Conventional Wisdom: The Last Loser In the 2002 Election

Posted September 22, 2003 at 5:40pm

One of the great difficulties for political analysts after the 2002 elections was the lack of exit polls to help sift through what was clearly a historic election upset for Republicans. Because of a computer glitch, Voter News Service, a media consortium that has provided immediate election data for networks, newspapers and wire services for the past decade, could not deliver results on election night. I know. I was there trying to call House races for VNS.

Last week, the VNS data was finally released, and the numbers blow away three key political assumptions that could have significant implications for the 2004 presidential and Congressional elections.

[IMGCAP(1)] Wrong Assumption No. 1: “When it’s the economy, stupid, Democrats win.”

The VNS data showed that, contrary to conventional wisdom, Republicans were able to win and win big when the issue agenda focused on domestic matters. The top five issues for voters in deciding how to vote for Congress were: the economy (29 percent), health care (14 percent), education (12 percent), terrorism (12 percent) and Social Security (11 percent). Iraq garnered 3 percent. If combined with terrorism, security issues would have placed second with 15 percent.

However, the economy, education, health care and Social Security represent a whopping 66 percent of the issues that voters said were most important to them. Many analysts explained the Republican victory by claiming security concerns eclipsed domestic issues, where Democrats have traditionally been stronger. The data shows this argument is dead wrong.

Among those voters who said the economy was the most important issue (29 percent), Republicans won by a margin of 51 percent to 46 percent; by 65 percent to 29 percent, voters said President Bush’s tax cuts, a key element of his domestic agenda, were a good idea. With the economy likely to be a dominant issue in 2004, the VNS data shows that an improving economy coupled with Republicans’ clear advantage on security issues could provide a positive election environment for the GOP.

Wrong Assumption No. 2: This is a 50/50 partisan-split country.

The VNS data shows that this is a center-right-leaning country, not the ideologically driven 50/50 electorate that has become the conventional wisdom after the razor-thin 2000 election. In fact, in the 2002 elections, Republicans won House races nationally by 51 percent to 46 percent, and the number of self-identified conservatives increased from 30 percent to 34 percent.

When it comes to assessing the political persuasion of voters, however, it is important to look at election trends over time — not a single election. But even there, the news isn’t good for Democrats.

In the 2002 elections, the number of self-identified liberals dipped to the lowest level in the past four elections — 17 percent. During that same time frame, moderates continued to be the dominant voting group, going from 45 percent of the electorate in 1994 to a high of 50 percent in 2000. In 2002, they dipped slightly, by 1 point, as conservatives reversed what had been a declining trend over the previous three elections (1996, ’98 and ’00), coming back to 34 percent after reaching a low of 30 percent.

Taken over time, the VNS data shows that the number of self-identified liberals is stagnating at best or declining at worst; that the nearly 2-1 ratio of conservatives to liberals gives Republicans a major advantage; but, most important, that the moderate middle remains the political force to be reckoned with for both parties.

Wrong Assumption No. 3: Republicans can’t attract minority voters in significant numbers.

The VNS 2002 data shows that Republicans are making steady and significant inroads against what has been the Democrats’ traditional strength with Hispanic voters. Comparing the 2000 and 2002 elections, the numbers show Republicans made a net gain of 7 points as Hispanic voters went from voting 35 percent for Republicans and 64 percent for Democrats to 38R/60D. (Data before the 1998 elections cannot be compared because of a change in the way VNS defines the Hispanic vote.) Not good news for the national Democrat coalition.

Clearly, the 2002 VNS exit poll data is positive news for Republicans, but each election has its own dynamics and 2004 will be no different. Whether Republicans can sustain the gains they have made with moderates and Hispanic voters in recent elections remains to be seen. But as we’ve learned the hard way this week, it’s no fun going against the prevailing winds.