Mineral royalty cash flow to PV-10

Once you've got a defensible decimal interest, the next problem is turning monthly cash flow into a number a buyer or auditor can sign off on. That number is the Net Present Value of forecast royalty cash flows — and it's where most mineral models leak value. This post walks through the inputs (decline curve, price deck, lease economics, discount rate), the monthly math, and the validation checklist that catches the recurring errors.

This is Stage 5 of the five-stage Texas mineral valuation pipeline. The earlier posts handle Stages 1–2 (parcel and ownership) and Stages 3–4 (well classification and decimal interest). The anchor post explains why the pipeline is built in stages in the first place.

The four inputs a defensible model needs

A royalty NPV model needs four input families: a production forecast per well, a price deck, the lease economics, and a discount rate. Get any one of them wrong and the NPV doesn't survive a buyer's diligence.

1. Production forecast per well

Modern Texas unconventional wells follow a hyperbolic decline curve through the early years, settling into an exponential terminal decline. The standard model is Arps, with three parameters:

Initial production rate (IP) — typically the highest monthly rate from the first 3–6 months of P-1 production data, expressed in bbl/d for oil and mcf/d for gas
Initial decline rate (D_i) — the slope of decline at month zero, typically 50–80% per year for unconventional Texas wells in year one
Hyperbolic exponent (b) — typically b ∈ [0.5, 1.5] for unconventional wells; pure exponential decline is b = 0

For a well with existing production history, fit these parameters to the actual P-1 data using scipy.optimize.curve_fit or equivalent. For undeveloped wells with no production yet, use a basin-typical type curve (Delaware Basin, Midland Basin, and Eagle Ford each have their own published type curves that have to be operator-adjusted because completion designs vary).

Two other parameters matter for the basins where they apply:

Estimated ultimate recovery (EUR) — the total volume the well is expected to produce over its life. Used as a sanity check against the integrated decline curve; if they don't match, one of them is wrong.
Gas-oil ratio (GOR) evolution — especially material for Eagle Ford wells, where the GOR rises sharply over the well's life and changes the economic mix.

2. Price deck

The price deck is a forward curve of expected oil and gas prices for the well's economic life. Three layers:

NYMEX forward strip — the listed forward prices for WTI crude and Henry Hub natural gas, typically 24 to 60 months out. Available daily from CME.
Long-term flat price — beyond the strip, pick a defensible flat assumption and document it. As of 2025–26, defensible ranges are $60–75 WTI and $3–4 Henry Hub. Auditors will ask why you picked your number; "consensus broker estimate" or "long-term marginal cost of new supply" are defensible answers.
Differential to local hub — the difference between the NYMEX benchmark and the price your operator actually realizes. For Permian wells, the Midland-to-Cushing differential is the canonical adjustment ($1–3/bbl in normal times, wider in pipeline-constrained periods). For Eagle Ford, Houston Ship Channel or Corpus Christi differentials apply.

The price deck has to be documented and dated. A valuation built on yesterday's strip is fine; one built on a deck that nobody can reconstruct is not defensible.

3. Lease economics

Three things from the lease and the state:

Royalty rate from the actual lease, not assumed. Modern Texas hot acreage is 20–25%; legacy leases run 15% or even 12.5%.
Post-production deductions per the lease's specific language. Some leases allow the operator to deduct gathering, processing, transportation, dehydration, and compression costs from the royalty before the check is cut. A nominal 25% royalty can become an effective 20–22% royalty in pay after deductions. State law varies on what's allowed: Texas leans operator-favorable; Oklahoma and Pennsylvania have had major class actions on this.
Severance tax — Texas charges 4.6% on oil and 7.5% on gas, levied on the market value at the wellhead. Various exemptions exist for low-producing leases but most have been price-triggered out of relevance.

4. Discount rate

The discount rate translates future cash flows into present value. The choice depends on what the model is for:

PV-10 (10% flat) — the SEC standard for reporting comparability. Use PV-10 for SEC-filed reserves disclosures and for any context where outside buyers will benchmark against industry filings.
Economic NPV for royalty interests: 8–12% for most actively-produced positions. Royalty interests are downside-protected (cost-free, lower-risk) and command a lower discount rate than working interests.
Non-operating working interests: 12–15%, depending on operator credit, basin maturity, and decline-curve confidence.

Document the rate. Auditors and diligence buyers will ask.

The math — monthly cash flow and monthly discounting

For each well, for each month from the valuation date forward, the calculation is:

revenue_month        = production_month × price_month × (1 − severance_rate)
royalty_cash         = revenue_month × decimal_interest
post_prod_deductions = royalty_cash × deduction_rate
net_cash_month       = royalty_cash − post_prod_deductions

NPV   = Σ ( net_cash_month / (1 + r/12)^month )
PV-10 = same with r = 0.10

One subtle but important point: discount monthly, not annually. Annual discounting applied to monthly cash flows undervalues by 4–5%, which matters at scale. The monthly discount factor uses (1 + r/12)^month, not (1 + r)^(month/12). The two are arithmetically close but not identical; the correct formula matches the cash-flow cadence.

