For marine contractor Dutra Group, the $8.9-million pier substructure repair project put out to bid by the Port of San Francisco last year fit precisely in the company’s sweet spot. As did other bidders, Dutra ponied up reams of paperwork on everything from local hiring to required environmental remediation work in hopes of winning the project.
In a rejection letter sent to the San Rafael, Calif.-based firm, port officials cited one small part of the requested information: Dutra’s workers’ compensation experience modification rating, or EMR. That number, determined by dividing a company’s actual workers’ comp claim losses by the expected losses for the company’s specific occupations, plays a vital role in determining the premium a company pays. The EMR is used to adjust the insurer’s premium charged for each occupation. It is designed for insurance underwriting and serves as an incentive to work more safely. An EMR of 1 is average for a work category and will have no effect on the insurance premium. An EMR higher than 1 adds to an employer’s premium, and lower than 1 saves the employer money.
But state and other public authorities, major utilities, private project owners and prime contractors for years now have used the rating as a one-size-fits-all way to gauge a contractor’s safety record. Three consecutive years of EMR are usually considered, excluding the most recent year. Some treat a company with one EMR blip as a contracting untouchable.
The port’s contracting staff saw Dutra’s EMR had skipped slightly above the very low maximum number it had set, and that was that. Staff also pointed out that Dutra was not the low bidder, so it wouldn’t have won anyway. But the injustice stung a couple of other bidders in addition to Dutra, and they sought exemptions from the port’s unbendable EMR threshold. They had all submitted an EMR just above the port number, which effectively took them out of contention.
A more comprehensive evaluation of safety performance usually includes other data and information—such as the reason for the higher EMR, style of the safety program, qualifications of the safety managers or U.S. Occupational Safety and Health Administration (OSHA) data used to calculate rate of injuries per hours worked—a different type of statistic. But the port “didn’t require other information like OSHA 300s,” says Cliff Hunt, Dutra risk manager and safety director, referring to the injuries log. The port’s procurement “focused on the EMR. It was black and white.”
Dutra’s experience is one of the more extreme examples of what employers and insurance agents say is regular misuse of EMRs, a tool for loss control and premium setting, to prequalify contractors.
In an environment now obsessed with information, simplifications and misuses of statistics shouldn’t be surprising. EMRs were born out of a need for better actuarial precision in workers’ compensation insurance—a no-fault form of coverage—but they now have become an informal proxy for the success and quality of an employer’s overall safety practices. And EMR misuse has often penalized smaller companies and those that have good safety records.
Across the country, the severity and frequency of workers’ comp claims have trended steadily downward. Benefits have been cut in many states and companies have worked hard on safety training and on return-to-work programs. Insurers have made profits on their workers’ comp underwriting for the past three years.
As contractors carefully managed insurance costs, EMRs—also called E-mods—became a focus of the effort to control bid prices. They are sometimes treated as vital to prequalification. One sign of that trend occurred this year when the National Council on Compensation Insurance (NCCI)—the insurer-owned workers’ comp rate-making entity covering most states—added to its published material a boilerplate disclaimer against using EMRs to prequalify employers.
“This new language is designed to raise awareness on this important topic and to reinforce the intended purpose of an experience rating worksheet” for using the rate only to adjust the premium, according to Kathy Antonello, NCCI’s chief actuary. “This is consistent with the information we have provided to the public and the industry that it’s not appropriate to use E-mods to compare the relative safety of employers. The E-mod should be used for its intended purpose.”
Changing the deeply entrenched practice of using EMRs to prequalify contractors may not be as simple as writing a few articles and publishing a disclaimer, however. Companies that understand the value placed by owners and prime contractors on safety know that a low EMR impresses them as much as do held-down costs. “It will take a generation” to get out of the habit of using EMRs as a quick take on a company’s safety record, says one industrial construction contractor’s safety director. “There are union shop stewards who talk about EMRs on a regular basis.”
At a recent construction conference of the International Risk Management Institute, where Antonello, in a presentation, repeated cautions against using EMRs to prequalify, an insurance director at a large construction firm turned to someone sitting near him and said, “We use our EMR to compete all the time.”
Smaller contractors are the biggest losers in the “EMR equals safety” mindset, insurance brokers report. Fewer hours worked maximizes the effect of a single claim.
“Setting these strict experience mod-factor thresholds can discriminate against smaller or midsized companies, disqualifying them from bidding on projects, despite their having an excellent safety and injury history,” wrote Sonja Guenther, a workers’ compensation specialist at Denver financial consultant IMA Financial Group, in a recent report on the issue. Larger contractors, with the help of consultants, can find ways to drive down their EMR that have little if anything to do with their actual safety record, critics add.
Still, the problem is not necessarily using EMRs to help evaluate a company’s safety record, but rather relying too heavily on the metric as a factor in a bid decision—or as the deciding one. A company’s EMR, considered in conjunction with accidents reported to OSHA and other data, can provide valuable insights, safety experts say. Since 1999, the California state code requires that a bidder’s EMR must be considered with other safety information.
