Have election boycotts ever successfully undermined the results?

Question

In recent news in Australia, the Government is planning a postal survey of the electorate's attitudes to Same-Sex Marriage. Attempts to make it a formal plebiscite failed.

There are people with objections to the process, claiming it is unconstitutional, unfair, unnecessarily expensive and/or divisive. Some of those people are calling for a boycott.

I am dubious that a boycott, even if well-followed, would successfully subvert the poll. I am wondering if such a boycott has ever been successful in undermining such a result.

So my question is about political history:

Has there ever been a successful boycott of a major election/vote?

• By major, I mean, say, 50,000 voters or more. Happy to hear about anything close.
• By successful, I mean undermined the public trust of the results sufficiently so the winner could not successfully claim a mandate, take leadership, pass the law - whatever the vote was about.
• Given this isn't a formal election, but a postal survey, I'll accept anything close to an election.

I want to understand if elector boycotts are a powerful tool of protest, or just wishful thinking.

Answer 1

In the United States (including territories), Puerto Rico recently held a referendum about officially declaring for statehood. 97% of the votes were to become a state. But turnout was only 23%, as most of the population boycotted the vote; essentially the only group that didn't call for a boycott were those who wanted to become a state. So the vote was entirely a reflection of that one opinion, and not the clear will of the people.

Note, however, that the referendum was intrinsically non-binding. Only the US congress can formally declare a new state. It was meant to get such a process rolling by making the interests of the populace clear, though Congress was well aware of this movement and had previously allocated funds for the referendum to be held. There has been no official response from the US government concerning this referendum.

However this is not the first referendum of this nature, as there was one in 2012 as well. And this one should fit your criteria pretty nicely. This referendum was also heavily criticized by much the same factions (the PPD namely), and they called for people to submit blank or invalid ballots. The 2012 referendum had a high turnout, but the vote ended up with approximately a half million blank ballots submitted. This made the will of the people less than clear. As a result, Congress decided to ignore the vote. Presumably the boycott of the latest referendum will be interpreted in a similar fashion.

Answer 2

Depending on the criteria, the following might apply.

In Italy, a referendum made to cancel/repel a law (or part of it) has a quorum equal to 50%+1 of the voters.

Since WW2, there have been 67 such referenda (link in Italian, as all others that will follow, sorry).

In the table the coloured column indicates if the quorum has been reached or not, green for a valid referendum, red otherwise.

You can see that after 1996 the turnout has been quite low on average, so the referendum of 2005 is not a significant outlier, but I still remember a vast and hard campaign from conservatives (and especially the church) that were calling for abstentionism.

As far as I remember, it was the first time that political entities and parties were openly calling for abstentionism, and this triggered some debate about whether advocating for abstentionism was a constitutional right or not.

In the end "abstentionism won", but with a turnout of 25.5% participation is not significantly different from the referendum of 2003 or 2009, so I'll leave it to you to decide if this example applies to your criteria.

Answer 3

On the 6th of April 2016 there was a referendum in the Netherlands about whether the European Union should sign an Association Agreement with Ukraine or not.

The "Against" front managed to quite succesfully change the tone of the referendum to "Should Ukraine be part of the European Union" and as such the more right-wing parties turned it into a "Are you in favour of expanding the European Union?"-discussion, akin to that in the UK before the Brexit.

Knowing that a lot of people would vote "Against" with that rhetoric, the advice emerged that "For" voters should not go to the polls, since a turnout lower than 30% would make the vote invalid. Eventually this confused a lot of voters, but arguably did convince a good number of "For" voters to stay at home during the vote.

Whether this changed the outcome is debatable, but it did leave quite a few Dutch people disillusioned about the process as it was mired in controversy and confusion and it's still highly debatable if the vote served its intended purpose, as the matter that actually ended up being discussed had little to do with what the vote was actually about.

In the end a turnout of 32% was reached and the "Against" front emerged victorious (even though he resolution was signed anyway, since the vote wasn't binding, and most other EU members had already signed it).

More information on this particular matter can be found here.

Answer 4

Another example of a boycott undermining a referendum result happened in Romania in 2012, when the President has been suspended by the Parliament. This required a referendum to validate the decision.

More than 8 million persons out of more than 18 million registered voters voted then, so this obeys your major criteria.

The results where clearly against the President (more than 80% against), but the voter turnout (46.24%) was less than 50% (this was required by law back in 2012 and this threshold was lowered afterwards or at least it became active after the referendum).

Some argued that asking for boycott was efficient:

Mr Basescu had initially urged Romanians to vote "no" to what he called "a coup", but later asked his supporters to boycott the vote altogether, a stance also adopted by the opposition Liberal Democrats.

This is debatable, but this turnout is clearly smaller than the turnout for the recent presidential elections (> 54%).