TSJ: mid-cycle EV/EBITDA implies limited upside; execution and liquidity risks offset cyclical recovery potential
Intrinsic value VND 25,893 vs market VND 21,000 — implied upside 23.3% (model confidence: low).
Business Overview
Công ty Cổ phần Du lịch Dịch vụ Hà Nội (TSJ) is a UPCOM-listed tourism & leisure company operating in the Vietnamese cyclical Du lịch & Giải trí sector. Revenue has been relatively stable over the past three years at VND 140–144 bn (VND 140.7 bn in 2023; VND 144.2 bn in 2024; VND 139.4 bn in 2025) while net profit recovered to VND 100.1 bn in 2025 from VND 75.4 bn in 2023. The shareholder base is highly concentrated: Tổng Công ty Du Lịch Hà Nội (an SOE) holds 45.2% and another institution holds 39.97%, giving the top two owners ~85.2% combined. The major SOE stake brings policy considerations (e.g., dividend/payout expectations and potential preferential access to state-related tourism projects).
Investment Thesis
TSJ's valuation rests on a mid-cycle EV/EBITDA approach with a fair EV/EBITDA of 150.39 derived from the company's own history, producing an intrinsic value of VND 25,893 per share — an upside of 23.3% to the current match price of VND 21,000. Strengths include low reported leverage (Debt/Equity 3.7%) and improving profitability: net profit rose to VND 100.1 bn in 2025 and EBIT margin is 15.5%. The company also shows an accounting net cash position in the model inputs (net_debt: negative VND 11,870,606,807), which supports enterprise-value driven valuation.
However, the valuation carries meaningful execution and data-quality caveats. The model confidence is low (calibrated via isotonic mapping), and sanity flags list illiquidity and mediocre earnings quality. The company's EV/EBITDA on market ratios is 63.4x, well above the sector median EV/EBITDA of 9.14x, signaling either capitalised non-operating assets or episodic earnings; the model instead uses an elevated historical fair EV/EBITDA (150.39) to justify the intrinsic value. Trading liquidity is thin (avg volume 2w = 261 shares) and the stock is listed on UPCOM, which raises the execution risk for larger institutional flows. Given concentrated ownership (top two >85%) and an SOE anchor, minority liquidity and governance signals deserve scrutiny.
Putting these together: the implied 23.3% upside sits in a zone where the reward may not fully compensate for low model confidence, illiquidity and earnings-quality concerns. The company could outperform if tourism demand strengthens further and management converts asset base into recurring EBITDA, but downside is material if earnings normalise or if minority holders cannot realise value due to low free float.
Valuation Commentary
We use a mid-cycle EV/EBITDA model anchored to the company's own historical fair EV/EBITDA (150.39) applied to a mid-cycle EBITDA (VND 15,583,838,359) then adjusted for net debt to derive per-share intrinsic value.
- Mid-cycle EBITDA: VND 15,583,838,359 (model input: mid_cycle_ebitda).
- Company historic fair EV/EBITDA: 150.39 (source: own_history).
- Net cash position in model: net_debt = negative VND 11,870,606,807 (reduces enterprise value).
- Sanity flags: illiquid trading and mediocre earnings quality (affecting confidence).
- Calibration: isotonic recalibration moved raw intrinsic value from VND 31,491.7 to VND 25,893 reflecting model weakness.
The 23.3% implied upside reflects a company-specific valuation that relies on an elevated historic EV/EBITDA multiple and a small net cash adjustment. Model confidence is low, so the intrinsic value should be treated as directional rather than precise. Liquidity and earnings-quality concerns reduce conviction; any price target sensitivity should stress lower EBITDA and a reversion to sector EV/EBITDA.
Bull vs Bear
- Tourism rebound drives above-cycle EBITDA: mid-cycle EBITDA input is VND 15,583,838,359; a stronger demand upswing would lift actual EBITDA above this base and materially increase equity value.
