The Internet as a Public Utility
- the Institute
- 2 days ago
- 14 min read
Technical Analysis of Community-Owned Mesh Networks
A Network Theory Applied Research Institute Whitepaper November 2025
1. Introduction
The internet was designed as a distributed network to survive nuclear war.[^1] Today, the top three cloud providers—AWS, Microsoft Azure, and Google Cloud—control approximately 65-66% of global cloud infrastructure, with the top ten providers controlling over 90% of the market.[^2] American cities and states offer hundreds of millions annually in tax incentives for data centers that create 20-50 jobs with minimal community wealth building. Recent analysis reveals that state and local governments pay an average of $1.4-6.4 million per data center job created, with total subsidies reaching $1 billion annually in states like Texas alone—representing a return on public investment of just 0.15-0.38x.[^3]
Rural electrification provides an alternative model. The Rural Electrification Administration (1935) enabled member-owned electric cooperatives that today serve 42 million Americans across 56% of U.S. landmass, supporting 623,000 jobs and contributing $111 billion annually to GDP—with $75 billion and 424,000 jobs occurring within local communities.[^4] This demonstrates that distributed, democratically-governed infrastructure succeeds where corporate models extract wealth.
This whitepaper analyzes whether internet infrastructure can operate as a profitable public utility using cooperative ownership. We demonstrate that community mesh networks generate superior technical performance (projected 10-20x latency reduction requiring pilot validation), economic returns (50-80x higher ROI than traditional subsidies based on verified cost data), and substantial community wealth building ($3,000-4,000 per household over 7 years with zero capital investment) compared to traditional data center subsidies.
Methodology Note: Economic projections are derived from comprehensive feasibility analysis conducted for Louisville, Kentucky metropolitan area (population 1.3 million), providing real-world grounding for the financial model. Technical performance projections are based on theoretical network analysis and comparative modeling, requiring empirical validation through pilot deployment.
2. Technical Architecture
2.1 Edge Computing and Market Growth
Edge computing distributes processing to network edges rather than distant data centers, reducing latency from 50-100ms to <5ms—enabling autonomous vehicles (<10ms required), remote surgery (<5ms for haptic feedback), and AR/VR (<20ms to prevent motion sickness).[^5] The global edge computing market has experienced rapid expansion, with 2024 valuations ranging from $24-28 billion and projections reaching $180-250 billion by 2030, reflecting sustained annual growth rates of 33-38%. This growth is driven by 5G deployment, IoT proliferation, AI-enabled edge applications, and smart city infrastructure.[^6]
2.2 Mesh Network Architecture
Mesh networks create redundant peer-to-peer connections enabling self-healing, scalability through network effects, and Byzantine fault tolerance.[^7] Each node connects to minimum 3 neighbors, eliminating single points of failure while achieving graceful degradation under stress.
Node Specifications:
CPU: 8-16 cores (AMD EPYC/Intel Xeon)
RAM: 32-64 GB ECC
Storage: 2-4 TB NVMe + 8-16 TB HDD
Network: 10 Gbps Ethernet
Power: 150-300W
Cost: $2,000-3,000 (bulk purchasing)
Software Stack:
Kubernetes: Container orchestration with automatic load balancing, self-healing, horizontal scaling[^8]
Ceph: Distributed storage providing object/block/file interfaces with self-healing and petabyte scalability[^9]
Security: Zero-trust networking, AES-256 encryption, TPM 2.0 hardware security modules[^10]
2.3 Projected Performance Advantages
Distributed architecture eliminates 40% energy overhead from centralized HVAC systems while enabling waste heat recapture in residential deployments.[^13]
2.4 Performance Modeling Methodology
The performance advantages cited in this whitepaper are based on theoretical network analysis and comparative modeling against centralized architectures. These projections require empirical validation through pilot deployment.
Latency improvements (10-20x):
Theoretical basis: Physical proximity eliminates 30-80ms round-trip to distant data centers
Assumption: Processing occurs within 5-mile radius vs. 500+ miles to regional data center
Validation requirement: Real-world latency testing under various load conditions during Year 1 pilot
Energy efficiency improvements (20-30%):
Theoretical basis: Distributed architecture eliminates centralized HVAC overhead (PUE reduction from 1.67-2.01 to 1.2-1.4) and transmission losses
Assumption: Residential/commercial building cooling vs. dedicated facility
Validation requirement: Comparative power consumption measurement over 12-month period
Local processing (90-99%):
Theoretical basis: Edge computing design keeps data near source
Assumption: Only synchronization and backup traffic crosses metro boundaries
Validation requirement: Network traffic analysis under real workloads
Critical Success Factor: Year 1 pilot deployment will validate these projections before scaling. If actual performance falls below modeled estimates, deployment plan and financial projections will be revised accordingly.
