Bogus BLS averages put larger wireless contractors at a greater risk of being red-flagged

In Featured News by Wireless Estimator

For years

For decades the U.S. Bureau of Labor Statistics’ (BLS) averages for Total Recordable Incident Rates to assess safety compliance have been the gospel. A scientific study claims they are garbage.

Avetta and other compliance vendors continue to use the U.S. Department of Labor’s Bureau of Labor Statistics (BLS) data to initially assess the safety culture and performance of wireless contractors through metrics such as the Total Recordable Incident Rate (TRIR) average. Compliance vendors red flag contractors if their TRIR exceeds the BLS’s current TRIR averages, even though they are aware that the TRIR is a statistically invalid, unsupportable metric which  forces clients, based upon an algorithm, to make business decisions that can impact a contractor and the company’s workforce.

Why? Possibly because contractor management through assessments of other lagging indicators, on-site reviews, outreach, and inspections would require compliance vendors to increase their staffing with individuals knowledgeable of wireless construction, at a considerable cost that the client or contractor might decline to absorb.

A short safety compliance history

In 1983 during the tremendous analog cellular system network buildouts across America, carriers, towercos, and general contractors assessed their subcontractors’ safety compliance the old-fashioned way: 1. Send in a certificate of insurance; 2. Mail us a copy of your training manual binder.

The manuals went into a file cabinet and were later dumped in a warehouse, and Acord insurance forms frequently rested without renewal for a year or so.

In the early 1990s when 2G debuted, compliance administrators were reviewing Experience Modification Rates (EMR), as did the contractor who oftentimes went into sticker shock when their workers compensation audit resulted in skyrocketing premiums, and on occasion were removed by clients when their rating went north of 1.0.

Today, Avetta has a near-monopoly

However, in past years, risk management executives bought Avetta’s pitch that it is more economical to have their compliance needs handled by the nation’s largest firm rather than hire more staff to improve safety.

In 2014, SBA said their current system of managing over 2,500 contractors wasn’t working since they didn’t have the database and platform and couldn’t rely upon manual checking. They then joined Avetta, acknowledging that they immediately were saving $100,000 in employee costs.

Coupled with an ambitious marketing team and a universal acknowledgment that fast-paced industry builds required outside compliance management, Avetta landed T-Mobile, Verizon, Vertical Bridge, InSite Wireless Group, Ericsson U.S., and others. Their most recent carrier is Dish Wireless. Also, Avetta acquired vendor management firm Browz, porting Crown Castle and others into their platform.

TRIR is weighted in favor of smaller contractors

A contractor’s TRIR score is based upon recordable incidents and the number of worker-hours.

As an example, if Mark’s Montana Maintenance has two recordable incidents over 140,000 worker-hours for his 49 employees, the company would have a safety performance TRIR of 2.86 per 200,000 hours. TRIR = Number of Recordable Incidents (2) x 200,000 divided by Number of Worker Hours (140,000).

The BLS’s 2020 average for NAICS 237130, Power and communication line and related structures construction, which is the NAICS most frequently used by contractors, for companies with 11 to 49 employees is 3.6. Therefore, Montana Maintenance’s 2.86 would not trigger a red flag, and it would appear that Mark has his employees focused on safety.

Tom gets taken over the coals for his size

If larger contractor Tom’s Towers, with 100 employees, has double the worker-hours at 280,000 as well as double the reported incidents at 4, they too would be at an acceptable 2.86. TRIR = Number of Recordable Incidents (4) x 200,000 divided by Number of Worker Hours (280,000).

Whereas 2.86 is a complimentary average for Mark, it’s red meat for Avetta’s algorithm that will red flag Tom’s Towers to every one of his company’s customers.

Why? The BLS’s 2020 NAICS 237130 TRIR average is 1.7 for Tom’s company size range (50 to 249).

One would presume that the BLS’s methodology is to access OSHA 300 logs and accurately identify TRIR averages to multiple decimal points, leaving some to embrace the belief that larger companies are safer.

That’s not true. and the BLS’s TRIR averages for wireless contracting, a small subset industry of construction, are flawed and meaningless and should be abandoned, according to many company owners and industry observers.

Huge TRIR dip in 2020 is questioned

The BLS does not access OSHA 300 logs. Instead, it relies upon, in part, OSHA-mandated surveys of companies randomly selected as part of a probability survey of about 230,000 establishments. The BLS says businesses selected for their survey are separated into groups according to their location, industry, and number of employees. They believe that they are getting a representative sampling.

Contractors have registered their concerns with Wireless Estimator as to the BLS’s 2020 TRIR averages, especially in the company employee range of 50 to 249, where the average dropped dramatically to 1.7 from 2019’s 2.9 which was close to the average for 2018 and 2017.

Although the BLS stated during 2020, there were 2.7 million workplace injuries and illnesses, down from 2.8 million in 2019, it was only 5.7 percent, and the drop was dwarfed by the 40 percent decline in NAICS 237130.

