Margin of Σrror

Margin of Σrror -

The US healthcare paradox: we like the Affordable Care Act but fear Obamacare | Harry J Enten

No wonder Republicans can campaign against the ACA when a plurality of Americans still believes it includes ‘death panels’

President Obama’s healthcare law is hated and loved by some so much that they are willing to shut down the government over it. I’m not sure I’ve ever seen so much passion over an issue about which so few (myself included) know as much as we should.

I wrote about this divide when Obamacare was in front of the US supreme court. Americans were opposed to “Obamacare”, or the Affordable Care Act, yet they were in favor of many of its provisions. Not surprisingly, Americans lacked knowledge of what exactly the law did.

So, as the political fight has intensified, on the eve of implementation of one of the ACA’s key provisions, the creation of new health insurance pools, how much has changed? Does the noisy debate on the ACA mean Americans are better-informed than before about Obamacare? Here are five ways Americans’ opinions about Obamacare have and have not evolved over the past year.

1. Americans have grown more negative in their views

The HuffPollster chart tells the story fairly well. It includes polls that ask about Obamacare and the Affordable Care Act (ACA).

Last June, the split was about 47% opposed to, and 40% in favor of, Obamacare. That gap narrowed, after the supreme court upheld the law’s constitutionality. The difference between those for and against the law dropped to only 2pt by the election.

Since the election, however, US public opinion has changed drastically: 52% of Americans now oppose Obamacare – that’s tied at the record high. A little less than 38% are in favor of it, according to the HuffPollster aggregate.

2. The Obamacare v the Affordable Care Act difference still exists, though may be overblown

A few weeks ago, a CNBC poll purported to show major differences when asked about the law in different manners: 46% were against Obamacare, while only 37% were opposed to the ACA. Importantly, support for the healthcare law also dropped from 29% to 22%. What’s happening is that Americans have heard a lot about “Obamacare”, but not much about the ACA. The key is the difference between those who favor and those who oppose for each question, which is about 15pt.

Most of the polls that show the best numbers for Obamacare (that is, the margin of opposition at 10pt or less) don’t mention Obama’s name. These include the Kaiser and ABC/Washington Post surveys. A Fox News poll found the gap was slightly wider, with the margin between favorable and unfavorable towards Obamacare at 26pt, and just 16pt for the ACA. Some, such as the UConn/Hartford Courant poll, use similar wording and find the largest gaps.

The bottom line is that many Americans oppose Obamacare no matter the wording. But Obama’s name probably makes them even less likely to like it.

3. There is a percentage of Americans who oppose the law for not going far enough, though this, too, is likely overblown

Democrats like to point out that even as a majority opposes Obamacare, a certain percentage of Americans think it’s because the law is “not liberal enough”. A new CNN poll puts that percentage at 11%. When you add those who favor Obamacare to those who regard it as “not liberal enough”, you have a near-majority.

The problem with this finding is that I don’t believe that Americans necessarily know what “too liberal” means in this context. My evidence for that is that the group with the highest percentage of those who say they are against the law because it is “too liberal” are, in fact, Republicans.

The actual percentage who don’t like Obamacare because it’s not liberal enough is probably closer to 7%, if not lower. That’s the percentage Kaiser found when they asked if Obamacare went “far enough” in changing the healthcare system. I caution, however, that some of that may be those who want radical conservative change – such as yearning for government to get completely out of healthcare.

4. Americans continue to like the individual provisions, except for the individual mandate

While only 37% of Americans viewed the ACA favorably in a March 2013 Kaiser poll, most liked what the healthcare bill is scheduled to do. Over 55%, and up to 88%, of Americans regard the following facets of Obamacare at least somewhat favorably: tax credits to small businesses to buy insurance, closing the Medicare “doughnut hole”, creating insurance exchanges, giving rebates to customers of insurance customers that spend too much on administrative costs, and the employer mandate. Even Republicans like all of them except the Medicaid expansion, increase in Medicare tax, employer mandate, and individual mandate.

Indeed, the only requirement of Obamacare most people didn’t like was the mandate for all people to join it.

5. Americans still don’t seem to know what Obamacare means for them

Given the discordance between Americans’ feelings on the individual parts of Obamacare and the law as a whole, it’s not that surprising that a striking 41% of Americans don’t feel they have enough information about the ACA, per the UConn/Hartford survey. Only 19% say they are very familiar with the law.

The individual provision questions strike the same chord. More than a third of people are unaware of the health insurance exchanges, subsidy assistance to individuals, or the Medicaid expansion. The latter two provisions of the law have actually seen a decrease in the percentage of people who knew these policies were in the bill, since it first passed. The only part of Obamacare that Americans seem to know really well is the individual mandate, which has also seen the largest percentage-point increase in awareness.

More worryingly, more people than not thought that Obamacare includes a public option, undocumented immigrant insurance, “death panels”, and cuts to Medicare. The Affordable Care Act contains none of these.

The fact is most of Obamacare is liked by the public. The issue is that the provision that is not liked is the best-known.

Conclusion: Americans are confused on Obamacare

There are lots of confused and confusing data here, and it’s difficult to say anything definitive about how Americans feel about the healthcare law signed by President Obama in 2010. As I found more than a year ago, they don’t like the law overall, even while they approve many of its measures.

