EVS: modest PB re-rating potential offset by weak earnings quality and thin profitability
Intrinsic value VND 6,002 vs market VND 5,700 — implied upside 5.3% (model confidence: low).
Business Overview
Công ty Cổ phần Chứng khoán EVS is a Vietnam-listed securities firm on HNX providing brokerage, proprietary trading, and financial services within the 'Dịch vụ tài chính' sector. The company has 164,800,618 shares outstanding and operates in a competitive peer set of 40 listed securities firms used in our cross-sectional PB calibration. EVS's balance-sheet scale is modest relative to national brokers: total assets were VND 2,347.5 bn in 2025 and fee-driven revenue was VND 275.9 bn in 2025.
Investment Thesis
EVS's valuation implies limited upside: our cycle-adjusted PB model yields an intrinsic price of VND 6,002 per share vs the market price of VND 5,700, an upside of 5.3%. The fair PB (0.6063) sits above EVS's current PB (0.4747), indicating some scope for multiple expansion if profitability normalises. However, current operating returns are very low: ROE is 0.6% and three-year average ROE is 1.07%, both far below the peer median ROE of 8.11%. This weak return profile limits conviction in a PB re-rating.
Valuation Commentary
Cycle-adjusted PB: blend of EVS's own median PB and peer median PB adjusted for peer ROE, calibrated with isotonic mapping on five years of data.
- Fair PB input: 0.6063 versus current PB: 0.4747
- Current BVPS: VND 12,007 per share and current PB imply market price ~VND 5,700
- Peer median ROE: 8.11% and ROE-adjusted peer PB: 0.7314 shift the blended fair value upward
- Model raw intrinsic value before calibration: VND 7,280.4; calibrated intrinsic value: VND 6,002
- Sanity flag: low earnings quality (score 11.3) reduces model confidence to 'low'
The implied upside of 5.3% is small and, given the model's low confidence and flagged low earnings quality, provides limited margin of safety. We place low conviction on the valuation — a positive re-rating would require sustained ROE improvement and clearer earnings quality.
Bull vs Bear
- Valuation gap: fair PB (0.6063) exceeds current PB (0.4747), supporting potential multiple expansion to the intrinsic value of VND 6,002.
- Relatively low leverage: Debt/Equity at 0.1864 gives balance-sheet flexibility for market-making or repo-driven revenue expansion.
- Large foreign room: 163,671,117 shares available to foreign investors could attract incremental demand if governance or results improve.
- Very low and falling profitability: ROE 0.6% and three-year avg ROE 1.07% imply limited ability to generate shareholder value versus peers (peer median ROE 8.11%).
- Deteriorating net income trend: net profit declined from VND 33.3 bn in 2023 to VND 12.1 bn in 2025.
- Weak earnings quality (score 11.3) flagged in model inputs reduces reliability of reported profits and our valuation.
- Rich P/E relative to returns: P/E 77.7 despite ROE 0.6% suggests price is sensitive to any downward earnings surprise; EV/EBITDA is also elevated at 33.5.
Sector Context
Vietnam's securities sector is cyclical and sensitive to market turnover, SBV liquidity and broker margining rules. Many listed brokers show volatile earnings tied to trading volumes and proprietary positions; peer median PB and ROE can swing materially across cycles. Regulatory factors — including SBV credit growth quotas (affecting bank-client flows), state ownership mandates for SOEs, and the presence of VAMC bonds on bank balance sheets — indirectly influence investor appetite for financial-service equities. For brokerage firms, VAS accounting and classification of trading securities, margin loans and client cash can create differences in reported equity and ROE versus regional peers. EVS sits in a crowded peer set of 40 firms; the sector median model upside is slightly negative (-1.1%), signaling a broadly challenged sector valuation environment.
Risk Factors
- Earnings-quality risk: model flagged 'low_earnings_quality' and earnings_quality score is 11.3; reported profit figures may be less reliable.
- Profitability shortfall: ROE 0.6% and net profit fell from VND 33.3 bn (2023) to VND 12.1 bn (2025), reducing internal capacity for dividends or buybacks.
- Valuation sensitivity: P/E 77.7 and EV/EBITDA 33.5 imply small earnings moves materially change implied value.
- Execution risk in revenue recovery: brokerage revenue declined to VND 275.9 bn in 2025 after VND 383.0 bn in 2023; recovery depends on market turnover and client flows which are volatile.
- Ownership concentration and governance: top five shareholders own ~29.5% combined (largest 9.71%), which can concentrate decision-making and may affect minority-holder outcomes.
- Liquidity and market risk: 1-year high/low range VND 9,300 / VND 4,400 indicates high price volatility; average 2-week volume 398,836 shares can limit large-block execution without market impact.
Catalysts
- Sustained recovery in trading turnover that lifts brokerage fees and proprietary trading results (revenue upside vs 2025's VND 275.9 bn).
- Clear improvement in ROE over one-to-two years toward peer levels (peer median ROE 8.11%).
- Corporate actions that improve earnings quality or capital returns (dividend or buyback announcements).
- Increased foreign buying utilising the large foreign room (163,671,117 shares) following governance/signalling improvements.
Forensic Assessment
M-Score is not available; there are no explicit forensic red flags flagged in the forensic summary. However, the model raised a 'low_earnings_quality' sanity flag and the earnings_quality score is low at 11.3, which is the primary forensic concern. Given this, reported profits (net profit VND 12.1 bn in 2025) should be interpreted cautiously and confirmed via notes on one-off items, related-party transactions, and classification of trading gains.
Track Record
The model's historical record spans nine years (2018–2026) with a hit rate of 50%, which is mediocre. Average historical upside for positive calls is very large (avg_upside_pct 206.2%), but this is skewed by outliers; the 50% directional hit rate suggests limited consistency, so past model signals should be treated with caution rather than high conviction.
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.