I’m not sure how to think about how the market will react to the election and subsequent turmoil.
The EMH tells us all the information should already be priced in, but I don’t quite see how. Either I’m wrong about what could possibly happen or the markets won’t be affected by these scenarios the way I think.
There are roughly three outcomes of the election:
Trump wins.
Biden wins in a landslide.
Biden wins in a non-landslide.
Trump probably has a chance of around 10-15 percent of winning. In which case Biden will concede and the left will have some protests (probably not more significant than the George Floyd protests).
But if Biden wins, it’s a whole different story. 538 puts a Biden landslide at around 30%, in which case Trump will likely concede. Protests will be not huge, I expect.
But the other 55-60% is where it gets interesting.
Trump has repeatedly said that he won’t concede as he can only really lose if fraud occurs. He laid the grounds for his followers to distrust the results of the election. Going as far as possibly planning to call the election for himself prematurely (although I’m not sure how trustworthy this particular information is).
Weird outcomes, where Trump looks like he’s winning on election night but ultimately loses are quite probable. This is due to democrats being more likely to vote by mail, which has become far more common because of the pandemic.
In the end, it’s quite possible that millions of die-hard followers of Trump will feel as if the election was stolen from them. And Trump will only stoke their anger.
(Which also scares me, since they are typically well armed and many of them have already sympathised with armed militias.)
Even if not likely: What would happen to stock markets, when 20% of the country with the largest economy are sent into a state of civil unrest?
No good business is going to stop being a good business no matter the election result and no bad business is going to be suddelnly healed. Just ignore the noise.
The way I understand the EMH (feel free to correct me), it tells us that the probability of each occurrence is already priced in. After the elections, using your numbers, a 10-15% of scenario 1, 30% of scenario 2 and 55-60% of scenario 3 will have turned into a 100% chance of one of them. That could have a (short term) impact on the markets (they’ll probably recover so planning on a reaction in either direction is only betting on one of these outcomes so, a bet).
A fringe scenario would be that we are witnessing history and the U.S. as we know them will no more be the U.S. as we’ve known them, in which case, all bets regarding the U.S. stock market are off.
It could be an argument for investing in a worldwide index rather than overweighting the U.S. As swiss investors, I doubt many of us are doing it (overweighting the U.S.) right now, so we should be ok (on the financial side of things, since other things could happen on the political and social sides).
So Biden has a 2:1 advantage over Trump, 66 vs 34%. If you seriously believe Trump has no more than 10%, go bet money against it, and you will be rich.
USA has a history of civil unrest, what is happening now is nothing new, I think. If Trump wins, we know what to expect. If Biden wins, we can expect more social welfare programs, more decisions that are friendly to big corporations (globalization), especially big tech. You can also expect over 60% marginal tax rate. Truly motivating for people to go the extra mile
No, prediction platforms are a bad place to look.
They are very far from efficient markets, as there are loads of fees and regulations (you can’t immediately pull your money from the platform, or you pay additional fees, etc.).
Nevertheless, I believe they are so far off that I would actually be betting money on it, but you can’t do it on PredictIt as a non-US citizen.
And I couldn’t find anything for Switzerland I felt comfortable with. With betting limits, the expected payoff of such a bet is likely below $100. So unlikely that it is worth my time to dig much deeper.
If you look purely at polling, this statistical model from Andrew Gelman is probably state of the art:
It gives Trump only a 5% chance, which I believe is quite overconfident.
If you want to allow for more uncertainty (of polling errors and in general) and fatter tails for the probability distribution, Nate Silver of 538 has arguably the best model.
They currently give Trump a 10% chance.
I kind of informs my point estimate the most. 538 has quite a good track record. And was e.g. called crazy in 2016 for giving Trump odds as high as 28% to win.
If you look at hobby-forecasters and superforecasters, you are at the around a 15% probability of Trump winning.
I lean on that interpretation aswell. If the unrest doesn’t reach unprecedented levels, the effects will likely be social and not economical.
But hasn’t the response of the stock market to the pandemic kind of shown that stock prices don’t really correlate that well with real economic developments?
Well, that’s weird. My chart comes from Smarkets.com. Odds for Biden are 1.52, against 1.53, and GBP 12 million were traded. That seems pretty efficient to me. Don’t know about any fine print, but if you’re so sure Trump only has 5% chance, maybe you should really consider to bet 1’000 on Biden, just for “fun”?
All the models are nice, but it’s all talk. Here people bet 12 million and put money where their mouth is. And I’m sure they know about these models etc.
I don’t believe he has a 5% chance, as I wrote.
If I had to give a point estimate it would currently be 12%.
I’m currently too busy to look into it (all the fees involved with depositing and withdrawal, regulations, how safe the bookie is, etc.) and not super liquid (with higher risk aversion) but thanks for the suggestion.
[When I looked into it a while ago you were typically only to make a return of below 10%. Which doesn’t earn you a lot with a betting limit (for example PredictIt has a limit of $800).]
Exactly, nobody can make any significant money off of the market if it were inefficient. No Bill Ackman can swoop in and fix it by betting a million dollars.
It’s basically just people playing with their money.
If you want, I’d be happy to bet with you two on this, just for fun and a small amount. (But I’m not sure if that would violate some terms of services of this forum or swiss law.)
Nah, it’s a zero sum game. One of us would be bitter in the end. But just to get this right, you would be willing to pay us 800 if Trump wins, and we pay you 100 if Biden wins. This implies Trump has 12.5% chance. Tempting
Fair enough! Betting makes me also a bit nervous. But I sometimes try to “put my money where my mouth is” in an attempt to get more accurate beliefs about the world.
Nah, I’d want more than that.
I have to incorporate the chance one of you flunk into my expected returns and I don’t have loads of money atm. So losing $100 hurts more than gaining them
Most probably true for many reasons.
But it’s also decreasing returns to wealth/spending. Hopefully won’t be such an issue in a year or so but at the moment I’m still a student.
Fun fact:
There’s actually a way to figure out how much of you money you should spend on any given bet.
By reading and partipating to this forum, you confirm you have read and agree with the disclaimer presented on http://www.mustachianpost.com/
En lisant et participant à ce forum, tu confirmes avoir lu et être d'accord avec l'avis de dégagement de responsabilité présenté sur http://www.mustachianpost.com/fr/
Durch das Lesen und die Teilnahme an diesem Forum bestätigst du, dass du den auf http://www.mustachianpost.com/de/ dargestellten Haftungsausschluss gelesen hast und damit einverstanden bist.
).
and I don’t have loads of money atm. So losing $100 hurts more than gaining them 
(neither would I)