Primary Elections Research Paper

1. History And Development

From the 1830s through the early 1900s, American political parties nominated their candidates for public office through a multitiered system not unlike those still used in most other democracies. The system began with gatherings of the party membership in small local units such as towns, wards, or precincts, which would select delegates to county or district conventions. These county and district conventions would nominate some candidates themselves and might also select delegates to participate in a state convention.

In the first decade and a half of the twentieth century, however, this system was almost entirely abandoned, as most states enacted legislation requiring the major political parties to nominate their candidates through primary elections. The rapid and widespread adoption of primaries was the confluence of two distinct trends within the American political system. The first was a movement to subject the parties to a greater measure of legal regulation. Though parties clearly played a major role in electoral politics, they were, in the beginning, regarded as entirely private entities, such that fraud, intimidation, and corruption in party conventions was seen as a purely internal matter. As Key (1958) put it, ‘it was no more illegal to commit fraud in the party caucus … than it would be to do so in the election of officers of a drinking club.’ Shortly after the Civil War, however, a number of states began passing laws aimed at outlawing bribery and fraud and preserving some measure of order and procedural integrity in party affairs. By the end of the nineteenth century, almost every state had enacted some legislation of this kind (for a detailed discussion of this development, see Merriam 1908).

Even as this trend was sweeping the field, however, a second demand was being made: that the parties needed to be ‘democratized.’ To quote Key again, ‘through the history of American nomination practices runs a persistent attempt to permit popular participation in nominations and thereby to limit or to destroy the power of party oligarchies.’ In the early 1900s, this viewpoint had a particular resonance with the Progressive movement, which saw popular government as being threatened by an alliance of party machines and powerful economic interests. The push for party ‘democracy’ was given further impetus by the increasing sectionalism of the American party system. By the late 1890s, many parts of the country had become so completely dominated by one party or the other that receiving the major party’s nomination was tantamount to being elected.

Though a number of southern Democratic parties had already used primary elections on their own initiative, and other states had made them available on a voluntary basis, the first comprehensive, mandatory, statewide primary law was adopted by Wisconsin in 1903. Wisconsin’s example was quickly emulated in a number of other western and mid-western states, and then spread to the rest of the country. By 1917, ‘the direct primary had been adopted for most nominations in all save a handful of states’ (Key 1958). A small number of states resisted the trend, but the tides of historical circumstance were clearly against them. The last holdout was Connecticut, which finally enacted a primary law in 1955. As of the early twenty-first century, primary elections are clearly the dominant way of making nominations throughout the USA.

It is perhaps worth adding that most state election laws require nomination by primary election only for so-called major parties (usually defined as parties that received some minimum percentage of the vote in the last election). Minor and third parties, such as the Libertarians or the Socialists, still generally use a caucus-convention process to nominate their candidates for public office.

Students of comparative politics have generally regarded primary elections as a uniquely American institution, strikingly different from the nomination systems employed in every other democracy (see, for example, Ranney 1981). While this still holds true as a generalization, there are some signs that other countries are beginning to experiment with primaries. In 1999, for example, the PRI in Mexico nominated its presidential candidate through an open, national primary. In these other countries, however, the decision to use a primary is generally made by the party itself; only in the USA is the primary imposed upon the major parties by law.

2. Presidential Primaries

The trend toward primaries also partially transformed the presidential nomination process. Since the 1830s, both major parties in the USA had nominated their presidential candidates at national conventions, to which each state sent its own delegation. These state delegations, in turn, were selected at state party conventions or by state party committees. Beginning in 1912, however, a sizable number of states started using primary elections to choose their national convention delegates. Some Progressives (including President Woodrow Wilson) wanted to go further, scrap the national conventions entirely (at least as a device for making nominations), and choose the parties’ presidential candidates through a national primary.

Yet, for a variety of reasons, primaries never became as thoroughly institutionalized in the presidential nomination process as they did in making state and local nominations. Proposals for a national primary never came close to being enacted; and while 25 states had passed presidential primary legislation by 1916, over the next 15 years eight of them repealed these statutes. The result, when all the dust had settled, was a distinctive system for conferring presidential nominations that modern scholars generally call the ‘mixed system.’ Under the mixed system, about one-third of the states selected their national convention delegations via presidential primary, while the remaining two-thirds used state conventions or state party committees. National conventions thus had a partly popular, partly organizational basis, and remained significant decision-making bodies, though the presidential nomination itself was increasingly decided in a single ballot (the best account of presidential primaries during this period is Davis 1967).

