Never has the phrase “a week is a long time in politics” been truer than during the past few months in the UK. A new prime minister has taken office, Parliament has been prorogued and Brexit has continued on its rocky path.
Forecasters have been tackling one of the most difficult political predictions of the year so far: “Between 3 May 2019 and 1 November 2019, which of the following will happen first in relation to the Brexit process?” Forecasters estimated the likelihood of the following possibilities:
a) The UK will leave the EU with the November 2018 Withdrawal Agreement in place
b) The UK will leave the EU with a new or renegotiated version of a Withdrawal Agreement in place
c) The UK will leave the EU without a Withdrawal Agreement in place
d) The UK will revoke Article 50
e) The UK’s Article 50 period will be extended beyond 31 October 2019
f) The UK will set a date for a general election
g) The UK will set a date for a referendum on Brexit
The results we have are for predictions made up until 4 September 2019.
The question has seen huge changes in predictions over time. In early May, option G, for the UK to have set a date for a referendum was given a 39% likelihood. But that relative certainty didn’t last. On 4 September 2019, forecaster confidence in this outcome had dropped to just a 6% likelihood.
Instead, the most popular prediction on 4 September 2019 was option C, that the UK would leave the EU without a deal. This option had only been assigned a 8% probability by the forecasters in early May, but that surged to a 30% likelihood by early September.