This week’s selection from the world of social science research.
Collidoscope is The Wire’s weekly newsletter on social science research, bringing together different views and ways of understanding and analysing society from across the world. You can subscribe to the Collidoscope newsletter here. If you missed the previous editions and would like to catch up, you can find them here.
Social media and financial markets
Can Twitter make the markets crash?
“Breaking: Two Explosions in the White House and Barack Obama is injured,” Associated Press tweeted on April 23, 2013. Of course now we all know that didn’t happen – what did happen was that AP’s Twitter account was hacked (the hack was later claimed by the Syrian Electronic Army). The tweet was retweeted about 4,000 times. But what else happened?
Within seconds of the tweet, the Dow Jones Industrial Average dropped 143.5 points and nearly $136.5 billion of the Standard & Poor’s 500 Index’s value was wiped out. Then, as more and more people questioned whether the information in the tweet was true and it was discovered that it was, in fact, the outcome of a malicious hack, the markets recovered just as quickly as they fell. All it took was one tweet. So how did that happen?
In an article in Theory, Culture & Society, Tero Karppi and Kate Crawford answer not only that but also what that could mean for financial markets – and unfortunately, the world in general. They have two possible theories on how things went down, given the secrecy around financial trading algorithms and decisions, it’s hard to say which one is true.
Case one: AP, a well-known news agency with a certain standing in terms of business news, tweeted something – and that tweet was retweeted (hence legitimised) by thousands of others, including people I recently learnt are called social media “influencers”. Softwares like Dataminr, RavenPack, Gnip and others, that mine social media for news that could affect the markets, had that information as soon as it was out there and sent it to traders, who reacted accordingly. A few minutes later, the softwares figured out that the tweet is fake and send that message out too. “Through real-time analysis of Twitter data, software packages like Dataminr assess emotion, importance and social meaning in order to ‘predict the present’ and thus transform social media signals into economic information and value.”
Case two: The same thing happens, except there are no intermediary softwares. Financial algorithms scanning Twitter came across the tweet and, given all the factors, considered it worth acting upon. These automated trading systems then went on to make trades based on the tweet – before they realised, a few minutes later, that the information was false and reversed their actions. “These arguments are based on the idea that trading algorithms are scanning Twitter data using forms of text and sentiment analysis, and then immediately placing trades… While opinions differ on how much high-frequency trading algorithms draw on Twitter data, it is clear that Twitter now has considerable power to produce effects in the market.”
Of course not just any old tweet would do this. If the tweet came from my handle, for instance, there would have been no impact whatsoever. “This capacity for certain Twitter accounts depends on their legitimacy as a ‘speaker’ – in Marazzi’s terms ‘it depends on the power and the legal designation of whoever “speaks monetarily”‘.”
What this instance does highlight, the authors say, is how the labour of trading has changed dramatically. A lot of the process is now computerised and automated – algorithms are written through which computers read through pages and pages of financial data in seconds and mine information from all over the world, which would otherwise require a huge network of human traders. “In 2012,” the authors write, “roughly 50 percent of the US equity trading volume was accounted for by high-frequency trading (HFT) firms which use complex algorithmic processes to trade securities.” The HFT systems can respond to anything that happens within micro-seconds.
But where does that leave us? It’s difficult to imagine that so many global transactions are based on what a computer thinks is important. The AP “hack crash”, the authors write, is an example of why this is a problem – it wasn’t an example of the system breaking, it was actually an example of the system doing exactly what is was supposed to do.
“These processes of imitative repetition moved and adapted from one system to another. Human and non-human traders responded to what appeared to be an authoritative news message, a performative utterance with monetary legitimacy. The discourses around the Associated Press hack crash reveal the power of networked social and financial systems to connect autonomously and to produce a present without human oversight or governance.”
But it’s not like there is no governance system at. The governance system, however, goes well beyond what we traditionally saw as the domain of finance. Modern systems of communication like Twitter – which are basically (if you disregard robot accounts, because that’s a whole other spiral) humans talking to humans – have found their way into the financial system, whether for better or for worse.
Liked what you read so far? Consider subscribing to this weekly newsletter.
Financial gambling in Spain
How do you go from being a ‘good’ financial trader to a looked-down-upon gambler?
If you’re living in post-crisis Spain, according to Jorge Núñez’s article in Cultural Anthropology, it’s pretty easy. Núñez argues that part of neoliberalism’s response to the 2007-2015 financial crisis in Spain was to shift the blame from the financial system to individuals, thereby protecting the system. Individuals working in the system were the problem, apparently, because they had a ‘gambling disorder’.
The system was good, the system had to be bailed out, Spanish authorities believed, according to Núñez’s article. It was the individuals working in the system (of course only the mid- and low-level ones, though) who were all wrong. This logic seems to believe that speculative practices are bad when individuals perform them – but okay for banks.
In fact not only is it ‘wrong’ for individuals to indulge in such practices, it’s classified as a mental illness, an addiction.
Several traders in Spain were diagnosed with a ‘gambling disorder’ during or after the crisis, according to Núñez. Many former independent traders, who used leverage (borrowed money) to play the market, now have criminal cases against them – like Núñez’s case study Sylvia.
Sylvia started working for a bank in 2000; by 2005 she had moved to the headquarters in Barcelona as a sales trader in charge of retail clients. She then moved to the investment division during the rise of the housing bubble – it has been argued that Spain mimicked the US subprime mortgage model. That bubble burst in 2007 and Sylvia’s bank was bought over by an Italian financial group. She didn’t like the new management and had also reached the “gender ceiling of high finance”, so she negotiated an exit package and moved back to her hometown to become a freelance trader. She did well; her confidence grew. By 2009, she was accepting money from other people to trade on their behalf, in exchange for a small commission.
