, not a manufacturer. The company doesn’t make panels. It buys them from tier-1 module suppliers, installs them on rooftops under lease or power purchase agreements (PPAs), bundles the contracts into asset-backed securities, and sells those ABS to institutional investors. The cash from ABS sales funds the next round of installations. The cycle repeats.

The model requires continuous origination. Panels go on roofs, contracts get signed, ABS get issued, cash comes in, and panels go on more roofs. Any disruption to panel supply doesn’t just delay installations. It starves the securitization pipeline that funds the entire operation.

Sunrun carries approximately $14.7 billion in total debt as of Q3 2025 (10-Q filing). Of that, the bulk is non-recourse project-level debt tied to specific solar asset portfolios. The company has over $2.6 billion in ABS issuances and $680 million in unused warehouse commitments. As long as originations flow, the machine works. When originations stall, the warehouse facilities sit underutilized and the ABS pipeline dries up.

Sunrun’s 10-K explicitly warns that “tariffs or other factors” could reverse the historic decline in solar component prices. The company also acknowledges that “if one or more [suppliers] that we rely upon to manufacture solar panels, inverters, or battery storage products reduces or ceases production, it may be difficult to quickly identify and qualify alternatives on acceptable terms.”

The Connecticut Attorney General filed a lawsuit against Sunrun in July 2024 alleging deceptive PPA sales practices, including impersonation, forged signatures, failure to obtain permits, and installation of non-functional systems. The case remains pending; Sunrun has disputed the allegations (CT.gov press release, July 2024). The company has also settled multiple TCPA class actions totaling over $5.5 million. More than 20 federal civil cases have been filed since 2023 (CourtListener docket search, February 2026).

The federal solar investment tax credit under Section 25D expired on January 1, 2026, eliminating the 30% credit for customer-owned systems with no phase-out. Section 48E credits for third-party-owned systems (Sunrun’s core model) remain available through 2027, which may actually push more customers toward Sunrun’s lease/PPA products rather than purchasing outright. RBC Capital cited this dynamic in maintaining its Outperform rating with a $22 price target. But the credit expiration still creates market confusion and sales friction that could slow originations for several quarters.

And the insider selling. On January 6, 2026, five Sunrun executives sold shares on the same day:

• Mary Powell, CEO: 8,754 shares at $17.80
• Danny Abajian, CFO: 7,190 shares at $17.80
• Jeanna Steele, Chief Legal and People Officer: 4,430 shares at $17.80
• Maria Barak, Chief Accounting Officer: 1,201 shares at $17.80
• Paul Dickson, President and Chief Revenue Officer: 6,119 shares at $17.80

(SEC Form 4 filings, January 7, 2026. Prices rounded from weighted averages.)

Director Lynn Jurich, Sunrun’s co-founder, sold 50,000 shares at $19.28 on January 2 and another 50,000 shares at $18.58-$19.26 in early February. Director Edward Fenster sold over 163,000 shares at $19.79-$20.72 on February 11.

At least some of these sales were disclosed as tax-withholding transactions tied to RSU vesting, not discretionary open-market reductions (StockTitan, January 2026). Maybe all of them are routine compensation mechanics. Maybe not. I can’t tell you which. I can tell you the timing.

Sunrun reports Q4 2025 and full-year results on February 26, 2026. Analysts expect revenue of $610.3 million and a loss of $0.08 per share. The stock trades at $19.99 as of February 24, with a 52-week range of $5.38 to $22.44 and a market cap of $4.66 billion.

The Handshake Supply Chain

If tariff concentration risk only affected solar, you could call it a sector-specific problem. But the same geographic vulnerability shows up in footwear, through a different tariff mechanism.

Solar faces product-specific trade remedies: AD/CVD orders with published rates for individual exporters and producers. Footwear faces country-level reciprocal tariffs: broad rates applied by country of origin regardless of the specific manufacturer. The legal regimes are different. The geographic chokepoint is the same.

Crocs Inc. (NASDAQ: CROX) has been shifting production toward Vietnam for years. As of fiscal year 2025, the company’s largest third-party manufacturer (primarily Vietnam-based) produced approximately 45% of all Crocs Brand units, down slightly from 50% in 2024. The second-largest manufacturer handled about 28%, also spanning Vietnam and China operations (CROX 10-K, filed February 12, 2026).

