Do Electoral Systems Affect Government Size?

June 9, 2004 • Commentary

In the midst of an election campaign heaving with policy platitudes, focus group‐​tested sound bites, and negative ads targeted at voters in purple states, one is reminded of Winston Churchill’s observation that “democracy is the worst form of government except all those other forms that have been tried.”

The same may be said about our electoral system, which allocates legislative seats on a first‐​past‐​the‐​post, winner‐​takes‐​all basis. The current system has many disadvantages, most notably its propensity to discriminate against minor parties operating outside the increasingly uncompetitive, cozy two‐​party system.

However, new research suggests that, in terms of policy outcomes, America’s winner‐​takes‐​all electoral system may be the least bad option for those seeking to limit government involvement in the nation’s economic life.

Economists Torsten Persson and Guido Tabellini recently studied how electoral rules influence government fiscal policy. Their findings were published in the American Economic Review. Specifically, they measured the effect of electoral rules on the size and composition of government spending. Persson and Tabellini compared policy outcomes in 140 democracies.

They found that electoral systems affect fiscal policy. In general, there is higher taxation and more government spending under a European‐​style proportional representation (PR) system, which allocates legislative seats based on a party’s percentage of the overall vote, than under a winner‐​takes‐​all system.

European PR systems tilt the composition of government spending toward programs that benefit large groups in the population, such as universal welfare programs. By contrast, the size of the minimal coalition of voters needed to win the election is smaller under a winner‐​takes‐​all system, which induces politicians to target smaller, but pivotal, constituencies.

According to the international evidence, a switch from PR to winner‐​takes‐​all reduces total government spending by approximately 5 percent of GDP. These effects are particularly pronounced in freer and older democracies. In PR democracies, in particular, government spending displays a “ratchet effect,” i.e., it goes up as a fraction of GDP during economic downturns, but it doesn’t come down during economic upturns.

The electoral system also affects the size of the budget deficit. Under a winner‐​takes‐​all system, deficits are smaller (about 3 percent of GDP) than under PR. A reform of the electoral system from winner‐​takes‐​all to European‐​style proportional rule would increase government spending by about 6 percent of GDP, financed by higher taxes and deficits.

Persson and Tabellini also ask whether winner‐​takes‐​all electoral rules cut welfare spending? They present evidence that winner‐​takes‐​all systems encourage a smaller welfare state than otherwise would be the case. Winner‐​takes‐​all elections cut welfare spending by 2 to 3 percent of GDP. Again, the effect is stronger in the older and freer democracies.

The fact that winner‐​takes‐​all electoral systems lead to smaller governments and smaller welfare programs than PR systems does not negate the deficiencies inherent in our current system. However, it does serve as a timely reminder that each electoral system produces a unique set of pros and cons.

As Churchill knew, there is no perfection in politics. The authors of our Constitution certainly didn’t provide us with a perfect electoral system. Nevertheless, our electoral system reinforces the Constitution’s central theme of limited government.

Let’s not spend time figuring out how to swap our electoral system for an equally flawed, if differently configured, European electoral system. Instead, let’s focus our reformist energies on disentangling the regulatory thicket that currently depresses the competitive nature of our own, increasingly undervalued, system.

About the Author