Democracy at a Crossroads
Why California Needs to Abandon "Winner-Take-All" Elections

By Steven Hill
Published January 1st 2005 in The California Journal
California's political institutions and practices are outdated and no longer reflect the vibrancy and diversity of our state in the 21st century.  Key political institutions are badly in need of an overhaul, making them better suited for the new California and its wide range of attitudes, demographics and geographic regions.

In particular, three ailing aspects of our representative democracy stick out. First, the most recent redistricting was nothing more than an "incumbent protection plan" in which party leaders all but did away with legislative elections in California, leaving voters from all partisan sides with choice-less elections.  Second, Californians' antiquated winner-take-all electoral system is increasingly hard-pressed to provide representation to California's diversity, and has exacerbated regional balkanization -- a stateside version of Red vs Blue California -- and is electing fewer bridge-building moderate legislators, with harmful consequences for state policy.  Third, winner-take-all elections for statewide offices combined with California's primary system discourage moderate and independent candidates and elect leaders who win with support from less than half of the electorate.

Fortunately, there are solutions to all of these democratic deficits, if the political will exists to think and act boldly outside the box of conventional thinking.

The Travesty of Redistricting:  Stopping Incumbent Protection Plans. What if you could pay $20,000, and for that modest sum end up with lifetime employment at a salary of $158,000 annually, with the best health and retirement benefits, frequent travel to Washington DC, and staff and paid expenses, all on the public's dime? What a deal, eh?

That's the cozy situation for members of California's congressional delegation as a result of gerrymandering their own legislative district lines. The 2001 redistricting in California was a travesty. Democrats dominated the process, since they had a majority in the state legislature and the governor's office.  So, according to one member of Congress, Democratic incumbents paid $20,000 apiece to the political consultant drawing the district lines -- who happened to be the brother of one incumbent -- to draw each of them a "safe seat" where they would easily win re-election. It was like paying protection money to a Mafia don for your turf.  Congresswoman Loretta Sanchez (D-47), knowing a bargain, told an Orange County Register reporter, "Twenty thousand is nothing to keep your seat. I usually spend $2 million every election."

Then, to the dismay of national Democrats, who with their party's control over the line-drawing were hoping to gain seats in California, state Democrats instead gave safe seats to the GOP incumbents too, in return for their acceptance. The fix was in. It was a bipartisan collusion against California democracy and the voters. And it worked. In the recent November election, 51 out of 53 congressional seats were won by huge landslide margins.

The Democrats also drew safe seats for state Senate and Assembly districts. Those too resulted in 90 percent of state legislative races won by landslide margins in November. The Democrats literally did away with most legislative elections in California, turning them into predictable farces. Moreover, the Democrats line-rigging allowed them to bite off more than their fair share, winning a disproportionate 63 percent of state Senate seats and 60 percent of state Assembly seats with only 54 and 53 percent of the statewide vote respectively in each chamber (in 2002, the Democrats won a whopping 70 percent of state Senate seats with only 54 percent of the statewide vote).  Forget about "money buying elections," most elections are decided during the line-rigging process, when politicians use sophisticated computers to handpick their voters before voters pick them.

But that's not all. This backroom redistricting has produced a government where hard-core partisans dominate the legislature and fewer moderates get elected. It has exacerbated a Red vs. Blue California marked by regional balkanization, where the high population coastal Blue areas are dominated by Democrats and the low population Red interior by Republicans. Not that there aren't Democrats in red areas and Republicans in blue areas -- as well as independents and third party supporters -- it's just that they are "orphaned voters" who never win any representation. Purple California gets smothered in the zero-sum game of winner-take-all elections, which grants representation only to the winners.

Governor Arnold Schwarzenegger, the Chamber of Commerce, Republican recall activist Ted Costa and others have proposed taking redistricting out of the hands of a partisan legislature. This makes sense, but the devil is in the details.  For instance, an initiative proposed by Costa would create an unwieldy process that requires any redistricting plan to receive voter approval. It also would immediately reopen redistricting instead of waiting until the end of this decade, as is customary. Both of these together would lead to bitter partisan battles that will disrupt any semblance of political peace for the remainder of the decade.  In particular, a mid-decennial redistricting is a bad idea because it would break the fragile truce between Democrats and Republicans that for decades has accepted a once-per-decade line redrawing. That truce was broken last year by GOP firebrand and congressional leader Tom DeLay who rammed through controversial mid-decennial Republican gerrymanders in Texas and Colorado. Attempting a re-redistricting in California with such clear partisan motivations would provoke bitter responses from Democrats, not only in California but probably in other states as well. It has not gone unnoticed that those pushing mid-decennial re-redistricting in California are mostly Republicans.

