David W. Brady is the Bowen H. and Janice Arthur McCoy Professor of Political Science and Leadership Values at Stanford University and the Deputy Director of the Hoover Institution.
While there is some disagreement in the profession of political science about how polarized the American electorate is, there is undeniably considerable polarization in the U.S. Congress. The post-World War II period of bipartisanship on both on foreign affairs and the conservative coalition (Northern Democrats and Southern Democrats) on domestic legislation, was replaced beginning in the late 1970s and accelerating in the 1980s by high levels of polarized voting. Legislative polarization is typically measured by roll call voting patterns, showing greater inter-party differences, and the data clearly show rises in polarized voting beginning in the late 1970s.
How does legislative polarization as measured in the U.S. stack up when we look at legislative bodies in other parts of the world? Surprisingly, the answer is we are not, relatively speaking, very polarized when compared to parliamentary systems. It is thus unclear why polarization is thought to be a special problem in the U.S. Moreover, polarization as measured in parliamentary systems conceals as much about disagreement as it reveals. A comparative analysis of polarization should cause us to go back to the drawing board to ask exactly why polarization is bad.
In the U.K., parliament polarization scores are basically 100 percent in that all of Labour votes against all of the Tories. In Japan, the parties of the left vote unanimously against the Liberal Democratic Party. One common measure of unified voting by parties is the Rice index of party unity, which is the absolute difference between the proportion of party members voting in favor and the proportion of members voting in opposition, multiplied by 100. In a survey of 16 European countries plus Australia and New Zealand, which includes 90 political parties, the lowest party unity scores were 88.63 for Finland and 93.17 for New Zealand, with the average for all sixteen countries being over 97. In contrast, Crespin and co-authors show that over time in the U. S. Congress, the index’s high score is 66 and, even when adjusted, is in the 75 range. In sum, if we compare the U.S. to other democracies, its polarization levels are not particularly high.
What then is there to worry about? There may be headlines in Italy and France about dysfunctional government, but they don’t blame it on polarized political parties. One thing that should generate concern: do such measures of polarization tell us much about intra-party ideology or differences? The answer is that we don’t know. Party unity scores, where parties of the left uniformly vote against parties of the right, could be the result of preferences or pressure. That is, the party unity scores could be the result of members agreeing with each other – having the same preferences – or the unity could be imposed by party leaders’ actions such as the three line whip in the U.K. House of Commons.
Research on European parties has turned to analyses of how political parties achieve unity. In a 2012 paper, Close and Lopez show that pre-floor vote disagreements within the parties are a common feature of European parliamentary parties, “contrary to the unitary assumption assumed by legislative studies relying on recorded votes analysis”(p. 26). How, then, do these parties achieve such high levels of unity and party voting? It could be that they have an internal party process where, after sufficient discussion, they agree to vote for the median position, or at the other extreme it could be that the party leaders decide the policy position and enforce unified voting. We know little about this because the processes are internal to the parties–neither the process nor the enforcement mechanism are public. Contrast this to the U.S. Congress, where all of the disputes and differences are played out in public, as in the recent case of the Tea Party and the final budget vote. In short, in the U.S., we know about polarization and internal policy differences, whereas in Europe and other democracies, we know the party’s final policy position and levels of cohesion and party voting but not how the cohesion was achieved. Before we go overboard on polarization in the U.S., we need a better understanding of where we stand relative to other democracies.
In addition to the difficulty of knowing how the polarized voting came about – it is from preferences, processes, or pressure? – knowing that party polarization on a left-right basis exists does not always tell you what policies will be enacted. The Bank of England, nationalized after World War II by the newly elected Labour government of Clement Atlee, was surprisingly made independent by the newly elected Blair Labour government, an act described by the May 6, 1997 BBC On This Day program “as the most radical shake-up in the bank’s 300 year history.” A left party which nationalized the bank achieved what numerous Conservative governments could not. Similarly, the recent shift by the PRI in Mexico in regard to foreign aid interests and PEMEX, the national oil company, ought to make us aware of the need to understand both policy shifts by parties and how they achieve unity in enforcing such shifts.
This is the latest post in our ongoing series on political polarization. The previous posts are listed below. -Dan Hopkins