3. Economic Model
3.1 Node Economics
Table 1: Per-Node Monthly Economics (60% utilization)
3.1.1 Financial Modeling Methodology and Louisville Case Study
The economic projections in this whitepaper are derived from comprehensive feasibility analysis conducted for Louisville, Kentucky metropolitan area (population 1.3 million, 478,000 households). This case study provides real-world grounding for the financial model and demonstrates practical viability.
Louisville Analysis Foundation:
Market Research:
Edge computing service pricing analysis across 20 major cloud and edge providers
Anchor customer needs assessment (Louisville Metro Government, University of Louisville, Norton Healthcare, Humana)
Competitive analysis of AWS, Azure, Google Cloud regional pricing with latency comparison
Result: $150/node monthly revenue at 60% utilization is conservative given projected 10-20x latency advantage enabling premium pricing for time-sensitive applications
Cost Modeling:
Electricity: Louisville Gas & Electric residential rates ($0.118/kWh) applied to 300W continuous draw = $26/month
Labor: Louisville metro area IT technician median salary ($58,600) applied to NOC staffing ratios (1 technician per 250 nodes) = $10/node/month
Equipment: Bulk purchasing quotes for 1,000+ unit orders from Dell, Supermicro, HPE with 5-year replacement cycle
Insurance: Quotes from cooperative insurance providers (NRECA-affiliated carriers) for liability and equipment coverage
Result: $40/node monthly operating cost validated by data from existing mesh network operators in similar markets
Demand Validation:
Survey of 120 Louisville small businesses: 73% expressed interest in lower-latency alternatives to cloud with cost parity
University of Louisville research computing: 200 active researchers with edge computing needs (genomics, medical imaging, AI training)
Louisville Metro Government: Cloud services currently cost $1.2M annually with documented latency issues affecting emergency response systems
Norton Healthcare system: HIPAA-compliant local processing requirements for medical imaging (2.4 petabytes annually)
Result: $2-3M in anchor customer contracts achievable by Year 2 based on documented need and procurement timelines
Household Participation Analysis:
Louisville demographic analysis: 127,000 owner-occupied households with suitable space (basement, utility room, garage)
Community organizing assessment based on comparable cooperative formation efforts: 10-15% participation rate achievable through education and outreach
Conservative model uses 5,000 nodes (4% of suitable households), well below demonstrated participation potential
Participation incentives: $25-50 monthly hosting stipend, zero capital investment, democratic governance
Result: Scaling projections are conservative given participation potential and wealth-building incentives
Sensitivity Analysis:
The model remains financially viable under stress testing:
Break-even occurs at 50% utilization (vs. modeled 60%), providing 10-percentage-point safety margin
Profitability achieved with 1,500 nodes (vs. modeled 2,000), allowing 25% shortfall in participation
ROI remains positive even with 25% cost overruns in operating expenses or equipment costs
Member benefits exceed S&P 500 returns (10% annually) even at 50% utilization, maintaining value proposition
Anchor customer loss tolerance: Model survives loss of any single anchor customer without returning to deficit
Generalizability to Other Metropolitan Areas:
Louisville represents a typical mid-sized U.S. metropolitan area. The model scales to:
Smaller metros (250K-750K population): 1,000-3,000 node deployment, break-even at 50-60% utilization, 12-18 month timeline to profitability. Requires 1-2 anchor customers minimum.
Larger metros (1.5M-5M population): 7,500-25,000 node potential, faster scaling due to anchor customer density and workforce availability. Break-even can occur within 18 months with aggressive deployment.
Rural areas: Hub-and-spoke hybrid model with community facility nodes (libraries, schools, fire stations) serving 50-100 households each via wireless last-mile. Requires modified economics but maintains cooperative ownership structure.
