How energy companies are valued

Why one toolkit does not cover the sector

Ask a generalist how to value a company and you get a discounted cash flow, trading comparables, and precedent transactions. That answer is not wrong in energy, but it is incomplete in a way interviewers probe immediately, because the sector contains businesses that break the standard tools in different directions.

A producer breaks them because it is a depleting asset, so a forward multiple assumes a production level the company cannot sustain without reinvestment. A refiner and a service company break them because their earnings swing so violently that a spot multiple points the wrong direction at both ends of the cycle. A midstream operator does not break them at all, which is itself worth knowing, since the correct answer there is conventional and candidates who over-complicate it look like they are pattern-matching rather than thinking.

So the useful mental model is: match the methodology to what the asset actually is. Depleting resource, contracted infrastructure, or cyclical processor. Everything below follows from that.

Upstream: net asset value first, multiples as cross-check

For an exploration and production company, the primary methodology is the net asset value model. You project the production profile of each reserve category over the life of the assets, apply a price deck adjusted to realized prices, subtract operating costs, taxes, and development capital, tax-effect, and discount each category at a rate reflecting its risk. Add undeveloped acreage and other assets, subtract net debt, divide by fully diluted shares. There is no terminal value, because the reserves deplete. The full construction is in reserves, PV-10, and the upstream NAV model.

The multiples run alongside it as cross-checks rather than as the answer, and each one has a specific weakness worth being able to name.

EV to EBITDAX is the sector's comparable multiple. The X adds exploration expense back, which neutralizes the difference between successful efforts and full cost accounting so two companies with identical operations become comparable, as explained in energy accounting: successful efforts versus full cost. What it does not neutralize is everything else: decline rate, reserve life, cost structure, production mix, and hedge position all remain sources of difference between two companies at the same multiple.

EV per flowing barrel of oil equivalent per day divides enterprise value by current daily production. It is a fast screen and is genuinely useful for a first pass on producers in the same basin with similar production mixes. It says nothing about how long that production lasts, nothing about cost structure, and it is distorted by mix, since a barrel of oil equivalent made mostly of gas generates far less revenue than one made mostly of oil.

EV per barrel of proved reserves takes the opposite view, valuing the resource base rather than current output. It corrects for reserve life but ignores the timing of production, so a company whose reserves are decades from being produced looks identical to one producing them now, when discounting says they are worth very different amounts.

PV-10 is a standardized disclosure, not a valuation. It uses a mandated trailing average price, covers only proved reserves, and is before tax, so it functions as a comparable reference point and a sanity check rather than an answer.

Midstream: conventional, with the work in the contracts

Midstream is the sub-sector where the standard toolkit works. EV to EBITDA against comparable operators, a discounted cash flow built on contracted volumes and fees, and for partnerships and other yield-oriented securities a distribution or distributable cash flow yield analysis reflecting how income-focused investors actually price them.

The analytical work is not in the methodology, it is in justifying the multiple. Two operators with the same EBITDA are not worth the same if one has long-dated take-or-pay contracts with investment grade shippers and the other has percent-of-proceeds arrangements with small private producers. So the questions are what percentage of revenue is fee-based or take-or-pay, what the weighted average remaining contract tenor is, how creditworthy the counterparties are, and how concentrated the acreage dedication is behind the system. Those distinctions are the subject of midstream contracts and MLP structures.

One diligence point specific to the metric: distributable cash flow subtracts maintenance capital, and the line between maintenance and growth capital is a management judgment. A company classifying aggressively can report a comfortable distribution coverage ratio while underinvesting, so compare maintenance capital against depreciation and cross-check with free cash flow after all capital spending.

Contracted export infrastructure follows the same logic pushed further, since a liquefaction project with most of its capacity sold forward under fixed fees is valued on a contracted discounted cash flow with sensitivities on construction cost and schedule rather than on a commodity price deck.

Downstream and services: the normalization problem

Refining and oilfield services share a valuation problem, and understanding it once solves both.

Both earn something that swings far more than the underlying commodity. A refiner earns a spread between two volatile prices, and a spread is more volatile than either component. A service company earns revenue from upstream capital budgets, which are themselves a discretionary derivative of producer cash flow, so service revenue amplifies the producer cycle. Both carry high operating leverage from largely fixed cost bases, which converts revenue swings into much larger earnings swings.

The consequence is that a trailing or spot multiple inverts. At a cycle peak, earnings are high, so EV to EBITDA looks low and the equity screens cheap, immediately before earnings normalize downward. At a trough, earnings are depressed or negative, so the multiple looks enormous or is undefined, at the point where the assets may be genuinely undervalued. A naive screen therefore recommends buying at the top and avoiding at the bottom.

