Modelling the 2020 Singapore General Election

Will this be a game of chess, or a game of twister?
Choose wisely when you play with Death!

When its prime minister Lee Hsien Loong called for parliament to be dissolved on 23 June 2020 for snap polls, Singapore joined an exclusive club of nations holding national elections during the global pandemic. South Korea's ruling Minjoo Party won its snap polls easily and even extended its majority in parliament. Taiwan's Tsai Ing-wen was handily reelected in its presidential polls. Will Singapore's People's Action Party do the same? Is it checkmate and a total wipeout for Singapore's opposition, which held just 6 seats out of 89 after the 2015 election?

How do voters decide in Singapore?

In 2015, we at Illusio modelled the Singapore general election. This year, we continue to believe that Dunleavy's multinational analysis of dominant party systems is the right model to forecast Singapore elections. The Dunleavy model has not broken down for elections post-2014, and Singapore continues to be a dominant party state.

Remember this graph?

As a refresher to our 2015 post on Dunleavy's model, the ruling party in a dominant party state enjoys a "halo of effectiveness". For a wide range of rational voters, the dominant party is a preferable choice to any other party that is closer to the voter's ideological position. Qualitative surveys such as those conducted in Japan show voters tend to choose the incumbent party because they see it as an effective deliverer of competent policy, even when those policies deviate with voter policy preferences.

In more practical terms, constituents in Japan and other dominant party states around the world see dominant political parties as a source of local spending projects and area improvements. This may also explain why the PAP has, in the light of policy failures since 2011, focused on unveiling URA's "town council masterplans" and taking credit for them.

How should the political contest be fought?

Under the Dunleavy model, elections in a dominant party state are essentially plebiscites on the ruling party. Has the ruling party delivered in its previous term? Is its vision for the nation in the next term what the people actually want? Has life improved for people in the past 4 years? Do they think the next 5 years will be better?

What would the Singapore/PAP report card for 2015/2020 look like?

Applying first principles to the development of the past 5 years, it's clear that any serious and competent party that is not the PAP would run a campaign attacking

1. the PAP's competence, focusing on highly publicised policy and governance failures (say national privacy failures, the failure to provide a sustainable 99-year HDB lease solution, international corruption of Singapore-branded companies like Keppel, the powerless job bank, the botched coronavirus strategy of the multiministerial task force and the massive covid19 dormitory outbreaks, amongst other failures)
2. the PAP's vision for Singapore. How many voters want a Singapore with 6.9 million people (or according to Liu Thai Ker, 10 or 20 million!), easy visas and immigration for foreign job-seekers, a depressed wage and low productivity economy, and more ineffective rounds of economic restructuring that do not address Singapore's addition to low productivity?

and downplaying 3. the importance of the sole thing the PAP did right: turn on the stimulus to keep businesses running and workers employed during the corona crisis.

What is the fly in the ointment?

Dunleavy notes a pure consequence of the dominant party model:

Only when some opposition parties adopt ‘convergent’ or ‘deeply convergent’ positioning strategies will support for dominant parties tend to be seriously eroded. Greater crowding of the ideological space is a key stimulus to some opposition parties adopting convergent or deeply convergent strategies.

But are there enough viable opposition parties this year to seriously erode the PAP's dominance, and its dominant share of the vote?

We note that the 2020 general election is the second time in Singapore's modern history where all seats in parliament are contested. Does this mean the field is "crowded" with sufficient parties adopting convergent or deeply convergent strategies? Unfortunately not since none of the opposition parties are entering contests in the same seats as each other. Dunleavy's Figure 3 describes contests where the dominant party is competing with an overcrowded field. This is not the same as in Singapore, where the dominant party is competing one-on-one with different challengers across the map, which is essentially Figure 2.

To put it in concrete terms: The Workers Party is contesting in East Coast Group Representative Constituency. The East Coast voter only has a choice between WP and PAP. The PAP is able to keep its 'halo' in the ward because not enough parties are contesting and overcrowding its position in the East Coast ward. Where the presidential election of the 4 Tans was overcrowded with multiple challengers, it resulted in a narrow victory by Tony Tan, the PAP's preferred candidate.

Schools and political parties are the same

And we are assuming that every opposition party is a viable party. According to the voter preferences of 2015: none of them are viable, but one was only fit to have a token voice in parliament. More research and surveys need to understand why, but Dr Tan Cheng Bock's Progress Singapore Party had, during its online party launch, hinted at an attempt to build a party capable of contesting every ward, i.e. a national party.

This is apparently a heretical idea in Singapore, that an opposition party should want to contest all the seats. But Dr Tan is a maverick. And also a realist. How do you ensure that your party doesn't get stymied or caught unprepared when constituency lines are redrawn every election? Just have a party huge enough, with volunteers and grassroots wide enough that you don't need to sweat to find assenters, polling agents and campaigners in any ward. That is a viable party, which Dr Tan has not succeeded in delivering, even though the PSP is the biggest challenger for seats in parliament this year (a total of just 24, out of 82).

In other words, we at Illusio see this as a difficult and uphill battle for the opposition and an easy victory for the People's Action Party, barring an upswell of dissatisfaction and unhappiness from Singapore's electorate or a newfound ability to puncture the PAP's aura of competence by the opposition during the campaign.