The South Carolina GOP Primary was held last night. After days of drama and political mudslinging that is typical of a primary in the Palmetto State, the results were more or less along expected lines. Donald Trump won with a 32.5% vote share; there was virtual tie for the second position with Marco Rubio emerging just slightly ahead of Ted Cruz. Rubio earned the support of 22.5% of the Republican electorate compared to Ted Cruz’s voting share of 22.3%. Jeb Bush, John Kasich and Ben Carson earned less than 10% of the votes each. Bush promptly dropped out of the race.
Here are some takeaways from the South Carolina results:
Ted Cruz is increasingly looking like a factional candidate:
When Ted Cruz won the state of Iowa, the betting market was unwilling to take him seriously. His odds of winning the nomination increased from 8%-9% to barely 13-14%. The odds seemed too low at that time; but, after two more states getting to vote, the odds make more sense now. The path for a Cruz nomination is looking increasingly narrow and constrained.
On paper at least, Ted Cruz looked a superior candidate to either Mike Huckabee or Rick Santorum, the winners of the Iowa caucus in 2008 and 2012 respectively. Huckabee and Santorum were poor at fund raising and as a result, ran campaigns on shoe string budgets, surviving one state at a time. Cruz, on the other hand, has an impressive fund raising machine with significant contribution from small donors. In fact, in terms of cash raised by a candidate and his super PAC combined, Cruz has been the second most successful candidate in the field, after Jeb Bush. This comfortable liquidity position also enabled the Cruz campaign to focus on the states which were slated to vote later in the schedule, especially the Southern states which looked natural fit for him.
If Cruz’s plan was to sweep the southern states and build up a formidable lead of delegates after the March 01st and March 05th primaries, this strategy is now lying in tatters. Trump has already won the South Carolina Primary comfortably and is leading in most of the other Southern states.
But why is Cruz struggling in the South. After all, the South has higher percentage of evangelical voters compared to Iowa. For example, 67% of the Republican voters in 2016 were evangelicals compared to 65% in Iowa. Also, as the following chart based on the 2012 exit polls demonstrates, Southern states typically have higher percentage of evangelicals among the Republican Primary voters compared to Iowa.
The answer lies in the fact that evangelicals in the South are relatively less conservative compared to evangelicals in Iowa. If we look at the percentage of very conservative voters in the 2012 exit polls, Iowa ranks almost at the top, after Nevada, and the southern states come below it.
This is because Iowa conducts a caucus while the Southern states conduct primaries. Caucus is a tedious, time consuming process which attracts a disproportionate share of the party faithful. As a result, the percentage of very conservative voters in GOP caucuses is generally higher compared to that in primaries. This also explains why candidates like Donald Trump, Newt Gingrich, John McCain, Bob Dole and George HW Bush, candidates who cannot be considered social conservatives, have won in South Carolina in the past.
Ted Cruz has done exceedingly well among very conservative voters. However, as the results of Iowa, New Hampshire and South Carolina show, his support level drops from very conservative voters to somewhat conservative voters by a factor of close to 50% and his support among moderate voters drop by a further 50%.
It may be pertinent to mention here that Cruz’s support level among very conservative voters was down from the level seen in Iowa, which is part of the reason why he struggled in South Carolina. But there is also not much room to grow his share among such voters. Further, somewhat conservative voters still constitute plurality of the Republican voters in most states, even in the more conservative states. Thus it will be very difficult for Cruz to win the nomination from here on unless he manages to significantly expand his base among these somewhat conservative voters.
Incidentally, Rick Santorum, a candidate similar in positioning to Cruz also dominated the very conservative voters in the 2012 primary elections. But, he also garnered sizeable support among somewhat conservative and moderate voters which allowed him to win Iowa and a few states in the South.
Santorum’s support level among somewhat conservative and moderate voters was indeed lower than his support level among very conservative voters; but, the fall in support was not as drastic as that of Cruz. Part of this may be on account of the fact that Santorum made a conscious appeal to the working class voters and presented himself as a compassionate conservative; Cruz, on the other hand, has positioned himself as a fiscal hawk and has made no attempt to appeal to the more moderate sections of the party. Even then, Santorum, with broader based support compared to Cruz, struggled to post any major victory outside the South and a few caucuses. This just goes on to show how difficult it is for a conservative purist to win a plurality of delegates in the Republican Primary.
Notice how the high support level of Santorum among very conservative voters in the mid-western states (i.e. Michigan, Ohio, Wisconsin and Illinois) did not translate into any victory there.
