Edited by David Leonhardt

The Upshot


Photo
Credit Ellen Porteus

Americans Say They Want Privacy, but Act as if They Don’t

Americans say they are deeply concerned about privacy on the web and their cellphones. They say they do not trust Internet companies or the government to protect it. Yet they keep using the services and handing over their personal information.

That paradox is captured in a new survey by Pew Research Center. It found that there is no communications channel, including email, cellphones or landlines, that the majority of Americans feel very secure using when sharing personal information. Of all the forms of communication, they trust landlines the most, and fewer and fewer people are using them.

Distrust of digital communication has only increased, Pew found, with the young expressing the most concern by some measures, in the wake of the revelations by Edward Snowden about online surveillance by the government. Yet Americans for now seem to grudgingly accept that these are the trade-offs of living in the digital age — or else they fear that it is too late to do anything about it.

“The reason is often they don’t have real choice,” said Marc Rotenberg, executive director of the Electronic Privacy Information Center. “It’s not like picking up the newspaper and realizing ice cream has too many calories and you can start eating frozen yogurt, information that people can act on.”

One reason is that once people are invested in a service — if they have all their social contacts on Facebook or years of email on Gmail, for instance — they have a hard time giving it up.

“It’s this modern economy that doesn’t really rely on price, but on connections and stickiness,” Mr. Rotenberg said. “The companies have done everything they can to make it impossible to go somewhere else.”

Eighty-one percent of people do not feel secure using social media to share private information. Sixty-eight percent feel that way about online chats, 59 percent about text messaging, 57 percent about email, 46 percent about talking on cellphones and 31 percent about talking on landlines.

In each case, those who said they were more aware of reports about government surveillance were more likely to say these communications were not secure.

People harbor equal distrust of advertisers and the government, Pew found. Eighty percent of users of social networks say they are concerned about advertisers or businesses gaining access to their information there, and 70 percent say they are at least somewhat concerned about the government doing so without their knowledge.

Yet highlighting the privacy paradox, 55 percent of people say they are willing to share information about themselves with web companies in order to use their services free, and 36 percent say they appreciate that these services are more efficient because they have access to this information.

The types of digital information that people consider to be most sensitive are their Social Security numbers, health information, the content of emails and phone calls and their location. They are least sensitive about their purchasing habits, media consumption, political and religious views, and the identities of their friends.

People with more education and higher incomes tend to be more sensitive about their online privacy, Pew found. And despite perceptions that young people care little about digital privacy, they often care more than older people. Email is an example. Just over half of all adults consider the content of email messages to be very sensitive. Fifty-nine percent of young adults feel that way, compared with 42 percent of older adults.

“There’s a pretty big, mounting body of evidence that suggests young adults are just as likely to care, if not more so, when it comes to awareness of government surveillance,” said Mary Madden, a senior researcher for Pew’s Internet Project and an author of the report.

Pew offered some evidence that people are inured to the trade-offs of using digital services: Ninety-one percent agree or strongly agree that consumers have lost control over how their personal information is collected or used by companies. They are unsure what to do about it, though.

Nearly two-thirds say they would like to do more to protect the privacy of their personal information online. About the same number think the government should do more to protect them.

There are technical methods for protecting information, like PGP, a data encryption tool. But broader fixes would most likely come through policy changes, privacy advocates say.

“The privacy survivalists will start exchanging PGP key prints, but that’s not going to work for most people,” Mr. Rotenberg said. “I think more needs to be done at the macro policy level to restore trust: update federal privacy laws, limit circumstances under which government gets access and mandate better security.”

The question is whether policy makers or their constituents will push for that — or whether they, too, have accepted privacy trade-offs as a fact of living life online.


Photo
Daisha Tanking, an intern, working at the St. Louis High School to College Center in St. Louis in June. The center helps low-income students make the transition to college by negotiating financial aid agreements, housing contracts and other matters.  Credit Jeff Roberson/Associated Press

Financial Aid, Simplified: A Better College Calculator

The financial-aid calculators maintained by colleges are a case study in both flawed bureaucracy and flawed customer service. As part of a 2008 higher-education law, Congress required colleges to build the calculators — online tools to let families estimate how much they’d really have to pay to attend a specific college.

The idea was a sound one. Many students and parents assume that attending college costs vastly more than it does, because they focus on the published price for tuition. For middle- and low-income families, the real cost tends to be far lower.

But the calculators are full of problems. Students are often asked to create an account to use one, which discourages many from doing so. And the calculators ask a battery of technical questions, many of which are unnecessary to provide a good estimate. Most of the students and parents who manage to start the calculator don’t even complete the process, research suggests.

In 2013, Wellesley — the women’s liberal arts college outside Boston, alma mater of Hillary Clinton, Madeleine Albright and Nora Ephron, among others — decided to take a different approach.

Led by an economist on the faculty, Phillip Levine, the college created a calculator that was much easier to use. Students didn’t have to create an account first. (Anyone can use it, even families with no plans of applying to Wellesley, which is significant because financial-aid offers at selective colleges are often similar.) The Wellesley calculator asks for only nine pieces of information, like annual income, house value, citizenship status and family structure.

In a paper published by the Brookings Institution on Wednesday, Mr. Levine reports the first year of results from the calculator, and they’re positive. They’re sufficiently good that Mr. Levine’s Brookings paper suggests that Congress consider changing the law to require colleges to simplify their existing calculators. Several senators — including Lamar Alexander and Chuck Grassley, both Republicans, and Michael Bennet and Al Franken, both Democrats — have introduced bills to simplify the financial-aid process.

