In 1948, pollsters and journalists learned that campaigns mattered. In 1984, the lesson may be that sometimes campaigns may not matter at all. The major question for analysis of this election, therefore, is not what did Ronald Reagan do to insure victory, but why did anyone think Ronald Reagan could lose?

This year is an awkward year for analysis for other reasons as well. This was a landslide election; landslides tend to be massive rejections of one candidate or massive affirmations of the winner's policies or performance. This year, the voters were clear in their approval of Ronald Reagan's economic programs. They were less clear in their support for increased spending on defense or in support of the Reagan social agenda. Straightforward analysis of the election provides indications of somewhat increased confidence in government and the prevalence of economic issues dominating the reasons for candidate support. The election seems a mandate for continued economic success, with little concern for how that is achieved. The groups who voted against Ronald Reagan were those who did not think they benefited from administration policies—blacks, Hispanics, and the poor—as well as self-described liberals and the traditional supporters of American liberalism, Jews.

Gerrymandering—the intentional manipulation of legislative boundaries for political advantage—is a venerable practice. Why, then, some might wonder, should we pay greater attention to it at this time? In particular, should judicial inquiry into constitutional issues of fair representation, intense for some two decades, now turn to what may well seem to comprise the heart of the “political thicket”? Throughout this period of reapportionment litigation, federal courts have alluded to the problem, with increasing concern shown by members of the Supreme Court of the United States, about its importance (e.g., Karcher v. Daggett, 103 S. Ct. 2653: 1983). Is the time ripe for a direct judicial examination of the gerrymander on constitutional grounds? And, if so, does California comprise an appropriate test case?

Prerequisite to answering such questions are: (1) an understanding of how and why gerrymandering, in magnitude, extent, and impact, has become an essentially new kind of issue rather than a mere extension of a traditional practice; and (2) a need to develop judicially manageable standards of identifying gerrymanders.

Prior to the reapportionment revolution of the 1960s, there existed a variety of constraints that conditioned boundary manipulation. For one thing, a large number of states simply failed to redistrict for several decades, the situation that triggered Baker v. Can (369 U.S. 186: 1962), Wesberry v. Sanders (376 U.S. 1: 1964), et al. This resulted in great disparities in population among districts, a form of “silent” or “status quo” gerrymander that in practice minimized periodic boundary manipulation. For example, district lines for Congress were typically redrawn only in states—usually a minority—that lost or gained seats.

1. Bruce Cain, declare as follows: …

2. Based on my study of redistricting, I believe that the attempt to define unbiased and objective standards of political fairness is futile. The reasons for this are several. First, there are many possible standards of political fairness, and political scientists do not agree as to which is best, as is demonstrated by the contradictory arguments in the declarations of Professors Grofman and Baker. Professor Baker argues for a compactness criterion while Professor Grofman disparages the use of what he terms “formal criteria” such as the shape of district lines and proposes instead a seats-votes measure. Secondly, the two standards of political fairness proposed by Grofman and Baker will prove contradictory in many circumstances, and this will force the courts to choose between them. Thirdly, the proposed standards are inconsistent with other redistricting criteria—particularly, with the court's efforts in the last decade to protect racial and ethnic minority voting strength. Finally, attempts to use fairness formulae are fundamentally flawed and impractical: they underestimate the many complex factors that determine the outcome of an election and overestimate the capacity of social scientists to predict future electoral outcomes.

However one wishes to characterize Walter Mondale's campaign for the presidency, his loss was only the latest in a series of Democratic presidential candidate defeats beginning in 1968. In 1968, Hubert Humphrey got 43 percent of the popular vote. In 1972, George McGovem received 38 percent of the popular vote. And in both 1980 and 1984, the Democratic presidential tickets got 41 percent of the popular vote. Only in 1976 did a Democratic presidential candidate receive a (very slim) majority of the popular votes cast. Indeed, Democratic presidential candidates have received only 42 percent of the total votes cast between 1968 and 1984.

Although Democratic presidential candidates have not been faring well for 16 years, party identification has remained about the same—with the Democrats as the majority party. Until 1984. And that is what makes the 1984 election interesting, for in this election the voters finally seemed to change their party identification to correspond with what now appears to be their habit of electing Republican presidents.

In 1980, when Jimmy Carter received the same proportion of the total votes cast as did Walter Mondale in 1984, self-styled Democrats were still in the majority. But, by 1984, Republicans and Democrats were at a virtual tie nationwide, as these figures from NBC News election day voter polls demonstrate.

