Some lessons from the political campaigning literature

At the event in Edinburgh recently, I was asked a question about polling. The question was along the lines of: if the Scottish people had overwhelmingly voted in the June 2016 referendum to Remain in the EU (62 per cent of the 67.2 per cent of eligible voters who voted) then why should the activists seeking independence not endorse joining the EU. Apart from the obvious reasons relating to the concept ‘independence’ and wanting to avoid membership of a neoliberal cabal, I replied by noting that if we conducted a poll about whether people thought taxes funded government spending, then we would find a much larger percentage agreeing with that proposition that the proportion that voted to remain in the 2016 Referendum. I then asked the audience: Would you consider that outcome legitimate or a symptom of a lack of education? The point is obvious. Polls play on ignorance as much as anything. The question of campaigning and polls also came up during the recent Australian federal election, where despite millions being spent on targetted advertising and activism, the results turned out very different to those expected and in most cases the dollars spent were largely ineffective (although note below). Further, there is a growing number of Modern Monetary Theory (MMT) groups forming around the world aiming to self-educate and push the public debate away from the mainstream economic narrative. The question that arises in each of these instances is how to actually push a new paradigm, a new way of thinking about concepts that permeate the very basis of our daily existence and have been ingrained in our perception in a particular way that this new way contests. That is no easy task. I have been doing some research and will report on the results in a series of blog posts starting today.

In one sense, to see these activist groups at work is heartening – when have ordinary citizens given their non-work time to get together and learn macroeconomics, stage events about macroeconomics and form groups to further educate their peers about macroeconomics?

That, in itself, is a pretty incredible part of the upsurge in community MMT action.

As an example, which I have some close connections with, is the great work being done in the UK by the GIMMS team – where a group of women without any particular background in economics have become organisers, writers, film producers, social media activists and lobbyists.

I was really pleased to participate and offer my thoughts on how they can advance their MMT work at a ‘Training the trainers’ workshop in London on May 14, 2019. A video will be available soon of that session.

On the weekend just gone I attended a newly-formed MMT discussion group in Melbourne and a good number of people of all ages gave up their afternoon to learn about and discuss MMT.

To help these groups I have been upskilling myself by studying the research literature on campaigning strategies.

I also have personal misgivings about associating our work with specific campaigning groups in Australia, which run the danger of stereotyping MMT as a fringe, Left idea and closing doors to other groups I am currently working with in the financial markets who are keen to learn more and who have, without doubt, access and influence at the top levels of the political process.

It is not an easy way forward. There is, in fact, no clear blueprint.

No other economic theory or framework has emerged and been developed as a body of work in the academy and then gained support in the general public largely via social media.

And moreover, that is in the context where MMT has not yet gained a dominant position in the academic hierarchy. That process takes a long time, given that PhD students have to be nurtured, gain jobs, and progress in their own careers up the academic hierarchy.

So we are all feeling our way in this endeavour which makes it interesting but also fraught with dead-ends and more.

How to progress?

Well if you are like me, then when in doubt, you do some research – that is, educate (in this case myself). As I will conclude in today’s blog post – education remains the key and is more important than short-term campaigning and polling.

This conclusion is supported by the research literature in the area of political engagement.

That literature has yielded some interesting results, which in the first instance surprised me, but then helped me understand why the recent federal election outcome in Australia, for example, was perverse when judged against what the opinion polls and campaigners were conditioning us to believe over the last few years.

A fairly definitive 2018 study of the campaign effectiveness – The Minimal Persuasive Effects of Campaign Contact in General Elections: Evidence from 49 Field Experiments – was published in the American Political Science Review by Joshua Kalla and David Broockman.

(The full reference is Kalla, J. and Broockman, D.E. (2017) ‘The Minimal Persuasive Effects of Campaign Contact in General Elections: Evidence from 49 Field Experiments’, American Political Science Review, 112(1), 148-166).

They sought to investigate whether political choices in the US can be manipulated by campaign advertising and direct-contact methods pursued by so-called “political elites”.

It is accepted that “framing effects” can influence choice, but the study wanted to focus on the usual campaigning tools deployed by political parties – direct canvassing etc.

They analysed “49 field experiments on the persuasive effects of campaign contact and advertising” which “rigorously estimate the effects of real campaigns’ choices about which voters to persuade and how to persuade them in the context of real elections”.

