It is Christmas 2020, late in the morning. I am thinking, sort of deeply. It is a dysfunctional tradition to make, by the end of the year, resolutions for the coming year. Resolutions which we obviously don’t hold to long enough to see them bring anything substantial. Yet, it is a good thing to pass in review the whole passing year, distinguish my own f**k-ups from my valuable actions, and use it as learning material for the incoming year.
What I have been doing consistently for the past year is learning new stuff: investment in the stock market, distance teaching amidst epidemic restrictions, doing research on collective intelligence in human societies, managing research projects, programming, and training consistently while fasting. Finally, and sort of overarchingly, I have learnt the power of learning by solving specific problems and writing about myself mixing successes and failures as I am learning.
Yes, it is precisely the kind you can expect in what we tend to label as girls’ readings, sort of ‘My dear journal, here is what happened today…’. I keep my dear journal focused mostly on my broadly speaking professional development. Professional development combines with personal development, for me, though. I discovered that when I want to achieve some kind of professional success, would it be academic, or business, I need to add a few new arrows to my personal quiver.
Investing in the stock market and training while fasting are, I think, what I have had the most complete cycle of learning with. Strange combination? Indeed, a strange one, with a surprising common denominator: the capacity to control my emotions, to recognize my cognitive limitations, and to acknowledge the payoff from both. Financial decisions should be cold and calculated. Yes, they should, and sometimes they are, but here comes a big discovery of mine: when I start putting my own money into investment positions in the stock market, emotions flare in me so strongly that I experience something like tunnel vision. What looked like perfectly rational inference from numbers, just minutes ago, now suddenly looks like a jungle, with both game and tigers in it. The strongest emotion of all, at least in my case, is the fear of loss, and not the greed for gain. Yes, it goes against a common stereotype, and yet it is true. Moreover, I discovered that properly acknowledged and controlled, the fear of loss is a great emotional driver for good investment decisions, and, as a matter of fact, it is much better an emotional driver than avidity for gain. I know that I am well off when I keep the latter sort of weak and shy, expecting gains rather than longing for them, if you catch my drift.
Here comes the concept of good investment decisions. As this year 2020 comes to an end, my return on cash invested over the course of the year is 30% with a little change. Not bad at all, compared to a bank deposit (+1,5%) or to sovereign bonds (+4,5% max). I am wrapping my mind around the second most fundamental question about my investment decisions this year – after, of course, of the question about return on investment – and that second question is ontological: what my investment decisions actually have been? What has been their substance? The most general answer is tolerable complexity with intuitive hedging and a pinch of greed. Complexity means that I have progressively passed from the otherwise naïve expectation of one perfect hit to a portfolio of investment positions. Thinking intuitively in terms of portfolio has taught me just as intuitive approach to hedging my risks. Now, when I open one investment position, I already think about another possible one, either to reinforce my glide on the wave crest I intend to ride, or to compensate the risks contingent to seeing my ass gliding off and down from said wave crest.
That portfolio thinking of mine happens in layers, sort of. I have a portfolio of industries, and that seems to be the basic structuring layer of my decisions. I think I can call myself a mid-term investor. I have learnt to spot and utilise mid-term trends of interest that investors in the stock market attach to particular industries. I noticed there are cyclical fashion seasons in the stock market, in that respect. There is a cyclically recurrent biotech season, due to the pandemic. There is just as cyclical a fashion for digital tech, and another one for renewable energies (photovoltaic, in particular). Inside the digital tech, there are smaller waves of popularity as regards the gaming business, others connected to FinTech etc.
Cyclicality means that prices of stock in those industries grow for some time, ranging, by my experience, from 2 to 13 weeks. Riding those waves means jumping on and off at the right moment. The right moment for jumping on is as early as possible after the trend starts to ascend, and jump just as early as possible after it shows signs of durable descent.
The ‘durable’ part is tricky, mind you. I saw many episodes, and during some of them I shamefully yielded to short-termist panic, when the trend curbs down just for a few days before rocketing up again. Those episodes show well what it means in practical terms to face ‘technical factors’. The stock market is like an ocean. There are spots of particular fertility, and big predators tend to flock just there. In the stock market, just as in the ocean, you have bloody big sharks swimming around, and you’d better hold on when they start feeding, ‘cause they feed just as real sharks do: they hit quickly, cause abundant bleeding, and then just wait until their pray bleeds out enough to be defenceless.
