In this week’s volume of Metaculus Mondays, we are going to forecast three questions of geopolitical significance relating to China. In particular, we are interested in predicting whether:
- The 2022 Olympics in China are boycotted by the United States and her allies in the Five Eyes and Quadrilateral Security Dialogue.
- China closes its Uyghur internment camps in Xinjiang by the end of 2021.
- China lands the next person on the moon, beating the United States, Russia, Japan, and others.
But before we get to these new forecasts, we want to first provide an update on two forecasts which recently resolved from past volumes. The first was our Suez Canal forecast from last week, which we gained 8 points on. Yay! The second forecast resolution was far more interesting and deserves some discussion: How many vaccine doses will the EU Administer by April?
If you’ve kept up with all our volumes, you’ll know we had little faith that we would get this one right. In volume 6, we said “we can probably kiss our closest EU vaccination prediction goodbye (75M doses by April 1).” And in volume 7, we reiterated this assessment, saying “we already discussed what likely went wrong with this forecast in last week’s volume of Metaculus Mondays.”
When the question finally resolved, we naturally ended up being bang-on the mark.
What’s the deal, then? Was our original forecast strong, but we had low-self confidence and over-reacted to news developments? Was our original forecast weak, but we just got lucky? While the answer is obviously somewhere in the middle, we have strong confidence our forecast ended up on the right side of maybe.
A few reasons:
- Our forecast’s lower 25% interval equalled or exceeded both the Metaculus and community median forecasts.
- The actual doses administered, 76.6M, exceeded both the Metaculus and community’s 75% upper intervals.
- The difference between the community and Metaculus median from the actual doses (10.6M - 11.6M) was less than the difference between our upper 75% interval and the actual doses (10.4M).
This only addresses whether our numbers were strong. So what about our forecasting process? Was that also correct or were we mere chimps lucking out with a copy of King Lear?
We’re inclined to say that our original forecast was fairly sound, and we instead (very clearly) over-reacted to the AstraZeneca developments in Europe. Why?
In our original forecast, we gave a starting point of analysis based on past rates of EU vaccinations as well as the Metaculus median forecast at the time. Then, we presented two hypotheses that either accepted or rejected the community’s bearish forecasts and examined the relevant bearish and bullish indicators to accept or reject either of those hypotheses.
What was the result of this process? We believe that by basing our original forecast off of Metaculus and base-rate vaccinations, we accounted for bearish possibilities with our Metaculus forecast (such as pausing vaccinations) while also accounting for bullish possibilities and signals that the community missed. Therefore, the factors we said we missed in volumes 6 and 7 were actually ones already built into our forecast, meanwhile the community and Metaculus aggregate forecasts missed facts we inputted.
With the past out of the way, let’s turn our attention to the future and get to some forecasting about China.
Question 1: 2022 Winter Olympics Boycott?
The first question is asking us whether at least 4 of these 7 countries will boycott the 2022 Olympics in China: Canada, USA, UK, Japan, Australia, India, New Zealand. In order to forecast this question, we are going to start our forecast assuming that if a boycott happens it would be significant enough to include at least 4 of these countries. At the end we’ll make a minor adjustment to account for the slight possibility of a boycott with fewer than 4 of these countries.
Why are we taking this approach? Firstly, China has stated it would sanction any country who boycotted the 2022 Olympics, meaning the costs for an individual state to boycott would be extraordinarily high. Meanwhile, analysts have written about the Quad–US, Japan, India, Australia–becoming increasingly coordinated in countering China and the United Kingdom has more-or-less tied itself to the US in terms of foreign policy according to their recent Global Britain plan.
The two states we could see acting independently are Australia (lone with respect to a boycott, due to history of standing up to China despite facing real consequences) and New Zealand (against a boycott, but unlike Australia this is mostly intuition and therefore barely weighted at all).
A good starting point for this forecast is determining what the base-rate for boycotting the Olympics is. According to Wikipedia, there have been 58 Olympic Games since 1896. Out of those 58 Olympic Games, 6 of them had boycotts useful for a base-rate calculation since they were due to human rights or geopolitical reasons.