I won’t go into their research design. You can read it if interested.

The extant research provides a good background to the study.

The literature tells us that “Americans’ political predispositions … [are] … highly durable and resistant to change”. I suspect the study has applicability outside of the US because it incorporates an understanding of framing and language, the influences of which are not US-centric.

In part, this is because the frames that people operate in when making decisions are fairly set. Confronting a person in this state with new information doesn’t really change much – they have all the “information, they will have chosen to retain”.

Further, typically, there is not a significant proportion of voters available for “meaningful persuasion”. This is because people have fairly rusted on beliefs when it comes to party identification and accept what the political machines tell them about a candidate regardless of what other sources of information might be suggesting.

There is some suggestion that early on in a cycle, frames can be altered. But once they become set, “providing … additional frames” or information “does little to affect their choices”.

It is clear that political parties think otherwise and “Campaigns spend a great deal of money advertising to voters” and the polling “firms and consultants who profit from these activities argue that their effects are large”.

But the research evidence suggests otherwise.

The study brings together 49 data sets drawn from real world campaigning examples to examine whether spending millions on campaigning is effective or not (and whether the effectiveness might be time-limited).

The results of their research are:

1. “The best estimate for the persuasive effects of campaign contact and advertising – such as mail, phone calls, and canvassing – on Americans’ candidate choices in general elections is zero. Our best guess for online and television advertising is also zero, but there is less evidence on these modes.”

2. “When campaigns contact voters long before election day and measure effects immediately, campaigns often appear to persuade voters. However, this early persuasion decays before election day and the very same treatments usually cease working close to election day.”

3. “We find campaigns are able to have meaningful persuasive effects in primary and ballot measure campaigns, when partisan cues are not present” – in other words, when there is no party identification frames in place.

4. If a politician changes a position on an issue and gets media coverage this can be effective in changing voter support.

5. Campaigning may influence participation (“turnout”) once opinions have been formed.

6. The evidence also allows for the possibility that heavy direct campaigning in the last few months before an election can “hurt candidates by 5.3 points”.

In terms of the recent Australian election, mining magnate Clive Palmer was estimated to have spent $A60 odd million in the recent federal election campaign in Australia, ostensibly to get his own candidates (including himself) elected.

His face was everywhere in the weeks leading up to the voting day – on huge billboards overlooking freeways, on TV, in the print media and more. Robo-SMS messages, blind mobile phone calls and the rest of accompanied his campaign.

The two major parties spent an estimated $A14.5 million (Liberal conservatives) and $A13.3 million (Labor).

Palmer’s outlay is a record in Australian political history by some distance and generated calls about billionaires buying elections. I won’t go into the nuances of allowed campaign expenses in Australia as they are beside the point here.

His campaign was targetted against the Labor Party and he exchanged preferences with the conservative government, which probably helped them retain office against all the polling estimates.

In narrow terms, he didn’t succeed in gaining a single seat and won around 3.4 per cent of the national vote. Even in his Queensland stronghold, he hardly recorded.

Apparently he was holidaying in Fiji during the last week of the campaign – not the typical behaviour of a politicians trying to squeeze out every opportunity to win votes.

But he also knew that if the Labor Party was elected there would probably would have been more climate action (read: increased constraints on coal and other mining). Equally, he knew that the government conservative coalition exhibits climate change denial and the Prime Minister had actually brought a lump of coal to Parliament (while Treasurer).

Palmer is a miner. He stands to gain many multiples of the campaign outlays in mining profits now the climate change deniers have been returned to office.

So we cannot judge the effectiveness of his campaign spending in terms of the success of his own candidates, which is why I used the qualifier ‘ostensibly’ (above).

However, the trade unions and progressive groups (such as GetUp) also spent a fortune using all the techniques that the academic research has found to be largely of zero effect. Most of this campaigning was aimed at electing the Labor Opposition and it failed dramatically.

GetUp campaigned heavily in selected seats with the aim of unseating the incumbent (usually some right-wing troglodyte). They had one success (that is, the incumbent was unseated). But in that case the candidate they promoted is a sound-finance, trendy-elitist (conservative) who happens to mutter a few words about climate change.

In effect, they blew it. The elected candidate will not support any fiscal shifts (surplus obsessed) and therefore to prosecute her climate change ambitions she will be looking for spending cuts elsewhere. That doesn’t sound like a progressive success to me.

