Key Fact	Detail
Report	Forensic Short — Research Case Study
Publication date	July 21, 2026
Issuer	Delixy Holdings Limited
Ticker	NASDAQ: DLXY
Domicile	Cayman Islands (operations: Singapore; PRC-linked trade flows)
Reference price	~US$0.45 (July 20, 2026 close)
52-week range	US$0.34 – US$7.00
IPO	July 9, 2025 at US$4.00 (Bancroft Capital, sole underwriter)
Estimated market cap	~US$7.4M (16,350,000 shares)
Industry (per filings)	Crude oil & oil-products trading (SIC 5172)
Rating	NOT RATED — retrospective case study; see Section 13 of the PDF
Nasdaq status	Bid-price deficiency (Rule 5550(a)(2)); cure deadline October 20, 2026

Important disclaimer — read first. This document is a private research exercise prepared for educational and pattern-study purposes, applying a forensic due-diligence checklist to public records. It is not investment advice, not a solicitation, and not published research. Every factual statement is drawn from public filings (SEC EDGAR), company press releases, or third-party market-data services, each cited in Appendix A of the PDF. Interpretations are expressly framed as opinion. Nothing in this report alleges fraud or unlawful conduct by Delixy Holdings Limited, its officers, or its shareholders. Every transaction described was disclosed by the company itself in its own SEC filings; the analysis concerns the economic pattern those disclosures form, not any concealed conduct. Where searches found nothing, the absence is stated rather than filled by inference.

---

Executive Summary

Delixy Holdings Limited is a Cayman Islands holding company for a Singapore oil-products trading business (Delixy Energy Pte. Ltd., incorporated 2007) that listed on the Nasdaq Capital Market on July 9, 2025 at US$4.00 per share, briefly traded as high as US$7.00, and has since collapsed ~89% below its IPO price to US$0.45, receiving a Nasdaq minimum-bid-price deficiency notice on April 23, 2026.

In our view, the public record is consistent with an offering whose economics primarily served pre-IPO insiders rather than public investors:

1. More cash flowed out to insiders before and during the IPO than the company raised from the public. Public filings show ~US$12.7M of dividends declared to pre-IPO owners in FY2023–FY2024 (against ~US$1.0–1.2M of annual net income), US$2.667M of dividend cash paid out in the six months immediately preceding the IPO, and US$2.6M of IPO proceeds going to two insider vehicles as selling shareholders — versus net primary proceeds to the company of only ~US$3.1M.
2. 36.66% of the company was transferred by the CEO to five BVI vehicles ~3 months before the F-1 was filed, and those shares were registered for resale concurrently with the IPO and expressly carved out of the lock-up — arming, from day one, a supply pool ~4× the size of the primary offering.
3. The financial statements show a razor-thin trading book with concentration patterns we consider forensically significant: gross margin of 0.8–1.4%; at FY2024 year-end a single "Customer I" accounted for ~100% of accounts receivable while a single "Supplier I" accounted for ~100% of accounts payable in near-identical amounts (US$17.5M vs US$17.1M); at FY2025 year-end the same mirror pattern repeated with a new pair ("Customer L" US$20.4M receivable / "Supplier L" US$20.2M payable). The company's own notes assert "no significant concentrations of risk related to major customers" directly beneath these tables.
4. Post-IPO governance moved in one direction — entrenchment: the CFO resigned six months after listing (December 31, 2025); an EGM in February 2026 converted all of the CEO's shares into Class B stock carrying 50 votes per share (≈98.5% of voting power on ~56% of the economics), adopted an equity incentive plan, and pre-authorized a reverse split of up to 1:500 at the board's sole discretion — two months before the Nasdaq deficiency letter arrived.
5. FY2025 results confirm the deterioration: revenue US$307.7M (−2.3%), net loss US$4.46M, operating cash flow −US$5.2M, year-end cash of US$1.79M, and a going-concern discussion in the annual report — for an entity whose market capitalization (US$7.4M) is now less than the dividends its insiders extracted pre-IPO.

We did not find: any litigation, regulatory enforcement, auditor turnover, late filings, or direct evidence of coordinated stock promotion. Those absences are stated in the relevant sections and materially temper the thesis. This is a pattern case, not a smoking-gun case.

---

The Thesis

In our view, DLXY at IPO represented a ~US$65M valuation (16.35M shares × US$4.00) placed on a business with:

• ~US$1.35M of total shareholders' equity at December 31, 2024 (post-dividend-extraction) — a price-to-book of ~48×;
• ~US$1.0M of net income (FY2024) — a P/E of ~64× for a commodity pass-through trader;
• a 1.4% gross margin, five traders, and one strata office unit in Singapore.

The subsequent 89% collapse has largely closed that gap. What remains forward-looking is the structural setup: a sub-US$1.00 stock with an October 20, 2026 compliance deadline, a pre-authorized reverse split of up to 1:500, US$1.8M of cash against a negative operating cash flow run-rate, 450M authorized Class A shares, a 50:1 dual-class structure that makes dilution costless to the controlling shareholder, and an FPI disclosure regime that requires no further financial disclosure until April 2027.

