The problem with Recovery.gov isn't the data, it is the politics

The White House has gotten a black eye for two key problems with the Recovery.gov stimulus database. The first problem was that the numbers of jobs were bogus. The Washington Examiner has concluded that at least 10% of the jobs were fabrications. The second problem was that the data about congressional districts were clearly garbage.

But I am not here to rehash this. There is a legitimate problem with the data. But the data wouldn't be such a big deal if the White House hadn't tried to politicize the data and claim victory.

Recovery.gov is a tremendous success for the transparency movement. Politicians lie. A key goal of the transparency movement is to give the people the power to keep the politicians accountable. And that's what this has done.

Here's what happened: The White House shared the data and lied about what it showed. And the data has been used to hang them.

First, let's go to the stimulus bill itself. Congress demanded that the White House report jobs created or saved numbers:

(8) The website shall provide a link to estimates of the jobs sustained or created by the Act.

But that legislative langauge says that "[t]he website shall provide a link to estimates" ... "by the Act". The bill doesn't require a per-contract or a per-district accounting of jobs.

So what's happening? The White House quickly learned that the stimulus bill was going to be a hot potato. So they started to use Recovery.gov to give them (and their Democratic allies in Congress -- recall no Republicans voted for this) cover.

Let's be clear what happened here: the White House politicized this data. And now they are getting hung with the politicization of it (not the bad data).

And this goes to the deeper point. Check out this story from the Washington Post from October 30th, the weekend before election day:

Reports to be released Friday on the government Web site Recovery.gov are expected to show that the $150 billion in grants and loans made so far under the economic stimulus package have created or saved about 650,000 jobs, White House officials said Friday morning.

White House officials said the reports -- which were filed by state and city governments and other recipients of stimulus grants and loans -- will confirm their recent estimates that the $787 billion package passed in February has so far saved or created about a million jobs, putting it on track to match their estimates of 3.5 million jobs created or saved over the three-year span of the stimulus. That calculation is based on the fact that today's reports do not include much of the package's spending -- tax cuts, safety net spending and fiscal aid to strapped states, which injected tens of billions more into the economy and, in the case of the state aid, forestalled layoffs of state workers.

The White House clearly leaked this information the weekend before election day to claim victory based on garbage data. Furthermore, they claimed that Recovery.gov confirmed their previous assessments, and didn't come to any new conclusions based on data. 

Fifteen days earlier, based on partial data, the White House also claimed victory based on incomplete data, using the same number. From the Hill's Walter Alarkon:

"All signs -- from private estimates to this fragmentary data -- point to the conclusion that the Recovery Act did indeed create or save about 1 million jobs in its first seven months, a much needed lift in a very difficult period for our economy," said Jared Bernstein, the chief economist for Vice President Joe Biden.

That same 1m had been their talking point for a while. In September the White House said:

This analysis indicates that the ARRA and other policy actions caused employment in August to be slightly more than 1 million jobs higher than it otherwise would have been.

In May, they said:

Our finding was that the ARRA would increase employment relative to the baseline in this quarter by approximately 3.5 million

Six months later, the White House was repeating these same numbers even though they had data to prove it false. That's the problem.

0
Your rating: None

Comments

The problem

I guess it's also a bit of a problem that even if the numbers are correct, each job would have cost roughly three average annual salaries.

I mean, it's not exactly rocket-science to "sustain or create a job" when you've got the salary paid for. "Creating" carries a notion of, well, creation of value.