Human Psychology Continues to Beat Data

Aug 27 , 2022
By Christian Tesfaye ( Christian Tesfaye (christian.tesfaye@addisfortune.net) is a researcher and Fortune's Deputy Editor-in-Chief whose interests run amok in the directions of political thought, markets, society and pop culture. )

---

No one really can predict the future. The advances in collecting, collating and analysing large amounts of data are impressive. But they have not taken humanity as near to clairvoyance as was hoped.

Take, for instance, the predictions on the aftermath of the invasion of Ukraine by Russia. The coverage all around was about doom and gloom. A potential nuclear strike being provoked was discussed; prices for a barrel of oil hitting 200 dollars; and famines in many places being caused by skyrocketing food prices.

The war was and continues to be ugly, but the worst did not happen. In fact, oil prices are below pre-war levels, as well as global food prices. Geopolitical tensions continue to simmer and food prices are still high by historical standards. But many dire predictions did not pan out. Consider even how every pundit agreed that Russia would take over Kyiv, the Ukrainian capital, within a few days until it became all too clear that it had no hope of doing as such.

Or consider the repeated grim assessments of China’s economy. Capital outflows shortly after the war in Europe were supposed to crash the economy. Before that, it was the default by the major real estate developer, Evergrande, which was supposed to be the Lehman-Brothers-collapse moment for China. The country is obviously in a property market crisis, but it is not nearly the calamity it was suggested to be at the outset; not to mention, Evergrande is still standing.

The cherry on the cake was predictions about COVID-19 cases in Africa at the outset of the pandemic. The most morbid reports came from Italy first, with hospitals barely coping with the influx. If Europe struggles so much to arrest the disease, how would Africa, which has the poorest health infrastructure in the world, ever cope? Biblical levels of suffering, it was said, were in store for African countries.

The pandemic was no joke, and its economic and public health effects continue to be felt. But it was nothing like it was predicted to be by the analysts. It has been a struggle to explain why, but the pandemic has been much more manageable in Africa than in the rest of the world, especially considering there have not been full lockdowns.

Why are we so bad at predicting events?

One reasonable explanation is bias. Most of the time, we are expressing what may be based on how we see the world and not necessarily on how it is in reality. We make assumptions that fit our worldviews and ideologies. Even experts have a hard time maintaining an entirely objective stance. The philosophical debate on whether research could be fully objective should be settled by now. It is most likely not the case.

It is also true that most of us are pessimistic. Look at most estimates about events over the past few years, and it is clear that they portray most occurrences as being worse than is otherwise the case. We see a Hitler, an economic calamity or major war in every nook and cranny. It is not that things are not bad. It is just that they are not as bad as advertised – not as grim as the activists and TV channels that want to keep us glued to our screens claim.

It probably has to do with our biological evolution. Our brains are better hardwired to record and respond to potentially alarming developments than they are to good news, because the former could be existential. It is our fight-or-flight psychology at work. The problem is that it was not designed for a world where so much bad news is easily accessible.

Our obsession with data and its analysis makes sense. It is an effort to take subjectivity out of predictions. But this has proven harder than anticipated. We need much more data to map events accurately, including human’s psychological tendency to overreact.

---

PUBLISHED ON Aug 27,2022 [ VOL 23 , NO 1165]

---