- 1956 (Australia): 8 countries out of 102 (8%), with 1 great power (13% of boycott)
- 1964 (Japan): 3 countries out of 143 (2%), with 1 great power (33%)
- 1976 (Canada): 36 countries out of 166 (22%), with 0 great powers (0%)
- 1980 (Soviet Union): 66 countries out of 171 (39%), with 2 great powers (3%)
- 1984 (United States): 18 countries out of 171 (11%), with 1 great power (6%)
- 1988 (South Korea): 7 countries out of 171 (4%) with 0 great powers (0%)
Note: Number of countries is determined by the total number of recognized sovereign-states by the end of the decade according to Wikipedia. This was used out of simplicity of gathering data.
Although 1936 is on the Wikipedia page, it was not included in our base-rate calculation because the two countries that boycotted had external reasons:
- Spain withdrew due to civil war
- The Soviet Union didn’t participate in any summer Olympics until 1952
That leaves us with 6 boycotts out of 58 Olympic Games, leading to a base-rate of boycott of ~10%.
Out of these 6 boycotts, 5 were due to geopolitics while 1–the 1976 boycott–was a mix of human rights and geopolitics reasons. Therefore, the base-rate for geopolitical boycotts is ~8.6% while the base-rate for human rights boycotts is ~1.4%.
Moving from the outside base-rate and to the specific case of boycotting the 2022 Olympics in Beijing, what are the reasons why countries would boycott broken down by category (geopolitical vs human rights)?
Geopolitical reasons to boycott Beijing Olympics:
- China starts a war or major cyber conflict with any number of countries (India, Taiwan, etc.)
- China annexes territory from another country (Taiwan)
- Convincing evidence emerges that COVID-19 originated in a lab and China covered it up
- Significantly escalated trade war (including sanctions / increased sanctions on countries such as Australia)
- Global South debt crisis, with geopolitical conflict due to Belt and Road loan conditions
Based on Metaculus forecasts (and our own), the likelihoods these happen on the magnitude needed for a boycott before the 2022 Olympics are:
- War or Cyber Conflict: given low forecasted likelihood of any lethal conflict between India and China by December 1, 2021, plus a subjective geopolitical analysis: 1%
- Annexes territory: Given an even distribution over the remaining years on this Metaculus forecast: 1.38%
- COVID Origins: Given our current forecast and adjusting for an end of 2021 resolution: 14.5%
- Trade War: 1%
- Major Debt Crisis: 1%
End Result: 19% Likelihood of a Geopolitics-related Boycott (or 2.25x the base-rate).
Human Rights reasons to boycott Beijing Olympics:
- Xinjiang and Uyghur Human Rights abuses (most likely)
- Hong Kong Crackdown (relatively likely)
- A violent crackdown in another region (unlikely)
- Any of the other alleged major human rights abuses by China (i.e. organ harvesting) going mainstream with evidence (unlikely)
Unlike the geopolitical boycott analysis, we instead consider the human rights boycott in relative terms of the base-rate given the low probabilities involved. We consider the following signals in our analysis: Google trends data in the relative 7 countries on a China boycott relative to the rest of the world; shift in global public opinion against China, including one poll which found 54% of Canadians were in favor of boycotting the 2022 Olympics (compared to 24% opposed); and the Biden Administration’s successful implementation of Western sanctions on China over the Xinjiang camps.
Let’s consider the impact these reasons have on the likelihood of a human rights boycott of the 2022 Olympics through three lenses:
- Conservative: 1x base-rate, considering strong economic ties to China, statements including those by the US Olympic Committee claiming boycotts are ineffective, and the fact that human rights were never used as a reason for a boycott against the Olympics in a Great Power country: 1.4%.
- Moderate: 3x base-rate, considering a significant but muted impact of recent attention on China’s human rights abuses: 4.2%.
- Aggressive: 5x the base-rate, considering a strong impact of recent positive signals: 7%.
At this moment, we will average the three lenses equally, leading to a blended probability of a human rights-related boycott of 4.2%.
Taken in aggregation, we give a 23.2% chance that four of the following countries–United States, United Kingdom, Canada, New Zealand, Australia, India, Japan–will boycott the 2022 Beijing Olympics.
