Referendums and Initiatives Can Hurt Democracy, Unless We Rethink How We Use Them

Opinion by Samuel Marks

Internationally, democratic satisfaction is on the decline—especially in well-established and long-standing democracies (LeDuc et al. 2002). This is to be expected after a period of global democratization. It is partly due to the increasing rise of “audience democracy,” in which parliament, parties, and the media are on the decline in the face of a disinterested public, leaving all involved malcontented (Urbaniti 2019). Citizens believe periodically sparse elections that elect increasingly unresponsive representatives are insufficient. In Europe, this phenomenon is further amplified, with the European Union creating a feeling of a democratic deficit as individual governments are losing power and sovereignty to a centralized and purportedly uncaring technocracy (LeDuc et al. 2002). The international plan to increase democracy has seemingly been the introduction of direct democratic items, such as referendums and initiatives, especially in cases of contentious policies or fundamental political changes (LeDuc et al. 2002). Some believe that by combining aspects of direct and representative democracy, democracy as a whole improves. However, direct democratic components can be dangerous to democracy in all but a few cases. 

Prior to delineating the potential dangers of direct democratic concepts such as referendums and initiatives, it is pertinent to define them. Although broadly used to describe all direct democratic measures, referendums are specifically a vote put forth by a governing party to the people, which might be binding, and, if so, authorized. Initiatives are votes facilitated through citizen petitions, which is why they are sometimes referred to as “citizen-initiated referendums” (LeDuc et al. 2002). The use of direct democratic items varies wildly internationally, with some using them extensively (such as Switzerland or Italy), others moderately/specifically (such as Brazil and Canada), a few subnationally (such as the United States, where individual states like California use them), and some not at all (such as Germany or India) (LeDuc et al. 2002).

The first issue with referendums and initiatives is their general complexity compared to “regular” democratic tasks and/or requirements of the “average” citizen. To most “average” citizens, voting in a yearly election is all they believe is asked of them to keep democracies afloat. Although this is normatively incorrect, it is the consensus of most people. The issue here rests on the concept that the “average” citizen, which constitutes a majority of people, is relatively uninformed in politics and policy. This issue tends to be circumvented by the fact that political parties act as informational shortcuts or cues for voters to allow for ease in elections (LeDuc et al. 2002). For example, in an election, a voter may not know the actual candidates but is familiar with the candidate’s party, thus voting with greater confidence. The problem with this system is that there are no informational shortcuts or cues on a referendum or initiative ballot, as neither parties nor candidates are present. Therefore, on what could be an extremely complex issue, there is no indication of what the “right” or “good” thing to do is for the individual voter. Even in cases with highly partisan matters, the lack of party cues leaves voters extremely confused (Hobolt 2006; LeDuc et al. 2002). California presents a possible solution, offering a condensed legislative debate on each direct democratic issue (for and against, with detailed arguments and analysis for both) for voters to review before casting their ballot (Bucy 2018). One issue with this system is that it forces the voters to become more well-informed without shortcuts or cues, which is a challenge on its own.

The lack of parties’ informational shortcuts and cues for the uninformed voter increases the importance of the subject’s publicity campaign. The problem with this added emphasis on the campaign is that there are no means of accountability, as there are in representative elections. For example, in an electoral campaign, if the promises the candidate/party makes are not actuated, they can be removed from office electorally in the next election, i.e. held accountable (Isiksel 2016). In direct democratic measures such as referendums and initiatives, however, once passed (even if not necessarily binding), it is seemingly irrevocable and can facilitate unforeseen issues. During the British referendum on whether or not to leave the European Union (known as Brexit), the campaign was full of disinformation and inflammatory remarks, which helped lead to the success of the “leave” vote. Since the campaign organizers of the “leave” vote knew no one could envision what leaving would look like, they could openly lie and spread misinformation to actuate their desires and then walk them back after the referendum passed (which is exactly what they did) (Bucy 2018). 

Furthermore, these campaigns can, and tend to be, dominated by populists—a danger in itself. Populism is defined as a “thin-centered ideology that considers society to be ultimately separated into two homogenous and antagonistic groups: ‘the pure people’ and ‘the corrupt elite,’ and argues that politics should be an expression of the general will of the people” (Mudde 2015). The dangers of populism in a referendum is that it can mobilize an uniformed mass to make a decision that may be normatively lamentable, for both society and democracy, without truly comprehending the damage they are inflicting. This is why the political class tends to be skeptical and challenge the democratic efficacy of referendums and initiatives (Schuck and de Vreese 2015). Furthermore, stealth democrats who distrust the political elites and have no interest in politics at all are typically supportive of direct democratic measures. Stealth democrats often directly support populists who they view as political outsiders to create “unfiltered” political results (Schuck and de Vreese 2015). This is compounded by the fact that referendums and initiatives are heavily influenced by short-term variables, such as the campaign itself, but also the current state of the economy, scandal, etc. (LeDuc et al. 2002).