Sources:
Louisville Community Data Center Cooperative Feasibility Study (NTARI, 2025)
Louisville Metro Government RFI Response Analysis (2025)
University of Louisville Research Computing Needs Assessment (2024)
Louisville Gas & Electric Commercial and Residential Rate Schedules (2024)
U.S. Bureau of Labor Statistics, Louisville-Jefferson County, KY-IN Metropolitan Statistical Area (2024)
American Community Survey 5-Year Estimates, Louisville Metro (2019-2023)
3.2 System-Level Financial Projections
Table 2 presents the 5-year financial trajectory modeled on Louisville metropolitan area deployment. These projections are conservative estimates based on validated anchor customer demand, demonstrated participation rates from other cooperative models, and existing market pricing for edge computing services.
Table 2: 5-Year Financial Trajectory
Note: Financial projections based on Louisville Community Data Center Cooperative feasibility analysis (2025). Utilization rates, pricing, and costs validated through market research, anchor customer assessments, and existing mesh network operator data. Pilot validation will occur in Year 1 before full-scale deployment.
3.3 Community Wealth Building
Over 7 years, 5,000-member cooperative generates:
Member hosting stipends: $15M ($3,000 per household)
Profit distributions (Years 4-7): $5M ($1,000 per household)
Total: $4,000 per household with zero capital investment
This matches S&P 500 returns ($2,844 gain on $3,000 invested 2015-2022 at 10% annual return)[^14] while ensuring 100% local wealth retention versus shareholder extraction.
3.4 Anchor Customer Strategy
Financial viability requires $2-3M in institutional contracts by Year 2 to ensure baseline utilization:
Municipal government ($500K-1M): Cloud services, emergency response systems, smart city infrastructure, public WiFi backbone
Healthcare systems ($500K-1M): HIPAA-compliant local storage, telemedicine low-latency connections, medical imaging processing, patient data sovereignty
Universities ($300K-600K): Research computing clusters, student services, campus network backbone, library digital collections
Regional businesses ($700K-1.5M): Edge computing for retail analytics, disaster recovery and business continuity, content delivery for local media, IoT device management
These contracts ensure 50-60% baseline utilization, making break-even achievable by Year 3 independent of residential utilization growth.
3.5 Environmental Revenue
20-30% energy efficiency improvement over centralized facilities (PUE reduction from 1.67-2.01 to 1.2-1.4) generates 3,850 metric tons CO₂e annual reduction at 5,000-node scale. At $10-15 per metric ton carbon credit prices, this produces $38,000-58,000 additional annual revenue through voluntary carbon markets or compliance programs.[^15]
Additional environmental benefits not monetized in model:
Elimination of water consumption for cooling (data centers use 1-5 million gallons daily)
Reduced transmission line losses (3-7% energy waste in long-distance transmission)
Waste heat recapture potential for residential heating (10-20% additional efficiency gain)
3.6 ROI Comparison: Traditional vs. Cooperative Models
Traditional Data Center Subsidy Model:
Analysis of data center megadeals reveals poor returns on public investment. Based on comprehensive research by Good Jobs First tracking economic development subsidies nationwide:
Typical large data center megadeal characteristics:
Public investment (tax incentives): $40-100 million
Permanent jobs created: 20-50 positions
Cost per job: $1.4-6.4 million[^16] (Good Jobs First 2016-2024 analysis)
Illinois data centers (2019-2024): $1.4 million per job average
Digital Realty, Franklin Park IL: $2+ million per job
Apple, North Carolina: $6.4 million per job
Google, Columbus OH: $2.7 million per job ($54.3M for 20 jobs)
Local hiring: Minimal; most technical positions filled by transferred staff
Community wealth retention: Near zero (profits extract to distant shareholders)
7-year wage benefits: ~$7-15 million (at $70K average annual salary)
ROI: 0.15-0.38x (community receives 15-38% return on public investment)
States with highest data center subsidies (2024-2025):[^17]
Texas: $1 billion annually in foregone tax revenue
Virginia: $732 million annually
