Monday, August 17, 2020

Economic collapse and data - Understanding long term cycles with data

 

As I was hopping from empty airport to empty airport in empty planes last week, a question came to mind: Is that "it"? Is this how civilizations fail to recover and enter terminal decline?  How could we know? Isn't a recovery just around the corner? A vaccine being readied as we speak? Will 2021 see the fastest bounce we have ever experienced or is "something" broken? And if that is the case, what exactly is that "something"? Why can't we learn from history? Look at past civilizations and avoid their fate? Why can't we understand long term cycles? 

 

In his deeply fascinating book, Collapse, published in 2005, Jared Diamond, analyzed how "societies choose to fail", a disturbing early insight into social and economic decline. But the book was written in the early part of the 21 century and as such, not surprisingly, gives an overwhelming role to external factors such as climate change and over use of natural resources. Although relevant to small scale societies such as the Anasazis of the South Western United States and the people of Easter Island in the Pacific Ocean, it is difficult to imagine that local factors could affect a more widespread civilization and that the lack of appropriate responses alone could be the reason for the collapse of all the civilizations that have preceded ours.

 

An earlier and more focused analysis came in 1972 with the report of the Club of Rome titled The Limits to Growth. This work, much ridiculed at the time, right in the middle of the "30 Glorious years" of rapid growth following the Second World War, was nevertheless a fundamental research based on early computer modeling, trying to understand the complex interaction of multiple variables on long term growth. Its conclusions would have been easier to dismiss if they did not predict that food, industrial output and services per capita would start to crash in 2020... 


But here we are in the later part of that year, and if the Central Banks had not decided earlier to break the thermometers (the markets), the momentous events earlier would have been plain to see. This is not a reference to the virus, but to the far more meaningful negative rates, break of the Repo market and QE infinity which preceded it around the end of 2019.

 

It is said that 100 years after the fall of Rome, the nobility in faraway provinces of the Western Empire were still living as if nothing had changed late in the 6th Century AD. Charlemagne saw himself as a late Emperor, reviving Constantine Christian Empire 450 years later. In-between, the complexity of Rome had become history. Its laws ignored, aquaducs and vias still used but in disrepair, monuments burnt down. But more than anything its financial and commercial infrastructure were gone. Based on the writings of the time, we know that the Romans saw it coming. But if they did see it coming and could do nothing about it, what is the chance that facing a similar fate, our civilization will fare better? This is the existential question we are facing today. And without a proper understanding of the problem, the chance that our answers will fare better may not be much higher. 



Early in the 20 Century, Kondratiev was the first in 1925 to analyze long term cycles which he concluded were due to technological advancements. He found two cycles of about 50 years each in 1790-1850 based on coal and cotton and in 1850-1900 based on steel and railways. This was a step beyond the short term "pork cycle" and the medium term financial cycles already recognized at the time but far short of the much longer periods of history with their regular rises and falls of cultures and civilizations. 

 

A further insight arrived in 1997 with the "Fourth Turning". In their book, Strauss and Howe, found long term cycles of about 80 years which their called Saeculum, each built on four generations of about 20 years, corresponding to a High, an Awakening, an Unraveling and a Crisis generation. As can be seen in the table above, the concept has now been expanded from the 15th Century to the present time.  It is most certainly an interesting idea but being "social" in nature, it lacks the hard data of the Club of Rome analysis and is still not long enough to explain our current predicament.

 

So, although we have made great progress in understanding short term economic, technological and societal cycles, longer term "civilization scale" cycles still elude us. The reason may be the same as for all the other complex systems we are confronted with today: be it the weather, the brain and artificial intelligence or the earth biosphere. The complexity of the systems overwhelms us and our ability to make sense of an unlimited number of factors. But not only!


If it was "just that" then sooner or later, progress in computer technology would help us solve the most intractable problems. This is not the case. This is not the case because, as we have slowly discovered over the years, scale matters. As systems grow in size, they undergo systemic changes called transition phase during which new phenomenons and rules emerge from the complexity. And without understanding these changes and emerging laws, we simply cannot grasp adequately the interaction of multiple variables. The "Big Data" curse is upon us!


It is in fact amazing how common the word "Big Data" has become and conversely how little grasp we have upon its meaning and consequences. Today, the largest companies, Google at its fore, are exploring this new continent of big data but instead of theories and hypothesis which were the hallmark of past scientific progress, they are groping in the dark. Unearthing new laws and principles one by one, each one completely revolutionizing its field whenever it emerges into our grasp and understanding. This was the case not so long ago with the amazing concept of backward propagation which by itself completely transformed Artificial Intelligence, enabling new technologies such as image processing and face recognition.


But the reason we stumbled into this new concept was the sheer number of engineers and brilliant people working in the field of AI, attracted by the lure of endless riches. Such is not the case with ecology and even less with social sciences. And as a consequence, the mysteries of their complexity will for a long time keep eluding our understanding, just beyond our grasp in the realm of possible but unachieved discoveries. 


This conclusion unfortunately does not answer our question: Can we analyze decline through big data and answer the fundamental question of societal collapse? Even less offer mitigation strategies to prevent such a collapse.  But expending on the concepts pioneered by the Club of Rome, it offers alleys to explore the question and build models which accurately represent the inner workings of our society. I find it amazing that we are progressing so fast to build smart cities based on complex integrated models while the global society around these cities remains misunderstood and left to its own fate, decaying at an accelerating pace. Is this the same hubris which doomed the Roman Empire?  


No comments:

Post a Comment

"Ukraine is Finished" US Army Colonel Reveals TRUTH About America's Failed War Against Russia (Video - 33mn)

  The scope of this video is not as broad as the one (erased by Google) from Macgregor but is very clear nevertheless.    The whole story ab...