Countering charges that the Katrina response was saturated in incompetence because of racism towards impoverished blacks, Bill O'Reilly has claimed multiple times that "75% of the poor in America are white." If he was feeling pithy enough, he could have said that the high percentage of underqualifed blacks in the New Orleans' government was a key factor in the ineffective rescue. But he's a genuinely good guy simply trying to reduce racial tension--clueless as he is about human biodiversity--and the thought police would try to run him into the ground for such acumen.
In any case, what he said is patently false. In 2003 there were approximately 35.9 million Americans in poverty. Roughly 16 million of them were white (8.2% of the white population)--44% of the total population of impoverished residents. Total poverty numbers for other groups are as follows: Blacks 8.8 million, 24.4% of black population and 24.5% of impoverished population; Hispanics 9.1 million, 22.5% of Hispanic population and 25.2% of impoverished population; Asians 1.4 million, 11.8% of Asian population and 3.9% of impoverished population.
Bill may have been attempting a sleight of hand to swell the numbers of the white impoverished. The Census considers Hispanics in the amalgamated category of "white". There are subsequently categories that break this down to "white, non-Hispanic" (to which I referred to as simply "white" above and will continue to do) and "Hispanic". By using the definition that includes Hispanics, whites comprise 67.7% of those in poverty--close to his 75% claim. Of course, this segment makes up over 85% of the US population so that percentage is still relatively low.
To cut through the obfuscation, I've run the numbers as an index. An index score of 1 for a segment means that as a group, the segment has exactly the same proportion of its members in poverty as does the entire national population as a whole. An index score of .5 means a group has only half as many members impoverished (proportionally) as does the entire country, and an index of 2 would indicate twice the poverty rate for the group compared to the country as a whole. In parentheses is the groups average IQ according to Wikipedia--depending on what data is looked at, the scores may vary by couple of points but center around these scores.
Thus, poverty certainly does come in all colors, which was Bill's point. But blacks are three times as likely to fall into poverty as are whites. Hispanics are nearly as likely as blacks, while Asians are much nearer to whites (as per usual). Poverty rates are inversely related to IQ, and the rates by race almost follow suit, except that Asians have higher average IQs than whites. A plausible cause for the higher poverty rates of Asians versus whites is that Asian Americans on average are losing their cognitive advantage over whites as increasing numbers of Asian immigrants are from South Asia (India, Vietnam, Cambodia, Indonesia, the Phillipines) where IQs tend to fall one standard deviation (15 points) or more below IQs of East Asians (China, Hong Kong, Japan, South Korea). The poverty data is more recent than the IQ data, and consequently the IQs have probably followed (inversely) the changing poverty rates. Grouping all Asian Americans into one category (as large classifications tend to) obturates the diversity within it. Still, this minority is not suffering that dastardly white hegemony and oppression that plagues other groups!