The mixed system came to an end in the early 1970s. In 1968, at a remarkably divisive and chaotic national convention, the Democrats voted to create a commission to re-examine their presidential selection process. In just four years, this Commission on Party Structure and Delegate Selection (often called the McGovern–Fraser Commission, after the two men who served as its chairmen) completely rewrote the party’s rules governing delegate selection—so much so that most scholars see the years since 1968 as constituting a new era in presidential nominations, often referred to as the plebiscitary system. While the plebiscitary system differs from the old, mixed system in a variety of ways, one of the most important differences has been a huge increase in the number of presidential primaries. Where the 1968 election featured 17 Democratic presidential primaries and 16 Republican primaries, by 2000 there were 39 Democratic primaries and 43 Republican primaries. Under the mixed system, it was quite possible for someone to win a presidential nomination without contesting a single primary; under the new system, the nomination essentially goes to whichever candidate wins the most primaries (on these points, see Hagen and Mayer 2000). Though both parties still hold national conventions, they are increasingly empty affairs, put on for the benefit of television, in which all the important questions have already been decided.

3. Who Can Participate?

Since primaries are a product of state rather than national legislation, primaries vary across states in a host of major and minor ways. One of the most important differences, for both its practical and theoretical consequences, concerns who is allowed to vote in the primaries. At one extreme are the so-called closed primaries, in which every registered voter is asked if they are supporters of or affiliated with any particular party, and then only those who have expressed an affiliation with a party in advance are allowed to vote in that party’s primary. (Even in this ‘extreme’ case, it should be noted, the level of commitment required from the voter is exceedingly light. Voters are only asked to affirm a kind of general, psychological tie to the party; they are not required, in any state, to show that they have performed any previous service for the party or to make a binding promise of future support.)

A somewhat more permissive regime, that might be called semiclosed primaries, prohibits registered Democrats from voting in Republican primaries and vice versa, but allows unaffiliated or independent voters to vote in either party’s primary. A related provision, found in many state statutes, allows voters to change their party affiliation when they show up at the polling place on election day. More permissive still are what might be called declarative systems, where there is no party registration and any voter may vote in either party’s primary simply by asking for that party’s ballot.

Declarative systems at least ask the voters to state publicly that they wish to participate in the Democratic or Republican party primary. Open primaries, by contrast, relieve the voters of even this encumbrance. In open primaries, each voter is, in effect, given a ballot for both parties and then allowed to make his or her decision about which primary to participate in in the privacy of the voting booth. Since open primaries at least require a person to vote in only one party’s primary on any given election day, a number of western states have recently sought to remove even this last imposition on the long-suffering voter. So-called blanket primaries allow voters to participate in both party’s primaries on the same day, casting a vote in, say, the Democratic primary for the US Senate, the Republican primary for the US House, the Democratic primary for state representative, and so on. (In June 2000, the US Supreme Court declared the California blanket primary law unconstitutional.)

States frequently change their primary legislation in large and small ways, and many of the systems that emerge are not easily placed in one of the categories listed above. But according to a recent analysis by Jewell and Morehouse (2001), 11 states had closed primary laws, 15 used semiclosed systems, 10 had declarative systems, 10 used open primaries, and 3 (including California) had opted for the blanket primary. As has already been suggested, none of these systems demand very much from the voter in the way of past or future support for the party, and none except the closed primaries even do much to prevent members of one party from helping select the other party’s candidates.

Does it matter what type of participation standard a state chooses to adopt? Much of the concern about having open or declarative primaries originally focused on the possibility of party ‘raiding’: cases where large numbers of persons affiliated with one party would vote in the other party’s primary, with the deliberate intention of nominating the weakest candidate. But there is little evidence that this sort of thing occurs: it is difficult enough to get most American voters to the polls at all, much less to convince them to carry out a complicated strategic maneuver. A more likely consequence of open primaries is that they make it more difficult for the parties to present a cohesive ideology and set of candidates. Independents and members of the opposition party tend to be less committed to a party’s core symbols and principles, and thus more willing to support a candidate who is out of step with large segments of his or her own party.

