Fairvote.org is currently undergoing an upgrade, and some features may not be working as usual. We apologize for any inconvenience, and expect to be back at full capacity soon.

2008's Shrinking Battleground and Its Stark Impact on Campaign Activity

// Published December 4, 2008

 Facts in Focus

* Of all 300 campaign events by major party presidential candidates tracked by the Washington Post between September 5 and November 4, 2008, 57% took place in only four states, representing just 17% of the nation’s eligible voters: Ohio, Florida, Pennsylvania and Virginia.
* 54.5% of all ads by the presidential campaigns, as tracked by CNN from September 24 to November 4, aired in those same four states.
* More than 98% of all campaign events and more than 98% of all campaign spending took place in only 15 states representing 36.6% of the nation’s eligible voter population, effectively sidelining nearly two-thirds of all Americans.
* Voter turnout in those 15 contested states was 67%, while turnout in the remaining states was 61%. Voter turnout declined in more a third of  states despite the public's high level of interest in the nation's first open-seat presidential election in half a century.
* Ohio and the 12 smallest states both have eight and a half million eligible voters, but Ohio had 62 campaign events (more than a fifth of all events) and the small population states had a total of 12, all in New Hampshire.


Both major party candidates in the 2008 presidential election made an ambitious promise upon effectively securing their party’s nominations —to wage nationwide campaigns and reach out to as many voters in as many states as possible. Barack Obama hired field staff in every state, while John McCain pronounced he would not give up on Democratic strongholds like California.
But the candidate's good intentions were undercut by the political reality created by the current Electoral College system and states’ use of the winner-take-all rule (i.e., awarding all of a state's electoral votes to the candidate who receives the most votes in each individual state). Under that winner-take-all rule, candidates have no reason to poll, visit, advertise, organize or pay attention to the concerns of states where they are comfortably ahead or hopelessly behind.

Indeed the 2008 election campaign was narrowly focused on the same small number of closely divided battleground states as in recent elections. According to the Washington Post’s candidate tracker, every campaign event with presidential or vice-presidential candidates in the two months after the Republican convention took place in one of 19 states, with nearly two-thirds of events (64%) in just five states. The states dominating candidates’ resources and attention (Florida, Ohio and Pennsylvania) were a repeat of the most sought-after states of 2004.

Similarly, more than 99% of ad spending by the major party candidates from September 24 to November 4, 2008 took place in 17 states, all but one of which (Montana, where less than a million
dollars in ads were aired) also were among the favored few that received campaign visits. Despite collectively representing nearly twice the eligible voter population as the battleground states, 31 states and the District of Columbia were effectively spectators to the campaign.

The historic nature of this campaign – with high public interest, a first-ever African American major party nominee, record-setting fundraising and a competitive primary season in both parties – was expected to generate a sharp increase in voter turnout, but ultimately turnout rose only 1.6% to 61.6% More than a third of states (18 in all) experienced a decline in turnout of eligible voters from 2004, including 14 of the spectator states. Overall, the top 15 states had a turnout of 66.9%, while the remaining states had 60.7% turnout.

“By 2012, we anticipate that the presidential election will be governed by the National Popular Vote plan for president where every vote is of equal weight,” commented FairVote’s executive director Rob Richie. “It couldn’t come a moment too soon.”

Here is a review of FairVote’s data about the peak campaign season of 2008.

Campaign Events

Once the general election began following the Republican convention, campaign events by major party presidential and vice-presidential candidates were overwhelmingly concentrated

Table 1. Campaign Events between September 5 and November 4, 2008*

Rank         State        Events        % of total    Cumulative %
1                OH           62            20.7               20.7
2                FL            46           15.3                36.0
3                PA           40            13.3                49.3
4                VA           23             7.7                57.0
5                MO          21             7.0                64.0
6                CO           20             6.7                70.7
7                NC           15             5.0                75.7
8                NV           12             4.0                79.7
9                NH           12             4.0                83.7
10              MI            10             3.3                87.0
11              IN              9             3.0                90.0
12             NM             8             2.7                92.7
13             WI              8             2.7                95.3
14             IA               7             2.3                97.7
15            MN              2             0.7                98.3
16            ME              2             0.7                99.0
17            DC              1             0.3                99.3
18            TN              1             0.3                99.7
19            WV             1             0.3               100.0
Total                        300           100        

* Source: Washington Post

The cumulative percentage column of this table is telling. Nearly half of all campaign events were held in just three large population swing states, Ohio, Florida, and Pennsylvania, and 75% of events were in only seven states. Only one of the 12 smallest population states had a single event, in contrast to Ohio, Pennsylvania and Florida together having 148 events.

