Energy Ponds – l’état des lieux

I drift away from investor-relations sites, and I go back to my ‘Energy Ponds’ concept. I feel like going through it once again. First thing first, I want to lay out the idea such as I have figured it out so far. The second of the first things is that I am describing a yet-unimplemented, tentative technological solution, which combines water management with the generation of renewable energies. Ram-pumps are installed in the stream of a river. Kinetic energy of the river creates a by-flow through the ram-pumps, endowed with its own kinetic energy and flow rate, derived from those of the river. That by-flow is utilized in two ways. Some of it, within the limits of environmental sustainability of the riverine ecosystem, is pumped into wetland-type structures, which serve the purpose of water retention. On the way to wetlands, that sub-stream passes through elevated tanks, which create a secondary hydraulic head and allow placing hydroelectric turbines on pipes leading from elevated tanks to wetlands, i.e. back to ground level. The remaining part of the ram-pumped by-flow, before going back into the river, is recirculated through hydroelectric turbines located close to the ram-pump. The basic idea is shown graphically in Figure 1.

The remaining part of the article develops a complex proof of concept for ‘Energy Ponds’. Component solutions are discussed against the background of relevant literature, and a quantitative simulation of its basic parameters is presented, for simulated location in Poland, author’s home country.

Figure 1

Energy Ponds’ are supposed to create a system of water retention with as little invasive change in the landscape as possible. Typical reservoirs, such as artificial ponds and lakes, need space, and that space ought to be taken away from other employments thereof. This is a dilemma in land management: do we use a given portion of terrain for a retention reservoir or do we use it for other purposes? In densely populated regions, that dilemma becomes acute, when it comes to mutual arrangement of human architectural structures, agricultural land, and something which, fault of a better word, can be called ‘natural landscape’ (quotation marks refer to the fact that many landscapes which we intuitively gauge as ‘natural’ are actually man-made and artificially maintained).  

Water management, as a distinct field of technology, with an environmental edge, could benefit from innovation-boosting tools typical for other technological fields (Wehn & Montalvo 2018[1]; Mvulirwenande & Wehn 2020[2]). Technological change as regards water management is needed both in rural areas and in cities. Interestingly enough, urban environments seem to be more conservative than agriculture in terms of water management (Wong, Rogers & Brown 2020[3]). There is a quest for scientifically based constants in water management, such as the required quantity of water per person per day; Hogeboom (2020[4]) argues it is roughly 3800 liters after covering all the conventional uses of water. On the other hand, Mohamed et al. (2020[5]) claim that the concept of ‘arid region’, so far reserved for desertic environments, is de facto a type of environmental context when we systematically lack accurate environmental data as regards quickly changing availability of water. Kumar et al. (2020[6]) go even further and postulate something called ‘socio-hydrology’: human societies seem to develop characteristically differentiated patterns of collective behaviour in different hydrological conditions. Other research suggests that societies adapt to increased use of water by visibly increasing the productivity of that use, whilst increased income per capita seems being correlated with increased productivity in the use of water (Bagstad et al. 2020[7]).

In the ‘Energy Ponds’ concept, retention of water is supposed to be a slow, looped recirculation through local ecosystems, rather than accumulation in fixed reservoirs. Once water has been ram-pumped from a river into wetlands, the latter allow slow runoff, via ground waters, back into the hydrological system of the river basin. It is as if rain was falling once again in that river basin, with rainwater progressively collected by the drainage of the river. In hydrology, such a system is frequently referred to as quasi-reservoirs (Harvey et al. 2009[8]; Phiri et al. 2021[9]). Groundwater seems being the main factor of the observable water storage anomalies (Neves, Nunes, & Monteiro 2020[10]). Purposeful control of the size and density in the patches of green vegetation seems to be a good practical regulator of water availability, whence the interest in using wetlands as reservoirs (Chisola, Van der Laan, & Bristow 2020[11]).

Ram-pumps placed in the stream of a river become distributed energy resources based on renewable energy: they collect and divert part of the kinetic energy conveyed by the flow of water. Of course, one can ask: isn’t it simpler to put just hydroelectric turbines in that stream, without ram-pumps as intermediary? A ram-pump, properly instrumented with piping, becomes a versatile source of kinetic energy, which can be used for many purposes. In ‘Energy Ponds’, the idea is to use that energy both for creating a system of water retention, and for generating electricity. The former has its constraints. The amount of water to adsorb from the river is limited by the acceptable impact on ecosystems downstream. That impact can be twofold. Excessive adsorption upstream can expose the downstream ecosystems to dangerously low water levels, yet the opposite can happen as well: when we consider wetlands as pseudo-reservoirs, and thus as a reserve of water, its presence can stabilize the flow downstream (Hunt et al. 2022[12]), and the biological traits of ecosystems downstream are not necessarily at risk (Zhao et al. 2020[13]). Strong local idiosyncrasies can appear in that respect (Xu et al. 2020[14]).

Major rivers, even those in plains, have a hydraulic head counted in dozens of meters and a flow rate per second in the hundreds of cubic meters per second. With the typical efficiency of ram-pumps ranging from 35% to 66%, the basic physical model of ram-pumping (Fatahi-Alkouhi et al. 2019[15];  Zeidan & Ostfeld 2021[16]) allows pumping from the river more water than it is possible to divert into the local wetlands.  

Thus, two sub-streams are supposed to be ram-pumped in the ‘Energy Ponds’ system: one sub-stream ultimately directed to and retained in the local wetlands, and another one being the non-retainable excess, to be redirected immediately back into the river. The latter can immediately flow through hydroelectric turbines placed as close as possible to ram-pumps, in order not to lose kinetic energy. The other goes further, through the system of elevated tanks. Why introducing elevated tanks in the system? Isn’t it better to direct the retainable flow of water as directly as possible to wetlands, thus along a trajectory as flat as is feasible in the given terrain? The intuition behind elevated tanks is the general concept of Roman siphon (e.g. Angelakis et al. 2020[17]). When we place an artificially elevated tank along the flow of water from the river to the wetlands, it allows transporting water over a longer distance without losing kinetic energy in that flow. Therefore, the wetlands themselves, as well as the points of discharge from the piping of ‘Energy Ponds’ can be spread over a greater radius from the point of intake from the river. That gives flexibility as regards adapting the spatial distribution of the whole installation to the landscape at hand. Elevated water tanks can increase the efficiency of water management (Abunada et al. 2014[18]; Njepu, Zhang & Xia 2019[19]).

A water tank placed at the top of the Roman siphon is a reserve of energy in pumped storage, and thus allows creating a flow with the same kinetic energy as was generated by ram-pumps, yet further away from the point of intake in the river. Depending on the volumetric capacity and the relative height of the elevated tank, various amounts of energy can be stored. Two points are to consider in that respect. ‘Energy Ponds’ is supposed to be a solution as friendly as possible to the landscape. Big water towers are probably to exclude, and the right solution for elevated tanks seems closer to those encountered in farms, with relatively light structure. If there is need to store more energy in a local installation of ‘Energy Ponds’, there can be more such elevated tanks, scattered across the landscape. With respect to the relative height, documented research indicates a maximum cost-effective elevation between 30 and 50 meters, with 30 looking like a pragmatically conservative estimate (Inthachot et al. 2015[20]; Guo et al. 2018[21]; Li et al. 2021[22]). Elevated tanks such as conceived in ‘Energy Ponds’ can be various combinations of small equalizing tanks (serving just to level up intermittence in the flow of water), and structures akin raingardens (see e.g. Bortolini & Zanin 2019[23]), thus elevated platforms with embedded vegetal structures, capable of retaining substantial amounts of water.

As an alternative to artificial elevated tanks, and to the extent of possibilities offered by natural landscape, a mutation of the STORES technology (short-term off-river energy storage) can be used (Lu et al. 2021[24]; Stocks et al. 2021[25]). The landscape needed for that specific solution is a low hill, located next to the river and to the wetland. The top of such hill can host a water reservoir.

The whole structure of ‘Energy Ponds’, such as conceptually set for now, looks like a wetland, adjacent to the channel of a river, combined with tubular structures for water conduction, ram pumps and hydroelectric turbines. As for the latter, we keep in mind the high likelihood of dual stream: one close to ram-pumps, the other one after the elevated tanks. Proper distribution of power between the generating units can substantially reduce the amount of water used to generate the same amount of energy (Cordova et al. 2014[26]).

Hydroelectricity is collected in energy storage installations, which sell electricity to its end users. The whole concept follows the general stream of research on creating distributed energy resources coupled with landscape restoration (e.g.  Vilanova & Balestieri 2014[27]; Vieira et al. 2015[28]; Arthur et al. 2020[29] ; Cai, Ye & Gholinia 2020[30] ; Syahputra & Soesanti 2021[31]). In that path of research, combinations of energy sources (small hydro, wind and photovoltaic) plus battery backup, seem to be privileged as viable solutions, and seem allowing investment in RES installations with an expected payback time of approximately 10 – 11 years (Ali et al. 2021[32]). For the sake of clarity in the here-presented conceptualization of ‘Energy Ponds’, only the use of hydropower is considered, whilst, of course, the whole solution is open to adding other power sources to the mix, with limitations imposed by the landscape. As wetlands are an integral component of the whole concept, big windfarms seem excluded from the scope of energy sources, as they need solid support in the ground. Still, other combinations are possible. Once there is one type of RES exploited in the given location, it creates, with time, a surplus of energy which can be conducive to installing other types of RES power stations. The claim seems counterintuitive (if one source of energy is sufficient, then it is simply sufficient and crowds out other sources), yet there is evidence that local communities can consider RES according to the principle that ‘appetite grows as we eat’ (Sterl et al. 2020[33]). Among different solutions, floating solar farms, located on the surface of the wetland, could be an interesting extension of the ‘Energy Ponds’ concept (Farfan & Breyer 2018[34]; Sanchez et al. 2021[35]).

Material and methods

The general method of studying the feasibility of ‘Energy Ponds’ is always specific to a location, and it unfolds at two levels: a) calibrating the basic physical properties of the installation and b) assessing its economic viability. Hydrological determinants are strongly idiosyncratic, especially the amount of water possible do adsorb from the local river and to retain in wetlands, and the impact of retention in wetlands upon the ecosystems downstream. With wetlands as a vital component, the conceptual scheme of the ‘Energy Ponds’ naturally belongs to plains, as well as to wide river valleys surrounded by higher grounds. That kept in mind, there are conceptual developments as regards artificially created wetlands in the mountains (Shih & Hsu 2021[36]).

The local feasibility of ‘Energy Ponds’ starts with the possible location and size of wetlands. Places where wetlands either already exist or used to exist in the past, before being drained, seem to be the most natural, as local ecosystems are likely to be more receptive to newly created or expanded wetlands. Conflicts in land management between wetlands and, respectively, farmland and urban settlements, should be studied. It seems that the former type is sharper than the latter. There is documented technological development as regards the so-called Sponge Cities, where urban and peri-urban areas can be transformed into water-retaining structures, including the wetland-type ones (Sun et al. 2020[37]; Köster 2021[38]; Hamidi, Ramavandi & Sorial 2021[39]). On the other hand, farmland is a precious resource and conflicts between retention of water and agriculture are (and probably should be) settled in favour of agriculture.   

Quantitatively, data regarding rivers boils down to the flow rate in cubic meters per second, and to the hydraulic head. Flow and head are the elementary empirical observables of the here-presented method, and they enter into the basic equation of ram-pumping, as introduced by Zeidan & Ostfeld (2021 op.cit.), namely:

HR*QR*η = HT*QT                     (1)

…where HR is the hydraulic head of the river, QR is its flow rate in m3/s, HT is the relative elevation where water is being ram-pumped, QT is the quantity of pumped water, and η is a coefficient of general efficiency in the ram-pumps installed. That efficiency depends mostly on the length of pipes and their diameter, and ranges between 35% and 66% in the solutions currently available in the market. Knowing that QT is a pre-set percentage p of QR and, given the known research, p = QT/QR ≤ 20%, it can be written that QT = QRp. Therefore, equation (1) can be transformed:  

η = [HT*QR*p] / [HR*QR] = HT*p / HR              (2)

The coefficient η is predictable within the range that comes with the producer’s technology. It is exogenous to the ‘Energy Ponds’ system as such unless we assume producing special ram-pumps in the project. With a given flow per second in the river, efficiency η natural dictates the amount of water being ram-pumped, thus the coefficient of adsorption p.  Dual utilization of the ram-pumped flow (i.e. retention in wetlands and simple recirculation through proximate turbines) allows transforming equation (1) into equivalence (3):

{[HRQRHTp / HR] = [HRQC + HTQW] } {QRHTp = [HRQC + HTQW]}         (3)

…where C stands for the sub-flow that is just being recirculated through turbines, without retention in wetlands, and QW is the amount retained. The balance of proportions between QC and QW is an environmental cornerstone in the ‘Energy Ponds’ concept, with QW being limited by two factors: the imperative of supplying enough water to ecosystems downstream, and the retentive capacity of the local wetlands. The latter is always a puzzle, and its thorough understanding requires many years of empirical observation. Still, a more practical method is proposed here: observable change in the surface of wetlands informs about changes in the amount of water stored. Of course, this is a crude observable, yet it can serve for regulating the amount of water conducted into the wetland.      

The hydraulic head of the river (HR) is given by the physical properties thereof, and thus naturally exogenous. Therefore, the fundamental technological choice in ‘Energy Ponds’ articulates into four ‘big’ variables: a) the producer-specific technology of ram-pumping b) the relative height HT of elevated tanks c) the workable fork of magnitudes in the amount of water QW to store in wetlands, and d) the exact technology of energy storage for hydroelectricity. These 4 decisions together form the first level of feasibility as regards the ‘Energy Ponds’ concept. They are essentially adaptive: they manifest the right choice for the given location, with its natural and social ecosystem.

A local installation of ‘Energy Ponds’ impacts the local environment at two levels, namely the retention of water, and the supply of energy. Water retained in wetlands has a beneficial impact on the ecosystem, yet it is not directly consumable: it needs to pass through the local system of supply in potable water first. The direct consumable generated by ‘Energy Ponds’ is hydroelectricity. Besides, there is some empirical evidence for a positive impact of wetlands upon the value of the adjacent, residential real estate (Mahan et al. 2000[40]; Tapsuwan et al. 2009[41]; Du & Huang 2018[42]). Thus comes the second level of feasibility for ‘Energy Ponds’, namely the socio-economic one. As ‘Energy Ponds’ is an early-stage concept, bearing significant uncertainty, the Net Present Value (NPV) of discounted cash flows seems suitable in that respect. Can the thing pay its own bills, and bring a surplus?

Answering that question connects once again to the basic hydrological metrics, namely head and flow. Hydroelectric power is calculated, in watts, as: water density (1000 kg/m3) * gravity acceleration constant (9,8 m/s2) * Net Head (meters) * Q (water flow rate m3/s). The output of electricity is derived from the power generated. It is safe to assume 50 weeks per year of continuous hydrogeneration, with the remaining time reserved for maintenance, which gives 50*7*24 = 8400 hours. Based on the previous formulations, power W generated in an installation of ‘Energy Ponds’ can be expressed with equation (4), and the resulting output E of electricity is given by equation (5):

W[kW] = ρ * g * (HRQC + HTQW) = 9,81 * (HRQC + HTQW)      (4)

E[kWh] = 8400 * W           (5)

The Net Present Value (NPV) of cash flow in an ‘Energy Ponds’ project is the residual part of revenue from the sales of electricity, as in equation (6).

