The NBR Formula
The Net Benefit Ratio expresses, as a single dimensionless number, whether the verified community benefit delivered by a compute deployment exceeds the resource costs it imposes on that community. A ratio greater than 1.0 means the deployment contributes more than it consumes by the defined metric categories. A ratio below 1.0 fails the threshold.
NBR = Net Benefit Ratio | Bv = Benefit Value (composite) | Cr = Cost / Resource draw (composite)
| Variable | Name | Definition | Status |
|---|---|---|---|
| NBR | Net Benefit Ratio | Composite ratio of verified community-facing benefit to verified resource cost for a specified compute deployment over a defined measurement period. | PROPOSED |
| Bv | Benefit Value | Aggregate of all measurable, attributable community benefits delivered by the deployment in the measurement period. See §4 for categories. Units must be normalized before aggregation; normalization methodology is subject to review. | PROPOSED |
| Cr | Cost / Resource Draw | Aggregate of all measurable resource costs imposed on the host region by the deployment in the measurement period. Includes energy, water, grid, land, and infrastructure impact. See §5 for components. | PROPOSED |
Bv — What Counts as Benefit
Bv is the numerator. It is the sum of all measurable, attributable benefits the deployment delivers to its host region during the measurement window. For a benefit to count toward Bv, it must meet three conditions:
- Measurable: the benefit must be expressible in a defined unit (financial value, volume, count, rate) that can be independently observed or metered.
- Attributable: a documented causal chain must connect the deployment's operation to the benefit. Correlation alone does not satisfy this requirement.
- Local: the benefit must accrue to the host municipality, county, watershed, or utility district — not primarily to the compute operator's distant shareholders.
Benefits that fail any of these conditions are excluded from Bv. Disputed or ambiguous items must be logged in the Pending Audit register and excluded from the active NBR calculation until reviewed.
Cr — What Counts as Cost
Cr is the denominator. It represents all resource costs the deployment places on the host region. Cost components must be measured at the source, not inferred from vendor estimates without independent metering.
| Component | Unit | Description | Status |
|---|---|---|---|
| E_total | MWh | Total electrical energy consumed by the deployment per measurement period, including idle overhead and cooling. Measured at utility interconnect, not estimated from hardware specs. | PROPOSED |
| W_consume | kL | Water consumed (not returned to source) by cooling systems per measurement period. Measured at meter, not calculated from WUE ratios alone. | PROPOSED |
| W_draw | kL | Total water drawn from aquifer or municipal supply, including water returned at lower quality. Captures stress on shared supply independent of consumptive loss. | PROPOSED |
| G_impact | MW-events | Measurable degradation to local grid reliability or capacity margin attributable to the deployment. Expressed as frequency and magnitude of grid stress events caused by load spikes. | PENDING METHODOLOGY |
| L_use | acres | Net land area committed to deployment footprint, including buffer, access roads, and cooling infrastructure. Excludes land returned to productive use as part of a documented remediation agreement. | PROPOSED |
Normalization of Cr into a single composite value requires a weighting methodology that is not yet independently validated. Operators using NBR must document their normalization choices and submit them for review under the GCTS-1 audit protocol.
Candidate Bv Categories
The following categories are the currently proposed list of benefit types eligible to contribute to Bv. All are subject to the measurability, attributability, and locality tests from §2. Categories marked PENDING METHODOLOGY do not yet have finalized measurement protocols.
PROPOSED
PROPOSED
PENDING METHODOLOGY
PROPOSED
PROPOSED
PROPOSED
PROPOSED
PENDING METHODOLOGY
Baseline and Counterfactual Logic
NBR requires comparison to a defined baseline — what would have occurred in the host region in the absence of the deployment. The counterfactual baseline must be:
- Documented before deployment begins. A pre-deployment characterization of the host region's resource state, economic baseline, and environmental conditions must be filed and retained.
- Conservative. When baseline data is incomplete, the counterfactual assumption must favor the more demanding interpretation — it must not be adjusted to make NBR appear favorable.
- Independently reviewed. The baseline characterization must be reviewed by an entity with no financial interest in the deployment outcome before the deployment's first NBR calculation is published.
- Updated periodically. Regional conditions change. The baseline must be reviewed at minimum annually and adjusted where verified changes in regional conditions warrant revision.
The counterfactual is not the worst-case scenario. It is the most reasonable, evidence-supported estimate of what the region's trajectory would have been without the deployment. Speculative negative counterfactuals (e.g., "without us the aquifer would have been depleted anyway") require documented evidence to support the claim.