Some Democrats may say that this points in their favor, but the same dynamic of a differential between the backing people will give for a broad proposition as opposed to their support for individual policies could be said to operate in the case of gun control. Thus two Colorado state senators were recently recalled over a gun control law whose individual provisions many said they liked.

Overall, Americans clearly don’t know enough about Obamacare. Of course, they know just as little about the Affordable Care Act – but to the extent that they are less hostile to a law that doesn’t bear President Obama’s name, it does appear that the embrace of the term “Obamacare” by Democrats and the White House was a tactic that has not worked out. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Chris Christie’s broad appeal sets up historic win in New Jersey. And then? | Harry J Enten

A Republican governor who can get re-elected in a blue state is something, but look at his coalition and start thinking about 2016

The fact that no one is talking about Chris Christie’s re-election campaign for New Jersey governor is about the most significant thing you can say about it. Many analysts, including this one, thought that Christie’s post-Sandy approval bump would ebb, so that we might get an at least somewhat competitive race with the Democratic candidate, Barbara Buono.

To put it simply, that hasn’t happened. Christie is ahead in the contest by a little more than 25pt, according to the latest Real Clear Politics average. That’s down from his 40pt lead at the beginning of the year, though it’s actually up slightly from the beginning of the month. Christie’s lead has held because his personality popularity has. He has a 68% favorable rating in the latest Stockton poll.

Christie’s dominating lead has pretty much silenced pundits like me: there just isn’t much to talk about in terms of the gubernatorial election. Christie’s going to win, and he’s going to win big.

Yet, it is worth recognizing how impressive Christie’s win looks like it’s going to be. Christie’s going to be the only Republican governor to win more than 50% of the vote in a state where President Obama won by at least 10pt over Mitt Romney. There’s a pretty good chance that Christie is going to be the only Republican governor east of Ohio and north of North Carolina by the end of next year.

Put another way, Christie is defying long-term trends. It wasn’t too long ago that the way a state went in the presidential election was not indicative of how a state would go in the gubernatorial election. That’s no longer the case. So, the fact that Christie is winning in a blue state now is as much of an anomaly as the Republican George Pataki winning in an equally blue New York state in 2002.

I’d go as far as to say that Christie’s win would be historic, if the 25pt margin holds. It’s the type of victory that people will remember because it’s so out of the normal historical range. How so?

President Obama won New Jersey by an average 13.9pt more than he won nationally over the past two elections. If the state was going to vote purely in line with its presidential vote, you’d expect Christie to lose by 13.9pt in a neutral year. Christie is running 39.2pt ahead of that pace.

I gathered the previous 156 non-recall gubernatorial elections since 2002 to see how this 39.2pt difference compares. Christie’s will be the best showing for a Republican candidate versus the presidential vote in the past five years. Of the prior 156 gubernatorial elections, and the two this year (assuming something crazy doesn’t happen in Virginia), since 2002, Christie’s “over-performance” will rank him seventh among Republican gubernatorial candidates.

The only candidates who beat him over the longer period are Jim Douglas in Vermont (a state that had been historically Republican) thrice, Connecticut’s Jodi Rell in 2006, Linda Lingle (who just lost by 20pt in a Senate race in 2012) in Hawaii, and Nevada’s Kenny Guinn in 2002. One thing all of these states had in common is that they didn’t have many black or Hispanic voters. Blacks and Hispanics tend to be less elastic in their patterns, so having more of them in a state makes a Republican’s job at running up the margin more difficult. At the time each of the six better performances occurred, the states where they happened had fewer blacks and/or Hispanics voting than the nation as a whole. That’s not the case in New Jersey.

That’s why it’s not surprising that Christie is only polling so well because he’s doing so well with blacks and Hispanics. He’s down only 19pt among blacks. He’s even with Hispanics. Of all the exit polls I could find over the past decade, no Republican gubernatorial candidate has scored better than that with blacks, and none outside of Florida has performed that well with Latinos.

Christie has shrunk the Democratic margin among blacks by about 70pt, against Obama’s share, in New Jersey. He’s done almost the same with Latinos, whom Obama carried in New Jersey by about 60pt – more than 15pt greater than he did nationwide. For a party looking to make inroads with minorities, Christie has done it.

Another thing none of these other over-performing Republicans had were presidential ambitions. Christie, of course, clearly does. It makes it more difficult to localize the race when this is the case, though Christie has.

Usually, a state’s voters don’t want their politicians to run for president. More Texans thought George W Bush shouldn’t run than should at this point in the 2000 cycle. More Tennesseans believed Al Gore shouldn’t run than should at this point in the 2004 cycle. More Illinoisans, Bay Staters, and New Yorkers thought Barack Obama, Mitt Romney, and Hillary Clinton respectively shouldn’t run than should at this point in the 2008 cycle.

Christie’s been so successful in his campaign that more New Jersey voters think he should run in 2016 than shouldn’t per a recent Quinnipiac poll. Christie has not only gone into a blue state as a Republican governor and convinced many he should be re-elected, but he’s done it even as they know he may run for president. More than that, they like the idea.

