America’s universities, increasingly strapped for cash, complain that funders are not paying them the full cost of doing needed research. The Council on Governmental Relations estimates “unrecovered” research expenses in 2012 alone at $4.6 billion, “a staggering multi-million dollar obligation per university.” Without full funding, colleges say, breakthroughs will go undiscovered and the nation will lose its competitive edge.

But I’ve come to believe that one reason universities don’t get back all their research expenses is the murky nature of accounting used to explain the “indirect costs” that make up a big part of research funding.

Indirect costs are facilities and administrative charges (F&A) such as building maintenance or library usage fees. In Nashville, for example, Meharry Medical College has a 45.5 percent indirect cost rate, which means that a $100,000 research grant actually costs $145,500. But less than two miles south, for similar research, Vanderbilt University has a 57 percent indirect cost rate.

Why? Even experts who support greater funding, such as the Council on Governmental Relations’ David Kennedy, admit there isn’t an easy answer and that the system for calculating indirect costs “is kind of cryptic.”

At times there’s no data at all. lists some $40 billion in annual federal grants to universities. How much do indirect costs add to that? Another $15 billion? More? The totals and breakdowns are not there.

“Indirect” percentage rates can range from zero to nearly 70 percent. Each college and the federal government negotiate them privately. The 103-page Office of Budget Management’s updated rules on federal grants, which take effect this December, does not explain the indirect cost-negotiating process or any “case-by-case” exception rules.

Texas A&M University professor Martha Sale agrees that the system may fail “to capture the actual costs” of research. But she also says there may be no one way to do that fairly. In 2009, her study found that indirect cost rates were higher at private universities than public ones, but the reasons were not clear.

Indirect costs are a federal issue for good reason: The federal government funds 63 percent of all university-based science and engineering. Other sources: the colleges themselves (19 percent), foundations (5.9 percent), businesses (5.1 percent), state and local governments (5.5 percent), and “other” sources, such as individuals (1.5 percent).

So it is the federal government that defines facilities and administrative as F&A fees. The facilities charges have no cap; they are the main subject of indirect cost negotiations, conducted by only two federal agencies, the Office of Naval Research or the Department of Health and Human Services. The government has capped administrative charges at 26 percent since 1991, after “unallowable costs” were found in research grants (including “maintaining and furnishing the homes of university presidents.”)

The Association of American Universities has campaigned against the 26 percent administrative cost cap, noting that other contractors—such as corporations and even NASA—can add indirect charges of more than 90 percent. Says vice president of research Tobin Smith, “We shouldn’t be treated differently.”

Universities hire consultants to negotiate the percentage they may add on to grants. Attain, a Virginia-based consulting group, has negotiated rates for schools like the University of Miami and Johns Hopkins University. Attain Vice President Mark Davis says that most universities do not have the in-house expertise to negotiate. His staff prepares rate proposals with university documents such as floor layouts for research space. The government comes back with a list of questions (when will the new building be occupied?) before the actual negotiation begins.

Why does all this happen privately? “I know to some people, [the system] might seem unorthodox,” says Davis, “but it is what it is.” Arif Karim, who oversees indirect cost negotiations at Health and Human Services, says it’s because rate proposals are proprietary in nature. “All universities want higher rates,” Karim says, but research space costs can be tough to pin down. Though negotiators sometimes visit campuses, they usually set rates over the phone.

Lawmakers have failed to make research awards more transparent. HR 3433, which died in Congress in 2011, would have required public release of full grant applications. Colleges opposed it, arguing that it would give people a “shortcut to further their own research or steal intellectual property.

Indirect costs are growing. A recent Government Accountability Office (GAO) report said that from 2002 to 2012, indirect costs outpaced direct costs within the National Institutes of Health (NIH), the primary research funder. NIH disagreed there was a problem. Yet GAO report author Linda Kohn calls rising indirect costs “a serious risk.”

Private foundations generally do not pay the federal rates. Something with indirect costs just “does not seem appropriate,” says Wendy Garen, CEO of the Ralph M. Parsons Foundation. Parsons pays no indirect costs. A scan of foundations shows that some pay no indirect charges, others a small fixed rate, and still others “up to” 10 or 20 percent on a “case-by-case” basis. “Up to” and “case by case” policies also lack transparency.

Universities allow private foundations to pay lower indirect costs. Why? Foundations make up less than 6 percent of research funding. Their grants do not usually demand complex science or medical equipment. They are, after all, engaged in charitable activities. Still, says Arthur Bienenstock, Stanford University’s assistant to the president of research, “I think the foundations should pay the negotiated government rate.”

But wouldn’t funders be more inclined to sweeten research dollars if universities were clear about the costs? I think so. If databases contained all the indirect expenses, general terms of negotiations were public, and rates were not just transparent but uniform, the arguments for payment would be stronger. Surely libraries, for example, could show as essential to cancer research without giving away a sensitive secret.

Jeremy Berg, a University of Pittsburgh scientist, looked into indirect costs out of sheer curiosity on his science policy blog, Datahound. Some grant information is rightfully kept private, Berg says. But since other data, such as accounting for research space, can be “tilted” to support a point of view, he adds, “I do think that everyone would be better off if there were more transparency.”