I am a curious person, who has a desire to challenge unsubstantiated assertions. So, when I heard comments that Jeremy Corbyn’s progressive approach to the economy and society would not secure enough votes for Labour to win in 2020, I wanted to test this.
Moreover, claims that Labour must accept the Conservatives’ narrative and shape its policies accordingly to win also needed to be assessed, as well as the need to win more constituencies in the south east of England, which can only be done with a centre right agenda.
Unfortunately, I do not have either a crystal ball or perfect foresight. But, being an economic modeller, I thought I would build a model to test these claims. Then, based on the conclusions arrived through this analysis, a decision on who is the best Labour leadership candidate can be made.
The method for exploring the battleground that I have used is Scenario Analysis. This analysis allows us to establish and analyse possible changes and the impact of these on the outcome of the general election in 2020. Scenario Analysis is a decision making tool that is used in economics extensively to make projections for the future.
I created a spreadsheet model in MS Excel and populated it with the downloaded 2015 General Election data from The Electoral Commission website. I derived new variables to estimate the assumptions at constituency level. The derived variables are the forecast vote for Labour, the Conservatives, the Lib Dems, the Greens, UKIP and the nationalist parties in Scotland and Wales based on the model assumptions relating to shifts to and from Labour. I also derived a variable to estimate the impact of attracting current non-voters to Labour. I have excluded Northern Ireland from the analysis.
Identifying the Battlefields
First, I wanted to assess the landscape and determine the possible target constituencies for 2020. I assumed that these target constituencies could be won if Labour managed to swing 5,000 votes from the winning party, assuming everything else remained constant. I know this is a bold assumption but it’s a model and you need to start somewhere. This resulted in 125 constituencies being identified. Tables 1 describes the target constituencies by the incumbent party, and table 2 shows the incumbent party by region.
Table 2 suggests that the constituencies are spread across the country and that Scotland is a priority area with 19 seats, and then the South East. Whilst, the Welsh and Scottish nationalists are in percentage terms the largest target with 36% of their constituencies in our grasp, we do need to be mindful that this suggests we are targeting almost one third of currently Conservative held seats. However, does this mean that we need to ape Conservative policies?
Trying to Win the Battles
With the scenario analysis model, you can enter an endless variety of examples of shifts from one party to Labour and from Labour to another party. I am going to highlight two scenarios. The first is what I term ‘Chasing the Tories’ and the second is ‘Progress to Success’.
Scenario 1: Chasing the Tories
The assumptions are current non-voters will never vote thus there is no need to try to engage, and the key to winning is attracting a significant share of current Conservative voters and some floating voters.
Table 3: Chasing the Tories model assumptions
Table 3 shows a screenshot of the model inputs for this scenario. It assumes there is no change in 2015 non-voters voting Labour in 2020; one in three Conservative and UKIP voters vote Labour in 2020. This is the level of conversion needed for Labour to win more votes than the Tories given the other assumptions. I have also assumed some modest gains in other floating voters. So, one in ten people, who voted either Green, Lib Dem or SNP/PC vote Labour is the assumed level. Lastly, it is assumed that 10% of 2015 Labour voters opt to vote for another party.
Table 4: Results from the Chasing the Tories
This scenario forecasts winning 63 out of the 97 identified Conservative seats, and one of the Conservative constituencies was not in the identified group. Also, seven of 11 identified constituencies these are in the South East. These assumptions indicate no gains in Scotland from the SNP.
The overall impact is that Labour secures the most seats, in 298 constituencies, but do not have a majority. However, Labour may be able to form a coalition.
Scenario 2: Progress to Success
The assumptions are current non-voters are nurtured through intrinsically-motivated local engagement and attracted to voting Labour by a progressive policy platform vote in 2020; these progressive policies are part of a credible economic plan that are attractive to a significant proportion of people, who in 2015 voted either SNP, UKIP, or Greens.
In this scenario, it assumes one in four previous non-voters vote Labour in 2020, and one in three former Green, SNP/PC, or UKIP vote Labour. Lastly, it is assumed that 10% of 2015 Labour voters opt to vote for another party. Table 7 details the assumptions.
Table 7: Progress to Success model assumptions
Table 8: Results from the Progress to Success
This scenario forecasts Labour to win an additional 87 seats. Forty-two seats are won from the Tories (38 in England including 4 in the South East and three in Wales). Thirty-eight seats are won in Scotland from the SNP. The impact is that Labour secures the most seats, in 319 constituencies, and a small majority.
The scenarios discussed here clearly suggest that as opposed to chasing the Tories, a more effective and realistic strategy is to reach out with a progressive and credible economic alternative that would re-engage non-voters and re-gain ones that we have lost to other parties.
People are recognising that Jeremey Corbyn’s policies offer this progressive and credible alternative, and anecdotal evidence suggests these are gaining traction and popularity with the target groups. Thus, Jeremy Corbyn’s plan offers the best chance of winning in 2020.
I am happy share the MS Excel model with interested parties. Can I ask that anyone who would like a copy please make a donation to my team’s (Keep Left) Cancer Research Relay for Life at http://bit.ly/1TZq1si? Thank you.