Posts Tagged ‘ron paul’
Per commenter FreeDem’s request, here are some other ways to interpret Ron Paul’s primary performances in each state, building on the statistical models estimated here.
The first exercise is to simulate what Paul’s percentage of the vote would have been in each state had they all voted at the same time with the same “institutions” (candidates on the ballot, primary vs. caucus, closed vs. open, and so on). For this exercise, I assumed that all calendar and institutional characteristics were at the median for the nation as a whole. Thus, I assume a primary not a caucus, a closed primary rather than an open or semi-open one, three other candidates on the ballot, a ballot date after Mitt Romney had built substantial momentum with his Michigan victory on Feb. 28 — but before his virtual mathematical clinching on Apr. 2, and no home state advantage for any candidate.
Under these conditions, Paul would have obtained 13.9% of the vote in the average state. Here is how he would be forecast to do in each state, in alphabetical order:
|District of Columbia||12.17%|
The rank order of states here is of course exactly the same as in the table of residuals from my earlier post, but the presentation of the numbers is perhaps more intuitive.
The next analysis looks at why Paul improved from 2008 so much in some states and little if any in others. In my last post I found that Paul’s electoral performance in 2012 is predicted by state opinion ideology, with more left-wing states giving more support to Paul. However, 2008 LP support is not at all correlated with Paul support in 2012. I inferred that Paul received a great deal of support from voters who would otherwise go Democratic and from liberal Republicans and independents, particularly since the association between ideology and Paul support strengthened conditional on an open primary.
However, I did not control for how “pro-Ron Paul” a state was in 2008. FreeDem reports seeing some odd state-level differences between Paul’s 2008 success and 2012 success when eyeballing the data. Using my analysis of the 2008 returns, I now try formally testing whether Ron Paul’s support in 2008 was correlated with his support in 2012. I regress the log of Paul support in 2012 on the institutional variables, Democratic-Green vote share in 2008, and the log of institution-corrected Paul vote share in 2008. Here are the results:
. regress lnrp caucus12 open12 closed12 ocb clinch home demgr08 lnrp08 Number of obs = 49 F( 8, 40) = 9.67 Prob > F = 0.0000 R-squared = 0.6592 Adj R-squared = 0.5911 Root MSE = .31484 ------------------------------------------------------------------------------ lnrp | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- caucus12 | .5744784 .1465334 3.92 0.000 .2783233 .8706335 open12 | .0946405 .1444686 0.66 0.516 -.1973414 .3866224 closed12 | -.1744016 .1367005 -1.28 0.209 -.4506836 .1018805 ocb | -.1192173 .0418184 -2.85 0.007 -.2037355 -.0346991 clinch | -.1881422 .1700441 -1.11 0.275 -.5318141 .1555297 home | .0832717 .2090593 0.40 0.693 -.339253 .5057963 demgr08 | .0091053 .004799 1.90 0.065 -.0005938 .0188044 lnrp08 | .4681954 .1376803 3.40 0.002 .1899331 .7464578 _cons | -.5945686 .6079726 -0.98 0.334 -1.823327 .6341899 ------------------------------------------------------------------------------
Now I do find evidence that Paul’s support in 2012 depended on the strength of the Paul constituency in 2008, controlling for everything else. Although the statistical significance is strong, the substantive import is in a sense strikingly low. Since the dependent and independent variable are both in logs, we can interpret the coefficient on Ron Paul support in ’08 — 0.468 — as an elasticity. In other words, for every 1% more votes Paul got in 2008, he got just 0.5% more votes in 2012. That tells us that while Ron Paul’s base turned out again for him in 2012, his additional success was probably not related to picking up a bunch of semi-libertarian marginal voters who might have gone with him had his campaign been stronger in 2008. Instead, he reached a whole new constituency. Paul more than doubled his vote share between 2008 and 2012. That mostly did not come from mobilizing more libertarians. He got the libertarians to vote for him again, but most of his new votes came from people we would not normally think of as libertarians.
Who exactly were these new Paul voters? We can find out more, I think, by looking at individual states where Paul improved or did not improve between elections. Here are some states where Paul did badly in both years (adjusted vote shares from 2008 and 2012 in parentheses): Alabama (2.9%/7.4%), Utah (3.0%/5.4%), Mississippi (1.9%/6.5%), Arizona (4.4%/8.7%), and Illinois (5.4%/9.7%). These are states without many libertarians and also without many social and foreign policy moderates and liberals. We could throw in West Virginia (2.9%/12.5%), Florida (3.8%/10.1%), New Jersey (4.9%/11.8%), Delaware (4.2%/12.5%), and California (6.3%/12.2%) as also being of this type.
