In political science, legitimacy is the right and acceptance of an authority, usually a governing law or a regime, … a system of government, … without which a government will suffer legislative deadlock(s) and collapse. … Unpopular régimes survive because they are considered legitimate by a small, influential élite. …
In moral philosophy, the term legitimacy is often positively interpreted as the normative status conferred by a governed people upon their governors’ institutions, offices, and actions, based upon the belief that their government’s actions are appropriate uses of power by a legally constituted government. (More)
Legitimacy is a common belief among the governed that they prefer their current system of government to possible alternatives. This is widely seen as a good thing, and in its absence many say that violent revolt or foreign influence is justified. So you might think that regimes would be eager to show their legitimacy to those they govern, and to the world.
Now the absence of a recent violent revolt is evidence for some degree of legitimacy. But let us define the degree of legitimacy of a regime as the cost that its governed would be willing to pay to keep that regime from changing. In this case, the absence of recent revolt only places a rather low and negative lower bound on the degree of legitimacy. So you might think regimes would be eager to show much higher degrees of legitimacy. Perhaps even positive degrees.
A second way to show legitimacy is to offer an official way to change the system. Many regimes have a constitution that can in principle be changed if enough people lobby hard and long enough to trigger the various official acts required by that constitution to effect change. But while this sets a higher (negative) lower bound than does the absence of revolt, honestly it isn’t usually that much higher. The governed could still strongly prefer an alternative system of government, and yet not care enough to coordinate to sufficiently push the usual constitutional process.
A third way to show legitimacy is to advertise the results of polls of the governed on the topic. But not only are such polls almost never done, observers can reasonably question their neutrality and relevance. Who is trusted to do them, and how well do citizen responses to random questions on the subject out of the blue indicate what they’d say if they thought about the topic more?
Regular referenda seem like a more informative approach. Hold elections at standard intervals wherein the governed is asked to endorse either the status quo or change. (In the system, not the people.) In this case, discussion leading up to the election could induce more thought, and give change advocates a better chance to make their case and persuade voters.
Voters might be asked to pick one of several directions of change, or they might just initiate a process that will soon generate more concrete alternatives and then offer them to the electorate. I’m sure that a lot could be said about the best way to run such referenda, but for today my focus is on the fact that almost no regimes ever hold such referenda. Not even bad ones intended to prevent regime change and produce the appearance of more legitimacy than actually exists.
Regimes the world over give lip service to the idea of regime legitimacy, saying both that it is important for regimes to have high legitimacy, and claiming that they in particular have high legitimacy. Yet in fact the most that regimes usually do is to include in their constitutions very slow difficult processes for regime change, processes that are rarely ever actually invoked. Regimes point to that plus the lack of recent revolts as sufficient evidence of their legitimacy. They do not institute regular legitimacy referenda.
Of course most ordinary people are not very upset about this fact. If they were to demand such referenda, then politicians might run on platforms which support them, and they might happen. Yet if asked these same ordinary people would also probably claim that it is important for regimes to have high legitimacy. Especially their own. It seems that both the governed and their governors pretend to care more about legitimacy than they do.
Part of the problem is one feature people often believe is desirable for a political system to be resistant to sudden unconsidered change. Indeed, I think that legitimacy (say as measured by just asking a poll) of the US governmental system is substantially increased by people's sense that it is very difficult to change certain aspects of the system.
All you need to believe to make this compelling is that either people are vulnerable to fads they would be disabused of on further reflection or (I think you'll find more persuasive) that there is some non-trivial error term in what voters choose.
This ultimately boils down to the point that as a matter of fact (at least in the US with constitution worship) people don't seem to support this kind of system so, on your own definition, it would seem to reduce not increase legitimacy.
Sure, I don't have opinions on what people what to hear, only on what equilibria are likely to be selected for.
I mostly believe that this level of signaling happens primarily on non-legible dimensions, and neither party is clear on what they actually want.