HEJ: mid-cycle EV/EBITDA implies modest upside but execution and liquidity risks weigh
Intrinsic value VND 12,339 vs market VND 11,000 — implied upside 12.2% (model confidence: low).
Business Overview
Tổng Công ty Tư vấn Xây dựng Thủy Lợi Việt Nam - CTCP (HEJ) is a UPCOM‑listed construction and engineering firm focused on hydraulic and civil works. The company reported revenue of VND 80.3 bn in 2025 after VND 68.6 bn in 2024 and VND 97.1 bn in 2023, reflecting a volatile top line typical of project-driven contractors. HEJ’s margins are mixed: a gross profit margin of 33.9% alongside a weak EBIT margin of 4.6% in the latest period, implying meaningful cost or operating-pressure below the gross line.
Investment Thesis
HEJ’s valuation using a mid‑cycle EV/EBITDA produces an intrinsic value of VND 12,339 per share, implying 12.2% upside to the current match price of VND 11,000. The model uses a mid‑cycle EBITDA of VND 6,760,163,418 and a fair EV/EBITDA multiple of 12.69 (own history), producing a fair‑value EV/EBITDA slightly above the sector median of 9.85. Net debt of VND 28,997,491,594 (model input) drives enterprise value adjustments.
Positives: HEJ’s EV/EBITDA of 11.4 is close to the model fair multiple, so the implied valuation is not detached from publicly observable earnings multiples. The firm’s P/B is 0.8 and BVPS is VND 13,782 per share, so market price sits below book, offering a margin of safety if asset values are realizable. Our historical model track record shows a high hit rate (88.9%) and large average realized upside (79.6%), which supports relying on systematic valuation bands — but see confidence caveat below.
Negatives / execution risk: HEJ reported a net loss of VND 13.9 bn in 2025 after modest profits in prior years (VND 0.4 bn in 2024, VND 1.4 bn in 2023), and revenue has been volatile (2023–2025: VND 97.1 bn → VND 68.6 bn → VND 80.3 bn). Profitability metrics are weak: ROE is -20.5%, ROA -6.6%, and net profit margin -16.8%, indicating losses despite a healthy gross margin. Debt leverage is material with Debt/Equity of 1.94 and reported net debt in the valuation inputs, increasing refinancing and covenant risk. Trading liquidity is thin (avg volume 149 shares over 2 weeks) and UPCOM illiquidity is flagged in the model’s sanity flags, which raises execution risk for larger flows.
Valuation upside of 12.2% is within a modest band: it does not sufficiently compensate for the combination of execution risk, recent loss generation, leverage, and low model confidence. Given the model’s stated confidence is low, we downgrade conviction relative to the raw upside figure.
Valuation Commentary
Mid‑cycle EV/EBITDA: we apply a fair EV/EBITDA multiple to a mid‑cycle EBITDA then subtract net debt to derive equity value per share.
- Mid‑cycle EBITDA: VND 6,760,163,418 (model input, own median over 7 years).
- Fair EV/EBITDA multiple: 12.69 (derived from firm history; sector median 9.85).
- Net debt: VND 28,997,491,594 reduces enterprise to equity value.
- Shares outstanding: 4,400,000 shares used to derive per‑share intrinsic value.
- Calibration: isotonic calibration reduced raw intrinsic VND 12,899.4 to reported VND 12,339; model confidence labelled low and illiquidity flagged.
The model implies VND 12,339 per share (12.2% upside). Confidence is low and UPCOM illiquidity is flagged, so the numerical upside should be treated cautiously: it is not large enough to overcome execution, profitability and leverage risks in our view. The valuation sits slightly above sector median EV/EBITDA, reflecting firm‑specific assumptions rather than clear market premium.
Bull vs Bear
- Valuation close to peers: EV/EBITDA 11.4 vs model fair 12.69 and sector median 9.85 — potential re‑rating if EBITDA stabilizes around mid‑cycle levels.
