Conversation

2) not investment advice. not political advice.

3) should you vote? let's do some math.

3:10 PM · Oct 18, 2020Twitter Web App

4) Structure: a) How much does the US president matter? b) How much does your vote influence the election? c) How much does it cost to vote? Put them all in $ Then compare a*b vs c

Replying to

5) (a) how much does the US president matter? I don't know! But let's ballpark it. The federal budget is something like $5T/year. That's $20T/presidency.

6) Maybe the president has 10% impact on it (the rest being congress, courts, electorate, advisors, etc.) Also maybe that's 50% of what the president does (?) So $4T impact. Are those right? IDK! But it seems like probably impact is between $1T and $50T.

7) (b) How much does your vote influence the election? Naively, there will be ~150m votes, and so each person has 1/(150m) impact. But in fact we think the election will be within a 20m vote band or so of 50/50; so each person is maybe 1/(20m).

8) We're not done yet! But actually we're pretty close. 1/20m is not a terrible estimate. Of course, it depends on your state. If you live in CA or WY, your vote will not swing the 2020 US presidential election. If you live in FL or OH, it might.

9) Florida will have ~10m voters, and will be within a ~6% band around 50/50; so your odds are ~1/600k. Florida has 29 electoral votes; overall total will be in a ~200 vote band. So maybe Florida has a ~15% chance of swinging the election. So a Floridian has a 15%/600k ~ 1/4m

10) This was really quick and dirty. How does it compare to 538? They have Florida as 13.8% to decide the election--pretty close! There could easily be a factor of ~2 error in this calculation, but seems like: --avg vote: 1/20m --vote in swing state: 1/4m

11) (c) ok, how much does it cost to vote? IDK, maybe an hour of your time? Median US household income ~$70k; say 2,000 hours/week of work. So that's something like $35 of opportunity cost to vote.

12) (a) * (b) ~ $4T * 1/4m ~ $1m in a swing state (and ~0 elsewhere) So that's $1m of gov't impact for $35 of your time.

13) Now you might think you can spend $ better than the government (and in particular better than the difference between Trump and Biden). How much better? Because if you live in a swing state, it's impacting something like $30,000 of gov't spending per $1 of your time.

13) You can adjust by your actual opportunity cost of time, and how close your state is projected to be. A very rough approximation: $30k * ($70k/income) * (% your state is close) (The electoral votes vs population terms roughly cancel out.)

14) Roughly speaking, the following states "matter": PA, FL, WI, MI, NC, AR, MN, NV, GE, CO, OH, NH, NM, NE-2 If you're in any of those, % your state is close is something within a factor of 2-3 of the above calcs. If not, probably you won't impact the presidential election.

Too Much Dark Money In Almonds

15) Draw your own conclusions from this. Mine are pretty similar to

slatestarcodex.com

Everyone always talks about how much money there is in politics. This is the wrong framing. The right framing is Ansolabehere et al’s: why is there so little money in politics? But Ansolabehe…