I mean't to blog on this before, but it got away from me. Jane Galt blogged about Mike Kinsley's article on Halliburton, and she makes much hay out of his "conflation" of income with revenue.
Revenue is every dollar you take in for providing goods or services to someone. Income is what's left after you subtract all the costs of providing the goods and services -- your profit. This distinction is not known to most readers, who assume that corporate income is the same as person income -- in other words, what you get paid. So comparing those two figures, those readers will figure that the manipulation accounted for a full 1/4 of income -- or, in other words, that Dick Cheney had to have known about a manipulation that increased his company's income by almost 33%.
Now, construction profit margins are typically slim, so as soon as I saw him compare that damning $100 million in revenue to the $438 million in income, I knew that this was a spurious comparison. But I didn't guess how spurious. In 1999, the year Kinsley selected for comparison (because the effects of this particular manipulation are one-off), Halliburton had over $10 billion in revenue. In other words, the change in accounting altered revenue by less than 1%. Suddenly "he must have known" doesn't sound as compelling, does it?
This critique is highly misleading, because the added 100 million is not, strictly speaking, proportional revenue. In other words, this accounting change doesn't add 100 million to the top line (revenue), and the proportional value (438 million/10 billion =about 4.4%) to the bottom line. It's free revenue, so it adds 100 million to both the top and bottom line. So Kinsley was right, and Megan wrong, in this instance (still love the blog, though!).