National Review Online ran an article yesterday (Dec. 8th, 2004) by Richard Nadler entitled "Bush's 'Real' Hispanic Numbers: Debunking the debunkers."

It's supposed to be an attack on my interpretation of the 2004 presidential election results. But a close reading shows that it largely supports my contention: the President's unlimited open borders plan would both be bad for the Republican Party (by importing more future Democrats than Republicans) and would not make the GOP more popular with current Hispanic voters.

Nadler's final estimate of the Hispanic results appears to be Kerry 60% - Bush 38%. That's awfully close to my best estimate (60-39). Nadler's number suggests that Bush's share of the Hispanic vote increased by three percentage points over 2000, equal to Bush's growth in the overall vote, and one point less than his growth in the non-Hispanic white vote (from 54% to 58%).

Nadler tries to debunk my claim that the Republicans' share of the Hispanic vote generally goes up and down in sync with the white vote (just skewed way to the left).

But he presents a table of numbers showing that, just as I've argued, in six of the last seven elections, the GOP's share of the Hispanic and white votes have moved in the same direction. (For statheads: the correlation coefficient of the white and Hispanic shares Nadler presents is a high 0.75, But I noticed that he made a typo in my favor. The real correlation is a lower but still strong 0.64. And that correlation is probably artificially lowered by the random errors in the reported Hispanic figures caused by small Hispanic sample sizes, especially in the past.)

In sharp contrast, there is essentially zero correlation (-0.08) between the movement of the GOP's share of the black and white votes—they've only gone up or down together in three of the last seven elections.

This distinction between the behavior of black and Hispanic voters has led to the common characterization of Hispanics as "swing voters," when a better characterization would be "flow voters," since on the whole they go with the overall national flow, just far more toward the Democratic side.

The simplest model of white, Hispanic, and black voting behavior is that voters (at least those who are less than well-to-do and are family-oriented) are on average torn between the Democrats' tax-and-spend policies and the Republicans' family values stances. The poorest ethnic group of voters, blacks, feels they can't afford to waste their vote on semi-symbolic family values issues when they need direct help on bread-and-butter issues. In contrast, the wealthiest ethnic group of voters, whites, can afford to vote for Republicans—both because some are so wealthy that GOP policies like eliminating the inheritance tax are in their self-interest; and because, for the majority, they can afford to vote for family values.

Hispanic voters fall in the middle. Hispanics, overall, are quite poor. But those who are citizens and regular voters tend to be a little better off than blacks, and somewhat more upwardly mobile. They are tempted by the GOP's family values rhetoric. But a large majority feel their pocketbooks demand they vote Democratic.

This suggests that Hispanics are most likely to become Republican voters when, on average, they aren't so poor. The most straightforward way to raise Hispanic average incomes is to stop taking in so many extremely poor Hispanics from south of the border.

(This also has the secondary effect of cutting out the depressing effect on Hispanic wages of the constant arrival of what Marx called "the reserve army of the unemployed" from Mexico.)

But we don't get that kind of broad analysis from Nadler because he has one fish to fry: getting Republicans to spend more money on ads on minority radio and TV stations.

This NRO article, like Nadler's effort in our Insight debate last spring on minority outreach (here is his essay and here is mine) reads like a sales pitch for why Republicans should spend more cash at his old advertising agency, Access Communications Group, an Overland Park, KS firm specializing in producing Republican ads for Hispanic and black radio stations.

In Insight, Nadler was forthright about admitting that he was the former president and a current consultant at Access, but NRO doesn't seem to mention this.

Knowing Nadler's business background makes it much easier to understand his otherwise inexplicably narrow focus on spending more on Hispanic advertising—which amusingly resembles Univision anchorman Jorge Ramos' recent book with its recurrent plugs for advertising on, well, Univision.

So, let me propose a Grand Compromise. I will support the Republican Party spending more money on ads on Hispanic and black radio stations—if Nadler will support a serious tightening up on immigration.

Nadler seems more interested in the quantity than the content of the radio ads, so I don't think he'd mind deep-sixing the President's open borders plan. (Unless he favors it as a way to boost the ratings of Hispanic radio stations.)

Indeed, Nadler actually points out that Republican advertisers have found that amnesty and guest worker programs are not a hot selling point with Hispanic voters.

In the 2002 and 2004 campaigns, only two of the GOP's 47 radio ad scripts aimed at Hispanics even mentioned immigration.

There's a reason for this. America's immigration crisis tends to hurt Hispanic-American citizens most directly (in terms of lower wages, crowding, and overstressed schools). So they tend to have sensibly mixed feelings about additional immigration—witness their 47% vote in favor of Arizona's Proposition 200 cracking down on illegal immigration.

Practical men like Nadler and I can do business on this issue.

The problem is that on this topic, there is a lone extremist. And his views are close to irrational.

He happens to be the President of the United States.

[Steve Sailer [email him] is founder of the Human Biodiversity Institute and movie critic for The American Conservative. His website www.iSteve.blogspot.com features his daily blog.]