---

The IPO: Small Raise, Large Insider Participation

Offering mechanics (424B4, July 9, 2025):

Item	Amount
Total offering	2,000,000 shares @ US$4.00 = US$8.0M gross
— Primary (company)	1,350,000 shares = US$5.4M gross
— Selling shareholders	650,000 shares = US$2.6M gross (Mega Origin 325,000; Novel Majestic 325,000)
Net proceeds to company	~US$3.125M (per prospectus estimate)
Underwriter	Bancroft Capital, LLC (sole lead; firm commitment)

Observations, in our view all individually disclosed and collectively telling:

1. 32.5% of the IPO was insiders cashing out. The CEO's vehicle and the largest outside vehicle each sold 325,000 shares into the offering itself.
2. Offering costs consumed ~42% of the company's gross raise (US$5.4M gross → ~US$3.1M net).
3. The registration was slow and heavily commented: confidential DRS June 2024 → seven F-1/A amendments → nine SEC comment-letter entries, with a three-month gap between first effectiveness (March 31, 2025) and pricing — consistent with a deal that initially could not get done.
4. Use of proceeds was generic (product expansion / working capital), with management retaining broad discretion.

---

Pre-IPO Value Extraction

All figures from the FY2025 Form 20-F and H1-2025 interim statements — i.e., the company's own disclosures.

Dividends out before the public came in. US$12,737,000 of dividends were declared in FY2023–FY2024 against combined net income of ~US$2.2M for the same period. The December 2023 dividend (US$9.7M) was declared against total shareholders' equity of ~US$11.9M — substantially the entire book value of the business, declared out to the controlling shareholder's BVI vehicle on one day. US$5.0M of that unpaid dividend was converted into an interest-bearing loan owed by the company to Mega Origin at 3.5% p.a. — the controlling shareholder's dividend claim became a creditor claim, maturing December 31, 2026 (balance at December 31, 2025: US$4.0M). The H1-2025 cash-flow statement shows a further US$2.667M of "Dividend paid" in the six months immediately preceding the IPO, while group cash fell from US$5.6M to US$1.8M.

The arithmetic that summarizes the offering: insiders received ~US$12.7M in dividends (FY23–24) + US$2.6M selling at the IPO ≈ US$15.3M out, while the company received ~US$3.1M in. The current market capitalization of the entire company (US$7.4M) is roughly half of what was extracted.

The August 2024 share transfers — the future supply pool. On August 21, 2024 (~3 months before the public F-1), Mr. Xie Dongjian transferred 36.66% of his shares to five BVI vehicles (Novel Majestic 18.62%; four "Independent Third Party" vehicles of 4.90%/4.90%/4.90%/3.34%). These five are the prospectus-defined "Resale Shareholders": a resale prospectus (424B3) was filed the same day as the IPO prospectus, and the underwriting lock-ups expressly excepted their registered shares. Scale: ~5.5M resale-eligible shares versus a 1.35M-share primary float — in our view the single most probable mechanical explanation for the price path (US$4 → US$7 → US$0.34). We have no trading records showing which parties actually sold, and make no claim that any particular holder did.

Ongoing related-party channels. Wisecome Oil Pte. Ltd., also wholly-owned by Mr. Xie, appears throughout the related-party notes (freight, office rental, a property disposal, an interest-free loan) — individually small; collectively showing, in our view, a group whose operating fabric runs through the CEO's private entities in both directions.

---

Financial Forensics

Scale without economics. Revenue ~US$300M/yr at 0.8–1.4% gross margin. At a 0.8% gross margin, the FY2025 book generates ~US$2.5M of gross profit against ~US$7.1M of G&A — in our view the business as currently configured cannot cover its own public-company cost base. FY2025: net loss US$4.46M, operating cash flow −US$5.2M, year-end cash US$1.79M, going-concern discussion in the annual report.

Receivables: the canonical fingerprint. Accounts receivable grew ~24× in FY2024 (revenue +9%) and a further 28% in FY2025 (revenue −2.3%); turnover days more than doubled; the company holds no collateral, charges no interest, and records no allowance.

The mirror pattern. From the FY2025 20-F concentration tables: FY2024 — Customer I ≈ 100% of AR (US$17.5M) against Supplier I ≈ 100% of AP (US$17.1M); FY2025 — Customer L ≈ 91% of AR (US$20.4M) against Supplier L ≈ 100% of AP (US$20.2M). Two consecutive year-ends, two different letter-pairs, same geometry — a pattern that appears consistent with back-to-back pass-through trades with matched counterparties, concentrating the company's entire balance-sheet risk in one anonymized counterparty pair per year. Directly beneath these tables, the notes state: "The Company does not have any significant concentrations of risk related to major customers" — while the risk-factor section of the same document states top-three customer concentration of 58.3%. Both cannot be right. The counterparties are anonymized in the filings; their identity, and whether any relationship exists between each customer/supplier pair, is unknown and unverifiable from public records — we state that explicitly rather than infer it.

