Countries criticized for their coronavirus response — the United States, the United Kingdom and France — instead use majoritarian electoral rules. Each district has only one seat available, and the seat goes to the candidate with the most votes. Majoritarian systems produce disproportional outcomes. The Liberal Democrat party, for example, got 23 percent of the U.K. vote in 2010, but only 8.8 percent of seats in Parliament.
My research shows electoral rules matter for health outcomes. Here’s what you need to know.
Proportional representation leads to more inclusion and more efficient health systems
In my book, “Coalitions of the Wellbeing: How Electoral Rules and Ethnic Politics Shape Health Policy” (Cambridge University Press 2015), I detail how in advanced industrial countries, proportional representation leads to broader, more inclusive coalitions that design more efficient health systems.
To win support in a proportional representation system, politicians need to cater to large swaths of the population. That’s because voters cast votes for parties, not individuals. Broad coalitions often form between multiple parties.
The resulting health systems in proportional representation countries have a wider geographic distribution of hospitals and medical personnel, less waste and better coordination of the health system across the country.
In contrast, majoritarian systems provide incentives for politicians to target goods and services to narrow segments of the population — especially districts where those in power might lose in the next election. Majoritarian systems also incentivize politicians to make policy and fund programs that allow them to claim credit, because when voters go to the polls, they select candidates rather than parties.
Politicians in majoritarian systems break the health budget up into highly visible goods they can easily claim credit for, in many cases. In 1990s Thailand, politicians allocated public funds to build hospital wings in their name, small health centers and even tiny 10-bed hospitals.
What we learn from New Zealand, which used to have a majoritarian system
My book details the case of New Zealand’s 1993 electoral reform, when the country changed from a majoritarian system to a mixed system. New Zealand now allocates 41 percent of seats via proportional representation and the remainder using majoritarian rules.
Since the switch, no New Zealand party has secured a majority of seats in parliament and broad coalitions have become the norm. Accordingly, health policy has become more inclusive of previously underserved groups, such as the Māori minority. In 2002, the Ministry of Health launched the Māori Health Strategy, and began including the Māori in policymaking and service delivery. In 2005, there were around 250 Māori health-care providers, compared with only 20 in the mid-1990s.
These changes had meaningful impact: Māori life expectancy increased at a higher pace than non-Māori in the 2000s. The inclusive health approach may have contributed to the success of New Zealand’s covid-19 response.
How do countries’ covid-19 experience reflect their electoral systems?
Looking at the 30 advanced-industrial countries with more than 1,000 coronavirus cases as of April 21, what do we see?
Majoritarian countries had an average death rate (deaths as a percentage of covid-19 cases) of 6.57 percent, compared to only 4.35 percent in proportional representation systems. Mixed systems — like New Zealand, South Korea and Taiwan — which incorporate elements of both majoritarian and proportional representation systems, had a death rate of 4.04 percent.
Of course, some countries were hit harder than others. But the covid-19 death rate also is associated with a key measure of health system efficiency: testing. Countries with lower death rates tested early and broadly. Majoritarian countries averaged only 10,847 tests per 1 million people, compared with 27,434 for proportional representation countries. Mixed systems averaged 17,915 tests.
Among 20 countries that publicly share data on ventilators, majoritarian countries average 13 ventilators per 100,000 people compared to 20 for proportional representation countries and 19 for mixed-system countries.
Some proportional representation systems may act more like majoritarian systems because of features that weaken the role of political parties and encourage politicians to showcase their individual accomplishments and characteristics as a vote-getting strategy. Political scientists use Joel Johnson and Jessica Wallack’s measure to differentiate among proportional representation systems according to how “personalist” they are.
Four proportional representation countries with significant death rates — Belgium, Sweden, the Netherlands and Spain — have an average personalist score that is more than twice (0.63) that of the four proportional representation countries with the fewest covid-19 deaths (Czech Republic, Iceland, Israel and Finland: 0.25).
These less personalist countries have more ventilators than the average proportional representation country, and twice as many intensive care beds. They have a testing rate that is almost four times that of the four proportional representation countries with the highest covid-19 death rates.
To be sure, electoral rules alone don’t determine the impact of covid-19. Australia is a majoritarian system with relatively low coronavirus death rate. Political science research on electoral rules and other health challenges seem to suggest citizens may benefit from proportional representation or mixed systems during this crisis. It’s too early to tell — but the pandemic may help reignite debates in majoritarian countries like the U.K. and Canada about reforming their electoral systems, and might even start such a debate in the United States.
Editor’s note: One nation was mischaracterized in an earlier version of this post. We have removed reference to that country’s political system. We regret the error.
Joel Selway holds a PhD in political science from the University of Michigan. He is an associate professor of political science at Brigham Young University. Data for this post can be found on his website.