The most encouraging statistics for the Wellesley calculator involve the completion rate. Among people who began entering information, 80 percent finished doing so and received an estimate — compared with only 30 percent who complete the process with Wellesley’s initial calculator, the one mandated by the 2008 law. The average time to enter the information and receive an estimate was only three minutes. “This statistic highlights the simplicity of the tool,” Mr. Levine notes.

Crucially, surveys conducted by the college show that many low-income families who used the calculator were surprised by how low their effective tuition would be. That’s the biggest goal of the project.

“People focus only on the sticker price,” Mr. Levine told me when the calculator was introduced. “The sticker price is a meaningful statistic for roughly 40 percent of our students. The majority of our students are receiving financial aid, and for them the sticker price is an irrelevant number.”

The sticker price at Wellesley — including tuition, fees, room and board — is close to $60,000. Yet a typical family making $50,000 a year would pay in the neighborhood of $6,000, for example.

The biggest downside to the calculator’s first year is the mix of students using it: Most are still not low- or even middle-income families. The users, according to the Brookings paper, “are more heavily tilted towards higher income financial aid applicants, particularly those with total family incomes between $80,000 and $225,000 and away from those with incomes below $50,000.” That last group, of course, has the most to gain from understanding the truth about financial aid.

One final note: When Wellesley first introduced the calculator, I criticized it for not including estimates of work-study wages or loans. Many students help pay for college by working and taking on debt, and excluding these figures gives an overly rosy view of the cost of college. Fortunately, Wellesley has since changed the calculator to include both.

In the past, the hypothetical family making $50,000 would simply have been told that it would need to pay $4,000 out of pocket. Now, it’s told that it will face a $4,000 direct bill and that the student will need to contribute an additional $2,000 from work-study — for a total of $6,000. (Wellesley would not ask the family to take out any loan, although it could choose to do so to cover part or all of the $6,000.)

The change makes the calculator more honest. At a time when the gap between what college graduates and everyone else earn is at a record high, the numbers still show that a degree from a college like Wellesley is a fabulous bargain for low-income students who manage to be accepted.


Why the Voters of 2016 Are Likely to Be Younger and More Diverse

A midterm election has come and gone, and now begins a new round of speculation about whether the Democrats can remobilize young and nonwhite voters in 2016.

The question hinges on how much of the growing nonwhite share of the electorate is the result of President Obama’s unique appeal.

This is a topic that will get a far deeper treatment in time, but for now, we can say that the preponderance of data on demographics and turnout suggests that voters will be at least as diverse as they were in 2012, when Mr. Obama won re-election.

The reason for such confidence? The data on turnout and demographics suggests that the growing nonwhite share of the electorate is primarily the result of demographics, not enthusiasm for Mr. Obama.

The white share of eligible voters has declined from above 75 percent in 2004 to 70 percent today. That five-point drop-off mirrors the 5.5-point decline in the white share of the electorate over the same period.

The white share of eligible voters will most likely slip further, to under 69 percent in 2016. The result is that the electorate will probably be as diverse as it was in 2012, even if nonwhite turnout rates drop and white turnout increases.

The Obama Electorate Isn’t Going Away

Even if nonwhite turnout falls to 2004 levels, the electorate will be as diverse as it was in 2012.

The non-Hispanic white share of the electorate under various turnout scenarios.
2000 Electorate
2004 Electorate
2008 Electorate
2016 Electorate, with 2004 turnout
2012 Electorate
2016 Electorate, with 2000 turnout
2016 Electorate, with average 2000–2012 turnout
2016 Electorate, with 2008 turnout
2016 Electorate, with 2012 turnout
80.7%
79.2%
76.3%
73.8%
73.7%
73.1%
72.7%
72.2%
71.6%
The non-Hispanic white share of the electorate under various turnout scenarios.
2000 Electorate
2004 Electorate
2008 Electorate
2016 Electorate, with 2004 turnout
2012 Electorate
2016 Electorate, with 2000 turnout
2016 Electorate, with avg. 2000–2012 turnout
2016 Electorate, with 2008 turnout
2016 Electorate, with 2012 turnout
80.7%
79.2%
76.3%
73.8%
73.7%
73.1%
72.7%
72.2%
71.6%

That’s not to say that Mr. Obama didn’t drive turnout higher, particularly among black voters. But even there, the contribution of Mr. Obama is smaller than one might expect. Sixty percent of eligible black voters turned out in 2004, compared with 66 percent in 2012.

There are a few confounding issues that will receive more attention in later posts, like the effect of the youth of the nonwhite population, which would tend to suppress nonwhite turnout rates, and the effect of an aging white population, which would tend to boost white turnout. But youth turnout was not particularly high in 2012, and that alone would not be enough to offset the steady diversification of the pool of eligible voters.


Here’s some of our best and most popular interactive work about politics and the economy:
We also like sports. Here's some of our best stuff on that topic:

More Posts

With a large majority and structural advantages, Republicans could keep the House for a generation, frustrating every major Democratic policy priority.

About the Upshot

The Upshot presents news, analysis and data visualization about politics and policy. It will focus on the 2014 midterm elections, the state of the economy, upward mobility, health care and education, and occasionally visit sports and culture. The staff of journalists and outside contributors is led by David Leonhardt, a former Washington bureau chief and Pulitzer Prize winner for his columns about economics.

Read more