We have argued before how previous polling shows the SEC states are not a firewall for Cruz. Further, most of such states award their delegates on a proportionate basis; as a result, even if Cruz ends up winning a few states including the delegate rich Texas, his home state, he would not be able to open up a significant lead in the delegate count.
It is also difficult to see a path for Cruz once the Southern states are done with their primaries. He may perform well in a few conservative states in the west. But even there, he will have to compete hard with Rubio who is almost as conservative as Cruz is, will have the backing of the party actors and was also a Mormon (Mormons constitute a significant percentage of Republican voters in states like Utah, Nevada, Arizona, etc.).
Overall, based on the results of South Carolina and based on his performance so far, it has become difficult to make a case for Cruz. May be, he shall be able to use his local roots and strong organization to turn things around just in time before the 01st March primaries. We may be wise to reserve the denouement till then.
Donald Trump has a high floor but a (relatively) low ceiling
After today’s result, the Donald Trump candidacy looks like an unstoppable juggernaut. After all, he has won in all but two of the 46 counties and has emerged winner in each of the seven congressional districts in South Carolina. He has made a sweep of the delegates being offered in the state. Even then, it will be gross oversimplification to consider Trump as an inevitable candidate.
The most worrying factor for Trump would be the fact that late voters are deciding against him. In each of the three states that have voted till date, Trump’s share among voters who decided earlier was much higher than those who decided late.
Roughly 45%-55% of the Republican voters have decided their votes more than one week before the primary or caucuses. Trump crushes the opposition among such voters. However, among late deciding voters, Trump fares much worse. In states where Trump already had a high support level (like South Carolina and New Hampshire), he could hold on to his lead to win. However, in Iowa where he had a much narrower lead, loss of the late deciding voters gave Cruz a win.
This shows that most of Trump’s supporters have already made up their mind. As a result, he is very unlikely to over-perform the polling numbers on the election days. Also, the non-Trump voters have been searching for options till the election days, trying to find the candidate they should support. In a divided field, it was not much of a worry for Trump. But with the field getting winnowed, when the contest moves to less favourable terrain for Trump, this may prove to be a headache for him.
This is partly related to the favourability numbers of Trump. While Trump has seen a significant increase in his favorabiluty numbers since his campaign started, he is still disliked by a large portion of Republican voters. However, at the same time, relative to other candidates, he has been able to convince a much larger portion of voters who view him favourably into supporting him. For comparison’s sake, Ben Carson today has the highest favourability rating among Republican voters, but is also ranked at the bottom when it comes to voters willing to vote for him.
This means a large proportion of Republican voters have already decided on Trump and shall vote for him no matter what. However, voters who are undecided mostly have a negative opinion about him and are willing to wait till the last moment to select a candidate other than Trump. In other words, Trump has a high floor but a low ceiling.
It is instructive that Trump has not been able to breach the 35% vote ceiling in any of the early three states. 35% may be enough to win states in a multi-candidate race, but may not be enough if the field gets reduce to two or three candidates.
The best hope for the Republican establishment is thus to reduce the race to a one on one between Trump and a candidate with high favourability and acceptability across sections of the party. Fortunately, they have this candidate in the form of Marco Rubio. However, they must also hope that it does not get too late before the winnowing happens.
Donald Trump is a polarising candidate
It is difficult to say this by looking at the map which he has more or less swept, but Trump is also a polarizing candidate. In fact, Trump’s performance among voters who want the candidate to share their values is abysmally low. He performs only slightly better among candidates who are concerned about electability. However, he does crush his rivals when it comes to voters who wants a candidate who tells it like it is and can bring needed change.
Value voters have preferred Cruz so far in this cycle, whereas candidates concerned with electability have tended to favour Rubio. Together such voters have constituted around 40%-55% of the total Republican voters. It is a little strange that the leading candidate in the Republican Primary has no appeal to the most idealistic voters who want the candidate to reflect the values that they have or to the most practical voters who want the Party to win in the general election. Instead, the Trump candidacy is being bizarrely driven by a revolt against the existing political system, votes for the politicians to come out of political correctness and votes for an outsider to change the system.
This same phenomenon is reflected in the answers to a separate question that asks the voters what they consider best preparation for presidency is – being from outside the system or having political experience. Rubio, in particular and Cruz, to a lesser extent does better among voters who agree with the latter while struggling among voters with preference for outsider candidates. Trump on the other hand has hardly any backing among voters who prefer political insiders while towering above rivals when it comes to selecting a voter with outsider credentials.