A true public interest redistricting should establish an independent nonpartisan commission with appointees from various sectors of California society, including minority representatives, business community and good government groups, rather than a panel of judges (as proposed by Costa) which, after the debacle of Bush v. Gore, practically no one sees as being nonpartisan anymore. Better to have a balance of politics, and have it out in the open. Most importantly, the commission should be guided by nonpolitical and legal criteria such as compactness of districts, respecting geographic boundaries, protecting minority voting rights, and enhancing competition. The initial districts then should be presented to the public via hearings around the state, with a goal of finalizing the lines after hearing public comment. But this only should occur at the start of the next decade, since a mid-decennial re-redistricting would fire the next salvo in the partisan war. California and the nation have had enough of that for this decade.

Winner-take-all makes losers of us all. But even the best-intentioned "public interest redistricting" will have limited impact in addressing the many ills of redistricting. That's because the problem is not just who draws the legislative lines, it's our antiquated, single-seat district, winner-take-all system. The Democratic vote has become highly urbanized and concentrated, so in a winner-take-all system even the fairest redistricting will make only a handful of districts more competitive unless the legislative district lines are drawn as spokes radiating outward from the blue Democratic urban hub. But doing so would clash with the Voting Rights Act, which helps minorities as a "community of interest" to elect their chosen representatives. The fact is, winner-take-all elections pit everyone against each other -- Democrats, Republicans, independents, different racial groups -- all trying to win a limited commodity-representation. That's why it's called "winner-take-all," because only one side wins.

So what can be done?  Political scientist Arend Lijphart from University of California-San Diego says the best solution is to "evolve from winner-take-all elections toward some form of proportional representation."  With proportional representation, multi-seat districts are used instead of single-seat districts, and blocs of like-minded voters win seats in proportion to their voting strength at the polls. You don't see disproportionate results where one party wins way more than their fair share of seats. For instance, in the state Senate, instead of electing 40 individual district seats we could elect 10 districts with four seats each, using a proportional method.  If a political party's team of candidates won twenty five percent of the vote in a four-seat district, one of its candidates would win one of these four seats, instead of nothing; fifty percent would win two seats, and seventy five percent would win three seats. There are different kinds of proportional systems, candidate-based or party-based, but all adhere to those basic rules.

According to Professor Lijphart, using such a "moderately proportional system" for California would make all parts of the state competitive for both major parties, occasionally even a third party. A Republican candidate could win 25 percent in a blue coastal area, and a Democratic candidate could win that in a rural area, giving those orphaned voters some representation. And moderates and independents running campaigns outside of the party machines would be elected as well. Purple California would have a voice, and we wouldn't see disproportionate results like in the current California Senate where Democrats have won more than their fair share of seats. Illinois' state legislature has used such a moderately proportional system, and currently Cambridge, Massachusetts, Peoria, Illinois and Amarillo, Texas use it for local elections. Ireland and Australia use such a method for national parliaments. Their experiences show it's a better way to foster competitive elections, elect more moderates, reduce regional balkanization, and provide opportunities for minority representation.

Moreover, proportional voting systems using multi-seat districts don't require redistricting. They allow voters from all political persuasions and races to control and define their representation, not the map-makers. Particularly in cities, where four major racial groups compete for representation, the drawing of race-conscious district lines usually is contentious and can reinforce some of the worst aspects of race and segregation.  Nonpartisan, candidate-based forms of proportional representation emerge as flexible and fairer methods to facilitate diverse representation and competitive elections for multiracial cities.

If Governor Schwarzenegger and others really want to do something about the ills of redistricting, simply changing who draws the district lines won't accomplish much. It's necessary to get rid of California's antiquated winner-take-all system, and adopt some version of the more modern proportional representation system. Incumbents have invested $20,000 apiece to maintain the status quo, but it's time for a real change.

Ensuring majority rule for statewide offices. In the absence of a requirement that California candidates win with a majority of the vote, and with eight political parties running candidates, the last two governor's races were won by candidates with support from less than half the voters. They also produced two of the lowest turnouts for a gubernatorial race in California's history, even the gubernatorial recall resulting in the third lowest turnout despite all the celebrity and hype (initial anecdotes reporting a surge in turnout actually were due to a decrease in polling places that created longer lines of voters).