The revenue is calculated as RE = PE*E , with PE standing for the price of electricity per 1 kWh. Investment outlays and the current costs of maintenance can be derived from the head and the flow specific to the given location. In that respect, the here-presented method, including parameters in equation (6), follows that by Hatata, El-Saadawi, & Saad (2019[43]). A realistic, technological lifecycle of an installation can be estimated at 12 years. Crossflow turbines seem optimal for flow rates inferior or equal to 20 m3 per second, whilst above 20 m3 Kaplan turbines look like the best choice. Investment and maintenance costs relative to ram pumps, elevated tanks, and the necessary piping remain largely uncertain, and seemingly idiosyncratic as regards the exact location and its physical characteristics. That methodological difficulty, seemingly inherent to the early conceptual phase of development in the ‘Energy Ponds’ concept, can be provisionally bypassed with the assumption that those hydraulic installations will consume the cost which would normally correspond to the diversion weir and intake, as well as to the cost of the powerhouse building. The variable IH corresponds to investment outlays in the strictly hydrological part of ‘Energy Ponds’ (i.e. ram-pumps, piping, and elevated tanks), whilst the ITU component, on the other hand, represents investment in the strictly spoken turbines and the adjacent power equipment (generator, electrical and mechanical auxiliary, transformer, and switchyard). The LCOS variable in equation (6) is the Levelized Cost of Storage, estimated for a discharge capacity of 6 hours in Li-Ion batteries, at €0,28 per 1 kWh (Salvini & Giovannelli 2022[44]; Chadly et al. 2022[45]). The ‘0,000714’ factor in equation (6) corresponds to the 6 hours of discharge as a fraction of the total 8400 working hours of the system over 1 year.

Case study with calculations

The here presented case study simulates the environmental and economic impact which could possibly come from the hypothetical development of the ‘Energy Ponds’ concept in author’s own country, namely Poland. The concept is simulated in the mouths of 32 Polish rivers, namely: Wisła, Odra, Warta, Narew, Bug, Noteć, San, Wieprz, Pilica, Bzura, Biebrza, Bóbr, Łyna, Drwęca, Barycz, Wkra, Gwda, Prosna, Dunajec, Brda, Pisa, Wisłoka, Nida, Nysa Kłodzka, Wisłok, Drawa, Krzna, Parsęta, Rega, Liwiec, Wełna, Nysa Łużycka (the spelling is original Polish). Flow rates, in cubic meters per second, as observed in those mouths, are taken as the quantitative basis for further calculations, and they are provided in Table 1, below. Figure 1, further below, presents the same graphically, on the map of Poland. The corresponding locations are marked with dark ovals. There are just 28 ovals on the map, as those located near Warsaw encompass more than one river mouth. That specific place in Poland is exceptionally dense in terms of fluvial network. Further in this section, one particular location is studied, namely the mouth of the Narew River, and it is marked on the map as a red oval.  

Table 1

River nameMouth opening on…Average flow rate [m3/s]River nameMouth opening on…Average flow rate [m3/s]
WisłaBaltic Sea1080ProsnaWarta17,4
OdraBaltic Sea567DunajecWisła85,5
SanWisła129Nysa KłodzkaOdra37,7
BiebrzaNarew35,3ParsętaBaltic Sea29,1
BóbrOdra44,8RegaBaltic Sea21,1
BaryczOdra18,8Nysa ŁużyckaOdra31

Figure 2

The hypothetical location of Energy Ponds installations at the mouths of rivers is based on the availability of hydrological data, and more specifically the flow rate in cubic meters per second. That variable is being consistently observed at the mouths of rivers, for the most part, unless some specific research is conducted. Thus, locating the simulated Energy Ponds at the mouths of rivers is not a substantive choice. Real locations should be selected on the grounds of observable properties in the local ecosystem, mostly as regards the possibility of storing water in wetlands. 

The map of location chosen for this simulation goes pretty much across all the available fluvial valleys in Poland, the physical geography of which naturally makes rivers grow as they head North, and therefore the northern part of the country gives the most water to derive from rivers. Once again, when choosing real locations for the Energy Ponds installations, more elevated ground is a plausible location as well. Most of the brown ovals on the map are in the broad vicinity of cities. This is another feature of the Polish geography: high density of population, the latter being clearly concentrated along rivers. This is also an insight into the function of Energy Ponds in real life. Such as it is simulated in Poland, i.e. in a densely populated environment, it is a true challenge to balance environmental services provided by wetlands, on the one hand, and the need for agricultural land, on the other hand. The simulation allows guessing that Energy Ponds can give more to city dwellers than to those living in the countryside.  

Another important trait of this specific simulation for Energy Ponds is the fact that virtually all the locations on the map correspond to places where wetlands used to exist in the past, before being drained and dried for the needs of human settlements. Geological and hydrological conditions are naturally conducive to swamp and pond formation in these ecosystems. It is important to prevent any ideological take on these facts. The present article is far from pushing simplistic claims such as “nature is better than civilisation”. Still, the draining and drying of wetlands in the past happened in the context of technologies which did not really allow reliable construction amidst a wetland-type environment. Today, we dispose of a much better technological base, such as comfortable barge-based houses, for example. The question of cohabitation between wetlands and human habitat can be reconsidered productively.   

Three levels of impact upon the riverine ecosystem are simulated as three hypothetical percentages of adsorption from the river through ram pumping: 5% of the flow, 10% and 20%, where 20% corresponds to the maximum possible pump-out as regards environmental impact. With these assumptions, the complete hypothetical set of 32 installations would yield 5 163 231 600 m3 a year at 5% of adsorption, and, respectively 10 326 463 200 m3 and 20 652 926 400 m3 with the adsorption rates at 10% and 20%.  In 2020, the annual precipitations were around 201,8 billion of m3, which means the 32 hypothetical installations of Energy Ponds could recirculate from 2,5% to 10% of that total volume, and that, in turn, translates into a significant environmental impact. 

Let’s focus on one particular location in order to further understand the environmental impact: the place where the Narew River mouths into Vistula River, north of Warsaw. The town of Nowy Dwór Mazowiecki, population 28 615, is located right at this junction of rivers. With the average consumption of water at the level of households being around 33,7 m3 a year, that local population consumes annually some 964 326 m3 of water. The flow rate in the Narew River close to its mouth into Vistula is 313 m3 per second, which amounts to a total of 9 870 768 000,00 m3 a year. Adsorbing 5%, 10% or 20% from that total flow amounts to, respectively, 493 538 400 m3, 987 076 800 m3, and 1 974 153 600 m3. From another angle, the same annual consumption of water in households, in Nowy Dwór Mazowiecki, corresponds to 0,0098% of the annual waterflow in the river mouth. The ‘Energy Ponds’ concept would allow to recirculate easily into the surrounding ecosystem the entire annual household consumption of water in this one single town.           

Let’s stay in this specific location and translate water into energy, and further into investment. The first issue to treat is a workable approach to using the capacity of ram-pumps in that specific location, or, in other words, a realistic estimation of the total pumped volume QC + QW . Metric flow per second at the mouth of the Narew River is 313 m3 on average. It is out of the question to install ram-pumps across the entire width of the stream, i.e. some 300 metres on average, as Narew is a navigable river. Still, what if we replaced width with length, i.e. what if a row of ram-pumps was placed along one bank of the river, over a total length of 1 km? Then, with the average efficiency of ram-pumps pegged at 50,5%, it can be assumed that 50,5% of the total flow per second, thus 50,5%*313 m3/s = 158,065 m3/s would flow through ram-pumps. With the baseline head of the Narew River being 92 meters, Table 2 below offers an estimation of the electric power thus possible to generate at the mouth, in an installation of ‘Energy Ponds’, according to equation (4).

Table 2 – Electric power [kW] possible to generate in an installation of ‘Energy Ponds’ at the mouth of the Narew River, Poland.

 Percentage of the total flow to be stored in wetlands (QW)
The relative height of elevated tanks5%10%20%
10 meters130 067,65117 478,4892 300,13
20 meters131 602,92120 549,0198 441,19
30 meters133 138,18123 619,54104 582,25

Source: author’s

When the highest possible elevated tanks are chosen (30 m), combined with the lowest percentage of the flow retained in wetlands (5%), electric power generated is the greatest, i.e. 133,138 MW. The optimal point derives logically from natural conditions. Comparatively to the artificially elevated tanks and their 30 meters maximum, the head of the river itself (92 meters) is an overwhelming factor. An interesting aspect of the ‘Energy Ponds’ concept comes out: the most power can be derived from the natural denivelation of terrain, with elevated tanks and their Roman siphon being just an additional source of potential energy. Further calculations, as regards the necessary investment outlays and the cost of storage, demonstrate that the lowest investment per unit of useful power – 987,16 Polish Zloty (PLN) per 1 kW – is reached precisely at the same point. Comparatively, the lowest power – generated at 20% of the flow adsorbed into wetlands and the lowest height of 10 meters of elevated tanks – is connected to the highest investment per unit, namely 1 111,13 PLN per 1 kW.       

The local urban population in Nowy Dwór Mazowiecki represents an annual consumption of electricity amounting to 828 752 048 kWh, and electricity hypothetically generated at that greatest power amounts to 1 118 360 718,72 kWh a year, thus covering, with an overhead, the local needs. This output of energy would be currently worth PLN 845,48 million a year at the retail prices of electricity[47] (i.e. when sold in a local peer-to-peer market). Should it be sold at wholesale prices[48], to the national power grid, it would represent PLN 766,08 million annually. Corrected with an annual Levelized Cost of Storage estimated at 107 161,99 PLN for Li-Ion batteries, that stream of revenue gives a 12-year discounted present value of PLN 5 134 million at retail prices, and PLN 4 647 million at wholesale prices.  With investment outlays estimated, according to the method presented earlier in this article, at some PLN 131,43 million, the project seems largely profitable. As a matter of fact, it could reach a positive Net Present Value already on the first year.

Comparatively, at the point of lowest power and highest investment per unit thereof, thus at 20% of adsorption into wetlands and 10 meters of height in elevated tanks, the 12-year discounted stream of revenue corrected for LCOS would be PLN 3 555,3 million (retail) and PLN 3 217,8 million (wholesale), against an investment of PLN 102,56 million.                   


The above-presented case study in the hypothetical implementation of the ‘Energy Ponds’ concept sums up optimistically. Still, the ‘Energy Ponds’ is still just a concept, and the study of its possible feasibility is hypothetical. That suggests caution, and the need to take a devil’s advocate’s stance. The case study can be utilized for both outlining the positive potential in ‘Energy Ponds’ and showing the possible angles of stress-testing the concept. The real financial value and the real engineering difficulty of investment in the basic hydraulic infrastructure of ‘Energy Ponds’ has been just touched upon, and essentially bypassed with a few assumptions. Those assumptions seem to be holding when written down, but confrontation with real life can bring about unpredicted challenges. This superficiality stems from the author’s own limitations, as an economist attempting to form an essentially engineering solution. Still, even from the economic point of view, one factor of uncertainty emerges: the pace of technological change. The method used for this feasibility study is a textbook one, similar to calculating the Levelized Cost of Energy: there is an initial investment, which we spread over the expected lifecycle of the technology in question. However, technologies can change at an unexpected pace, and the actual lifecycle of an installation – especially a highly experimental one – might be much shorter than expected. In the author’s (very intuitive) perspective, technological uncertainty is a big pinch of salt to add to the results of the case study.

Another factor of uncertainty is the real possibility of restoring wetlands in densely populated areas. Whilst new technologies in construction and agriculture do allow better overlapping between wetlands, cities, and farming, this is still just overlapping. At the bottom line, wetlands objectively take land away from many other possible uses. Literature is far from decisive about solutions in that respect. The great majority of known projects in the restoration of wetlands aim at and end up in restoring wildlife, not in assuring smooth coexistence between nature and civilisation. Some serious socio-economic experimentation might be involved in projects such as ‘Energy Ponds’.

Hydrogeneration in ‘Energy Ponds’ belongs to the category of Distributed Energy Resources (DER). DER systems become more and more popular, across the world, and they prove to be workable solutions in very different conditions of physical geography (McIlwaine et al. 2021[49]). Connection to local markets of energy, and into local smart grids, seems critical for the optimization of DER systems (Zakeri et al. 2021[50]; Touzani et al. 2021[51]; Haider et al. 2021[52]; Zhang et al. 2022[53]). How necessary is the co-existence – or pre-existence – of such a local network for the economically successful deployment of ‘Energy Ponds’?  What happens when the local installation of ‘Energy Ponds’ is the only Distributed Energy Resource around?

[1] Wehn, U., & Montalvo, C. (2018). Exploring the dynamics of water innovation: Foundations for water innovation studies. Journal of Cleaner Production, 171, S1-S19.

[2] Mvulirwenande, S., & Wehn, U. (2020). Fostering water innovation in Africa through virtual incubation: Insights from the Dutch VIA Water programme. Environmental Science & Policy, 114, 119-127.

[3] Wong, T. H., Rogers, B. C., & Brown, R. R. (2020). Transforming cities through water-sensitive principles and practices. One Earth, 3(4), 436-447.

[4] Hogeboom, R. J. (2020). The Water Footprint Concept and Water’s Grand Environmental Challenges. One earth, 2(3), 218-222.

[5] Mohamed, M. M., El-Shorbagy, W., Kizhisseri, M. I., Chowdhury, R., & McDonald, A. (2020). Evaluation of policy scenarios for water resources planning and management in an arid region. Journal of Hydrology: Regional Studies, 32, 100758.

[6] Kumar, P., Avtar, R., Dasgupta, R., Johnson, B. A., Mukherjee, A., Ahsan, M. N., … & Mishra, B. K. (2020). Socio-hydrology: A key approach for adaptation to water scarcity and achieving human well-being in large riverine islands. Progress in Disaster Science, 8, 100134.

[7] Bagstad, K. J., Ancona, Z. H., Hass, J., Glynn, P. D., Wentland, S., Vardon, M., & Fay, J. (2020). Integrating physical and economic data into experimental water accounts for the United States: Lessons and opportunities. Ecosystem Services, 45, 101182.

[8] Harvey, J.W., Schaffranek, R.W., Noe, G.B., Larsen, L.G., Nowacki, D.J., O’Connor, B.L., 2009. Hydroecological factors governing surface water flow on a low-gradient floodplain. Water Resour. Res. 45, W03421,

[9] Phiri, W. K., Vanzo, D., Banda, K., Nyirenda, E., & Nyambe, I. A. (2021). A pseudo-reservoir concept in SWAT model for the simulation of an alluvial floodplain in a complex tropical river system. Journal of Hydrology: Regional Studies, 33, 100770.

[10] Neves, M. C., Nunes, L. M., & Monteiro, J. P. (2020). Evaluation of GRACE data for water resource management in Iberia: a case study of groundwater storage monitoring in the Algarve region. Journal of Hydrology: Regional Studies, 32, 100734.

[11] Chisola, M. N., Van der Laan, M., & Bristow, K. L. (2020). A landscape hydrology approach to inform sustainable water resource management under a changing environment. A case study for the Kaleya River Catchment, Zambia. Journal of Hydrology: Regional Studies, 32, 100762.

[12] Hunt, J. D., Nascimento, A., ten Caten, C. S., Tomé, F. M. C., Schneider, P. S., Thomazoni, A. L. R., … & Senne, R. (2022). Energy crisis in Brazil: Impact of hydropower reservoir level on the river flow. Energy, 239, 121927.

