A group of researchers have conducted some experiments to find out. Initially they ran their experiment in the US, but they then expanded the scope by crossing into other nations such as the UK, Belgium, and Spain to see if they obtained similar results.
What exactly did they do?
The study paper sums it up as follows …
Using evidence from Great Britain, the United States, Belgium and Spain, it is demonstrated in this article that in integrated and divided nations alike, citizens are more strongly attached to political parties than to the social groups that the parties represent.
What they actually did was to use games to measure trust. Their trust game consisted of a two player game designed to measure how confident players felt about the other player and what degree of trust they had for their opponent.
- Players are given some cash
- They are told that they can give ‘some, all or none’ of the money to a second player and they are also of course shown a profile that describes the other player.
- They are told that the researchers will triple any amount given by Player 1 to Player 2
- Player 2 could, at his/her discretion, return some, all, or none of the money back to Player 1
The more Player 1 trusts or expects reciprocity from Player 2, the more Player 1 should allocate to Player 2.
So what happened?
In the US version the Player 2 profiles that got the least amount of cash given to them consisted of profiles that represented a different political tribe (democrat or republican). Interactions with people of a different race did not have the same impact when they shared a vote for the same political tribe.
They concluded that politics was the more potent force for generating division.
Having completed the study within the US they decided to widen the scope to other nations to see if they got a different result.
- In the UK the parameters they played with compared political ideology to religion
- In Belgium the comparison was between politics and people coming from a different region
- In Spain, the Basque region specifically, the comparison was between political ideology and ethnic origin
In every case, political divisions were demonstrated to be the most potent divider that generated the least trust.
Britons demonstrated a significant degree of discrimination against opposing partisans in the trust game.
The pattern of discrimination was not symmetric; prejudice against partisan opponents far exceeded in-group favouritism. The out-group penalty proved dramatically larger than the in-group bonus.
The United States also represented a case of relatively weak social divides, with the exception of race; however, unlike the United Kingdom, there is very strong political polarisation.
They found that American partisans, like their counterparts in Britain, demonstrate animosity toward their opponents; however, they are more generous with co-partisans.
This is a divided society with three very distinct regions. Each region has traditional left/right political party divides, and no party exists in all regions. Naturally they explored interregional divisions …
Our design consists of intraregional comparisons of political parties of differing ideologies and interregional comparisons of political parties of both similar and dissimilar ideologies, allowing us to demonstrate that the party divide is more powerful than the regional divide.
Belgians consistently behaved generously toward co-partisans and discriminated against members of other parties. Belgian participants also treated co-partisans preferentially, to a degree greater than that observed in the United States.
What is truly fascinating here is that Leftists give more to fellow leftists, and rightists give more to fellow rightists, regardless of Player 2’s regional/ethnic identity.
Spain – Basque region
The big political issue is those who strive for a united Spain vs those who want an independent Basque region. Obviously the traditional left/right divide also exists.
Despite the history of conflict between ethnic Basques and ethnic Spaniards, it is partisanship rather than ethnicity that exerts the largest effect on trust. For all four parties represented in the study, respondents treated a Basque co-partisan no differently than a Spanish co-partisan. Partisanship and agreement on devolution contribute to interpersonal trust in the Basque Country far more than does ethnicity.
on average, we find higher levels of partyism; in absolute terms, the number of people willing to discriminate on party lines is larger than those willing to discriminate on religious/ethnic/racial lines. This does not mean that party extremists are more violent or dangerous than religious/ethnic/racial extremists. Nevertheless, partisanship is a source of measurable bias in the behaviour of citizens.
Why is it like this?
You might perhaps wonder what is going on here. The answer is apparently quite simple.
Social pressure exists to reduce specific types of discrimination such as racial or gender discrimination, and so we observe an ongoing reduction of discrimination on that basis.
As a contrast, there is no social pressure to discriminate on the basis of political ideology, hence such divisions continue to not only thrive but to also increase.
The study itself notes this …
unlike race, gender and other social divides where group-related attitudes and behaviours are constrained by social norms, there are no corresponding pressures to moderate disapproval of political opponents. In fact, the rhetoric and behaviour of party leaders suggests to voters that it is perfectly acceptable to treat opponents with disdain. In this sense, individuals have greater freedom to discriminate against out-party supporters.
One other interesting observation here is that unlike your ethnic origins or culturally inherited religious beliefs, individuals tend to be responsible for their political allegiances. It is very much an expression of who they are and so it is a lot easier to hold individuals accountable for such choices, hence such discrimination is generally socially acceptable.
It is not definitive
It is however worth noting that they explored using a very specific set of parameters. I do wonder what might happen if they used such games to measure other aspects such as Atheism. Does a believer being faced with an out atheist who is also a member of their own political tribe invoke even less trust that a believer from another political tribe?
It is not simple and may be quite complicated.
You can test your own bias.
Think about the specific political ideology you identify with. Now think about those that represent the other alternative ideology. Do you find within your heart that you quite literally hate and loath “them”?
It might indeed be justifiable due to the utterly appalling moral stance they take. Clearly, as illustrated by the cartoon at the top you are wholly aware of my specific bias, and you might or might not agree. It is worth remembering that they truly do hate and loath us as well.
I, like many others, am a product of the age we now live within. Such feelings about political opponents are encouraged, nurtured, and are to some degree socially acceptable. If you expressed similar feelings in a racial or even a religious context, then there would be social pressure that frowns upon that, and so it is political divisions that do very strongly divide us in a manner that is accepted by society.
The paper itself can be found here – The tie that divides: Cross-national evidence of the primacy of partyism.
(Unfortunately there is a paywall for the above).