On November 6, I set out to cast my vote for the President of the United States. Throughout the day I was asked if I had voted yet and I would reciprocate. One friend explained to me that he did not vote because he lives in a state that historically votes Democrat; therefore, any vote for a Republican would not affect the outcome. I agreed. In California this year there were 3,685,045 votes for Romney and yet all the state’s 55 electoral votes went to Obama. In Texas, there were 3,294,400 votes for Obama and that did not alter the fact their state’s 38 electoral votes went to Romney. Looking at those numbers could lead one to conclude that a singular vote doesn’t count for much.

Later that night as I was watching the election results, I heard something that caught my interest.  Some political pundit said, “this just in… Maine’s 4th electoral vote will be for President Obama.”  Wait, what did he say?  As it turns out, both Nebraska and Maine do not follow the “winner take all” electoral college system. Instead, they award an electoral vote to the winner of each congressional district and then an additional two electoral votes to the candidate that wins the popular vote. For example, Maine has 4 total electoral votes and a total of 2 congressional districts. This concept of not using a winner take all system got my mind to thinking – this system gives people in a state with strong political affiliation a chance to influence the election even if they don’t win the popular vote.

So I decided to perform a little exercise with the great state of Ohio. When looking at the voting results by county, the map appears predominately red even though the result of the popular vote favored Obama by 103,519 votes. The race is a lot closer than the 18 electoral votes would indicate.  Although the results by district are available nationwide for the 2004 & 2008 elections it has not been released for the 2012 election as of this writing. So, I took this one step further and attempted to apportion the votes to the boundaries of Ohio’s 16 congressional districts.*  According to this map,  Romney would have won nine of the sixteen electoral votes if the ME (Maine) model was used. Obama would have also won an additional two electoral votes since he won the popular vote in Ohio. This would have yielded a 9 – 9 split of Ohio’s electoral votes.

Data by congressional districts is available nationwide for the 2004 & 2008 elections. The below table summarizes what the results would have been assuming all states awarded electoral votes by congressional districts. Please note the numbers in the district column also include the two additional votes the candidate was awarded for winning the popular vote in that state.

**Won more districts than the candidate that won the popular vote in that state

Although the end results of the election did not change; the margins by which the victorious candidates won did shift. The 2004 election was not nearly as close as it seemed and in 2008, McCain put up a much better fight than the electoral college gave him credit.

While this “new” system seems to improve the importance associated with a single vote, this method of awarding votes is deeply flawed…especially in a state like Ohio.  The biggest problem is district gerrymandering – fabricating strange district boundaries to favor one party over the other. The best way to resolve this would be to create districts using established geographic boundaries in conjunction with Census data. Maine, for example, uses county and township borders.  A similar system would likely need adapted to make this system feasible. That topic will be address by us at another time. Coincidentally, Ohioans voted down a measure to change the State’s constitution to create a board to oversee district boundary creation.

What do you think?  Is there a better way to work within the constructs of the electoral college or is it simply an idea that has outlived its usefulness?

-Matt Hamaide, Senior Consultant, Urban Decision Group (UDG), LLC

Footnotes and References

**In an effort to reapportion the district data from the 2012 county data, I complied a list of all the counties either full or partially within their respective district boundaries. I then added up all the votes reported for each county to get district totals. The first method included all the votes in a partial county even though not all of them would fall within the district. The second time, I only included 50% of the votes of any county that was only partially in the district boundary. I then compared both. If both methods’ totals awarded the district to the same candidate, then that is who was given credit for that district. Only in Districts 5 & 6 did the totals disagree. For both, the district was awarded to the winner of the second method in which only 50% of the total votes of partial counties were included. For example, District 5, the City of Toledo (the bulk of the votes for Lucas County) was not included in the actual boundary of the district. The same was true for District 6 – Youngstown was not included within the boundary (the bulk of the votes for Mahoning County). When considering this, it appears the partial totals were more accurate.

Data and Boundary Sources:
Ohio Secretary of State
New York Times