Some, especially on the right of the GOP, may say Christie has sold out conservative principles to get where he is in the campaign. The funny thing is that Republicans in his home state don’t agree. A recent survey had him winning a majority of them in a hypothetical 2016 matchup. It’s the only survey done so far this year in any state where a candidate has won the majority of the vote in a Republican primary.

When you look at the whole picture, you can see what I mean when I say Chris Christie’s re-election prospects are historic. He’s winning by a wider margin than nearly every other Republican gubernatorial candidate before him since 2002, compared to the state’s presidential leaning. He’s done so with a wide coalition and has gotten them to go along with his presidential ambitions.

It makes you think that Christie might win a certain election in 2016, too. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

The Public Policy Polling controversy | Harry J Enten

Public Policy Polling’s methodology is under attack. But since PPP is prolific, cheap and accurate, criticism is unlikely to dent it

Public Policy Polling (PPP) and Nate Cohn are involved in the Nerd Fight of 2013. For those who don’t know, Cohn has pointed out that PPP hadn’t fully revealed its true weighting methodology and were potentially weighting their results towards the polling averages. I strongly suggest you read Cohn’s arguments and PPP’s Tom Jensen’s rebuttals, if you haven’t already.

I’d weigh in on the methodological debate myself, though I think Cohn’s and Jensen’s views speak for themselves. The question that I want to answer here is whether there are any long-term ramifications for PPP. The answer, as far as I can tell, is that there won’t be. Why?

1. PPP’s sponsors are sticking by them

To conduct polling, you need money. PPP’s main political world patron is the liberal blog Daily Kos. Despite harangues from Cohn and Nate Silver, Kos’ leader, Markos Moulitsas, took to Twitter to defend PPP in the aftermath of Cohn’s first article. David Nir, who runs the indispensable Daily Kos elections blog, called Cohn’s article “a bit curious”.

My own conversations with Daily Kos writers reveal that they believe that Cohn is engaged in a vendetta. They believe PPP is accurate, and they think all pollsters engage in unusual weighting techniques. PPP is only being attacked because it releases a lot more of its internal data into the public sphere, which allows people to see how it comes to the numbers it does.

2. People see PPP as accurate

Quotes put together by Mark Blumenthal are telling. Accuracy is what people care about. Most people don’t care if you only interview tourists in SoHo to figure out who will win a US election, even if that is terrible methodology. Thus, Cohn’s arguments about how PPP is accurate, which are quite clever, just don’t matter to a lot of people.

What does matter is that Mark Blumenthal’s readjustment of Nate Silver’s 2010 pollster rankings to ensure no penalty for not following AAPOR disclosure requirements found that PPP ranked sixth out of 30 major pollsters. Silver’s analysis of the 2010 election put PPP in the middle of the pack for accuracy. Steve Singiser’s 2012 pollster overview had PPP fifth out of 17 very active pollsters.

I tend to think that accuracy is overrated, as my chart on special elections over the past ten years shows. I get the feeling though that most give credit to PPP just for trying, even if they are worse than the average pollster in special elections.

3. PPP is cheap

This is an extension of points one and two to a degree. When you’re cheap, people are willing to put up with a lot of stuff they may not otherwise. PPP, which uses recorded instead of live voices in their calls, costs about 4% the amount of most polls you see television networks sponsor.

The fact that PPP is cheap allows it to produce a lot of polls. Over 18% of the polls done last cycle were PPP surveys. They can poll races that few care about. They can poll Ryan Braun’s favorability in Wisconsin. In other words, they can flood the zone. They can get their brand out, so that even if their percentage of polls that get news stories is lower than other pollsters, the absolute number is higher.

In the three days since Cohn’s first article on PPP, the news archive site News Library shows that PPP has been cited 70 times by newspapers and television stations throughout the country. That’s actually more than in the same period during August.

4. Rasmussen still gets cited

Rasmussen Reports didn’t do a very good job in 2010 or 2012. Most people don’t think their methodology is exactly top-notch. Yet, they still produce polls, and people still talk about them. They got 41 hits on News Library in the same period PPP had 70. That’s despite producing fewer state data than PPP.

If an inaccurate pollster with a methodology that some deplore continues to make the news, then I just don’t see how one that many people view as accurate, even if dubious is going away anytime soon.

5. The aggregators who used to use PPP are still going to use it

Blumenthal’s HuffPollster, Drew Linzer, Real Clear Politics, Silver, and Sam Wang produced polling aggregates that became must-reads for many political junkies in 2012. Had anyone pledged not to use PPP in 2014 or 2016, it would be big news. It turns out that none of them is eliminating PPP from their datasets.

Linzer and Wang have actually defended PPP’s methods. Silver has been harshest about PPP, though PPP is still going into his models. Blumenthal has been more even-handed, but will continue to put PPP in the HuffPollster aggregates. Real Clear hasn’t said anything; but they continue to put PPP surveys into their database.

They all seem to think that PPP adds more than it detracts from their aggregates. Indeed, therein lies the rub. No one thinks PPP’s data is phony, even if people believe it to be massaged at times. Even PPP’s most ardent critics, like my friend Republican pollster Logan Dobson, will cite PPP; especially when it shows a result they like.


