Joko Widodo and Jusuf Kala secured 53.1 per cent of the vote in the 2014 Presidential Election. Photo by Widodo S Jusuf for Antara.


It is now less than a year before Indonesia holds its first ever simultaneous national legislative and presidential elections on 17 April 2019. Given the possibility of a rematch between President Joko “Jokowi” Widodo and Prabowo Subianto, his opponent in 2014, campaign teams and political operatives will no doubt be looking back to the 2014 election for lessons that can be applied to the rapidly approaching 2019 race.


In a recently published research paper, Nurdien Aji and I examined the factors that were important for Jokowi’s victory in 2014. In particular, we investigated the effect of socioeconomic conditions at the village level on voting behaviour.


Our study combined data from the 2011 Village Potential Census (Podes) (which captures limited information on village-level infrastructure and local economic development) with data from polling booths uploaded through the crowdsourced Kawal Pemilu vote-counting initiative to assess how village-level economic conditions affected voting behaviour. Our study found that, in general, villages with good economic conditions were more likely to vote for Jokowi.


Our study also attempted to assess the impact of media access and religious and ethnic identity. In a biased media environment, access to television and the internet (through smartphones) seemed to be important factors influencing voters to back Prabowo. Religion also played a role.


The 2014 presidential election marked Indonesia’s transition into a mature democracy. It was Indonesia’s third successful direct presidential election, and it involved peaceful transition to a new president from outside the political and military elite.


Following the 9 April 2014 Legislative Election, parties coalesced around two coalitions. Joko Widodo was backed by the Great Indonesia Coalition, consisting of the Indonesian Democratic Party of Struggle (PDI-P), the National Democratic Party (NasDem), the National Awakening Party (PKB), the People’s Conscience Party (Hanura), and the Indonesian Justice and Unity Party (PKPI). Prabowo, meanwhile, was supported by the Red and White Coalition of Gerindra, Golkar Party, the National Mandate Party (PAN), the Prosperous Justice Party (PKS), the United Development Party (PPP), and the Crescent Star Party (PBB). As Prabowo had the backing of four Islamic parties to Jokowi’s one, Prabowo’s team presented their candidate as more truly representing the interests of the Muslim community.


Despite the fact that the Red and White Coalition secured 48.9 per cent of the vote in the legislative election, compared to the Great Indonesia Coalition’s 40.9 per cent, Jokowi emerged victorious, with 53.1 per cent of the presidential vote (compared to 46.9 per cent to Prabowo).


We developed an econometric model to determine whether there was any relationship between economic conditions, ideology, access to information, and religion and ethnicity, and voting behaviour.


Our results showed that in villages with more positive economic conditions, such as good infrastructure (like paved roads) and economic access (availability of banks and markets), voters were more likely to back Jokowi. Our model found that the probability of Jokowi securing more than 50 per cent of votes increased by almost 2 per cent in villages with good infrastructure.


This suggests that despite the fact that village level infrastructure such as banks, markets and roads are under the authority of district or city-level governments, voters still interpret them as being under the central government’s authority.


Villages with sizeable middle-class populations were more inclined to vote for Jokowi, while villages that were dominated by working class residents (agricultural labourers and factory workers) tended to prefer Prabowo.


In a 2015 study of the 2004, 2009 and 2014 elections, Higashikata and Kawamura used city and district level gross regional domestic product (GRDP) data to demonstrate that districts with higher economic growth tend to support ruling parties. Mujani, Liddle and Ambardi also found that voters tend to favour incumbents if economic conditions are good, and vote for the opposition when they are poor.


It is possible voters who experienced strong economic conditions under Susilo Bambang Yudhoyono were more inclined to associate Jokowi with continuing the economic policies of his predecessor. Jokowi was relatively accommodating of the market and displayed a preference for limited government intervention in the economy, policy positions more consistent with the policies of Yudhoyono. By contrast, Prabowo’s strongly nationalistic campaign advocated more radical change to the structure of the Indonesian economy.


Our results also confirmed the vital role played by the media. The probability of Jokowi winning in villages with a good quality private television signal decreased by 11 per cent.


It is possible that this outcome was a result of the polarised media coverage of the 2014 race. During that election, there was a sharp divide in how television media reported on the election, because of the political affiliations of media owners. TVOne, ANTV, MNCTV, RCTI and Global TV all favoured Prabowo, while only MetroTV was associated with the Jokowi campaign.


Finally, religion and ethnicity also appeared to be an important factor in voting behaviour. If we look at the partial effect of ethnicity and religion, majority Muslim villages were more likely to vote for Prabowo. Assuming that all other variables are constant, the probability of Jokowi securing more than 50 per cent of votes decreased by 38 per cent in villages with a Muslim majority.


On the other hand, villages with majority Javanese populations were more inclined to vote for Jokowi. The probability of Jokowi winning in villages with majority Javanese populations increased by about 36 per cent. This is significant, given that Javanese represent almost half the Indonesian population. Combining both ethnicity and religion, this study confirmed that religion was a more dominant factor than ethnicity in influencing voter behaviour in the 2014 presidential election.


These findings suggest that a carefully targeted media campaign, focusing on economic performance, ethnic identity and religious issues could play a significant role in determining the outcome of next year’s presidential race.


This post was based on the article “Behind Jokowi’s victory: did economic voting matter in the 2014 Indonesian presidential election?”, recently published in the Asia-Pacific Journal of Regional Science.



, , ,

We acknowledge and pay respect to the Traditional Owners of the lands upon which our campuses are situated.

Phone:13 MELB (13 6352) | International: +(61 3) 9035 5511
The University of Melbourne ABN:84 002 705 224
CRICOS Provider Code:00116K (visa information)