[13] Zhao, P., Li, Z., Zhang, R., Pan, J., & Liu, Y. (2020). Does water diversion project deteriorate the water quality of reservoir and downstream? A case-study in Danjiangkou reservoir. Global Ecology and Conservation, 24, e01235.

[14] Xu, D., Lyon, S. W., Mao, J., Dai, H., & Jarsjö, J. (2020). Impacts of multi-purpose reservoir construction, land-use change and climate change on runoff characteristics in the Poyang Lake basin, China. Journal of Hydrology: Regional Studies, 29, 100694.

[15] Fatahi-Alkouhi, R., Lashkar-Ara, B., & Keramat, A. (2019). On the measurement of ram-pump power by changing in water hammer pressure wave energy. Ain Shams Engineering Journal, 10(4), 681-693.

[16] Zeidan, M., & Ostfeld, A. (2021). Hydraulic Ram Pump Integration into Water Distribution Systems for Energy Recovery Application. Water, 14(1), 21,   

[17] Angelakis, A. N., Voudouris, K. S., & Tchobanoglous, G. (2020). Evolution of water supplies in the Hellenic world focusing on water treatment and modern parallels. Water Supply, 20(3), 773-786

[18] Abunada, M., Trifunović, N., Kennedy, M., & Babel, M. (2014). Optimization and reliability assessment of water distribution networks incorporating demand balancing tanks. Procedia Engineering, 70, 4-13.

[19] Njepu, A., Zhang, L., & Xia, X. (2019). Optimal tank sizing and operation of energy-water supply systems in residences. Energy Procedia, 159, 352-357.

[20] Inthachot, M., Saehaeng, S., Max, J. F., Müller, J., & Spreer, W. (2015). Hydraulic ram pumps for irrigation in Northern Thailand. Agriculture and Agricultural Science Procedia, 5, 107-114.

[21] Guo, X., Li, J., Yang, K., Fu, H., Wang, T., Guo, Y., … & Huang, W. (2018). Optimal design and performance analysis of hydraulic ram pump system. Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, 232(7), 841-855.

[22] Li, J., Yang, K., Guo, X., Huang, W., Wang, T., Guo, Y., & Fu, H. (2021). Structural design and parameter optimization on a waste valve for hydraulic ram pumps. Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, 235(4), 747–765.

[23] Bortolini, L., & Zanin, G. (2019). Reprint of: Hydrological behaviour of rain gardens and plant suitability: A study in the Veneto plain (north-eastern Italy) conditions. Urban forestry & urban greening, 37, 74-86.

[24] Lu, B., Blakers, A., Stocks, M., & Do, T. N. (2021). Low-cost, low-emission 100% renewable electricity in Southeast Asia supported by pumped hydro storage. Energy, 121387.

[25] Stocks, M., Stocks, R., Lu, B., Cheng, C., & Blakers, A. (2021). Global atlas of closed-loop pumped hydro energy storage. Joule, 5(1), 270-284.

[26] Cordova M, Finardi E, Ribas F, de Matos V, Scuzziato M. Performance evaluation and energy production optimization in the real-time operation of hydropower plants. Electr Pow Syst Res 2014;116:201–7.

[27] Vilanova, M. R. N., & Balestieri, J. A. P. (2014). Hydropower recovery in water supply systems: Models and case study. Energy conversion and management, 84, 414-426.

[28] Vieira, D. A. G., Guedes, L. S. M., Lisboa, A. C., & Saldanha, R. R. (2015). Formulations for hydroelectric energy production with optimality conditions. Energy Conversion and Management, 89, 781-788.

[29] Arthur, E., Anyemedu, F. O. K., Gyamfi, C., Asantewaa-Tannor, P., Adjei, K. A., Anornu, G. K., & Odai, S. N. (2020). Potential for small hydropower development in the Lower Pra River Basin, Ghana. Journal of Hydrology: Regional Studies, 32, 100757.

[30] Cai, X., Ye, F., & Gholinia, F. (2020). Application of artificial neural network and Soil and Water Assessment Tools in evaluating power generation of small hydropower stations. Energy Reports, 6, 2106-2118.

[31] Syahputra, R., & Soesanti, I. (2021). Renewable energy systems based on micro-hydro and solar photovoltaic for rural areas: A case study in Yogyakarta, Indonesia. Energy Reports, 7, 472-490.

[32] Ali, M., Wazir, R., Imran, K., Ullah, K., Janjua, A. K., Ulasyar, A., … & Guerrero, J. M. (2021). Techno-economic assessment and sustainability impact of hybrid energy systems in Gilgit-Baltistan, Pakistan. Energy Reports, 7, 2546-2562.

[33] Sterl, S., Donk, P., Willems, P., & Thiery, W. (2020). Turbines of the Caribbean: Decarbonising Suriname’s electricity mix through hydro-supported integration of wind power. Renewable and Sustainable Energy Reviews, 134, 110352.

[34] Farfan, J., & Breyer, C. (2018). Combining floating solar photovoltaic power plants and hydropower reservoirs: a virtual battery of great global potential. Energy Procedia, 155, 403-411.

[35] Sanchez, R. G., Kougias, I., Moner-Girona, M., Fahl, F., & Jäger-Waldau, A. (2021). Assessment of floating solar photovoltaics potential in existing hydropower reservoirs in Africa. Renewable Energy, 169, 687-699.

[36] Shih, S. S., & Hsu, Y. W. (2021). Unit hydrographs for estimating surface runoff and refining the water budget model of a mountain wetland. Ecological Engineering, 173, 106435,

[37] Sun, Y., Deng, L., Pan, S. Y., Chiang, P. C., Sable, S. S., & Shah, K. J. (2020). Integration of green and gray infrastructures for sponge city: Water and energy nexus. Water-Energy Nexus, 3, 29-40.

[38] Köster, S. (2021). How the Sponge City becomes a supplementary water supply infrastructure. Water-Energy Nexus, 4, 35-40.

[39] Hamidi, A., Ramavandi, B., & Sorial, G. A. (2021). Sponge city—an emerging concept in sustainable water resource management: A scientometric analysis. Resources, Environment and Sustainability, 5, 100028.

[40] Mahan, B. L., Polasky, S., & Adams, R. M. (2000). Valuing urban wetlands: a property price approach. Land economics, 100-113.

[41] Tapsuwan, S., Ingram, G., Burton, M., & Brennan, D. (2009). Capitalized amenity value of urban wetlands: a hedonic property price approach to urban wetlands in Perth, Western Australia. Australian Journal of Agricultural and Resource Economics, 53(4), 527-545.

[42] Du, X., & Huang, Z. (2018). Spatial and temporal effects of urban wetlands on housing prices: Evidence from

Hangzhou, China. Land use policy, 73, 290-298.

[43] Hatata, A. Y., El-Saadawi, M. M., & Saad, S. (2019). A feasibility study of small hydro power for selected locations in Egypt. Energy Strategy Reviews, 24, 300-313.

[44] Salvini, C., & Giovannelli, A. (2022). Techno-economic comparison of diabatic CAES with artificial air reservoir and battery energy storage systems. Energy Reports, 8, 601-607.

[45] Chadly, A., Azar, E., Maalouf, M., & Mayyas, A. (2022). Techno-economic analysis of energy storage systems using reversible fuel cells and rechargeable batteries in green buildings. Energy, 247, 123466.

[46] The Pregoła river is located in the Russian Kaliningrad district, whilst much of the Łyna river is located in Poland. The hypothetical location of Energy Ponds is assumed to be in Poland, thus upstream from the mouth into Pregoła. 

[47] last access May 17th, 2022

[48] last access May 17th, 2022

[49] McIlwaine, N., Foley, A. M., Morrow, D. J., Al Kez, D., Zhang, C., Lu, X., & Best, R. J. (2021). A state-of-the-art techno-economic review of distributed and embedded energy storage for energy systems. Energy, 229, 120461.

[50] Zakeri, B., Gissey, G. C., Dodds, P. E., & Subkhankulova, D. (2021). Centralized vs. distributed energy storage–Benefits for residential users. Energy, 236, 121443.

[51] Touzani, S., Prakash, A. K., Wang, Z., Agarwal, S., Pritoni, M., Kiran, M., … & Granderson, J. (2021). Controlling distributed energy resources via deep reinforcement learning for load flexibility and energy efficiency. Applied Energy, 304, 117733.

[52] Haider, R., D’Achiardi, D., Venkataramanan, V., Srivastava, A., Bose, A., & Annaswamy, A. M. (2021). Reinventing the utility for distributed energy resources: A proposal for retail electricity markets. Advances in Applied Energy, 2, 100026.

[53] Zhang, S., May, D., Gül, M., Musilek, P., (2022) Reinforcement learning-driven local transactive energy market for distributed energy resources. Energy and AI (2022), doi:

Stratégies financières

Je continue un peu dans la foulée d’analyse des rapports courants des sociétés de ma liste « technologies nouvelles en énergie ». Je fais de mon mieux pour développer sur les premières observations que j’ai déjà présentées dans « Mes lampes rouges » ainsi que dans « Different paths ». Comme je résume partiellement ce que j’ai lu, ma première conclusion est une confirmation de mes intuitions initiales. Ce que nous appelons l’industrie de l’hydrogène est en fait une combinaison des technologies de pointe (piles à combustible à la base d’hydrogène, par exemple) avec des technologies bien établies – quoi que sujettes à l’innovation incrémentale – comme l’électrolyse ou le stockage des gaz volatiles. Il semble y avoir d’importants effets d’échelle, probablement en raison de la complexité technologique. Les sociétés relativement plus grandes, comme Plug Power ou Fuel Cell Energy, capables d’acquérir d’autres sociétés et leurs technologies, semblent être mieux placées dans la course technologique que des indépendants qui développent des technologies propriétaires de façon indépendante. C’est un truc que j’ai déjà remarqué dans le photovoltaïque et dans l’industrie de véhicules électriques : oui, il y a des petits indépendants prometteurs mais la bonne vieille intégration industrielle, surtout en verticale, semble revenir comme stratégie de choix après de décennies de bannissement. 

J’ai remarqué aussi que la catégorie générique « technologies d’hydrogène » semble attirer du capital de façon un peu inconsidérée. Je veux dire qu’il semble suffisant de dire « Eh, les gars, on invente dans l’hydrogène » pour que les investisseurs se précipitent, peu importe si le modèle d’entreprise est viable et transparent, ou pas-tout-à-fait-vous-comprenez-c’est-confidentiel. Je vois dans l’industrie d’hydrogène le même phénomène que j’observais, il y a encore 4 ou 5 ans, dans le photovoltaïque ou bien chez Tesla : lorsqu’une technologie nouvelle commence à prendre son envol en termes de ventes, les organisations qui s’y greffent et développent sont un peu démesurées ainsi qu’exagérément dépensières et il faut du temps pour qu’elles se fassent vraiment rationnelles.    

Pour gagner un peu de distance vis-à-vis le business d’hydrogène, je commence à piocher dans les rapports courants d’autres sociétés sur ma liste. Tesla vient en tête. Pas sorcier, ça. C’est la plus grosse position dans mon portefeuille boursier. Je lis donc le rapport courant du 4 août 2022 qui rend compte de 13 propositions soumises à l’assemblée générale d’actionnaires de Tesla le même 4 août, ainsi que de l’opération de fractionnement d’actions prévue pour la seconde moitié d’août. Ce dernier truc, ça m’intéresse peut-être le plus. La version officielle qui, bien entendu, sera mise à l’épreuve par le marché boursier, est que le Conseil D’Administration souhaite rendre les actions de Tesla plus accessibles aux investisseurs et employés et procédera donc, le 17 août 2022, à un fractionnement d’actions en proportion trois-pour-une en forme d’une dividende-actions. Chaque actionnaire enregistré ce 17 août 2022 verra le nombre de ses actions multiplié par trois. Les nouvelles actions fractionnées entreront en circulation boursière normale le 24 août 2022.

Il est vrai que les actions de Tesla sont plutôt chères en ce moment : presque $900 la pièce et ceci après la forte dépréciation dans la première moitié de l’année. Formellement, le fractionnement en proportion trois-pour-une devrait diviser cette cotation par trois, seulement le marché, ça suit les règles d’économie, pas d’arithmétique pure. Je pense que par la fin de 2022 on aura trois fois plus d’actions de Tesla flottantes et cotées à plus qu’un troisième du prix d’aujourd’hui encore qu’entre temps, il y aura des turbulences, je vous le dis. J’attache donc ma ceinture de sécurité – en l’occurrence c’est une position en Apple Inc., bien plus stable et respectable que Tesla – et j’attends de voir la valse boursière autour de ces actions fractionnées.

A part cette histoire de fractionnement, les autres 12 propositions couvrent 4 qui ont été acceptées – dont une relative au fractionnement déjà signalé – ainsi que 8 propositions non-acceptées. Les 4 acceptées sont relatives à, respectivement :

>> la nomination de deux personnes au Conseil D’Administration

>> l’accès par procuration, proposition sans engagement présentée par actionnaires en minorité

>> la ratification du choix de PricewaterhouseCoopers LLP comme auditeur financier de Tesla pour l’année comptable 2022

>> l’accroissement du nombre d’actions ordinaires de Tesla par 4 000 000

Les 8 propositions rejetées se groupent en deux catégories distinctes d’une façon intéressante. Il y en a donc deux qui viennent des cadres gestionnaires de Tesla et qui postulaient de modifier l’acte d’incorporation de Tesla de façon à éliminer la règle de majorité qualifiée de 66 et 2/3% dans les votes, ainsi qu’à réduire à 2 ans le mandat des directeurs du Conseil d’Administration. Ces deux propositions-là ont perdu car elles n’avaient pas… de majorité qualifiée de 66 et 2/3%. Les 6 propositions restantes parmi les non-acceptées étaient toutes des propositions sans engagement de la part d’actionnaires minoritaires et toutes les 6 demandaient des rapports additionnels ou bien des changements afférents à, respectivement : la qualité de l’eau, travail forcé d’enfants, le lobbying, la liberté d’association, arbitrage dans les affaires d’emploi, la diversité au sein du Conseil d’Administration, les politiques internes contre le harassement et la discrimination.

Dans le vocabulaire politique de mon pays, la Pologne, nous avons l’expression « compter les sabres ». Elle désigne des votes qui sont perdus d’avance mais qui servent à compter la taille de la coalition possible que le proposant donné pourrait rallier pour quelque chose de plus sérieux. Je bien l’impression que quelqu’un chez Tesla commence à compter les sabres.

Je passe au rapport courant de Tesla du 20 juillet 2022 qui, en fait, annonce leur rapport financier du 2ème trimestre 2022. Ça a l’air bien. Le bénéfice net pour la première moitié de 2022 a triplé par rapport à la même période de 2021, le flux de trésorerie se fait plus robuste. Rien à dire.                

Je tourne vers un modèle d’entreprise beaucoup plus fluide, donc celui de Nuscale Power ( ). Lorsqu’on lit la présentation générale de ce business ( ), tout colle à merveille : NuScale Power fournit des petits réacteurs nucléaires innovatifs, où un module peut fournir 77 mégawatts de puissance. Seulement, lorsque je commence à lire leur rapport annuel 2021, ça se corse, parce que le rapport est publié par l’entité nommée Spring Valley Acquisition Corporation, qui se présente comme une société coquille incorporée dans les îles Cayman, sous la forme légale de société exonérée. Le management déclarait, dans le rapport annuel 2021, que le but de Spring Valley Acquisition Corporation est de conduire une fusion ou bien une acquisition, un échange d’actions ou bien leur achat, une acquisition d’actifs, une réorganisation ou bien une autre forme de regroupement d’entreprises. Au mois de mars 2021 ; Spring Valley Acquisition Corporation est entrée en un accord tripartite, accompagné d’un plan de fusion, avec sa filiale en propriété exclusive, Spring Valley Merger Sub, Inc., incorporée dans l’état de Delaware, ainsi qu’avec Dream Holdings Inc., une autre société incorporée dans le Delaware, celle-ci sous la forme de société d’utilité publique. C’est une nouveauté dans la loi des sociétés dans le Delaware, introduite en 2013. Un article intéressant à ce sujet est accessible sur « Harward Law School Forum on Corporate Governance ».

