Politics and Asset Prices
Applying probability theory like PTJ
The other day a reader was kind enough to send me a video of a terrific interaction between two hedge fund legends. Last month, Ray Dalio and Paul Tudor Jones sat down at the Greenwich Economic Forum to discuss a variety of topics.
There are so many aspects of this video to explore, but I want to drill down on one particularly interesting comment that Jones made.
"We did this poll internally about where the S&P would trade if Elizabeth Warren would become President, and then Biden, Buttigieg, Kobalchar, etc. And then we took the election probabilities. And so, our poll said that if Elizabeth Warren would become President, the S&P would trade around 2,250. It's at 3,050 now. But I am just saying that her policies would probably give you something like that.
Then if you just take election probabilities, that would imply... if you anchor that, I think Buttigieg and Biden were probably 2,700 - cause again, they are going to come and raise taxes and that's going to cause some economic contraction, so that would imply that Donald Trump, if he gets re-elected, the S&P is worth 3,600."
This is a fascinating intellectual exercise that the folks at Tudor did. Let's take a moment to understand their thinking, and then examine what it might mean for markets going forward.
This interview was conducted at the beginning of November 2019. I am unsure which election market Paul was using, but let's go with PredictIt:
On November 1st, the price breakdown of the leading contenders was Trump $0.40, Warren $0.23, Biden $0.14, Buttigieg $0.11 and Sanders $0.09.
Let's plug these into a spreadsheet and see if we can get the same numbers as Tudor.
I used Tudor's SPX forecasts for Warren, Biden, Buttigieg and Sanders. I assumed a Michael Bloomberg Presidency might be mildly positive, and for everyone else, unchanged.
When I put those numbers into my model, to make it balance, I actually need a Trump value of 3,942 - which contrasts Tudor's 3,600 call.
Now I am not sure whether I have different market election odds than Tudor, or if Paul was using extreme forecasts for effect in his chat. It doesn't really matter. The point is that the market is a discounting mechanism, and if the performance of the S&P 500 will truly be that catastrophic under an extreme-left Democratic candidate win, then it might mean that a Trump win needs to be euphoric to balance it out.
Here is my pushback to this model. Too often the market is not a good discounting mechanism. How many times have you seen a risk staring everyone straight in the face, yet the market ignores it? And then when that risk comes to be, everyone freaks out. I have spent the last 25 years trading against the market, and if it was so perfect, there wouldn't be those individuals (like Tudor) who have consistently beat it. So my complaint with this analysis is that it assumes participants are properly discounting the market risk if the far-left Democratic candidates win.
Another possibility? Maybe the 2,250 forecast for Sanders and Warren is too pessimistic. If we change the decline to a 20% bear market (2,454) and move Biden and Buttigieg to a 10% correction (2,760), we get a 3,741 balancing figure for Trump which is approaching more reasonable levels.
I think these assumptions are more realistic, so let's dig into this model a little more.
Since Tudor made this presentation, Michael Bloomberg has jumped into the race and both far-left candidates' odds have sunk. Could this be why the stock market has rallied?
Let's keep our stock market forecasts unchanged and see how much the model moves with the new odds:
By my rudimentary calculations, the S&P 500 should be 63 points higher (3173-3110) due to the change in election odds.
In this time frame, the S&P 500 is 43 handles higher (3066.91 to 3110). Now obviously this analysis is back-of-the-napkin voodoo. It's not supposed to be precise by any means.
Yet I think it's illustrative of how the election might be moving markets.
But even more importantly, as Paul Tudor Jones says, "as an investor, you have to have a view on the election because the outcomes are so extreme."
Lest your dislike of politics encourage you to ignore the election, Jones has a warning;
"[the President] does make a difference. Ronald Reagan, when he became President, he was a huge difference to the stock market. And I would say, who the next President is will also have a huge impact on the economy, the stock market and particularly asset prices cause clearly asset prices today; whether it be US stocks, interest rates, whether it's the dollar, because it's all priced off of, in my opinion, a 5% budget deficit with this incredibly stimulative fiscal policy, combined with this overly stimulative monetary policy is creating this US exceptionalism that one day - like if we normalized our deficit to where Europe is right now, we would have completely different valuations for the stock market, valuations for the dollar - the dollar would be substantially lower.
So, this next Presidential election, and what policies they pursue afterwards - this one is going to be more meaningful than any in my lifetime."
Paul Tudor Jones likes to apply probability calculations to his stock market forecast, where Ray Dalio claims to be more "mechanistic" about this analysis.
"If you take the way corporate taxes would be changed (the undoing of that), then that's worth 7% or 8%. If you take the changes in GAAP accounting, that's worth about 15% of earnings and would therefore drop [the stock market] on the basis of that alone. If you take then the wealth tax, which is $2.8 trillion over 10 years, and so that will have an effect, and then that doesn't have to do with regulation. I think we are dealing with the question - again like the 30's - is it populism of the left or populism of the right? And how is that going to work? So we have an issue in that politics will certainly matter to asset prices."
Either way you approach it, there is no denying politics matter. More than ever. I just doubt the market is doing a good job discounting it...
Thanks for reading,