Anatomy of a Labour Landslide: The Constituency System and the 1997 General Election BY RON JOHNSTON, CHARLES PATTIE, DAVID ROSSITER, DANNY DORLING, HELENA TUNSTALL AND IAIN MACALLISTER* ONE noted characteristic of the UK electoral system is that a party's percentage of the votes cast at a general election is usually substantially different from the percentage of the seats that it wins in the House of Commons. In general, parties with more than 30% of the votes get a greater percentage than that of the seats and are over-represented, whereas those with fewer votes are under-represented, in some cases very substantially so, as has been the experience of the Liberal Democrats over the last twenty-five years. There are some exceptions to this generalisation, however; small parties whose votes are geograph- ically concentrated in a few constituencies (such as Plaid Cymru and several of the Northern Ireland parties) tend to get a number of MPs consistent with their share of the vote; in 1951 and 1974 (February) the party with most votes did not also get most seats; and then in 1997 the Conservative Party won 30.7% of the votes cast but only 25% of the seats. It is generally appreciated that this bias in the translation of votes into seats is created through the interaction of two geographies —the geography of support for the parties and the geography of the map of constituencies. The latter is overlaid on the former, which tends to be very consistent in its structure over time, even if its relief varies as party fortunes wax and wane. But can the bias be manipulated, by the parties or other agents, to partisan ends? To answer that question, we first define and then decompose the nature of bias in the UK electoral system and then look at the most recent general election in detail. Bias and the electoral system Bias in electoral systems has been defined in a variety of ways. We define it here as the difference between the number of seats won at an election by two main parties if the only change were in their relative shares of the votes cast. For example, in 1997, Labour won 43.2% of the votes cast in the UK and obtained 419 seats in the House of Commons, whereas the Conservative Party won 30.7% of the votes, * The authors arc respectively Professor of Geography, University of Bristol; Senior Lecturer in Geography, University of Sheffield; Research Fellow in Geography, University of Bristol; Lecturer in Geography, University of Bristol; Research Assistant in Geography, University of Bristol; and Research Assistant m Geography, University of Bristol. © Oxford University Press