Many professional pollsters have long thought that Public Policy Polling’s methods were suspect. Nate Cohn’s recent expose confirms for many their suspicions. That said, in a world in which polling data can be hard to come by, there’s no real sign that usage of PPP’s data has or will slow, as long as their polls continue to be seen as accurate. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Facing government shutdown, Obama’s party can’t bank on 1996 mythology | Harry J Enten

Convention says Republicans got blamed for the last shutdowns, helping Clinton win in 1996. In reality, it was the economy, stupid

We are less than a week from a possible government shutdown, thanks to the inability of congressional Republicans and President Obama to reach a budget compromise. Much of the disagreement stems from the determination of some Republicans use the budget bill to defund Obamacare. Given the imminence of the threat, much reference has been made to the previous government shutdowns in 1995 and 1996.

For those who don’t remember, 1995-96 featured congressional Republicans led by Newt Gingrich taking on Democratic President Bill Clinton. The conventional wisdom now is that Clinton won the political battle over the shutdowns. Some have taken that a step further and believe Gingrich’s “defeat” cost Republicans in the 1996 election.

The former is definitely true. Republicans clearly took more blame for the shutdowns 17 years ago. Today, though, the “margin of blame” is 16pt smaller – with Americans surveyed only 3pt more likely to blame congressional Republicans than the president (the margin was 19pt in 1995-96). That suggests that Republicans are much in better shape now than they were then.

But even if the polling today did look like 1995-96, I would argue that this looming shutdown will offer nowhere such a clear win for Obama and the Democrats as it did for Clinton. The 1996 elections didn’t differ at all from what you’d expect – given the state of the economy and the outcomes of congressional elections in presidential years when there is split government.

Take a look at presidential and congressional approval from 1995-1996. This allows us to see what impact the budget had on the different parties’ overall stature because of the shutdowns.

You would have expected Congress to see a steep decline in 1995-1996 because of the budget shutdown, but that simply didn’t happen. Check out this graph from Charles Franklin, with the key points of budget shutdown included.

As now, congressional approval was already in the can back in 1995. There was perhaps a slight decline in congressional approval going into 1996, but it’s a point or two at most.

The same pattern held with President Clinton. Here’s a chart from the same time period, created by the Monkey Cage‘s John Sides.

Clinton’s approval rating just after the shutdowns was, if anything, slightly lower than before it. In other words, he really didn’t win much in terms of his standing. He didn’t gain ground in his approval rating, and didn’t lose less than Congress.

Clinton’s major increase in presidential approval occurred in the months after the shutdown. Those ratings corresponded very well with a major increase, also, in congressional approval. That’s not surprising, given that both approval ratings tend to move in unison with one another. Congressional and presidential approval in this case moved up – because the economy was improving.

Perhaps counterintuitively, both Clinton and congressional Republicans actually saw their standing improve in the ballot test for the November 1996 elections. Clinton opened up about a 6pt edge on Republican Bob Dole in the immediate aftermath of the shutdown, when he had been tied prior to it. Congressional Republicans closed a 5pt deficit, to a 1pt deficit, in the national House vote ballot.

Both of those margins pretty much held through the election. Congressional Republicans would gain a little bit on congressional Democrats. Both Clinton and Dole ran away from Reform party candidate Ross Perot (Clinton slightly more so). It looks as though most people did not determine their vote based on their view of the government shutdown.

Indeed, only 10% of Americans said the government shutdown was their greatest reservation about Republicans, following the 1996 vote, per a post-election poll. The exit polls didn’t even ask about it.

There just isn’t much sign that 1996 differed from what you’d expect, given the fundamentals. Clinton won the national vote by a little less than 9pt over Dole. One would think that if the shutdown had really hurt Republicans over the long term, then Clinton would have done far better than the economy would suggest. That simply didn’t happen.

Of the seven economic fundamental models displayed by Brendan Nyhan, two underestimated Clinton’s vote, three overestimated it, and two pretty much nailed it. That’s what you would expect to happen if there were no big event that overrode the 1996 election.

The same holds for the House. House Democrats gained two seats over their 1994 showing, but that’s well within expectations. The result was less of a loss than Republicans went on to suffer in 2008 or 2012, or then Democrats sustained in 1992, for instance. It’s equal to the loss Republicans took in 2000. Only once since 1952 has the majority party gained more than three seats in a presidential election year, when the other party controlled the White House.

In short, there’s just no clear evidence that House Republicans suffered, even if they were largely blamed for the shutdown.

In fact, Senate Republicans actually picked up two seats in 1996. Some might say that Democrats would gladly settle for a two-seat Republican gain in 2014. While that’s true, you have to know the baseline going into the 1996 elections: Republicans controlled 56% of the class up for re-election; they ended up winning 62% of the class thanks to wins in the south.

The reverse will be true in 2014. Democrats will control 60% of the seats up for election in 2014. Republicans have a lot more opportunity to pick up seats. They are playing offense mostly in the south, as they were in 1996. If Republicans were to win 62% of the seats in play in 2014, they’d pick up eight seats.

Now, I don’t think Republicans will gain eight seats in 2014. To me, one would be wise not to project too much correlation between the 1995-96 shutdown and a possible one in 2013. This is a midterm election, not a presidential election year. Congressional and presidential approvals are both in worse shape now than they were then. And polling puts Obama in worse shape than Clinton was at this point, as he faces a possible shutdown.