The defeated candidate, a former Prime Minister knifed by his own party, which retained government, was clearly a climate-change denier. But it would have been better for him to remain in office (given what has happened) because his corrosive, revenge-seeking behaviour has been creating havoc within the conservative Government – having him in office on the backbench was an asset to progressive causes!

So the millions that they spent campaigning could have been donated to a food or homeless charity with better progressive effect.

It is also highly likely, and this is consistent with the academic research evidence, that the strident campaigning in conservative seats actually reinforced the likelihood of the candidate they were trying to unseat of being elected. One of the chief objects of hate in Australian politics, who was instrumental in promoting the massive government instability last year, was returned with an increased majority and his seat has moved from marginal to safe.

What appears clear is that this group did not advance the progressive cause at all and came out appearing as a marginal (fringe) left-wing group with little influence at all.

Certainly not a group that one would want to align with in the task of promoting Modern Monetary Theory (MMT) to the broader population. It will just reinforce the ‘MMT is crazy kooky stuff’ narrative that our critiques are promoting.

There is another research strand that bears on this topic – the so-called “politically motivated reasoning” studies.

A good introductory paper in this field is from Dan Kahan – “The Politically Motivated Reasoning Paradigm, Part 1: What Politically Motivated Reasoning Is and How to Measure It”.

(Reference: Kahan, D. (2016) ‘The Politically Motivated Reasoning Paradigm, Part 1: What Politically Motivated Reasoning Is and How to Measure It’, Emerging Trends in Social & Behavioral Sciences. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2703011).

We learn that:

1. “motivated cognition refers to the unconscious tendency of individuals to fit their processing of information to conclusions that suit some end or goal” – so we see what we want to see. In general, we do not identify that we are behaving in that way.

2. “Motivated cognition is best understood as a description or characterization of a process and not an explanation in and of itself” – we do not necessarily know what the motivates the “mental processing” or the “precise cognitive mechanism or mechanisms through which it operated to generate the goal-supporting perceptions or beliefs.”

Some things are obvious – such as financial motivations.

But other things are less so – “self-image”, “protect connections with others” etc.

Further the:

… mechanisms are also diverse … biased information search … seeking out (or disproportionally attending to) evidence that is congruent rather than incongruent with the motivating goal; biased assimilation … tendency to credit and discredit evidence selectively in patterns that promote rather than frustrate the goal; identity-protective cognition … the tendency of people to react dismissively to information when accepting it would cause them to experience dissonance or anxiety.

3. “motivated cognition can make us stupid, but it is not a consequence of stupidity” – research has shown that even “expert scientists … conform the performance of their reflective and deliberate evaluations of evidence to the desire they have to see exciting conclusions vindicated and disfavored ones rejected.”

4. “motivated cognition … tends to focus more on the need for maintaining a valued identity, particularly as a member of a group.”

Which then links into the Groupthink literature that I have written a lot about in the past.

This sort of behaviour can drive what Kahan has called “ideological polarization” in politics which is not surprising where values are concerned but make no sense where there are “conflicts over facts that turn on empirical evidence” (Source).

Social psychologists have found that even though the facts should appear obvious, conflict continues.

Part of the story is that:

… members of the public rely on heuristics or mental shortcuts that can generate systematic biases in their perceptions of risk and similar facts. They also tend to seek out and assess evidence in biased patterns that reinforce the positions that they, or those who share their ideological predispositions, already hold. Some psychologists maintain, too, that these effects are intensified by dogmatism, aversion to complexity, and like traits that correlate with political conservativism and that make politically conservative individuals distinctively resistant to revising their beliefs based on empirical evidence.

So there are several things going on.

First, a dominance of “fast, associative” modes of information processing “based on low-effort heuristics” – over “high-effort systematic reasoning”.

So we are dominated by “visceral, emotion-guided modes of perception” which cause us to “overestimate the incidence and harm associated with more sensational risks”.

At the discussion group on Saturday there were people who blurted out the ‘deficit-hyperinflation’ link without considering the complexity of the situation under focus (Zimbabwe!).

We get all upset about “terrorist acts” relative to “climate change” threats.

And the influenced by “experts” is minimised because it requires people “to engage in the more effortful, more dispassionate “System 2″ style of reasoning suited to understanding the technical evidence that experts use”.

If I have 60 minutes I can usually get a person to understand the basics of MMT. But most people demand 2 minute snapshots.

