By Rep. Joey Sarte Salceda (2D-Albay)
Chairman, House Committee on Ways and Means
Co-Chair, Economic Stimulus and Recovery Cluster,
House Defeat-COVID-19 Committee
1. Summary. This analysis analyzes the strength of a Duterte-Marcos or a Marcos-Duterte ticket considering four key measures, namely (1) core strength, or the loyal base of voters with which a candidate will start, (2) transferability of core strength, or the ability of one candidate to carry their running-mate among their loyal base, (3) aversion, or the degree to which a certain candidate could polarize the electorate to the detriment of the ticket, and (4) national appeal, or the breadth of favorability of a candidate among various sectors.
2. The dynamic of the race for President. In surveys for President where Mayor Sara Duterte is mentioned as an option, the Davao City chief executive takes the lead; Former Senator Ferdinand “Bongbong” Marcos takes second or third place to Mayor Duterte. This dynamic holds for both June and September 2021 Pulse Asia Surveys.
a. Notably, in both surveys,no candidate breaches the 39% threshold, the threshold breached by the winning candidates in the 1998, 2004, 2010, and 2016 elections. Even at her peak, Mayor Duterte remained 11 points away from this threshold.
b. As will be discussed later, given the strength of these candidates’ core areas, and given their non-overlap, having both candidates compete for the Presidency will likely lock the other out of one’s core areas, leading to a scenario closer to the 1992 elections, where multiple administration-associated and opposition-identified candidates competed.
i. In 1992, the winning candidate, then-Defense Secretary Fidel V. Ramos, won 23.58% of the vote, a margin of only 3.84% versus runner-up Miriam Defensor Santiago
c. On the other hand, if the bases of both Duterte and Marcos could be combined without alienating too many voters in non-core areas, such that the ticket is not locked out of other areas where they could still compete for votes to add to their combined bases, such a scenario would almost certainly be enough to win the Presidency.
3. Analysis of core areas. Areas that voted 50% for Former Senator Bongbong Marcos comprise 15.64% of the vote, while areas that voted for President Duterte at more than 50% comprise 24.71% of the vote. Duterte has a bigger core area.
a. President Duterte clearly appears to have a bigger core base than Senator Marcos. Assuming Mayor Sara Duterte works with the same base as that of her father, she will be starting with a bigger base of core voters than Senator Marcos would be. On this measure, it would appear that a Duterte-Marcos ticket would have higher winnability than a Marcos-Duterte ticket.
b. Among areas where Duterte won more than 50% of the vote (29 provinces and HUCs), his average performance was 67.35%. Marcos’s core areas, while smaller (17 provinces and HUCs), appeared to have voted slightly more strongly for him, at an average performance of 71.09%. The Duterte core areas are bigger, but just slightly softer, while the Marcos core areas are smaller, but slightly harder than the Duterte core areas.
c. Overlap in core areas. The likely overlap of support between Duterte and Marcos voters in their core areas is significant, but not enough to win the Presidency. In simpler terms, the overwhelming majority of each candidate’s voters in their respective core areas do not overlap. In theory, a combination of their respective bases will add significantly to the votes of the running-mate. Whether this transferability effect can materialize is discussed in the paragraph on this matter.
i. Duterte core areas. In Duterte core areas, Marcos performed at 27.58%, below his national performance of 34.47%. In fact, if Marcos performed in these areas just on par with his national performance, he would have netted an additional 1.69% (or 694,000 votes) of the national vote, enough to have made him the winner of the 2016 elections for Vice President.
ii. Marcos core areas. Duterte’s performance in Marcos’s core areas incidentally almost mirror’s Marcos’s performance in Duterte core areas, at 27.60% of the vote. If, in Marcos core areas, Duterte performed on par with his national performance, he would have gained an additional 760,000 votes.