For those who look to the 1995-96 shutdown as a sign that it will have major electoral implications, look again. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

The approval rating of Congress is up. Yes, you read that right | Harry J Enten

Their stand against a war in Syria has pushed their approval up to 20%, which could make a difference in the midterm election

Trying to find an American who likes the job Congress is doing is like trying to find a parking space in New York City. A few months ago Congress’ approval rating was at historic lows. Today, it still stinks, but there are signs that, as President Obama flounders, Congress may be making a slight and potentially important comeback.

The HuffPollster aggregate has Congress’s approval rating up to 20%. It’s not just one outlier poll, either. CBS/NY-Times, CNN, Gallup and Reason-Rupe have Congress’ approval at its highest this year.

To be sure, 20% is quite low. It is, however, higher than congressional approval going into the 2012 election. Just 12% approved of Congress per the HuffPollster very sensitive trend among adults aggregate, which is used for consistency and to catch any last minute movement in the polls.

The reason for the increase could be any number of reasons, but the sudden increase is most likely from Syria. Congress has expressed a lot of skepticism over a possible war. This stance is in line with the American people, while President Obama’s initial position on attacking Syria was not. While Obama has changed his stance, a majority of Americans disapprove of his handling of the situation.

There’s no way to know if this is a temporary blip or the start of a new trend, though it certainly underestimates how well individual congress members are doing.

When we look at approval of a person’s own representative, polls show 30pt or higher approval than that of the Congress as a whole. Democrats generally like their Democratic congress member, and Republicans the same for their own. Republicans and Democrats may not even like their own caucus, yet still will vote for their own congressman. It’s why re-election rates for congress members are so high.

That’s not to say the 8pt improvement in overall congressional approval isn’t important. It definitely is.

David Jones and Monika McDermott found that, once controlling for a host of factors including the president’s approval and the state of the economy, a 10pt decrease in congressional approval costs majority party incumbents about 4pt in their re-election margin, while helping minority incumbents by about a point. Overall, a 10pt drop in congressional approval would lead on average to a 17 seat loss for the majority party.

Usually, this rise in congressional approval would be mitigated by a rise in the president’s approval. The president’s and Congress’ approval ratings tend to run in concert with each other, as demonstrated by Jim Stimson. That’s why it’s odd to see major gains for either party in Congress when the two branches are controlled by different parties (even if the US senate has a slim Democratic majority).

When one rises and the other falls, it can lead to historical anomalies. Consider 2012: President Obama’s approval was decent, even if not great. Congressional approval was at its all time low of 12%. And while a host of variables such as gerrymandering and urban packing kept Republican seat losses to a minimum, they lost the national House vote by over a point. It was the worst loss for the majority party in the national vote for a presidential year House election since 1948.

For 2014, it may be the opposite. Congress’s approval seems to be rising, while the president’s is dropping. Presidential approval is, in my opinion, more instrumental in determining the results of congressional elections, so a low Obama approval would be worse for the presidential party than a dropping congressional approval. But with congressional and presidential approval in concert against the Democrats, it could provide quite the punch back in the other direction from 2012.

Of course, we don’t know if the current congressional approval rating rise will hold. It could sink back into the deep abyss by 2014. If it does stay up, it doesn’t guarantee anything. Again, 20% approval isn’t great.

Still the rise in Congress’s approval rating, even if it is only to 20%, does in conjunction with the usual movement against the president’s party in midterms make Republican gains more likely. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

President Obama’s flailing approval ratings hurt his party | Harry J Enten

Obama’s approval rating is unlikely to rise by the 2014 midterm or 2016 presidential election, spelling danger for Democrats

President Obama’s standing with the American people is flagging. The Real Clear Politics average at the time of this writing has his net approval rating at -7.3pt. But does Obama’s approval rating matter now given that he never has to face re-election? And if it does matter, can he recover?

The answers are that it does matter – Obama’s approval rating can greatly affect the 2014 midterm elections and, to a lesser extent, the 2016 presidential election – and the historical odds of it recovering much seem to be slim.

In midterms, electorates often take out their frustration with the president on the their party’s congressional members. A poor presidential approval rating will only add to that frustration. A president likely needs an approval rating in the mid 60s, like Bill Clinton in 1998 and George W Bush in 2002, to avoid the curse of “midterm loss”.

In every non-wartime midterm election since 1938, simply knowing how many seats the president’s party controlled and the president’s approval rating goes a long way in determining how the midterm is going to shake out. Not counting 1974, because Richard Nixon resigned and Gerald Ford took his place, more than 75% of the variation between the seats won in the House by the president’s party in the midterm is explained by the two aforementioned variables.

If the president’s approval rating were to hold, this very simple regression finds the Democrats would lose 30 seats in the House. Now, no one I’ve spoken to thinks that the president’s party is going to lose that many seats. The margin of error on this regression is large enough that no seat loss is possible with the president’s net approval rating of -7.5pt. Still, the chances of a major House loss for the president’s party are perhaps better than one might think, if the president’s position holds.

The effect of the president’s poor standing extends into the Senate elections. It’s much more difficult to model Senate elections over the long-term because not every senator is up for re-election in a given year. That said, the Senate is likely to be decided in states where President Obama lost in 2012. Democratic Senate candidates are likely going to try very hard to localize their races.