For Dutra, a large company with a long history, disqualification from the San Francisco port project came as an unpleasant surprise. The contract project manual for the substructure repair work on piers 29 and 31.5 stated that, in addition to having the specified maximum EMR, the bidders’ safety history and qualifications would be considered. The manual also directed bidders to submit the firm’s plan for accident prevention and managing subcontractor safety. And the manual mentioned that a negative safety review could disqualify the bidder. The issue came up at the pre-bid meeting in January 2018, when one potential bidder asked if there would be flexibility and a port official told bidders to submit questions individually. Dutra says it had an excellent series of EMRs for the years leading up to the pier project—the company declines to reveal them—and was only slightly above the very low 0.8 requirement the port had set.
In discussing the situation with port officials, Dutra staff felt assured that EMRs would not be used as an instant disqualifier. Company managers told the port officials that Dutra’s actual workload is a mix of mostly marine construction, with only 15% on land. Work on water is covered by an entirely different workers’ comp system under the U.S. Longshoreman and Harborworkers Act. More to the point, Dutra’s EMR says nothing about its safety record on marine projects like the pier repairs. But when the rejection letter came, the port’s chief harbor engineer informed Dutra it was disqualified because the firm’s EMR did not meet the 0.8 criteria over the previous three years. At least two other bidders were rejected for the same reason. Port officials declined to comment on the issue to ENR.
Prime contractors have used EMRs as prequalification litmus tests, too. The realization that it was unfair jolted Vincent DeLeonardis, president of Pontiac, Mich., general contractor George W. Auch Co., to change its practices. Until then, his company automatically disqualified subs with an EMR higher than 1.0.
For DeLeonardis, the turning point came when he was forced to disqualify a tile subcontractor when it turned out its EMR was above the threshold Auch was using at the time.
“I remember I was on vacation and calling this guy from the airport,” DeLeonardis recalled. “He was madder than a hornet as he explained why his EMR was above” what Auch required. The sub’s explanation turned out to be eye-opening: It had been hit with three claims, one of which was a repeat and another potentially fraudulent.
“Three incidents … and boom, this guy is over 1.0 and can’t do our work,” DeLeonardis said. After that incident, he dropped the practice of relying on the EMR as a decisive element to prequalify its subcontractors. Under the new policy, Auch treats an EMR of 1.9 as a red flag but not “a poison pill,” he said.
The injustice has been examined by others and mentioned in workers’ comp studies and research for decades. But so far only Virginia legislators, lobbied by insurance agents, have passed a law to stop EMR misuse for both public and private construction (see p. 24).
Indiana tackled the problem before Virginia voted on the subject. Indiana investigated EMR misuse in 2011, raising awareness of the problem surrounding it.
A report by the Indiana Compensation Rating Bureau warned against EMR use as a make-or-break metric, noting it had become common practice for construction owners and managers to use a rating of 1.0 or even 0.90 or lower to prequalify contractors.
The report highlighted stories of firms boxed out of bidding for jobs because of the practice. One contractor said his firm saw its superior EMR rating in the 0.82-range jump to 1.07 after a single claim—for a worker suffering a hernia.
Prime contractors that use an EMR of 1.0 or above to screen out subs in the prequalification process could “ruin” a company, one sub stated in the Indiana report.
Another contractor told the story of suddenly facing automatic disqualification and financial disaster after his site supervisor suffered back and neck injuries in a vehicle rear-ending. The firm’s workers’ comp insurer picked up the tab, pushing the contractor’s EMR past 1.0.
“Through no fault of my business, and through no lack of safety management by me and my employees, we are virtually out of business,” the contractor wrote. “No one has yet explained to me how that’s fair or even reasonable.” While no legislative remedy was attempted, the Indiana report—along with a series of bureau workshops and discussions about the issue—helped raise awareness.
NCCI helps set EMR rates for 35 states, with others, like California and Indiana, relying on their own quasi-public agencies. But because insurance is regulated at the state level, each one applies a different rule. “Every state is uniquely different with different cultures and workers’ comp benefit structure,” said Guenther. An EMR calculation rule might be determined in a state insurance regulation or in a filing by NCCI or another rating entity. For example, a contractor that hops from Colorado to Kansas can “drive a lower mod rate,” Guenther pointed out, because Kansas has a plan where 70% of every dollar for medical-only injuries spent by the insurer is removed from the EMR calculation. States that operate in this fashion are called “rating adjustment states.”
While there are many permutations in calculating EMRs, with several weighting and adjustment factors added to actual and expected losses, the crucial point is this: A contractor’s EMR measures a particular company’s losses against those of others whose staffers work in the same trades or occupations. That, in turn, can lead to some statistical distortions that can that can make a firm’s safety record appear weakened when, in fact, there has been no change at all.
For example, if the pool of all company losses improves, one company’s EMR may deteriorate simply because it no longer is rated as highly as it was before against the losses for the entire trade. On the other hand, there is a myth that as the EMR goes up, the premium rises in tandem. Not so, according to Joe Tiernan, a shareholder in the Kansas City, Mo., office of Holmes Murphy, an insurance broker. “If you’re an organization that displays a culture of safety, a desire to continually improve and invest in the staff’s well-being, a good carrier and broker will work with you leading up to your renewal to offset the financial impact of an increased EMR. A higher EMR doesn’t unequivocally equate to an increased workers’ comp rate per man hour.”
