Census numbers uncensored

The Census aims to be every man’s hero. It promises an economic stimulus, a reduction in unemployment, and greater funds for every community. Of course, the reality is much closer to a game of musical chairs with your money. And guess who will be left standing?

The most immediate impact of the[intlink id="625703" type="post"]Census is that it distorts unemployment rates[/intlink]. With 1.2 million hired temporarily during the fall, the Census is already skewing the unemployment numbers in the government’s favor. Specifically, the fall data shows unemployment at 9.8 percent (September), 10.1 percent (October), and 10 percent (November).

Who can forget the hoopla over the November reduction from 10.1 to 10? To government officials, it was as if the clouds had parted after a relentless hurricane, “proof” that the massive stimulus spending was working.

This one-tenth-of-a-percent drop was exalted as the beginning of the end for the recession. Of course, a closer look at the numbers revealed that the decrease was largely explained by those leaving the workforce and dropping out of the government’s statistics. Similarly, the January drop from 10 percent to 9.7 percent was praised as a sign from the heavens, a sign that was subsequently tarnished by February’s slight increase in unemployment.

In an attempt to get a clearer picture of the effect from the Census on unemployment data, we evenly subtracted the 1.2-million Census bump across fall’s unemployment rates and found the new numbers ringing in at 10.1 percent (September), 10.4 percent (October), and 10.1 percent (November). If a one-tenth-of-a-percent drop in November was a reason to celebrate, then a three-tenth-of-a-percent October upward revision is a reason to cringe.

In the months ahead, expect the same number games.

The Census already hired 1.2 million workers in the fall, as mentioned earlier; now they’re planning for another 1.2 million in the spring. February began with an additional 15,000 Census workers; the hiring will peak in April-May with 800,000 workers hired in just two months. The peak alone is projected to temporarily push unemployment downward by half a percent.

Digging a bit deeper, we find that this year, 723,000 door-to-door Census takers will be needed in comparison to 2000’s 604,000, a 16.6 percent increase despite the lack of an equivalent rise in the population.

The hiring numbers are pushed upward by the low total working hours per employee. On average, each temporary Census employee will work 19 hours a week for six weeks. But “work” may not be the most appropriate word. The Census estimates that only 47.8 million houses will require a Census worker to visit. Divide this by 723,000 census takers to get 66 houses per worker. Since each temp will be employed for six weeks, this translates into less than two houses visited per workday.

The spending side of the Census equation doesn’t make much sense either. According to a Census Bureau press release, mailing back the form costs the government only 42 cents, while visiting a house costs 56 dollars.