When I see, for example, a company like the German Biontech (https://investors.biontech.de/investors-media) suddenly losing value in the stock market, whilst the very vaccine they ganged up with Pfizer to make is being distributed across the world, I am like: ‘Wait a minute! Why the stock price of a super-successful, highly innovative business would fall just at the moment when they are starting to consume the economic fruit of their innovation?’. The only explanation is that sharks are hunting. Your typical stock market shark hunts in a disgusting way, by eating, vomiting and then eating their vomit back with a surplus. It bites a big chunk of a given stock, chews it for a moment, spits it out quickly – which pushes the price down a bit – then eats back its own vomit of stock, with a tiny surplus acquired at the previously down-driven price, and then it repeats. Why wouldn’t it repeat, as long as the thing works?
My personal philosophy, which, unfortunately, sometimes I deviate from when my emotions prevail, is just to sit and wait until those big sharks end their feeding cycle. This is another useful thing to know about big predators in the stock market: they hunt similarly to big predators in nature. They have a feeding cycle. When they have killed and consumed a big prey, they rest, as they are both replete with eating and down on energy. They need to rebuild their capital base.
My reading of the stock market is that those waves of financial interest in particular industries are based on expectations as for real business cycles going on out there. Of course, in the stock market, there is always the phenomenon of subsidiary interest: I invest in companies which I expect other investors to invest to, as well, and, consequently, whose stock price I expect to grow. Still, investors in the stock market are much more oriented on fundamental business cycles than non-financial people think. When I invest in the stock of a company, and I know for a fact that many other investors think the same, I expect that company to do something constructive with my trust. I want to see those CEOs take bold decisions as for real investment in technological assets. When they really do so, I stay with them, i.e. I hold that stock. This is why I keep holding the stock of Tesla even amidst episodes of while swings in its price. I simply know Elon Musk will always come up with something which, for him, are business concepts, and for the common of mortals are science-fiction. If, on the other hand, I see those CEOs just sitting and gleaming benefits from trading their preferential shares, I leave.
Here I connect to another thing I started to learn during 2020: managing research projects. At my university, I have been assigned this specific job, and I discovered something which I did not expect: there is more money than ideas, out there. There is, actually, plenty of capital available from different sources, to finance innovative science. The tricky part is to translate innovative ideas into an intelligible, communicable form, and then into projects able to convince people with money. The ‘translating’ part is surprisingly complex. I can see many sparse, sort of semi-autonomous ideas in different people, and I still struggle with putting those people together, into some sort of team, or, fault of a team, into a network, and make them mix their respective ideas into one, big, articulate concept. I have been reading for years about managing R&D in corporate structures, about how complex and artful it is to manage R&D efficiently, and now, I am experiencing it in real life. An interesting aspect of that is the writing of preliminary contracts, the so-called ‘Non-Disclosure Agreements’ AKA NDAs, the signature of which is sort of a trigger for starting serious networking between different agents of an R&D project.
