Earlier, I wrote a story on the latest report by the Special Inspector General for the Troubled Asset Relief Program. The story drew heavily on the comments of Linus Wilson, an assistant professor at the University of Louisiana at Lafayette, who said his analysis showed that the Treasury Department left $375 million on the table in its dealings with Morgan Stanley (MS).

    Wilson has published a number of papers tracking Treasury's progress in selling the warrants it received from banks in the bailout, and I have quoted him frequently on this issue. He has often been critical of Treasury's procedures, but his estimates have generally been trustworthy, in my view. His mid-range estimate of the proceeds from the sale of Treasury's JPMorgan Chase (JPM) warrants was $900 million, to take one example; the actual sale at auction raised $950 million.

    But Treasury takes issue with Wilson's analysis, saying he vastly overstates the value of the Morgan Stanley warrants — and that this is of a piece with his earlier work:

    “Assistant Professor Wilson’s statements are inaccurate and without merit," according to the Treasury statement from David Miller, chief investment officer for the Office of Financial Stability. "Prior to the first three auctions of warrants, Wilson predicted each issuer’s warrants could be worth almost three times the prices actually realized in the auction. Even the low point of his range was more, in each case, than the realized auction prices which reflect fair market value. This shows his model has consistently and dramatically overvalued the warrants."

    As I say, I have been keeping an eye on his estimates on the big banks and they have been pretty good; they were very much in line with actual payments in the cases of Goldman Sachs (GS) and Bank of America (BAC). But it is true that the early Wilson estimates, on three small banks, have come in for criticism before.

    Treasury concludes that "Assistant Professor Wilson’s valuation of Morgan Stanley’s warrants is similarly inaccurate and he has made baseless accusations. SIGTARP concluded that the price Treasury received from Morgan Stanley was 6 percent above Treasury’s composite fair market value, and SIGTARP did not question Treasury’s methodology."