reverse-dcf

name	reverse-dcf
description	Reverse DCF valuation tool. Uses market price to infer market-implied growth expectations, replacing the forward DCF "tweak assumptions to fit conclusions" approach. Suitable for mature companies with positive FCFF (A-shares/H-shares/US stocks). Triggered when user says "Reverse DCF", "implied growth rate", "market implied expectations", "price in", "valuation reverse", "use DCF to calculate", or "what is the market pricing in". Not suitable for companies with negative FCFF (use PS implied revenue reverse instead).
version	1.0.0

Reverse DCF -- Reverse-Engineer Market Implied Expectations

Core Concept

Forward DCF: give assumptions -> calculate target price (parameters are arbitrary, becomes "serving the conclusion") Reverse DCF: give market price -> reverse-engineer implied growth rate (translates "expensive/cheap" into a verifiable proposition)

Essential difference:

• Forward DCF outputs "fair value" -- abstract, you can always be "right" but your wallet won't get thicker
• Reverse DCF outputs "market expectations" -- concrete, verifiable numbers that can be compared with reality

Applicability Conditions

Prerequisite: base year FCFF must be positive. Companies with negative FCFF (early high-growth, heavy asset expansion phase) will produce meaningless results with standard two-stage DCF. -> Alternative: use PS implied revenue reverse (see "Special Cases" below)

Data requirements:

• Latest full annual report (income statement, balance sheet, cash flow statement)
• Current stock price / market cap
• Interest-bearing debt details (short-term borrowings, long-term borrowings, lease liabilities, etc.)
• Risk-free rate (10-year government bond yield)
• Industry beta, ERP (Damodaran data recommended)

Practical Steps

Step 1: Determine Enterprise Value (EV)

EV = Market Cap + Net Debt
Net Debt = Interest-bearing Debt - Cash & Equivalents - Trading Financial Assets

Note:

• For "net cash" companies (cash > interest-bearing debt), EV will be smaller than market cap
• A-share companies need to check: whether trading financial assets include wealth management products (count as cash equivalents)
• STAR Market companies often have large amounts of raised funds sitting on the books, EV may be much smaller than market cap

Step 2: Calculate Base Year FCFF (Core, Most Error-Prone)

Recommended Method A: Reverse from CFO (Most Reliable in Practice)

FCFF = CFO - Capex + After-tax Interest Expense

• CFO = Cash flow from operations (audited real cash)
• Capex = Cash paid for fixed assets, intangible assets
• Net cash companies: after-tax interest ≈ 0

Verification Method B: Bottom-up from NOPAT

NOPAT = Operating Profit x (1 - Effective Tax Rate)
FCFF = NOPAT + D&A - Capex - Delta WC

• Effective tax rate ≠ statutory tax rate (e.g., South Korea statutory 24%, SK Hynix actual only 14.9%)
• D&A = Depreciation & Amortization
• Delta WC = Change in working capital (inventory + receivables + prepayments - payables - contract liabilities)
• Cross-validate both methods, take the average

Important: Base year selection determines everything

• Using peak-cycle FCFF -> implied CAGR underestimated
• Using mid-cycle FCFF -> implied CAGR overestimated
• This is a judgment call, not a calculation

Step 3: Build WACC

Re = Rf + beta x ERP
WACC = Re x We + Rd x (1-t) x Wd

Recommended parameter sources:

Parameter	Source
Risk-free Rate Rf	10-year government bond yield of company's home country
Equity Risk Premium ERP	Damodaran annual data (varies by country)
Beta	Damodaran industry unlevered beta -> re-lever by D/E (more stable than single stock regression beta)
Pre-tax Cost of Debt Rd	Average IG corporate bond yield in company's home country
Effective Tax Rate	Company's actual effective tax rate (from financial statements, not statutory rate)

Step 4: Two-Stage DCF Model

EV = Sum [FCFF_0 x (1+g1)^t / (1+WACC)^t] + [FCFF_10 x (1+g2) / (WACC-g2)] / (1+WACC)^10

Where:

• g1 = explicit period (10-year) FCFF annualized growth rate reverse-engineering target
• g2 = terminal growth rate (usually 2.5-3%, ≈ nominal GDP growth)
• FCFF_10 = FCFF_0 x (1+g1)^10 x (1+g2)

Forward DCF: given g1 and g2, calculate EV Reverse DCF: given EV and g2, solve for g1

Step 5: Build Sensitivity Table

              Terminal Growth g2
            1.5%  2.0%  2.5%  3.0%  3.5%
     18%     xxx   xxx   xxx   xxx   xxx
CAGR 20%     xxx   xxx   xxx   xxx   xxx
g1   22%     xxx   xxx   xxx   xxx   xxx
     25%     xxx   xxx   xxx  [1189] xxx  <- Target EV
     28%     xxx   xxx   xxx   xxx   xxx