Then we have states like Idaho (10.8%/11.5%), D.C. (7.8%/12.2%), Montana (7.8%/14.8%), North Dakota (7.1%/15.5%), South Dakota (8.6%/15.4%), and Washington (8.2%/15.8%), where Paul did well in 2008 but did not improve much in 2012. He seemed to hit a ceiling in places like these. My hypothesis is that these states have a lot of libertarians but not many social and foreign policy moderates who are willing to take a Republican ballot. In all the states I just mentioned except perhaps D.C., social conservatives tend to dominate the state Republican party. Washington may be a partial exception, but only a partial one. Washington certainly does not have the tradition of moderate Republicanism that, say, Massachusetts and Connecticut do.
Then there are states where Paul underperformed in 2008 and then improved significantly in 2012: Connecticut (4%/15.9%), Indiana (4%/15.9%), Iowa (3.8%/19.7%), Kentucky (3.8%/14.8%), Massachusetts (3%/17.7%), Oklahoma (4%/16.5%), Rhode Island (6.1%/27.1%), South Carolina (4.3%/16.3%), and Vermont (6.4%/31.4%). New York (6.7%/17.3%) is a state where Paul went from slightly above average to well above. What at least some of these states have in common is a lot of former Republicans and independents who are moderate to liberal on social and foreign policy issues. (I can’t explain Oklahoma, but South Carolina’s improvement might have to do something with a bounce from New Hampshire, and Kentucky with Rand Paul’s prominence.) In Vermont, especially, I will bet my bottom dollar that most Ron Paul voters were centrists or even progressives ticked off at wars, banker bailouts, and civil liberty infringements, but not necessarily signing onto Paul’s whole economic program. His outstanding improvement there may also have something to do with spillover from New Hampshire.
Finally, there are only a few states where Paul did very well in both years: Maine (6.9%/19.5%), Michigan (7.2%/16.8%), New Hampshire (11.2%/31.7%), and Pennsylvania (7.8%/26.4%). These seem like states with both quite a few libertarians and lots of moderates and progressives willing to take a Republican ballot. While economic insecurity might be another hypothesis for the increase in Paul support, that wouldn’t explain why Paul did so poorly in Nevada (10.1% adjusted), New Jersey (11.8% adjusted), or Georgia (12.2% adjusted), all of which have suffered more than the national average from the bursting of the housing bubble.
After the 2008 primary season, I analyzed Ron Paul’s performance in each state to see how institutional factors such as caucus and primary form affected his electoral success. This exercise turned out to be useful for estimating the size of the pro-liberty electorate in each state. In this post, I do the same with the 2012 results.
The dependent variable in this analysis is the percentage of the vote obtained in each state’s statewide primary or caucus. If a state held both a primary and a caucus or convention, I used the primary results. The independent variables are as follows. (All variables were taken from Dave Leip’s Atlas of U.S. Presidential Elections.)
First, I use a dummy variable for whether the state had a caucus or convention versus a primary. As an outsider candidate with a committed band of activist supporters, Paul tended to do much better in caucuses than primaries.
Next, I also included dummy variables for who is eligible to vote: a dummy for an open caucus/primary, in which all voters are allowed to vote, and a dummy for a closed caucus/primary, in which only registered Republicans are allowed to vote. The excluded category consists of “modified-open” elections, in which independents and Republicans are allowed to vote. Since Paul was more popular among self-identified independents than among self-identified Republicans, it stands to reason that he would do best in open primaries and worst in closed primaries.
I also tried variables for the number of other candidates running active races and appearing on the ballot. In this regard, I counted only “significant” candidates that might have had a chance of winning, that is, the following six (other than Paul): Mitt Romney, Rick Santorum, Newt Gingrich, Rick Perry, Jon Huntsman, and Michele Bachmann. Bachmann dropped out of the race after Iowa and Perry and Huntsman after New Hampshire. Thereafter, Santorum withdrew only after the April 10 primaries and Gingrich held on until May 2. However, in many states – almost entirely primary states – candidates that had withdrawn remained on the ballot and often received significant support. We should expect both of these variables to have a negative influence on Paul’s vote share (for instance, in Virginia, where only Romney and Paul were on the ballot, Paul benefited from tactical voting by supporters of other candidates and received 40%), but in fact the number of candidates in the race had a statistically significant, positive association with Paul’s support (candidates on the ballot has the expected negative association, as shown below). This result is theoretically implausible and apparently simply captures Paul’s otherwise unexplained success in Iowa and New Hampshire. Therefore, in the estimations reported here, I have included solely the variable for number of non-Paul candidates on the ballot.
