Caution on the generic ballot, Palin's presidential plans, Obamacare drags down the Dems, and more...
6:30 AM, Sep 2, 2010 • By JAY COST
Welcome. This is a regular feature I'll be offering every weekday, first thing in the morning. Basically, what I'll do is flag the most notable stories of the 2010 midterm campaign, and provide my two cents on what's really happening.
President Barack Obama discusses health care legislation in a speech delivered to a joint session of Congress.
(1) Generic ballot. Still making the rounds a few days after publication is Gallup's out-sized Republican lead on the generic ballot: 10 points, 51-41. Wowsa. That would translate into a 55-45 Republican victory in the two-party vote. That's larger than 1994. Heck, that's larger than 1946. That's 1928 territory, which was back when the GOP won every congressional district in Philadelphia, Pittsburgh, Detroit, Milwaukee, and even San Francisco. And that's GOP +10 among registered voters. Over at Pollster, Harry Enten argues, "the Republican margin on the likely voter model could be 5-10% greater than on the registered voter model."
I think the appropriate posture here is one of caution.
I mentioned a few weeks ago that the Gallup poll is bouncy. Obvious question: Just recently, the ballot had a Democratic lead of 6 points, so has enough happened to justify a 16-point shift so quickly? Similarly, President Obama has gone from -11 to +2 in his net job approval in the last two weeks. This is a bounce that nobody else has found. It's appropriate not to get hung up on the day-to-day or even week-to-week changes in the Gallup poll.
The generic ballot has other issues. My sense of it is that when one party has a lead in the real world, the generic ballot will often overstate that lead in the poll. So, the Dems win by 13 points in 1958, but the generic ballot that fall has them winning by 23. They win by 17 in 1974, but the generic ballot has them winning by 27.
Another point. To get a good sense of what these results mean, we need a decent historical trendline. We don't really have that here. It's not coincidental that this is the first time the generic ballot has the GOP up by this much. It's only recently that the Republicans have become competitive in the race to control the House. From 1932 to 1994, the Democrats held the House all but twice by splitting Northern districts and winning 70-90 percent of Southern districts. Since 1994, the GOP has remained competitive in the North, but now roughly splits Southern districts. That's why a year like 2010 could produce a Republican House majority, but the GOP couldn't even take the House in 1972 when Richard Nixon won 61 percent of the vote. The South was just off the table.
So, the fact that the GOP lead of +10 being "unprecedented" doesn't mean exactly what we might want to think it means. It's as much a testament to Republican non-competitiveness from 1954-94 as much as Democratic doldrums in 2010. In other words, the shape of the congressional elections has fundamentally shifted in the last 20 years. So, I wouldn't take a 10-point lead literally. I'd say a 10-point lead points to a solid House majority for Republicans, but these numbers should be read cautiously.
What's more noteworthy than GOP +10 on the generic ballot is that Republicans in Congress now outpoll Democrats on seven of nine issues, according to Gallup. What's amazing about this is that Gallup tested "Republicans in Congress" versus "Democrats in Congress" and gave voters the potential response of "No Difference." And even with this, the GOP is at or above 50 percent on terrorism, immigration, and federal spending. And on the economy, the GOP hits 49 percent. That's amazing because "Republicans in Congress" has long been a surefire way to generate terrible GOP polling numbers. Not anymore, apparently. Two theories as to why: (a) Republican candidates nationwide are rebranding the party's congressional image; (b) Democrats have really shot themselves in the foot. My money is on (b) with a twist of (a).
(2) Is Sarah Palin running for president? This is a taste of what will begin almost immediately after the midterm:
Her every movement is going to be analyzed and reviewed to divine her motivations. They've already roped Allahpundit in. It's only a matter of time before the speculation fever hits the rest of us!
I'll just say this: people go to Iowa. Lots of people. It's a nice place.
(3) Is Gillibrand beatable? The new Quinnipiac numbers for New York Democratic senator Kirsten Gillibrand are terrible. They have her hovering around 45 percent against all three of her potential Republican opponents. Big reason why: she's running behind in the NYC 'burbs. This is consistent with a developing theme of the Obama Democratic Party: It isn't doing so well in the suburbs that Bill Clinton brought into the party in the 90s. We saw that in Virginia, then in New Jersey. Is it happening in New York, too?