Campaign events are piled on to a handful of heavily contested battleground states, while the rest of the nation is completely ignored by candidates. 31 states did not receive any campaign events, according to data collected by the Washington Post. Campaign consultants consider these states safely Republican or Democrat in a competitive year and therefore will not waste candidates’ time and resources on them.

Even the few campaign events in non-competitive states were part of the swing state pattern. Tennessee, for example, was a safe state in this election cycle, yet it still received one visit. The same is true for West Virginia and Washington, D.C. However, the events in these states had less to do with gaining voters in each state than with reaching out to voters in neighboring states. The event in Tennessee was held in Blountville, which sits close to Virginia and North Carolina, two key battleground states. The speech by Joe Biden in Washington, D.C. can be explained by two factors: 1) it is where Sen. Biden reports for his work in the U.S. Senate, and 2) it is right next to a key battleground state, Virginia. Similarly, the campaign event in West Virginia was held only an hour outside Ohio, the star state for getting candidate attention on the 2008 campaign trail.

(Note that, as discussed in our methodology section below, there almost certainly were campaign events that were overlooked by the Washington Post, such as ones hastily organized outside candidate schedules like Joe Biden’s campaign stop in Pierce County, Washington and Sarah Palin’s decision to go to Nebraska. But overall the Washington Post data is strong and unbiased.)

Campaign Spending

The ranking of states by amount of advertising spending looks quite similar to the campaign event ranking of states. Florida, Pennsylvania and Ohio lead the way, just as they did in 2004. The table below lists the top recipients of ad spending, accounting for 99.3% of all spending between September 24 and November 4, 2008.

Table 2. Ad spending by candidates from September 24 to November 4, 2008*

Rank    State    Ad $ spent    % of total    Cumulative %
1          FL    $29,249,985    18.2                18.2
2          PA    $24,903,675    15.5                33.7
3          OH    $16,845,415    10.5                44.1
4          VA    $16,634,262    10.3                54.5
5          NC     $9,556,598      5.9                 60.4
6          IN    $8,964,817        5.6                 66.0
7          WI    $8,936,200       5.6                 71.5
8          MO    $7,970,313      5.0                 76.5
9          CO    $7,944,875       4.9                 81.4
10        NV    $7,108,542       4.4                 85.9
11        MI    $5,780,198       3.6                  89.5
12        MN    $4,262,784      2.6                  92.1
13        IA    $3,713,223        2.3                  94.4
14        NM    $3,134,146      1.9                  96.4
15        NH    $2,924,839      1.8                  98.2
16        MT    $971,040        0.6                  98.8
17        ME    $832,204        0.5                  99.3

* Source: CNN.com

This table shows that over half of all campaign ad spending by the candidates was in only four states, and over 99% of all spending was in only 17 states. If we combine the total ad spending of the 25 states (including D.C.) at the bottom of the ranking, these states account for one-hundredth of a percent (.01%) of the total spending over this period. In order to reach just 1% of the total ad spending, we need to combine the bottom 36 states, and nine states did not receive a single penny in ad spending during this period.

The Shrinking Battleground – and a Roadmap for Change

The number of states that fall within the competitive range has been dwindling since 1992, with a large jump in the number of completely uncompetitive states and a decline in the number of competitive states. Consequently, the proportion of the U.S. population with a meaningful vote keeps growing smaller each election cycle

Table 3. Increasing partisanship of states over time

        States within each two-party partisanship range
                    (60%+)              (55%+)       (under 55%)
Year             Landslide         Comfortable   Competitive
1960                 6                    11                33
1964                12                   15                24
1968                 6                    14                31
1972                 7                    16                28
1976                 8                    10                33
1980                10                    11               30
1984                 5                    16                30
1988                 2                    23                26
1992                 3                    16                32
1996                10                    14                27
2000                15                    14                22
2004                15                    18                18
2008                16                    19                16

As the table above indicates, the number of battleground states has been nearly cut in half from an average of 30 states between 1960 and 1980 to an average of only 18 between 2004 and 2008, Despite the talk of Barack Obama’s candidacy expanding the Electoral College map, both parties increased their number of safe states. In fact, 2008 marked a record low in the number of competitive states since 1960 and a record high in the number of completely non-competitive states. Much of the perception of the election creating new battlegrounds was tied to Barack Obama winning the popular vote b y more than 7%. In an evenly divided year, the number of truly competitive states would have been remarkably small.

This election cycle continued the same trends as every election cycle since 1992, with a steadily declining number of states in the competitive range where campaign activity might change the results. This is a natural result of the current winner-take-all system of allocating electors under the Electoral College system – one established by state statute, and one that became the national norm decades after the framers established the U.S. Constitution and delegated power over election of the president to the states. Instead of having a single nationwide popular vote, the United States has 56 separate contests in states, Washington, D.C, and congressional districts in Maine and Nebraska, and the closest races are the ones that decide the election.