Ainsi donc, en mars 2021, Spring Valley Acquisition Corporation, Spring Valley Merger Sub, Inc. et Dream Holdings Inc. avaient convenu de conduire un regroupement d’entreprises avec AeroFarms. Dream Holdings fusionne avec Spring Valley Merger Sub. En octobre 2021, l’accord en question a été résilié. En décembre 2021, Spring Valley Acquisition Corporation entre en un nouvel accord tripartite, encore une fois avec la participation de Spring Valley Merger Sub. Cette fois, Spring Valley Merger Sub est introduite comme une LLC (société à responsabilité limitée) incorporée dans l’état d’Oregon. La troisième partie de l’accord est NuScale Power LLC, aussi incorporée en Oregon. Poursuivant cet accord, Spring Valley Acquisition Corporation change de lieu d’incorporation des îles Cayman pour l’état de Delaware, pendant que Spring Valley Merger Sub LLC fusionne avec et en NuScale Power LLC. Après la fusion, Spring Valley change de nom et devient NuScale Power Corporation.

Comment a marché la combine ? Eh bien, voici une annonce courante de NuScale Power, datant d’hier (10 août), où NuScale donne un aperçu de leurs résultats pour le 2nd trimestre 2022. La perte d’exploitation pour cette première moitié de l’année 2022 était de 44,75 millions de dollars, un peu moins que dans la première moitié de 2021. Leurs actifs ont presque triplé en 12 mois, de $121,2 millions à $407,3 millions. Côté exploitation, une nouvelle entité opérationnelle est créé sous le nom de « VOYGR™ Services and Delivery (VSD) » avec la mission d’organiser les services, les fournitures et la gestion clients pour la technologie VOYGR™. Cette dernière est la technologie pour bâtir et exploiter des centrales nucléaires à puissance moyenne sur la base de « NuScale Power Module™ », soit avec 4 modules dedans et une puissance de 924 mégawatts de puissance électrique (VOYGR-4) soit avec 6 modules (VOYGR-6).

Cette comparaison rapide d’évènements relativement récents chez Tesla et NuScale Power me conduit à la conclusion que si je veux comprendre à fond un modèle d’entreprise, il faut que je m’intéresse plus (que je l’avais fait jusqu’à présent) à ce qui se passe dans les passifs du bilan. Je vois que des différentes phases d’avancement dans le développement d’une technologie s’accompagnent des stratégies financières très différentes et le succès technologique dépend largement du succès de ces mêmes stratégies.

Different paths

I keep digging in the business models of hydrogen-oriented companies, more specifically five of them: 

>> Fuel Cell Energy

>> Plug Power

>> Green Hydrogen Systems

>> Nel Hydrogen

>> Next Hydrogen (previously BioHEP Technologies Ltd.)

I am studying their current reports. This is the type of report which listed companies publish when something special happens, which goes beyond the normal course of everyday business, and can affect shareholders. I have already started with Fuel Cell Energy and their current report from July 12th, 2022 ( ), where they disclose a deal with a group of financial institutions: Jefferies LLC, B. Riley Securities, Inc., Barclays Capital Inc., BMO Capital Markets Corp., BofA Securities, Inc., Canaccord Genuity LLC, Citigroup Global

Markets Inc., J.P. Morgan Securities LLC and Loop Capital Markets LLC. Strange kind of deal, I should add. Those 10 financial firms are supposed to either buy or intermediate in selling to third parties parcels of 95 000 000 shares in the equity of Fuel Cell Energy. The tricky part is that the face value of those shares is supposed to be $0,0001 per share, just as it is the case with the ordinary 837 000 000 shares outstanding, whilst the market value of Fuel Cell Energy’s shares is currently above $4,00 per share, thus carrying an addition of thousands of percentage points of capital to pay.

It looks as if the part of equity in Fuel Cell Energy which is free floating in the stock market – quite a tiny part of their share capital – was becoming subject to quick financial gambling. I don’t like it. Whatever. Let’s go further, i.e. to the next current report of Fuel Cell Energy, that from July 7th, 2022 ( ). It brings updates on two projects:

>> The Toyota Project: a 2,3 megawatt trigeneration platform for Toyota at the Port of Long Beach, California.

>> The Groton Project: a 7.4 MW platform at the U.S. Navy Submarine Base in Groton, Connecticut.

Going further back in time, I browse through the current report from June 9th, 2022 ( ). It is the official version of a press release regarding financial and operational results of Fuel Cell Energy by the end of the 1st quarter 2022. As I am reading through it, I find data about other projects:

>> Joint Development Agreement with ExxonMobil, related to carbon capture and generation, which includes the 7,4 MW LIPA Yaphank fuel cell project

>>  a carbon capture project with Canadian National Resources Limited

>> a program with U.S. Department of Energy regarding solid oxide. I suppose that ‘solid oxide’ stands for solid oxide fuel cells, which use a solid, ceramic core of fuel, which is being oxidized and produces energy in the process.     

I pass to the current reports of Plug Power ( ). Interesting things start when I go back to the current report from June 23rd, 2022 ( ). This is a summary presentation of something which looks like the company’s strategy. Apparently, Plug Power plans to have 13 plants with Green Hydrogen running in the United States by 2025, with a total expected yield of 500 tons per day. In a more immediate perspective, the company plans to locate 5 new plants in the U.S. over 2022 (total capacity of 70 tons per day) and 2023 (200 tons per day). Further, I read that what I thought was a hydrogen-focused company, has, in fact, a broader spectrum of operations: eFuel and methanol, ammonia, vehicle refueling, blending and heating, refining of natural oil, and the storage of renewable energy.  

As part of its strategy, Plug Power announces the acquisitions of companies supposed to bring additional technological competences: Frames Group ( ) with power transmission systems and technology for building electrolyzers, ACT (Applied Cryo Technologies: ) for cryogenics, and Joule ( ) for the liquefaction of hydrogen. My immediate remark as regards those acquisitions, sort of intellectually straight-from-the-oven-still-warm-sorry-but-I-told-you-still-warm, is that Plug Power is acquiring a broad technological base rather than a specialized one. Officially, those acquisitions serve to enhance the Plug Power’s capacity as regards the deployment of hydrogen-focused technologies. Yet, as I am rummaging through the websites of those acquired companies, their technological competences go far beyond hydrogen.

Sort of contingent (adjacent?) to that current report is the piece of news, still on the Plug Power’s investors-relations site, from June 8th, 2022. It regards the deployment of a project in Europe, more specifically in the Port of Antwerp-Bruges ( ). This is supposed to be something labelled as a ‘Gigafactory’.

A little bit earlier this year, on my birthday, May 9th, Plug Power published a current report ( ) coupled with a quarterly financial report ( ). Apparently, in the 1st quarter 2022, they had revenues 96% higher than 1Q 2021. Nice. There are interesting operational goals signaled in that current report. Plug Power plans to reduce services costs on a per unit basis by 30% in the 12 months following the report, thus until the end of the 1st quarter 2023. The exact quote is: ‘Plug remains focused on delivering on our previously announced target to reduce services costs on a per unit basis by 30% in the next 12 months, and 45% by the end of 2023. We are pleased to report that we have begun to see meaningful improvement in service margins on fuel cell systems and related infrastructure with a positive 30% increase in first quarter of 2022 versus the fourth quarter of 2021. The service margin improvement is a direct result of the enhanced technology GenDrive units that were delivered in 2021 which reduce service costs by 50%. The performance of these enhanced units demonstrates that the products are robust, and we expect these products will help support our long-term business needs. We believe service margins are tracking in the right direction with potential to break even by year end’.

When a business purposefully and effectively works on optimizing margins of profit, and the corresponding costs, it is a step forward in the lifecycle of the technologies used. This is a passage from the phase of early development towards late development, or, in other words, it is the phase when the company starts getting in control of small economic details in its technology.

I switch to the next company on my list, namely to Green Hydrogen Systems (Denmark, ). They do not follow the SEC classification of reports, and, in order to get an update on their current developments, I go to their ‘Announcements & News’ section ( ).  On July 18th, 2022, Green Hydrogen Systems held an extraordinary General Meeting of shareholders. They amended their Articles of Association, as regards the Board of Directors, and the new version is: ‘The board of directors consists of no less than four and no more than nine members, all of whom must be elected by the general meeting. Members of the board of directors must resign at the next annual general meeting, but members of the board of directors may be eligible for re-election’. At the same extraordinary General Meeting, three new directors have been elected to the Board, on the top of the six already there.

To the extent that I know the Scandinavian ways of corporate governance, appointment of new directors to the Board usually comes with new business ties of the company. Those people are supposed to be something like intermediaries between the company and some external entities (research units? other companies? NGOs?). That change in the Board of Directors at Green Hydrogen Systems suggests something like the broadening of their network. That intuition is somehow confirmed by an earlier announcement, from June 13th ( ). The three new members of the Board come, respectively, from: Vestas Wind Systems, Siemens Energy, and Sonnedix ( ).

Still earlier this year, on April 12th, Green Hydrogen Systems announced ‘design complications in its HyProvide® A-Series platform’, and said complications are supposed to affect adversely the financial performance in 2022 ( ). When I think about it, design normally comes before its implementation, and therefore before any financial performance based thereon. When ‘design complications’ are serious enough for the company to disclose them and announce a possible negative impact on the financial side of the house, it means some serious mistakes years earlier, when that design was being conceptualized. I say ‘years’ because I notice the trademark symbol ‘®’ by the name of the technology. That means there had been time to: a) figure out the design b) register it as a trademark. That suggests at least 2 years, maybe more.

I quickly sum up my provisional conclusions from browsing current reports at Fuel Cell Energy, Plug Power, and Green Hydrogen Systems. I can see three different courses of events as regards the business models of those companies. At Fuel Cell Energy, broadly spoken marketing, including financial marketing, seems to be the name of the game. Both the technology and the equity of Fuel Cell Energy seems to be merchandise for trading. My educated guess is that the management of Fuel Cell Energy is trying to attract more financial investors to the game, and to close more technological deals, of the joint-venture type, at the operational level. It further suggests an attempt at broadening the business network of the company, whilst keeping the strategic ownership in the hands of the initial founders. As for Plug Power, the development I see is largely quantitative. They are broadening their technological base, including the acquisitions of strategically important assets, expanding their revenues, and ramping up their operational margins. This a textbook type of industrial development. Finally, at Green Hydrogen Systems, this still seems to be the phase of early development, with serious adjustments needed to both the technology owned and the team that runs it.

Those hydrogen-oriented companies seem to be following different paths and to be at different stages in the lifecycle of their technological base.

Mes lampes rouges

Me revoilà, je me suis remis à blogguer après plusieurs mois de pause. Il fallait que je prenne soin de ma santé et entretemps, je repensais mes priorités existentielles et cette réflexion pouvait très bien avoir quelque chose à faire avec les opioïdes que je prenais à l’hôpital après mon opération.

Redémarrer après un temps aussi long est un peu dur et enrichissant en même temps. C’est comme si j’enlevais de la rouille d’une vieille machine. J’adore des vieilles machines que je peux dérouiller et réparer. J’ai besoin de quelques vers d’écriture pour m’orienter. Lorsque j’écris, c’est comme si je libérais une forme d’énergie : il faut que j’y donne une direction et une forme. Je suis en train de travailler sur deux trucs principaux. L’un est mon concept d’Energy Ponds : une solution complexe qui combine l’utilisation des béliers hydrauliques pour accumuler et retenir l’eau dans des structures marécageuses ainsi que pour générer de l’électricité dans des turbines hydroélectriques. L’autre truc c’est ma recherche sur les modèles d’entreprise dans le secteur amplement défini comme nouvelles sources d’énergie. Là, je m’intéresse aux entreprises dans lesquelles je peux investir via la Bourse, donc les véhicules électriques (comme investisseur, je suis in fidèle de Tesla), les systèmes de stockage d’énergie, la production d’hydrogène et son utilisation dans des piles à combustible, le photovoltaïque, l’éolien et enfin le nucléaire.

Intuitivement, je concentre mon écriture sur le blog sur ce deuxième sujet, donc les modèles d’entreprise. Raison ? Je pense que c’est à cause de la complexité et le caractère un peu vaseux du sujet. Le concept d’Energy Ponds, quant à lui, ça se structure peu à peu comme j’essaie – et parfois je réussis – à y attirer l’intérêt des gens aux compétences complémentaires aux miennes. En revanche, les modèles d’entreprise, c’est vaseux en tant que tel, je veux dire au niveau théorique, ça me touche à plusieurs niveaux parce que c’est non seulement de la science pour moi mais aussi une stratégie d’investissement en Bourse. Par ailleurs, je sais que lorsque je blogue, ça marche le mieux avec de tels sujets, précisément : importants et vaseux en même temps.

Voici donc une liste de sociétés que j’observe plus ou moins régulièrement :

>> Tesla

>> Rivian

>> Lucid Group

>> Nuscale Power 

>> First Solar

>> SolarEdge

>> Fuel Cell Energy

>> Plug Power

>> Green Hydrogen Systems

>> Nel Hydrogen

>> Next Hydrogen (précédemment BioHEP Technologies Ltd.)

>> Energa

>> PGE

>> Tauron

>> ZPUE    

La première différentiation sur cette liste c’est ma propre position comme investisseur. Je tiens des positions ouvertes sur Tesla, Nuscale Power, Energa, PGE, Tauron et ZPUE. J’en ai eu dans le passé sur Lucid Group, First Solar et SolarEdge. En revanche, Rivian, Fuel Cell Energy, Plug Power, Green Hydrogen Systems, Nel Hydrogen ainsi que Next Hydrogen – ceux-là, je regarde et j’observe sans y toucher.

La deuxième différentiation est relative aux flux opérationnels de trésorerie : il y en a des profitables (Tesla, First Solar, SolarEdge, Energa, PGE, Tauron et ZPUE) et des pas-tout-à-fait-et-ça-va-venir-mais-pas-encore profitables (Rivian, Lucid Group, Nuscale Power, Fuel Cell Energy, Plug Power, Green Hydrogen Systems, Nel Hydrogen, Next Hydrogen).   

Comme je viens de faire ces deux classifications, il me vient à l’esprit que j’évalue les modèles d’entreprise selon le critère de revenu propriétaire tel que défini par Warren Buffett. A ce propos, vous pouvez consulter soit le site relations investisseurs de son fonds d’investissement Berkshire Hathaway ( ) soit un très bon livre de Robert G.Hagstrom « The Warren Buffett Way » (John Wiley & Sons, 2013, ISBN 1118793994, 9781118793992). Les entreprises qui dégagent un surplus positif de bénéfice net et amortissement sur les dépenses capitalisées en actifs productifs sont celles qui sont déjà mûres et stables, donc financièrement capables de lancer quelque chose comme une nouvelle vague de changement technologique. En revanche, celles où cette valeur résiduelle « bénéfice net plus amortissement moins dépenses capitalisées en actifs productifs » est négative ou proche de zéro sont celles qui ont toujours besoin de venir à termes avec la façon dont ils conduisent leur business et sont donc incapables de lancer un nouveau cycle de changement technologique sans assistance financière externe.