1. Each cell = EV for the (CAGR, g2) combination
2. Find the cell closest to actual EV
3. The corresponding g1 is the market-implied explicit period CAGR expectation

Key observations:

• Pricing is much more sensitive to explicit period CAGR than terminal g
• Same row (same CAGR) valuation varies little, same column (same terminal g) valuation can differ by 2x
• So the focus of questioning the valuation should be "can the 10-year CAGR be achieved", not "terminal 3% is too aggressive"

Step 6: Translate CAGR into Plain Language

After calculating implied CAGR, do three translations:

Translation 1: Company size in 10 years

FCFF_10 = FCFF_0 x (1+g1)^10
Corresponding revenue approx = FCFF_10 / (current FCFF/revenue conversion rate)

Translation 2: What needs to happen at the industry level

Current TAM -> revenue needed in 10 years -> implied market share change
Or: how many times TAM needs to grow to support it

Translation 3: Compare with history

Company's actual CAGR over past 10 years vs implied CAGR
Industry comparable companies / global tech giant historical CAGR comparison
(TSMC's 20-year CAGR ≈ 18%, already an industry miracle)

Step 7: Honestly Face Limitations

1. Terminal Value (TV) as % of EV is high -- cyclical stocks especially dangerous, TV > 50% is a red flag
   • Alternative: Exit Multiple method (exit at year 10 using industry median EV/EBITDA)
2. Base year FCFF selection determines everything -- peak vs trough cycle differs by multiples
3. D&A and Capex are estimates -- precision limited by public disclosure granularity
4. Reverse DCF tells you what the market expects, not whether the market is right or wrong
   • 25% CAGR may or may not be reasonable, requires independent judgment
   • It's not a conclusion, it's the starting point of a discussion

Special Cases

Companies with Negative FCFF (e.g., Cambricon)

Cannot use standard two-stage DCF reverse. Use PS implied revenue reverse instead:

1. Assume a reasonable terminal PE (e.g., mature stage 30x)
2. Reverse-engineer net profit needed in 10 years
3. Assume a reasonable net profit margin, reverse-engineer revenue needed in 10 years
4. Calculate implied revenue CAGR

Implied Revenue_10 = Market Cap / Terminal PE / Net Profit Margin
Implied CAGR = (Implied Revenue_10 / Current Revenue)^(1/10) - 1

Cyclical Stocks

Recommend using mid-cycle FCFF instead of single-year FCFF as the base year. Alternatively, use Exit Multiple method instead of Gordon Growth Model for terminal value.

Data Source Recommendations

Data	Recommended Acquisition Method
Stock price / market cap	Navigate to Sina Finance or Xueqiu stock page, extract text
Financial data	Navigate to https://emweb.securities.eastmoney.com/PC_HSF10/NewFinanceAnalysis/Index?type=web&code=<stock-code> (income statement/balance sheet/cash flow statement three tabs have complete info)
Interest-bearing debt details	Same as above, balance sheet tab -- short-term borrowings, long-term borrowings, lease liabilities, bonds payable each item
CFO/Capex	Same, cash flow statement tab -- net cash from operations, cash paid for fixed assets
Industry beta/ERP	Damodaran Online (pages.stern.nyu.edu/~adamodar/)
10-year government bond yield	TradingView or central bank websites, navigate to fetch
Historical revenue/CAGR	East Money PC_HSF10 income statement tab, pull 5-10 years of revenue data, calculate manually

Note: All data should be fetched in real-time via browser_use, do not rely on training data. A-share stock codes in emweb format are SH600519 (Shanghai) or SZ000651 (Shenzhen), Hong Kong stocks use HK00700.

Practical Case References

Series of articles (author: monokuro, Manager at a Tokyo Big 4 valuation team):

1. Core case: "What does the market expect from SK Hynix? Reverse DCF gives a scary number" -- Using Reverse DCF, SK Hynix's current stock price implies FCFF CAGR of 25% over the next 10 years, requiring FCFF of 208 trillion KRW in 10 years (9x 2025 revenue), double the actual growth rate of the past 10 years (13%).
2. Method correction: "The previous Reverse DCF of five AI chip giants has three corrections needed" -- Points out practical pitfalls like Exit Multiple replacing GGM, mid-cycle FCFF replacing single-year FCFF, data source corrections, can serve as supplementary reference for the Reverse DCF limitations section.

Methodology core reference: Aswath Damodaran's valuation framework ("Investment Valuation" and annually updated country risk premium data).