Illinois: $370 million annually ($468M total since 2019)
At least 10 states: Over $100 million annually each
Economic analysis findings:
Georgia 2022 study: Revenue forgone from exemptions exceeded tax revenue generated, resulting in net negative state fiscal impact[^18]
Good Jobs First conclusion: "At that price, taxpayers will always lose"
Local business impact: "Data centers use few goods or services that can be provided by local businesses"
Most subsidies are unnecessary: Site selection surveys show only 3% of data center owners cite tax incentives as primary factor (cheap electricity and available land dominate)[^19]
Community Mesh Network Cooperative Model:
Based on Louisville metropolitan area deployment analysis with validated cost and revenue data:
Investment and Structure:
Public investment: $350K-500K (federal broadband grants, municipal seed capital)
Member investment: Zero (equipment provided by cooperative LLC)
Permanent jobs created: 15-20 full-time technical staff
Member participation: 5,000 households hosting nodes
Community wealth retention: 100% (cooperative ownership structure)
7-year community wealth generation:
Member hosting stipends: $15 million ($3,000 per household)
Profit distributions (Years 4-7): $5 million ($1,000 per household)
Local wages (technical staff): $4.7-7.9 million (15-20 positions at $45K-75K)
Total community benefit: $24.7-27.9 million
ROI: 50-80x (community receives 50-80X return on public investment)
Table 3: Public Investment Returns Comparison
The cooperative model generates returns 131-533 times higher per dollar of public investment because:
Wealth stays local: 100% retention vs. shareholder extraction to distant investors
Value distribution: Member stipends distribute value to 5,000 households, not 20-50 employees
Democratic ownership: Profit sharing replaces executive compensation and shareholder dividends
Architectural sovereignty: Distributed architecture eliminates infrastructure monopolies and vendor lock-in
Aligned incentives: Member-owners benefit from network success through both usage and ownership
Sources:
Tarczynska, K. (2016). Money Lost to the Cloud: How Data Centers Benefit from State and Local Government Subsidies. Good Jobs First.
Good Jobs First. (2024). New Data on Data Center Subsidies, Same Old Problems.
LeRoy, G., & Tarczynska, K. (2025). Cloudy with a Loss of Spending Control: How Data Centers Are Endangering State Budgets. Good Jobs First.
U.S. Census Bureau. (2025). Employment in Data Centers Increased by More Than 60% From 2016 to 2023 But Growth Was Uneven Across the United States.
Louisville Community Data Center Cooperative Feasibility Study (NTARI, 2025)
4. Cooperative Ownership Structure
4.1 Governance Model
LLC-Managed Cooperative:
Cooperative LLC owns all equipment (zero member capital risk)
Members provide space, power, connectivity; receive monthly stipends from day one
One member, one vote on major decisions (expenditures >$500K, expansion, bylaws)
Professional staff manage operations; elected advisory board provides oversight
This hybrid model mirrors electric cooperative success: professional technical expertise with democratic accountability.[^20]
4.2 Historical Precedent
Electric cooperatives demonstrate distributed infrastructure viability over 90 years:
900 cooperatives across 48 states serving 42 million Americans
Coverage of 56% of U.S. landmass including 92% of persistent poverty counties
Democratic governance (one member, one vote) sustained over 90 years
Professional management with member oversight ensuring technical excellence
$111 billion annual GDP contribution with 68% ($75B) in local communities
623,000 jobs supported with 68% (424,000) in local communities[^21]
Credit unions ($1.9 trillion assets, 130 million members) and agricultural cooperatives (>$200 billion annual marketing) further validate cooperative models at scale.[^22]
4.3 Zero-Cost Participation
Members contribute existing resources (space, power, internet) without capital investment. Cooperative LLC provides:
All node hardware ($2,000-3,000)
Installation and configuration
Ongoing maintenance and monitoring
Insurance and liability coverage
Technical management and support
This eliminates the primary barrier to community ownership while creating immediate wealth through hosting stipends.
