When making predictions, it’s worth noting that accurate forecasts are notoriously difficult to make.
To borrow from Yogi Berra, predictions are extremely difficult to make, especially when it is about the future. Ever since man has been thinking about the future, literature has been replete with forecasts that have gone awry. In 1911 for example, Thomas Edison predicted that everything would be made of steel within the next 50 years.
In 1950, the Associated Press claimed that all women would be 6 feet tall by the turn of the century, citing “overwhelming scientific evidence” for this claim. In 2007, Steve Ballmer said that the iPhone had “no chance” of getting any significant market share. Wherever you look, the science of forecasting has been disastrous.
In economics and business, part of the problem has been the iterative nature of the forecast itself, otherwise known as the feedback loop. Forecasts of about a 3% rate of inflation causes labor to demand at least a 3% increase in wages; given that the basis of the forecast has now caused a reactionary effect in the ecosystem, inflation is now likely to be higher. The fact that the forecast exists, changes the reality that it is trying to predict. Given that most forecasts are based on linear thinking (projecting straight lines into a future time period), this “reflexivity” is never captured.
Yet it is also true that the internet continues to develop in patterns that can be predicted. For example, we have aggregators like TripAdvisor and Rotten Tomatoes that agglomerate large amounts of data into simple statistics that give a reasonably reliable signal to the customer. In the same way, it is but inevitable that the same discipline will be extended to the field of economic forecasting, with a “reliability” or “good judgement” indicator for economists and analysts, such that the investing public can ascertain what weightage to give to certain predictions based on a track record.
What we know about the future is that the higher the level of confidence in any particular trajectory, the more likely it is to be wrong. A sample study done by the New York Times indicated that across all fields, less than 5 percent of forecasts turned out to be prescient. Yet, underneath the ashes, what was discovered was that good forecasters turned out to get better more than 60 percent of the time. This then implies that we need to be looking for models that can build in the non-linearity, and until then, look for the people who are doing it already.
It is curious that in an age of increasing transparency, the number of failures are increasing, and rather than having open ended conversations, the world seems to be moving towards the politics, and the economics of populism. This was heralded as a new dawn in 2016; in 2018 the dark clouds have gathered overhead, and alongside it, the increasingly ominous predictions about a dystopian future. In times past, populism has been a trend that has played out over a number of decades, and it is by looking at these patterns of history that forecasters have projected the same going forward; a dour view of the world that will generate increasing amounts of anxiety and angst, as the demons of time past strike back.
However, we know that change is accelerating, and that for the most part, the intellectual rigor that globalization has brought is a genie that is unlikely to be put back into its bottle. The UAE, perhaps more so than most countries in the world has embraced this trend, and despite the headwinds, the pace of reform towards openness is likely to accelerate as it seeks to assert its prowess in the economic and scientific spheres. I predict therefore that most predictions will turn out to be wrong; rather a mechanism for measuring these predictions will gather momentum, and will yield surprising results in terms of where the skill sets emerge from. We think that for reliable economic forecasts, the natural thing is to ask an economist. This, however, has already been proven to be wrong.
Forecasting is an art that needs to be separate from the specific subject knowledge. What needs to be isolated are the people that are reliably good at forecasting, such that the models that they have built can be studied and hopefully replicated. In the final analysis, this is perhaps the forecast that itself is the greatest source of optimism: the necessarily open nature of predictive dialogue ensures that there can only be a democracy of intellect, which by its very nature is global, regardless of when it is accepted.
The post The Perils of Prediction appeared first on Bloomberg Businessweek Middle East.
from WordPress https://ift.tt/2BXIiax
via IFTTT