One autumn out of four, election forecasting surpasses baseball as America's national pastime. Then, everyone wants to know who will win, and everyone has a guess. Now, with the ballots carefully counted, forecasters await their awards. Below, I evaluate the quality of a wide range of 1984 presidential and congressional forecasts. The evaluation proceeds from nonscientific to scientific approaches, although this distinction is sometimes blurred. To lower the level of suspense, I should say that some forecasts turned out to be quite good. By way of conclusion, I offer a set of rules for selecting a high-quality forecasting instrument.

Lucky Guesses

Many popular election forecasting rules take advantage of chance, which has been working in their favor. Perhaps the most famous is the World Series forecast, which says, “If the American League wins the World Series, then the Republican presidential candidate will win.” This technique was accurate from 1952 to 1976, missed in 1980, but worked again in 1984 with the victory of the Detroit Tigers. A lesser known rule of this type, which is my personal favorite, is based on the Beaujolais wine harvest. Accordingly, “If the Beaujolais vintage looks bad, then the Republican will take the presidency.” This has held post-1960, and continues to do so with the poor 1984 crop (yielding a wine too light, with little color). There are other such rules that relate more directly to the candidates themselves and, in that sense, have more verisimilitude.

1. I offer the Court this declaration because I believe that it is helpful for the Court to consider the extent to which the reapportionment process is inescapably political and value-laden. The drawing of district boundaries requires the weighing of many different and often conflicting interests. Any result will be a political result pleasing to some and not to others. Tests that could be used by the courts to manage cases, like this one, in which a partisan group petitions the courts to impose a judicial result in a state in which one political party claims to be disadvantaged by the outcome of the political process, are not neutral but political in their outcomes, and trade-offs between and among the various tests that might be used require the exercise of political judgment.

2. In the early 1960s, the United States Supreme Court affirmed the principle that members of Congress must represent districts containing equal numbers of individuals (regardless of whether they vote or are even eligible to vote) as measured by the United States census. The application of this standard required only that the courts determine the respective populations of districts, a readily manageable task. The plaintiffs in this case, however, are asking the Court to do something much more difficult, namely protect the interests of one of many political or social groups, and have suggested that there may be an ideal degree of collective “effectiveness” of votes cast by a particular group. This is an entirely different matter.

With the exception of 1976, all national elections since 1964 have generated commentary among both media analysts and political scientists about the possibility of realignment. Reports have varied from the straightforward—yes or no—to the contrived—realignment has been realized at the presidential but not the congressional level. In this essay, we outline our view of those factors that are necessary for a realignment, and we evaluate the 1984 elections with respect to those factors. Our focus in this analysis is on the tripartite structure of American party systems: party in the electorate, party in government, and party as organization. In addition, we discuss the policy consequences associated with realignments.

Theories of Realignments

Schlesinger (1984: 371) reminds us that “[the parts of parties] are treated as though each leads a life of its own with little attention to what if anything holds them together.” This admonition also holds for students of realignments. One school of thought, centering around the University of Michigan's Survey Research Center, focuses attention on party in the electorate. In this view, a realigning election is one in which the normal party vote shifts from one party to another, and because of the influence of individual partisan identification on electoral outcome, this new majority party dominates elections for a generation or more. This emphasis on the distribution of party identification in the electorate is beneficial in that it gives us an operational definition of realignment and allows us to assess both critical and secular (gradual) realignments.

The 1980s California congressional plans are not shown to be gerrymanders by the fact that the 1982 results were grossly disproportional to the votes of the electorate, nor by the fact that these plans each have only a very small number of seats where both parties have reasonable chances for the victory, nor by the fact that each plan is characterized by tortuously shaped districts, e.g., in San Diego and Northern California, which have no justification in terms of minority protection or in preserving city boundaries intact, although all three of these statements are true. It is well recognized that single member districts cannot be expected to yield proportionality between a party's vote percentage and the percentage of seats it wins (indeed, defense experts cite my own work on just this point); and it is also recognized that neither the absence of competition, nor the existence of ill-compact districts, make a gerrymander in and of themselves. Each of these factors, rather, is an indicator of possible gerrymandering efforts. When all three of these factors are found together with others (e.g., a plan passed in frantic haste by a partisan majority with no input from the minority party, a plan with clear discriminatory impact on the incumbents of the minority party) we have evidence for the conclusion that a plan that looks like a gerrymander, was intended to be a gerrymander, and has the effects of a gerrymander, is a gerrymander.