Should San Francisco Abolish Elections in Odd-Numbered Years?

by Paul Hogarth on June 28, 2007

On June 26th, Supervisor Jake McGoldrick proposed a Charter Amendment at the Rules Committee to abolish city elections in odd-numbered years. The measure would move San Francisco’s elections for Mayor, District Attorney, and Sheriff to coincide with presidential elections, and City Attorney and Treasurer with gubernatorial elections. The Amendment was tabled due to lack of support, but it’s actually a good idea. More people vote in even-numbered years, so holding elections in odd-numbered years effectively disenfranchises thousands of voters – including a large number of renters and people of color. While many cities have municipal elections in odd-numbered years, it’s a vestige of the urban “reform” movements of the early Twentieth Century when business elites feared that political machines were giving too much power to the poor and immigrants. And when cities have moved their municipal elections to even-numbered years, we have seen a higher voter turnout and, consequently, a more progressive electorate.

Under McGoldrick’s Charter Amendment, the last odd-numbered Mayor’s race will be in 2011. Afterwards, the next Mayor’s race will be in November 2016 to coincide with the Presidential election. While this would give whoever gets elected Mayor in 2011 a five-year term, it will not help Gavin Newsom because – assuming he gets re-elected in November – the Mayor will not be eligible to run again because of term limits.

It could, however, help Kamala Harris if she chooses to run for a third term in 2011 because District Attorneys are not covered by term limits (as well as Sheriff Michael Hennessey.) If Dennis Herrera and Jose Cisneros get re-elected in November 2009, they will also get a five-year term to move the City Attorney and Treasurer’s races to November 2014 to coincide with the Governor’s Race.

In San Francisco, it’s clear that a substantial number of voters “check out” in odd-numbered years, leaving many decisions in the hands of a smaller pool of people. While turnout last November was at 60%, it was at 54% in 2005 – an unusually high rate for an odd-numbered year because it coincided with Schwarzenegger’s special election. The Mayor’s race in 2003 had a decent turnout in December (54%) after a 46% turnout in November, but it was a full twenty points lower than in the 2004 Presidential race (74%.)

Likewise, the 1999 Mayor’s race had a 45% turnout in November (and 49% in the Brown-Ammiano run-off), but that paled in comparison to the 66% turnout in the following year’s presidential race. It gets even worse when you have an odd-numbered election year without a Mayor’s race – in November 2001, the turnout for a highly competitive City Attorney’s race was at less than 30 percent.

Of course, major cities like New York, Los Angeles and Chicago also have their municipal elections in odd-numbered years – but does that mean it’s good public policy? Many of these cities moved to separate municipal elections in the early 1900’s as part of a package of reform measures that included non-partisan elections to “depoliticize” city affairs and promote a “short ballot” so that voters can focus on fewer races at a time.

While these were “good government” measures designed to prevent the corruption of machine slate politics, it was also an elitist movement of business owners who wanted more government efficiency. Some, like New York reformer Andrew D. White, went so far as analogizing the city to a corporation and feared a city government that was run by immigrants and the urban poor. Breaking up municipal races into separate elections, they reasoned, would make it harder for party “bosses” to tell uneducated people how to vote.

When cities have moved away from odd-numbered election years, they have seen a higher turnout of infrequent voters – more renters, more students, more minorities, and more low-income people. In Berkeley, for example, city elections were once held in April of odd-numbered years. During the 1970’s, progressives from Berkeley Citizens’ Action suffered repeated defeats because their coalition depended largely on infrequent voters showing up. After a stunning loss in April 1981, everyone knew that something had to be done.

So in 1982, rent control activist Marty Schiffenbauer single-handedly collected all the signatures for a charter amendment to move city elections to November of even-numbered years – effectively killing municipal elections in odd-numbered years. His campaign, with the clever slogan “November beats April,” prevailed with little fanfare because few people at the time appreciated the magnitude of such a change.

With the new change in place, Berkeley progressives went on to win the November Mayor’s race that year, and then swept the City Council in 1984. Berkeley has since had all its municipal elections consolidated with the presidential and gubernatorial races, and I doubt you’ll hear anyone advocate a return to the old system. Recently, Berkeley voters agreed to move the Mayor’s race to presidential election years (rather than gubernatorial), which should increase voter participation – and it will go into effect in 2008.

At the June 26th Rules Committee hearing, Supervisor Sean Elsbernd said he didn’t like the idea of abolishing odd-number election years, because it was important to have voters focus on the Mayor’s and D.A.’s race. But McGoldrick said that concern is unwarranted. “I think people can walk and chew gum at the same time,” he told me.

Progressive campaign veterans have expressed concern about consolidating municipal elections because they will not be able to devote their full resources behind a mayoral campaign if there’s a presidential campaign going on at the same time. But that’s a small price to pay if it means a higher voter turnout in the Mayor’s race – and thus a better chance at electing progressive candidates. After all, it could allow like-minded campaigns who are running for different offices to make alliances and coordinate.

According to Controller Ed Harrington, eliminating municipal elections in odd-numbered years would save the City about $3.7 million. So if you’re a “fiscal conservative,” you should support it too. McGoldrick says this wasn’t the primary reason for promoting the Charter Amendment – “the real issue is voter turnout” – but with the current budget fight it’s always good to find efficient ways to cut costs.

EDITOR’S NOTE: Before moving to San Francisco, Paul Hogarth lived in Berkeley for eight years and was involved in local electoral politics. He was an elected member of the City’s Rent Board, and briefly served on the Steering Committee of Berkeley Citizens’ Action. Send feedback to

November 1995 (Mayor and DA Race) – 51.9%
December 1995 (Brown-Jordan Runoff) – 45.5%
March 1996 (Presidential Primary) – 40.5%
November 1996 (Presidential Election) – 62%
June 1997 (49ers Special Election) – 44%
November 1997 (City Attorney Race) – 30.4%
June 1998 (Gubernatorial Primary) – 46%
November 1998 (Gubernatorial Election) – 56%
November 1999 (Mayor and DA Race) – 45%
December 1999 (Brown-Ammiano Runoff) – 49%
March 2000 (Presidential Primary) – 45%
November 2000 (Presidential Election) – 66.66%
December 2000 (Board of Supes Runoff) – 32.6%
November 2001 (City Attorney Race) – 29.6%
December 2001 (City Attorney Runoff) – 17%
March 2002 (Gubernatorial Primary) – 34%
November 2002 (Gubernatorial Election) – 50%
December 2002 (Dufty-Ma Runoff) – 38.4%
October 2003 (Gray Davis Recall) – 59%
November 2003 (Mayor and DA Race) – 45.67%
December 2003 (Newsom-Gonzalez Race) – 54.5%
March 2004 (Presidential Primary) – 42.5%
November 2004 (Presidential Election) – 74%
November 2005 (City Attorney Race) – 53.6%
June 2006 (Gubernatorial Primary) – 37%
November 2006 (Gubernatorial Election) – 60%

Filed under: Archive

Translate »