Starting with bad data

I found an article about cybercrime at Wired.com an it included the following:

It's been estimated that last year alone cyber criminals stole intellectual property from businesses worldwide worth up to $1 trillion," said President Obama in his first major address on cybersecurity, back on May 30, 2009. "In short, America's economic prosperity in the 21st century will depend on cybersecurity."

$1 trillion is an impressively large, perfectly round number. The only trouble is, it's not correct. McAfee, the very organization that came up with the original estimate, announced Monday that the actual and extrapolated losses from online espionage, hacking, and cybercrime probably fall closer to $100 billion--one-tenth of the original figure. That's less than 1% of the U.S. GDP and in line with other minor costs of doing business, like losses from employee "pilferage." That's the cost to both businesses and governments combined, by the way.

Of course the press is not giving the corrected numbers the same play they gave that beautiful 1 trillion number. After all, corrections are not exciting or sexy and 100 billion just doesn’t have the same eye catching punch as that $1 trillion.

The tag line from the article made the whole thing worth reading:

One thing is clear, though: It's hard to make good policy when you start with bad data.

Might this be a key to the largest problems we face now, the misinterpretation of all that data floating around out there?