If there are no fair elections in a country, the opposition may choose to still participate and try to defend its result or to boycott an unfair election altogether. Are election boycotts effective against an authoritarian regime (by 'effective' I mean "effective in expediting a transition to democracy")? What do studies show?
Are election boycotts effective against an authoritarian regime?
What do studies show?
In her PhD dissertation, Protesting the Contest: Election Boycotts around the World, 1990-2002, 2006, Emily Ann Beaulieu1, concludes:
The main themes and findings presented here take us a great distance in understanding the causes and consequences of election boycotts, but they beg the question of how we are to understand countries that hold problematic elections where the incumbent commits fraud and the opposition boycotts. Are such countries democracies? Are they democratizing?
Certainly countries experiencing major election boycotts have not yet attained the “gold standard” of democracy—although even countries who are considered to have reached this standard continue to face problems such as election fraud. And yet, the findings of this project suggest that countries holding fraudulent elections and experiencing major election boycotts are not hopeless autocracies. They are clearly in a state of flux, which is not to say that they are on an automatic path to democracy (or dictatorship) but that they are still engaged in some process of better-defining how their country will be governed to the satisfaction of its citizens. Furthermore, boycotts, although they carry some negative short-term implications, appear to help such countries proceed in a more democratic direction.
For specific effects, see these chapters:
1Doctor Emily Beaulieu Bacchus, Associate Professor in Comparative Politics in the department of Political Science and Director of the International Studies program at the University of Kentucky and author of Electoral Protest and Democracy in the Developing World (Cambridge University Press), 2014.
Short answer: No.
Firstly you would have to achieve sufficient (non-)participation as to put the turnout far beyond some normal variation as to substantially undermine the result. Not least because boycotts can be a tactical way to explain an expected loss while claiming popular support.
Secondly, it has to be effective. The meaning of the turnout will always be contested so a lot would depend on the wider domestic and international backing achieved by the opposition - which is the boycott is probably not the key factor.
The Brookings Institute published a paper on the analysis of 171 boycotts around the world, of which 4% were classified as effective - and then they were accompanied by other factors that may have been more influential.
Brookings Institute - "Threaten but Participate: Why Election Boycotts are a Bad Idea"
It is by its nature difficult to define whether a non-action has causal impact. There are many things to consider, and to my knowledge “election boycott” as a means of exercising democratic freedom, is not considered standard, normal or effective in general.
From a principal point of view, boycotting an election has consequences firstly for actual election turnout. In semi-free autocratic states where elections happen, but fraud is present or strongly suspected, for instance Belarus in their recent election, turnout is nominally high, and ostensibly in favor of the incumbent.
As such, abstaining from the vote itself, without being part of a chain of political action, would not tip the scale other than to legitimize the incumbent, whose job of fabricating a credible result is made easier the fewer instances of actual voting do occur.
The more evidence, the harder to hide the crime.