Proposition 62, an initiative on the November ballot, sought to address some of the shortcomings of our elections by instituting a primary in which the top two candidates, regardless of party affiliation, would advance to the November election.  It was a flawed approach that would have severely restricted voters' choices in November, and quite rightly was rejected by voters. But the open-style primary, allowing voters to choose from all candidates regardless of party, points the way toward another reform that will ensure majority winners, voters' choice, and cost-effective elections.

The solution is simple. Get rid of the taxpayer-funded political party primaries altogether, and hold one election in November in which the majority winner in a multi-candidate field is elected using instant runoff voting.  Instant runoff voting (IRV) allows voters to rank their candidates, first choice, second choice, third choice, and uses a series of "instant runoff" rounds to determine the majority winner in a single election.  If your first-choice candidate is dropped from contention, your vote goes to your second choice as your backup candidate. Last-placed candidates are eliminated round by round, until one candidate ends up with support from over 50 percent of the voters, and he or she is elected

Using IRV, voters are liberated to vote for the candidates they really like instead of always picking the "lesser of two evils" or worrying about spoilers. In the 2000 presidential election, if Florida had used IRV, the nearly 100,000 Ralph Nader voters would have had the option of ranking a second choice.  No question thousands of them would have ranked Al Gore second, Gore would have been the recipient of all those runoff votes, and he would have won Florida and the presidency. History would have been changed.

IRV also could help moderate candidates like Sens. John McCain (R.-Ariz.), John Edwards (D-N.C.) or former Los Angeles mayor Richard Riordan to break the stranglehold that partisan voters have on the primary process. Instead of a November election dominated by the most partisan Democratic and Republican nominees, centrist candidates could build a broad coalition by trying to win the second and third rankings from the supporters of other candidates.

Last November, San Francisco became the first jurisdiction in California to use instant runoff voting for the election of local offices like mayor and supervisor. It had an immediate impact on the quality of campaigns, fostering more positive campaigns that stuck to the issues instead of mounting nasty attacks trying to tear each other down. That's because candidates sought to win by attracting the second or third rankings from the supporters of other candidates. This caused some candidates to build coalitions together, even hold joint fundraisers and rank each other on campaign literature.  IRV also will save San Francisco taxpayers millions of dollars each election cycle, the cost of administering low-turnout December runoff elections.

The Utah Republican Party uses IRV to nominate their congressional and gubernatorial candidates, and Ireland and Australia have been using IRV for decades to elect their highest offices. Santa Clara County and Berkeley also have passed pro-IRV measures, and IRV has been drawing national attention, with legislative bills introduced into 22 states. IRV is being watched as a method that can accommodate the reality of diverse candidate fields without resulting in unintended consequences like spoiled races and non-majority winners.

Getting rid of the low-turnout primary elections in California also will save tens of millions of tax dollars spent to administer them. Why should the public pay for the political parties' private primary?  Since the US Supreme Court has ruled that a political party's primary is a private affair and that California cannot force parties to open up their primary to voters not in their party, then why should California taxpayers foot the bill for these parties' private shindig?  Let the parties pay for a primary or a caucus themselves. Or the parties can choose not to have a primary or a caucus, they can deliver to the November election as many or as few of their nominees as they wish. And then the voters will rank their candidates, 1, 2, 3, and the candidate supported by a majority of California voters will be elected.

Another badly needed improvement is universal voter registration. With universal voter registration, all California citizens would be automatically registered to vote when they turn 18.  Previously this was not technically feasible, but the federal Help America Vote Act has mandated that all states must create statewide voter databases by 2006, which will make universal voter registration possible. Universal voter registration is used by nearly every established democracy in the world because it makes voter registration guaranteed, nonpartisan, and shielded from the shenanigans that plagued the recent presidential election, like Republican partisans in Nevada tossing out Democratic registration cards, accusations of Democratic fraud in cities, and Republican challenges of minorities at the polls. If deployed throughout the U.S., 50 million voters instantly would be added to the voting rolls, a disproportionate of them young people and minorities. California should lead the way on this, since so many Californians would benefit.

That's how democracy is supposed to work.  That's how it can work in California, if the political will is in enough supply, and the state's leaders become unstuck from the fly paper of old ideas.

Steven Hill is Irvine Senior Fellow of the New America Foundation, and author of "Fixing Elections:  The Failure of America's Winner Take All Politics" (