Case 055Relative valuationHard
The street values Velmora Refining at 6x FY27 EBITDA built on a peak refining margin. You normalise the margin to mid-cycle. Defend your variant view of the multiple.
1The situation
Velmora Refining processes 130 million barrels of crude a year. At an assumed Rs 80 to the US dollar, each USD 1 a barrel of gross refining margin is worth Rs 1,040 crore of EBITDA, and fixed and other costs net to Rs 1,360 crore.
The street's FY27 EBITDA of Rs 8,000 crore uses a margin of USD 9 a barrel and applies 6x EV/EBITDA. Velmora's margin over the last ten years averaged USD 6.5, with a low of 3.5 and a high of 8.5. Net debt is Rs 12,000 crore, there are 150 crore shares, and the stock trades at Rs 235.
2Your task
What EBITDA and what multiple would you use, what value does that give, and how do you defend being away from the street?
Quick check
The street's EV of Rs 48,000 crore is what multiple of mid-cycle EBITDA?
Worked solution
Try it on paper, then open one step at a time.
30-second answerThe answer to give first
Apply 6x to mid-cycle EBITDA of Rs 5,400 crore, not to the peak Rs 8,000 crore: an enterprise value of Rs 32,400 crore and about Rs 136 a share, or about Rs 149 crediting one peak year's extra cash. That is the same as 4.05x peak EBITDA. The street's Rs 48,000 crore is 8.9x mid-cycle, a price that assumes USD 9 margins last.
Step 1Why is a normal multiple on a peak year a mistake?
A mango seller has a stellar month in May. Nobody would value her stall at a normal multiple of May's profit, because June will not look like May. A multiple is a shorthand for many years of earnings, so it only works on earnings that are typical of those years. Refining is a cyclical business: the margin between crude and products swings with global capacity, and a margin above the whole ten-year record is a peak, not a base.
Step 2How do you get to mid-cycle EBITDA?
Build EBITDA from the margin so the link is visible. 130 million barrels at Rs 80 a dollar means each USD 1 a barrel is Rs 1,040 crore. At USD 9, gross margin is Rs 9,360 crore and EBITDA Rs 8,000 crore after Rs 1,360 crore of costs. At the USD 6.5 average, EBITDA is Rs 5,400 crore, a third lower, because costs do not shrink when the margin does. The street's number is a single year; yours is the average year.
Step 3Which multiple, and why is it the same answer as a low multiple on peak?
Six times is a sensible multiple for a refiner through a cycle. You can apply it to mid-cycle earnings, or apply a lower multiple to peak earnings, and the two must give the same value: here 4.05x peak. Saying both versions shows the interviewer you understand the multiple and the earnings are not independent choices. The street is effectively paying 8.9x for a normal year.
Give the peak its due. If FY27 really delivers USD 9, Velmora earns Rs 2,600 crore more EBITDA than a normal year, about Rs 1,950 crore after 25% tax, which is about Rs 13 a share of extra cash. That lifts the variant value to about Rs 149, still far below the street's Rs 240 and the Rs 235 price. One good year is worth one year of excess cash, not a permanently higher value.
| Approach | EBITDA | Multiple | EV, Rs crore | Per share, Rs |
|---|---|---|---|---|
| Street | 8,000 | 6.0x | 48,000 | 240 |
| Mid-cycle | 5,400 | 6.0x | 32,400 | 136 |
| Mid-cycle plus one peak year's extra cash | 5,400 | 6.0x | 34,350 | 149 |
Close with what would prove you wrong. If new refining capacity is delayed for years or old capacity closes for good, the mid-cycle itself moves up, and the ten-year average becomes the wrong anchor. Name the capacity data you would watch; that is the honest limit of a normalised view.
Where candidates lose it
Candidates defend a variant view by arguing for a different multiple, 5x instead of 6x, without touching the earnings. The interviewer wanted to hear that the disagreement is about which year's earnings the multiple belongs to.
The second loss is treating mid-cycle as the ten-year average without asking whether the cycle itself has shifted. Say it before the interviewer does.
What the interviewer asks next
- How would you value Velmora if the margin stays at USD 9 for three years and then reverts?
- Why might a refiner's mid-cycle multiple be lower than a chemical company's?
- What would make the ten-year average the wrong mid-cycle anchor?
Asked at Franklin Templeton, Oil & Gas, San Mateo, 2024 (Wall Street Oasis): Why did I have a variant view of the multiple I applied to a refiner company relative to street expectations
Company names and figures are illustrative.