Attribution Rules
When benefits or costs cannot be fully attributed to a single deployment — for example when multiple operators share infrastructure — the following attribution principles apply:
| Situation | Attribution Method | Notes |
|---|---|---|
| Sole-operator deployment | Full attribution to operator | All Bv and Cr assigned to single NBR calculation. |
| Shared infrastructure (multiple operators) | Pro-rata by measured load share | Allocation based on metered consumption ratios, not contracted capacity. |
| Benefit delivered by third party (e.g., utility) in response to operator incentive | Partial attribution — maximum 60% of benefit value | Operator may claim only the incentive-attributable fraction. Baseline third-party activity excluded. |
| Cost caused by external factor (drought, grid event) coinciding with deployment | Adjusted Cr — subtract confirmed external component | Requires regulator or independent verifier confirmation of external cause. No self-certification. |
| Benefit not yet delivered but contractually committed | Excluded from current NBR; logged as pending | Contractual commitments are Proposed, not Verified. Future reporting may include upon delivery. |
Operator NBR vs. System NBR
The framework distinguishes two scopes of NBR calculation:
- Covers one physical deployment location
- Measured over one reporting period
- Reported by the operator, verified by accredited auditor
- Used for G-level certification determination
- Required for regulatory compliance under proposed ordinances
- Primary NBR for public disclosure
- Aggregates multiple operators or deployments
- Used for municipal or regional policy reporting
- Not a substitute for Operator NBR at the individual site level
- Computed by governance body from audited Operator NBRs
- Published in annual public compliance reports
- Cannot be used to offset a failing Operator NBR
A deployment with NBR below the required threshold cannot use a favorable System NBR to argue compliance. Each site is evaluated individually. System NBR is a planning and policy metric, not a compliance mechanism.
NBR Thresholds by G-Level
| G-Level | Min NBR | Deployment Scale | Audit Frequency | Public Report |
|---|---|---|---|---|
| G-0 | NBR ≥ 1.0 | < 5 MW | Self-reported annually | Summary disclosure |
| G-1 | NBR ≥ 1.2 | 5 – 20 MW | Third-party annually | Full report required |
| G-2 | NBR ≥ 1.5 | 20 – 100 MW | Third-party semi-annually | Full report + data files |
| G-3 | NBR ≥ 2.0 | > 100 MW | Continuous telemetry + annual audit | Real-time dashboard + annual report |
The proposed Texas County SYNC 3.0 pilot is a 50 MW deployment that falls in the G-2 range under this proposed classification. Its projected NBR has not been measured; it is a scenario projection. See Pilot Status for current deployment phase.
Confidence and Verification Levels
Every published NBR claim must carry an explicit confidence designation. Claims without confidence designation are considered unverified and must not be cited in public-facing materials.
The current public NBR framework operates entirely at L5 (Proposed) and L3 (Modeled) confidence levels. No L1 or L2 NBR calculations have been published. First verified NBR data is expected from the Texas County pilot if and when the pilot advances to Phase 1 (active deployment).
Documentation Expectations
Operators claiming NBR compliance must maintain and make available, on request by the oversight body, the following documentation:
- Pre-deployment baseline report — characterization of host-region resource state before deployment began.
- Measurement plan — documented methodology for metering each Bv and Cr component, including sensor specifications, meter locations, data logging format, and audit trail.
- Bv attribution records — source documentation for each claimed benefit item: contracts, invoices, utility bills, payroll records, or regulatory filings.
- Cr measurement logs — time-stamped utility meter readings, water meter readings, and grid-interconnect telemetry for the full measurement period.
- Normalization methodology disclosure — the specific weighting or conversion factors used to aggregate heterogeneous Bv and Cr components into a single ratio.
- Conflict-of-interest declaration — signed disclosure by any party involved in data collection, analysis, or reporting of any financial interest in the deployment outcome.
- Auditor report — signed findings from an accredited third-party auditor (G-1 and above), including exceptions and qualified items.
- Annual update — updated calculation for each measurement period, with changes from prior period explained.
Changelog
- Published formula, variable definitions, and attribution rules
- Added Operator vs. System NBR distinction
- Added confidence-level taxonomy (L1–L5)
- Added Jevons adjustment note to Bv categories
- Documented all items as Proposed; no verified data yet
- First public release of GCTS-1 document including NBR reference
- Threshold values introduced as working proposals
- Formula concept introduced in design documents
- Not yet published for external review