Those pundits who think there is no differentiation between establishment and outsider candidates should take a look at the chart above. It clearly delineates the so-called lanes in the Republican Primary – Trump is the clear favourite of voters who want a rank outsider, whereas, for voters who want an experienced politician, Rubio is the undisputed choice. Cruz is occupying the somewhat uncomfortable position of having limited appeals to both sections of the party.
The bifurcation among voters who want an establishment candidate and those who want an outsider has been close to 50-50 in both Iowa and South Carolina, indicating that the race shall continue to be closely fought among two sections of voters who want very different things from their candidate.
The County Wise Results Mirrored those of 2012, but not entirely
For people who have been confused by the unpredictability of this year’s Republican Primary, yesterday’s South Carolina election must have come with a much needed air of familiarity. Donald Trump, Marco Rubio and Ted Cruz’s performance in the state were pretty similar to those of Newt Gingrich, Mitt Romney and Rick Santorum respectively in 2012. The countywise comparison of the candidates’ performances has been mentioned in the following scatter plots:
Trump vs Gingrich:
Rubio vs Romney:
Cruz vs Santorum:
The relationships are by no means perfect. The field this year is yet to settle down, with Bush and Kasich competing hard with Rubio for establishment votes and Carson winning some of the value votes. However, Cruz like Santorum in 2012 and Huckabee in 2008 did well in the more social conservative northern parts of the state whereas Rubio did well in the heavily urban counties, including winning in Charleston and Richland counties. Trump like Gingrich before him did well pretty much across the state.
Presence of Kasich and Carson in the race shall help Trump
The first casualty of the South Carolina election night was Jeb Bush who had to end his campaign, a bit ignominiously after burning through more than 120 Million USD of cash in a futile and often embarrassing bid. But, as on today, John Kasich and Ben Carson are still there in the race. And the continuing presence of these candidates shall continue to benefit Trump.
It would, of course, be simplistic to argue that all of Kasich’s voters shall go to Rubio whereas all Carson voters would henceforth support Cruz, if Kasich and Carson drop out. But a look at the exit polls suggests that like Cruz, Carson is primarily drawing support from very conservative and value voters. One may argue that the voters for Kasich are more moderate compared to those of Rubio, but on the question of experience, they are firmly in the establishment category.
In the last section, we discussed how the county wise votes garnered by Trump, Rubio and Cruz reflected trends of previous candidates in 2012. In fact, if we add the vote share of Cruz and Carson and plot it against the vote share of Santorum in 2012 or Huckabee in 2008, the fit (i.e. R-squared) improves. Similarly, the addition of vote shares of Rubio, Bush and Kasich and plotting it against the vote share of Romney in 2012 generates a more meaningful curve.
Cruz + Carson vs Santorum:
Cruz + Carson vs Huckabee:
Rubio + Bush + Kasich vs Romney:
This indicates that Cruz and Carson on the one hand and Rubio, Bush and Kasich on the other hand are drawing from the same well of voters who supported Santorum and Romney respectively in 2012. The longer Carson and Kasich stay in the race, the more votes they will continue to siphon off, votes which would have otherwise disproportionately gone to Cruz and Rubio respectively.
Carson may have already become a forgotten footnote in this year’s primary cycle; but, his voting share has approached the high single digits in Iowa and South Carolina. If he performs as well in the Southern states slated to vote on 01st March and 05th March, he may cost Cruz a number of delegates. First of all, these states are expected to be contested strongly between Cruz and Trump and Carson’s vote share may end up being the difference between these two candidates in a number of states. Secondly, most of these states award three delegates to each congressional district with the winner getting two and the second placed candidate getting one, provided he clears a minimum cut off. In a number of such congressional districts, Carson may turn out to the difference between a first and second place appearance or between a second and third place appearance for Cruz. In other words, Carson has the potential to cost Cruz plenty of delegates in the coming primaries.
John Kasich, at the other end of the spectrum, is doing relatively well among moderates. Such voters will come plenty in the blue states of east and west coast. As a result, Kasich may be expected to perform better in such states, like he did in New Hampshire. This may be problematic to Rubio, especially in the North Eastern states where Trump is likely to be very strong and a division of votes between Rubio and Kasich, like in New Hampshire, may lead to Trump winning a number of states in the delegate rich region. Further, Kasich may also perform well in Middle East, especially in Ohio, where he is the sitting governor and which is almost a must win state for Rubio.
Thus while Trump may not be unstoppable right now, the longer Carson and Kasich stays in the race, the higher the odds of a Trump nomination will get.