Tory think tank Onward released a report on the 30th October detailing an important new voter group labelled ‘Workington Man’. Like Mondeo Man or Worcester Women before, the label defines the sort of voter that will win the Conservatives the election. John Curtice, renowned bespectacled psephologist has pointed to “nearly 50 Labour seats that could fall into the Tory column… disproportionately located in the north of England and the Midlands, where many a voter backed Leave”, these seats are defined by older, white male Brexit voters — now grouped within the idea of ‘Workington Man’.
Given that this sort of voter archetype is central to victory, one would expect that the Conservative Party alongside Labour and the Brexit Party will invest heavily in pushing advertising to these groups. The Brexit Party is currently fitting that mould. However, the Conservatives and Labour (the two main contenders for the seat) are not. These parties are currently via Facebook advertising more interested in reaching Workington Woman rather than Workington Man.
Using analysis of Facebook adverts in Workington and other major Brexit voting target seats, the old school definition of Workington Man doesn’t actually fit what is actually happening in campaign reality.
Image 1 — Portland Square in Workington — via Alexander P Kapp: licensed for reuse under Creative Commons.
Workington Man is labelled by Onward as:
This voter is likely to be over 45 years old, white, does not have a degree and has lived in his home for over 10 years. He voted to Leave the EU in 2016 and thinks the country is moving away from his views both economically and culturally. The typical “Workington Man” favours security over freedom across both social and economic axes, but leans much more towards security on social issues.
Workington Man is more likely to think that crime is a major issue facing the country and twice as likely as the rest of the population to think that immigration is a major issue. He is particularly sceptical about the benefits of globalisation and thinks that we have a special responsibility to protect local institutions such as pubs and post offices from closure.
Workington man is one of a few general voter groups who are currently ‘volatile’ in their vote choice. With high levels of volatility and vote switching caused by modern political malaise, economic problems and Brexit, some segments of the voterbase are ripe for campaigning. As the British Election Study reported at the 2017 election, it featured “the highest levels of switching between the Conservatives and Labour since the BES started in 1964” (BES data).
Graph 1 — BES data showing vote share of parties over time
In order to examine the targeting of Workington man in reality, it is important to understand what a place like Workington is like both offline and online.
According to incumbent Labour MP Sue Hayman, who is defending a majority of 3,925 in the Workington Parliamentary constituency. A constituency of 59,361 people also featuring the towns of Cockermouth and Maryport. 60% voted to leave and the constituency is ranked 177th out of 650 in the governments index of multiple deprivation.
Graph 2 — 2017 General Election results in Workington
Examining Workington town via Facebook audience insights, some of these realities are also visible through the region’s Facebook residents. The real-world population of Workington is held to be 25,207 (2011), with a Facebook population of 20–25,000 people (2019). It is important to note here that Facebook has been found to overestimate population statistics including having more users on Facebook than local populations caters for. The Facebook users of Worthington are 50% more likely to state their job as a community or social service role than the UK Facebook average, -25% less likely to work in the arts, entertainment, sport and media and -36% less likely to work in food or restaurants. Residents on Facebook are far less likely to have a postgraduate degree and far more likely to be high school educated only.
The constituency, like the rest of the UK is also a desktop friendly location, Workington Facebook users overwhelmingly use phones to access the app, 76% of access is mobile only, against 22% for desktop/mobile. The town is also not a fan of Apple products, 60% of users use android, while only 33% of users access Facebook through iPhones, 13% lower than the UK Facebook population. Workington’s favourite Indian on Facebook is Masala Corner, their favourite community page is Whats on in Workington and their favourite brewery is Tractor Shed Brewing. The town online is reflective of its offline realities, this has important implications for how parties can campaign using Facebook to the users.
The Facebook users of Workington are perfectly representative of the UK Facebook population’s gender profile, skewed towards female users at 53% and 47% male. However, of note is that the town is also much more active on Facebook than the UK average especially in ad’s clicked, posts liked and comments placed.
Graph 3— Workington town’s Facebook users actions compared to UK average
If one was to be sending political adverts to the town one can be assured that they will be more impactful than normal. Equally using Facebook appears to be a legitimate approach to garner engagement, as Workington residents are more active than your average UK Facebook user.
Within the constituency three parties have so far sent Facebook adverts. The Conservatives tested 1 advert 6 times, each time spending under £100. It was then made live from the 2nd November to the 5th November. £900 was spent reaching 60000+ Facebook users in the constituency. Labour has used local pages to send advertising, with 5 adverts from two pages that spent little reaching 16,000 people. One of these adverts is still running. Finally, the Brexit Party started running an event advert on the 5th November, it so far has only spent £100 and has reached 4000. Table 1 below outlines the party approaches seen in Workington so far.