HEYDUDE is the sharper case. In 2023, just 5% of HEYDUDE production was in Vietnam. By 2025, that figure was 44%. The majority of HEYDUDE production shifted from China to Vietnam in a single two-year window, and the company already flagged “disproportionate impact of tariffs on HEYDUDE Brand products” as a factor in Q2 2025 results.

Crocs guided flat to slightly lower revenue for 2026, citing tariff headwinds. The stock is at $96.69 with a market cap of $4.86 billion and a forward P/E of 7.27 based on FY2026 consensus estimates (stockanalysis.com, February 24, 2026). The year-to-date decline of 12.8% suggests some tariff risk is priced in, but the market is treating the headwind as manageable. Analysts have a consensus Hold rating with a $96 target, essentially flat to current price.

Crocs carries $1.18 billion in term loan debt on a SOFR-based variable rate, plus $1.48 billion in senior notes (4.25% due 2029, 4.125% due 2031). The indentures restrict dividends and share repurchases if leverage thresholds are breached. A sustained margin compression from tariff costs, particularly on the HEYDUDE brand where the Vietnam shift was most aggressive, could push the company toward those thresholds.

In the company’s older 10-K filings (2011-2016), Crocs disclosed that it had “no written supply agreements with our primary third-party manufacturers in Asia” (SEC EDGAR, CROX 10-K FY2015). The most recent filing references multi-year distribution agreements. Whether the current agreements are binding manufacturing commitments or softer commercial arrangements, the question is the same: how much protection does Crocs actually have if a manufacturer diverts capacity under tariff pressure?

The pattern extends beyond solar and footwear. YETI Holdings (NYSE: YETI) moved roughly 80% of production to Vietnam, Thailand, and Malaysia (Supply Chain Dive, 2023), then pivoted its U.S. import routing through the Port of Houston. Gorilla Technology Group (NASDAQ: GRRR) manufactures edge-AI hardware through a joint venture with Lanner Electronics in Thailand, while a Seeking Alpha bull case described the company as “unlikely to be affected by potential US-imposed tariffs” because its revenue is international. The revenue is international. The supply chain is not.

Four industries, four companies, one geographic chokepoint.

The Tariff Cascade, Eight Months In

The AD/CVD orders took effect in June 2025. Eight months have passed. Why hasn’t the sky fallen?

Partially because it has. Sunrun’s stock hit $5.38 in 2025 (52-week low, per stockanalysis.com), a drawdown of roughly 75% from its 52-week high. It has since recovered to $20, fueled by the Section 48E credit argument, a HASI joint venture for 300MW of capacity, and a management narrative of $200-500 million in 2025 cash generation.

But the recovery assumes the tariff impact is priced, managed, and fading. Three factors suggest it may be accelerating instead.

First, reciprocal tariff rates on Southeast Asian countries have been volatile since their April 2025 announcement, with multiple pauses, increases, and modifications. Vietnam’s rate was initially scheduled at 46%, negotiated down to approximately 20% by August 2025, and has continued shifting. Each policy revision adds uncertainty to procurement planning and makes long-term supply contracts harder to price.

Second, inventory buffers are finite. Multiple solar companies, including Sunrun, executed “safe harbor” equipment purchases in late 2024 and early 2025, pre-buying panels at pre-tariff prices (Sunrun’s Q4 2024 safe harbor purchase was approximately $350 million). Those stockpiles are drawn down over the course of 2025. When they run out, new panel procurement happens at post-tariff prices. The question for Q4 2025 earnings (reported February 26) is whether that transition has started showing up in the unit economics.

Third, AD/CVD rates are additive, not substitutive. Each new tariff layer stacks on top of the existing ones. A solar installer that priced a 20-year PPA based on a panel cost of $0.25/watt faces a very different business case if the landed cost is $0.40/watt or $0.55/watt. At some point, the PPA rate required to maintain margins exceeds what residential customers will sign.

The Wagers

Two option trades, one for each company. Correlated through the tariff mechanism, uncorrelated at the company level. If one fails on company-specific factors (Sunrun solves its supply chain, Crocs absorbs the margin hit), the other can still fire.

