The financial crash of the late 2000s was the most severe since the 1920s. It saw close to $3 trillion lost from the stock market, 10 million people lost their homes in the US alone, and the average wage had its biggest fall in recorded history. The repercussions are still being felt today and many people never recovered from it, in many areas wage growth in real terms has stagnated for over a decade and there has been a huge increase in inequality as the richest in society recovered quickly whilst those who earn less are yet to get to the same level as before the crash.
In wider society this has created an atmosphere of protectionism and blame that has seen people in the UK vote for Brexit, leading former Prime Minister Sir John Major to say ‘to break off and head off into splendid isolation doesn’t seem to me to be in our interests now, or perhaps more important, in the interests of our children and our grandchildren and future generations.’ Similarly it saw Donald Trump win the presidential vote (admittedly without a popular mandate) in the US under the banner of US protectionism, which threatens to do huge damage to the world economy, with a CNBC poll finding that 51% of those asked saying that protectionist policies were the number one danger to the US economy.
However, in the past decade, we have also seen the explosion of data use, which has created significant opportunities for companies and individuals to leverage it to solve some of the most pressing problems the society faces. So we took a look at the 5 most important ways that data could prevent the next financial crisis.
Keeping Bad Behavior In Check
There is little argument that one of the central reasons for the crash was at worst illegal, at best immoral behavior from those within the banking sector. The practice of rigging rates, mis-selling products, and taking huge risks with other people’s money was common, but there was little accountability for the practice, which saw only one banker, Kareem Serageldin, jailed for the widespread criminal actions.
It is worth noting that it was not the action of any one person that caused the financial crash, but a large number of people who all acted badly. This collective damage is what caused the crash, rather than the actions of one or two nefarious characters, which is what has made monitoring and punishment difficult in hindsight.
However, the way we work, the way we trade, and even the way we interact with customers is completely different to 10 years ago. For one, almost everything that we do is monitored and recorded. With these kind of powerful compliance systems in place, it is no longer an excuse to say ‘I didn’t know what they were doing’ because company leaders have access to everything. This knowledge of being observed or even the illusion of being observed has also been shown to improve behavior. A study from the University of Newcastle found that even just a poster with a pair of eyes saw people tidy up after themselves in a cafeteria 100% more than when the poster wasn’t there. If people are aware that their communications and trades are being closely scrutinized, they are considerably less likely to take nefarious actions as a group.
The market crashes that saw close to $3 trillion lost may not have been predictable 1 or 2 months before, but early signs would have certainly started to show around that time if people had been looking for them. The issue in 2008 was that the vast majority of markets did not use data in the same way we do today, where algorithms can pull information from a huge variety of areas to inform traders of decisions in real-time before they make a move.
If traders can effectively use data from these areas today, such as looking for frequencies of keywords on social media, fluctuations in other connected markets, and even online share rumours can be used to create more accurate models, making markets less volatile. There are even examples of investment houses using satellite images of parking lots to predict future revenues for companies, which means more predictability in markets and fewer shocks that could cause huge dives in share prices.
Preventing Emotive Decisions
Companies today often preach the merits of being data-driven, giving them the ability to quickly make the best decisions based on data rather than gut feeling. This was not the case in the late 2000s, where company leaders would often make decisions for short term gain, which inadvertently ended up causing long term damage.
The amount of data available today means that modeling allows companies to see what impact their decisions will have in the future and for their wider industry landscape. If they lower their interest rates, for instance, they can predict how that will impact their company, their rivals, suppliers, and even the wider economy. This allows them to make decisions with a broad knowledge of how this will impact further down the road, rather than just looking at the next quarter or even the next 12 months.
Bubbles are caused by unsustainable systems being created, inflating values, and then eventually collapsing. This is one of the primary reasons for the financial crash which was broadly a result of the housing market falling due to inflated prices and mortgages which couldn’t be repaid.
There is such a huge amount of data available that could have predicted this, but at the time these weren’t looked at in detail, with only a few people noticing and, rather than warning against it, they betted against the market in order to make money themselves. Michael Lewis’ The Big Short documented roughly how the data pointed to what was going to happen, and those who took advantage of this data made billions.
At present there are signs that several inflated markets are beginning to turn and investors within them are acting more conservatively precisely because they have data today that they did not have in 2007. Few would argue that there isn’t a significant tech bubble in Silicon Valley amongst the many startups who are valued at billions of dollars having posted no profits in their history. For instance, Uber is reportedly worth close to $70 billion, but it has never made a profit. It, in fact, reportedly lost around $3 billion in 2016 alone. These kind of strange valuations have caused a 60% decrease in VC funding in the past year. Similarly, London house prices, which have historically been one of the safest bets for foreign investment, have seen their prices decrease by up to 7% in some of the city’s most affluent areas. This has reportedly seen fewer purchases in that time, putting the bubble at risk of bursting.
Data lets companies see when a bubble is genuinely forming and helps to ease it, rather than simply letting it burst where people can lose considerable amounts of money as we saw in the early 2000s with the tech bubble and late 2000s with the housing bubble.
Accurate Lending Calculations
One of the key drivers for slumps and crashes is dangerous lending to millions of people who couldn’t afford to repay loans, mortgages, and finance options, leaving those who initially lent the money out of pocket and those who borrowed it financially ruined. These poor lending practices are essentially what caused the 2007 financial crash, with banks giving mortgages to those who could clearly not afford to repay them, then losing the value of the mortgage when the customer couldn’t keep up with repayments.
It is still an issue today, with Alex Brazier, director for financial stability at the Bank of England claiming that there are ‘classic signs’ that loans such as car finance, credit cards, and personal loans are becoming unsustainable for many in the UK, which may cause another crash in the future.
Predictive analytics, statistical modeling, and big data are the most powerful tools that banks and financial regulators have to protect lenders and customers against making potentially harmful decisions. Simple credit ratings are only useful to a certain extent, but through looking at a variety of data from a number of different sources, better decisions can be made about whether certain loans are safe or dangerous.