We’ve now forecasted five times whether the US will return to the JCPOA (Iran Nuclear Deal) by the end of 2021. In this week’s volume of Metaculus Mondays we want to do a different kind of forecast—one designed to make us rethink and question our current assumptions. So instead of updating for the sixth time, we are going to make a new forecast explaining a discrete, theoretical world where we very much got this question wrong—and there is no blaming the #fattails.
Far from the world that gives a meager 30% chance, this is the world where the best forecast would give an 80% likelihood the US returns to the JCPOA by the end of the year.
But before we get started, in case you missed it, we published two great podcast interviews last week.
- Want to learn from a top Metaculus and Good Judgement COVID Forecaster? The Right Side of Maybe with Juan Cambeiro (transcript available)
- Interested in listening to a former Australian diplomat and co-founder of a new geopolitical media company? Global Guessing Weekly Podcast with John Fowler
Our Original Case
On February 15, 2021 in the second volume of Metaculus Mondays, we gave a 20% likelihood that the United States either lifts or waives sanctions on Iran greater or equal to those mandated under the original JCPOA or negotiates a new deal with Iran containing sanctions relief. At the time, the community median forecast on Metaculus was 55%; on Good Judgement Open it was 80%.
Our original forecast was primarily based on the following factors:
- Given the time it took to negotiate the original JCPOA (2 years) and the time it took Trump to pull out of the JCPOA (18 months), we found it highly unlikely ( < 5%) a new deal would be negotiated by the end of the year — Result: Only worth focusing on whether the Biden Administration would lift or waive sanctions on Iran.
- We next looked at the constraints on both Iranian and United States decision-makers. On the highest level, these were well telegraphed. For Iran: The US lifts all sanctions unilaterally. For the US: Iran returns to the terms under the original JCPOA unilaterally.
- After highlighting these first-order constraints, we looked at three second-level constraints that could dictate the outcome:
- Political: What can Biden and Rouhani accomplish before the June elections? How might the election outcome affect the chances? What can Biden reasonably accomplish unilaterally?
- Time: How long would it take to rejoin should both sides agree to a return to the deal?
- Geopolitical: How have Iran’s geopolitical priorities and positioning within the Middle East changed since 2015 and how might that affect a return to the deal?
Given all of this, now might seem like an odd time to make a pre-mortem forecast about a world where we are off on this forecast. Since our original forecast, a lot of the constraints we identified came into play—and at the same time, we’ve seen both the Metaculus and especially the GJOpen communities approach our forecast (which we’ve actually increased over time) rather than increase upwards.
Yes, for months diplomats have been saying “we’re almost there with a deal” before following up with a “but, actually, just FYI...a lot of difficult points still need to be addressed”, and yes, hard-liner Raisi won in the recent Iranian “elections”. However, we feel it is precisely because of these developments that it is best to take a hard look and see if we’ve missed any big pieces.
Point: Hardliners in Iran have complete control of the government. Their recent election was likely influenced in their favor, and Raisi has been reported as Khomeini’s successor putting him in a very strong domestic political position, allowing him to act unilaterally and get an early win in his global political career.
- Counterpoint: Raisi doesn’t need a political win given his strong position, making a deal with the United States an unnecessary political risk.
Point: China has shown an interest in working with Iran in recent times. This is likely due to the strong economic and geopolitical potential of Iran, as well as its adversarial orientation towards the United States. It is possible that lifting sanctions would more quickly allow Iran to pivot towards China and later Russia.
- Counterpoint: Alternatively, Chinese interest in partnering with Iran may encourage a scorched earth approach to diplomacy from Iran, leaving the United States negotiations entirely.
Point: Iran is in a very strong regional position in the Middle East right now: they beat Saudi Arabia in Yemen and have engaged diplomatically with the Saudi kingdom recently; they drummed up anti-Israeli sentiment in the US with the recent Gaza-Israel war; they’ve held talks with Iraq and conducted clear signs of statecraft. The economic boost from sanctions relief could help make these geopolitical wins, long-term sustainable.