Next, I included a three-point variable for the extent to which Mitt Romney had clinched the Republican nomination when the primary in question was held. After February 28, it became very unlikely that Romney would lose, as Nate Silver noted here. After April 3, it became a near mathematical certainty that Romney would win. Therefore, the “clinching” variable takes on values of 0.5 for contests held after February 28 through April 3 and values of 1.0 for contests held after April 3.
Finally, I include a variable intended to capture home-state effects for active candidates. It is coded zero for all states except Massachusetts, Georgia, and Texas. For the former two, home states of Romney and Gingrich, respectively, it is coded -1, and for the last, Paul’s home state, it is coded 1. (Santorum dropped out of the race before the Pennsylvania vote, for which reason it is coded zero.)
The dependent variable is rather skewed, as the following histogram shows:
As a result, statistical analysis with this variable revealed the presence of heteroskedasticity, which violates the assumptions necessary for Ordinary Least Squares regression. Therefore, I transformed the variable using the natural log, which gave it a roughly normal distribution and eliminated heteroskedasticity in the regression analysis:
The results of the regression analysis are reported below the fold:
My original forecasts for Ron Paul’s primary performances are here. Those forecasts were based simply on the Iowa result, so it was quite possible that there would substantial error, and indeed there has been. Paul significantly overperformed his forecast in New Hampshire and South Carolina, the forecast was dead on in Florida, and then Paul underperformed significantly in Nevada. In yesterday’s elections, Paul did significantly worse than expected in Colorado, slightly worse than expected in Minnesota, and slightly better than expected in Missouri. In general, he seems to be doing worse than expected since Florida.
Why is that? It could be that my forecast model was an unbiased model at the time, but that circumstances have changed unfavorably for Paul’s candidacy. Certainly, recent good economic news probably doesn’t help an antiestablishment candidate like Paul. Perhaps his poor Florida performance, although it should have been anticipated, demoralized some of his supporters. On the other hand, my forecast model could have been wrong, particularly in assuming that Paul’s vote shares would continue to feature overdispersion. It’s possible that with a broadening voter base, Paul’s caucus advantage has declined. Thus, Paul should improve on his 2008 performances everywhere, but not in a manner proportionate to his 2008 performances: there will be some apparent regression to the mean.
To see how Paul’s 2012 performances are stacking up against his 2008 performances, I ran a regression on the states with results so far. First, I regressed 2012 performance against 2008 performance linearly. Here are the results:
Call: lm(formula = vote12 ~ vote08) Residuals: Min 1Q Median 3Q Max -5.0465 -3.8563 0.8463 2.0316 6.8799 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 6.9290 3.4156 2.029 0.0888 . vote08 1.1807 0.3632 3.251 0.0175 * --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 4.444 on 6 degrees of freedom Multiple R-squared: 0.6378, Adjusted R-squared: 0.5775 F-statistic: 10.57 on 1 and 6 DF, p-value: 0.01745
These results suggest that for every 1% in 2008 vote share in a state, Paul is now receiving 1.2% in 2012, in addition to a base of 6.9% everywhere – so getting 5% in a 2008 primary would be associated with a forecast of about 12.9% in 2012. With these eight data points, the simple model explains 63.8% of the variance in 2012 performance.
Next, I turn to a log-linear model, which would be more appropriate if Paul’s performances continue to experience overdispersion. Here are the results:
Call: lm(formula = lnvote12 ~ lnvote08) Residuals: Min 1Q Median 3Q Max -0.36483 -0.22250 0.06901 0.15519 0.35117 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 1.5376 0.3628 4.238 0.00545 ** lnvote08 0.6086 0.1768 3.442 0.01377 * --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.2805 on 6 degrees of freedom Multiple R-squared: 0.6638, Adjusted R-squared: 0.6078 F-statistic: 11.85 on 1 and 6 DF, p-value: 0.01377
Although the coefficient estimate is not so easily interpreted, this model actually does a slightly better job than the simple linear model. (I also test various transformations of the independent variable to get at other nonlinearities, but none of those models improves significantly over this one.) So I use these estimates to get new forecasts of the remaining contests. Here they are:
|District of Columbia||3-Apr||16.6%|
In general, these new forecasts are lower for Paul in his best states and higher in his worst states. (So yes, his support is less overdispersed this time around, suggesting that his new support is less enthusiastic than his core support – not really surprising.) With the new forecasts, it’s looking unlikely that Paul will win any states outright, although Idaho, North Dakota, and Maine present possibilities.