Given that this question opens on Wednesday, if you disagree with any part of our analysis or feel we missed something, please comment below or reach out to us on social media! We’d love to have as strong of an initial forecast as possible.
Based on recently developing news that the United States is considering a 2022 Beijing Olympics boycott with allies, we've updated our forecast.
Method: We are currently weighing this threat as a 1.4x multiplier on our base forecast for a 2022 boycott. Re-normalizing our probabilities returns us with a 29.7% likelihood.— Global Guessing (@GlobalGuessing) April 6, 2021
(We are still debating the overall effect of this news and watching to see further developments.)
Question 2: Uyghur Camps to Remain Open?
Staying on the topic of Chinese human rights, the second question in today’s Metaculus Monday is about the Uyghur internment camps being run by the PRC government.
More specifically, this question asks whether these Uyghur internment camps will still be open after January 1st, 2022. Before we get into this prediction, however, it’s important to have an understanding of what is happening in China with respect to what has been called by some, a Uyghur ‘genocide’.
The Uyghurs are a Muslim minority that live in live in Xinjiang, an autonomous region in northwest China.
- Since the PRC was established in 1949, China has had control over the region.
- With a population of roughly 10 million, the Uyghurs are culturally discrete from Chinese society, however, as they speak their own language and are, predominately, practitioners of Islam.
Much of the tension between China and the Uyghurs comes from the geographical advantages of Xinjiang.
- Xinjiang is rich in both oil and other natural resources. It is also situated between Kazakhstan, Kyrgyzstan, and Mongolia making it extremely advantageous for trade.
- As such the PRC has encouraged migration of Chinese citizens from China proper into Xinjiang in order to better leverage these advantages.
Tensions began to escalate in 2013 when the PRC announced their ‘Belt and Road Initiative’ which resulted in a more concerted effort to gain greater control in Xinjiang, and 2014 tensions resulted in riots in Xinjiang’s capital, Urumqi, leading to hundreds of deaths.
Since then, China has employed an incredibly repressive regime in Xinjiang to control the Uyghur population, painting them as extremists and separatists who need to be reeducated. And this leads us to our question.
- The main pillar of this reeducation has been the use of internment camps (now referred to as Vocational Education and Training Centers by the Chinese government) where Uyghurs are subjected to rough living conditions, physical and sexual abuse, and other forms of religious persecution.
- And despite these alleged abuses, China formally legalized the existence of these camps soon after their creation, in 2018.
Since these camps became operational in 2017, there has been waves of international pressure against China to close these camps, without much recourse. So, this question is asking us if this trend will be bucked, and if China will close their camps by the end of this year.
The current community median prediction for this question on Metaculus is that there is a 95% likelihood that the Uyghur internment camps in Xinjiang remain open past January 1st, 2022, with 43 predictions registered thus far. Given the extent of the pressure already put on China to close these camps, and the country’s seeming apathy in response, led to to believe that this number is in the ball-park of what we would intuitively predict.
Compared to the community perspective, however, we felt that there were slightly more factors drawing down the likelihood of camps remaining open, namely our forecast on the chances of an Olympic boycott in 2022, and the odds that Covid-19 originated according to the ‘Lab Leak’ theory.
If China both believed that a boycott of its Olympics in 2022 was a credible possibility, and cared enough about the potential of a boycott, we think that there is a small chance (read very small) that China would acquiesce to international demands to open their camps. Hosting the Olympics is a massive honor, and China might not want to jeopardize the integrity of the event, or its ability to host it altogether.
With respect to the ‘Lab Leak’ theory, if truly damning evidence comes out this year that Covid-19 came from a Chinese lab, it is feasible to imagine that China may want to deflect attention away from its missteps by dismantling the Uyghur internment camps in a ploy to garner favor with the public.
While both of these possibilities are viable, there are also a number of factors that would draw down the likelihood of either from occurring, including public pressure from Chinese citizens to keep the internment camps operational out of a desire to retain national identity.
Given the multitude of factors that could also play a role, we decided to attribute a 1% likelihood to the two most probable scenarios, being the Olympics boycott and “Lab Leak’ theory, and discount the community median accordingly.
