The Magazine

You've Got Spam

From the June 16, 2003 issue: The inundation of unsolicited e-mail advertising, and what to do about it.

Jun 16, 2003, Vol. 8, No. 39 • By CHRISTOPHER CALDWELL
Widget tooltip
Single Page Print Larger Text Smaller Text Alerts

WE ARE GOING TO NEED a new way to think about spam, those importunate unsolicited e-mails advertising products, pandering to vices and insecurities, and bearing headers like GET LOLITA OUT OF DEBT BY ADDING THREE INCHES TO YOUR MORTGAGE! The problem is changing before our very eyes. Shortly after the turn of the year, I logged on to America Online's spam report and read that the company's new blocking software had for the first time diverted 1 billion unwanted e-mails in a single day. As this article went to press, I signed on again and found that AOL now routinely has days on which it blocks 2.3 billion pieces. When I wrote an article on spam in late April of this year, the freshest data showed that 70 percent of AOL's e-mail was spam. By the time David Streitfeld of the Los Angeles Times wrote a piece on the same subject a month later, the figure had risen to 80 percent. (It had been 50 percent in January.) Brightmail, which makes spam-blocking software, estimates that between January 2002 and March 2003, the percentage of e-mails that are spam nearly tripled, from 16 to 45 percent. Yahoo!, according to Streitfeld, has seen its spam level quintuple in the past year.

My mailbox at The Weekly Standard, spam-free a year ago, is now more than 98 percent unsolicited mail, entailing a 15-minute discard operation at the beginning of every working day. No one writing me a letter to the address listed on our website should count on its reaching me. I've begun a gradual retreat from office e-mail, and will soon conduct my correspondence the way I did when The Standard was founded in 1995--via telephone and the U.S. Postal Service. Many Americans have made a similar retreat, finding that the extraordinary efficiencies that e-mail brought in the first half-decade they used it have evaporated--and in some cases have turned into inefficiencies. According to Britain's New Statesman, 13 percent of e-mail users have changed their addresses since the start of the year, in order to escape spam.

As the size of the problem changes, so does its nature. Two years ago spam was a joke. A year ago it was an annoyance. And a few weeks ago, Earthlink executive David Baker told a reporter that spam "has the potential to render the Net virtually unusable."

Spam is increasing because it is an easy way for dumb people--and a safe way for dishonest people--to make money. Once you have a reliable mailing list (and it is possible to buy target lists for as little as $500 per million names), then you can operate a permanent spamming operation at marginal cost approaching zero. This, in turn, means that response rates approaching zero can still turn a profit. A figure commonly bandied about is that only 1 in every 100,000 targets need respond to allow a spammer to make money. One occasionally reads of spammers getting rich off a 1-per-200,000 rate. When a spammer sends out a billion unwanted e-mails a day, as Alan Ralsky of Michigan and Boca Raton does, according to Computerworld magazine, he can get very rich indeed.

The addresses are practically all the spammer pays for. All the other costs are externalized, falling on consumers, other businesses, and government. Business groups estimate $9 billion in productivity will be lost to spam this year. Consumers pay for spam through time lost deleting and through phone bills while they do it, telecom companies subsidize spam through the bandwidth they build, Internet service providers must install new machines to accommodate it, and everyone pays for spam in the slowdown of Internet traffic.

What a line of work! Many businesses receive invisible subsidies, but probably none defends its privilege by shoveling out such a steaming-hot pile of libertarian malarkey as the spam trade. One would expect this from lobbies like the Direct Marketing Association and EMarketersAmerica.org, who decry the "mob mentality" of anti-spam activists. But there is also an extraordinary self-righteousness on the part of the mass e-mailers themselves. Spammer Bob Dallas told the New York Times that blocking mass, unsolicited e-mails "is against everything America stands for. The consumer should be the one in control of this." In congressional hearings last month, Ronald Scelson, the "Cajun Spammer," who sends over 200 million pieces of junk e-mail a day, complained that AOL and other Internet service providers were trammeling our rights as Americans. "The carriers right now," Scelson warned, "are deciding and filtering . . . whether you're going to read and see your mail or not. This is censorship. I was brought up and fought for this and still fight for this because I believe in freedom." The guy telling you WHERE TO MEET HORNY CHICKS IN YOUR NEIGHBORHOOD only looks like a pornographer. He's actually Washington crossing the Delaware.

The market will take care of things, of course. Eventually it will be glutted with spammers, profits will fall, and the amount of spam will stabilize. Unfortunately it seems likely to stabilize at a level higher than that at which people are comfortable being on the Internet at all.

POTENTIAL SOLUTIONS to the spam crisis tend to be of two kinds: technological and legal. Those who want to maintain the libertarian ethos of the Internet urge that we simply put as much technological expertise to work fighting spam as the spammers do producing it. Paul Graham in the Internet magazine Network World, for instance, rightly says there are two ways of stopping spam: Keep it from entering your inbox, or keep spammers from sending it. Since spam relies on profit-making, Graham says, "if you solve the first problem, you also solve the second."

Unfortunately, Graham's optimism will be totally unfounded as long as the cost of sending spam to a non-existent or blocked mailbox is zero. What's more, it is by no means certain that the sophistication of spam-blocking technology is outstripping the sophistication of spam-sending technology. Russian-invented programs called Jeems can enter vulnerable home computers, transform their own identities periodically, and send out hundreds or thousands of spam packets in a session, totally undetected. Vast efforts have already gone into technology to block spam, and they appear insufficient to deter it. The primary tool that exists today is the "Bayesian" filter, which seeks out words like "Viagra" and phrases like "online gambling." Spammers have long been able to evade such filters with subtle misspellings (TURN HER ON WITH HERBAL VIARGA!).