Je me concentre sur les sociétés spécialisées dans l’hydrogène :  Fuel Cell Energy, Plug Power, Green Hydrogen Systems, Nel Hydrogen, Next Hydrogen. Les technologies de production et d’utilisation d’hydrogène semblent être le matériel pour la prochaine vague de changement technologique en ce qui concerne l’énergie. Encore, il y a hydrogène et hydrogène. Le business de production d’hydrogène et de sa fourniture à travers des stations de ravitaillement c’est la technologie d’électrolyse et de stockage des gaz volatiles, donc quelque chose de pas vraiment révolutionnaire. Il y a espace pour innovation, certes, mais c’est de l’innovation incrémentale, rien qui brise les murs de l’ignorance pour ainsi dire. En revanche, l’utilisation d’hydrogène dans les piles à combustible, ça, c’est une technologie de pointe.

Dans ces deux cas de développement de technologie des piles à combustible (donc piles à hydrogène), soit Fuel Cell Energy et Plug Power, je passe en revue leur bilans et je les compare avec les autres trois sociétés, orientées plus spécifiquement sur l’électrolyse et le ravitaillement en hydrogène. Je m’arrête à leurs passifs. Quatre trucs m’intéressent plus particulièrement : est-ce qu’ils ont un capital social positif, la proportion « dette – capital social », la structure dudit capital social et les pertes accumulées dans le bilan.

Comme ces 5 sociétés n’ont pas toutes publié leurs rapports du 2nd trimestre 2022, je compare leurs rapports annuels 2021.    

>> Fuel Cell Energy :  capital social de $642,4 millions, fait 79% des passifs du bilan ; la source principale du capital social est la prime d’émission de $1,9 milliards, ce qui permet de compenser un déficit accumulé de $1,266 milliards.

>> Plug Power : capital social de $4,6 milliards, soit à peu de choses près 78% des passifs et alimenté par une prime d’émission de $7,07 milliards qui compense un déficit accumulé de $2,4 milliards.

>> Green Hydrogen Systems ; avec ceux-là, je commence à convertir les monnaies ; Green Hydrogen Systems est une société danoise et ils rapportent en couronnes danoises, soit 1 DKK = 0,14 USD ; le capital social ici est de $164,06 millions, fait 90% des passifs, vient surtout d’une prime d’émission de $243,71 millions et contient un déficit accumulé de $96,87 millions.   

>> Nel Hydrogen ; cette fois, c’est la Norvège et les couronnes norvégiennes à 1 NOK = 0,1 USD ; le capital social monte à $503,87 millions ce qui donne 84% des passifs et se base sur une prime d’émission de $559,62 millions et compense avec surplus un déficit accumulé de $97,16 millions.

>> Next Hydrogen (précédemment BioHEP Technologies Ltd.) ;  cette fois, ce sont les dollars canadiens – à 1 CAD = 0,77 USD – et les dollars canadiens propriétaires de Next Hydrogen font un capital social de $29,1 millions qui, à son tour, fait 78,6% des passifs et – surprise – vient surtout du capital-actions pur et simple de $58,82 millions et contient un déficit accumulé de $32,24 millions.

Je commence à voir un schéma commun. Toutes les 5 sociétés ont un modèle d’entreprise très propriétaire, basé sur le capital social beaucoup plus que sur la dette. Cela veut dire Peu de levier financier et beaucoup de souveraineté stratégique. Dans les quatre cas sur cinq, donc avec l’exception de Next Hydrogen, cette structure propriétaire est basée sur une combine avec les prix d’émission des actions. On émet les actions à un prix d’appel follement élevé par rapport au prix comptable basé sur les actifs. Seuls les initiés savent pourquoi c’est tellement cher et ils payent, pendant que le commun des mortels est découragé par cette prime d’émission gigantesque. Tout en entrant en Bourse, les fondateurs de la société restent maîtres du bilan et donnent à leurs participations une liquidité élégante, propre au marché financier public.

Dans le cinquième cas, donc avec Next Hydrogen, c’est plus transparent et moins tordu : c’est le capital-actions qui pompe le capital social et ça semble donc plus ouvert aux actionnaires autres que les fondateurs.

Dans tous les cas, le capital social sert à compenser un déficit accumulé de taille très importante et en même temps sert à créer un coussin de liquide sur le côté actif du bilan. Les actifs autres que l’argent liquide et ses équivalents sont donc largement financés avec de la dette.

Prendre contrôle propriétaire d’une entreprise profondément déficitaire indique une détermination stratégique. La question se pose donc, c’est une détermination à faire quoi au juste ? Je rétrécis mon champ d’analyse à Fuel Cell Energy et je commence à passer en revue leurs rapports courants. Le Rapport courant du 12 Juillet 2022 informe que Fuel Cell Energy est entrée en contrat de vente sur marché ouvert (anglais : « Open Market Sales Agreement ») avec Jefferies LLC, B. Riley Securities, Inc., Barclays Capital Inc., BMO Capital Markets Corp., BofA Securities, Inc., Canaccord Genuity LLC, Citigroup Global Markets Inc., J.P. Morgan Securities LLC and Loop Capital Markets LLC dont chacun est designé comme Agent et tous ensemble sont des « Agents ». Le contrat donne à Fuel Cell Energy la possibilité d’offrir et de vendre, de temps en temps, un paquet de 95 000 000 actions (contre les 837 500 000 actions déjà actives) à valeur nominale de $0,0001 par action (soit la même valeur nominale que les actions déjà en place). Ces offres occasionnelles de 95 000 000 actions peuvent se faire aussi bien à travers les Agents qu’aux Agents eux-mêmes. Cette dualité « à travers ou bien à » se traduit en une procédure de préemption, ou Fuel Cell Energy offre les actions à chaque Agent et celui-ci a le choix de d’accepter et donc d’acheter les actions, ou bien de décliner l’offre d’achat et d’agir comme intermédiaire dans leur vente aux tierces personnes. Fuel Cell Energy paiera à l’Agent une commission de 2% sur la valeur brute de chaque transaction, que ce soit l’achat direct par l’Agent ou bien son intermédiaire dans la transaction. Par le même contrat, Jefferies LLC et Barclays Capital Inc. ont convenu avec Fuel Cell Energy de mettre fin à un contrat similaire, signé entre les trois parties en juin 2021.

Intéressant. Fuel Cell Energy entreprend d’utiliser son capital social comme plateforme de coopération avec une sorte de club d’institutions financières. Ces paquets de 95 000 000 actions à valeur nominale de $0,0001 par action font nominalement $9500 chacun, soit à peu près les dépenses voyage demi-mensuelles d’un PDG dans les organisations signataires du contrat. Pas vraiment de quoi déstabiliser un business. La commission de 2% sur un tel paquet fait $190. Seulement, l’émission publique de 837 000 000 actions existantes de Fuel Cell Energy s’était soldée par une prime d’émission de 5 157 930%. Oui, une prime d’émission de plus de 5 millions de pourcent. Ça fait beaucoup de points de pourcentage. Le moment quand j’écris ces mots, le prix boursier d’une action de Fuel Cell Energy est de $4,15 (soit 4149900% de plus que la valeur comptable). Par ailleurs, le volume d’actions en circulation est de 19 722 305, qui fait un free float d’à peine 19 722 305 / 837 500 000 = 2,35%. Chacun de ces paquets de 95 000 000 actions convenus par le contrat en question fait plus que ça et il peut donner occasion à une prime d’émission de plus de $394 millions et une commission de presque $8 millions.

Je n’aime pas ça. Comme investisseur, j’ai toutes me lampes rouges qui clignotent lorsque je pense à investir dans Fuel Cell Energy. Ce contrat du 12 juillet 2022, c’est carrément du poker financier. Je sais par expérience que le poker, c’est divertissant, mais ça ne va pas de pair avec une stratégie d’investissement rationnelle. Il me vient à l’esprit ce principe de gestion qui dit que lorsque les gestionnaires d’une société ont trop de liquide inutilisé à leur disposition, ils commencent à faire des trucs vraiment bêtes.

The real deal

I am blogging again, after months of break. My health required some attention, and my life priorities went a bit wobbly for some time, possibly because of the opioid pain killers which I took in hospital, after my surgery. Anyway, I am back in the game, writing freestyle.

Restarting after such a long break is a bit hard, and yet rewarding. I am removing rust from my thoughts, as if I were giving a new life to an old contrivance. I need to work up to cruise speed in my blogging. Currently, I am working on two subjects. One is my concept of Energy Ponds: a solution which combines ram pumps, hydropower, and the retention of water in wetlands. The other one pertains to business models in the broadly spoken industry of new sources of energy: electric vehicles (I am and remain a faithful investor in Tesla), technologies of energy storage, hydrogen and fuel cells based thereon, photovoltaic, wind and nuclear.

As I am thinking about it, the concept of Energy Ponds is already quite structured, and I am working on structuring it further by attracting the attention of people with knowledge and skills complementary to mine. On the other hand, the whole business models thing is foggy theoretically, and, at the same time, it is important to me at many levels, practical strategies of investment included. I know by experience that such topics – both vague and important – are the best for writing about on my blog.

Here comes the list of companies which I observe more or less regularly with respect to their business models:

>> Tesla

>> Rivian

>> Lucid Group

>> Nuscale Power 

>> First Solar

>> SolarEdge

>> Fuel Cell Energy

>> Plug Power

>> Green Hydrogen Systems

>> Nel Hydrogen

>> Next Hydrogen (précédemment BioHEP Technologies Ltd.)

>> Energa

>> PGE

>> Tauron

>> ZPUE   

Two classifications come to my mind as I go through that list. Firstly, there are companies which I currently hold an investment position in: Tesla, Nuscale Power, Energa, PGE, Tauron et ZPUE. Then come those which I used to flirt with, namely Lucid Group, First Solar and SolarEdge. Finally, there are businesses which I just keep watching from a distance: Rivian, Fuel Cell Energy, Plug Power, Green Hydrogen Systems, Nel Hydrogen, and Next Hydrogen.

The other classification is based on the concept of owners’ earnings such as defined by Warren Buffett: net income plus amortization minus capital expenses. Tesla, PGE, Energa, ZPUE, Tauron, First Solar, SolarEdge – these guys generate a substantial stream of owners’ earnings. The others are cash-negative. As for the concept of owners’ earnings itself, you can consult both the investor-relations site of Berkshire Hathaway (  ) or read a really good book by Robert G.Hagstrom « The Warren Buffett Way » (John Wiley & Sons, 2013, ISBN 1118793994, 9781118793992). I guess the intuition behind hinging my distinctions upon the cash-flow side of the house assumes that in the times of uncertainty, cash is king. Rapid technological change is full of uncertainty, especially when that change affects whole infrastructures, as it is the case with energy and propulsion. Besides, I definitely buy into Warren Buffett’s claim that cash-flow is symptomatic of the lifecycle in the given business.

The development of a business, especially on the base of innovative technologies, is cash-consuming. Cash, in business, is something we harvest rather than simply earn. Businesses which are truly able to harvest cash from their operations, have internal financing for moving to the next cycle of technological change. Those in need of cash from outside will need even more cash from outside in order to finance further innovation.

What’s so special about, cash in a business model? The most intuitive answer that comes to my mind is a motto heard from a banker, years ago: “In the times of crisis, cash is king”. Being a king means sovereignty in a territory, like “This place is mine, and, with all the due respect, pay respect or f**k off”. Having cash means having sovereignty of decision in business. Yet, nuance is welcome. Cash is cash. Once you have it, it does not matter that much where it came from, i.e. from operations or from external sources. When I have another look at businesses without positive owners’ earnings – Nuscale Power, Rivian, Fuel Cell Energy, Plug Power, Green Hydrogen Systems, Nel Hydrogen, and Next Hydrogen – I shift my focus from their cash-flow statements to their balance sheets and I can see insane amounts of cash on the assets’ side of the house. These companies, in their assets, have more cash than they have anything else. They look almost like banks, or investment funds.

Thus, my distinction between business models with positive owners’ earnings, on the one hand, and those without it, on the other hand, is a distinction along the axis of strategic specificity. When the sum total of net income and amortization, reduced by capital expenses, is positive and somehow in line with the market capitalization of the whole company, that company is launched on some clear tracks. The business is like a river: it is predictable and clearly traceable in its strategic decisions. On the other hand, a business with lots of cash in the balance sheet but little cash generated from operations is like lord Byron (George Gordon): those guys assume that the only two things worth doing are poetry and cavalry, only they haven’t decided yet the exact mix thereof.      

That path of thinking implies that a business model is more than a way of conducting operations; it is a vehicle for change through investment, thus for channeling capital with strategic decisions. Change which is about to come is somehow more interesting than change which is already there. Seen under this angle, businesses on my list convey different degrees of vagueness, and, therefore, different doses of intellectual provocation. I focus on the hydrogen ones, probably because in my country, Poland, we have that investment program implemented by the government: the hydrogen valleys.

As I have another look at the hydrogen-oriented companies on my list – Fuel Cell Energy, Plug Power, Green Hydrogen Systems, Nel Hydrogen, and Next Hydrogen – an interesting discrepancy emerges as regards the degree of technological advancement. Green Hydrogen Systems, Nel Hydrogen, and Next Hydrogen are essentially focused on making and supplying hydrogen. This is good old electrolysis, a technology with something like a century of industrial tradition, combined with the storage and transport of highly volatile gases. Only two, namely Fuel Cell Energy and Plug Power, are engaged into fuel cells based on hydrogen, and those fuel cells are, in my subjective view, the real deal as it comes to hydrogen-related innovation.

Ugly little cherubs

I am working on my long-term investment strategy, and I keep using the Warren Buffet’s tenets of investment (Hagstrom, Robert G.. The Warren Buffett Way (p. 98). Wiley. Kindle Edition.).

At the same time, one of my strategic goals is coming true, progressively: other people reach out to me and ask whether I would agree to advise them on their investment in the stock market. People see my results, sometimes I talk to them about my investment philosophy, and it seems to catch on.

This is both a blessing and a challenge. My dream, 2 years ago, when I was coming back to the business of regular investing in the stock market, was to create, with time, something like a small investment fund specialized in funding highly innovative, promising start-ups. It looks like that dream is progressively becoming reality. Reality requires realistic and intelligible strategies. I need to phrase out my own experience as regards investment in a manner, which is both understandable and convincing to other people.

As I am thinking about it, I want to articulate my strategy along three logical paths. Firstly, what is the logic in my current portfolio? Why am I holding the investment positions I am holding? Why in these proportions? How have I come to have that particular portfolio? If I can verbally explain the process of my so-far investment, I will know what kind of strategy I have been following up to now. This is the first step, and the next one is to formulate a strategy for the future. In one of my recent updates (Tesla first in line), I briefly introduced my portfolio, such as it was on December 2nd, 2021. Since then, I did some thinking, most of all in reference to the investment philosophy of Warren Buffett, and I made some moves. I came to the conclusion that my portfolio was astride a bit too many stocks, and the whole was somehow baroque. By ‘baroque’ I mean that type of structure, where we can have a horribly ugly little cherub, accompanied by just as ugly a little shepherd, but the whole looks nice due to the presence of a massive golden rim, woven around ugliness.

I made myself an idea of what are the ugly cherubs in my portfolio from December 2nd, and I kicked them out of the picture. In the list below, these entities are marked in slashed bold italic:

>> Tesla (,

>> SA ( ),

>> Alten ( ),

>> Altimmune Inc ( ),

>> Apple Inc ( ),

>> CureVac NV ( ),

>> Deepmatter Group PLC ( ), 

>> FedEx Corp ( ),

>> First Solar Inc ( )

>> Inpost SA ( )

>> Intellia Therapeutics Inc ( )

>> Lucid Group Inc ( )

>> Mercator Medical SA ( )

>> Nucor Corp ( )

>> Oncolytics Biotech Inc ( )

>> Solaredge Technologies Inc ( )

>> Soligenix Inc ( )

>> Vitalhub Corp ( )

>> Whirlpool Corp ( )

>> Biogened ( )

>> Biomaxima ( )

>> CyfrPolsat ( )

>> Emtasia ( )

>> Forposta ( )

>> Gameops ( )

>> HMInvest ( )

>> Ifirma ( )

>> Moderncom ( )

>> PolimexMS ( )

>> Selvita ( )

>> Swissmed ( )  

Why did I put those specific investment positions into the bag labelled ‘ugly little cherubs in the picture’? Here comes a cognitive clash between the investment philosophy I used to have before I started studying in depth that of Warren Buffet and of Berkshire Hathaway. Before, I was using the purely probabilistic approach, according to which the stock market is so unpredictable that my likelihood of failure, on any individual investment, is greater than the likelihood of success, and, therefore, the more I spread my portfolio between different stocks, the less exposed I am to the risk of a complete fuck-up. As I studied the investment philosophy of Warren Buffet, I had great behavioural insights as regards my decisions. Diversifying one’s portfolio is cool, yet it can lead to careless individual choices. If my portfolio is really diversified, each individual position weighs so little that I am tempted to overlook its important features. At the end of the day, I might land with a bag full of potatoes instead of a chest full of gems.

I decided to kick out the superfluous. What did I put in this category? The superfluous investment positions which I kicked out shared some common characteristics, which I reconstructed from the history of the corresponding ‘buy’ orders. Firstly, these were comparatively small positions, hundreds of euros at best. This is one of the lessons by Warren Buffet. Small investments matter little, and they are probably going to stay this way. There is no point in collecting stocks which don’t matter to me. They give is a false sense of security, which is detrimental to the focus on capital gains.  

Secondly, I realized that I bought those ugly little cherubs by affinity to something else, not for their own sake. Two of them, FedEx and Allegro, are in the busines of express delivery. I made a ton of money of their stock, just as on the stock of Deutsche Post, during the trough of the pandemic, where retail distribution went mostly into the ‘online order >> express delivery’ pipeline. It was back then, and then I sold out, and then I thought ‘why not trying the same hack again?’. The ‘why not…?’ question was easy to answer, actually: because times change, and the commodity markets have adapted to the pandemic. FedEx and Allegro has returned to what it used to be: a solid business without much charm to me.  

Four others – Soligenix, Altimmune, CureVac and Oncolytics Biotech – are biotechnological companies. Once again: I made a ton of money in 2020 on biotech companies, because of the pandemic. Now, emotions in the market have settled, and biotech companies are back what they used to be, namely interesting investments endowed with high risk, high potential reward, and a bottomless capacity for burning cash. Those companies are what Tesla used to be a decade ago. I kept a serious position on a few other biotech businesses: Intellia Therapeutics, Biogened, Biomaxima, and Selvita. I want to keep a few of such undug gems in my portfolio, yet too much would be too much.

Thirdly, I had a loss on all of those ugly little cherubs I have just kicked out of my portfolio. Summing up, these were small positions, casually opened without much strategic thinking, and they were bringing me a loss. I could have waited to have a profit, but I preferred to sell them out and to concentrate my capital on the really promising stocks, which I nailed down using the method of intrinsic value. I realized that my portfolio was what it was, one week ago, before I started strategizing consciously, because I had hard times finding balance between two different motivations: running away from the danger of massive loss, on the one hand, and focusing on investments with a true potential for bringing long-term gains.

I focus more specifically on the concept of intrinsic value. Such as Warren Buffet used it, intrinsic value was based on what he called ‘owner’s earnings’ from a business. Owner’s earnings are spread over a window in time corresponding to the risk-free yield on sovereign bonds. The financial statement used for calculating intrinsic value is the cash-flow of the company in question, plus external data as regards average annual yield on sovereign bonds. The basic formula to calculate owner’s earnings goes like: net income after tax + amortization charges – capital expenditures). Once that nailed down, I divide those owner’s earnings by the interest rate on long-term sovereign bonds. For my positions in the US stock market, I use the long-term yield on the US federal bonds, i.e. 1,35% a year. As regards my portfolio in the Polish stock market, I use the yield 3,42% for Polish sovereign bonds on long-term.

I have calculated that intrinsic value for a few of my investments (I mean those I kept in my portfolio), on the basis of their financial results for 2020 and compared it to their market capitalisation. Then, additionally, I did the same calculation based on their published (yet unaudited) cash-flow for Q3 2021. Here are the results I had for Tesla. Net income 2020 $862,00 mln plus amortization charges 2020 $2 322,00 mln minus capital expenditures 2020 $3 132,00 mln equals owner’s earnings 2020 $52,00 mln. Divided by 1,35%, that gives an intrinsic value of $3 851,85 mln. Market capitalization on December 6th, 2021: $1 019 000,00 mln. The intrinsic value looks like several orders of magnitude smaller than market capitalisation. Looks risky.

Let’s see the Q3 2021 unaudited cash-flows. Here, I extrapolate the numbers for 9 months of 2021 over the whole year 2021: I multiply them by 4/3. Extrapolated net income for Q3 2021 $4 401,33 mln plus extrapolated amortization charges for Q3 2021 $2 750,67 minus extrapolated capital expenditures for Q3 2021 $7 936,00 equals extrapolated owner’s earnings amounting to $4 401,33 mln. Divided by 1,35%, it gives an extrapolated intrinsic value of $326 024,69 mln. It is much closer to market capitalization, yet much below it as for now. A lot of risk in that biggest investment position of mine. We live and we learn, as they say.

Another stock: Apple. With the economic size of a medium-sized country, Apple seems solid. Let’s walk it through the computational path of intrinsic value. There is an important methodological remark to formulate as for this cat. In the cash-flow statement of Apple for 2020-2021 (Apple Inc. ends its fiscal year by the end of September in the calendar year), under the category of ‘Investing activities’, most of the business pertains to buying and selling financial assets. It goes, ike:

Investing activities, in millions of USD:

>> Purchases of marketable securities (109 558)

>> Proceeds from maturities of marketable securities: 59 023

>> Proceeds from sales of marketable securities: 47 460

>> Payments for acquisition of property, plant and equipment (11 085)

>> Payments made in connection with business acquisitions, net (33)

>> Purchases of non-marketable securities (131)

>> Proceeds from non-marketable securities: 387

>> Bottom line: Cash generated by/(used in) investing activities (14 545)

Now, when I look at the thing through the lens of Warren Buffett’s investment tenets, anything that happens with and through financial securities, is retention of cash in the business. It just depends on what exact form we want to keep that cash under. Transactions grouped under the heading of ‘Purchases of marketable securities (109 558)’, for example, are not capital expenditures. They do not lead to exchanging cash money against productive technology. In all that list of investment activities, only two categories, namely: ‘Payments for acquisition of property, plant and equipment (11 085)’, and ‘Payments made in connection with business acquisitions, net (33)’ are capital expenditures sensu stricto. All the other categories, although placed in the account of investing activities, are labelled as such just because they pertain to transactions on assets. From the Warren Buffet’s point of view they all mean retained cash.

Therefore, when I calculate owner’s earnings for Apple, based on their latest annual cash-flow, I go like:

>> Net Income $94 680 mln + Depreciation and Amortization $11 284 mln + Purchases of marketable securities $109 558 mln + Proceeds from maturities of marketable securities $59 023 mln + Proceeds from sales of marketable securities $47 460 mln – Payments for acquisition of property, plant and equipment $11 085 mln – Payments made in connection with business acquisitions, net $33 mln + Purchases of non-marketable securities $131 mln + Proceeds from non-marketable securities $387 mln = Owner’s earnings $311 405 mln.

I divide that number by the 1,35% annual yield of the long-term Treasury bonds in the US, and I get an intrinsic value of $23 067 037 mln, against a market capitalisation floating around $2 600 000 mln, which gives a huge overhead in the former over the latter. Good investment.

I pass to another one of my investments, First Solar Inc. ( ). Same thing: investment activities consist most of all in moves pertinent to financial assets. It looks like:

>> Net income (loss) $398,35 mln

>> Depreciation, amortization and accretion $232,93 mln

>> Impairments and net losses on disposal of long-lived assets $35,81 mln

… and then come the Cash flows from investing activities:

 >> Purchases of property, plant and equipment ($416,64 mln)

>> Purchases of marketable securities and restricted marketable securities ($901,92 mln)

>> Proceeds from sales and maturities of marketable securities and restricted marketable securities $1 192,83 mln

>> Other investing activities ($5,5 mln)

… and therefore, from the perspective of owner’s earnings, the net cash used in investing activities is not, as stated officially, minus $131,23 mln. Net capital expenses, I mean net of transactions on financial assets, are: – $416,64 mln + $901,92 mln + $1 192,83 mln – $5,5 mln = $1 672,61 mln. Combined with the aforementioned net income, amortization and fiscally compensated impairments on long-lived assets, it makes owner’s earnings of $2 339,7 mln. And an intrinsic value of $173 311,11 mln, against some $10 450 000 mln in market capitalization. Once again, good and solid in terms of Warren Buffet’s margin of security.

I start using the method of intrinsic value for my investments, and it gives interesting results. It allows me to distinguish, with a precise gauge, between high-risk investments and the low-risk ones.

Je dois faire gaffe à la valeur intrinsèque

Me revoilà sur mon blog et je me concentre sur un truc : ma stratégie d’investissements boursiers. Je veux optimiser ma stratégie et à cette fin je me réfère à Warren Buffett et à sa philosophie d’investissement telle que vous et moi pouvons la trouver dans les rapports annuels de Berkshire Hathaway Inc. ( ). Je prends donc les principes de Warren Buffett et je les applique comparativement à deux compagnies : Tesla ( ), la plus grande position dans mon portefeuille d’investissement, d’une part, et Selvita ( ), une société polonaise de biotechnologie, sur les actions de laquelle je commence à développer un investissement sérieux.   

Avec Tesla, j’ai déjà entamé une analyse façon Warren Buffett (Tesla first in line) et maintenant je continue de manière comparative. C’est un truc qui marche : lorsque je veux comprendre quelque chose de complexe, je peux comparer cette chose complexe avec une autre chose complexe. Comparaison est une stratégie cognitive fondamentale. Elle me permet d’apercevoir les différences et les similarités entre des phénomènes complexes (tout est complexe, en fait) et de comprendre ainsi ce que la science cognitive désigne comme « saillance ».

Le genre de saillance sur laquelle je me concentre maintenant sont les traits distinctifs (et donc saillants et importants) de ces deux sociétés – Tesla ( ) et Selvita ( ) – en ce qui concerne les piliers conceptuels de la stratégie de Warren Buffett, qui sont :

>> l’entreprise en tant que telle : le modèle d’entreprise est-il simple et compréhensible ? l’entreprise a-t-elle une histoire cohérente d’exploitation ainsi que des perspectives favorables à long terme ?

>> la gestion : est-ce que la gestion de l’entreprise semble rationnelle ? Les gestionnaires semblent-ils agir dans le meilleur intérêt des actionnaires ? Les gestionnaires résistent-ils les

 modes et les pressions institutionnelles externes ?

>> la finance : quel est le retour sur capitaux propres ? quels sont les bénéfices agrégés pour les actionnaires ? Quelle est la marge de bénéfice dans les produits de l’entreprise ? Quelle est la concordance entre rétention de trésorerie d’une part et l’accroissement de valeur boursière ?   

>> le marché boursier : quelle est la valeur économique de la société en question ? comment cela concorde-t-il avec sa valeur boursière ?

Je me concentre sur la concordance entre la valeur économique de, respectivement, Tesla et Selvita, d’une part et leur valeur boursière d’autre part. Dans la stratégie modèle de Warren Buffett, la valeur économique d’une entreprise est égale au flux prévisible de trésorerie d’activités d’exploitation, escompté avec un taux de retour sur investissement sans risque. Pour donner un exemple pratiqe de cette méthode de base, je cite et traduis un passage du livre « The Warren Buffett Way » par Robert G. Hagstrom, plus précisément le fragment des pages 136 – 137 de l’édition Kindle, où la méthode de Buffett est démontrée dans son achat de Washington Post en 1973. Alors : « Nous commençons par calculer les revenus propriétaires pour l’année fiscale : bénéfice net de $13,3 millions plus dépréciation et amortissement de $3,7 millions moins les investissements capitalisables de $6,6 millions, ça donne un revenu propriétaire de $10,4 millions. Si nous divisons ce revenu par le taux de rendement des obligations souveraines long-terme de la Trésorerie Fédérale des États-Unis (6,81%), la valeur de Washington Post atteint $150 millions […]. Buffett dit qu’avec le temps, les investissements capitalisables d’un journal vont être égales au flux d’amortissement et de ce fait le bénéfice net devrait être une bonne estimation du revenu propriétaire. »

J’applique le même raisonnement à mes deux cas particuliers, donc Tesla et Selvita. Je vais chez et je sélectionne le rapport annuel pour 2020, soit Je vais droit au rapport des flux de trésorerie .

Le bénéfice net de Tesla pour 2020 était de $862 millions, la charge d’amortissement montait à $2 322 millions, et le solde d’investissements capitalisables fût $3 132. J’obtiens un revenu propriétaire de $52 millions.  J’utilise deux taux de rendement comme référence : celui des obligations souveraines du Trésor Polonais, soit 3,242% (puisque j’investis à partir de Pologne), ainsi que celui des obligations souveraines long-terme de la Trésorerie Fédérale des États-Unis (1,35%), puisque mon résultat sur Tesla est déterminé par la valeur intrinsèque de Tesla telle qu’estimée par le marché financier pour Tesla se trouve aux États-Unis.

Après avoir divisé le revenu propriétaire de Tesla pour 2020 par ces deux taux alternatifs, j’obtiens une fourchette de valeur intrinsèque entre $1 603,95 milliards et $3 851,85 milliards. La capitalisation boursière de Tesla est couramment de $1 019 milliards, mais tout récemment, le 4 novembre, elle atteignait $1 248,43 milliards.

Je répète le même exercice – donc basé sur les résultats financiers pour l’année 2020 – pour Selvita. Je prends leur rapport financier annuel 2020 ( ) et sur la page 8 je trouve le rapport des flux de trésorerie. Le bénéfice net était de PLN19 921 919, avec les charges d’amortissement de PLN13 525 722.

En ce qui concerne les d’investissements capitalisables, ça se corse. Je trouve un investissement en actifs matériels et immatériels de valeur totale de PLN15 003 636, ainsi que « l’acquisition d’autres actifs financiers » qui monte à PLN10 152 560. Selon la logique de Warren Buffett, l’investissement strictement dit, donc celui qui est déductible du revenu propriétaire, est celui en actifs productifs afférents à l’exploitation. L’acquisition d’actifs financiers est un placement, pas un investissement en actifs d’exploitation.

