The Journal has more bad news on the banks and some top-notch analysis that cuts through the government line on TARP.

It reports that lending by the nineteen biggest TARP recipients is down 19 percent since the Treasury implemented the program six months ago.

And—sit down before you read this— the government has been fluffing up the numbers:

The Journal’s analysis paints a starker picture of the lending environment than the monthly snapshots released by the government and is a reminder of the severity of the credit contraction. One reason for the disparity: The Treasury crunches the data in a way that some experts say understates the lending decline.

By more than double, at least in February:

Using the same raw data, the Journal’s analysis focused on the total amount of new loans by the 21 banks, a more comprehensive measure. In February, that total fell 4.7% from January, more than double the government’s estimate of the decline in the median.

The total decline in lending from October to February was 19 percent by the Journal’s tally. Even worse for the economy, the WSJ calculates that commercial lending by the big banks plummeted by 40 percent between October and February. That means the folks that do the hiring aren’t getting money to start or expand businesses.

Though the Journal is excellent to follow that point with information that undermines it. Lesser reporters wouldn’t have made this connection:

One factor that may have depressed commercial borrowing is a partial thawing of bond markets, where some big companies raise money instead of borrowing it from banks. About $70 billion of corporate bonds were issued in February, up from $21.4 billion in October, but still only about half the level of last May, according to Thomson Reuters.

And I didn’t quite understand the Journal methodology on the first pass-through, but it helpfully explains it down low giving a concrete example:

The Treasury’s conclusion that lending has dropped only modestly is partly the result of its focus on the median monthly change at the top banks, rather than the change in average or total lending. Thus, for the month of February, the Treasury based its report on the 2.2% drop in loan origination at PNC, which fell at the midpoint of the 21 banks that the Treasury surveyed.

And it gives time to those who—unconvincingly, but not without merit—defend the Treasury’s methods:

Statisticians sometimes use median figures to avoid having data distorted by outliers on either end of the spectrum. Treasury officials have been concerned about the month-to-month volatility in the bank data and are still developing how best to present it.

Tom Fullerton, an economics professor at the University of Texas at El Paso, defended the use of the median, citing the monthly volatility in bank-lending data.

But other experts say the average would provide a more accurate measure lending. Economist David Boyum says using the median does not answer the fundamental question: “What has happened to overall lending?”

The point is: Overall lending by the bigs has been much lower than the government has said it’s been, and the Journal does an excellent job figuring it out and carefully presenting the information. It also gives nice context, noting that the numbers would be even worse without the TARP funds and writing that these numbers mean it’s less likely that the government will let those itching to repay the money do so.

Good stuff.

If you'd like to help CJR and win a chance at one of 10 free print subscriptions, take a brief survey for us here.

Ryan Chittum is a former Wall Street Journal reporter, and deputy editor of The Audit, CJR's business section. If you see notable business journalism, give him a heads-up at rc2538@columbia.edu. Follow him on Twitter at @ryanchittum.