Equity research memo on one listed Indian company

C1 · Equity research memo on one listed Indian company#

Objective#

Produce the memo a careful fundamental investor would write before committing capital to one NSE- or BSE-listed company of your choice: what the business is, what it earns on the capital it uses, what it is worth under explicit assumptions, what the current price already assumes, what could be wrong with the accounts, and how the answer changes across scenarios. The output is a decision framework, not a recommendation. You must be able to defend every number from a filing, a dataset or a stated assumption.

Choose a non-financial company (Beneish and the FCFF DCF of M06 are not valid for banks and NBFCs; if you choose a lender, you must use the bank-valuation approach of M06A and replace the forensic section accordingly). Prefer a company with at least ten years of listed history and an annual report you can read in full.

Deliverables#

1. The memo (4,000–7,000 words, markdown or PDF) following the outline below.
2. A model workbook or Python notebook that reproduces every number in the memo from raw inputs: ten years of statements, the DCF, the reverse DCF, the forensic scores and the scenario table. Inputs on one sheet or in one cell block, with the source and date of each.
3. A one-page summary: thesis in three sentences, value range, implied expectations, the three things that would change your mind, and a pre-committed review date.
4. A source log: every annual report, results release, shareholding filing and dataset used, with URL and access date.

Time budget and minimum viable scope#

Done in full (ten-year reconstruction from annual reports, all twelve analyses, a 4,000–7,000-word memo), this capstone takes realistically 25–40 hours. The full-track plan budgets about 24 hours for a minimum viable memo; the practitioner and core tracks budget about 4 hours to start it (choice of company, filings, first reconstruction and reverse DCF) and leave the rest for after the course. The minimum viable memo keeps the analyses that carry the decision:

• five (not ten) years of NOPAT, invested capital, ROIC, reinvestment and FCFF, reconciled to reported cash flow (analyses 2–3);
• cost of capital with a range, a two-stage DCF with one sensitivity table, and the reverse DCF (analyses 4–6);
• Beneish and Piotroski for three years and the red-flag checklist (analysis 8);
• three scenarios with probabilities and the margin of safety as a probability (analysis 10), and the index baseline (analysis 11);
• a 2,000–3,500-word memo, the one-page summary and the source log.

Analyses 7, 9 and 12 and the ten-year history are the extension to the full scope; acceptance criteria 2, 5 and 6 then apply to the years actually covered.

Required analyses (with the methods from the modules)#

1. Reading the company from the outside (M06 · 6.1). Segment revenue and profit, customers and suppliers, capital intensity, promoter holding and pledge history, related-party transactions, auditor and any qualifications or emphasis-of-matter paragraphs. Shareholding pattern trend over at least eight quarters (M06A on promoter and insider disclosures).
2. The three statements as one system (M06 · 6.2, Propositions 6.1–6.2). Rebuild ten years of NOPAT, net operating assets, reinvestment and free cash flow to the firm. Reconcile reported cash flow from operations to your FCFF; explain every year where CFO/PAT is below 0.8.
3. Growth, reinvestment and return on capital (M06 · 6.4, Propositions 6.5–6.7). Compute ROIC and the reinvestment rate each year; check $g \approx$ reinvestment rate × RONIC against realised growth. State whether growth has created value (ROIC vs WACC).
4. Cost of capital (M06 · 6.5). Derive the rupee risk-free rate from the 10-year G-sec (state the date and the source, RBI/CCIL), the equity risk premium you use with its source (for example Damodaran's India ERP, dated), beta estimated yourself against the Nifty 50 or Nifty 500 on weekly returns over 2–5 years (report the standard error), and the WACC. Show the WACC range you consider defensible.
5. Two-stage DCF (M06 · 6.3, equation for $V_0$ in 6.6). Explicit forecast of at least five years and a terminal value with $g_T$ below the long-run nominal GDP growth you assume. Report the share of value in the terminal value. A two-way sensitivity table (WACC × terminal growth, and WACC × RONIC in the terminal period).
6. Reverse DCF (M06 · 6.6, Proposition 6.10). Solve for the growth rate implied by the current market price under at least three WACCs and two terminal RONICs. Compare implied growth with the company's own ten-year record and with industry growth. This is the centre of the memo.
7. Convexity of value in the discount rate (M06 Proposition 6.11). Show numerically that averaging values across a WACC distribution gives more than valuing at the average WACC, and say why this matters for your margin of safety.
8. Forensic accounting (M06 · 6.8). Beneish M-score for each of the last five years with all eight indices shown (flag if $M > -1.78$); Piotroski F-score with the nine signals shown; the India red-flag checklist of 6.8 (pledging, related parties, auditor changes, contingent liabilities, capitalised expenses, receivable days, cash-and-debt both high). A paragraph per flag that fires.
9. Quality framework (M06 · 6.9). Apply the Coffee Can filter exactly as defined (revenue growth and ROCE thresholds every year), and one qualitative framework (moat sources, Fisher, Kedia's SMILE or Prasad's "avoid big risks"). State which criteria fail.
10. Scenario valuation and margin of safety (M06 · 6.7; M12 · 12.5 for scenario trees). At least three scenarios (bear, base, bull) with explicit, internally consistent drivers (volume, price, margin, reinvestment, RONIC, WACC), a probability for each and the reasoning behind the probabilities. Report the probability-weighted value, the distribution of value, and the margin of safety as a statistical statement: the probability, under your scenario distribution, that value is below the current price.
11. Index baseline (M06 · 6.11). What return does the Nifty 50 or a relevant index fund offer under the same assumptions? Your thesis must beat this baseline after costs and taxes (M01 · 1.8) to justify a concentrated position.
12. Sizing logic (M03, M14 · 14.1–14.4, 14.13). If you were to hold the stock, what weight would your scenario distribution and a fractional-Kelly or risk-budget argument support, and what is its contribution to portfolio ES? This is an exercise in method, not an instruction to trade.