Je pense donc que je peux calculer le revenu propriétaire de Selvita de deux façons différentes. La première variante c’est « bénéfice net plus amortissement moins l’investissement en actifs matériels et immatériels » et dans la deuxième variante je considère l’acquisition d’actifs financiers comme un flux additionnel de trésorerie et je l’ajoute au solde du premier calcul. J’obtiens ainsi un revenu propriétaire façon Warren Buffett dans la fourchette entre PLN18 444 005 et PLN28 596 565. Je divise par le taux de rendement comme des obligations souveraines du Trésor Polonais (3,242%) et j’obtiens une fourchette correspondante de valeur intrinsèque entre PLN568 908 235 et PLN882 065 545. La dernière capitalisation boursière de Selvita est de PLN1 478 millions, avec un maximum sur les 12 mois derniers noté le 5 juillet 2021, égal à PLN2 894,7 millions.

Ma conclusion provisoire est que, sur la base des résultats financiers audités pour 2020, Tesla reste sous-valorisée par le marché boursier et ça donne des opportunités intéressantes. En revanche, Selvita semble être un peu gonflée en Bourse et je dois être sur mes gardes. Maintenant je passe à l’extension de l’exercice précèdent avec la méthode de MRQ ou « Most Recent Quarter », soit avec les résultats financiers non-audités de deux sociétés pour le troisième quart de 2021. Je fais un truc très primitif, qui est néanmoins utilisé fréquemment en analyse financière, donc j’extrapole les résultats de trois quarts de l’année fiscale en les multipliant par « 4/3 ». Oui, c’est simpliste et ça donne juste une estimation très provisoire de ce que les résultats annuels audités pour 2021 peuvent bien être. Néanmoins, cette méthode permet de simuler l’état d’esprit d’autres investisseurs qui – tout comme moi – utilisent la méthode de valeur intrinsèque façon Warren Buffett.

 Je commence par Tesla, encore une fois  ( . Bénéfice net $3 301 millions plus charge d’amortissement $2 063 moins investissements capitalisables de $5 952, ça donne… merde… – $588. Embarrassant, n’est-ce pas ? Revenu propriétaire négatif veut dire valeur intrinsèque négative. Esquive élégante : « Buffett dit qu’avec le temps, les investissements capitalisables d’un journal vont être égales au flux d’amortissement et de ce fait le bénéfice net devrait être une bonne estimation du revenu propriétaire. » Bon, Tesla, c’est presque comme un journal, quoi. Sauf que ça n’a rien à voir. Ce n’est même pas le même type fondamental de bien économique. Enfin, essayons avec l’équivalence « bénéfice net = revenu propriétaire ». J’extrapole le bénéfice net pour les 9 mois de 2021 sur les 12 mois de l’année fiscale et ça donne $4 401,33 millions. Je divise par le taux de rendement des obligations souveraines long-terme de la Trésorerie Fédérale des États-Unis (1,35%) et j’ai $326 024 milliards.

Je commence à comprendre la danse folle autour des actions de Tesla. Vous regardez le bénéfice net et ça a l’air de décoiffer (positivement). Vous jetez un coup d’œil sur les dépenses capitalisables d’investissement et vous commencer à vous poser des questions. Si le calcul très simple de Warren Buffett donne autant de doute, pas étonnant que plusieurs petits investisseurs se laissent prendre au jeu des grands fonds d’investissement futés.

Je passe à Selvita : . Bénéfice net sur les 9 mois de 2021 fait PLN8 844 005, les charges d’amortissement sur la même période montent à PLN17 764 894, acquisition d’actifs productifs est de PLN9 655 884 et celle d’actifs financiers c’est PLN3 172 566. Je somme des deux façons alternatives décrites plus tôt, j’extrapole sur les 12 mois, je divise par le taux d’intérêt sur les obligations de Trésor Polonais (3,242%)  et j’obtiens une valeur intrinsèque entre PLN660 936 257 et PLN784 623 041. C’est toujours très en-dessous de la capitalisation boursière de Selvita. Je dois faire gaffe.

Tesla first in line

Once again, a big gap in my blogging. What do you want – it happens when the academic year kicks in. As it kicks in, I need to divide my attention between scientific research and writing, on the one hand, and my teaching on the other hand.

I feel like taking a few steps back, namely back to the roots of my observation. I observe two essential types of phenomena, as a scientist: technological change, and, contiguously to that, the emergence of previously unexpected states of reality. Well, I guess we all observe the latter, we just sometimes don’t pay attention. I narrow it down a bit. When it comes to technological change, I am just bewildered with the amounts of cash that businesses have started holding, across the board, amidst an accelerating technological race. Twenty years ago, any teacher of economics would tell their students: ‘Guys, cash is the least productive asset of all. Keep just the sufficient cash to face the most immediate expenses. All the rest, invest it in something that makes sense’. Today, when I talk to my students, I tell them: ‘Guys, with the crazy speed of technological change we are observing, cash is king, like really. The greater reserves of cash you hold, the more flexible you stay in your strategy’.

Those abnormally big amounts of cash that businesses tend to hold, those last years, it has two dimensions in terms of research. On the one hand, it is economics and finance, and yet, on the other hand, it is management. For quite some time, digital transformation has been about the only thing worth writing about in management science, but that, namely the crazy accumulation of cash balances in corporate balance sheets, is definitely something worth writing about. Still, there is amazingly little published research on the general topic of cash flow and cash management in business, just as there is very little on financial liquidity in business. The latter topic is developed almost exclusively in the context of banks, mostly the central ones. Maybe it is all that craze about the abominable capitalism and the general claim that money is evil. I don’t know.

Anyway, it is interesting. Money, when handled at the microeconomic level, tells the hell of a story about our behaviour, our values, our mutual trust, and our emotions. Money held in corporate balance sheets tells the hell of a story about decision making. I explain. Please, consider the amount of money you carry around with you, like the contents of your wallet (credit cards included) plus whatever you have available instantly on your phone. Done? Visualised? Good. Now, ask yourself what percentage of all those immediately available monetary balances you use during your one average day. Done? Analysed? Good. In my case, it would be like 0,5%. Yes, 0,5%. I did that intellectual exercise with my students, many time. They usually hit no more than 10%, and they are gobsmacked. Their first reaction is WOKEish: ‘So I don’t really need all that money, right. Money is pointless, right?’. Not quite, my dear students. You need all that money; you just need it in a way which you don’t immediately notice.

There is a model in the theory of complex systems, called the ants’ colony (see for example: (Chaouch, Driss & Ghedira 2017[1]; Asghari & Azadi 2017[2]; Emdadi et al. 2019[3]; Gupta & Srivastava 2020[4]; Di Caprio et al. 2021[5]). Yes, Di Caprio. Not the Di Caprio you intuitively think about, though. Ants communicate with pheromones. They drop pheromones somewhere they sort of know (how?) it is going to be a signal for other ants. Each ant drops sort of a standard parcel of pheromones. Nothing to write home about, really, and yet enough to attract the attention of another ant which could drop its individual pheromonal parcel in the same location. With any luck, other ants will discover those chemical traces and validate them with their individual dumps of pheromones, and this is how the colony of ants maps its territories, mostly to find and exploit sources of food. This is interesting to find out that in order for all that chemical dance to work, there needs to be a minimum number of ants on the job. In there are not enough ants per square meter of territory, they just don’t find each other’s chemical imprints and have no chance to grab hold of the resources available. Yes, they all die prematurely. Money in human societies could be the equivalent of a pheromone. We need to spread it in order to carry out complex systemic changes. Interestingly, each of us, humans, is essentially blind to those complex changes: we just cannot wrap our mind around quickly around the technical details of something apparently as simple as the manufacturing chain of a gardening rake (do you know where exactly and in what specific amounts all the ingredients of steel come from? I don’t).  

All that talk about money made me think about my investments in the stock market. I feel like doing things the Warren Buffet’s way: going to the periodical financial reports of each company in my portfolio, and just passing in review what they do and what they are up to. By the way, talking about Warren Buffet’s way, I recommend my readers to go to the source: go to first, and then to as well as to . For now, I focus on studying my own portfolio according to the so called “12 immutable tenets by Warren Buffet”, such as I allow myself to quote them:

>> Business Tenets: Is the business simple and understandable? Does the business have a consistent operating history? Does the business have favourable long-term prospects?

>> Management Tenets: Is management rational? Is management candid with its shareholders? Does management resist the institutional imperative?

>> Financial Tenets Focus on return on equity, not earnings per share. Calculate “owner earnings.” Look for companies with high profit margins. For every dollar retained, make sure the company has created at least one dollar of market value.

>> Market Tenets: What is the value of the business? Can the business be purchased at a significant discount to its value?

(Hagstrom, Robert G.. The Warren Buffett Way (p. 98). Wiley. Kindle Edition.)

Anyway, here is my current portfolio:

>> Tesla (,

>> SA ( ),

>> Alten ( ),

>> Altimmune Inc ( ),

>> Apple Inc ( ),

>> CureVac NV ( ),

>> Deepmatter Group PLC ( ), 

>> FedEx Corp ( ),

>> First Solar Inc ( )

>> Inpost SA ( )

>> Intellia Therapeutics Inc ( )

>> Lucid Group Inc ( )

>> Mercator Medical SA ( )

>> Nucor Corp ( )

>> Oncolytics Biotech Inc ( )

>> Solaredge Technologies Inc ( )

>> Soligenix Inc ( )

>> Vitalhub Corp ( )

>> Whirlpool Corp ( )

>> Biogened ( )

>> Biomaxima ( )

>> CyfrPolsat ( )

>> Emtasia ( )

>> Forposta ( )

>> Gameops ( )

>> HMInvest ( )

>> Ifirma ( )

>> Moderncom ( )

>> PolimexMS ( )

>> Selvita ( )

>> Swissmed ( )   

Studying that whole portfolio of mine through the lens of Warren Buffet’s tenets looks like a piece of work, really. Good. I like working. Besides, as I have been reading Warren Buffett’s annual reports at , I realized that I need a real strategy for investment. So far, I have developed a few efficient hacks, such as, for example, the habit of keeping my s**t together when other people panic or when they get euphoric. Still, hacks are not the same as strategy.

I feel like adding my own general principles to Warren Buffet’s tenets. Principle #1: whatever I think I do my essential strategy consists in running away from what I perceive as danger. Thus, what am I afraid of, in my investment? What subjective fears and objective risks factors shape my actions as investor? Once I understand that, I will know more about my own actions and decisions. Principle #2: the best strategy I can think of is a game with nature, where each move serves to learn something new about the rules of the game, and each move should be both decisive and leaving me with a margin of safety. What am I learning as I make my moves? What my typical moves actually are?

Let’s rock. Tesla (, comes first in line, as it is the biggest single asset in my portfolio. I start my digging with their quarterly financial report for Q3 2021 ( ), and I fish out their Consolidated Balance Sheets (in millions, except per share data, unaudited: ).

Now, I assume that if I can understand why and how numbers change in the financial statements of a business, I can understand the business itself. The first change I can spot in that balance sheet is property, plant and equipment, net passing from $12 747 million to $17 298 million in 12 months. What exactly has happened? Here comes Note 7 – Property, Plant and Equipment, Net, in that quarterly report, and it starts with a specification of fixed assets comprised in that category. Good. What really increased in this category of assets is construction in progress, and here comes the descriptive explanation pertinent thereto: “Construction in progress is primarily comprised of construction of Gigafactory Berlin and Gigafactory Texas, expansion of Gigafactory Shanghai and equipment and tooling related to the manufacturing of our products. We are currently constructing Gigafactory Berlin under conditional permits in anticipation of being granted final permits. Completed assets are transferred to their respective asset classes, and depreciation begins when an asset is ready for its intended use. Interest on outstanding debt is capitalized during periods of significant capital asset construction and amortized over the useful lives of the related assets. During the three and nine months ended September 30, 2021, we capitalized $14 million and $52 million, respectively, of interest. During the three and nine months ended September 30, 2020, we capitalized $13 million and $33 million, respectively, of interest.

Depreciation expense during the three and nine months ended September 30, 2021 was $495 million and $1.38 billion, respectively. Depreciation expense during the three and nine months ended September 30, 2020 was $403 million and $1.13 billion, respectively. Gross property, plant and equipment under finance leases as of September 30, 2021 and December 31, 2020 was $2.60 billion and $2.28 billion, respectively, with accumulated depreciation of $1.11 billion and $816 million, respectively.

Panasonic has partnered with us on Gigafactory Nevada with investments in the production equipment that it uses to manufacture and supply us with battery cells. Under our arrangement with Panasonic, we plan to purchase the full output from their production equipment at negotiated prices. As the terms of the arrangement convey a finance lease under ASC 842, Leases, we account for their production equipment as leased assets when production commences. We account for each lease and any non-lease components associated with that lease as a single lease component for all asset classes, except production equipment classes embedded in supply agreements. This results in us recording the cost of their production equipment within Property, plant and equipment, net, on the consolidated balance sheets with a corresponding liability recorded to debt and finance leases. Depreciation on Panasonic production equipment is computed using the units-of-production method whereby capitalized costs are amortized over the total estimated productive life of the respective assets. As of September 30, 2021 and December 31, 2020, we had cumulatively capitalized costs of $1.89 billion and $1.77 billion, respectively, on the consolidated balance sheets in relation to the production equipment under our Panasonic arrangement.”

Good. I can try to wrap my mind around the contents of Note 7. Tesla is expanding its manufacturing base, including a Gigafactory in my beloved Europe. Expansion of the manufacturing capacity means significant, quantitative growth of the business. According to Warren Buffett’s philosophy: “The question of where to allocate earnings is linked to where that company is in its life cycle. As a company moves through its economic life cycle, its growth rates, sales, earnings, and cash flows change dramatically. In the development stage, a company loses money as it develops products and establishes markets. During the next stage, rapid growth, the company is profitable but growing so fast that it cannot support the growth; often it must not only retain all of its earnings but also borrow money or issue equity to finance growth” (Hagstrom, Robert G.. The Warren Buffett Way (p. 104). Wiley. Kindle Edition).  Tesla looks like they are in the phase of rapid growth. They have finally nailed down how to generate profits (yes, they have!), and they are expanding capacity-wise. They are likely to retain earnings and to be in need of cash, and that attracts my attention to another passage in Note 7: “Interest on outstanding debt is capitalized during periods of significant capital asset construction and amortized over the useful lives of the related assets”. If I understand correctly, the financial strategy consists in not servicing (i.e. not paying the interest due on) outstanding debt when that borrowed money is really being used to finance the construction of productive assets, and starting to service that debt only after the corresponding asset starts working and paying its bills. That means, in turn, that lenders are being patient and confident with Tesla. They assume their unconditional claims on Tesla’s future cash flows (this is one of the possible ways to define outstanding debt) are secure.   

Good. Now, I am having a look at Tesla’s Consolidated Statements of Operations (in millions, except per share data, unaudited: ). It is time to have a look at Warren Buffett’s Business Tenets as regards Tesla. Is the business simple and understandable? Yes, I think I can understand it. Does the business have a consistent operating history? No, operational results changed in 2020 and they keep changing. Tesla is passing from the stage of development (which took them a decade) to the stage of rapid growth. Does the business have favourable long-term prospects? Yes, they seem to have good prospects. The market of electric vehicles is booming (EV-Volumes[6]; IEA[7]).

Is Tesla’s management rational? Well, that’s another ball game. To develop in my next update.

[1] Chaouch, I., Driss, O. B., & Ghedira, K. (2017). A modified ant colony optimization algorithm for the distributed job shop scheduling problem. Procedia computer science, 112, 296-305.

[2] Asghari, S., & Azadi, K. (2017). A reliable path between target users and clients in social networks using an inverted ant colony optimization algorithm. Karbala International Journal of Modern Science, 3(3), 143-152.