(4) Is Pennsylvania seeing red? This week's Reuters/Ipsos poll has the GOP gubernatorial candidate, Tom Corbett, up 15 points on Democrat Dan Onorato. Plus, Republican Pat Toomey is up 10 on Democrat Joe Sestak in the Senate race. This is a state that hasn't gone Republican for president since 1988 -- and it looks as though there might be no compelling statewide races this cycle. Meanwhile, Charlie Cook has a whopping eight of twelve Democratic Pennsylvania House seats as at least somewhat vulnerable, with four of them as toss-ups (including two toss-ups in the Philly suburbs, which speaks to the previous item).
(5) From McKees Rocks to Columbus? It is looking increasingly likely that John Kasich will be the next governor of Ohio. Rasmussen shows the Republican up 12 against incumbent Ted Strickland, with the latter at just 40 percent. PPP has Kasich up 10, again with Strickland mired at 40%. The GOP's time in the Ohio wilderness after the Taft debacle appears to have been fairly short.
Tom Jensen of PPP summarizes the state of the Midwestern Democratic Party:
(6) Health care. Nate Silver and Jonathan Chait disagree with my recent assessment that the health care law has been a factor in the political decline of the Democrats. Both of them make essentially the same point: you can't prove it! Well...yeah! Absent a poll asking people if their main reason for opposition to the Democrats is health care, the best we can do is make a circumstantial argument.* This kind of argumentation happens all the time, especially over at FiveThirtyEight: Every time Silver offers up a statistical correlation, he's making a circumstantial argument. Nothing wrong with that. And while correlation does not necessitate causation (and all that jazz), there is a very strong circumstantial argument to be made here. Consider the contrary assertion: The president and the Democrats' numbers dropped sharply between Memorial Day and Labor Day of last year, right when the health care debate heated up, then declined again between November and December as each chamber passed their versions of it; yet while the bills were manifestly unpopular, it was not a reason for the decline. Does that really make sense?
For his part, Silver says that my argument is "underdetermined" but also that it's "implausible that (health care) hasn't played some role." I suppose that both of these statements could be true at the same time, but that really requires some nuance, doesn't it?
I could bore you with more polling data to back up my argument, or we could just let Ron Wyden (D-Ore.) break the tie:
Wyden is up for reelection this year. So far, he looks fairly safe. But it is peculiar, isn't it? An incumbent Democrat up for reelection is petitioning to get his state waived from one of the major provisions in the bill.
Somebody needs to tell Senator Wyden that the argument that health care has hurt his party's prospects is underdetermined!
In all seriousness, we can look to the actions of politicians to get a sense of the political effects of health care reform. Are Republicans running against it? Yes. Are Democrats in vulnerable districts running in support of it? Not really. Are some Democrats even running away from it? Yes. Is there a strong correlation between House Democrats who voted no and McCain's share of the district vote? Oh, most definitely.
* In July, Democracy Corps did actually ask people why they disapproved of the president, allowing them to give their own answer. Guess what item was in a statistical tie for number one? (Hint: It rhymes with "realth rare.")
(7) NRSC to outspend DSCC in Illinois. This item from The Hill caught my eye:
A largely-unnoticed development in this campaign season is that the NRCC -- after several cycles of falling far behind the DSCC -- has caught back up in the race for dollars. The DSCC has lots of places to defend, and it looks like they have been forced to make some tough choices.
(8) A Burr in the GOP's saddle? PPP shows Senator Richard Burr (R-N.C.) pulling some weak numbers against Democrat Elaine Marshall, besting her by just 43-38. My former colleague at RealClearPolitics, Sean Trende, opines:
Indeed. North Carolina has long been one of the most diverse states of the South, being one of the few places in the old Confederacy that still had sizeable Republican support (in the west). The Tar Heel state was also a rare Southern exception in that it invested in education and technology much sooner than the rest of the region. It retains a great deal of diversity to this day. The inevitable problem for any senator is that he or she has to harmonize the interests of the voters. The more diverse those voters are, the harder that is to do. Burr has long had trouble with this task, and he's lucky that he's up for reelection in a cycle that is so favorable to his own party.
Jay Cost is a staff writer at THE WEEKLY STANDARD.