- Net cash position reduces enterprise value: model net_debt is negative VND 11,870,606,807 which supports per-share intrinsic value.
- Low leverage (Debt/Equity 3.7%) provides balance-sheet flexibility for investments or shareholder distributions if governance permits.
- SOE shareholder (45.2%) may enable preferential access to state-linked tourism projects, supporting revenue stability.
- Earnings-quality concerns: model flags 'mediocre_earnings_quality' and earnings_quality score is 45.1, increasing the risk that reported EBITDA is volatile or contains non-recurring items.
- Illiquid stock: average 2-week volume is only 261 shares and listing on UPCOM impairs ability of large investors to enter/exit positions.
- Valuation mismatch vs sector: market EV/EBITDA is 63.4x while sector EV/EBITDA is 9.14x, suggesting the observed multiples may reflect accounting/one-off items rather than sustainable operations.
- Highly concentrated ownership: top two shareholders hold ~85.2% combined (45.2% + 39.97%), limiting free float and raising minority investor governance risk.
Sector Context
The Vietnamese tourism & leisure sector is cyclical and sensitive to domestic and inbound travel flows; macro shocks or policy changes (e.g., visa regimes, air connectivity) materially affect near-term earnings. Accounting under VAS can recognise land use rights, revaluation gains or other non-cash items differently from IFRS peers, which may inflate asset values and create dispersion in EV/EBITDA comparisons. Many tourism firms are small-cap and struggle with liquidity on UPCOM/HOSE; TSJ's trading profile (1y high VND 49,300; 1y low VND 18,000) shows episodic price swings. Policy angles matter: SOE shareholders often face payout mandates and strategic objectives that can influence capital allocation and minority returns. Peers show a wide range of outcomes — the sector median upside is ~5.6%, while top peers can show >40% upside, highlighting cross-sectional dispersion driven by scale, assets and earnings quality.
Risk Factors
- Illiquidity risk: avg volume 2w = 261 shares; large orders would move price materially.
- Earnings-quality risk: earnings_quality = 45.1 and model sanity flagged 'mediocre_earnings_quality', raising the chance of episodic or non-recurring EBITDA.
- Ownership concentration: top two shareholders combine for ~85.2% of shares, limiting free float and increasing governance/expropriation risk for minorities.
- Valuation multiple reversion: the model depends on an elevated historical fair EV/EBITDA (150.39); reversion toward sector EV/EBITDA (9.14) would compress value significantly.
- Cyclical demand exposure: tourism revenues fell in the 3-year window (2023–2025 revenue moved from VND 140.7 bn → VND 139.4 bn) and remain sensitive to macro and travel patterns.
- UPCOM listing and disclosure: secondary listing venue and thin trading can delay price discovery and amplify volatility for corporate actions.
Catalysts
- Stronger-than-expected tourism rebound (domestic + inbound) that drives EBITDA above the modelled mid-cycle level.
- Corporate actions improving liquidity or governance (partial free-float increase or listing transfer) which would materially raise free float.
- Publication of audited statements or disclosures addressing earnings-quality concerns.
- SOE-driven project awards or contracts that expand recurring revenue streams.
Forensic Assessment
M-Score is not available (null). Given the lack of M-Score, the primary forensic concerns are earnings quality and ownership concentration: earnings_quality at 45.1 and model sanity flags indicate mediocre reported earnings quality, which is the headline forensic concern. No explicit red flags or positive signals are present in the provided forensic array, so the assessment defaults to caution on reported profitability and on the thin free float given the dominant SOE/institutional stakes.
Track Record
Model track record spans 10 years (2017–2026) with a hit rate of 66.7% but a low average upside of 1.4% historically. The hit rate is acceptable but not exceptional; the low average upside implies the model often produces modest price moves in practice. Given the current model confidence is 'low' and the model was recalibrated (isotonic), past performance supports using the output as a directional input rather than a high-conviction price target.
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.