Tag: census bureau

It’s Illegal to Say ‘None of Your Damn Business’

The government’s troops are rallying behind the Census Bureau’s American Community Survey. “After the House voted this month to defund a major part of the U.S. Census Bureau, the agency is taking the threat very seriously,” reports the Washington Times, “with its supporters in both business and government rallying to preserve the annual questionnaire.”

Wait. Who could be against the Census Bureau? Its constitutional charter is to enumerate citizens every ten years for the purpose of apportioning representation in Congress. This is a necessary and unremarkable administrative function.

Oh, wait—again. Government bloat is a law of gravity, and the Census Bureau does far, far more than count noses. Its American Community Survey has made the Census Bureau the research arm for the welfare/redistribution state and a source of corporate welfare in the form of demographic data about Americans.

So Census goes around asking people dozens of questions that have nothing to do with the agency’s constitutional purpose.

The ACS is controversial enough among the strongly principled that Census has a Web page entitled: “Is the American Community Survey legitimate?” Their answer: “Yes. The American Community Survey is legitimate. It is a survey conducted by the U.S. Census Bureau.” (Did you know there’s a whole class on the “appeal to authority” at Fallacy University?…)

The real authority they cite is Title 13 of the U.S. code, which, in section 221, allows the government to fine people who refuse to answer the Census Bureau’s questions. It’s illegal to say “none of your damn business” when a government official comes around asking about your toilet. I’ve written many times, in long form and short, that the helping hand of government strips away privacy before it goes to work.

So it’s nice to see that Rand Paul (R-KY) in the Senate and Ted Poe (R-TX) in the House have introduced a bill to make the American Community Survey voluntary, unless it’s a question that the Census actually needs for its constitutional purposes. Reading public comments on the House bill is particularly interesting. There is a good number of people who want to be left well enough alone. They shouldn’t be subject to penalties for saying so. It’s a matter of principle and privacy.

‘Useful’ Idiots and the American Community Survey

Since 2005, the Census Bureau has administered a set of probing questions to a large random sample of Americans each year. It’s called the American Community Survey, participation is mandatory, and the penalty for failing to comply is a fine of up to $5,000. The questions are so prying (“Do you have trouble bathing?”, “When do you leave for work in the morning?”) that many people find it hard to believe the government would have the temerity to demand answers to them, and so conclude that it must be a scam operated by unscrupulous con artists. Others simply lack confidence in the federal government’s promise to keep such information confidential (perhaps not surprising when it has shown itself incapable or disinclined to keep its mouth shut about even life-and-death anti-terrorism operations.)

The anti-ACS backlash has garnered the attention of some in Congress. The House has voted 232 to 190 to eliminate the survey altogether and Sen. Rand Paul has introduced legislation in the Senate to make it voluntary. They argue that it is inappropriate for the state to compel citizens to divulge such personal details, and that these questions are in no way required to fulfill the constitutional role of the Census, which is to apportion political representation based on population counts.

These bills have consternated a great many people from New York Times economics reporters and Forbes.com contributors, to associate editors for Bloomberg Businessweek and multitudes in the left-of-center blogosphere. Their counter-argument is fairly simple: this information is useful both to the state and to select private corporations and therefore the state should have the right to compel citizens to divulge it.

That so many people think this is a proper basis for determining the legitimacy of state powers is the most compelling evidence I have yet seen contradicting the Flynn Effect.

It would undeniably be useful for law enforcement officials to have finger prints, photos, and DNA samples for every person in the United States, and to track their movements at all times using GPS devices. And it would surely be useful to the medical community if citizens were randomly and forcibly assigned to participate in drug trials. Nevertheless, even ardent believers in the beneficence of government must be able to see that these would not be desirable practices in a free society.

Americans rightly expect to live their lives with a reasonable level of privacy. The Fourth Amendment to the Constitution protects them from government searches unless there is some compelling reason for them. But the reasons for making the ACS mandatory are not at all compelling. Medical and social science researchers manage to deal with their inability to forcibly assign subjects to treatment and control groups. Political pollsters manage to produce reasonable forecasts without being able to compel citizens to reveal how they will vote. And even when government _is_ in possession of all the necessary and relevant data, it quite frequently makes idiotic decisions anyway—so it isn’t even clear that these ACS data would necessarily lead to more effective government programs.

At the very least the ACS should not be mandatory. It probably shouldn’t exist at all.

The Census Asks Too Much

Everyone in America, I presume, has just received a letter from the U.S. Census Bureau urging us to fill out our Census forms. Seems like a very expensive way to tell us to watch for the form to arrive in the mail. But I’m particularly interested in why they say we should promptly fill out the form:

Your response is important. Results from the 2010 Census will be used to help each community get its fair share of [federal] government funds for highways, schools, health facilities, and many other programs you and your neighbors need. Without a complete, accurate census, your community may not receive its fair share.

Obviously this is a zero-sum game. If my neighbors and I all fill out the form, then you and your neighbors will get less from the common federal trough. But at least we’ll be getting our “fair share,” as the letter tells us twice in three sentences.