- Counterpoint: Iran has been able to accomplish all of this under United States sanctions, perhaps giving up long-term nuclear potential could limit the country from leveraging its improved geopolitical position. Perhaps its geopolitical wins of late are predicated on a long-term vision of Iran that is nuclear.
Point: The short-term economic benefits of lifting sanctions would likely outweigh the short-term benefits of a nuclear weapon for Iran; the Iranian situation is different from North Korea’s.
- Counterpoint: However, short-term gains may result in long-term vulnerabilities for Iran’s position in the Middle East from a security perspective.
Point: Biden wants a foreign policy achievement that stands out from the Trump administration, and is committed to getting the deal done regardless of long-term benefits.
- Counterpoint: Examining this point through a Papic-ian lens, Biden’s preferences may not outweigh his ability to work with congress, Iran, and the other negotiating parties to get a deal across the line in time.
Blending the Forecasts
So how do we blend our current JCPOA forecast with our pre-mortem on this question?
We are currently at 30% on this forecast and we initially planned to drop our forecast to 20%. We feel, however, that the lack of material changes in the situation on the ground regarding JCPOA negotiations, and the impact of our pre-mortem are not significant enough to warrant a 10% drop.
Instead, we are dropping our forecast for this question to 25% and will continue to evaluate developments on this issue.
Ultimately, the effect of this forecast vis a vis our pre-mortem is that we are lowering our forecast at a slower rate than we otherwise would have. We also want to be on the lookout for any signals that would indicate this alternative forecast is the correct one (or more of a correct forecast than we are currently saying).
What would those signals look like?
- New signs of Iranian economic vulnerability that would require a stable solution such as Iranian-United States diplomacy as opposed to illicit financing from a ‘revisionist’ power.
- Signals from Raisi that he is genuinely interested in working with the United States to support Iran’s revival, as not as a political victory
- Either Iran or the United States lowering their bar for cooperation
What would signals look like indicating this new forecast is less correct than we are currently saying?
- China or Russia making a strong financial commitment to supporting Iran which would preclude the need for United States sanction relief
- More time passing without material updates on the JCPOA negotiation process
- Joe Biden’s domestic political positioning requiring a political win i.e. playing a tough hand with Iran
If there are other signals that we should be mindful of, be sure to comment below or reach out to us on Twitter!
COVID Origins Update
Finally, we want to conclude this volume by updating our Covid origins forecast due to a similar rethinking process. Although there have not been any material changes since our last update, we have decided to lower our overall Covid origins forecast from 82% to 71.4%. This does not change our current Metaculus forecast, however, due to us also decreasing the political discount factor in our prediction (50% -> 75%). While we actually have little remaining confidence that political considerations will prevent a declaration of lab-related origins should the relevant health agencies reach that conclusion, we still want to discount our overall forecast to account for the limited time before resolution.
So why the decrease? Because Zeynep’s recent post made us rethink our overall confidence in our forecast. As she reminds us, with the exception of during the initial months of the pandemic there has been an information drought due to an active cover-up by the Chinese government. The result, she says, is that “going forward, we should consider all possibilities as potential pathways and think about what can be done. I people don’t [sic] understand the evidentiary basis on which people are assigning precise(ish) likelihoods to the paths leading to a rare, perhaps once-in-a-century rare, event when a lot of paths have been demonstrated as viable.”
While we disagree with the notion of assigning precise likelihoods (it’s kind of our schtick), we now believe our previous 1-in-5 likelihood for non-lab origins was too bearish given the circumstances. As a result, we reduced our forecast by ~15% and remain unconvinced a lab-related explanation is 50-50 or sub-50% likelihood given the signals currently driving our forecast.
As news around this question is constantly changing and evolving, we will be keeping a close eye on new developments. As far as the impact of the questions which we forecast go, this one ranks among the highest, and as such deserves both ample attention and a critical eye.