In the final installment of our election mini-series, we move from the African continent towards Europe where we will be predicting whether incumbent President of Portugal Marcelo Rebelo de Sousa will win at least 50%+1 of the vote during the Sunday, January 24th election.
Marcelo de Sousa is currently the overwhelming favorite to win the Sunday, 1/24, election for president in Portugal. With a calculated polling average over 63%, Global Guessing gives President Marcelo de Sousa a 99.42% chance to win, making him nearly certain to win.
On average, President Marcelo de Sousa will win 59.94% of the vote with an 89% confidence interval [55%, 65%].
At the same time, de Sousa tested positive for COVID-19 6 days ago, Tuesday 1/12. According to Hoffmann and Wolf (2020), the case fatality ratio of COVID-19 in men over the age of 70 in Portugal is ~12%. There is ample evidence that heads of state and high-ranking politicians have access to medical care that undoubtedly lowers their risk of death. Moreover, there have been no reports that de Sousa has developed severe symptoms. Nevertheless, de Sousa is still not without legitimate risk of death.
Because of de Sousa's current infection of COVID-19, Global Guessing predicts there is a 97.67% he is sworn in for a second term.
Should de Sousa not develop severe symptoms by election night, we would put his odds of being sworn in at >99%. (And were de Sousa without infection from COVID, his odds of victory and being sworn in would be 99.67%.)
Marcelo de Sousa won the presidency in 2016 as an independent, appealing to moderate perspectives on Portuguese society. He ran a platform of mending the damages done during the austere years of Portugal’s economic history, between 2011 and 2014. His conservative politics have resulted in several public wins during his first term in office, chiefly his handling of the Covid-19 pandemic and his calling for a State of Emergency in November. Although Portugal has handled the crisis relatively well for Europe, de Sousa recently warned that stricter measures might be necessary as the crisis worsens.
Andre Ventura is currently polling in second place. Ventura is a Member of the Assembly of the Republic, Portugal's parliament, and is president of the CHEGA Party, a nationalist party. Andre's candidacy is supported by the French nationalist Marine Le Pen.
Ana Gomes is a former diplomat and a member of Portugal’s Socialist Party. She also held the role of Member of European Parliament from 2004 to 2019 when she stepped down to pursue domestic politics in Portugal. Gomes’ policy focus has largely revolved around issues of immigration and human rights, both of which have come to the fore of Portuguese politics in recent weeks.
Portugal’s geopolitical significance is framed largely by its geographic context. Portugal is historically a sea-faring nation facilitated by its Western border which sits adjacent to the North Atlantic Ocean. In the 15th and 16th centuries Portuguese exploration was so comprehensive that it reached not only nearby Spain and North Africa, but also India and Brazil as well. Today, Portugal’s geopolitical situation is, in many ways, characterized by this same spirit of maritime exploration as well as its colonial past.
Portugal’s proximity to the Atlantic Ocean has made it a prime partner for maritime trade and increased its sovereign territory. Portugal maintains control over two autonomous zones off their Atlantic coast, namely the Azores Islands, a nine-island archipelago, and Madeira, a nearby-four-island archipelago. Both have been the focus of Portuguese foreign policy for decades, and will likely be at the fore of Portugal’s next administration.
The Azores Islands have been a tool of diplomacy between Portugal and the United States since World War I, with the United States establishing a military base there that smoothed Portugal’s entry into NATO after World War II in 1949. In recent years, however, the islands have been used by Huawei to test new 5G technology, putting Portugal in a proverbial ‘geopolitical pickle’.
Madeira, while less directly-related to Portugal’s bilateral relationships, still holds geopolitical significance. Madeira and the Azores Islands are both part of recent plans announced by Portugal to extend its maritime borders, solidifying its status as the third-largest Exclusive Economic Zone in the European Union.
One of the early members of the European Union, joining only after Denmark and Greece, Portugal has long been a key ally in Europe. While in 2020, Portugal’s GDP ($175 billion) was only ~1.17% of the total European Union GDP ($15000 billion), its Portugal’s strategic relevance that makes it such an important player in the organization.
