Using the results of Amendment 43 as indicating the proportion of opposition to gay marriage (56% to 44%) then using demographics it can be determined approximately when a majority will support the amendment’s repeal.
To do this I first use the report P20-556 by Kelly Holder of the U.S. Census Bureau studying the characteristics of the voting population. From it I approximate the voting population in Colorado as 9% of voters are age 18-24, 14% are 25-34, 19% are 35-44, 21% are 45-54, and 19% are over 65. This, of course, carries the risk that Colorado’s voting demographics are different due to our younger population.
According to the paper “Explaining Rising Support for Same-Sex Marriage in California” by Gregory Lewis of Georgia State University and Charles Gossett of California State Polytechnic University, Pomona almost all of the change in support for same sex marriage since 1997 is caused by cohort change. That is one generation dying off and new ones coming of age. They found this by studying the data of the Field Poll of Californians about the subject, which had been repeated six times since 1985 at the time of their study.
From an article the Sacramento Bee on most recent Field Poll in California I got the data for the most recent poll done this year. It found 68% of voters 18-29 years old said they favored allowing same-sex marriage. 58% of voters 30 to 39 and 51% of voters 40-49 favored gay marriage. That compared with 47% of voters 50-64 and 36% of those over 65 who supported the idea.
Now I do some back of the envelope estimations on Colorado’s population as of 2006. I estimated 69% support by those under 18 years of age, 67% for 18-24, 57% for 25-34, 50% for 35-44, 46% for 45-55, 35% for 54-65, and 27% for those over age 65. I got these percentages by using the actual vote in 2006 on Amendment 43, 855,126 against gay marriage and 699,030 for it, and then running that through my age cohorts with percentages until the numbers balanced (after rounding).
Then I started replacing age groups with younger ones with two-year intervals in my spreadsheet. (How I love spreadsheets!) From this I found that the break over point using these assumptions comes in election year 2014 when a very small majority supported gay marriage. I would guess to take Amendment 43 out of the constitution will take a few years beyond that, though.
My numbers gets to about 54% for to 46% against in 2020, twelve years from now, so I would guess that would be the safe date for a referendum or initiative given the random fluctuations of public opinion. I then ran this in a different way. My second run found a very, very slightly higher percentage for gay marriage, in no case was it more than 0.5%. The reason for this is that the first run each subsequent set of numbers depended upon the previous election cycles set. That meant rather than being exactly 69% for the youngest cohort in 2016 it was instead 68.6281311%. I knew this would happen and figured it would just make my numbers a bit more modest, so I’m inclined to go with my first run.
So my guess is that gay marriage will be legalized in Colorado in about six years at the very soonest and twelve years at the latest.
Estimate of numbers Against and For Same Sex Marriage: