Student loans, specifically the interest rates on some of those loans, have been in the news in recent months. After some debate and a lapsed July 1 deadline, the U.S. Senate this week reached a bi-partisan compromise plan to limit the increase in interest rates charged on subsidized Stafford loans.

The debate focused on whether to let interest rates on the government loans to college students nearly double to 6.8 percent or remain frozen at 3.4 percent. Advocates for lower interest rates seized on a report by the Congressional Budget Office suggesting the federal government makes a profit — a $50 billion profit — on student loans. Sen. Elizabeth Warren, D-Mass., called those profits “obscene.”

Warren’s comments made for good sound bites and headlines; except that the profit figure she quoted is highly questionable, or simply wrong.

It turns out that there are two different accounting methods that can be used to evaluate the cost, or profit, of a government loan program. Congress mandated the use of an accounting method in 1990 that does not factor in risk of default or collection costs.

The end result, most experts agree, is that costs and potential losses are understated and profits are overstated — by a lot. The CBO report notes the problems with failing to account for market risk. In fact, the CBO says the post-1990 accounting method obscures tens of billions of dollars of potential losses due to defaults.

Responding to Republican requests for an alternative valuation of the student loan program, the CBO used “fair value accounting,” which does account for market risk, including the cost of defaults. Using the two different approaches on six education loan programs, the new CBO estimates go from $36 billion in profit to $5.5 billion in profit. A story in the Atlantic magazine on this issue notes that big profit swing is for just one year and does not include any administrative costs.

In another analysis of Stafford loans for 2013, the CBO estimate goes from $4.6 billion profit using the older, mandated accounting method to a loss of $3.5 billion, using fair-value accounting.

With an estimated default rate on Stafford loans of 23 percent, it would be foolish to not factor in default costs. If the program is to be self-sustaining, the interest rate has to be high enough to earn additional income to cover default losses.

Warren, a former Harvard law professor and financial expert, was disingenuous in her comments about obscene profits. It’s hard to believe that, given her expertise, she didn’t know about the different accounting methods — and thus the different profit (or loss) pictures for the loan program.

Putting her comments about “obscene” profits on student loans to the truth test, the Washington Post’s fact-checker department awarded Warren “one or two Pinocchios” for her representation of the student loan program.

The Post concluded that Warren and other defenders of lower interest rates on student loans “should use more caution in citing figures — especially when the source Warren cites, CBO, says there are problems.”

Given her education and background in financial matters, Warren should know better. She probably did, but she’s a politician now.

Student loans play an important role in helping more people attend college, but the debate should stick to facts, not questionable figures. The student loan debate should use the most accurate accounting — and should go further to explain why college costs have risen faster than inflation for more than a decade.