One race that they won’t be able to localize is the 2016 presidential election. To be sure, the incumbent president’s approval rating matters less when he is not running for re-election. Clinton’s high approval rating didn’t save Al Gore, while Hubert Humphrey nearly won even as Lyndon Johnson slumped. The factor that will matter most will be economic growth during 2016.

However, the president’s approval plays a role in the election to find his successor. Once we control for the economy, every 5pt increase in a president’s net approval rating increases his party’s candidate’s margin by 1pt in the presidential election per Drew Linzer. An election his party might have won by 1pt had the incumbent president had a +5pt net approval rating becomes an election the incumbent party loses by 1pt with a -5pt rating.

Of course, we’re still a long way from the midterm election and an even longer way from the presidential election in 2016. But the chances that the president can pick up ground before the midterm and even the next presidential election are not as good as one might think.
You may remember that Obama’s approval was in a similar position after the debt ceiling crisis in 2011, and he recovered. The difference is that we were heading into a president’s re-election year, when presidents regularly see their approval rating rise and fall. Going into a midterm or an election when the president does not run for re-election is a different story.

The president’s approval rating has never increased by more than 7pt from this point after re-election until the midterm election.

The greatest increase in a second term was 7pt for Bill Clinton in from this point until the 1998 midterm. The only other increase at all was 4pt for Ronald Reagan in 1986. Both of these years featured much stronger economies than we have now. All other presidents saw a decline in their approval rating. The average previously re-elected president loses about 10pt on their net approval from now until the midterm.

When we include presidents who were in their first term, 14 out of 16 saw a decline in their approval rating from now until the midterm. It is, in other words, unlikely that the president sees an increase in his approval. Add on the fact that most polls now are of adults and the likely electorate is probably going to lean more Republican, and it doesn’t look good for the Democrats.

2016 will almost certainly feature better turnout among the Democratic base of minority and youth voters than 2014. The issue is that of the six presidents who had won re-election and weren’t running in the next election, none have seen more than a 2pt improvement in their net approval rating from this point until the next presidential election. Clinton saw a 1pt increase and Eisenhower a 2pt increase. Neither of those increases would put Obama in positive net approval territory.

Now all that said, it’s certainly possible that President Obama’s approval rating will rebound. There have been great deviations in a president’s approval from this point in his presidency forward. It’s just that almost all movement has been in the negative direction. Maybe this time will be different.

If historical patterns hold, however, President Obama’s approval ratings will matter in the upcoming midterm and next presidential election, and they will likely be an albatross around the neck of the Democratic party. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

Bill de Blasio’s diverse coalition could clinch contest for New York mayor | Harry J Enten

Given New York’s ethnically divided politics, De Blasio leads the Democratic mayoral primary with historically broad support

Long-time followers of New York City politics know that the vote in city-wide elections usually breaks down along racial and ethnic lines. In this year’s comptroller race, for instance, Eliot Spitzer is winning black voters and losing white voters by a wide margin. The mayoral race, however, is a far different scenario.

The latest trio of mayoral polls puts Bill de Blasio just south of the 40% needed to avoid a runoff with likely second-place finisher Bill Thompson, who is 15pt to 20pt behind. De Blasio has made a late charge, but what’s truly surprising to me is how De Blasio is getting to 40%.

De Blasio is trying to hold together the most diverse coalition in modern history to win the Democratic primary for mayor. In the latest Quinnipiac poll (which is far from perfect, though will have to do), De Blasio is at 44% among Hispanic voters, 40% among white voters, and 37% among black voters. This is despite Thompson being black. So, given the margin of error on subsamples, we could say that De Blasio looks to be scoring equally well among all the main racial and/or ethnic groups.

Compare this racial coalition to that of the last white Democrat to win a mayoral primary, Mark Green in 2001. Green took 83% of whites, 29% of blacks, and only 16% of Latinos on his way to winning a runoff against Freddy Ferrer, who is Hispanic, 51% to 49%. The same racial divisions were evident to some degree in 2005 when Ferrer beat Anthony Weiner, in 1997, when Ruth Messinger defeated Al Sharpton, and in 1989, when David Dinkins topped Ed Koch.

Put another way, De Blasio is trying to achieve something unprecedented. A non-Jewish white candidate has not finished first in the Democratic New York mayoral primary in 44 years. De Blasio may be many things, but he’s not black or Jewish. For those us who use history as a guide, the lesson is that the only tradition that endures is the tradition of change.

Many have ascribed De Blasio’s winning coalition to the fact that he’s become the anti-Mike Bloomberg. De Blasio has run to the left in this Democratic primary – seemingly a smart move in a field crowded with competitive centrists. Yet, the data indicate that De Blasio’s edge is not necessarily down to being an anti-Bloomberg.

De Blasio is doing about as well with Bloomberg backers as he is with those who dislike the departing mayor. In a Public Policy Polling survey completed on Sunday night, De Blasio is at 37% among those who approve of Bloomberg and at 39% with those who disapprove. That matches a Marist poll conducted just a few days earlier.

My own guess is De Blasio has masterfully parlayed a mixture of biography and political positioning into broad appeal. De Blasio’s a white Brooklynite who promises to pay attention to the outer boroughs; this allows him to be competitive with moderate and conservative outer borough whites. He’s the most liberal of the major contenders: hence his backing from white progressives. His stances on policing, and astute ads featuring his biracial son, allow him to bring minorities into his coalition.

