Jews are about 2% of the population, but as one of the oldest age groups, and with high voter participation rates, Jews represent a larger share of the electorate, probably between 2.5% and 3%. Jews are concentrated in states Democrats usually win, with over 60% of the nation’s Jewish population in New York, Massachusetts, New Jersey, California, Illinois, and Maryland, and another 20% in the five battleground states of Florida, Pennsylvania, Nevada, Ohio, and Virginia.
Do the results in New York 9 signify a major shift in Jewish voting patterns? If they do, how will this impact the 2012 presidential and congressional races?
The national exit polls and earlier surveys suggest that the last Republican who won the Jewish vote (by a plurality, not a majority) was Warren G. Harding in 1920. In that race, the Socialist, Eugene Debs, garnered 38% of the Jewish vote, doubling the Jewish share for the Democratic candidate.
In more recent times, the last Republican to run close to the Democrat among Jewish voters was Ronald Reagan against Jimmy Carter in 1980 (Carter 45%, Reagan 39%, John Anderson 15%). Carter, of course, suffered a humiliating defeat in his re-election run, winning only six states, 49 Electoral College votes, and just 40% of the national popular vote. Carter’s Jewish percentage was only 5% higher than his national vote percentage.
In 2008, Barack Obama won 53% of the national popular vote, but was estimated to have won 78% of the Jewish vote, a 25% difference. If approval levels are indicative of intent to vote for a candidate, then the current gap is 11% (55% approval for Obama among Jews, 44% among all voters).
The New York 9 results suggest that the higher number of Orthodox Jews in the district is not sufficient explanation for the strong performance by Turner among Jewish voters. In heavily Orthodox and Russian precincts in New York in 2008, Jews voted 2 to 1 for McCain over Obama. Since these two groups comprise, at most, 15% of Jewish voters nationally (Orthodox Jews are a younger than average age cohort, with larger average family size), their votes among the total Jewish vote would have been 10% for McCain and 5% for Obama. All other Jews would have voted 73% to 12% for Obama over McCain (or by a six to one margin) to get to the 78% to 22% overall vote among the Jewish community.
Assume that the Orthodox plus Russian vote (estimated at 7% of Jews in the district) was 40% of the total in New York 9, and assume it went by 3 to 1 for Turner — a stronger result than for McCain — or 30% for Turner and 10% for Weprin, who is an Orthodox Jew himself. If the overall Jewish vote in the district was 50% for each candidate, then the non-Orthodox, non-Russian vote among Jews was 40% for Weprin and 20% for Turner — a 2 to 1 margin, down from the 6 to 1 margin among this group for Obama in 2008.
Now take these results by subgroup and apply them to the national Jewish voter distribution in 2012, assuming the Orthodox/Russian group is up to 16% of the total thanks to a slightly faster population growth than all Jews. If this group goes 3 to 1 for the Republican, that is a 12% to 4% split. If the remaining 84% of Jewish voters break 2 for 1 for Obama, that is a 56% to 28% split. In total, the Jewish vote for Obama would be 60% and for the Republican it would be 40%. The last Republican to earn 40% of the Jewish vote was Dwight Eisenhower in 1956. No wonder messaging to Jews has become a big issue for Democratic strategists.
One can adjust Obama’s percentage within each category by a point or two, without impacting the overall results all that much. If Obama receives 65% of the Jewish vote, it would still be the worst performance for a Democratic presidential candidate since Walter Mondale in 1984.
There is a long way to go until the 2012 vote. I think Waxman is right that some Republican candidates would do better among Jewish voters than others (Romney in particular). Jewish liberals seem to have an allergic reaction to any GOP candidate who is too visibly or outspokenly Christian. But if Obama’s margin among Jewish voters were to drop sharply (from 56% in 2008 to 20-30% in 2012), that would represent a net shift of about 100,000 to 125,000 votes in Florida, and close to half that range in Pennsylvania. Elections have been decided by a lot less than this.