Why do you think it's on the test?

Given this from the Nation's Newspaper of Record:

Because of an editing error, an article on Saturday about the concern of some economists that a failure to extend the payroll tax cut could undercut the country’s fragile economic recovery described incorrectly the increase in taxes that would occur without such Congressional action. Taxes on the payroll income of America’s 160 million wage earners would increase to 6.2 percent from 4.2 percent; the change would not amount to a 2 percent tax increase.

And this, from the Foremost Newspaper of the Carolinas:

Employees have paid a 4.2 percent Social Security tax this year, a 2 percent cut.

... three questions (rounded for your convenience):

• What's the percentage increase when a tax rises from 4 percent to 6 percent?
• What's the percentage decrease when a tax falls from 6 to 4 percent*?
• If a tax rate is cut by 50 percent one year and that result is raised by 50 percent the following year, is the resulting tax rate the same as the initial tax rate?

These will continue to be on the final until there's evidence from the field that America's Newspapers -- whether international, regional or hyperlocal, whether delivered in print or through some yet-unimagined satanic tablet device developed in some garage in California -- can get the answers right at better than chance levels.

* Your audience is significantly -- χ² (2 df) 53.25, p = .001, v = 0.648 -- less bothered by "30 to 40 percent" than by "$2 to $3 million," if you're looking for style rules to reconsider.