Last week, about 55 per cent of voters in Scotland decided that they did not want the country to go independent of the United Kingdom. I predicted this result much earlier, saying that it would be No victory, with between 50-55 per cent of voters deciding to remain in the United Kingdom. I was confident in this result mainly because of what happened 19 years ago in Quebec, and the difficulty of convincing the majority of people to go with a big change.
It was going to be close largely because you need to elect someone willing to push independence as an issue. Any party that wants to break from the nation is going to make it a cornerstone of their policy, so they will need at least the majority of seats in a parliamentary system. However, that's relying on the voters, which is a much smaller subset of the overall population. The voter turnout for a referendum, however, tends to be a much larger proportion than a general election.
So what you have is a significant portion of the population who wants to break away, as well as a portion of the population which is fine with the status quo. The people who want to break away can make enough of an impact to vote in a party that wants to hold a referendum, separate from the larger nation, and become independent. However, now they have to actually hold that referendum, which means they have to convince a big portion of the population, the one that normally doesn't get politically active, that they should go along with the plan to separate.
These people are the big swath of undecided voters. Now, there are going to be undecided voters who like to just not tell people how they're voting because of a desire to be a bit difficult when someone is making up poll results. I am, admittedly, one of those people, but I'm in the minority. The majority of undecided voters are legitimately undecided, they have not made up their mind on the issue. However, when tabulating results, you might as well count the undecided as going with the status quo.
This is because if someone is undecided on a matter like sovereignty, they are choosing between something that has some risks attached and something which they already know. The majority of the time, people are going to consider the risky path before choosing the one they know already, the status quo is low risk. If you're going back and forth up to the election day, chances are very good that you're not going to go with risk. These are also likely the less politically active people, so their main issues are going to be personal, how the relative turmoil will affect their job, or their family, and risk in those areas is unappealing to most.
The problem the Yes side has in every referendum is that they have to convince the risk-averse that their alternative is the best one. Even in cases where the country is extremely badly off, it's still a massive risk to actually choose to break away, and it comes with logistical hurdles, major changes in commerce and business, many big decisions that have to be made and have wide ranging consequences. If life is relatively peaceful, like it is in a place like Quebec or Scotland, the number of people who are going to be unwilling to take a risk increases. If you have it good, you're going to be less willing to make a drastic change, and thus unions are held together.