Before the specific trades, a piece of market color. The September 2026 $15 RUN put currently carries 40,023 contracts of open interest (Polygon.io, February 24, 2026). That’s 4 million shares of put exposure at a strike 25% below the current price, priced at $2.13 per contract. Someone, or some group, has built a large position betting on (or hedging against) a significant Sunrun decline within the next seven months. The implied volatility on that contract is 87.6%. For context, RUN’s trailing 30-day realized volatility is roughly 65%. The options market is pricing in more turbulence ahead than what the stock has shown recently.

The speculative trade sits further out of the money. The September 2026 $10 put asks $0.76 (Polygon.io, February 24, 2026), roughly 50% below the current stock price. At $76 per contract, the maximum loss is small. The payoff math: if RUN drops to $5 (which it hit in 2025), the contract is worth $424 after subtracting the premium, roughly 5.6x. A move to $3 pays $624, about 8.2x. Implied volatility on this strike is 87.5%, and the delta is -0.08, meaning the market assigns roughly an 8% probability of finishing in the money.

Sunrun reports earnings on February 26. The September 2026 expiration gives the trade roughly seven months of runway, spanning the February earnings release, any reciprocal tariff adjustments, and the Q2 2026 earnings that would reflect the full tariff impact on unit economics.

The CROX trade: September 2026 $70 put at $2.90 ask (Polygon.io, February 24, 2026), 27.6% out of the money, with 424 contracts of open interest. Cost per contract: $290. Delta: -0.14. Implied volatility: 51.3%, roughly in line with CROX’s recent realized volatility, suggesting the market isn’t pricing in an unusual downside event for Crocs specifically. If the stock drops to $50 (a tariff-driven margin compression scenario), the contract pays roughly $1,710, about 5.9x.

The CROX trade aligns with the spring/summer production cycle when Vietnam-sourced inventory for fall retail hits the tariff wall. HEYDUDE’s 44% Vietnam concentration makes it the more exposed brand, and management already flagged the impact.

Fair warning on both: option chains move. By the time you read this, the numbers will be different. These are illustrations of convexity, not recommendations. The point isn’t that these exact contracts are mispriced today. The point is that deep OTM puts let you structure a position on a conditional scenario without putting meaningful capital at risk.

The Cascade

The scenario for Sunrun, spelled out step by step. Every link is conditional. Some have already happened. I’m not predicting the rest will follow. I’m saying if they do, the stock isn’t priced for it.

Some friction that could slow or break the chain. Sunrun’s $680 million in unused warehouse commitments provides a liquidity cushion. The company could renegotiate supplier contracts to share tariff costs, though the AD/CVD rates on Trina and Jinko specifically (not just “the industry”) limit how much flexibility exists. Exclusion petitions are possible but take months to process and don’t pause tariff collection during review. The industry could receive a legislative carve-out to protect Inflation Reduction Act deployment targets, but congressional appetite for tariff exemptions in the current trade-hawkish environment is uncertain at best.

Sunrun navigated COVID-era supply disruptions. The company has operational experience managing panel shortages. The difference is scale. COVID disruptions were temporary and broadly shared. AD/CVD tariffs are structural, company-specific (varying by producer and country of origin), and scheduled to increase.

Conviction and Humility

Two questions.

Will the tariff impact on Southeast Asian solar manufacturing remain severe? Likely. The ITC ruling is final, the AD/CVD orders are published, and the rates are set. Court challenges by Jinko Solar Vietnam and Trina Solar Thailand are pending but don’t pause duty collection. Reciprocal tariffs remain volatile and have been directionally hawkish despite interim negotiations. U.S. wafer production has been absent since 2016, and domestic cell manufacturing remains negligible relative to import volumes (DOE Solar Supply Chain Factsheet, December 2024). IRA-funded factories are under development but years from producing at scale. The industry can’t reshore its way out of a geographic tariff in 12 months.

Will it cascade through Sunrun’s balance sheet to the point of equity destruction? Lower probability. Sunrun managed through previous disruptions. The company projects $200-500 million in cash generation for 2025. The TPO model may benefit from Section 25D expiration. RBC maintains Outperform at $22. The $14.7 billion debt load is mostly non-recourse project-level debt, which limits contagion to the parent. Management has demonstrated capital markets access (over $1 billion raised in non-recourse financing in 2024).