Caitlin Zaloom in her book Out of the Pits: Traders and Technology from Chicago to London talks about the feeling that digital trading instils in traders – the numbers you’re dealing in no longer seem real, everything happens so fast that it feels more like a high-adrenaline game that anything else. (When I first read Zaloom’s book I also watched the documentary Floored: Into the Pit, that looks at the hyper-masculine, competitive culture among traditional floor traders and the crisis in identity that affected some with the move to digital trading – I still have a hard time convincing myself that people in the documentary are real, but it’s a fascinating, if scary, watch.)
Back to Sylvia – according to Núñez, what Zaloom describes is very similar to what happened with traders in Spain.
“This scholarship resonates deeply with the experience of freelance traders in Barcelona. These traders also stitch their attention to on-screen markets, un-flaggingly tracking prices that yo-yo up and down in psychedelic colors. They too develop trading habits around idealized states of awareness. Silvia, for her part, lived in a quiet neighborhood near a park, but rarely left her apartment to wander. Her routine was carefully organized around the opening and closing times of the German and Spanish stock exchanges. She woke up very early in the morning, had breakfast, read every single financial newspaper, and sat at her workstation. By 9 a.m. she was placing her first round of trades. …As Silvia noted, underscoring the rigor of her timetable, “I had to schedule my meals and snacks—otherwise I forgot to eat.”
Silvia recalled that her trading behavior became more erratic and sloppier with time. “I started acting like a bolsera,” she said. In Spanish market argot, the label bolsero/a has negative social and behavioral connotations. It indicates lower class status, on the one hand, and an almost irrational approach to financial markets, on the other.”
Soon, things “started to fall apart”, according to Sylvia. She lost her mother’s retirement money. Lawsuits began to pile up. Her sister connected her to Nuria – a psychiatrist at the Catalonia University Hospital. Sylvia has been seeing Nuria for more than two years now and lives with her mother; her sister given her a small amount of money and keeps a close eye on what she’s spending on.
Sylvia’s case is far from unique – though she is one of the few women to be in this situation, given the gendered nature of stock markets. But it represents a pattern in Spain, according to the author – “small fry” individuals are handed the guilt of a system that is broken, the guilt is associated with a mental health problem or addiction in certain individuals without looking at the issue as larger, both politically and systemically, Núñez writes. The fix, too, is seen in medical terms pertaining to individuals. The unchecked financial system which breeds the culture that would have celebrated individuals like Sylvia (before her downfall), however, is not questioned.
Friends and gamblers in urban India
Can you be friends with the people you gamble against?
“My friends are all gamblers and among gamblers you cannot be friends,” someone at the Delhi Race Course told Stine Simonsen Puri while he conducted research for his article in the Journal of South Asian Studies. The purpose of Puri’s study isn’t to say how many bettors he met have a gambling addiction (though he does indicate that most of them would fall under that category). He, instead, looks at how the men who place bets at the Delhi Race Course interact and engage with each other, what their relationships are like.
Puri also notes the relationship that Núñez pointed to in Spain: “The ‘work’ of betting, as it is called by bettors, has clear similarities with the ‘work’ of speculation in stock exchange futures markets; I have found strong correlations in betting strategies and the social environment between betting at the Delhi racecourse and speculation in futures markets in the United States as described by Caitlin Zaloom.”
Close to 300 bettors come to the Race Course every day, according to Pur, betting anything between Rs 10 and Rs 5 lakh. Given that for many of them betting is somewhat of a full-time activity – their job are either related to it or secondary – much of their social interactions are also at or around the races. But do these interactions qualify as a friendships?
For the bettors Puri spoke to, they didn’t. They called these interactions ‘hanging out’ and ‘having fun’, but would not go as far as to classify them as friendships. They know all about each other’s monetary situations – what they’ve won or lost, how empty their bank accounts are looking, how upset their family is with them for wiling away the money. But they also hid things from each other – they would never meet each others’ families or even explicitly ask about them. Puri describes these unconventional relationships:
“What I observed in Delhi was gambling in an urban setting, which was innately a social act that was all about establishing relationships. These were nevertheless relationships limited to a single domain. What I find interesting is how the Delhi racecourse exists as a place of hyper-sociality, where everyone pays attention to everyone else and explores the possibilities for friendship with others, yet where gambling is not done primarily to establish relationships, but where relationships are used to enable gambling. One way that friendships enable gambling is through loans.”
This money was loaned interest free, in contrast with the moneylenders who also sit at the Race Course and charge interest between Rs 100 and Rs 200 a day. Repayments to friends could also happen by making bets on their behalf – so the amount they got back could be very different from what they had given. But being “indebted to a friend at the Delhi racecourse did not translate into an emotional bond of trust and generosity that extended beyond the gambling context. At the racecourse, debt between friends concerned the possibility for credit. It was not so much the economic practice of making loans that supported the social practice of friendship or alliance, but, rather, the social practice of friendship that supported the economic practice of giving interest-free loans.”
Puri’s description of these relationships is hard to relate to – they do seem to be friendships, and I guess similar to what people call ‘work friendships’, that are centred around your jobs. But here there is the added intimacy of knowing a lot about each others personal financial lives. Puri’s conclusion makes sense: “Friendships at the racecourse are immersed in the precariousness associated with the game the bettors are sharing in.”
That’s it for this week! If you liked what you read, please consider subscribing to this weekly newsletter.
If you have any comments or suggestions on what could be carried in this column, write to me at firstname.lastname@example.org