Federal Housing Finance Agency analysis used to prevent principal reduction on Fannie Mae and Freddie Mac loans was seriously flawed, according to one leading analyst.

"We have reviewed the study and have a number of very substantial objections," said Amherst Securities Senior Managing Director Laurie Goodman before a Senate subcommittee Thursday, who gathered additional data via telephone.

The FHFA used the report to determine a wide-scale principal reduction program would cost more than forbearance, while increasing the risk of strategic default among borrowers seeking relief.

This information was not provided in the analysis. Sen. Jeff Merkley, D-Ore., was perplexed and he asked her how she received more information on the principal reduction study than lawmakers in the Senate.

"I made about 50 phone calls. The info was not available in one place. We were unable to construct what was done in the study and made a bunch of phone calls to figure this out," Goodman said. "We were very persistent." Goodman called several sources within the FHFA.

Goodman added the report also did not differentiate loans based on mortgage insurance, another critical metric.

The mortgage insurer does not cover the amount forgiven, she explained. She gave an example. If a borrower holds a $100,000 mortgage on a house worth $75,000, the mortgage insurer would cover any loss down to $70,000. If the GSE reduces the loan amount down to $80,000 and the borrower redefaults – though redefault rates on principal reduction mods are roughly 25% – the GSE loses the $20,000 given up.

The insurer would then pay the GSE $10,000 – the difference between the new loan balance of $80,000 down to $70,000 covered – and the resulting REO is only sold for $70,000, the GSE still loses the $20,000 it gave up in principal reduction and the insurer is off the hook.

"The problem with principal reduction on loans with mortgage insurance is that Fannie and Freddie are essentially subsidizing the mortgage insurer," Goodman said. Gauging the benefits of reducing principal for borrowers without the insurance could swing the verdict, she said.

And roughly 32% of the GSE portfolio of seriously delinquent loans carry mortgage insurance, meaning more than two-thirds of troubled Fannie or Freddie borrowers could potentially stand a better chance of securing a principal reduction, Goodman said. But the study didn't dig that deep.

She found other problems as well.

The FHFA used state-level home prices, not indices on the metropolitan statistical level, allowing the analysis to pick up fewer borrowers in severe negative equity. Higher Home Affordable Modification Program incentives were not factored in – though the FHFA has said it was reconsidering the Treasury Department's new offer. And borrower characteristics at origination were used, not current information like FICO scores.

Still, Mark Calabria, director of financial regulation studies at the Cato Institute, doubted the government's ability to build a program that would allow Fannie and Freddie to consider principal reduction on specific, targeted amount of loans.

"I would not use 'specific' and 'targeted' to describe our previous foreclosure efforts," Calabria said.

John DiIorio, CEO of 1st Alliance Lending, which works with investors and banks to conduct principal reductions agreed with Goodman in that the FHFA study was flawed, but he doubted a government program would be streamlined enough to do any good.

"I don't think there is a one size fits all solution," DiIorio said.

But Goodman did call for allowing Fannie and Freddie to go through their portfolios and determine on a loan-by-loan basis if principal reduction was worth doing.

"We would urge the FHFA to re-run their results, using the new model which incorporates the triple incentives, correcting the technical flaws in their analysis, and breaking out loans with and without mortgage insurance separately," Goodman said in her written testimony. "We believe when this is done, it will be clear that forgiveness is the better solution for the bulk of the two-thirds of their book of business without mortgage insurance."