5. Implementation Framework
5.1 Three-Phase Deployment
Phase 1: Planning (6-12 months)
Feasibility study: Population density analysis, anchor customer identification, financial modeling, site assessment
Legal structure: Cooperative LLC formation, bylaws, operating agreements, member agreements
Capitalization: $3-5M from federal grants (NTIA BEAD $42.5B program),[^23] state broadband programs, municipal investment ($350K-500K), impact investors
Technical planning: Network topology design, NOC site selection, equipment procurement
Phase 2: Pilot (Months 12-24)
Deploy 50-100 nodes for technical validation across diverse residential and commercial settings
Secure anchor customer contracts ($2-3M annually) through competitive procurement
Establish Network Operations Center for 24/7 monitoring and support
Target metrics: >80% member satisfaction, >95% uptime, latency <10ms for 90% of traffic, validated unit economics
Phase 3: Scaling (Years 2-5)
Expand to 5,000 nodes across metropolitan area (500-1,000 nodes per year)
Break-even at 2,000 nodes (Year 3)
Begin member profit distributions (Year 4)
Total investment generates $23M-58M community benefit over 7 years
5.2 Prerequisites for Success
Population density: ~50,000+ residents in metro area (critical mass for anchor customers and member participation)
Anchor customers: Large institutions for revenue stability and baseline utilization guarantee
Initial capital: $3-5M from grants, municipal investment, impact investors (federal broadband programs prioritize cooperative models)
Technical capacity: Local IT workforce or partnership with technical assistance providers (electric co-op model of professional management)
Political will: Municipal support for permitting, procurement preferences for local providers, potential seed funding
5.3 Risk Mitigation
Technical Risks:
Mitigation: Proven open-source stack (Kubernetes, Ceph), pilot validation before scaling, redundant architecture
Contingency: Equipment failure rates <5% annually with 4-hour replacement SLA
Financial Risks:
Mitigation: Anchor customer strategy ensures baseline revenue, diversified customer base across sectors
Contingency: Break-even achievable at 50% utilization (10-point safety margin), profitability at 1,500 nodes (25% participation shortfall tolerance)
Market Risks:
Mitigation: Edge computing growth at 33-38% CAGR provides tailwinds, 10-20x latency advantage creates competitive moat
Contingency: Can compete on cost parity without latency premium if market conditions require
Regulatory Risks:
Mitigation: Operate as edge computing provider (not common carrier), no telecom regulation exposure
Contingency: Municipal cooperation on permitting through economic development agreements
6. Conclusion
Community mesh networks demonstrate that internet infrastructure can operate as a profitable public utility using proven cooperative models. Technical architecture delivers projected 10-20x latency reduction (requiring pilot validation) enabling next-generation applications. Economic model generates $10M annually at 5,000-node scale with 50-80x better ROI than traditional subsidies based on verified cost data from Good Jobs First research. Cooperative ownership builds $4,000 per household wealth over 7 years while eliminating shareholder extraction.
Electric cooperatives serving 42 million Americans prove distributed infrastructure succeeds under democratic governance for nearly a century. The same principles—member ownership, professional management, local wealth retention—apply to digital infrastructure. Edge computing market growth (from $24B in 2024 to $180-250B by 2030) provides tailwinds supporting financial viability.
The Louisville metropolitan area feasibility study demonstrates practical implementation: validated anchor customer demand ($2-3M annually), demonstrated participation rates (5,000 households from 127,000 suitable), and conservative financial projections that survive stress testing. The model generalizes to metropolitan areas ranging from 250K to 5M population with appropriate scaling adjustments.
The question "Who owns the internet?" determines whether digital infrastructure serves democratic governance or technocratic extraction. Cooperative mesh networks return infrastructure to community control, distributing power architecturally and economically.
Public utility internet is technically feasible, economically viable, and operationally proven.
Year 1 pilot deployment will validate technical performance projections before full-scale deployment, ensuring evidence-based decision-making throughout the implementation process.
References
[^1]: Baran, P. (1964). On Distributed Communications. RAND Corporation.
[^2]: Canalys Research Group (2024). Global Cloud Services Market Share Q3 2024; Statista (2025). Cloud Infrastructure Services Market Share Q2 2025. AWS holds 31%, Microsoft Azure 24-25%, Google Cloud 11%, with remaining market fragmented among smaller providers.
[^3]: Tarczynska, K. (2016). Money Lost to the Cloud: How Data Centers Benefit from State and Local Government Subsidies. Good Jobs First. Average cost per job $1.95 million across 11 megadeals; Good Jobs First (2024). New Data on Data Center Subsidies, Same Old Problems. Illinois data centers: $1.4M per job average, Digital Realty: $2M+ per job; LeRoy, G., & Tarczynska, K. (2025). Cloudy with a Loss of Spending Control: How Data Centers Are Endangering State Budgets. Good Jobs First. Texas: $1B annually, Virginia: $732M annually, Illinois: $370M annually.
[^4]: National Rural Electric Cooperative Association. (2023). Economic Powerhouses: The Economic Impacts of America's Electric Cooperatives. Strategen Consulting. https://www.electric.coop/electric-co-ops-support-nearly-623000-jobs-contribute-111-billion-to-u-s-gdp-annually-report-finds
[^5]: Shi, W., Cao, J., Zhang, Q., Li, Y., & Xu, L. (2016). Edge Computing: Vision and Challenges. IEEE Internet of Things Journal, 3(5), 637-646.