Data sources#

• Annual reports and results: company investor-relations pages; NSE and BSE corporate filings (financial results, shareholding pattern, pledge disclosures under SAST, related-party transaction filings). Read at least the last three annual reports in full, including the notes and the auditor's report.
• Ten-year financials: build them from the annual reports and the results filed with NSE and BSE; document any reclassification between years. The annual report is the source of truth.
• Choosing the company: a free screener such as https://oicharts.com/screener is enough to narrow a universe; the company's page (e.g. https://oicharts.com/stocks/TITAN) shows the current price and, for an F&O stock, its derivatives positioning.
• Prices: NSE bhavcopy (equity) or Yahoo Finance (.NS tickers) for adjusted prices; adjust for splits and bonuses (M01 · 1.10).
• Rates and ERP: RBI DBIE or CCIL for the 10-year G-sec; Damodaran's country risk data (dated) for the ERP and default spread; FRED for US comparisons.
• Industry data: industry associations, regulator reports, competitor filings. Cite each with date.

Acceptance criteria#

1. Every number in the memo is reproducible from the workbook; a reviewer changing one input (say WACC) sees every dependent number update.
2. The FCFF reconciliation explains at least 90% of the difference between ten-year cumulative PAT and cumulative FCFF.
3. Beta is reported with its standard error and sample; WACC with a range, not a point.
4. The terminal value share of DCF value is reported; if it exceeds 75%, the memo discusses why the explicit period is too short or the terminal assumptions too generous.
5. The reverse DCF gives a unique implied growth for each WACC (Proposition 6.10) and the memo compares it with at least one base rate: the distribution of ten-year revenue or earnings growth for comparable Indian companies (for instance, the fraction of Nifty 500 companies that sustained that growth), with source.
6. Beneish and Piotroski are computed for at least five years with every component visible; any flag has a written explanation.
7. Scenario probabilities sum to 1 and each scenario's drivers are mutually consistent (growth funded by reinvestment at the stated RONIC).
8. Every changeable fact (prices, rates, holdings, lot sizes, tax rates) carries an as-of date and a source.
9. No price targets presented as forecasts and no recommendation language; the conclusion is stated as "at price P, the market implies X; my distribution gives Y; the decision rule is Z."

Template outline for the write-up#

1. Summary (one page, the deliverable 3 content).
2. The business: what it sells, to whom, with what capital; segment table.
3. Industry structure and competitive position; moat evidence and its absence.
4. Ten-year financial history: NOPAT, ROIC, reinvestment, FCFF, CFO/PAT; what changed and why.
5. Capital allocation and governance: dividends, buybacks, acquisitions, related parties, promoter behaviour.
6. Forensic review: Beneish, Piotroski, checklist; findings.
7. Cost of capital.
8. DCF valuation and sensitivities.
9. Reverse DCF: what the price implies and how plausible it is against base rates.
10. Scenarios, probabilities and the value distribution; margin of safety.
11. Comparison with the index baseline.
12. Risks, what would change my mind, and the pre-committed review triggers.
13. Limitations of this analysis.
14. Appendix: source log, model description, data checks.

Common failure modes#

• Anchoring on a forward DCF. Building the DCF first and then defending it. Do the reverse DCF first and ask whether you believe what the price implies.
• Terminal value doing all the work. A 90% terminal share with $g_T$ close to WACC means the valuation is a guess about the denominator.
• Growth without reinvestment. Forecasting 15% growth with a falling reinvestment rate implies a RONIC the company has never earned. Check Proposition 6.5 every year of the forecast.
• Beta from a short or illiquid sample. Daily beta on an illiquid mid cap is biased towards zero (non-synchronous trading); use weekly returns and report the error.
• Taking screener exports on trust. Data-vendor and screener exports reclassify items (other income, exceptional items, lease liabilities). Reconcile two years to the annual report.
• Ignoring the notes. Contingent liabilities, related-party balances and capitalised interest live in the notes, not the summary statements.
• Scenario probabilities chosen to hit a target. Write probabilities before computing the scenario values, and record them in the source log with a timestamp.
• Story over base rates. Very few companies compound revenue at 20% for ten years; the memo must state how rare the implied path is.