Three Reasons the Tariff Thesis Could Be Wrong

1. Exclusion rates and administrative reviews significantly reduce the effective tariff burden on Sunrun’s specific suppliers. JinkoSolar’s Malaysia rate (40.30%) is much lower than its Vietnam rate (244.95%). If Jinko shifts production to Malaysia or qualifies for company-specific exclusions, the actual cost impact narrows.

2. Sunrun pre-staged enough inventory through safe harbor purchases to bridge until either the tariffs are reduced, legislative relief arrives, or domestic manufacturing scales. The $350 million Q4 2024 pre-buy provides meaningful runway.

3. Congressional intervention. The Inflation Reduction Act set aggressive solar deployment targets. AD/CVD tariffs that make residential solar installations uneconomic directly undermine those targets. A tariff exemption or phase-in for residential solar equipment is politically plausible, particularly if utility-scale installations slow visibly.

Why It Might Be Closer Than the Market Thinks

1. The tariff rates are extraordinary. Even the “favorable” JinkoSolar Malaysia rate of 40.30% (before reciprocal tariffs) adds roughly $0.08-$0.12/watt to panel costs. The unfavorable rates (Jinko Vietnam at 245%, Trina Thailand at 375%) make those supply channels commercially unviable for the U.S. market. Section 201 expired in February 2026, which helps, but the AD/CVD rates alone exceed what most installers can absorb.

2. Every major Sunrun supplier manufactures in the targeted countries. “Multiple suppliers” is geographic concentration disguised as vendor diversity. Jinko, Trina, Canadian Solar, and LONGi all have significant operations in Vietnam, Malaysia, or Thailand.

3. Reciprocal tariff policy remains volatile and directionally hawkish. Each escalation stacks on top of AD/CVD rates that already exceed 200% for some producers. The trend line points up.

4. Sunrun’s debt structure amplifies supply disruptions. The ABS model requires continuous origination. A two-quarter installation slowdown doesn’t just defer revenue. It disrupts the securitization pipeline that funds operations. The non-recourse structure protects lenders but concentrates equity risk.

5. Five executives sold shares on the same day (January 6, 2026). The CEO, CFO, CLO, CAO, and President all filed Form 4s reporting sales at approximately $17.80/share. The co-founder sold 50,000 shares the week before. The stock is now at $19.99.

The strongest argument against the full cascade: federal intervention. If the administration decides that protecting IRA solar deployment targets outweighs trade enforcement, a tariff exemption for residential solar equipment could eliminate the catalyst. A 90-day exemption window alone would push the impact past near-term option expirations and give the industry time to adjust procurement. The prior administration used exactly this mechanism in June 2022 (a two-year tariff moratorium) when the original SE Asia circumvention inquiry threatened to freeze solar imports. That moratorium expired in June 2024. The Trump administration has shown no inclination to renew it, and its broader trade posture has been toward escalation, not exemption. The precedent for intervention exists. The political conditions for it do not.

The Rubber Swan Scorecard

I rate every thesis across seven dimensions.

What to Watch

Sources

AD/CVD Tariff Proceedings

1. Commerce Department Final Determinations: Crystalline PV Cells from Cambodia, Malaysia, Thailand, and Vietnam (Apr 21, 2025): AD/CVD rates by producer and country
2. USITC Final Determination: Material Injury, Inv. Nos. 701-TA-722-725 and 731-TA-1690-1693 (May 20, 2025): affirmative injury finding
3. Federal Register: AD/CVD Orders Published (Jun 24, 2025)
4. Norton Rose Fulbright: Updated Solar Import Tariffs (Apr 2025): rate summary and legal analysis
5. Solar Power World: Commerce Reveals Final Tariff Amounts on SE Asian Solar Imports (Apr 21, 2025)
6. Solar Power World: End of an Era: Sec. 201 Tariffs on Imported Solar Panels Expire (Feb 2026)
7. American Alliance for Solar Manufacturing Trade Committee: Commerce Gives U.S. Solar Manufacturers Decisive Victory (Apr 21, 2025)
8. DHL Vietnam: US-Vietnam Tariffs, A Guide for Exporters (2025): reciprocal tariff rate history

