Politics

Opinion · Politics

Political analysis and opinion that goes beyond the horse race: elections, policy, and the ideas underneath the day's partisan fights.

Latest Politics

Politics Economics Ideas

About Politics

Analysis, punditry and the difference

Political writing spans two things that look alike on the page. Analysis explains how a system produces an outcome: what an institution can and cannot do, where a decision actually gets made, which constraints bind whom. Punditry predicts and evaluates: who is winning, who erred, what a result portends. The first ages well because institutions change slowly. The second ages badly and is produced in far greater volume, because prediction is more entertaining than procedure. This section is aimed at the first, which means preferring the pieces that explain a mechanism over the pieces that score a contest.

Institutions constrain outcomes more than intentions do

A great deal of political commentary treats outcomes as products of will, when they are usually products of structure. Electoral systems shape how many parties are viable and whether coalitions are necessary. The division of authority between national and local government determines which promises a given office can actually keep. Courts, civil services, central banks and upper chambers all constrain what a winning party can do and on what timescale. Understanding where a power actually sits explains far more disappointments than any account resting on the character of the people involved.

Reading polls without being misled

A poll is a sample, and its headline number carries a margin of error that widens for subgroups, so a finding about a small demographic slice is often too imprecise to mean anything. Question wording and order measurably change answers. Who declines to respond is a persistent source of error, and adjustments made to correct for it involve judgment calls that differ between pollsters. Averages across many polls are more reliable than any single one. The most common misreading is treating a poll as a forecast rather than as a snapshot of stated intentions, which is a different thing.

Where policy is actually made

Legislation is the most visible route to policy and often not the most consequential one. A great deal of what governments do is decided through regulation written by agencies under authority a legislature delegated years earlier, through budget allocations that fund some programs and starve others, and through litigation that settles what an ambiguous statute means. Enforcement discretion matters as much as the rule: a law applied rarely and a law repealed can look similar from the ground. Coverage that follows only the legislative chamber misses most of the machinery.