That is a team of voters who have brought De Blasio to the verge of winning the primary. The question is whether or not he’ll actually get to the magic 40%, to win outright in the first round. With De Blasio at 36%, 38%, and 39% in recently released polls, and with somewhere between 8% and 10% of voters undecided, it seems quite possible that he’ll make it – should the undecided vote break his way.

I still urge caution. In the last five competitive mayoral primaries, one of the two leading contenders received what they polled in pre-election polls but got no more. The other leading contender picked up the vast majority of undecideds. Normally, it’s the leading candidate of color who picks up the most support. The complicating factor is that Thompson is trailing among blacks by 10pt to 15pt, depending on the survey. So, who knows if history will hold?

Adding to the confusion is potential under-the-radar momentum for Bill Thompson. Thompson’s 25% in the Quinnipiac poll and 20% in the Marist survey are his two highest percentages in those surveys to date. He’s picked up 5pt in the last week per Quinnipiac, while De Blasio has dropped 4pt from 43% to 39%. If that is real momentum (and I don’t know if it is), then it could lead to a much closer election night than most predict.

Finally, there have been two instances in the past 16 years where a candidate fell short of 40% on election night yet reached it once absentee votes were counted. It took weeks before Messinger was declared to be over 40% in 1997. By then, they even held a runoff debate between Messinger and Sharpton!

De Blasio likely has got be more than 0.3pt on either side of 40.0% on election night for us to be confident that absentees would clinch it. The fact that De Blasio’s best numbers have come in the final weeks suggests that absentees may, in fact, be less likely to go for him than ballots cast on election day. That means that, unlike those who straddled 40% in years past, De Blasio is more likely to fall back, than spring forward in post-election day counts – if it’s on a knife edge.

Of course, none of this will matter if Bill de Blasio reaches 41%. If he does, it will cap a remarkable two months for the public advocate. If he doesn’t, Thompson’s likely to give him a good fight in the runoff, regardless of early polling. His favorables are as good as De Blasio’s, and Thompson has a tendency to close well.

But enough with my analysis, let’s hear what the voters have to say. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

It may be Labor Day weekend, but union power is waning | Harry J Enten

Union membership is decreasing in America and so is the political influence of unions in elections

It’s Labor Day weekend. For most, the weekend marks the end of the summer. A last hurrah before football starts and the kids go back to school. You’ll notice I haven’t mentioned anything about “labor”. That’s cause while labor’s political influence is still poignant, it’s clearly not the powerhouse it once was.

Why do I say that?

1. Union household percentage is dropping

Politicians would fear labor if they made up a large percentage of the electorate. The 2012 exit poll put labor at 18% of the electorate. That’s not small, especially considering that union households went for Obama by 18pt. Indeed, the gap between union and non-union households is about the same as the gender gap we always here about.

The issue is that 18% is a far cry from the 34% unions made up in the 1976 election.

The percentage unions have made up in exit polls have dropped by 8pt since 2000 alone. 2000 marked a momentary spike that soon abated and the gradual decline continued. It’s been over a point drop per election on the whole. Year alone predicts 88% of the variation in union membership from election to election – meaning that there’s a powerful relationship between years gone by and the decline of labor as a percentage of the electorate.

As union households make up less of the electorate, there’s less of a reason that politicians need to adhere to their arguments. Republicans have to make less of an outreach, and Democrats don’t need to pander nearly as much. The populist appeal of Al Gore in 2000 may seem like a distant memory for future generations.

2. Union backing can be a bad thing … in New York City

No one expects that in a Republican state like Mississippi that being backed by a union is a good thing. Few people belong to unions in that south. One would think, however, that union backing would be a good thing where unions grow off the trees. In the largest city in the state with the most union members, labor should dominate.

That simply isn’t the case in New York City. The (Democratic) candidate backed by labor for New York City mayor has lost in every election since 1993. Part of that is no doubt because of crime and not labor negotiations. A good sized portion of it is, however, people who don’t belong to unions are overwhelmingly fearful of too much union influence and overly generous city contracts.

Republican (or whatever Mike Bloomberg is calling himself these days) mayors won’t bend, and people love it. While Democrat Bill Thompson won union households by 20pt in 2009, non-unions voted for Bloomberg by about 15pt.

In 2013, all the major newspapers in New York City passed over Thompson for the Democratic nomination. They didn’t do so because they thought he was unqualified. They did so because they didn’t like that he was backed by the United Federation of Teachers (UFT). Did I mention the Democratic candidate backed by the UFT hasn’t won a primary in a generation?

If unions can’t win in New York, where can they win?

3. Wisconsin Governor Scott Walker said “screw you” and won

The Wisconsin budget dispute of 2011 and subsequent recall of 2012 was perhaps the biggest blow to labor in a long while. Republican Governor Scott Walker led a Republican controlled legislature to make major changes to Wisconsin law. They limited collective bargaining of all non-emergency personnel to close a budget shortfall*.

Democrats and unions went crazy. They protested and staged sit-ins. Walker and the Republicans didn’t care. The Democrats and unions went to court and lost. They decided to force Walker into a recall election.