The states within each partisanship range vary in terms of region and size. The system holds no bias in favor of or against large or small states, nor does it have any particular tendencies towards eastern or western states. The problem with the Electoral College has little to do with geography and everything to do with the implications of the winner-take-all system established by local statute in 48 states and the District of Columbia under the authority granted to the states by the U.S. Constitution to establish rules for presidential elections.

FairVote does not support states taking action to replace the winner-take-all rule with systems that would divide that states’ electoral votes; our 2007 analysis (Fuzzy Math: Wrong Way Reforms for Allocating Electoral College Votes, www.fairvote.org/wrongwayreforms) shows that those approaches are fraught with partisan implications and perverse consequences. We instead urge state legislators to join Maryland, Illinois, New Jersey and Hawaii in enacting the National Popular Vote bill  (www.nationalpopularvote.com). The National Popular Vote (NPV) plan calls for states to pass identical legislation to enter them into an interstate agreement to award their electoral votes to the winner of the popular vote in all 50 states and the District of Columbia once the number of participating states represents a majority of the Electoral College.

Under the National Popular Vote, a vote in a safe state will carry the same weight as a vote in a swing state, and the nationwide popular vote winner will be guaranteed to win the election. FairVote's executive director Rob Richie commented "In the current Electoral College system, when your state isn't close you don't count. It's time to respect every vote in every election by states adopting the National Popular Vote plan by 2012."

A national popular vote, on the other hand, would make every vote across the nation equally important and equally valuable to the candidates, regardless of where the vote is cast. Candidates would have to vie for every vote throughout the nation, not just for votes in battleground states. The result would be a truly national campaign, one that caters to the true electorate and not simply to the lucky few in Ohio and Florida. Not only would the candidates be forced to widen their range of issues to address those that matter to the wider electorate, but they would also be forced to distribute their resources more equitably across the nation. The vast amount of money that rolls into swing states from TV ad spending, event expenditures, and supporting state-based campaign headquarters ought to be shared among the nation at large.

However, money and attention aren’t all that is at stake in this flawed system. Voter engagement suffers when citizens feel that their votes don’t count and are ignored by the candidates and their backers. A Republican voter in a strongly Democratic state knows that his or her vote won’t change the outcome in that state, so voting may be seen as a waste of time – creating a cycle of safe states becoming even safer. But a Democratic in that state also has little incentive to vote, just like thir Republican counterparts in safe Republican states. However, if voters in safe states knew that their votes would meaningful contribute to the candidates’ nationwide vote totals, this disincentive to vote and volunteer on behalf of their candidate would be removed.

FairVote has posted its full candidate trackers at http://fairvote.org/tracker. Visitors http://fairvote.org/president will find information on the National Popular Vote plan, which promises to make 2008 the last year with the current Electoral College system, along with our 2006 report Presidential Elections Inequality that provides detailed information on the last four decades of presidential elections and the impact of the current Electoral College system on the 2004 elections. In 2009 we will release an updated version of our Presidential Elections Inequality report that will include a detailed analysis of campaign attention based on visits, spending, and advertisements in each state by each of the two major party campaigns and their independent backers and showcase what states are the most likely battlegrounds in 2012 – and what states are certain to be ignored in an nationally competitive year.

Notes and Methodology

The information on campaign events was compiled using data from the Washington Post and is based on publicly available information from media reports, campaigns, and other sources. Events not reported by the Washington Post are therefore not included in this analysis. Some events – particularly fundraisers – are not included in the analysis, as they are often unannounced and do not serve to expand the candidates’ support bases. A campaign event is carefully defined to exclude any event that is not strictly for the purpose of reaching voters and gaining votes. Excluded events include Sarah Palin's two visits to her home state of Alaska, Joe Biden's four visits to his home state of Delaware, and the visits by each candidate corresponding to the debates in Mississippi, Missouri, Tennessee, and New York. Non-campaign events that are excluded from this analysis include private meetings (such as those between Governor Palin and world leaders in New York) and events at which both candidates were invited guests.

The information on campaign ad spending was compiled from CNN.com. Ad spending figures included in this analysis are by the candidates themselves and not by the parties and Democratic and Republican leaning interest groups. The data is based on the cumulative spending totals on September 24, 2008 and those on November 4, 2008. We took the difference between these two figures to extract the spending totals for each state throughout this period.

Founded in 1992 and with a ten-member Board of Directors that includes former Congressman and presidential candidate John B. Anderson, New Yorker essayist Hendrik Hertzberg and musician Krist Novoselic, FairVote is a non-profit, non-partisan organization that studies the impact of electoral rules and systems on turnout, representation and electoral competition. For more information, contact FairVote’s Rob Richie at (301) 270-4616 or [email protected] Thanks to FairVote’s Democracy Fellow Laura Kirshner for her leadership in assembling this report.