This was raised at the discussion meeting on Saturday. It was stated: We need short, sharp 2 minute videos and primers.

Yet the danger then is that we produce a version of MMT that is so simplistic that we run the danger it just gets converted into the dominant frames (the System 1 reasoning – visceral) and so we get countless newspaper attacks on us claiming MMT is about printing infinite quantities of money so the government can buy whatever it wants.

And at that point we lose the person to the dominant frame.

Second, as noted above – motivated reasoning generates systematic bias.

When the dominant players in a group hold a particular view it is difficult for an individual to break out of that paradigm.

When I was a graduate student, a senior professor took me aside and advised me to leave the program and take up sociology because I didn’t have a career in economics with the views that I had!

Community membership is a massive conformity mechanism.

People engage in “identity self-defense” to resist empirical facts that “run contrary to the dominant belief within their groups”.

So trying to start an engagement process with a ‘fact’ war – look at this fact, or that shocking fact – will not get us very far because unless the frame changes the fact will be either rejected or reinterpreted to suit the dominant frame.

For example, we might say to a group: Look at the massive rise in unemployment to 12 per cent, the government has to do something about it.

Some will deny the figure and question the basis of the estimate and even claim the statistical agency is a stooge designed to advance specific agendas.

Others will just say something like – yeh, it is disgusting that there is that much laziness in the society and the government is not helping by subsidising it with income support.

Third, there is an “association between ideological or cultural values and cognitive-reasoning styles”.

The research evidence shows that:

… right-wing ideology is a manifestation of settled intellectual traits such as dogmatism, aversion to complexity, and a craving for certainty or “closure” in argumentation. The cognitive style that comprises these dispositions, it is surmised, generates reflexive closed-mindedness toward empirical evidence hostile to the factual premises of policies that reflect ideologically conservative values or policy preferences …

That means that progressives have to be clever in how they approach debates about matters that the right-wing ideology promote – such as mainstream macroeconomics.

These debates will typically be conducted on System 1 type reasoning (visceral) which is resistant to logic and complexity.

It is hard to engage in that sort of environment.

Last week, there was an very unedifying Twitter conversation that I was unfortunately copied into. I contributed two Tweets to a very long dialogue, primarily to correct the public record rather than convince the principle protagonist who hated MMT and the Job Guarantee.

But it was clear the person, who had strident views, had read very little of the literature and proudly stated he didn’t need to read it because it was just garbage.

One cannot progress when confronted with that situation.

Trying to work out which of these three drivers of fact conflict are dominant is also essential for working out a strategy.

For example, if “identity-protective motivated reasoning” was dominant, then we know the facts will be interpreted according to the dominant frame.

So I give examples of the way I talk about the Job Guarantee depends on the audience.

For example, if I am in a group where individualist, corporatist values might be dominant I will never start talking about equity, opportunity, fairness or human rights.

I might introduce the discussion of the Job Guarantee in terms of asking which sector the unemployed are currently in.

I usually get quizzical looks. But soon enough I can get the group to understand that the unemployed are in the public sector because the private sector doesn’t currently want them.

Next question: And what are they doing in the public sector?

Answer eventually comes: Not much.

Question: Would it not be better that they were doing something for society?

Answer: resounding agreement.

And at that stage, the principle is established and I can usually get that group to conclude that a Job Guarantee is superior to mass unemployment without talking at all about the terrible social costs of unemployment and all the rest of the progressive reasons why we should introduce a Job Guarantee.

From the framing literature perspective, we should not be surprised by these results.

Please see the following blog posts (among others) for some background:

1. The ‘truth sandwich’ and the impacts of neoliberalism (June 19, 2018).

2. Correcting political ignorance and misperceptions (August 5, 2015).

As I have noted previously, if we attack a dominant frame using the language and concepts of that frame, it is highly likely we will just reinforce the frame.

Conclusion

This research is on-going and I will report further when I have digested the next tranche of articles.

It certainly points to education being the key – we have to alter the frames first and not get involved in ‘fact wars’.

We have to create new language because we know that language triggers frames.

We have to try to understand the way groups process information (System 1 or 2) and approach them accordingly.

What is likely is that spending millions berating people about a coal mine or blasting them with facts will have zero impact.

That is enough for today!

(c) Copyright 2019 William Mitchell. All Rights Reserved.