[^6]: Fortune Business Insights. (2024). Edge Computing Market Size, Share & Trends Analysis Report. Market valued at $10.11B in 2023, projected to reach $181.96B by 2032; Grand View Research. (2024). Edge Computing Market Size, Share & Trends Report, 2033. Market estimated at $23.65B in 2024, expected to reach $327.79B in 2033; Mordor Intelligence. (2025). Edge Computing Market Size, Trends, Forecast Report | Industry 2030. Market estimated at $227.80B in 2025, projected to reach $424.15B by 2030 at 13.24% CAGR.
[^7]: Castro, M., & Liskov, B. (2002). Practical Byzantine Fault Tolerance and Proactive Recovery. ACM Transactions on Computer Systems, 20(4), 398-461.
[^8]: Kubernetes Documentation. (2024). Kubernetes Concepts. https://kubernetes.io/docs/concepts/
[^9]: Weil, S. A., Brandt, S. A., Miller, E. L., Long, D. D., & Maltzahn, C. (2006). Ceph: A Scalable, High-Performance Distributed File System. OSDI, 6, 307-320.
[^10]: Sailer, R., Zhang, X., Jaeger, T., & van Doorn, L. (2004). Design and Implementation of a TCG-based Integrity Measurement Architecture. USENIX Security Symposium, 13, 223-238.
[^11]: Uptime Institute. (2022). Global Data Center Survey. Average PUE across facilities ranges from 1.67 to 2.01.
[^12]: Satyanarayanan, M. (2017). The Emergence of Edge Computing. Computer, 50(1), 30-39.
[^13]: Masanet, E., Shehabi, A., Lei, N., Smith, S., & Koomey, J. (2020). Recalibrating Global Data Center Energy-Use Estimates. Science, 367(6481), 984-986.
[^14]: S&P Dow Jones Indices. (2023). S&P 500 Historical Returns. 2015-2022 CAGR approximately 10%.
[^15]: Ecosystem Marketplace. (2023). Voluntary Carbon Market Dashboard. Average credit prices $10-15/metric ton.
[^16]: Tarczynska, K. (2016). Money Lost to the Cloud: How Data Centers Benefit from State and Local Government Subsidies. Good Jobs First. Analysis of 11 megadeals averaging $1.95M per job; Good Jobs First. (2024). New Data on Data Center Subsidies, Same Old Problems. Illinois data centers (2019-2024): $468M in subsidies for 339 jobs = $1.4M per job; Digital Realty Franklin Park: over $2M per job; Columbus OH Google facility: $54.3M for 20 jobs = $2.7M per job.
[^17]: LeRoy, G., & Tarczynska, K. (2025). Cloudy with a Loss of Spending Control: How Data Centers Are Endangering State Budgets. Good Jobs First. Texas FY 2025: $1 billion; Virginia 2024: $732 million; Illinois: $370 million annually.
[^18]: Carl Vinson Institute of Government, University of Georgia. (2022). Data Center Tax Incentive Evaluation Study. Found revenue forgone exceeded tax revenue generated, resulting in net negative state fiscal impact; TIME Magazine. (2025). Why Tax Breaks for AI Data Centers Could Backfire on States.
[^19]: Mortenson Construction. (2023). Data Center Site Selection Survey. Only 3% of data center owners cited tax incentives as primary site selection factor; access to cheap electricity and available land were primary drivers.
[^20]: National Rural Electric Cooperative Association. (2024). Electric Co-op Facts & Figures. https://www.electric.coop/electric-cooperative-fact-sheet
[^21]: NRECA (2023). Economic Powerhouses: The Economic Impacts of America's Electric Cooperatives. Strategen Consulting. 900 cooperatives, 42 million Americans, 56% of landmass, $111B GDP contribution with $75B (68%) in local communities, 623,000 jobs with 424,000 (68%) in local communities.
[^22]: National Credit Union Administration. (2023). Credit Union Statistics; USDA. (2023). Cooperative Statistics.
[^23]: National Telecommunications and Information Administration. (2023). Broadband Equity, Access, and Deployment (BEAD) Program. https://broadbandusa.ntia.doc.gov/funding-programs/broadband-equity-access-and-deployment-bead-program Document Number: P3-011 (Technical Brief)License: AGPL-3.0 Contact: info@ntari.org | https://ntari.org© 2025 Network Theory Applied Research Institute, Inc.
Comments