Walker not only won the recall in 2012, but he won it rather easily. He expanded his 6pt win in the prior election to 7pt in the recall. This was despite a very turnout among union households. They made up 32% of the recall electorate versus 27% in the prior gubernatorial election. In other words, a pretty large chunk. Walker was able to overcome this larger turnout by expanding his margin among non-unions households from 13pt to 22pt.

And make no mistake, the result was all about collective bargaining. 52% of voters approved of limiting collecting bargaining for government workers and of Walker’s handling of it. Walker won pretty much all of these voters, while losing the rest.


The nation will celebrate labor this weekend, yet labor has to be worried politically. They are still a force in American politics, though their power is steadily waning.

*Note that this was a major difference between changes made in Wisconsin and the less successful efforts in Ohio. The lesson is people are less likely to stand for cuts to firefighters and policemen. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds

The Democratic advantage in electoral college may be waning | Harry J Enten

President Obama and Hillary Clinton are polling much worse in Colorado and Iowa than nationally

President Obama won the 2012 presidential election by 3.9pt. He took the 2008 election by 7.3pt. Yet, these margins underestimated his true strength. United States presidential elections are decided in the electoral college. Obama gained an extra advantage there, but history and new polling suggests that the next Democratic nominee might not be able to hold it.

What do I mean by “extra advantage”? Obama won Colorado by a little less than 5.4pt in 2012, which is 1.5pt greater than his nationwide margin. If one had given every state Obama won by less than his margin in Colorado to Mitt Romney, Obama still would have won the electoral college by a 272-266 vote margin. That is, he won the “tipping point” state by a wider margin than the nationwide vote. Romney could have the national vote by a point, and he still would lost the election.

This could be a major problem for Republicans in upcoming elections. It would mean that they’d start off 1.5pt in the hole. The question is whether it will hold.

One could reasonably expect it to given that most states move the same percentage point to the right or left as the nation as a whole. So that as Obama lost 3.4pt off his nationwide margin in 2012, he lost, for example, 3.6pt off his margin in Colorado. Once you control for home state effect (eg John McCain did better in Arizona than a Republican from a different state would), over 97% of the differences in Obama’s margin between states in 2012 is predicted by just knowing his margin between states in 2008.

The issue is that what is reasonable isn’t always what is true. Check out this chart from Nicholas Miller. When the black bar ends above the middle line, the Democrats have a disadvantage in the electoral college relative to the national vote. When the black bar ends below the middle line, the Democrats have an advantage relative to the national vote.

It goes back and forth over time. If anything Republicans tend to do better in the electoral college than the national vote over the long haul. Over the past eight elections, however, Republicans have done better four times, while Democrats have done better four times (note that the chart was produced before the 2012 election).

How does current polling fit into this? Now, no one can expect that polling in 2013 to be too predictive of what will happen in 2016. President Obama’s approval rating may rise or fall. Hillary Clinton’s polling may rise or (most likely) fall to some degree. That said, there’s likely a higher correlation between how states rise and fall relative to the national rise and fall than the overall swing. So if one state is showing worse (or better) numbers for Obama or Clinton now compared to nationally, it has a decent chance of holding.

That’s why data from Colorado and Iowa could signal the end of the Democratic advantage in the electoral college. Colorado was, as mentioned, a tipping state with 9 electoral votes, while Iowa with 6 electoral votes has been slightly more Democratic than the nation as a whole.

Quinnipiac has conducted polls nationally and in both states recently. They looked at both Obama’s job approval rating, and how Hillary Clinton is faring against potential GOP challenger Chris Christie. The polling from both of those states say the Democratic grip on them may be waning.

We see that President Obama’s net approval rating is not great in the latest national Quinnipiac survey, but it’s over 10pt worse in both Colorado and Iowa. We know that presidential approval rating does at least a somewhat decent job at predicting presidential elections, and in this case, it seems to be doing so for 2016.

Christie is running well ahead of where he is nationally in both Iowa and Colorado. Instead of trailing Clinton, he’s either tied or in the lead. Again, the margins will change, but the point here is relative to the national vote, he’s doing better than he is nationally.

Some might point out that polling in Colorado has a tendency to undershoot the Democratic candidate. That’s true, though not to the extent that even if there were a bias here that it would erase the Republican lean compared to the national as a whole. We also have other polls from these states suggesting that it’s not a one time phenomenon. More than that, these numbers make sense.

Iowa is chock full of whites without a college degree – the group that has abandoned Obama the most over the past few months. This, as I’ve noted, is part of a longer term trend. The state’s voters are quite elastic in their voting choices per Nate Silver’s index, which means that they would be more likely to shift harder to the right (or left) if there is a shift nationally.

Colorado doesn’t have as many whites without a college degree, yet it’s not the state that many think it is. It’s actually whiter than the nation as a whole by 8pt, and its voters are more elastic. The voters there seem perturbed over new gun control laws. In addition, Clinton has always polled poorly in the west harkening back to her to 2008 run.

Thus, it seems that this polling data may very well be correct. If it is, then the historical trend will continue. The Democrats edge in the electoral college versus the national (popular) vote may not sustain itself through 2016. © 2013 Guardian News and Media Limited or its affiliated companies. All rights reserved. | Use of this content is subject to our Terms & Conditions | More Feeds