As I am wrapping my mind around those questions, I meditate over the words written by Joseph Schumpeter, in his Business Cycles: “Whenever a new production function has been set up successfully and the trade beholds the new thing done and its major problems solved, it becomes much easier for other people to do the same thing and even to improve upon it. In fact, they are driven to copying it if they can, and some people will do so forthwith. It should be observed that it becomes easier not only to do the same thing, but also to do similar things in similar lines—either subsidiary or competitive ones—while certain innovations, such as the steam engine, directly affect a wide variety of industries. This seems to offer perfectly simple and realistic interpretations of two outstanding facts of observation : First, that innovations do not remain isolated events, and are not evenly distributed in time, but that on the contrary they tend to cluster, to come about in bunches, simply because first some, and then most, firms follow in the wake of successful innovation ; second, that innovations are not at any time distributed over the whole economic system at random, but tend to concentrate in certain sectors and their surroundings”. (Business Cycles, Chapter III HOW THE ECONOMIC SYSTEM GENERATES EVOLUTION, The Theory of Innovation). In the Spring, when the pandemic was deploying its wings for the first time, I had a strong feeling that medicine and biotechnology will be the name of the game in technological change for at least a few years to come. Now, as strange as it seems, I have a vivid confirmation of that in my work at the university. Conceptual balls which I receive and which I do my best to play out further in the field come almost exclusively from the faculty of medical sciences. Coincidence? Go figure…
I am developing along two other avenues: my research on cities and my learning of programming in Python. I have been doing research on cities as manifestations of collective intelligence, and I have been doing it for a while. See, for example, ‘Demographic anomalies – the puzzle of urban density’ or ‘The knowingly healthy people’. As I have been digging down this rabbit hole, I have created a database, which, for working purposes, I call ‘DU_DG’. DU_DG is a coefficient of relative density in population, which I came by with some day and which keeps puzzling me. Just to announce the colour, as we say in Poland when playing cards, ‘DU’ stands for the density of urban population, and ‘DG’ is the general density of population. The ‘DU_DG’ coefficient is a ratio of these two, namely it is DU/DG, or, in other words, this is the density of urban population denominated in the units of general density in population. In still other words, if we take the density of population as a fundamental metric of human social structures, the DU_DG coefficient tells how much denser urban population is, as compared to the mean density, rural settlements included.
I want to rework through my DU_DG database in order both to practice my programming skills, and to reassess the main axes of research on the collective intelligence of cities. I open JupyterLab from my Anaconda panel, and I create a new Notebook with Python 3 as its kernel. I prepare my dataset. Just in case, I make two versions: one in Excel, another one in CSV. I replace decimal comas with decimal points; I know by experience that Python has issues with comas. In human lingo, a coma is a short pause for taking like half a breath before we continue uttering the rest of the sentence. From there, we take the coma into maths, as decimal separator. In Python, as in finance, we talk about decimal point as such, i.e. as a point. The coma is a separator.
Anyway, I have that notebook in JupyterLab, and I start by piling up what I think I will need in terms of libraries:
>> import numpy as np
>> import pandas as pd
>> import os
>> import math
I place my database in the root directory of my user profile, which is, by default, the working directory of Anaconda, and I check if my database is visible for Python:
It is there, in both versions, Excel and CSV. I start with reading from Excel:
>> DU_DG_Excel=pd.DataFrame(pd.read_excel(‘Dataset For Perceptron.xlsx’, header=0))
I check with ‘DU_DG_Excel.info()’. I get:
RangeIndex: 1155 entries, 0 to 1154
Data columns (total 10 columns):
# Column Non-Null Count Dtype
— —— ————– —–
0 Country 1155 non-null object
1 Year 1155 non-null int64
2 DU_DG 1155 non-null float64
3 Population 1155 non-null int64
4 GDP (constant 2010 US$) 1042 non-null float64
5 Broad money (% of GDP) 1006 non-null float64
6 urban population absolute 1155 non-null float64
7 Energy use (kg of oil equivalent per capita) 985 non-null float64
8 agricultural land km2 1124 non-null float64
9 Cereal yield (kg per hectare) 1124 non-null float64
dtypes: float64(7), int64(2), object(1)
memory usage: 90.4+ KB
Cool. Exactly what I wanted. Now, if I want to use this database as a simulator of collective intelligence in human societies, I need to assume that each separate ‘country <> year’ observation is a distinct local instance of an overarching intelligent structure. My so-far experience with programming opens up on a range of actions that structure is supposed to perform. It is supposed to differentiate itself into the desired outcomes, on the one hand, and the instrumental epistatic traits manipulated and adjusted in order to achieve those outcomes.