That's why several federal lawmakers urge legislation. Most of it is toothless. The RID-Spam Act, sponsored by Republican congressmen James Sensenbrenner of Wisconsin and Billy Tauzin of Louisiana, would require commercial e-mailers to offer an "opt-out" choice to their targets. A spammer could be fined for continuing to solicit consumers who had explicitly requested to be left in peace. Anti-spam activists are unimpressed. Given the galaxy of addresses and identities used by most spammers, this would be far less effective than the "opt-in" model used in the European Union--where no company is supposed to send e-mail solicitations at all unless a customer has explicitly asked to be sent them.

Senators Ron Wyden, a Democrat from Oregon, and Conrad Burns, a Montana Republican, have their own bipartisan opt-out bill that adds a demand for honesty in tag lines. Thus, a spammer couldn't pitch online gambling under the headline, BAD NEWS ABOUT YOUR MOTHER. Two thirds of spam mailings are fraudulent in some way, according to the Federal Trade Commission--including 44 percent that come from fictitious addresses. (Apologies to HotKandySue78904@yahoo.com if I'm being unfair here.) The merit of the Wyden-Burns approach is that it allows us to prosecute many Internet abuses under the existing trade laws. Maybe that would help, since it is estimated that 90 percent of spam comes from about 200 practitioners. But maybe it wouldn't.

What's more, there are two separate problems with spam--the quantity of it and the quality of it--and the Wyden-Burns measure deals mostly with the former. A parent upset that his children are getting hundreds of solicitations to GET QUAALUDES WITHOUT A PRESCRIPTION every year is not going to be delighted when they're receiving "only" dozens. Which brings us to one of the great disappointments--perhaps the central disappointment--of the computer age. Americans were sold on the Internet as an educational tool, and have been cajoled by Al Gore and like-minded politicians into spending vast sums of money to subsidize it as such. But thanks to spam, the "information superhighway" has become a rather racy place for children to travel unsupervised.

Under such circumstances, draconian steps are in order. Of the measures being discussed in Congress, the most sensible--sensible because it is what the spam problem will inevitably lead the public to demand within a year or two--is that of Democratic senator Charles Schumer of New York. Schumer wants a federal no-spam list like the no-call telemarketing list the FTC will have up-and-running by this summer. This is "opt-out" with a vengeance--consumers will be able to opt out of the unsolicited commercial e-mail cesspool altogether. Questions have been raised about the measure's feasibility. One enforcement question--how to identify the source of e-mails, which are harder to trace than telephone calls--will have to be addressed technologically, perhaps by the computer industry. But most of the quibbles about a do-not-spam list have the phony-baloney quality that is the hallmark of arguments traditionally made against Internet taxation. First comes feigned helplessness. Behind a lot of cant about the "ethereal" nature of human relations in the telecommunications age, the solution is deemed technologically unworkable. Then when taxation, or spam-blocking, as the case may be, is shown to be quite feasible, the argument is advanced that enacting it would "undermine the free-for-all spirit of the Internet."

But this libertarian argument, particularly when mustered on behalf of spam, is the most phony-baloney argument of all. If we could only "unleash" this and "untrammel" that, politicians used to argue in the 1990s, we'd get higher profits and more freedom. So, by bipartisan consensus, a no-tax, low-regulation regime was devised for the Internet. It was market Rousseauianism, and for several years, the Internet economy has allowed us to conduct a long experiment on how the noble savage comports himself in cyberspace. Libertarianism has proved an attractive creed for the Internet generation in its lifestyle variant of live-and-let-live. But as a market system it has proved a flop.

The Internet economy, as spam shows, turns out to be like a garden: Leave it alone and you will not get (as you might assume in theory) a profusion of wild and interesting growth. No--you'll get the entire space choked off by the most noxious and aggressive weed. And spam has reached the point where it calls for a mighty pesticide. An entire range of federal regulations is going to be necessary if the Internet is to be kept usable; and enacting such regulations responsibly will take legislative prudence and care. A do-not-spam list is a first imperative. But it is also a social necessity that the principle of taxing the Internet be established soon. This will mean retiring the (in retrospect) absurdly named Internet Tax Freedom Act of 1998, which placed a moratorium on certain Internet taxes, and was extended in 2001 until November of this year.

It was always unfair not to tax business on the Internet, of course. There is no reason that Amazon.com should enjoy a pricing advantage (a de facto government subsidy) over a corner bookstore. But the most damaging part of the moratorium turns out to have been the most innocent-looking: that it banned charges for Internet access. Something like e-mail "postage" will be required if we are going to change the economic incentives that have invited pornographers, snake-oil salesmen, and other social predators into Americans' living rooms, in some cases hundreds of times a week. There are reasonable ways such postage can be collected. A penny-per-e-mail charge would drive most spammers out of business, subject them to jail time for tax evasion if they hid their operations, and cost the average three-letter-a-day Internet user just ten bucks a year. If even that seems too hard on the small user, then an exemption could be made for up to 5,000 e-mails per annum. If the postage were decried as a tax hike, then it could be used to fund one-to-one tax cuts in other areas--like sales taxes for the brick-and-mortar retail stores that have labored under such an unfair tax disadvantage for the past half decade.

Such programs can be argued over and adjusted in the coming months. But there is no chance that the Internet will return to its old level of user-friendliness until lawmakers recognize that the decision to leave it unregulated was a serious, ideologically driven mistake.

Christopher Caldwell is a senior editor at The Weekly Standard.