Hello. Thank you for having me. I regret that I can't be with you in person today. I'm in Washington at the moment.
I've been asked to speak with you about what voters know about referenda and some of the implications for strategies and outcomes. Democracies around the world use referenda to offer legitimacy and elevated legal status to a range of statutory and constitutional proposals. This is similar to what the gentleman from New Brunswick was saying with regard to looking for legitimacy.
That's the starting principle. For people like us, who know a lot about referenda, and for people like you, who have been involved in drafting referenda and thinking about all the possible things that could be in it or might not be in it, and about what language to use to describe it, from that perspective many referenda are very complicated instruments. Yet when the same question is brought to voters, they necessarily see it differently, because we don't ask them to rewrite the proposal. By the time we bring it to them, the question is simply “yes” or “no”. Not surprisingly, they're going to think about referenda differently from us.
What I'd like to do is review a couple of basic facts about what people know about referenda. There's a question about whether citizens are competent to make this type of decision, so I'll give you an argument for and against.
The argument against it is that referenda typically deal with complex topics to which many voters pay little or no attention. Moreover, when you run surveys about referenda, and you ask people what seem to be basic questions about the content, it's often the case that voters answer the questions incorrectly. From that perspective it looks like they're not qualified to vote in a referendum.
However, there's also evidence in favour of them voting competently. Let me first say what I might mean by “competence”. What I'm talking about is a voter who casts a vote that's consistent with a set of facts and values that they care about. The values may pertain to the life they want for their family, their community, or their nation; and people in different situations might vote differently. By competence I mean the vote that someone would cast if they knew a lot about the referendum in question. Typically they don't, so the question is whether they can still vote competently. In many cases the answer is that they can. The reason they can do this is, again, they only have two choices, “yes” or “no”. If there was a correct vote for, let's say, a particular voter, and they used a coin to cast a vote, they would get the correct answer 50% of the time.
There are a number of situations where voters can do better than that. The way they do is to look for simple environmental cues, which in referenda often come in the form of interest group or party endorsements. Suppose you have a well-known entity or person who has a political history. You know their stands, and they come out and say that they're in favour of a change or they're against it. People use those. They calibrate to try to figure out what they would do if they knew as much as that person. If there are well-informed people who share the values of voters, they can use various endorsements to figure out how they would vote if they knew more.
This is controversial, because you might think that voters should still know a lot. However, in the report I sent, there are actually many cases where all of us make what seem like very complicated decisions using a very simple environmental cue. The example used in my report is driving. If you think about a busy intersection in a city at rush hour, there may be four lanes of traffic each way on each road. There can be 150 or 200 cars at the intersection at a time, and the engineering problem is how to get everyone safely through a relatively small space in 90 seconds.
To try to solve that problem from an engineering perspective is very complicated, because you have to think about the speed and acceleration potential of each of the cars. You have to know something about the intention of every driver, and you have to know something about what every driver believes about every other driver. It's a very complicated problem, and yet all of us solve that problem every day with a nearly 100% success rate, because we have a simple environmental cue we can use to make the right decision. We look at the traffic light to determine whether it's red, yellow, or green, and we look at the car in front of us. That's the simple rule we use to make a decision about when to press the accelerator.
In politics, things like political parties and interest group endorsements serve the role of traffic lights. If you know that your interest is aligned with a particular group or individual, and they say, for instance, “I've looked at the proposition, I've looked at the referendum, and for people with our values, this is a good thing”, we've shown that what happens over and over again is that people cast the same votes they would have cast if they had known more.
Of course, there's a downside to this, too, because if you're a voter, and your values don't match the values of people speaking out and giving their point of view, you can get lost. You can also be subject to manipulation. Someone could represent themselves as sharing your values when in fact they don't. This often happens in the form of mailers. A person such as the leader of an interest group wouldn't get up in public and lie about their position, but maybe someone would send out a mailer saying that the Liberals in the country have a particular point of view, when in fact they don't.
To summarize, most people do not obtain detailed information about referenda. Instead they look for interest group or simple cues to tell them how people who have values like theirs are likely to vote.
An additional point I'll make is that if you want to know whether people will take the time to read the fine print of a referendum, most people won't. That's because if you put the referendum online, you're competing with hockey games, Pokémon GO, and cat videos, which many people find very appealing. Very few people in any country will put those things aside to read a piece of legislation. The Internet allows some people to gain more information than has ever been possible about these things, but most citizens have their daily lives. While some are watching hockey games, for others it takes all of their energy through the day to feed their families, take care of elderly parents, or do things for their community. They don't have the time to invest in legislation, so most people in referenda look for these simpler cues to try to figure out what they would do.
A related point is about how people think once they gain this information. In the case of Brexit, there was a sense that there was this intellectual argument about trade liberalization that wasn't part of the campaign. The question is why was that. One of the answers is that for voters, a referendum isn't an intellectual argument. The question for them is quite simple: Is yes better for me or is no better for me? Those are the only two choices they have. You could say there are all these complicated aspects of Brexit that they should have thought about, but they weren't in the legislature. Their choice was very simple, and the only thing they could do in that situation was figure out which would be better for them, yes or no. That was what they could act on. For other reasons, they might have wanted to know more, but for the act of voting competently, that was sufficient.
The last point I'll make is with regard to campaigns. In referendum campaigns, the “no” side has a huge advantage, regardless of the legislation. This is true throughout the world, and you might ask why. It's because with a no campaign, you're running against change, and people don't know what life is going to be like under that change. A typical no campaign is when you think about a worst-case scenario, and you make your whole campaign about that.
With a “yes” campaign, you have to describe this new world and convince people that even though there are scary possibilities, their life is going to be better. I have a statistic that I use just to tell you how skewed this is. In California, where there's a professional referendum industry, and people care about their win-loss records, most people will not touch a yes campaign unless it's polling 70% or more a year in advance of election day. The reason is that people recognize that no campaigns are easier to run. Everybody believes that the yes support will fall over time, the mystery being whether it will be above or below 50% on election day. There are very few cases where support for yes actually goes up during a campaign. It happens, but it's quite rare.
In sum, in many cases, if voters have clear interest group endorsements, they can make the same choices they would have made had they known more. As a general matter, though, many times they learn that change is scary, so if they're confused, you see more support for the status quo than you might expect.