The standard fix is mid-cycle EV to EBITDA: estimate what the business earns at a normalized margin across a full cycle and apply a multiple to that. This does not eliminate the judgment, it relocates it, which is the point. The argument moves to what mid-cycle actually means for this asset, which is the argument worth having.

The supporting methods differ slightly by sub-sector. For refining, replacement cost per barrel of capacity, adjusted for complexity so a more capable facility is worth more, plus free cash flow yield to capture the recurring cost of turnarounds. Those mechanics are in downstream refining and crack spreads.

For services, replacement value of the fleet is the natural anchor in asset-heavy segments like rig contracting, where the equipment has an observable market value, alongside EV per active unit as a screen. The variables you interrogate are day rates, utilization, and contract backlog with its tenor, since a long-dated backlog at fixed rates is worth far more than spot exposure heading into a downturn. The asset-heavy versus asset-light split matters more here than anywhere else in the sector: a company supplying mostly people and consumables can flex costs down as activity falls, while a fleet owner carries fixed costs against equipment that cannot be repurposed, which is why idle capacity in that segment is so damaging.

Why two identical-looking producers are worth different amounts

This is the sector's best comparison question, and the answer set is worth memorizing because it recurs in several forms.

Production mix. Identical barrel of oil equivalent volumes generate very different revenue depending on the liquids share, because the roughly six-to-one conversion is an energy equivalence rather than an economic one.

Decline rate and inventory depth. A company with a shallower decline and years of remaining development locations sustains production with less reinvestment, so more of its EBITDA converts to free cash flow. A company that must spend heavily to stay flat is worth less per unit of current EBITDA even though the multiple looks the same.

Cost structure. Lease operating expense per barrel, transportation, and general and administrative costs per barrel all vary widely and determine what the company keeps.

Realized pricing. Location matters, because a producer in a basin with constrained takeaway capacity receives a price materially below the headline benchmark. Two companies at the same benchmark price can realize very different revenue.

Balance sheet and hedges. Leverage changes the risk of the equity, and a hedge book fixes near-term cash flows at levels that may be well above or below the current market, as covered in commodity hedging and price decks.

Operatorship and acreage contiguity. An operated position with control over the development pace is worth more than a scattered non-operated interest, and contiguous acreage supports more efficient development, which also drives what an acquirer will pay for a property package.

What interviewers push on

The first push is the terminal value question for a producer, where the correct answer is that there is none, because the reserves deplete, and that what appears at the end of the model is separately estimated undeveloped acreage value plus salvage net of plugging and abandonment liabilities.

The second is why the sector uses EBITDAX, where the answer they want is the exploration accounting divergence specifically, not a general statement about capital intensity.

The third is a trap disguised as a screen: given a table of comparables, which company is cheapest. The correct move is to refuse the direct answer and ask what is behind the multiples, because in this sector a low multiple usually reflects something real about decline rate, reserve life, basis differentials, or leverage rather than a mispricing.

The fourth is asking you to sanity-check a NAV against the market. When the NAV and the market capitalization disagree sharply, that gap is the interesting finding, and it typically points at a specific thing: a price deck the market does not share, an acreage position the market is not valuing, or a hedge book being priced differently.

Practice question

How would you value an exploration and production company, a midstream partnership, and a refiner differently?

They need three different frameworks because they're three different kinds of asset. For the producer, the primary tool is a net asset value model, because it's a depleting asset and a forward multiple would assume a production level it can't sustain without reinvestment. I'd project each reserve category over the life of the assets, apply a price deck adjusted to realized pricing, subtract operating costs, taxes, and development capital, discount producing reserves at a lower rate than undeveloped ones, then add acreage, subtract net debt, and divide by shares. No terminal value, because the reserves run out. I'd cross-check with EV to EBITDAX and EV per flowing barrel. For the midstream partnership, the standard toolkit works: EV to EBITDA, a discounted cash flow on contracted volumes, and a distributable cash flow yield. The real work there isn't the methodology, it's justifying the multiple with contract quality, so what percentage of revenue is fee-based or take-or-pay, the weighted average tenor, and counterparty credit. For the refiner, I'd use mid-cycle EV to EBITDA rather than a spot multiple, because refining earns a spread between two volatile prices and the trailing multiple inverts, looking cheap at the peak and expensive at the trough. I'd support that with replacement cost per barrel of capacity adjusted for complexity.

What the interviewer is listening for: Whether you match methodology to asset type rather than reciting one framework three times. The two details that carry the answer are the absence of a terminal value for the producer and the inverted multiple problem for the refiner, since both prove you understand why the standard tools fail rather than just that they do.