Portugal has maintained positive relationships with its formerly colonies, namely Brazil and Angola. These ties provide the European Union with influence in areas of Africa and South America where their reach was previously insufficient. The European Union has strong ties to North Africa through France, but Southern Africa has been.
Portugal’s role in the EU has also been highlighted by its past economic plight. A combination of high debt due to the introduction of the Euro in 1999, and the global economic crisis in 2008, left Portugal in dire economic straits. The country required a massive $78 billion bail-out package from the IMF and EU in 2011, in addition to austerity measures for economic adjustment, to set the country on a correction course to stability. These measures included raising taxes, cutting wages for civil servants, and the implementation of additional measures Brussels, the headquarters of the EU.
These measures, while viewed by some as necessary for the survival of the country, were also extremely harmful. Although Portugal ultimately exited its bail-out in 2014 without much dissent, unlike its peers Greece and Italy, issues did result from the country’s austerity measures. Portugal’s economic austerity led to massive emigration between 2011 and 2014, equaling nearly 5% of the country’s population at the time. Much of this was due to rampant unemployment and privatization, creating a tense political environment leading up to the 2015 elections.
In 2015 Portuguese voters showed contempt for the austere monetary regime they were under, and voted in Antonio Costa as Prime Minister, a left-leaning Portuguese lawyer and politician who ran as part of the Socialist party. Under Costa, Portugal’s austere policy suite was dismantled, making way for a laissez faire approach to domestic and foreign markets.
Since Portugal’s liberal shift in 2015, the country has become the poster child for European economic recovery. An dramatic increase in direct foreign investment, and a greater focus on tourism and exports, saw the country’s GDP rise from $216.2 billion in 2012 to $242.3 billion in 2018 (2.83% CAGR), and unemployment drop from 17.5% in 2013, in the midst of the austerity regime, to only 7.9% in 2018.
While today Portugal’s economy is viewed as both stable and growing, this perspective is not the result of Portugal’s fiscal and monetary policies alone. The growth of the European Union has much to do with investor sentiments regarding Portugal, as made evident by the country reaching negative-yield bonds in December. This milestone grouped Portugal with other European economies like The Netherlands, Germany, and France, all economies in which investors are buying bonds with the understanding that their returns will be less than their original investment. What stands out with Portugal, however, is how quickly the Portuguese bond rates improved downwards in the past year.
Some investors are making this bet with the expectation that demand for these bonds will continue to rise, thereby increasing bond prices and lowering yields even further. This would allow for a profitable sale of these negative-yield bonds at some point in the future. Others are investing in these bonds purely to diversify their holdings, with the losses in these rather stable bonds being offset by more lucrative, higher-returning investments in a portfolio.
Regardless of intent, these negative-yield bonds are a signal of investor relative confidence in Portugal’s economic future, which can also be viewed as a proxy for confidence in the outcome of the upcoming election. The re-election of de Sousa would keep Portugal on a stable growth trajectory, and it seems this is what markets are betting on.
Finally, one of the largest prospective boon’s to Portugal’s economy stems from its large supply of lithium. Lithium is one of the most important components in the manufacturing of electronics, and its price tripled between 2015 and 2018. And while production levels could be raised to mitigate price hikes, most production methods currently used are damaging to the environment, and thus have been heavily regulated globally.
Like many minerals and precious metals, the majority of global production for lithium comes out of Australia (42,000), Chile (18,000), and China (7,500), while Portugal (1,200) is sixth. What is notable, however, is that Portugal is the largest lithium producer out of all European countries. With lithium being a key part of the development of rechargeable batteries, allowing the use of carbon-neutral energy sources to power vehicles as opposed to gasoline, Portugal could become an important player in the European green-energy movement. This is something that would heavily align with progressive Prime Minister Antonio Costa’s vision for Portugal’s future, especially with his new leadership role in the European Union.
