U.S. Rep. Elijah Cummings, D-Md. – the former chairman and current ranking Democrat on the House Oversight and Government Reform Committee – has some issues with American International Group. And generally speaking, we can’t blame him.

Back in 2009, he was one of the loudest congressional critics of the insurance giant’s federal bailout package, and particularly of compensation and perks that continued to flow to AIG executives and associates even after it was saved from extinction by the Federal Reserve and the U.S. Treasury.

But amid his crusade to lambaste the company, Cummings has stumbled across one particular business area where AIG has been a dominant player – providing workers’ compensation insurance to overseas government contractors – that seems to have led him down a misguided path. Concerned that workers comp insurers “have reaped enormous profits” by providing coverage to contractors who do business in Iraq and Afghanistan, Cummings is now proposing the federal government completely take over that market by providing coverage directly through a self-insurance program.

The federal government has been requiring government contractors to provide workers’ comp to their employees at overseas military bases going back to 1941’s Defense Base Act, which has since been expanded to require coverage for nearly all overseas contractors and subcontractors of any government agency. For decades, the DBA workers’ comp program was a tiny and insignificant one, with total premiums of just $18 million as recently as 2002. But with the enormous build-up of contractors in Iraq and Afghanistan over the past decade, premiums grew to more than $400 million by 2007.

So on what basis does Cummings make his assertion that insurers in the DBA program, which is dominated by four large carriers, are overcharging for their services? Mostly, it’s by comparing apples and pomegranates.

A breathless 2008 report commissioned by Cummings from Oversight Committee staff noted that:

The top four DBA insurance companies received $1.5 billion in premiums from 2002 through 2007 from contracts negotiated with private contractors. They will pay out $928 million in claims and expenses for injuries incurred under these policies and earn net underwriting gains of $585 million.

Someone even modestly familiar with property and casualty insurance will recognize that what is being described there – losses and loss-related expenses compared to premiums – are the components of the companies’ claims ratio, which, in this case, comes out to an aggregate 62%. In general, claims ratios of between 60% and 70% are pretty standard for workers’ comp, so this is hardly an unusual tally. But the report then offers a very strange point of comparison:

Collectively, the four companies reported underwriting losses of 1% since 2002 on workers’ compensation insurance other than DBA insurance. According to A.M Best, the largest DBA insurer, AIG, has retained underwriting gains of just 1% since 2002 from its workers’ compensation insurance. In comparison, AIG reported underwriting gains under the DBA program of 38% since 2002.

Here, the Oversight Committee staff appears to be reporting not the companies’ claims ratios, but their combined ratios — that is, the claims ratio combined with the expense ratio, which represents companies’ operating costs as a percentage of premiums. Workers’ comp operating ratios tend to run between 30% and 40%, so it would make sense that combined with claims ratios of between 60% and 70%, combined ratios would tend to hover around the break-even point of 100%.

Of course, looking at a small snapshot of workers’ comp claims for what are mostly new risks doesn’t exactly provide answers as to whether the rates being charged are “excessive.” Workers’ comp claims are long-tailed. Employees of overseas contractors are mostly young and healthy. Disabling injuries sustained by covered employees could last decades. Until the business is significantly more seasoned, there’s no way to say whether it will even prove profitable at all to these carriers, much less “excessively” so.

And there are other reasons it would be reasonable to expect that workers’ comp insurers covering DBA contractors would charge more for coverage than on domestic risks. Namely, that they are facing obstacles that domestic workers’ comp insurers generally don’t have to deal with. Among them is that overseas contractors are frequently operating in areas that lack basic medical care, thus often necessitating costly medical evacuations for treatment. Adjusting claims in overseas districts with foreign languages, customs and legal systems also can be costly and labor-intensive.

DBA insurers also are being asked to provide more or less round-the-clock coverage, as an overseas contractor’s “work day” doesn’t ever really end. And they are operating in areas and covering professions for which there is not a tremendous amount of credible actuarial data to construct rates. Given large areas of “known unknowns,” it would be irresponsible not to build in a risk load that anticipates potential problems.

While it is true that DBA insurers do not cover true “war risks,” which are covered by the federal government under the War Hazards Coverage Act, they do have to pay claims arising from war risks upfront and wait for reimbursement. There can be significant lags in reimbursement times, and there may not be total clarity about whether certain injuries are the war hazards or workers’ comp hazards. The reimbursements also don’t accrue interest, and don’t compensate insurers for the time value of money.

We wouldn’t disagree with Cummings if his point were simply that there are some clear problems in how the DBA program is administered. But these appear to have equally clear solutions.

For one, rather than simply requiring contractors to obtain workers’ comp from whatever carrier they like, contractors instead must choose from a list of carriers approved by the Department of Labor.  Besides the predictable effect that artificially limiting the pool of competitors is going to result in less robust competition, the DBA program’s structure also denies contractors the ability to take full advantage of scale and to try to secure the best rates through a package of coverage from a carrier that also covers their domestic risks.

The program also allows most contractors to essentially bill the federal government directly for the cost of their workers’ comp coverage. It’s hard to see any advantage to such a structure. Insurance coverage, like labor or technology or any other input, is a cost of doing business. Companies should bear their own workers’ comp costs, and those costs should be built into the bids they submit to secure the contract in the first place. As currently structured, the program leaves contractors with no incentive whatsoever to control their costs or even to shop around.

But none of these problems suggest that having the federal government directly provide workers’ comp insurance to private companies is an appropriate or worthwhile solution. Among the purposes of risk-based prices in workers’ compensation is to force employers to internalize the costs of unsafe workplaces. Having the federal government provide the coverage directly further reduces the incentive for contractors to invest in safety, which should be of utmost concern when it comes to men and women who agree to take work abroad under what are already some pretty hazardous conditions.

In support of his legislative goal, Cummings also raises a point that is not a bug, but a feature, of private insurance: that private carriers subject claims to scrutiny in an effort to weed out fraud. He cites analysis finding that private insurers have denied a significant portion of serious injury claims, particularly those for psychological damage and post-traumatic stress.

Insurers must operate in good faith, and it is of course important to always ensure that they are paying legitimate claims. But it is also important to bear in mind that, as a “no fault” system, workers’ comp is particularly prone to fraud, and questionable claims must be submitted to scrutiny. Government insurance programs like Medicare do not exactly have a sterling record for policing fraud.

Cummings hasn’t laid out any plan for how this program would do any better, and he appears unconcerned that it should even try. Adding insurance fraud – in some cases, committed by foreign nationals hired on by overseas contractors – to the list of costs borne by the American taxpayer doesn’t seem like such a good plan to us.

Featured Publications