Case 029DCF and intrinsic valueCore

A junior's DCF of Pravolt Batteries puts over 90% of value in the terminal value, which implies an exit multiple of 28x EBITDA against peers at 14x. What is wrong, and how do you fix it?

The situation

Pravolt Batteries makes lead-acid and lithium batteries for two-wheelers and inverters. A junior analyst's DCF forecasts five years: revenue of Rs 1,000 crore growing 12% a year, EBITDA margin 15%, depreciation 2.2% of revenue, capex 3.5% of revenue, working capital at 8% of each year's new revenue, tax 25% and a discount rate of 12%. The five years of free cash flow are worth Rs 372 crore today.

For the terminal value the junior grows year-5 after-tax operating profit at 9.5% forever, with capex equal to depreciation, and divides by 12% less 9.5%. The model gives an enterprise value of Rs 4,577 crore. Listed battery peers trade at about 14x current EBITDA.

Your task

What is wrong with the terminal value, how do you show it quickly, and what does a corrected DCF give?

Quick check

Which single check exposes the problem fastest?

Worked solution

Try it on paper, then open one step at a time.

30-second answerThe answer to give first

The terminal value implies 28x EBITDA because it grows at 9.5% forever against a 12% discount rate while reinvesting nothing. Growth that close to the discount rate makes the denominator tiny, and growth with no capex is free growth. Using 6% growth and reinvesting enough to fund it gives a terminal value of about 7.9x EBITDA and an enterprise value near Rs 1,559 crore, with 76% in the terminal value.

Step 1Why translate a terminal value into a multiple at all?

If a broker values your flat at Rs 3 crore, you check it against what similar flats sold for last month before you believe it. A terminal value is the same kind of claim: the price someone would pay for the business at the end of year 5. Dividing the terminal value by year-5 EBITDA turns an abstract growth formula into a number you can compare with the market, and a number far above peers is a red flag before any other check. Here Rs 7,410 crore over Rs 264 crore is 28x, and it carries 92% of the value.

Translate the terminal value into the multiple it implies, Rs crorePV of years 1 to 5: Rs 372 crore, the same in bothTV 4,20592% of valueEV 4,577Junior's DCFTV 1,18776% of valueEV 1,559Fixed DCFImplied EV / year-5 EBITDAJunior28.0xPeers today14.0xFixed7.9xPeers price years of fast growth still ahead
In the junior's DCF of Pravolt the terminal value is 92% of an Rs 4,577 crore enterprise value and implies 28.0x year-5 EBITDA, twice the 14x peers trade on today; the fixed version implies 7.9x and gives about Rs 1,559 crore.
Step 2What are the two errors inside the terminal value?

The first is the growth rate. In the Gordon growth formulaTerminal value equals next year cash flow divided by the discount rate less the perpetual growth rate; named after Myron Gordon., dividing by 12% less 9.5% multiplies next year's cash flow by 40, and every tenth of a point of growth moves the answer sharply. A company cannot grow faster than the economy forever, so perpetual growth belongs near long-run nominal growth. The second error is quieter: capex equal to depreciation means no net investment, yet the model grows 9.5% a year. Growth has to be paid for. At a 20% return on new capital, 6% growth needs 30% of profit reinvested every year.

The relationship
TV=NOPAT5(1+g)(1−gRONIC)WACC−g=169.2×1.06×0.700.12−0.06≈2,092\text{TV} = \frac{\text{NOPAT}_5 (1+g)\left(1 - \tfrac{g}{\text{RONIC}}\right)}{\text{WACC} - g} = \frac{169.2 \times 1.06 \times 0.70}{0.12 - 0.06} \approx 2,092
NOPAT 5year-5 after-tax operating profit, Rs 169.2 crore
gperpetual growth, 6%
RONICreturn on new invested capital, 20%
g / RONICshare of profit that must be reinvested to grow at g
What it says in wordsThe terminal value is the cash left after paying for growth, divided by the gap between the discount rate and that growth.
Step 3What does the corrected DCF give, and how do you sanity-check it?

With 6% growth and matching reinvestment the terminal value is Rs 2,092 crore, 7.9x year-5 EBITDA, and the enterprise value falls to about Rs 1,559 crore. An implied multiple below today's 14x is correct here, because peers are priced for years of fast growth and Pravolt's terminal value describes a business that has stopped growing fast. As a cross-check, an exit at 10x gives Rs 1,872 crore. The honest answer is a range between those two, with the terminal value still 76% of the total, which is typical for a growing company.

The multiple a terminal value implies explodes as growth nears the discount rate7x14x21x28x35xPeers today, 14xJunior: g 9.5%, 28xFixed: g 6%, reinvesting, 7.9xNo reinvestmentReinvestment matched to growth3%5%7%9%10%Terminal growth rate (discount rate 12%)
At a 12% discount rate Pravolt's implied exit multiple rises from single digits at 4% growth to 28x at 9.5% growth with no reinvestment, while 6% growth with reinvestment matched to it implies 7.9x.

Close by saying what the fix does not settle. If Pravolt's lithium business really can grow at double digits for a decade, the right repair is a longer explicit forecast, ten years with growth fading, not a higher perpetual growth rate. Growth belongs where you can see and test it.

Where candidates lose it

The common loss is attacking the discount rate or the explicit forecast first, because those feel like the difficult inputs. The error is in one line of the terminal value, and the interviewer wants to see you find it with one division.

The second is fixing the growth rate but leaving capex equal to depreciation. That still values growth that nobody pays for, and a sharp reviewer will catch it on the next question.

What the interviewer asks next

  • What terminal value share would make you comfortable for a mature utility, and why?
  • If you use an exit multiple instead, how do you choose it without circularity?
  • Pravolt's lithium unit is growing 40% a year. How would you model it separately?

Company names and figures are illustrative.