As I pass in review my past research on the topic, a few big manifestations of collective intelligence in cities come to my mind. Creation and development of cities as purposeful demographic anomalies is the first manifestation. This is an otherwise old problem in economics. Basically, people and resources they use should be disposed evenly over the territory those people occupy, and yet they aren’t. Even with a correction taken for physical conditions, such as mountains or deserts, we tend to like forming demographic anomalies on the landmass of Earth. Those anomalies have one obvious outcome, i.e. the delicate balance between urban land and agricultural land, which is a balance between dense agglomerations generating new social roles due to abundant social interactions, on the one hand, and the local food base for people endorsing those roles. The actual difference between cities and the surrounding countryside, in terms of social density, is very idiosyncratic across the globe and seems to be another aspect of intelligent collective adaptation.
Mankind is becoming more and more urbanized, i.e. a consistently growing percentage of people live in cities (World Bank 1). In 2007 – 2008, the coefficient of urbanization topped 50% and keeps progressing since then. As there is more and more of us, humans, on the planet, we concentrate more and more in urban areas. That process defies preconceived ideas about land use. A commonly used narrative is that cities keep growing out into their once-non-urban surroundings, which is frequently confirmed by anecdotal, local evidence of particular cities effectively sprawling into the neighbouring rural land. Still, as data based on satellite imagery is brought up, and as total urban land area on Earth is measured as the total surface of peculiar agglomerations of man-made structures and night-time lights, that total area seems to be stationary, or, at least, to have been stationary for the last 30 years (World Bank 2). The geographical distribution of urban land over the entire land mass of Earth does change, yet the total seems to be pretty constant. In parallel, the total surface of agricultural land on Earth has been growing, although at a pace far from steady and predictable (World Bank 3).
There is a theory implied in the above-cited methodology of measuring urban land based on satellite imagery. Cities can be seen as demographic anomalies with a social purpose, just as Fernand Braudel used to state it (Braudel 1985) : ‘Towns are like electric transformers. They increase tension, accelerate the rhythm of exchange and constantly recharge human life. […]. Towns, cities, are turning-points, watersheds of human history. […]. The town […] is a demographic anomaly’. The basic theoretical thread of this article consists in viewing cities as complex technologies, for one, and in studying their transformations as a case of technological change. Logically, this is a case of technological change occurring by agglomeration and recombination. Cities can be studied as demographic anomalies with the specific purpose to accommodate a growing population with just as expanding a catalogue of new social roles, possible to structure into non-violent hierarchies. That path of thinking is present, for example, in the now classical work by Arnold Toynbee (Toynbee 1946), and in the even more classical take by Adam Smith (Smith 1763). Cities can literally work as factories of new social roles due to intense social interactions. The greater the density of population, the greater the likelihood of both new agglomerations of technologies being built, and new, adjacent social roles emerging. A good example of that special urban function is the interaction inside age groups. Historically, cities have allowed much more abundant interactions among young people (under the age of 25), that rural environments have. That, in turn, favours the emergence of social roles based on the typically adolescent, high appetite for risk and immediate rewards (see for example: Steinberg 2008). Recent developments in neuroscience, on the other hand, allow assuming that abundant social interactions in the urban environment have a deep impact on the neuroplastic change in our brains, and even on the phenotypical expression of human DNA (Ehninger et al. 2008; Bavelier et al. 2010; Day & Sweatt 2011; Sweatt 2013)
At the bottom line of all those theoretical perspectives, cities are quantitatively different from the countryside by their abnormal density of population. Throughout this article, the acronymic symbol [DU/DG] is used to designate the density of urban population denominated in the units of (divided by) general density of population, and is computed on the grounds of data published by combining the above cited coefficient of urbanization (World Bank 1) with the headcount of population (World Bank 4), as well as with the surface of urban land (World Bank 2). The general density of population is taken straight from official statistics (World Bank 5).
The [DU/DG] coefficient stays in the theoretical perspective of cities as demographic anomalies with a purpose, and it can be considered as a measure of social difference between cities and the countryside. It displays intriguing quantitative properties. Whilst growing steadily over time at the globally aggregate level, from 11,9 in 1961 to 19,3 in 2018, it displays significant disparity across space. Such countries as Mauritania or Somalia display a [DU/DG] > 600, whilst United Kingdom or Switzerland are barely above [DU/DG] = 3. In the 13 smallest national entities in the world, such as Tonga, Puerto Rico or Grenada, [DU/DG] falls below 1. In other words, in those ultra-small national structures, the method of assessing urban space by satellite-imagery-based agglomeration of night-time lights fails utterly. These communities display peculiar, categorially idiosyncratic a spatial pattern of settlement. The cross-sectional variability of [DU/DG] (i.e. its standard deviation across space divided by its cross-sectional mean value) reaches 8.62, and yet some 70% of mankind lives in countries ranging across the 12,84 ≤ [DU/DG] ≤ 23,5 interval.