Smaller contractors in all cases are more vulnerable to such ebbs and flows, and losses and claims loom larger for them compared to more sizable peers. In general, one loss can put a contractor with $3 million to $15 million in annual revenue above a 1, says Tiernan.
Others also noted that size makes a difference. A big construction management firm probably doesn’t have a lot of at-risk employees, notes a retired construction executive. “Their employees are in the trailers,” while smaller companies are likely to have leaner administrative staffs with more workers in the field.
Business cycles alone can shift EMRs. In a downturn, fewer hours worked make any claim loom larger. In some cases, there is simply a mathematical limit to how low a smaller contractor can push down its EMR compared to larger competitors.
Guenther, the workers’ compensation specialist, noted in a company paper that one client wound up disqualified from bidding because its EMR was just over the very low 0.8 required by the project owner. But the contractor had not had an injury in six years. “Under the current actuarial formula, this contractor was at their lowest possible mod. It was loss-free,” she wrote.
Still another complication involves the use of online prequalification systems. While they provide a valuable service to owners and prime contractors, they are also a mystifying bureaucratic barrier that stands between a subcontractor or contractor and the entity to which it is bidding.
That can be a big problem if the EMR needs explaining, according to Tiernan. “When this occurs, figuring out how to have a one-to-one conversation with a real human representative of a big manufacturer or industry company is very difficult, particularly in a reasonable timeframe,” he contends. Several of the prequalification companies contacted by ENR did not respond to requests for comment.
Gaming the System
There are also methods for larger contractors, with the help of consultants, to game the system and drive down their EMR in ways that have no relationship to their actual safety record. The most important way is what insiders call “buying down the mod.” This refers to the unfair advantages some contractors gain in net reporting states—where their payment of an insurance deductible is not counted towards the EMR. Guenther notes that 15 to 20 states have net reporting, in which the deductible paid by the employer is taken out of the losses before the rate is calculated. In Colorado, under a recently passed statute, employers can absorb $17,000 as a deductible and “keep that out of their mod factor,” she says.
What’s unfair, Guenther adds, is that one company may have a $17,000 deductible while another only can absorb $1,000 of the cost of a claim. Still, although it puts contractors on a different footing, net reporting, or taking out the deductible before calculating the EMR, does mean that employers at least “have a financial stake in their injuries,” she says.
There are deceptions possible—such as when an employer changes its name, buys or sells the firm or creates a holding company. That can reset a contractor’s EMR back down to 1, according to Tim Ervine, manager for environment, health and safety at Duke Energy.
“You could have someone with an EMR of 5, and once that name change takes place, you are back to 1,” he says. “You could inadvertently be bringing in a bad apple.”
Some companies unfairly take advantage of such potential loopholes, says Ervine, who helps scrutinize the safety records of contractors hired by the Charlotte, N.C.-based utility in its six-state region.
Dutra’s Hunt says there is another way companies try to improperly improve an EMR—concealing accidents and trying to take care of them off the books. But that also risks severe state fines and penalties for violating the law.
The level of effort is high for a contractor that wants to assure the accuracy of its workers’ comp. The firm must check that the insurer properly classifies its operations and trades and must review open claims with insurers to make sure the settled amount is accurately reflected in the EMR. One alternative to EMRs was proposed in 2012 by Bradley Hartmann, president of Red Angle, a training company. He proposed banishing EMR “to the quants” at NCCI and shifting industry practice toward tracking individual worker histories and calculating “RIDs,” a rate of injury and death.
When it comes to fairer prequalification, there are hopeful signs. Some state agencies are using a higher threshold when it comes to the cut-off EMR while also looking at other safety performance evidence.
California’s Dept. of Water Resources has established a 1.25 EMR as a goal, but not as an absolute qualifier. Contractors that come in above that number—which is considerably higher than the 0.8 used by the Port of San Francisco—have to hire a full-time, on-site safety official for the project, bidding documents show. The Maine Dept. of Transportation sets the goal at 1.5.
An EMR above that still doesn’t automatically bar a contractor from successfully bidding on state transportation projects, although contract conditions might include having an independent safety consultant on the worksite at all times, according to George MacDougall, contracts and specifications engineer for the Maine DOT. A rate of 1.25 to 1.49 would also be “concerning,” he noted, and would require a sit-down meeting with the company to understand what it is doing “to address the poor safety experience.” The contractor isn’t automatically knocked out of contention, however.
Indiana rating bureau President Ron Cooper says that while its study and workshops helped raise awareness in the state, he still gets four or five complaints each year from contractors who say they were unfairly boxed out of bidding by their EMR. But he is optimistic that the outreach helped raise awareness about extenuating EMR circumstances. “At least,” he says, “the door was opened to having a conversation—where it wasn’t before.”