Table 1 — Workington constituency adverts (excluding test ads)
The adverts sent can be seen below, all bar 2 (both Labour ads) are not longer running, however many more are likely to come soon.
Image 2 — the ads that have run in Workington constituency
So who are the parties trying to reach with these Workington adverts? Well the answer is surprising given the narrative of Workington Man. In reality, when using the targeting parameters of Facebook both Labour and the Conservatives are targeting women over men. Only the Brexit Party’s strategy fits the mould.
Graph 4 — The demographics reached by the different parties in Workington constituency
The Workington Man narrative appears misplaced, as it is Workington women that are most heavily sought after. However is this also the case for other constituencies that also feature voter groups like Workington man? Are the Conservatives that are reliant on winning similar seats also not interested in reaching this supposedly vital voter group? Is the reality in these seats the same as within Workington?
Recently the Conservatives have been spending £27,100 on 30 target seats via a Facebook ad similar to the one seen for Workington. It features the following message:
“Your vote in X will be the difference between a majority government that will Get Brexit Done and another hung Parliament.
It only takes X people to switch their vote in X to elect a Conservative MP that will Back Boris to Get Brexit Done”.
Only the place names and vote count needed are altered, with the numbers subject to controversy over incorrect figures. Below 4 examples can be seen, each ad targets a key marginal seat that the Conservatives think they can win based upon the seats Brexit vote, demographics and previous Conservative support.
Image 3 — Examples of the Tory adverts
The seats in question are mapped out below, mostly they are situated in the north and midlands, alongside some Welsh seats.
Image 4 — Where the adverts were sent
As seen below this spend has led to enormous reach across these areas, with 2,560,000 people receiving the adverts. Some seats saw greater reach, including Penistone and Stocksbridge and Blackpool South, while others such as Bury South and Delyn performed very poorly despite the money spent. The average money spent across these seats was £903 (using Facebook’s lower estimate band) with average reach of 85333 people.
Graph 5 — The spend, location and reach of the adverts
The demographics these adverts were sent to are shown below. Again against what one would expect, all of the adverts sent across these seats were delivered to women ahead of men. This was especially so towards women over 45.
Graph 6 — Who was targeted across the 30 seats
Seats like Wrexham, Wakefield and Newport West saw content more heavily targeted at women, with trends seen less in seats such as Newcastle-under-Lyme, Coventry South and Bishop Auckland. Only two seats saw the content targeted at the under 25’s, Derby North and Bury South. The targeting of younger people in Chris Williamson’s Derby North alongside Ivan Lewis’s seat of Bury South may be because of the previous incumbents’ nefarious activities leading to their suspensions from the Labour Party. As such the Conservatives see a wider opportunity to influence younger demographics in these seats.
Graph 7 — The gender of those targeted across the 30 seats
Graph 8 — Overall targeting of women vs. men across age groups
So why are the Conservatives, whose own think tank Onward promoted to them the importance of targeting Workington Man, actually targeting Workinton Woman? A possible answer lies in a Survation constituency poll undertaken in Workington from the 30th-31st of October. Of the 506 people polled the results were as follows.
Graph 9 — Survation Poll of 506 residents in Workington
One must always take a pinch of salt with constituency polls due to difficulties in socio-demographically accurate recruitment and small sample sizes. However, overall the Conservatives feature a 9% lead over Labour, this looks incredibly healthy until you examine the polling by gender. As the following data are sub-samples there is a much larger likelihood of a higher margin of error. However, what the data shows is that the Conservative’s have a Workington Woman problem.
Graph 10 — Survation poll broken down by gender (258 Women / 248 Men)
This is problematic for them as it is these voters who eventually swung back behind Labour, as well as many don’t knows also breaking for the party at the 2017 General Election. This gave Labour a stronger result than expected and meant the optimistic targets set by Tory HQ were not met.
So what does this mean? Firstly it is clear that the real battle across many of these seats is for Workington Woman not Workington Man. Secondly, this is a battle that is being fought across gender demographics, the archaic soundbite nature of “Workington Man”, like “Mondeo Man” before it does not align with political reality even at the base level. Instead campaigns are much more complex, and although overall trends can be discerned, such as the targeting of women in these 30 target seats, the limited data Facebook offers to us may hide the real parameters determining approach. As such it is vital we force Facebook to greater transparency so we can appreciate how we the voters are being campaigned to.