But where does the government get the authority to ask me my race, my age, and whether I have a mortgage? In fact, the Constitution authorizes the federal government to make an “actual enumeration” of the people in order to apportion seats in the House of Representatives. That’s all. Not to define and count us by race. Not to ask whether we’re homeowners or renters. Just to ask how many people live here, so they can apportion congressional seats.

I’m not interested in getting taxpayers around the country to pay for roads and schools and “many other programs” in my community. All the government needs to know from me is how many people live in my house. And I will tell them.

More on the census and the Constitution here.

The Census and the Constitution

The Washington Post profiles Daniel Weinberg, assistant director of the Census, who says:

“Since the decennial census is in our Constitution, it is the most important task a government statistician can undertake. The census is key to our democratic society by making sure that our congressional districts are equal in size so that we have representative democracy. To be involved in something that is central to our democracy is pretty exciting.”

Good point. The census is indeed in the Constitution, Article I, Section 2. The Constitution provides that every ten years an enumeration of the population of each state shall be made in order to allocate members of the House of Representatives.

Unfortunately, the census has been loaded down with intrusive questions not authorized in the Constitution and bearing no relation to the constitutional necessity of reapportionment. This year the Census Bureau is boasting of “one of the shortest forms in history,” which is all to the good. Still, it does ask respondents to list their race, which really should be irrelevant to government. And to tell whether they own their home or have a mortgage, in order “to administer housing programs and to inform planning decisions.” (That’s worked out well!) And of course they need age and sex data, in order to facilitate various government programs and mandates and to assist “sociologists, economists, and other researchers who analyze social and economic trends.”

Through the American Community Survey, the Census Bureau continues to ask Americans many more questions, from whether you’re on food stamps to how many bathrooms you have. All very interesting to sociologists and planners, of course, but hardly what Madison anticipated when he and his colleagues provided for an “actual enumeration” of the constituents of Congress.

Writing in Slate back in 2000, Tom Palmer complained that the Census Bureau was selling the census as a kind of Super Lotto: You can’t win if you don’t play! “The numbers are used to help determine the distribution of hundreds of billions of dollars in federal and state funds. We’re talking hospitals, highways, stadiums and school lunch programs.” Come on! Get your piece of other people’s tax dollars!

In 1990 David Kopel reviewed the Census Bureau’s promise of confidentiality.


How Many Uninsured Are There?

The Wall Street Journal’s Numbers Guy tackles the question:

The Census Bureau estimates that the number of uninsured amounts to 45.7 million people. But the agency might be over-counting by millions due to faulty assumptions…

Even though legislation won’t cover many of them, illegal immigrants are especially difficult to enumerate: Few raise their hands to be counted. Prof. [Jonathan] Gruber estimates they make up about 13% of the uninsured today, or nearly six million people of that 45 million number…

Of the rest, some people are eligible for health insurance but don’t know it and many can afford it but don’t want it. About 43% of uninsured nonelderly adults have incomes greater than 2.5 times the poverty level, according to a report released Tuesday by the business-backed Employment Policies Institute.

He left out a few things, though.

The estimate of 46 million uninsured, which comes from a less-than-ideal government survey, has been the occasion of a fraud on the public.  For 20 years, the Church of Universal Coverage told us that 40-some million Americans are uninsured for the entire year.  Then, experts including the non-partisan Congressional Budget Office said that no, 40-some million is the number who are uninsured on any given day, and a lot of those people quickly regain coverage.  The number of Americans who are uninsured for the entire year is actually 20-30 million.  Yet the Church of Universal Coverage kept using that 40-some million estimate as if nothing had happened – even though the meaning of that estimate had completely changed.

The Congressional Budget Office also reports that as many as 15 percent of those 20-30 million chronically “uninsured” are eligible for government programs, so they’re effectively insured.

According to economists Mark Pauly of the University of Pennsylvania and Kate Bundorf of Stanford, as many as three-quarters of the uninsured could afford coverage but choose not to purchase it.  Again, according to the Congressional Budget Office, 60 percent of the uninsured are under age 35, and 86 percent are in good-to-excellent health.

Government intervention has made health insurance unnecessarily expensive for them, so these folks quite sensibly don’t want to be ripped off.  Mandating that they buy coverage is really about hunting them down and taxing them.

Robert H. Frank, A 200% Tax Even Socialists Will Hate

In the latest issue of Forbes, Cornell University economist Robert H. Frank is pushing “A Tax Even Libertarians Can Love.” I hope he wasn’t counting on this libertarian’s support.

What he advocates is “replacing the income tax with a progressive tax on spending. …A family’s income minus its savings is its consumption, and that amount minus a large standard deduction – say, $30,000 a year for a family of four – would be its taxable consumption. …Rates would start low, perhaps 20%, then rise gradually with total consumption. …With savings tax-exempt, top marginal tax rates on consumption would have to be significantly higher than current top rates on income.”

