His conclusion:

I examine the price of Iraqi state bonds, which the Iraqi government is currently servicing, on world financial markets. After the Surge, there was a sharp decline in the price of those bonds, relative to alternative bonds. This decline signals a 40% increase in the market's expectation that Iraq will default. This finding suggests that, to date, the Surge is failing to pave the way toward a stable Iraq and may in fact be undermining it.

Then, in October, the W.S.J. ran a piece with some criticism from professional money managers on Greenstone's study:

It's "a bit of a stretch" to draw political conclusions from a market where bonds are thinly traded and accurate prices are hard to detect, says Tom Cooper, a portfolio manager in Boston for GMO LLC's emerging-markets bond fund.

Mr. Cooper and Michael Conelius, a T. Rowe Price portfolio manager, said in interviews that Mr. Greenstone failed to take into account broader market performance. This summer's credit crisis led investors to reprice a range of assets around the world, pushing up yields on the riskiest bonds.

The MIT professor's conclusions make about as much sense as saying "the surge caused the subprime crisis," Mr. Conelius says.

Annual yields on Iraq's bonds were 9.6% in February, climbed to 11.5% during August's market chaos, then rallied to 10.5% last week, Mr. Conelius said.

...

Hedge funds and mutual funds are among the big holders of these bonds. At the end of June, the bonds made up 4% (roughly $26 million) of Mr. Conelius's T. Rowe Price Emerging Markets Bond Fund.

He likes Iraqi bonds because they're "uncorrelated" to other assets and because of Iraq's vast oil reserves. If Iraq stabilizes and its economy recovers, he says, the bonds' values will rise.

Mr. Conelius says Iraq's bonds aren't comparable to Qatar's, and Mr. Greenstone agrees but thinks there is no apples-to-apples comparison. Qatar is a stable country with a strong credit rating.

Better, Mr. Conelius said, is to compare Iraq's bonds to bonds issued by Lebanon, a country with a similar economic outlook. Iraqi bonds have outperformed Lebanon's since February, Mr. Conelius said.

Mr. Greenstone said it's significant, though, that data showing investors' risk assessments on Iraqi bonds didn't improve from February through June, before the global-markets crisis hit.

Now, in an interview on M.I.T.'s web site, Greenstone explains himself.

A couple of key excerpts:

Q: What inspired you to focus particularly on the 30,000-person U.S. troop surge?

A: As an economist, I found the evidence used by supporters and opponents of the surge less than convincing. An egregious example was Sen. John McCain's announcing on TV news that things in Iraq were improving--as he walked through a Baghdad market, surrounded by armed guards and camera crews.

The fact is that the war is terribly costly both in human and financial terms: Nearly 80 people (Iraqis plus coalition troops) are dying per day, and the U.S. cost of the war may exceed $1.2 trillion. When the stakes are so high, we must have reliable information to make decisions with.

Q: How did you gauge the long-term effects of the Surge?

A: I looked at the prices of Iraqi government bonds that trade on world financial markets. Bond prices give a good idea of how traders view the chance that the institution or government offering the bond will be around when the bond comes due. Unlike the other measures, they provide a direct estimate of whether there will be a stable Iraq in the future.

My analysis found that the price of the Iraqi bonds never increased after the Surge, indicating that traders did not foresee the Iraqi government would last long enough to pay its debts. In fact, there is some evidence that the Iraqi bonds declined relative to comparison bonds. This finding suggests that, to date, the Surge is failing to pave the way toward a stable Iraq.

The appeal of using financial markets is that traders' only concern is to make profitable decisions. There is no room for politics or justifying previous decisions in these markets.

Q: What is your response to critics interviewed by The Wall Street Journal who say your analysis did not account for the recent credit market crisis?

A: The Journal noted that the decline in the price of Iraqi bonds largely coincided with this summer's credit crisis. But the price of Iraqi bonds did not increase from February through June, a period when the Surge was in force but the credit market crisis hadn't hit yet. The point is that the bond market has never been convinced that the Surge would help Iraq's long-term future.

Further, the finding of a decline in Iraqi bonds remains even when their prices are compared to the price of bonds from other emerging markets.

My sense is that Greenstone is on to something here. Besides his paper, two other points stand out. One, as the W.S.J. article notes, T. Rowe Price's Conelius holds these bonds in one of the company's mutual funds, so it's in his interest to make them look like a "buy." And two, if, as Conelius says himself, Iraqi bonds are "uncorrelated" to other assets, there is much information to be gleaned from studying the bonds' movements.