---

Post-IPO Governance Timeline

Date	Event
Jul 9–10, 2025	IPO priced/closed; US$3.1M net to company; US$2.6M to insider sellers
Dec 30, 2025	H1-2025 results published — 6 months after period-end
Dec 31, 2025	CFO Tianshu Chu resigns (as CFO and director), ~6 months post-IPO; new CFO from logistics sector
Feb 23, 2026	EGM passes: dual-class (all 9,176,000 Mega Origin shares → Class B @ 50 votes/share); 2026 Equity Incentive Plan; reverse-split pre-authorization, 1:2 up to 1:500, board's sole discretion
Mar 11, 2026	Stock closes below US$1.00 — start of the 30-day deficiency clock
Apr 23, 2026	Nasdaq minimum-bid-price deficiency letter; cure by October 20, 2026
Apr 30, 2026	FY2025 20-F filed: net loss US$4.46M; going-concern discussion

The EGM vote counts (~11.36M FOR on every proposal) approximate the combined Mega Origin + Novel Majestic holdings — public records indicate the two insider vehicles alone carried every proposal. The 50:1 dual-class conversion eight months post-IPO moved Mr. Xie from 56% voting to ~98.5% voting; combined with 450M authorized Class A and a fresh equity incentive plan, future dilution of Class A holders carries no control cost to the insider.

---

What We Did Not Find (Thesis-Tempering)

Reported in full because it cuts against the short archetype:

• No auditor churn: Onestop Assurance PAC (Singapore, PCAOB-registered) across FY2023–FY2025; the FY2025 20-F was filed on time; no NT 20-F, no restatement.
• No litigation or enforcement involving the company, its subsidiaries, or named principals was located in the sources searched.
• No direct evidence of promotion targeting DLXY — no named Telegram/WhatsApp/Discord campaigns, no stock-promoter coverage, no paid-newsletter trail — though on ~24,000 shares/day of volume the float is thin enough that modest coordinated flow could produce both legs of the US$4 → US$7 → US$0.34 path.
• A genuine, 18-year-old Singapore operating business with real revenue scale — this is not a shell in the classic sense.

---

Catalysts

Catalyst	Window	Mechanics
Reverse split	Before ~Aug 22, 2026 (EGM authority expires) or renewed; Nasdaq cure by Oct 20, 2026	Board pre-authorized up to 1:500
Dilutive raise	Any time post-split	US$1.79M cash, negative OCF, US$4.0M shareholder loan maturing Dec 31, 2026; F-3 shelf becomes available July 2026 — i.e., now
Receivables event	FY2026 audit cycle	US$20.4M owed by one anonymized counterparty vs US$7.4M market cap and no allowance
Going-concern escalation	FY2026 20-F (Apr 2027)	A second loss year with sub-US$2M cash would, in our view, make an emphasis-of-matter paragraph difficult to avoid
Shareholder-loan maturity	Dec 31, 2026	Repayment would consume >2× the company's cash
Disclosure vacuum	Now → Apr 2027	As an FPI with no interim obligation, the market will reprice on corporate actions, not results

---

Why This Is Not Rated

At US$0.45 versus a US$4.00 IPO, ~89% of the short outcome is already realized. This report documents the anatomy; it does not identify an attractive entry. Borrow is likely unavailable or punitive (~24k shares/day of volume, ~2M-share genuine float; no options), and post-reverse-split share counts in the low millions under 98.5% insider voting control can produce violent upward dislocations unrelated to value. The bull/benign reading also exists and is stated in full in the PDF: a real 18-year-old trading business, stable auditor, on-time filings, back-to-back matched trades as a normal structure in physical commodity intermediation, and lawful, disclosed pre-IPO dividends. For most participants the actionable output of this report is avoidance — of the equity, of any future post-split promotion cycle, and of any dilutive re-offering — not a short position.

---

Conclusion

Applying our forensic checklist to Delixy Holdings produces a clean instance of the small-float Asia-based Nasdaq IPO extraction archetype — not through any single smoking gun, but through the checklist's signature accumulation: capital flowing predominantly out to insiders across the IPO event (~US$15.3M out vs ~US$3.1M in); a lock-up-exempt, concurrently-registered insider supply pool ~4× the primary float, assembled by personal share transfers three months before filing; a trading book whose entire balance-sheet exposure sits with one anonymized counterparty pair per year, in mirror-image amounts, beneath a notes-level assertion that no concentration risk exists; post-IPO governance that consolidated 98.5% voting control, pre-armed a 1:500 reverse split before the deficiency letter, and swapped the CFO within six months; and a disclosure regime (FPI, controlled company, EGC) that minimizes every ongoing transparency obligation.

Set against that: no litigation, no auditor churn, no late filings, no promotion evidence, and a genuinely old operating business. In our view the record supports the conclusion that public investors funded an exit, not an expansion — as a pattern judgment about disclosed economics, not an allegation of unlawful conduct. The checklist item that would have caught this pre-IPO, in one line: compare pre-IPO dividends to intended net proceeds (US$12.7M out vs US$3.1M in) — when the owners take out 4× what the public puts in, the offering's purpose is, in our view, self-describing.

Full analysis, source citations (SEC EDGAR, CIK 0002025218), and methodology in the PDF.