Mar 11th 2004
From The Economist Global Agenda


In the biggest citation yet of a new American anti-spam law, Microsoft and three other industry giants have filed suit against alleged junk e-mailers. But the war on spam can only be won using technology as well as the courts








AT THE World Economic Forum’s get-together in Davos in January, Bill Gates predicted that unsolicited e-mail would cease plaguing the world by 2006. A foolhardy forecast, no doubt—what global problems ever get fixed in two years?—but this week Mr Gates took a small step towards keeping his word. On Wednesday March 10th, Microsoft and three other technology giants—Yahoo!, America Online (AOL) and EarthLink—announced that they had filed lawsuits against more than 100 people they accuse of sending out countless junk marketing e-mails, known as spam. The lawsuits, filed in federal courts in California, Virginia, Georgia and Washington state, were among the first to cite a CAN-SPAM law that took effect across America in January. The law, which was overwhelmingly passed by Congress last year, makes it a crime to send junk e-mail using misleading subject lines or false return addresses. It creates more uniform protections than the patchy laws passed by individual states.

Spam is a headache for almost everyone. Some 62% of all e-mails are unsolicited junk promising cheap university diplomas, penis enlargement and other delights, according to Brightmail, an anti-spam group (see chart). This week’s lawsuits are just the latest courtroom assault against the spammers. Last June, Microsoft, whose MSN/Hotmail e-mail service is one of the world’s most popular (and junk-ridden), filed civil lawsuits in state courts against 15 alleged spammers. AOL, likewise prodded by angry customers, filed suits of its own last year. And EarthLink has already collected millions of dollars in anti-spam settlements.

However, legal action is of limited use. Spammers crop up like mushrooms after rain, so it will be impossible to prosecute everyone. Moreover, even if most American spammers are somehow shut down, overseas ones will have few inhibitions, out of reach of American law (though Microsoft has already gone after one or two junk e-mail senders in other countries). So e-mail providers are scrambling to find solutions in technology as well as the courtroom. They are weeding out those mass e-mails they can, or asking senders to confirm their identities.

The trouble is that most anti-spam measures hamper e-mail use in other ways—on top of which, spammers are a wily bunch who keep coming up with new ways to evade detection. Talk is now turning to ways of forcing spammers to pay to reach recipients (spammers find e-mail so alluring because it is free, unlike regular post, the fax or the telephone). In Davos, Mr Gates said that Microsoft planned to create an electronic stamp of sorts that would allow friends and relatives through for free, while charging other e-mailers. Other companies are also working along those lines (see article).

This week’s lawsuits may be most notable for bringing together four often-bitter rivals. AOL, Earthlink and Yahoo!—all of which are internet service providers—have little love for each other, or for Microsoft, which is a known bully; it has been rapped on the knuckles by an American judge for building its own web browser into its Windows operating system and thus crushing makers of rival browsers. Such domineering tactics may also get the company into trouble with the European Commission, which is expected to issue a ruling this spring on—among other things—Microsoft’s treatment of RealNetworks, the company behind a rival music player.

But the feuding seems to stop when it comes to spam. The technology giants have banded together before against the scourge—AOL, Microsoft and Yahoo! agreed last year to share information about spammers. This week’s lawsuits, an extension of that co-operation, signal the industry’s determination to get on top of the problem.
Make 'em pay

Feb 12th 2004 | SAN FRANCISCO
From The Economist print edition


The dismal science takes on spam

THE short history of society's fight against spam—usually defined as unwanted commercial e-mail—may be about to pass into a significant third phase. In the first phase, it was geeks who led the resistance, using techie weapons such as e-mail filters with fancy Bayesian mathematics. In the second phase, politicians joined in, eager to get their names on to new legislation—in America, for instance, 36 states and Congress have passed laws of some sort against spam. Now, in the third phase, the economists are taking over.

The market opening for the economists is obvious. Both the geeks and the politicians are widely seen to have failed miserably. E-mail filters, though helpful, often block genuine e-mails, while still letting through some spam. And legislation, so far, seems to have had no effect at all, probably because hard-core spammers have moved offshore. More than half of all e-mail is now spam, and the share is rising.