As aforementioned, another issue with direct democratic measures such as referendums and initiatives is their inflexibility. Referendums and initiatives have no means of specifying alternatives, weighing reasons, ranking preferences, and/or indicating the costs of trade-offs (and finding those that are tolerable) (Isiksel 2016). In contrast, elected representatives have to be flexible when deciding policy and weigh the costs and benefits while adjusting accordingly. This allows for maneuverability in power and accountability in the future, neither of which exist in direct democratic concepts. Therefore, referendums and initiatives break the feedback loop that ensure true democratic strength (Isiksel 2016).

The final issue with direct democratic items such as referendums and initiatives is their treatment as “second-order” elections. The reason they are referred to as “second-order” elections is because their voter turnout tends to be far lower than that of a regular election. The results are less of a reliable measure of the general public’s actual desires, with possibly only a minority of citizens voting on a matter that controls the majority (LeDuc et al. 2002).

Potentially Beneficial Referendums

Despite these dangers and possible failures of referendums and initiatives, there are a few specific cases where they might be beneficial. When a majority or governing party is internally divided on a relatively low-scale issue, having the public decide helps to save face (Robin Best, personal communication, May 2022). This tends to be true of issues such as the legalization of cannabis for recreational use. Referendums can help settle this issue and do not require parties to endorse the measure, as there is still some taboo around cannabis legalization (Robin Best, personal communication, May 2022). 

Another potentially beneficial use for referendums is in determining whether to enfranchise long-term, non-citizen residents in local elections. By holding a local referendum (city-wide, county-wide, etc.), the people of a small area can determine if they desire to allow non-citizen residents to have the ability to affect low-scale politics. Tangential to elections, a referendum could be used to enact a compulsory voting rule as a type of election reform. In such a referendum, citizens would determine whether or not they feel their fellow citizens should or should not be forced to be politically involved.

Another possible but far less preferable case would be to use a referendum when there is gridlock over an extremely partisan issue, thus leaving the public to decide. However, since the matter is already clearly partisan and the parties have a clear position on it, the selection of a representative to deliberate on such matters should already be enough (Robin Best, personal communication, May 2022). 

Direct democratic components like referendums and initiatives can be harmful and/or dangerous to democracy, except in rare cases. Their complexity, inflexibility, lack of accountability, possibilities for populism, and relative indifference illustrated by the population, put direct democratic components below representative democracy in the toolbox for democratic governance. In the case that a direct democratic measure is utilized, the outcome should be non-binding and act as recommendations to elected representatives to prevent the aforementioned pitfalls.

Samuel Marks is a senior political science major from Poughkeepsie, NY. He is planning to get his Master’s in Public Administration and a Juris Doctor degree. Sam has previously written on politics in the past and has had papers published. He likes to run, watch the Mets and Jets, and anime. Sam also has a unique upbringing, as he grew up in Asia for 13 years, which gives him a unique insight into the global political sphere.


Bucy, Erik P. 2018. “The Brexit mess shows how the UK’s referendum process could learn from California’s ballot initiatives.” LSE Blog, December 6.

Hobolt, Sara B. 2006. “How Parties Affect Vote Choice in European Integration Referendums.” Party Politics, 12(5): 623-47.

Isiksel, Turkuler. 2016. “The British people have spoken. But what exactly did they say? The Monkey Cage.” The Washington Post, July 1.

LeDuc, Lawrence, Richard G. Niemi and Pippa Norris. 2002. Comparing Democracies 2: New Challenges in the Study of Elections and Voting, 2nd ed. London, UK; Thousand Oaks, CA: Sage.

Mudde, Cas. 2015. Transformations of Populism in Europe and the Americas: History and Recent Trends. London, UK: Bloomsbury Publishing.

Schuck, Andreas R.T. and Claes H. de Vreese. 2015. “Public support for referendums in Europe: A cross-national comparison in 21 countries.” Electoral Studies, 38: 149-158.Urbinati, Nadia. 2019. Me the People: How Populism Transforms Democracy. Cambridge, MA: Harvard University Press.