Why examine the contracting chain?
A tower project can pass through several companies before the crew reaches the site. The accountability question is concrete: can the people commissioning and supervising that work identify the crew, the applicable requirements and the evidence supporting the assignment? Federal guidance has described weaknesses in this chain. A useful institutional response should make those responsibilities inspectable, then test whether the proposed system works.
This brief separates documented concerns from claims the evidence does not yet establish. It proposes a bounded records study for ATG; it does not report a completed field audit, demonstrate a safety benefit or announce an operating program.
What the fatality record supports
OSHA’s tower overview reports 13 communication tower-related fatalities in 2013 and 12 fatalities at tower worksites in 2014. Those are historical counts, not a current annual rate. They show why tower safety drew sustained federal attention, but cannot establish how likely a climber is to die on the job today. (OSHA tower overview)
A widely repeated comparison also needs its original date and denominator. In February 2014, OSHA’s David Michaels described tower workers’ fatal-injury risk as perhaps 25 to 30 times that of the average American worker. The agency presented a qualified estimate, not a current comparison with construction workers. This brief does not adopt that multiplier as a present-day headline. (DOL, February 2014)
Newer data do not remove the classification problem. The BLS 2024 industry table records 29 fatal occupational injuries in power and communication line and related structures construction, NAICS 23713. That category covers more than tower work. It cannot be substituted for a tower-climber count or divided by a differently defined workforce to produce a credible tower rate. (BLS fatal injuries, 2024)
BLS warns that detailed counts may not sum to broader categories and that classification changes require care in comparisons. A defensible trend needs a stable scope, reference period and exposure denominator. Until those are established, ATG should state what each source counts rather than manufacture a single industry risk statistic. (BLS definitions)
The concern is also organizational
The 2017 joint OSHA–FCC Communication Tower Best Practices guide describes contracting arrangements in which carriers and tower owners may not know who is doing the work or when. It recommends clear responsibilities across contracting tiers, contractor verification and access to training records. Its account supplies a reason to examine traceability; it does not measure how widespread that problem is today or prove that a registry alone prevents deaths. The guide is advisory, not a new regulation. (OSHA–FCC guidance, 2017)
A record must answer a decision
For this brief, traceability means being able to connect a defined work assignment to the responsible organizations, the applicable evidence and the decision made from it. This is an analytical definition, not a claim that ATG already provides this service.
The distinction matters. A credential can show that its issuer recorded a particular qualification under specified conditions. It does not, by itself, establish every requirement for a particular site, task or assignment. BLS describes several different occupations under telecommunications technicians and notes both formal education and on-the-job training. A single undifferentiated label would hide that variation. (BLS telecommunications occupations)
OSHA’s recommended practices for multiemployer worksites emphasize exchanging hazard information before work starts and when conditions change, coordinating policies, and identifying who implements the relevant parts of the safety program. A records system should support that coordination without displacing employers’ responsibilities or qualified site-specific judgment. (OSHA multiemployer guidance)
The following questions define a proposed documentary test, not a safety procedure or a readiness authorization.
- Who is doing the work
- Named employing organization and responsible contracting tier, linked to the particular assignment. Use protected identifiers where personal names are unnecessary.
- What requirements apply
- Dated task and site scope, the source of each requirement, and a record of who determined applicability.
- What evidence was accepted
- Issuer, scope, currency and verification result for relevant records, with exceptions visible rather than silently treated as complete.
- Who decided and why
- Decision-maker, date, evidence considered, restrictions and the route for correction or review.
- What changed afterward
- Revisions, unresolved concerns and the handoff to the next responsible party, preserving the prior record.
A missing record is a finding about the reviewed file, not automatic proof that training never occurred. An apparently complete file is not proof of competent performance. Both distinctions belong in the published results.
Where this lands in ATG’s design
ATG’s Public Charter reading guide connects workforce development with safety and traceability. It describes independent evaluation and a separate appeal process, alongside prohibitions on apprentice participation charges, personal advancement-evaluation charges, wage deductions funding ATG and placement fees. The proposed funding model is intended to cover instruction, equipment, supervision and administration without apprentice charges. Those are design commitments to test against implementation, not evidence of an operating funded intake. (ATG Public Charter; ATG funding model)
ATG’s apprenticeship is proposed and not yet registered; enrollment is not open. Existing educators, employers and workforce organizations are essential to testing the design. The practical question is how evidence can travel reliably between them while protecting workers’ access and review rights.
A bounded study before a larger claim
The next useful step is a documented, consent-based study of a clearly defined set of tower work packages. Before collecting anything, the study would need a written protocol, participating organizations’ permission, minimum necessary data access, privacy controls and an agreed review method. No collection or outreach is initiated by this brief.
- Define the sample and limits before reviewing files. Record how projects were selected, the period covered, the contracting tiers represented and exclusions. Do not call a convenience sample representative of the industry.
- Use a shared coding sheet. Two reviewers should independently assess a subset, record disagreements and reconcile them against published definitions. Separate missing, contradictory, out-of-scope and unverifiable evidence.
- Measure record retrieval time, the proportion of sampled decisions with a reconstructable evidence trail, and unresolved discrepancies. Report denominators and limitations. Do not translate those measures into a claimed reduction in injuries.
- Invite participating workers and organizations to correct factual errors before any attributable publication. Report aggregate findings unless explicit permission supports identification. Document changes and preserve an appeal route.
A later evaluation could test whether improved traceability changes assignment quality, administrative effort or other predefined outcomes. Safety impact would require a suitable design, credible exposure data and enough observations. The first deliverable should be an honest account of what can and cannot be reconstructed from the evidence, not a promised result.
Next evidence questions
A follow-on credential map should identify requirements by role, buyer, source and date rather than present nineteen as a universal count. Research into safety outcomes would need to establish which mechanisms were involved, not attribute fatalities to a missing credential by assumption. For now, a developing institution can set a concrete standard for its own design: make decisions reconstructable, responsibility visible and errors correctable.
Sources and scope
Sources checked 3 November 2026. Government material is identified by its publication or data year. ATG guides describe its proposed design; they are not independent evaluations of that design.
OSHA Communication Towers Overview. Historical tower fatality counts and hazard context. The page does not supply a current tower fatality rate. Open source ↗
U.S. Department of Labor news release February 25 2014. David Michaels’s qualified 25–30 comparison was with the average American worker. Retained as dated context, not a current risk estimate. Open source ↗
BLS Table A1 Fatal occupational injuries by industry and event 2024. NAICS 23713 reports 29 deaths in power and communication line and related structures construction. This is broader than tower work. Open source ↗
BLS Census of Fatal Occupational Injuries Definitions. Classification, coding and aggregation cautions. Occupational and industry categories are not interchangeable. Open source ↗
OSHA and FCC Communication Tower Best Practices June 2017. Printed pages 1–5 cover contracting structure and verification. Advisory guidance, not a new standard or an estimate of current prevalence. Open source ↗
BLS Occupational Outlook Handbook Telecommunications Technicians. Occupational differences and training context. Not a buyer-specific credential inventory. Open source ↗
OSHA Communication and Coordination for Host Employers Contractors and Staffing Agencies. Recommended practices for exchanging information and assigning responsibilities at multiemployer worksites. Open source ↗
ATG Public Charter public reading guide source v3. ATG’s stated commitments. Public guide updated October 15, 2026; implementation and operating status are separate questions. Open source ↗
ATG How the Model Is Funded public reading guide source v3. Proposed funding design, not proof of funds awarded or an approved operating intake. Open source ↗