- Book value cushion: P/B 0.8 with BVPS VND 13,782 provides downside protection if assets and working capital convert at or near balance sheet values.
- High historical model hit rate (88.9%) suggests the model framework has been effective for this company/universe historically.
- Profitability deterioration: net loss of VND 13.9 bn in 2025 (after small profits in 2023–24) with net margin -16.8% and ROE -20.5% — earnings recovery is uncertain.
- Leverage and cash risk: Debt/Equity 1.94 and model net debt VND 28,997,491,594 increase refinancing and covenant pressure, especially if project receipts slip.
- Market and liquidity constraints: UPCOM listing and low average volume (149 shares/2w) — selling pressure could move price materially and foreign ownership room is unknown.
- Concentrated ownership: largest shareholder holds 32.75%, which can limit free float and corporate governance scrutiny; minority liquidity and exit risk rise as a result.
Sector Context
Construction and building materials in Vietnam are cyclical and project‑driven; revenue recognition and timing under VAS can produce lumpy top‑line and margin patterns for contractors (work‑in‑progress and progress billing norms). Banks’ lending to construction is sensitive to SBV macroprudential guidance and any credit growth quotas, which can tighten working‑capital financing for contractors carrying high receivables. Peer universe median implied upside is 9.6%; HEJ’s 12.2% sits slightly above that but with lower confidence. For construction companies with land or real‑estate exposure, land use rights valuation and payment timing are frequent forensic levers; HEJ’s public filings do not disclose material land‑bank signals in the dataset provided. UPCOM illiquidity and concentrated ownership are common in the segment and often compress attainable market prices relative to theoretical intrinsic values.
Risk Factors
- Profitability risk: net loss in 2025 (VND -13.9 bn) after thin profits prior years — continued losses could erode equity and trigger covenant breaches.
- Refinancing and leverage: Debt/Equity 1.94 plus model net debt VND 28,997,491,594 — rising rates or tighter bank lending could increase interest burden or restrict working capital.
- Liquidity / marketability: trading on UPCOM with avg volume ~149 shares (2w) and model sanity flag 'illiquid' — price may gap on modest flows and large shareholders control ~32.8%.
- Model / valuation confidence: model confidence labelled low; isotonic calibration reduced raw intrinsic value and inputs rely on own_history fair multiple.
- Revenue concentration and project timing: project‑based business breeds lumpy cash flows and revenue volatility (2023–25 revenue range VND 68.6–97.1 bn).
- Corporate governance / shareholder concentration: largest owner 32.75% may influence strategic decisions and related‑party contracting risks exist inherently in VAS‑accounted companies.
Catalysts
- Stabilization in profitability: positive net profit and margin improvement in a reported quarter/year would materially support the mid‑cycle EBITDA assumption.
- Debt reduction or re‑financing at favorable terms that reduces net debt materially below the model input of VND 28,997,491,594.
- Visibility on new contract wins or backlog conversion that raises mid‑cycle EBITDA above the current model input of VND 6,760,163,418.
- Improved liquidity or a listing upgrade that reduces the UPCOM illiquidity discount and increases free float.
Forensic Assessment
No Beneish M‑Score or other forensic metric is available in the input (mscore: null). The dataset shows no explicit forensic red flags; however, the combination of lumpy revenue recognition common under VAS, concentrated ownership and small UPCOM free float merits scrutiny of related‑party contracting, receivable ageing and WIP recognition in disclosures. Earnings quality is moderate at 61.3/100, suggesting reported earnings have some support from cash/operating items but are not pristine.
Track Record
The valuation model’s historical performance for this coverage universe is strong on paper: a 10‑year hit rate of 88.9% and average upside of 79.6%. That said, past model success does not eliminate company‑specific operational and liquidity risks; the model itself flags low confidence for this specific estimate, so apply historical track record as supportive context rather than definitive proof.
Written by a language model on 2026-08-10 from this page’s own model outputs and financial statements, and may quote figures from that date. Descriptive analysis, not investment advice — no buy, sell or hold recommendation is given or implied.