To an economist, the reason for this mess is obvious. “At the moment, both the sender and the receiver of an e-mail pay a marginal price of zero,” says Benjamin Hermalin, a professor at Berkeley. “That must be wrong,” for e-mails do have marginal costs (such as electricity, storage and network congestion). For spammers, this presents a sure-win strategy: mail as much as you can, because even one hit out of a million is profitable.

An obvious solution, then, is to charge for e-mail, or at least for the sort that spammers send. This idea has even been endorsed by the über-geek, Bill Gates of Microsoft. At the World Economic Forum in Davos last month, he predicted that fancy technology and payment schemes would together defeat spam within two years. (Sceptics noted that Microsoft could also help by fixing security flaws in its products—the latest confessed to this week—that can be exploited by spammers.)

To libertarians, among the earliest internet enthusiasts, the notion that e-mail should be un-free is anathema. They would fiercely oppose any economist daring to suggest charging “ordinary” people. Besides, any system that would rely on processing millions of tiny “micropayments” would be an administrative nightmare. In short, a good system has to remedy the perverse incentives that attract spammers, but leave everyone else alone, and thus better off.

Goodmail Systems, a start-up in Silicon Valley, hopes to launch such a solution this summer. Its idea, says Richard Gingras, its boss, is to create “a new, trusted class of e-mail from volume senders” which recipients would instantly spot as such in their inboxes. An example: Amazon, keen that its automatic order-confirmations not be filtered out as spam by mistake, would buy a million e-mail “stamps” for $0.01 each from Goodmail, which would share the $10,000 with participating internet service providers (ISPs). Goodmail fixes the stamps to Amazon's e-mails as encrypted e-mail headers, and sends the decryption key to the ISPs. The ISPs can then identify Amazon's e-mails as they arrive, and pass them on to the inboxes of consumers.

Under Goodmail's egalitarian system, spammers would also be free to buy stamps, just as senders of junk-mail buy postage from the post office. This, however, would introduce into their equation the normal economics of marketing. For most hawkers of penis enlargements, one can assume, the numbers would not add up, so they would not buy postage. They would therefore be left in a shrinking subculture of illegitimate spammers, which ISPs can filter out more easily. Thus, says Mr Gingras, Goodmail's system would reduce (though not eliminate) spam, even if it is adopted only partially.

A different approach is not to charge senders for every message upfront, as Goodmail will, but only for those that lead to complaints from recipients, and in arrears. This is the route taken by IronPort Systems, another firm in Silicon Valley. For a bit over a year, it has been offering a service in which legitimate bulk e-mailers, such as newsletters, post a financial bond in return for assured delivery past spam filters into inboxes. Senders have to honour the bond and pay only if recipients complain. “We've designed the financial incentives to be insurmountable for spammers,” says Craig Sprosts, a director. The idea, as with Goodmail, is that this whitelist eventually grows so large that spammers will in effect be quarantined, as though they were on a blacklist. About 18,000 ISPs have already joined the system, and over 200 senders, including Google, a search engine, have posted bonds.

Some people want to go even further. Balachander Krishnamurthy, a boffin at AT&T Labs, the research arm of the phone company, proposes a system in which ISPs would establish a consortium—similar to banks creating Visa—that would act as a clearing house. ISPs would then give all of their subscribers credit limits. From then on, every time a recipient declares an e-mail “unwanted” to the clearing house, the sender is charged, say, $1. Once his credit limit—$200, for instance—is reached, the ISP shuts down his account.

This neatly addresses another big problem. The most insidious way of spreading spam these days is by writing software viruses, such as last week's MyDoom, that turn the computers of innocent consumers into spam machines. In Mr Krishnamurthy's system, however, the credit limits of infected users would max out within seconds of a new virus spreading, stopping it from travelling further. Presumably, ISPs will know enough to realise that grandma was the victim, not the hacker, and will eventually reconnect her