Cost of 2010 Census a Whopping $14.7 billion
Policy + Politics

Cost of 2010 Census a Whopping $14.7 billion

Alex Rader/The Fiscal Times

Counting the country’s resident population is a monumental, high stakes task for the U.S. Census Bureau. The data it collects determines key political and economic benefits: district voting lines, reapportionment of the U.S. House of Representatives, and trillions of dollars of federal aid to state and local governments. But the cost of acquiring this data is skyrocketing with each passing decade--$14.7 billion for the 2010 Census, more than double the bureau’s $6.5 billion budget in 2000.

In fact, the Census’s budget has ballooned every decade since 1950, and by the bureau’s own admission could reach as high as $22 billion for the 2020 Census. The bureau did save $1.87 billion of its FY2010 budget, but roughly half came from the contingency fund, set aside for last minute infrastructure meltdowns and severe weather, neither of which occurred.

There was the embarrassing handheld computer debacle, which resulted in $3 billion in losses when the Census Bureau unsuccessfully implemented a $595 million contract to develop the handhelds. The computers were supposed to replace the millions of costly paper forms and maps that enumerators must carry when going door-to-door. But there were constant setbacks including contractual issues, cost overruns, missed deadlines and lack of an overall recovery plan should the devices fail.

Cut the Census, Skew the Numbers?
The Census budget is arguably the most challenging federal program for the Commerce Department to cut because of the constitution’s rigid deadline, the rising cost of non-response follow up, and the untimely shifts in leadership that occur with every new presidential administration.

According to the Constitution, the Bureau is obligated to deliver a Census on time and in compliance with burdensome counting methods, which are budget busters. They include visiting a person’s home as many as six times if they do not respond to the mail-in questionnaire.

These methods are time consuming and require extraordinary spending to find a small percentage of the country’s population of immigrants and ethnic minorities who move frequently, double up in single-family dwellings more often than others, and are less likely to respond to questionnaires. More than 585,000 temporary Census workers knocked on the doors of over 42 million households for non-response follow ups. Meanwhile, each percentage point increase in the mail-back response rate last year saved the Census Bureau $85 million because workers did not have to make door-to-door follow up trips.

How good is good enough, how much money
should we spend to get closer to an elusive goal?

In spite of the due diligence, the Census doesn’t always get it right. The 1990 Census missed an estimated 8 million people, mostly immigrants and urban minorities. During the 2000 Census, the National Academy of Sciences estimated as many as 17 million people were counted twice, and as many as 14 million people were missed all together. As for the accuracy of the 2010 Census, undercount and overcount estimates won’t be released for another two years or so.

“Somehow we need a way that this society can start to engage in a trade-off discussion,” said Groves, during an interview with The Fiscal Times. “Many of these follow up activities are hundreds of millions of dollars that we do in order to try and count these elusive last cases [those hardest to track down] and somehow we need to try to ask ourselves in a real way how good is good enough, how much money should we spend to get closer to an elusive goal. And we’ve never had that kind of discussion before.”

Groves’s “trade-off discussion” has long been overlooked by lawmakers, according to University of Wisconsin-Milwaukee history and urban studies professor Margo Anderson, an expert on the Census. “So far as Congress can shed light on issues and make people focus--that hasn’t happened,” she said.

“Congress as the political body that oversees the Census needs to take the lead in starting that debate and overseeing that conversation, and defining the end goal,” added Terri Ann Lowenthal, former staff director of the House Subcommittee on Census, Statistics, and Postal Personnel.

A bill, cosponsored by Sens. Tom Coburn, R-Okla., and Tom Carper, D-Del., and Reps. Carolyn Maloney, D-N.Y., and Charlie Dent, R-Pa., and supported by the last seven U.S. Census directors, would have given the Census Bureau more autonomy from the Department of Commerce and the White House. It also would have made the Census Bureau director a presidential term appointment of five years, with the 10-year decennial cycle split into two, five-year phases – planning and operational. A Census director could serve no more than two terms, thus creating continuity across administrations. As it stands, the Census director is a political appointee who can be removed from duty at any time. The Senate passed the bill, but it was opposed by House Republicans and the Obama administration.

“If you gave me the entire federal budget I could
go out and count everyone in the United States.”

The Census Bureau, meanwhile, continues to seek new efficiencies and cost savings, by using technology and the Internet to reduce administrative expenses, and statistical sampling costs.

To be sure, there will always be costly external factors outside of the counters’ control, including a population that continues to increase and spread farther away from major cities. The population will also continue to diversify, making it markedly more multilingual.

“If you gave me the entire federal budget I could go out and count everyone in the United States,” quipsDr. Kenneth Prewitt, former director of the U.S. Census Bureau. “However if you gave the Department of Federal Safety the entire federal budget we wouldn’t have highway fatalities.”

There are also costly internal controls that have plagued the Census in recent years. In March 2008, the Government Accountability Office (GAO) added the 2010 Census to its “high-risk list” of agencies that are vulnerable to fraud, waste, abuse and mismanagement.

The reasons, according to GAO: “shortcomings in the Census Bureau’s management of information technology, weak performances by technology that the Bureau plans to use for data collection, uncertainty of cost estimates, and the elimination of several dress rehearsal activities.” The Census Bureau remains on the GAOs list today.

In spite of the bad report card, 2010 marked a number of successes, and could be a turnaround year. The Census Bureau:

  • Delivered the 2010 decennial Census to President Obama on time and under budget.
  • Scored a 74 percent mail-in response rate to its questionnaire, equal to the 2000 decennial Census, and halting a three decade decline.
  • Provided Census participation support to residents in 59 different languages.

And, importantly during one of the worst recessions in U.S. history, the Census created roughly 1.4 million temporary hourly-wage jobs during the height of the financial crisis in 2009 and 2010. Moreover, given the larger pool of workers to choose from, the bureau was able to hire workers with more skills, and more experience than in the past. Those workers proved to be highly dependable, and didn’t quit or disrupt the Census taking – which also was a money saver.

U.S. Commerce Secretary Gary Locke noted during a news conference last month that in 2009, the Commerce Department’s own inspector general and the Government Accountability Office both ranked the 2010 Census as one of the federal government programs most likely to fail. “That did not happen,” said Locke, who had been sworn into office just nine months prior to the Census’s release, nearly the same time U.S. Census Bureau Director Robert M. Groves had spent on the job.