Contractors also said that the decline could result from contractors not reporting minor injuries to avoid being red-flagged.

The BLS’s sister agency, OSHA, also believes that injuries and illnesses are being underreported.

A Monday article in Wireless Estimator on a proposed new OSHA rule for heat stress highlighted an underlying concern that applies to TRIR. OSHA believes that BLS recordable incidents are vastly unreported.

One company owner said, “In the era of third-party vetting, the industry has begun non-reporting, especially the small incidents where the worker went right back to work. My concern is what will happen when the TRIR average keeps dropping, and then even those not reporting the small stuff have to make decisions on not reporting bigger incidents that push them over the new lower industry average?”

Wireless contracting NAICS is shared with larger non-industry companies

Although cellular phone tower construction, radio transmitting tower construction, and communication tower construction are examples of businesses operating under NAICS 237130, they are primarily the smaller companies providing wireless infrastructure services other than Mastec North America Inc. with 7,250 employees, many of them in their power and underground utilities division, and Dycom Industries, Inc. with 11,902 employees, a greater portion of them providing electrical contracting services.

There are over 2,700 businesses using the NAICS 237130 code.

It’s been estimated by industry observers that approximately 40% of all wireless contractors are below the 11-employee threshold, and their illnesses and injuries are not reported, further skewing TRIR averages.

Scientific study states that TRIR is useless for analysis of safety performance

A November 2020 scientific study of the validity of using TRIR as a measure of safety performance by the Construction Safety Research Alliance (CSRA) concluded that the use of TRIR metrics is not appropriate or meaningful.

The report’s key take-aways after analyzing 17 years of data and 3.2 trillion worker hours were:

1. There is no discernible association between total recordable incident rates and fatalities;
2. The occurrence of recordable injuries is almost entirely random;
3. TRIR is not precise and should not be communicated to multiple decimal points of precision; and
4. In nearly every practical circumstance, it is statistically invalid to use TRIR to compare companies, business units, projects or teams.

Since TRIR has remained the most prevalent measure of safety for nearly 50 years, the researchers said that they stand by their conclusions but welcome and recommend that other researchers should propose alternate hypotheses about TRIR.

Dr. Matthew Hallowell, the lead author of the study and the executive Director of CSRA as well a President’s Teaching Scholar and Endowed Professor of Construction Engineering at the University of Colorado, informed Wireless Estimator that to date, there have been no studies submitted.

However, he said that it takes many months to generate different models and results. CSRA’s study took about a year, he said.

There are remedies for red flags but they’re not always easy

Avetta’s platform will red flag a contractor when a TRIR exceeds the threshold set by a client.

Each client can designate a percentage that triggers a non-compliant condition and red flag. Companies range from less than or equal to 100% of the industry average to 120% of the industry average.

Some clients have representatives that a contractor can contact to discuss what the incident was that triggered non-compliance. If it’s acceptable, they then become compliant under a conditional waiver.

It sounds like a streamlined system, but frequently it’s not.

Recently, according to a number of contractors, they received red flags, reportedly due to new TRIR data being uploaded that affected the system and in some cases removing waivers.

Certificate of insurance issues regarding renewals, required new safety procedures, not in the contractors safety manual, and other compliance matters will trigger a red flag.

In some cases, contractors receive advance notices of a pending non-compliance issue.

Avetta’s policy is to respond to an issue in three days or less.

However, if a contractor is shut down for one day, it could cost the company thousands of dollars while administrators and project managers work towards mitigating the problem.

There will be an educational session at NATE UNITE 2022 in Las Vegas on February 23, addressing third-party compliance issues that will be moderated by Kathy Stieler, NATE Director of Safety, Health & Compliance. Ample time  has been set aside for questions from the audience and it is likely that issues with Avetta will be explored since they’re the industry’s largest compliance company.


In a Q&A exchange with Wireless Estimator yesterday, Scott Spencer, Avetta’s SVP of Global Strategic Accounts, provided some illuminating answers to contractors’ TRIR concerns and provided suggestions that could relieve TRIR averages from being a stand-alone indicator.

Spencer said incident rates should be accompanied with additional indicators, both lagging and leading, to more accurately measure safety performance. “

“We recently completed a joint project with NATE to improve the quality and scope of a safety management system audit that ties industry codes to risks and regulation. We’re also working with a key wireless client to perform site safety observations to ensure that what’s in the Safety Management System is reflected in the field,” Spencer said.

Avetta-Q1.  If it wasn’t underreporting, can Avetta identify what caused the NAICS 237130 (Power and Communication Line and Related Structures Construction) TRIR 2020 average to drop dramatically to 1.7 (Size: 50 – 249) from 2019’s 2.9, which has been close to the average for 2018 and 2017?