That’s right; in addition to the 23% of the Republican vote he took, Paul took 4% of the Democratic vote as a write-in candidate, good for second place, according to the NH Secretary of State. (Note: the NH SOS website is down right now, so I’m relying on descriptions of what it says given to me by my colleagues in New Hampshire.)
Last night, Ron Paul increased his percentage of the Iowa caucus vote from 10.0% in 2008 to 21.4%. If we can expect this same kind of increase from Paul in the remaining states, what would we expect his performance to be? I have found that Ron Paul’s primary vote shares are best modeled logarithmically, due to overdispersion in the data. Put another way, Paul increased his vote share from 2.3 log points to 3.1 log points last night, an increase of 0.765 log points. What if Paul experienced that same gain in the remaining states in the primary calendar?
I’m going to do something pretty simple here. I’m going to add that figure to Paul’s 2008 performances for each state and then convert back into percentages. However, Montana has switched from a caucus to a primary, which should hurt Paul, so for Montana I’ve subtracted from Paul’s score the estimated benefit he received from caucus states in 2008. I’ve arranged the list according to each state’s place in this year’s calendar.
|District of Columbia||3-Apr||17.5%|
I think that this forecast underestimates Paul’s support in next week’s New Hampshire primary, because it does not take into account the increase in Free State Project activists in that state. Nevertheless, it should be clear what a daunting task Paul faces. Even if he hits 20-22% in New Hampshire, he is not likely to be competitive in South Carolina and Florida, both of which are dominated by socially conservative defense hawks in the former case and older voters in the latter, and both of which are primaries, where Paul has usually done far less well.
However, if Paul can manage to beat expectations there, he stands a decent chance of winning the Nevada, Minnesota, and Maine caucuses and placing second in Colorado. A string of strong performances just might set him up to beat these forecasts in the next few states, especially if he becomes the anti-Romney anti-Santorum by default. After Super Tuesday, Paul faces a potential long dry spell of primary states in which he does not do particularly well. It’s hard to see how he does not get eliminated from contention for the nomination during this period, unless he really manages to build momentum out of New Hampshire and the early February contests.
Ron Paul is still surging. I have an article forthcoming in the next issue of The American Conservative forecasting the New Hampshire primary and the role that participants in the Free State Project and other libertarian activists may play therein. At the time I wrote the article, I made the fairly bold forecast that Paul will score between 15 and 30 percent, likely closer to the former number. That forecast is now looking less and less bold.
Two polls today show Paul in the lead in Iowa: PPP has him at 23%, three percentage points beyond Romney, and Insider Advantage has him at 24%, six percentage points over Mitt. Meanwhile, PPP‘s poll of New Hampshire has Paul at 19%, good for second place behind favorite son Romney. CNN/Opinion Research puts Paul at a record-high 14% nationally, while Gallup has him at a record high for their polls, 11%.
If Paul wins Iowa, which looks like at least a 50-50 proposition right now, then all bets are off in New Hampshire. The conventional wisdom is that a Paul win hurts Gingrich and helps Romney, but if Paul can use a win in Iowa to put a scare into Romney in New Hampshire, where Romney has always been expected to run away with it, Romney comes out badly bruised as well. Mainstream commentators are finally waking up to the possibility that Paul could win Iowa and New Hampshire. Right now, I’d put the probability of that occurrence at somewhere around 15%, but if it happens, it would be an earthquake.
Incidentally, the cross-tabs on these polls are enlightening. In the PPP poll of New Hampshire, Paul’s support among those who are strongly committed is 21%, indicating his firmer base. (Romney, however, is at 41% among firmly committed voters, implying he may be able to limit damages from an Iowa loss.) Paul is viewed overwhelmingly favorably in New Hampshire (53-38), which talking heads tell us is not the case most other places. Paul is the second choice of 49% of Gary Johnson supporters (who pulls in 1% himself), 30% of Michele Bachmann supporters, and 25% of Jon Huntsman supporters. Since Huntsman is doing well in New Hampshire, this seems to confirm my suspicion that to a certain degree he and Paul are struggling over a similar pool of voters. Paul is also the second preference of 23% of Romney voters, indicating the degree to which Paul’s appeal has broadened to moderates and independents. A final point of interest is that Paul is leading the field with 28% among those who view foreign policy or national security as the most important issue in the election.
Paul’s path to victory in New Hampshire, it would seem, would require a win in Iowa and an unexpectedly poor finish for Romney. If Romney’s core supporters started to drift away, Paul and Huntsman could expect to benefit.