Correlations which the [DU/DG] coefficient displays at the globally aggregate level (i.e. at the scale of the whole planet) are even more puzzling. When benchmarked against the global real output in constant units of value (World Bank 6), the time series of aggregate, global [DU/DG] displays a Pearson correlation of r = 0,9967. On the other hand, the same type of Pearson correlation with the relative supply of money to the global economy (World Bank 7) yields r = 0,9761. As the [DU/DG] coefficient is supposed to represent the relative social difference between cities and the countryside, a look at the latter is beneficial. The [DU/DG] Pearson-correlates with the global area of agricultural land (World Bank 8) at r = 0,9271, and with the average, global yield of cereals, in kgs per hectare (World Bank 9), at r = 0,9858. That strong correlations of the [DU/DG] coefficient with metrics pertinent to the global food base match its correlation with the energy base. When Pearson-correlated with the global average consumption of energy per capita (World Bank 10), [DU/DG] proves significantly covariant, at r = 0,9585. All that kept in mind, it is probably not that much of a surprise to see the global aggregate [DU/DG] Pearson correlated with the global headcount of population (World Bank 11) at r = 0,9954.
It is important to re-assume the meaning of the [DU/DG] coefficient. This is essentially a metric of density in population, and density has abundant ramifications, so to say. The more people live per 1 km2, the more social interactions occur on the same square kilometre. Social interactions mean a lot. They mean learning by civilized rivalry. They mean transactions and markets as well. The greater the density of population, the greater the probability of new skills emerging, which possibly translates into new social roles, new types of business and new technologies. When two types of human settlements coexist, displaying very different densities of population, i.e. type A being many times denser than type B, type A is like a factory of patterns (new social roles and new markets), whilst type B is the supplier of raw resources. The progressively growing global average [DU/DG] means that, at the scale of the human civilization, that polarity of social functions accentuates.
The [DU/DG] coefficient bears strong marks of a statistical stunt. It is based on truly risky the assumption, advanced implicitly by through the World Bank’s data, that total surface of urban land on Earth has remained constant, at least over the last 3 decades. Moreover, denominating the density of urban population in units of general density of population was purely intuitive from the author’s part, and, as a matter of fact, other meaningful denominators can easily come to one’s mind. Still, with all that wobbly theoretical foundation, the [DU/DG] coefficient seems to inform about a significant, structural aspect of human societies. The Pearson correlations, which the global aggregate of that coefficient yields with the fundamental metrics of the global economy, are of an almost uncanny strength in social sciences, especially with respect to the strong cross-sectional disparity in the [DU/DG].
The relative social difference between cities and the countryside, measurable with the gauge of the [DU/DG] coefficient, seems to be a strongly idiosyncratic adaptative mechanism in human societies, and this mechanism seems to be correlated with quantitative growth in population, real output, production of food, and the consumption of energy. That could be a manifestation of tacit coordination, where a growing human population triggers an increasing pace of emergence in new social roles by stimulating urban density. As regards energy, the global correlation between the increasing [DU/DG] coefficient and the average consumption of energy per capita interestingly connects with a stream of research which postulates intelligent collective adaptation of human societies to the existing energy base, including intelligent spatial re-allocation of energy production and consumption (Leonard, Robertson 1997; Robson, Wood 2008; Russon 2010; Wasniewski 2017, 2020; Andreoni 2017; Heun et al. 2018; Velasco-Fernández et al 2018).
It is interesting to investigate how smart are human societies in shaping their idiosyncratic social difference between cities and the countryside. This specific path of research is being pursued, further in this article, through the verification and exploration of the following working hypothesis: ‘The collective intelligence of human societies optimizes social interactions in the view of maximizing the absorption of energy from the environment’.
