The latest is 14.1. But Mitt Romney’s tax rate is only one of many touchstone percentages doing the rounds. The U.S. presidential hopeful has also contributed 47, the percentage of Americans he says are dependent on government handouts. Both those numbers have irritated a lot of people. The challenger would prefer that the focus shift to a number which embarrasses President Barack Obama. A good candidate is 8.1 - the percentage of the labor force which is unemployed. Percentages frame many current debates. The Occupy Wall Street movement contributed the 1 percent of excessively rich Americans and economists Carmen Reinhart and Kenneth Rogoff offered the 90 percent ratio of government debt to GDP as a borderline of fiscal ill-health. Investors have decided that euro zone sovereign bond yields below 7 percent, or depending on the day 5 percent, mark the end of the crisis. In China, the GDP growth rate is the magic number - 7.5 percent is the current talisman. This percentage-mania is basically a good thing. To start, it is more sophisticated than a round number fixation - oil at $100 a barrel, the U.S. national debt at $16 trillion or the S&P 500 at 1500. Percentages are ratios, so they provide at least twice the information as a single number. More sophisticated mathematical measures might be even more helpful - second derivatives and standard deviations, anyone? But most investors, let alone voters, will struggle to understand why. More importantly, these percentages provide helpful markers in a more complicated world. Sure, it’s arbitrary to say that 14.1 is low, while 25 or 40 is a reasonable percentage tax rate, and it’s silly to reduce the development of China’s to a single number. But the magic ratios make it harder to ignore problems and deny failures. Still, accountants are justly criticized for choosing to be precisely wrong rather than roughly right. Any numerical obsession in the inevitably murky world can easily degenerate into sloppy thinking. The effort to meet a clear percentage target can encourage statistical legerdemain, warp judgments or distort policies. Do the math: There is a 100 percent need not to rely too much on the numbers.