Avetta-AHere is an analysis of NAICS 237130 from the Avetta member network: 2019 – 3.2 TRIR Avetta average compared to BLS for all sizes at 2.5. 2020 – 2.96 TRIR Avetta average compared to BLS for all sizes at 1.9.

Our numbers are more closely aligned with the averages for 11-49 employee count. We aren’t seeing the dramatic drop in recordable incident rates that is being reported by the BLS.

Avetta-Q2.  Does an Avetta client such as SBA Communications, T-Mobile or Vertical Bridge have the ability to allow contractors to use multiple NAICS codes, or is NAICS 237130 utilized for convenience or continuity?

Avetta-AThe Avetta reporting system allows for a primary and secondary NAICS code. The BLS comparison is based on the primary NAICS code selected by the participating company. The primary code can be changed annually based on the evolving nature of a company’s portfolio of services. The participating company chooses the appropriate industry code.

Avetta-Q3.  As an example, if a multi-disciplined contractor also performs foundation services and they have a reportable incident, why should they be penalized by exceeding their 1.7 NAICS 237130 TRIR when some foundation-related TRIRs are at 4?

In that example, if the contractor’s primary industry was “foundation services” they would be measured against that BLS threshold.

Avetta-Q4.  Is it possible that any year’s BLS TRIR average is inaccurate since approximately 40% of the nation’s wireless contractors are below the 11-employee threshold and are not included in any BLS surveys or data collection, yet those contractors are – through subcontracting – performing services for Avetta’s clients?

Avetta-AWe cannot speak to BLS methodology. As a part of our new Avetta One product release we think we can provide some additional insights. New Subcontractor Management functionality will give our clients and prime contractors the ability to gather incident and accident data on all subcontractor performed services. Transparency into the next tiers of contractors is critical to ensuring better risk awareness and a safer work environment.

Avetta-Q5.  Since some contractors report their recordable incidents under NAICS 238120 (Structural Steel and Precast Concrete Contractors), and are paying a three times higher workers compensation rate to protect their workers, are those contractors allowed to use the TRIR average of 4.2 for 2020 (Size: 50-249) if they do cellular work? If not, why not?

Avetta-ASelection of the appropriate industry classification is at the discretion of the contractor. Neither Avetta nor its clients dictate which industry code is used. To help ensure that the correct code is used Avetta does verify, through audit and assessment, that the risks and regulation associated with the industry code selected align with the safety management system, training protocols, etc. We also verify the code against their public website and a listing of services provided outside of the NAICS code selected.

6.  Are there any modifiers that Avetta or its clients use to differentiate between employee size? If so, please explain.

Avetta-AAvetta does not employ a modifier, but our flagging system gives clients the ability to stratify or add weighting to compliance requirements based on employee counts or other variables. Instead of a strict hurdle rate that requires 100% of the BLS average = a pass (green flag), clients can add layers to the compliance calculation.

For example, a contractor with 10 or fewer employees could have a hurdle rate of 150 or 200% of the industry average. Multiple clients are also using our Scorecard functionality to bring TRIR in as one of many inputs that drives an overall scorecard flag. This allows them to target suppliers who have multiple deficiencies instead of relying solely on the TRIR figure for flagging.

Avetta-Q7.  Since Avetta is always assessing improvements for contractor safety, the company is likely aware of a scientific study published November 2020 by the Construction Safety Research Alliance (CSRA) that clearly states measuring safety performance by using TRIR averages is invalid, and recommended that new approaches to safety measurement are needed. Do you agree? Do you have any suggestions and/or alternatives like assessing workers compensation loss runs?

Avetta-AWe agree. The CSRA study states, “TRIR is almost always communicated as a precise number as if it was the only possible outcome… TRIR should be accompanied by a range of potential outcomes that the safety system could have reasonably produced.”

TRIR cannot be a stand-alone indicator of safety performance. Avetta believes that the TRIR must be accompanied by other lagging indicators like LTIR, DART, EMR, Loss Runs, Citation histories and severity. More importantly, leading safety indicators should be considered as future safety performance potential and awareness are measured. A company that limits safety qualification to the TRIR isn’t getting an accurate assessment of performance for organizations they wish to employ.

Avetta-Q8.  You’ll be announcing a new and improved solution to simplify supply chain risk management. Please identify how it will benefit one of the country’s most important, but smallest industries, wireless construction.



Here are the advantages of Avetta One that help the wireless supplier experience.

  1. Improved flagging and compliance tools like the scoring and weighting methods listed in our response above.
  2. Better visibility into worker qualifications and increased transparency to the worker level. The new workforce management tool will give wireless contractors a solution for employee certification and training management.
  3. Automated upload of insurance documentation. No longer will contractors have to upload and update insurance information. Simply send the Accord form or proof of insurance to Avetta, and we’ll take care of the rest.
  4. Subcontractor management, making it easier for suppliers to manage.