 World Bank 1: https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS
 World Bank 2: https://data.worldbank.org/indicator/AG.LND.TOTL.UR.K2
 World Bank 3: https://data.worldbank.org/indicator/AG.LND.AGRI.K2
 Braudel, F. (1985). Civilisation and Capitalism 15th and 18th Century–Vol. I: The Structures of Everyday Life, Translated by S. Reynolds, Collins, London, pp. 479 – 482
 Royal Institute of International Affairs, Somervell, D. C., & Toynbee, A. (1946). A Study of History. By Arnold J. Toynbee… Abridgement of Volumes I-VI (VII-X.) by DC Somervell. Oxford University Press., Section 3: The Growths of Civilizations, Chapter X.
 Smith, A. (1763-1896). Lectures on justice, police, revenue and arms. Delivered in the University of Glasgow in 1763, published by Clarendon Press in 1896, pp. 9 – 20
 Steinberg, L. (2008). A social neuroscience perspective on adolescent risk-taking. Developmental review, 28(1), 78-106. https://dx.doi.org/10.1016%2Fj.dr.2007.08.002
 Ehninger, D., Li, W., Fox, K., Stryker, M. P., & Silva, A. J. (2008). Reversing neurodevelopmental disorders in adults. Neuron, 60(6), 950-960. https://doi.org/10.1016/j.neuron.2008.12.007
 Bavelier, D., Levi, D. M., Li, R. W., Dan, Y., & Hensch, T. K. (2010). Removing brakes on adult brain plasticity: from molecular to behavioral interventions. Journal of Neuroscience, 30(45), 14964-14971. https://www.jneurosci.org/content/jneuro/30/45/14964.full.pdf
 Day, J. J., & Sweatt, J. D. (2011). Epigenetic mechanisms in cognition. Neuron, 70(5), 813-829. https://doi.org/10.1016/j.neuron.2011.05.019
 Sweatt, J. D. (2013). The emerging field of neuroepigenetics. Neuron, 80(3), 624-632. https://doi.org/10.1016/j.neuron.2013.10.023
 World Bank 4: https://data.worldbank.org/indicator/SP.POP.TOTL
 World Bank 5: https://data.worldbank.org/indicator/EN.POP.DNST
 World Bank 6: https://data.worldbank.org/indicator/NY.GDP.MKTP.KD
 World Bank 7: https://data.worldbank.org/indicator/FM.LBL.BMNY.GD.ZS
 World Bank 8: https://data.worldbank.org/indicator/AG.LND.AGRI.K2
 World Bank 9: https://data.worldbank.org/indicator/AG.YLD.CREL.KG
 World Bank 10: https://data.worldbank.org/indicator/EG.USE.PCAP.KG.OE
 World Bank 11: https://data.worldbank.org/indicator/SP.POP.TOTL
 Leonard, W.R., and Robertson, M.L. (1997). Comparative primate energetics and hominoid evolution. Am. J. Phys. Anthropol. 102, 265–281.
 Robson, S.L., and Wood, B. (2008). Hominin life history: reconstruction and evolution. J. Anat. 212, 394–425
 Russon, A. E. (2010). Life history: the energy-efficient orangutan. Current Biology, 20(22), pp. 981- 983.
 Waśniewski, K. (2017). Technological change as intelligent, energy-maximizing adaptation. Energy-Maximizing Adaptation (August 30, 2017).
 Wasniewski, K. (2020). Energy efficiency as manifestation of collective intelligence in human societies. Energy, 191, 116500.
 Andreoni, V. (2017). Energy Metabolism of 28 World Countries: A Multi-scale Integrated Analysis. Ecological Economics, 142, 56-69
 Heun, M. K., Owen, A., & Brockway, P. E. (2018). A physical supply-use table framework for energy analysis on the energy conversion chain. Applied Energy, 226, 1134-1162
 Velasco-Fernández, R., Giampietro, M., & Bukkens, S. G. (2018). Analyzing the energy performance of manufacturing across levels using the end-use matrix. Energy, 161, 559-572