This approach yielded a 93% chance that the Uyghur internment camps remain open by the end of the year.
Question 3: China's Race to the Moon
For our third and final China-related question this week, we decided to predict if China will be the next country to land a person on the moon.
While the least-tied to geopolitics directly of the questions we approached this week, space travel still has great importance for countries. There are already treaties in place regarding space exploration that have been signed by several countries, and in 2019 the United States Space Force was established to handle security issues in space on behalf of the U.S. So while the future of space travel might seem a bit esoteric, it does have geopolitical implications that will become increasingly important in the future.
Other than China, our preliminary research showed that the United States was the only other true individual competitor to be concerned about. The other actors that we factored into our analysis were Russia, India, Japan, and SpaceX, but it seems more likely that in the time frame of the question, these actors will engage in orbiter landings as opposed to human landings. That’s not to say they don’t matter, but as of now we only assign a 10% combined likelihood that these groups land the next human on Mars.
been announced by the United States and China (Russia also a mission planned for the 2030s, but . The United States announced in 2019 that it would return humans to the moon by 2024, a timeline that has been viewed as highly doubtful from the beginning.
Then the Artemis program was announced, a U.S.-funded human spaceflight mission to send the first woman and next man to the moon. Led by NASA and the United States, the mission has already experienced some delays, however. NASA announced in January that it would be putting a pause on its search for a lunar lander developer which will likely cement the Artemis program’s delay beyond 2024.
Competing with the Artemis program is China, who has been discussing an unnamed mission back to the moon in the early 2030s. Unlike the Artemis program, which heavily involves international partners, China’s plans leverage only their own technology and research. This could be a boon to their space ambitions, as fewer actors means less negotiating and fewer potential hurdles. Working alone also means no outside support, which could slow down the process as well.
We also considered the potential of China working alongside Russia to beat out the United States’ timeline for the Artemis program as the two countries have discussed extensive collaboration efforts in space, leaving NASA and the United States out.
To build a forecast perspective we looked at some questions on Metaculus that have covered related questions.
Our first task is to understand how far delayed NASA’s upcoming moon landing is. This question on Metaculus asks if NASA’s Space Launch System (SLS) will launch a person to the moon before 2024. With over 200 predictions registered, the current community median forecast is 3%, which aligns with our view that the deadline proposed by Trump in 2019 is not realistic.
The second question asks whether NASA’s SLS will land a human on the moon before 2025. With over 1,000 predictions, the median forecast is 10%--indicating a potential 7% likelihood NASA will land a person on the moon in 2024 alone.
The final question on Metaculus directly asks when NASA’s SLS will launch a person to the moon. With 125 forecasts, the median community forecast is April 19, 2028.
Taken in aggregation, we estimate NASA’s SLS will land a human on the moon between 2027 and 2028.
How does China compare? Given China’s goals of reaching the moon in the early 2030s, and the few improvements they would need to speed up that schedule, we believe that China could put a human on the moon by 2029-2030. We are aware that there was a report released by Reuters in 2016 stating that, according to a senior space official, China has goals to put astronauts on the moon by 2036. However, according to spacenews.com, China is expected to release a new space white paper by the end of 2021 which “will more clearly lay out China’s plans for the next five years and may detail the status and extent of crewed lunar plan developments.” Depending on what is in the white paper, our forecast is subject to change.
As of now, we want to favor the outside chance to account for uncertainty and China’s rapidly growing technological capabilities. As a result, we give the United States 2:1 odds over China in terms of landing the next person on the moon.
As stated earlier, we give a 10% combined likelihood that SpaceX, Japan, India, Russia, or any other entity land a human on the surface of the moon before the US or China. Next, we split the remaining probabilities between the US and China by a 2:1 ratio.
We therefore forecast a 30% likelihood that China lands the next person on the moon.
If you made a forecast on any of the questions we covered in this volume of Metaculus Mondays, feel we made a mistake in our analysis, or believe we missed an important signal, please comment down below or reach out to us on social media! And don’t forget to sign up with your email address below to get Metaculus Mondays hitting your inbox every week.