There’s a pattern of falsifying statistics throughout the entire Census Bureau. And anyone who attempts to blow the whistle on the fraud is either retaliated against or ignored, according to two new sources who have experienced the process firsthand.
As I’ve been writing for more than six months, the Census Bureau office in Philadelphia uncovered a case of fraud in 2010, did nothing about it and allowed the practice to continue.
In that instance, a data collector named Julius Buckmon was faking reports that went into the nation’s all-important jobless tally and consumer-inflation survey.
Because the Census Bureau’s surveys are scientific — meaning each answer, in the case of the jobless survey, carries the weight of about 5,000 households — Buckmon’s actions alone would have given inaccurate readings on the economic health of 500,000 families.
Buckmon alleged that he was told to fudge the data by higher-ups. There was no formal probe back then into what Buckmon was doing or what he was alleging, although a Census investigator — who is now under indictment for other crimes against the bureau — did question a few people.
A source told me from the start of my investigation last October that Buckmon’s actions weren’t isolated and that falsification continued in the Philadelphia office right through the 2012 presidential election, only stopping when I exposed the practice last fall.
Now others who work at Census in different areas of the country are stepping forward to tell me similar stories about data being changed at the whim of supervisors who are more concerned about making quotas than protecting the integrity of information that is used for everything from cost-of-living adjustments for Social Security recipients, monetary policy decisions by the Federal Reserve and business plans by companies in the US.
“I can tell you that waste, falsification and fraud are rampant,” says one of my new sources, who works as a Census supervisor in the Midwest and handles a number of surveys, including those on jobs, health and crime.
When this source complained, higher-ups “told me to shut my mouth.” When that didn’t happen, the source was deprived of work.
Says another source who works for Census in the North Central part of the country: “I’ve said more than once (to superiors), ‘You have to watch; they are falsifying data.’ It goes nowhere. I am thoroughly disgusted with it.”
I’ve agreed to keep confidential the identity of these two sources as well as others who want to speak with me. In all, I’m told that a dozen or more Census workers are willing to talk about data falsification.
But all are in fear of their jobs, especially since a lot of workers have already been fired in budget-cutting moves that halved the number of Census regions in the US.
I’ve suggested that these two new sources speak with the Commerce Department’s Office of Inspector General and the House Oversight Committee about their allegations. They agreed.
The OIG, Oversight Committee and several others are looking into the operation of Census since I broke the Buckmon story.
If investigators do want to speak with any of the Census sources who have now come forward, I am insisting that the workers be protected under federal whistle-blower laws.
The data fabrication takes a number of forms.
My Midwest source says it is not unusual in that region for 800 out of roughly 2,000 interviews for the Current Population Survey (which is used to get the unemployment rate) to be incomplete, called “refusals,” on the last day of the monthly collection period.
Then, magically, only 100 will be unfinished when results are turned in next day to headquarters, which surveys for the Labor Department. “It’s statistically impossible,” my source says, “to complete the number of refusals we have in the last few hours.”
So supervisors are either filling out the surveys themselves or lying that houses are vacant — which also counts as a completed survey. Either way, any kind of falsification would obviously give a misrepresentation as to whether or not people in the household have a job.
And in an obvious conflict of interest, the wives of two of the supervisors in this Midwest region, according to my source, have been hired to check the results.
Up to 25 percent of the thousands of surveys that go into the jobless report may be fake, this source estimates. Falsification practices, obviously, also include field representatives like Buckmon who fill out the whole survey themselves (called at Census “curbstoning”).
Or reps will simply reach a member of a household and — rather than conduct a legitimate interview — simply ask if anything has changed from last month. If the answer is “no,” the old interview will be handed in again and be falsely counted as completed.
My sources say that there is no way to tell if there is political bias in the falsified surveys — for instance, whether those who are inappropriately filling out the forms lean toward making the unemployment rate go up or down.
But my sources say that Census workers seem to lean Democratic. “When you are in a regional office in a group of Census workers, it’s painfully obvious,” says the supervisor from the Midwest, whose own biases tilt toward conservative Republican.
I’ll tell you more when I speak with the rest of the Census workers who now feel free to confess their organization’s failings.
No comments:
Post a Comment