SE Asia Manufacturing Concentration

9. DOE Solar Energy Technologies Office: Solar PV Supply Chain Deep Dive (2024): global wafer/cell production data, 97% China wafer share
10. DOE: Solar Energy Supply Chain Report (Dec 2024): “Achieving American Leadership in the Solar Supply Chain”
11. JinkoSolar 20-F: manufacturing locations (Malaysia, Vietnam, Florida)
12. Canadian Solar 20-F: manufacturing locations (Vietnam, Thailand, Indonesia)
13. Trina Solar 2022 Sustainability Report: manufacturing locations (Vietnam, Thailand, Indonesia, UAE). Trina delisted from NYSE in 2017; no SEC filings available post-2017.

Sunrun Capital Structure and Operations

14. Sunrun 10-K FY2024: $14.7B total debt, ABS issuances, warehouse commitments
15. Sunrun 10-Q Q3 2025 (filed Nov 6, 2025): safe harbor program balance, Q3 2025 financials
16. Sunrun: Q4/FY2025 Earnings Date Announcement: Feb 26, 2026
17. RBC Capital: Outperform rating, $22 price target (broker report, no public URL)
18. stockanalysis.com: RUN: price, market cap, 52-week range as of Feb 24, 2026

Sunrun Insider Trading

19. SEC Form 4: Mary Powell (CEO), filed January 7, 2026, 8,754 shares at $17.80
20. SEC Form 4: Danny Abajian (CFO), filed January 7, 2026, 7,190 shares at $17.80
21. SEC Form 4: Jeanna Steele (CLO), filed January 7, 2026, 4,430 shares at $17.80
22. SEC Form 4: Maria Barak (CAO), filed January 7, 2026, 1,201 shares at $17.80
23. SEC Form 4: Paul Dickson (Pres/CRO), filed January 7, 2026, 6,119 shares at $17.80
24. SEC Form 4: Lynn Jurich (Director), filed January 6, 2026, 50,000 shares at $19.28
25. SEC Form 4: Edward Fenster (Director), filed February 12, 2026, 163,844 shares at $19.79-$20.72
26. StockTitan: Sunrun CEO Rule 144 Filing (Jan 2026): RSU tax-withholding context

Crocs Supply Chain

27. CROX 10-K FY2025 (filed Feb 12, 2026): Vietnam production (45% Crocs Brand, 44% HEYDUDE)
28. CROX 10-K FY2024: Vietnam production (51% Crocs Brand, 20% HEYDUDE)
29. CROX 10-K FY2015: “We do not have written supply agreements with our primary third-party manufacturers in Asia”
30. StockTitan: CROX 10-K Annual Report Summary: tariff impact on HEYDUDE noted as Q2 2025 factor

Broader Tariff Exposure

31. Supply Chain Dive: Yeti Will Move 80% of Production Out of China (2023): Vietnam/Thailand/Malaysia manufacturing shift
32. Gorilla Technology and Lanner Electronics JV Announcement: Thailand manufacturing; Seeking Alpha bull case on tariff non-impact

Legal and Regulatory

33. CT Attorney General: Lawsuit Against Sunrun (Jul 2024): deceptive PPA sales practices; case pending
34. Energy.gov: Federal Solar Tax Credits for Businesses: Section 25D expired January 1, 2026 (no phase-out for customer-owned systems)
35. 26 U.S.C. § 48E: Section 48E ITC for third-party-owned systems, available through 2027
36. Commerce Department FAQ: Expiration of Presidential Proclamation 10414: Biden two-year tariff moratorium on SE Asian solar (Jun 2022; expired Jun 2024)

Stock prices, option chain data, open interest, and implied volatility figures sourced from Polygon.io on February 24, 2026 and may have changed since publication.

Disclosure: I hold small positions in the option contracts described above. They’re speculative positions, not portfolio allocations.

Not financial advice. This is a stress-test scenario published for educational and entertainment purposes. It describes a conditional sequence of events, not a prediction. Options involve risk of total loss. Do your own research. Consult a financial advisor. I’m a rubber swan, not a fiduciary.