His concept of “significantly higher” includes tax rates of 100-200% on marginal income that isn’t saved.  This is about minimizing affluence, not maximizing revenues.  There is ample evidence from Emmanuel Saez and others that the amount of reported income drops sharply as marginal tax rates rise above 25-30% (and even less on capital gains).

In his 2007 book, Falling Behind: How Rising Inequality Harms the Middle Class, Frank suggests marginal tax rates of 50% above $220,000  and rising to 200%.  Since seniors (like me) commonly finance retirement from past savings, Frank’s tax scheme amounts to rapid confiscation of past savings.

For young people, Frank’s tax can’t possibly encourage savings because it discourages earning any income in the first place.  Consumption is, after all, the motive for both earning and saving.   The prospect of facing future consumption taxes of 50-200% would surely discourage saving much, because the rewards from invested savings (namely, future consumption) would be subjected to such prohibitive tax brackets. Under this steeply progressive tax on unsaved income, any income exempt from taxes today would be subject to brutal taxes whenever folks wanted to buy anything of value, like a car or house, or to retire on their accumulated savings.

In another April 25 piece in The New York Times, Mr. Frank shifts from promoting confiscatory taxes on consumption to defending small tweaks to the current tax regime. “The current [tax] system is much fairer than many people believe, and the president’s proposal will make it both fairer and more efficient.” That comment was aimed at the tea parties.  Yet tax party protesters clearly understood, as Frank does not, that the president’s first wave of proposed tax increases come nowhere near paying for his grandiose spending plans.  My estimate of last October, that Obama’s plans would add $4.3 trillion to the deficits over ten years is now looking much too generous, if not wildly optimistic.

In the New York Times piece, Frank argues that income differences are mainly a matter of luck.  As he often does, Frank pretends to possess evidence about this topic that other economists have missed.  He says, “economists have only begun to realize [that] pay differences often vastly overstate differences in performance.”

In his book, whenever Frank alludes to what “the evidence suggests,” his sources are usually suspect, obsolete or invisible. He claims “regulations, like cartoons are data.”  He cites an unpublished master’s thesis, unidentified surveys and “casual impressions.”

Frank  claims “happiness can be measured reliably” by brain waves.  Explaining this better in the Economic Journal in 1997, he noted that people who say they are happy show “greater electrical activity in the left prefrontal region of the brain” which “is rich in receptors for the neurotransmitter dopamine, higher concentrations of which been shown independently to be correlated with positive affect.”  If we accept the amount of dopamine in the brain as the gauge of happiness, however, then the happiest people are those who routinely abuse crack and meth.

In the second chapter of Falling Behind, his first graph lists a Census Bureau URL as the source for household income data from 1949 to 1979.  Click on that link and you will find the data only go back to 1967.   In reality, all of Frank’s income and wealth graphs actually came from Chris Hartman at inequality.org. Hartman is not an economist or statistician, but a “researcher, writer, editor, and graphic designer with experience in politics, higher education, and publishing.”  Hartman’s non-facts used in Robert Frank’s first graph actually came from a 1994 book from the Economic Policy Institute, reflecting the “authors’ analysis…  of unpublished census data.” Frank’s comparison of CEO pay with “average wages” came from Hartman’s flawed calculations for United for a Fair Economy, which were critiqued on page 131 of my textbook Income and Wealth. And Frank’s demonstrably false claim that “asset ownership has become even more heavily concentrated during recent years” is likewise from inequality.org.

In short, Professor Frank often bases his remarkably strong opinions on fragile facts.

Homeless Scare Numbers

The National Center on Family Homelessness has generated headlines today by releasing a report that claims “one in 50 children is homeless in the United States every year.” That would be a total of 1.5 million homeless children, a truly shocking figure. The number is all the more shocking because the U.S. Department of Housing and Urban Development says there actually only 671,000 people were homeless in 2007 (the last year for which data is available), of which only about 249,000 were people in families. Assuming even one adult per family would mean there were around 166,000 homeless children, far too many, but also far fewer than 1.5 million.

What accounts for the discrepancy? First, the National Center uses an incredibly broad definition of homeless. For example, in addition to those we usually think of as homeless (those living in shelters or on the streets), they also include people “Sharing the housing of other persons due to loss of housing, economic hardship, or a similar reason.” Under this definition, when your out-of-work in-law crashes on your couch, he’s homeless. The National Center also includes people “living in motels, hotels, trailer parks, or camping grounds,” children awaiting foster care placement, and children of migratory farm workers. And, a child needs only to fall into one of these categories for a single day to qualify as homeless.

Second, this study, like the HUD study as well, are not actual counts of the homeless, but estimates and extrapolations based on reports by various government agencies. The Census Bureau does attempt to do an actual head count of the homeless (170,000 in 2000), but that estimate is both out-of-date and generally criticized as an undercount. Still, going from that estimate to 1.5 million homeless children seems quite a stretch.

Homelessness is clearly a problem, and for the children involved, a tragedy, but scare headlines are a poor substitute for thoughtful public policy.