Referendum Turnout and Legitimacy
A town of 38,000 held a referendum on closing the central square to private cars. Turnout was 31%, and 54% voted yes. The town's research office reported: residents under 30 are 30% of the population, over-60s 25%, and 30–59s 45%. Estimated turnout was 14% among under-30s, 58% among over-60s, and 35% among 30–59s. Among voters, 71% of under-30s, 46% of over-60s, and 52% of 30–59s voted yes. The square currently has 120 parking spaces and two bus stops; the council has already spent 800,000 CZK on design work for pedestrianisation. Councillor Bauer says the result is illegitimate because young people were underrepresented. Councillor Dvořák replies that a majority is a majority and the referendum is binding. A third member proposes online voting to 'fix youth turnout' and wants to rerun the vote.
1. Approximate the yes share if each age group had turned out at the same rate. Does the underrepresentation reverse the outcome?
Read the worked answer
Equal-turnout yes share = 0.30×0.71 + 0.25×0.46 + 0.45×0.52 = 0.213 + 0.115 + 0.234 = 0.562, or about 56%. The observed result was 54% yes, so equalising turnout would slightly increase the yes share, not reverse it. The underrepresentation of young voters therefore does not explain the outcome as a narrow majority produced only by old voters. It does, however, mean the referendum electorate was not representative. A 56% counterfactual still leaves a substantial minority opposed, so the legitimacy question is about representation and intensity, not about whether the majority would flip. The calculation also relies on self-reported or estimated group turnout and vote shares, which carry uncertainty.
2. Distinguish the empirical claim 'turnout was 31%' from the normative claim 'the result is illegitimate.' Which evidence would support the latter?
Read the worked answer
'Turnout was 31%' is an empirical statement: it can be checked against ballot records. 'The result is illegitimate' is normative: it depends on a principle such as equal representation, a quorum, or the idea that a decision should reflect the affected population. Differential turnout is evidence relevant to the normative claim, but it does not by itself prove illegitimacy. To support the latter, one would need a stated threshold (for example, a minimum turnout or a requirement that all age groups be represented within a margin) and evidence that the outcome would likely differ under a fairer process. The 56% counterfactual suggests the outcome would not differ, which weakens the illegitimacy claim even if it supports a critique of the process.
3. Evaluate the proposal to introduce online voting and rerun the referendum. What would you need to know?
Read the worked answer
Online voting might raise youth turnout, but that is an empirical prediction, not a guarantee. It could also advantage households with reliable internet and digital skills, and it raises security, identity-verification, and cost questions. Rerunning the referendum would require a legal basis and would set a precedent for re-voting whenever turnout is uneven. Before adopting the proposal, the town would need pilot evidence on turnout effects by age, an audit of the digital divide, and a judgment about whether the expected change in outcome justifies the cost. If the equal-turnout estimate is 56% yes, the proposal is unlikely to change the result; it may still be worth doing for participation reasons, but that is a separate argument.