Increasing demand for lithium has been reflected in capital markets, as lithium producers have received high levels of investment recently, yielding equally high returns. Sociedad Quimica y Minera de Chile, or SQM (NYSE:SQM), the world’s second-largest lithium producer, returned 86% in 2020. The Chilean company has also been making inroads with the local community they mine near, making future mining and discoveries much more probable.
Electric vehicles have likely been the most visible driver for lithium demand. Tesla’s stock price has reached astronomical heights on the back of its electric card operations, and Warren Buffet now owns 25% of BYD, a Chinese-based electric vehicle company. As the shift to less carbon-reliant modes of transportation progresses, and the demand for lithium-supported rechargeable batteries grows, the prospects of Portugal leveraging its lithium supplies to fund domestic growth becomes increasingly likely.
As renowned investor and former early Facebook senior executive Chamath Palihapitiya has stated, “Tesla is a distributed energy business...they are figuring out how to harness energy, how to store it, and then how to use it in a way to allow humans to be productive. Cars are a manifestation [of this]…” Electric vehicles are just a manifestation of a larger green energy wave, at the center of which lithium will likely be an important player.
Should Marcel de Sousa win, Portugal will likely continue its positive growth trajectory, all else equal, with the president helping to ensure domestic stability. We will be curious to see what share of the vote de Sousa wins. He will likely clear the 52% of the vote he won in 2016, but will he be able to clear 60 of the vote? We are also watching to see the performance of Andre Ventura to help see whether nationalism is on the rise or fall in Europe, which might provide more insights on the upcoming Franco and German elections in 2021, 2022.
Our approach to the Portugal presidential election is similar to our previous election predictions, using Bayesian statistics. We first calculated the base-rate for Marcelo de Sousa by using the historical performance of the Social Democratic Party (PSD), the party de Sousa used to be president of. Next, we calculated the current position of de Sousa in the 2020-2021 pre-election surveys. We only calculated these values for round 1 of voting due to the extremely low chances of a second round vote taking place (famous last words).
With the historical base-rate and current position, we ran a Bayesian prediction model in R with 50,000 simulations. We then sorted the simulations and calculated the percentage of instances where de Sousa cleared 50%+1 of the vote. Due to the extremely high probability of victory and our confidence in de Sousa’s chances of victory, we did not create a 89% confidence interval this time for his likelihood of winning. However, we did create an 89% confidence interval on our forecasted vote total. Our code and data available at the end.
Determining the base-rate was extremely simple for Portugal. We used the historical performance of the PSD in the past three elections–2006, 2011, and 2016–weighted at 0.03, 0.17, and 0.80 respectively. The base rate was therefore calculated at 52.125%.
However, calculating the standard error was more complicated due to the nature of its operation in our model. Instead of the ~0.04 that was dictated by our data, we instead inputted a tau of 0.06, a value which was between our human-centric value and our data-driven value.
Calculating the bias was simple and was done by comparing the 2016 polls in the month prior to the election to the election results.
Determining the current position of de Sousa was equally simple since there is a fair deal of polling in Portugal. As a result, we selected the latest poll from any pollster conducted in the last month about the Portuguese election and, based on examination of historical polling performance and accuracy, weighted them equally.
Factoring in COVID
The only other change to this prediction was incorporating the answer to: “What if de Sousa dies of COVID-19?”
This question was answered two different ways:
- Considering if his death would cause him to lose the election
- Considering if he would die at all before he is sworn in for re-election
As hinted in the introduction, we did not simply calculate what the odds are that a Portuguese male, aged 70+ dies from COVID. Instead, we also attempted to factor in the increased survivability of significant politicians while also not discounting de Sousa’s age or general health. We calculated there was a ~2.00% chance de Sousa dies from COVID-19 and subtracted that from what our model originally outputted to answer the second point.
We also calculated there was only a ~1.00% chance he would die before the election. To answer the first point, therefore, we decided there was an 80% chance that even if de Sousa died, he would still win the election. We then calculated how that would affect the rest of the vote and adjusted accordingly, lowering his odds of winning the election from 99.69% to 99.46%, or 0.23%.