4. Who Does Participate?

In numerical terms, voter turnout in primary elections is substantially greater than the turnout at local caucuses (where these are still held), but smaller than the turnout at general elections. The best venue for comparing turnout rates is presidential elections, for besides having reliable data on the number of votes cast in presidential primaries and in the November general election, there are, as noted earlier, still about 10–15 states that continue to select their national convention delegates through a caucus-convention procedure. In general, contested presidential primaries attract about 20–35 percent of a state’s voting age Democrats or Republicans, compared to a 50–55 percent turnout in the general election and a participation rate of just 1–3 percent in the presidential caucuses. In addition, there is far greater variation in primary election turnout rates, depending on how many seriously contested races are on the ballot.

One major exception to these generalizations occurred in the South when that region was still a one-party enclave. Since winning the Democratic nomination virtually guaranteed a candidate’s triumph in the general election, southern Democratic primaries regularly drew a far larger turnout than the general elections that followed.

It was once widely believed that the people who voted in a party primary were not very representative of the party’s typical voter or identifier. More recent research, however, suggests that this fear has been exaggerated. While there are some differences between primary voters and general election voters (the former are, for example, somewhat older and better educated), the differences are relatively small and do not appear to have much effect on either the type of candidate who gets nominated or the ideological views that get expressed.

5. Consequences

The widespread use of primary elections as a nominating device has had several major consequences for American politics. First, it greatly weakened party organizations by taking away from them the single most important function they once performed. Prior to the Progressive Era, it was very difficult for a candidate to get nominated by claiming to be an ‘outsider’ or insurgent, running in avowed opposition to the party establishment. Since the adoption of the primary, however, such candidacies have become quite common. This is not to say, of course, that candidates endorsed or supported by the regular party organization invariably lose primaries; but primaries have greatly increased the odds that a non- or anti- organization candidate will be successful.

Primaries also greatly lengthen the public phase of American election campaigns, since they generally guarantee that, in one party or the other, there will be two distinct campaigns: one for the nomination and one for the general election. Primary campaigns are also more expensive than those required in a caucus-convention system, since they must be targeted at a much larger and less attentive audience.

Though the point is disputed, there is also a longstanding concern that primaries make it more difficult for the parties to present a united front during the general election. According to this argument, often referred to as the ‘divisive primary’ hypothesis, primaries require the parties to air all their grievances and divisions in public, rather than settling them behind closed doors. In some cases, indeed, candidates may go out their way to create or emphasize divisions, as a way of distinguishing themselves from the other contenders in a primary. Primaries, it is further argued, may be good at assessing the first choices of party voters, but are not very good at discovering compromise candidates that are acceptable to a variety of factions, or at promoting ‘balanced tickets’ that offer some measure of reward or recognition to a diverse set of interests. Other scholars, however, claim that primaries do not so much create divisions as reveal the ones that already exist (for a review of these issues, see Mayer 1996). At the very least, one can say that sharp intraparty conflict was certainly not absent from the days when American parties selected their candidates through caucuses and conventions, and that the primary has not prevented most Republican and Democratic candidates from receiving a substantial majority of the votes cast by their fellow partisans.

Bibliography:

1. Davis J W 1967 Presidential Primaries: Road to the White House. Thomas Y. Crowell, New York
2. Hagen M G, Mayer W G 2000 The modern politics of presidential selection. In: Mayer W G (ed.) In Pursuit of the White House 2000: How We Choose Our Presidential Nominees. Chatham House, New York
3. Jewell M E, Morehouse S M 2001 Political parties and elections in American States, 4th edn. Congressional Quarterly, Washington, DC
4. Key V O 1958 Politics, Parties, and Pressure Groups, 4th edn. Thomas Y. Crowell, New York
5. Mayer W G 1996 The Divided Democrats. Westview Press, Boulder, CO
6. Merriam C E 1908 Primary Elections. University of Chicago Press, Chicago
7. Ranney A 1981 Candidate selection. In: Butler D, Penniman H R, Ranney A (eds.) Democracy at the Polls. American Enterprise Institute, Washington, DC