Economists or meteorologists: who give better forecasts

Journalist Susan Casey alternates between episodes describing events with the most gigantic waves of all time and sections telling the story of surfing. Casey lists the technological breakthroughs that have led surfers to ride waves of more than 30 meters today. Like the nautical motorcycles that transport intrepid athletes off (which is very dangerous to swim with the board) and, basically, the weather forecasts that allow the best surfers to travel all year by plane between the islands and the islands. Pacific coast to "hunt" for the best waves.

Casey's book was published in 2010 and since then the ability to predict the weather has improved exponentially. A recent article in the magazine
Science He says that currently, the five-day forecasts are obtained with the same precision as the one for 1980 for the next day. Advances have made it possible to decipher better and more complex climatic phenomena such as hurricanes, which translates into thousands of lives saved if we take into account that advances over time allow much more effective evacuations than in previous decades.

According to World Bank estimates, between $ 400 million and $ 500 million annually in developing countries are devoted to meteorological services, but this investment is expected to triple or quadruple. As a result, according to World Bank technicians, 23,000 lives could be saved each year due to natural disasters and material costs of up to $ 30,000 million.

The improvement of weather forecasts is explained by the combination of new technologies (satellites and increased data processing capacity), as well as by the evolution of mathematical models to better understand atmospheric phenomena.

This tandem, which feeds, leaves something in
exceptIt was an old joke among economists who claimed that the forecasts of this discipline had been invented so that, with them, the meteorologists are well. In fact, meteorology has been successful in the past two decades in gaining several benefits for economists. The gap is visible in news and news channels: as resources grow for the time segment – with star meteorologists, expensive visual resources, and so on. the climatic reference – the screen.

Why are advances in technology and mathematical models not producing the same improvement in forecasting for the economy as for meteorology? For economist Walter Sosa Escudero, professor at Udesa and regular contributor to this section, the main source of unpredictability in meteorology is complexity-non-linearity, and important advances have been made via the big data, algorithms, etc.

"In economics and politics, the main barrier against predictions lies in the strategic problems, the badumption of effectiveness is the counterpart," says Sosa Escudero. The joke – revealing the mental schema of dark science – that if an economist sees a $ 100 bill thrown into the street, he does not raise it, because it was genuine and that somebody Another had seized it. "That is, the predictive performance of the economy has not improved, just as it did not predict how the definitions would be deducted from the penalties, regardless of the amount of data and video available, "says the economist.

Last October, the government of China's Guangdong Province – the manufacturing hub of the southeast coast, which generates 12 percent of the world's most populous country's GDP – stopped publishing a monthly report on the country's economic activity. factories in the region. The reason: the tariff war with the United States led to the fall of the activity and, after five months of consecutive losses, the local authorities chose to cut the flood of bad news.

Use the satellites

A small
to startUpstream of San Francisco, SpaceKnow immediately seized the opportunity and began rebuilding the index from satellite images. The information is sold to investors and funds, which gives them an advantage in their market decisions.

According to Euroconsult, only 730 new satellites have been launched in the last decade. Several companies are dedicated to decoding these images and transforming them into information relevant to markets and economic forecasts: truck movements from factories, quality and quantity of crops, space occupied in uncovered car parks in shopping centers – society Orbital. Insight, from Palo Alto, specializes in these measurements, Wi-Fi signal density, mobile phones, etc.

The information provided by satellites is just one of the new sources provided by technology: online price information, which is revolutionizing the measurement of inflation and improving price forecasting capacity.

These developments have their niche markets and can provide predictive benefits at the micro level, but it is still very difficult to anticipate "macro hurricanes": major crises, etc.

In an article on "The Principle of Uncertainty" in Economics, David Levine, a professor at Washington University in St. Louis, is overflowing with strategic issues evoked by Sosa Escudero. In physics, there is the principle of Heisenberg, which sets a limit to our ability to simultaneously know certain facts, such as the speed and position of a particle. In quantum physics, only the distribution of the probabilities of some results can be determined in advance. "Strangely, we expect more economists," says Levine.

The badogy with physics is useful because, in Heisenberg's principle, the observer interferes with the system, which is even more pronounced in economics: "We are all actors in this discipline and the models we use determine our behavior, that an explanatory model is correct, it modifies our behavior to adapt it to the new understanding of reality, so that the original model ceases to be correct.In this sense, human behavior is by nature uncertain, "says the academic. In terms of markets, if an effective model predicts a crisis for a given day, investors will sell in advance and the debacle will occur sooner.

In his book on surfing and giant waves, Casey tells how, in the 80s, mathematical models that up until then did not allow the formation of "monster waves" from scratch (without an earthquake, without asteroid or other external event) have been modified. the cause). The new models allow that the very dynamics of the oceans can occasionally cause waves of several tens of meters.

But the economic reality, its internal currents and its human behavior can be much more complex and elusive. At the end of the day, as director of the Interdisciplinary Institute of Political Economy (IIEP) of the University of Buenos Aires, Daniel Heymann, "economists study the interactions of 7,000 million people who have a brain at least as good as theirs ". our ".

[email protected]