A walked-through example

Consider a well in its first month after production start:

Input	Value
Monthly production	17,000 bbl
Realized price (after differential)	$68/bbl
Decimal interest	0.01776 (from the allocation-well math)
Texas severance rate	4.6%
Post-production deduction rate	5%
Discount rate	10% (PV-10)

Month 1 net cash to the mineral owner:

revenue_month        = 17,000 × $68 × (1 − 0.046)   = $1,102,936
royalty_cash         = $1,102,936 × 0.01776          =  $19,588
post_prod_deductions = $19,588 × 0.05                =     $979
net_cash_month       = $19,588 − $979                =  $18,609

That $18,609 then gets discounted by the month-1 factor 1 ÷ (1 + 0.10/12)^1 = 0.99174 for a present-value contribution of $18,455. The same calculation repeats for every month over the well's economic life, with production declining per the Arps curve and prices following the deck. The sum is the NPV.

What the model should output

A defensible model produces five things, every time, in the same shape:

NPV at the chosen discount rate (8–12% for royalty interests). The headline number.
PV-10 for SEC-comparable reporting. Always include this even when the headline NPV uses a different rate, because buyers and auditors benchmark against PV-10.
$/NMA implied — NPV ÷ NMA. Useful for comparing against market comps (current Permian PDP-heavy acreage trades at roughly $25,000–$45,000 per NMA).
Sensitivity table — price ±20%, decline curve ±20%, discount rate ±300 bps. The single most useful output for a buyer doing diligence; it shows where the model is exposed.
Monthly cash flow detail exported to CSV — so the model can be re-run by downstream tools (PHDWin, ARIES, Excel, or a custom dashboard) without rebuilding the inputs.

Validation checklist for Stage 5

Run any NPV through this checklist before signing off. Each line catches a known failure mode:

☐ Decline curve fit to actual P-1 data for producing wells (don't apply a basin type curve when actual data exists)
☐ Type curve documented for non-producing wells, with the basin, operator class, and completion-design assumptions explicit
☐ Price deck sourced and dated — strip + long-term flat + differential all named
☐ Royalty rate read from the lease, not assumed
☐ Post-production deductions modeled per lease, not assumed
☐ Texas severance tax included (4.6% oil, 7.5% gas)
☐ Discount rate documented with rationale
☐ Monthly discounting — not annual
☐ Sensitivity table included: price ±20%, decline ±20%, discount rate ±300 bps
☐ $/NMA implied reported alongside NPV for benchmarking
☐ PV-10 reported alongside the economic NPV
☐ Division order cross-check — model decimal interest vs. operator-paid interest, flag deltas above 5%
☐ Audit log — every NPV record links back to its input data snapshot (decline curve fit, price deck, lease) so the number is defensible 18 months later

Benchmarking $/NMA against the market

The $/NMA implied by the model is most useful when compared against current market comps. Permian mineral acreage in 2025–26 trades at roughly:

Drilled and producing minerals (PDP-heavy): $25,000–$45,000 per NMA
Undeveloped but near permitting activity: $10,000–$25,000 per NMA
Inactive / no near-term drilling: $2,000–$8,000 per NMA

If your model says a PDP-heavy position is worth $80K/NMA, you've probably over-modeled. If it says inactive acreage is worth $50K/NMA, ditto. The comps are a sanity check against the headline NPV; they don't replace the model, but they catch order-of-magnitude errors before the report leaves the room.

The takeaway

Stage 5 is where the four upstream stages — surface acreage, ownership, well classification, and decimal interest — turn into a number that means something. The math is monthly cash flow discounted at the chosen rate, summed across the well's economic life. The four inputs that matter are the decline curve, the price deck, the lease economics, and the discount rate. Get any of those wrong and the NPV is wrong by the same factor; get them all right and the model survives diligence.

The 13-line validation checklist above catches the recurring failure modes. Most of them — annual instead of monthly discounting, assumed royalty rate, generic decline curve where real data exists — are not subtle. They show up in models because the analyst is moving fast, not because the math is hard. The discipline is in the checklist, not in any single arithmetic step.

One way to read the whole five-stage pipeline: each stage has one or two recurring errors that account for most of the value leakage in real-world mineral valuations. Stage 1's biggest error is computing area in Web Mercator. Stage 2's is assuming mineral acres equal surface acres. Stages 3–4's is applying pooled math to allocation wells. Stage 5's is annual discounting and undocumented price decks. The audit of any valuation is, in essence, a walk through those five recurring errors — and the validation checklists across the four posts in this series are the complete map.

Sources

Need to turn this engineering logic into a production-ready tool? Explore custom petroleum software development.