What are the reasons for not having a voter turnout threshold for a nation-wide referendum?

Question

Romania held a referendum that failed to be validated due to low voter turnout and the 30% threshold limit (this was lowered from 50% a few years ago). The leader of the main party in power thinks that not having a threshold would be a good idea, but this seems to just be a justification of his party failure.

It is not clear if having a voter turnout for referendums is more frequent than not having it, but this article provides a few examples of both cases:

(..) the Brexit vote (for which no minimum turnout was specified) would have passed, as 37% of the British electorate voted to leave the European Union. However, the UK vote is considered valid while the referendums in Romania and Hungary are not.

For me having not having a threshold for referendums seems a little strange, as referendums can have serious consequences (e.g. Brexit) and a minimum legitimacy level seems natural.

Question: What are the reasons for not having a voter turnout threshold for a nation-wide referendum?

Answer 1

The goal of the Brexit vote was to legitimize David Cameron's position that the Conservatives (Tories) should be against Brexit. He offered that vote to stem defections to Labour and UKIP. If he had made the vote threshold arbitrarily high, it would have been seen as cheating and would have failed to legitimize his position. He needed a majority of the vote for the legitimacy he wanted.

The truth is that Brexit passed not because of that referendum, which was non-binding, but because Parliament passed it. If voters had not wanted Brexit, they could have supported one of the two anti-Brexit parties, the Liberal Democrats and the Scottish National Party. Instead, they voted overwhelmingly for the pro-Brexit parties, the Conservatives and Labour. The Conservatives' losses were more than offset by Labour's gains and Liberal Democrat gains were more than offset by SNP losses. The 2017 election reconfirmed Brexit.

Due to parliamentary sovereignty, elections for Parliament are effectively what matters in the UK. The reason why they did not block Brexit is that if they had tried, they lacked the votes. Many Members of Parliament would have voted for Brexit.

In Romania as I understand it, referendums are actually binding. Therefore it makes sense for them to be given explicit thresholds. If the UK actually had a binding referendum mechanism, it would make sense to have a threshold there as well. But they don't. And it would have been actively counterproductive for them to have tried to introduce one just for the non-binding Brexit vote, as it would have reduced the legitimacy of their planned win. It would have been much simpler to have just ignored the referendum, except that they lacked the votes.

TL;DR: the Brexit "referendum" was more of an official poll than a real referendum. In and of itself, it had no power. It was given power by Parliament respecting that decision. And there are more pro-Brexit MPs than not. Otherwise it would not have passed.

Answer 2

One reason not to have a voter turnout threshold is that generating interest on non-controversial issues can be hard.

For example, suppose that there is 98% support for a measure to remove the king from the constitution of a nation after the king abdicates his throne with no successor in place. But, since it is uncontroversial, only 10% of the population votes. If there was a higher than 10% threshold, this uncontroversial measure might not pass, even though all proper processes were followed.

Another possibility is that most people don't care, even though some do.

For example, suppose that there is a measure to determine if the nation should transfer a small, uninhabited island to another nation, and 90% of the population doesn't care, but 10% of the population is hotly divided and cares a lot. A referendum allows the decision to be made in a manner that doesn't taint the eyes of sitting politicians for the 10% of the population that cares a lot in a definitive way that may require a referendum for constitutional reasons. A threshold leaves an issue of importance for some purposes unresolved.

More generally, referendums are usually held because some issue needs at definitive resolution that, for whatever reason, ordinary measures don't resolve. Often getting a definitive resolution is more important then how it is resolved. A threshold requirement thwarts this purpose.

A third possibility is to discourage the political tactic of a referendum boycott.

A threshold allows opponents of a measure that has majority support among people who would usually vote into a defeat, by not voting at all, rather than showing up and voting against a measure, in a country where voter turnout isn't all that high relative to the threshold.

Suppose that the threshold is 30% and typical voter turnout is 40%. If 25% of people support a measure, 15% of people oppose it, and 60% of people aren't bothered to vote, the opponents can win by boycotting even though they couldn't win by voting. And, you generally want to design a system in which the incentive in the political system is to participate, rather than to boycott, an election.

The last problem could be cured with a threshold of affirmative votes for the measure only, rather than a threshold of total votes, and a requirement that a majority of votes cast support the measure. This is a system that is used in the U.S. in many elections of workers on the question of whether a workplace should be unionized.

For example, if the threshold is 25% of eligible voters voting in favor, and a majority of the votes cast also being in favor, boycotting won't work, but decisions of importance can't be made via referendums unless there is a decent affirmative base of support for the measure.

Answer 3

There's a mathematical logic at work. Let's say it's an election cycle with low turnout. (In the USA at least, voter turnout is very high every fourth year in November, when there's a Presidential election, lower on alternating even-numbered years, when there's an election for a Senator or Governor in most states, lower still on odd-numbered years, and yet lower on local or special elections that aren't held in November.) For our example, let's say that 40% of voters have turned out to vote.

Let's say the referendum is popular, and is likely to succeed with 21 "yea" to every 19 "nay" votes. That means 52.5% of the voters who turned out are voting "yea".

Let's say there's a threshold that 30% of eligible voters have to participate.

Here's the problem: the minority of "nay" voters can win the referendum by abstaining from the vote. That would lower the participation level to 21%, and the referendum would fail. This despite the fact that turnout is normal for that year, and the fact that "yea" on the referendum is indisputably the choice of the voters.

Logically, your requirement that 30% of voters turn out in fact requires that 30% of all eligible voters turn out and vote yes. If the referendum is a close one and most people are 50/50, that means in practice you need extraordinary turnout of 60% to even have a fair referendum at all.