AP's Hananel Grudgingly But Cryptically Notes House Investigation of Allegedly Juiced 2010-2012 Jobs Reports

November 20th, 2013 12:39 PM

In a mild surprise, the Associated Press, aka the Administration's Press, hasn't totally ignored John Crudele's Monday evening blockbuster story at the New York Post about how fabricated Census Bureau information fed a pretty clearly cooked September 2012 Employment Situation report. But the wire service's Sam Hananel ruined the surprise by spending five terse paragraphs making sure that relatively disengaged readers would learn as little as possible.

Most crucially, Hananel never told readers that the alleged manipulation may have been the main reason why the reported September 2012 unemployment rate fell below 8 percent for the first time since President Barack Obama took office in January 2009. At the time, former GE CEO Jack Welch was among those who strongly questioned the rate drop.


Hananel's report, which has not been updated since Tuesday evening at the AP's national site, was completely numbers-free:

APonHouseCensusUnempInquiry112013

While he at least acknowledged the "manipulation" charge, Hananel shed no light on what was allegedly done, how widespread the manipulation might have been, how long it had been occurring, or its possible effect.

I'll let the Post's Crudele provide some of the specifics:

Census ‘faked’ 2012 election jobs report

In the home stretch of the 2012 presidential campaign, from August to September, the unemployment rate fell sharply — raising eyebrows from Wall Street to Washington.

The decline — from 8.1 percent in August to 7.8 percent in September — might not have been all it seemed. The numbers, according to a reliable source, were manipulated.

And the Census Bureau, which does the unemployment survey, knew it.

Just two years before the presidential election, the Census Bureau had caught an employee fabricating data that went into the unemployment report, which is one of the most closely watched measures of the economy.

And a knowledgeable source says the deception went beyond that one employee — that it escalated at the time President Obama was seeking reelection in 2012 and continues today.

... The Census employee caught faking the results is Julius Buckmon, according to confidential Census documents obtained by The Post. Buckmon told me in an interview this past weekend that he was told to make up information by higher-ups at Census.

... But, Buckmon says, he was never told how to answer the questions about whether these nonexistent people were employed or not, looking for work, or have given up.

But people who know how the survey works say that simply by creating people and filling out surveys in their name would boost the number of folks reported as employed.

Census never publicly disclosed the falsification. Nor did it inform Labor that its data was tainted.

AP itself, in commenting on last month's 204,000 gain in seasonally adjusted jobs, may have partially confirmed the unusual claim that the mere existence of more surveys can affect results (even though, to be clear, the following relates to the Establishment Survey of employers and not the Household Survey:

THE SHUTDOWN'S SURPRISE EFFECT ON JOBS NUMBERS

"It seems that when the initial response rate is high, the initial payroll number is often, though certainly not always, stronger than the prior trend," Ian Shepherdson, chief economist at Pantheon Macroeconomics, wrote in a research note.

In other words, if the jobs numbers in prior months were based on a lower participation rate, a stronger participation rate would skew the number up.

Thus, the mere existence of more data may have inflated the results.

The seasonally adjusted increase in total employment reported in the Household Survey reported on October 5, 2012 – just 4-1/2 weeks before Election Day and shortly after early voting began – was originally 873,000 (since revised down to 810,000). There was also a somewhat suspect drop of 456,000 in the number of unemployed (since revised to -401,000). With the exception of a few Januarys (that's when the Census Bureau adjusts its population controls), the revised employment increase is the largest seen since June 1983 (991,000) and May 1984 (857,000), during quarters when the economy was growing at annual rates of 9.3 percent and 7.1 percent, respectively.

Given how long Crudele reports that data fabrication has been occurring, it's reasonable to question whether it was done to meet quotas regardless of results, or whether it was done to achieve a desired result.

As usual, the "if this had occurred during a conservative or Republican administration, the coverage would be widespread" bromide certainly applies. It also would be far more specific than the AP report above.

Cross-posted at BizzyBlog.com.