4. Analysis of transferability of votes in bailiwicks. This measure analyzes the ability of the candidate to carry their running-mate in the race. This measure could have implications on (1) who will deliver more for the ticket and (2) whose voters are more obedient to the candidate and could therefore rely on a post-election governing coalition of supporters.
a. Core Duterte areas. In areas where President Duterte carried more than 50%, his running-mate, Senator Alan Peter Cayetano performed at an average of 31.03% per area. This is compared to his national performance of 14.38% of the total vote, or an overperformance of nearly 17%. The correlation of their performances in Core Duterte areas was very strong, at 86%. In other words, Duterte voters are more likely not only to support their candidate but also to follow their candidate’s choice of running-mate.
i. We saw the power of this transferability in 2019, the only legislative “or midterm” election in the post-1987 history of the country where the official opposition was locked out of winning any Senate seat.
ii. In contrast, even at the peak of President Marcos’s popularity in the 1967 midterm elections, the opposition Liberal Party was not locked out of office, and the most vocal opposition candidate Senator Benigno Aquino, Jr., ranked second in the Senate race, with 49.5% of the vote.
b. Core Marcos areas. In areas where Senator Marcos carried more than 50%, his running-mate, Senator Miriam Defensor-Santiago carried 7.7% of the vote on average, double her national average of 3.4%, but not enough to make her the leader in any of the core areas. In fact, in terms of correlation, their performance demonstrated significantly weaker correlation in core Marcos areas than nationwide. Contrast this with Cayetano’s performance, winning 6 of Duterte’s core areas.
c. By a wide measure, Duterte voters are likely to carry the running-mate of their candidate than Marcos’s voters are. This makes a Duterte-Marcos ticket more favorable to Marcos’s chances of winning than a Marcos-Duterte ticket is for a Duterte.
d. In terms of outcomes, given the combination, if a Duterte wins by a large margin for President, the winning Vice President will likely be Marcos, whereas if Marcos wins for President, the winning Vice President may not necessarily be Duterte.
5. Analysis of voter aversion. Another key measure is the likelihood that certain voters may be turned off by having one candidate on top of the ticket over another.
a. We measured ‘aversion’ by how weak a candidate is in a certain province or HUC. We consider areas where a candidate performed at 10% or weaker as “averse,” where performance is 10% to 15% as “mid-averse” and otherwise as “not averse.”
b. The results of this analysis are as follows:
c. A notable figure is that provinces that account for 10.54% of voter turnout in 2016 may not be inclined to vote for Marcos but does not have the same disinclination towards Duterte. The converse is true only for areas that account for 1.40% of the voter turnout in 2016. On this measure, a Duterte-Marcos ticket may generate less aversion than a Marcos-Duterte ticket.
d. In strategic terms, a Duterte-Marcos ticket can be sold to more voters and more areas than a Marcos-Duterte ticket would be.
6. National appeal. This measure seeks to analyze the overall or aggregate strength of the candidates using sectoral analysis and aggregate polling numbers.
a. Aggregate results of the Pulse Asia surveys. In referring to surveys, our analysis will make use of the Pulse Asia surveys conducted from June 7 – 16, 2021 and September 6-11, 2021. We use these surveys as they (1) both mention Mayor Duterte and Fmr. Senator Marcos, and are conducted by the same firm and are therefore comparable in methodology.
i. Both polls show Mayor Duterte to be leading, initially with a 15-point lead over Senator Marcos, and a smaller 5-point lead in the latter survey.
ii. Both results consider Mayor Duterte denying that she will run for President; hence these numbers may conceal actual support for Mayor Duterte if she were to declare herself a candidate.
b. Insights on class from the June and September 2021 surveys.
i. The June Pulse Asia survey shows Mayor Duterte winning all income classes (ABC, D, and E). This would have been weaker, but consistent, with now President Duterte’s performance in the 2016 elections, where, according to the SWS exit poll, he won all income classes.
ii. The September Pulse Asia survey still shows Mayor Duterte leading among classes D and E, and within the margin of error for the lead in classes ABC (16%), behind Manila Mayor Isko Moreno (20%) who actually lost 1 percentage point from his June performance in the same class, and Fmr. Senator Marcos (18%, a gain of 1 pp from June).
iii. Both surveys appear to show Mayor Duterte having broader inter-class appeal than any other option in the surveys.
7. Conclusion. Based on various measures, a Duterte-Marcos would have a stronger chance of having both candidates win because (1) Duterte voters are more likely to carry the running-mate with their candidate on top; (2) Duterte-solid areas account for a larger share of the national vote than Marcos-solid areas; (3) there are more Marcos-averse areas than Duterte-averse areas, with the former being potentially more emphasized if a Marcos-Duterte ticket were to be agreed upon. In other words, a Duterte-Marcos tandem would have a bigger core, a bigger neighborhood of non-core areas that can be appealed to.