[3] Emdadi, A., Moughari, F. A., Meybodi, F. Y., & Eslahchi, C. (2019). A novel algorithm for parameter estimation of Hidden Markov Model inspired by Ant Colony Optimization. Heliyon, 5(3), e01299.

[4] Gupta, A., & Srivastava, S. (2020). Comparative analysis of ant colony and particle swarm optimization algorithms for distance optimization. Procedia Computer Science, 173, 245-253.

[5] Di Caprio, D., Ebrahimnejad, A., Alrezaamiri, H., & Santos-Arteaga, F. J. (2021). A novel ant colony algorithm for solving shortest path problems with fuzzy arc weights. Alexandria Engineering Journal.



Laisse tomber les modèles d’entreprise

Une fois de plus, j’ai eu du décalage entre mes mises à jour. Remarquez, ça a été un décalage productif. J’ai développé un peu sur le sujet des voitures électriques. J’ai discuté le sujet avec quelques-uns de mes collègues, que j’avais invité à participer au projet et j’ai changé un peu d’optique. Le premier truc est abjectement opportuniste : la formulation de sujet. L’un de mes collègues a dit : « Écoute, le modèle d’entreprise, c’est occupé. Á l’École Supérieure de Commerce, à Varsovie, il y a une équipe qui a déjà défini plus de 100 modèles distincts d’entreprise et leur truc, ça a l’air solide côté méthode. Vaut mieux pas s’engager dans de territoire. Lorsqu’il sera question d’évaluation de notre demande de financement au ministère, ces gars-là, ils vont mettre nos têtes coupées sur leurs piques. Laisse tomber. Donnons à notre projet un sujet que presque personne ne veut toucher : l’innovation et la théorie de changement technologique dans le secteur des voitures électriques ». 

J’y a réfléchi et je suis bien d’accord. De ma part, toute cette histoire de modèles d’entreprise c’était un peu forcé. J’essayais d’emballer mes idées – qui par ailleurs sont 100% dans l’étude de changement technologique – dans le paquet étiqueté « modèle d’entreprise » parce que je pensais que ça se vendrait mieux. Je viens donc de (ré)apprendre une leçon importante : reste fidèle à tes idées et ton intuition. Prendre en considération les idées des autres est une chose, mais te déguiser en ces idées en est une autre et c’est une erreur stratégique.

Puisque je rejette le déguisement et je décide de rester fidèle à mes idées, je reviens à une idée dont il est même difficile de dire que je l’ai. C’est plutôt cette idée qui m’a et son essence est la suivante : les automobiles de toute sorte, à combustion interne ou électriques, peu importe, à part de mouvoir par eux-mêmes, ont un autre trait intéressant, celui d’être des réservoirs d’énergie par eux-mêmes. Une population d’automobiles est un réseau mobile de réservoirs d’énergie. Ceci dit, voilà quelques observations intéressantes.   

Premièrement, il y a de plus en plus d’automobiles nouvelles par chaque personne nouvelle née. Les nombres de base, je les montre dans le tableau ci-dessous. Le nombre d’automobiles produites annuellement vient de , pendant que la démographie est basée sur les données de la Banque Mondiale ( ). Plus loin, en-dessous du tableau numérique, je présente le même sujet graphiquement, en forme d’index 2010 = 1,00.

AnnéeAutomobiles produites globalementPopulation mondiale
1999               39 759 847       6 034 491 778 
2000               41 215 653       6 114 332 537 
2001               39 825 888       6 193 671 774 
2002               41 358 394       6 272 752 973 
2003               41 968 666       6 351 882 363 
2004               44 554 268       6 431 551 642 
2005               46 862 978       6 511 748 367 
2006               49 918 578       6 592 734 537 
2007               53 201 346       6 674 203 645 
2008               52 726 117       6 756 917 893 
2009               47 772 598       6 839 574 286 
2010               58 264 852       6 921 871 603 
2011               59 897 273       7 002 860 245 
2012               63 081 024       7 085 762 277 
2013               65 745 403       7 169 638 368 
2014               67 782 035       7 254 226 881 
2015               68 539 516       7 338 964 954 
2016               72 105 435       7 424 286 143 

Deuxièmement, il y a de plus en plus de puissance électrique installée par personne dans le monde ( ). C’est un fait qui apparemment a peu en commun avec les automobiles mais j’explique un peu plus loin. En tout cas, voilà les données numériques, ci-dessous. La troisième colonne, avec « Delta puissance / delta population » en tête est une valeur marginale. C’est la croissance annuelle de la puissance électrique installée divisée par la croissance annuelle analogue dans la population. Ça montre combien de kilowatts nouveaux de puissance électrique correspondent à chaque être humain nouveau sur la planète. Nous pouvons voir que ça bouge, ce quotient des deltas. Au fil des années, chaque homo sapiens reçoit plus de puissance électrique en cadeau. Enfin, pas tout à fait en cadeau : les parents de cet homo sapiens doivent payer pour l’énergie fournie par la puissance en question.     

AnnéePuissance électrique installée dans le monde (MW)Puissance électrique par tête d’habitant (kW)Delta puissance / delta population
19993 363 432,460,561,13
20003 472 904,930,571,37
20013 576 290,560,581,30
20023 754 233,590,602,25
20033 908 520,310,621,95
20044 043 439,790,631,69
20054 175 549,310,641,65
20064 367 561,020,662,37
20074 550 173,790,682,24
20084 748 664,030,702,40
20094 950 686,180,722,44
20105 182 394,400,752,82
20115 443 572,280,783,22
20125 664 728,480,802,67
20135 893 934,280,822,73
20146 170 181,660,853,27

Maintenant, j’interpose les bagnoles avec la puissance électrique et voilà : troisièmement, il y a de moins en moins d’automobiles nouvelles par chaque MW nouveau de puissance électrique installée. Si on inverse le raisonnement, il y a de plus en plus de puissance électrique nouvelle par chaque voiture nouvelle.

Nous avons donc la situation suivante : nos réserves des carburants fossiles – surtout calculés par tête d’habitant – se rétrécissent, pendant que la puissance électrique installée par tête d’habitant est en train de croître. Point de vue jus en réserve, il y a une transition vers l’électricité à partir de la combustion directe des carburants fossiles.

Je pose donc l’hypothèse que le développement des voitures électriques est une fonction émergente de notre civilisation et dans cette fonction émergente, la technologie de stockage d’énergie, dans les batteries aussi bien que dans les stations de chargement, est l’intermédiaire naturel entre la puissance électrique installée et le développement d’électromobilité.

The batteries we don’t need anymore

I continue on the thread I started to develop in my last update in French, titled ‘De quoi parler à la prochaine réunion de faculté’, i.e. I am using that blog, and the fact of writing, to put some order in the almost ritual mess that happens at the beginning of the academic year. New calls for tenders start in the ministerial grant programs, new syllabuses need to be prepared, new classes start. Ordinary stuff, mind you, this is just something about September, as if I were in Vivaldi’s ‘Four seasons’: the hot, tumultuous Summer slowly folds into the rich, textured, and yet implacably realistic Autumn.

My central idea is to use some of the science which I dove into during the summer holidays as an intellectual tool for putting order in that chaos. That almost new science of mine is mostly based on the theory of complex systems, and my basic claim is that technological change is an emergent phenomenon in complex social systems. We don’t know why exactly our technologies change the way they change. We can trace the current technologies back to their most immediate ancestors and sometimes we can predict their most immediate successors, but that’s about it. Futuristic visions of technologies that could be there in 50 years from now are already some kind of traditional entertainment. The concept of technological progress, when we try to find a developmental logic in the historically known technological change, is usually standing on wobbly legs, on the other hand. Yes, electricity allowed the emergence of medical technologies used in hospitals, and that saved a lot of human lives, but there is no way Thomas Edison could have known that. The most spectacular technological achievements of mankind, such as the Egyptian pyramids, the medieval cathedrals, the Dutch windmills from the 16th century, or the automobile, seen from the historical distance, look ambiguous. Yes, it all solved some problems, but it facilitated the emergence of new problems. The truly unequivocal benefit of those technological leaps, which could have been actually experienced by the people who made them, was to learn how to develop technologies.

The studies I did during the Summer holidays 2021 focused on four essential, mathematical models of emergent technological change: cellular automata, flock of birds AKA particle swarm, ants’ nest, and imperfect Markov chains. I start with passing in review the model of cellular automata. At any given moment, the social complexity can be divided into a finite number of social entities (agents). They can be individual humans, businesses, NGOs, governments, local markets etc. Each such entity has an immediate freedom of movement, i.e. a finite number of one-step moves. The concept is related to the theory of games and corresponds to what happens in real life. When we do something social, we seldom just rush forwards. Most frequently, we make one step, observe the outcomes, adjust, then we make the next step etc. When all social agents do it, the whole social complexity can be seen as a collection of cells, or pixels. Each such cell (pixel) is a local state of being in society. A social entity can move into that available state, or not, at their pleasure and leisure. All the one-step moves a social entity can make translate into a trajectory it can follow across the social space. Collective outcomes we strive for and achieve can be studied as temporary complex states of those entities following their respective trajectories. The epistemological trick here is that individual moves and their combinations can be known for sure only ex post. All we can do ex ante is to define the possible states, and just wait where does the reality go.

As we are talking about the possible states of social complexity, I found an interesting mathematical mindf**k at quite an unexpected source, namely in the book titled ‘Aware. The Science and Practice of Presence. The Groundbreaking Meditation Practice’ by Daniel J. Siegel [Penguin Random House LLC, 2018, Identifiers: LCCN 2018016987 (print), LCCN 2018027672 (ebook), ISBN 9780143111788, ISBN 9781101993040 (hardback)]. This is a mathematical way of thinking, apparently taken from quantum physics. Here is the essence of it. Everything that happens does so as 100% probability of the given thing happening. Each phenomenon which takes place is the actualization of the same phenomenon being just likely to happen.

Actualization of probability can be seen as collision of two vehicles in traffic. When the two vehicles are at a substantial distance from each other, the likelihood of them colliding is zero, for all practical purposes. As they converge towards each other, there comes a point when they become sort of provisionally entangled, e.g. they find themselves heading towards the same crossroads. The probability of collision increases slightly, and yet it is not even the probability of collision, it is just the probability that these two might find themselves in a vicinity conducive to a possible collision. Nothing to write home about, yet, like really. It can be seen as a plateau of probability slowly emerging out of the initial soup of all the things which can possibly happen.

As the two cars drive closer and closer to the crossroads in question, the panoply of possible states narrows down. There is a very clear chunk of reality which gains in likelihood, as if it was a mountain range pushing up from the provisional plateau. There comes a point where the two cars (and their drivers) just come on collision course, and there is no way around it, and this is a peak of 100% probability. Boom! Probability is being consumed.

What do those cars have in common with meditation and with the emergence of technological change? As regards meditation, thought can be viewed as a progressively emerging actualization of something that was just a weak probability, sort of a month ago it was just weakly probable that today I would think what I think, it became much more likely yesterday, as the thoughts from yesterday have an impact on the thoughts of today, and today it all comes to fruition, i.e. to the 100% probability. As regards emergent technological change, the way technology changes today can be viewed as actualization of something that was highly probable last year, just somehow probable 10 years ago, and had been just part of the amorphous soup of probability 30 years ago. Those trajectories followed by individual agents inside social complexity, as defined in the theory of cellular automata, are entangled together precisely according to that pattern of emergent probabilities. Two businesses coming up with two mutually independent, and yet similar technologies, are like two peak actualizations of 100% probability in a plateau of probable technological change, which, in turn, has been slowly emerging for some time.

Those other theories I use explain and allow to model mathematically that entanglement. The theory of particle swarm, pertinent to flocks of birds, assumes that autonomous social agents strive for a certain level of behavioural coupling. We expect some level of predictability from others, and we can cooperate with others when we are satisfactorily predictable in our actions. The strive for social coherence is, therefore, one mechanism of entanglement between individual trajectories of cellular automata. The theory of ants’ nest focuses on a specific category of communication systems in societies, working like pheromones. Ants organize by marking, reinforcing and following paths across their environment, and their pheromones serve as markers and reinforcement agents for those paths. In human societies, there are social pheromones. Money and financial markets make probably the most obvious example, but scientific publications are another one. The more scientific articles are being published on a given topic, the more likely are other articles being written on the same topic, until the whole thing reaches a point of saturation, when some ants (pardon me, scientists) start thinking about another path to mark with intellectual pheromones.

Cool. I have (OK, we have) complex social states, made of entangled probabilities that something specific happens, and they encompass technology. Those complex states change, i.e. one complex state morphs into another. Now, how the hell can I know, as a researcher, what is happening exactly? Such as the theory of complex systems suggests it, I can never know exactly, for one, and I need to observe, for two. As I don’t know exactly what is it exactly, that thing which I label ‘technological change’, it is problematic to set too many normative assumptions as for which specific path that technological change should take. I think this is the biggest point of contention as I apply my theory, such as I have just outlined it, to my main field of empirical research, namely energy economics, and technological change in the sector of energy. The more I do that research, the more convinced I am that the so-called ‘energy policies’, ‘climate policies’ etc. are politically driven bullshit based on wishful thinking, with not much of a chance to bring the positive change we expect. I have that deep feeling that setting a strategy for future innovations in our business/country/world is very much like that Polish expression ‘sharing the skin of a bear which is still running in the woods’. First, you need to kill the bear, only then you can bicker about who takes what part of the skin. In the case of innovation, long-term strategies in that domain consist in predicting what we will do when we have something we don’t even know yet what is it exactly.

I am trying to apply this general theory in the grant applications which I am in charge of preparing now, and in my teaching. We have that idea, at the faculty, to apply for funding to study the market of electric vehicles in Europe and in Poland. This is an interesting situation as regards business models. In the US, the market of electric cars is clearly divided among three categories of players. There is Tesla, which is a category and an industry in itself, with its peculiar strategy of extreme vertical integration. Then there are the big, classical car makers, such as Toyota, General Motors etc., with their business models based on rather a short vertical chain of value added inside the business, and a massive supply chain upstream of the house. Finally, there is a rising tide of small start-ups in the making of electric vehicles. I wonder what I could be in Europe. As our European market of electric vehicles is taking off, it is dominated by the incumbent big manufacturers, the old school ones, with Tesla building a factory in Germany, and progressively building a beachhead in the market. There is some timid movement towards small start-up businesses in the field, but it is really timid. In my home country, Poland, the most significant attempt at starting up an electric vehicle made in Poland is a big consortium of state-controlled companies, running under the name of ‘Electromobility Poland’.  

I have that intuition, which I provisionally express as a working hypothesis, namely that business models are an emergent property of technologies which they use. As regards the market of electric vehicles, it means that Tesla’s business model is not an accidental explosion of Elon Musk’s genius mind: it is an emergent characteristic of the technologies involved.

Good. I have some theory taking shape, nice and easy. I let it ripen a bit, and I start sniffing around for facts. What is a business model, in my mind? It is the way of operating the chain of value added, and getting paid for it, in the first place. Then, it is the way of using capital. I noticed that highly innovative environments force businesses to build up and keep large amounts of cash money, arguably to manage the diverse uncertainties emerging as technologies around morph like hell. In some cases, e.g. in biotech, the right business model for rapid innovation is a money-sucker, with apparently endless pay-ins of additional equity by the shareholders, and yet with a big value in terms of technological novelty created. I can associate that phenomenon of vacuum cleaning equity with the case of Tesla, who just recently started being profitable, and had gone through something like a decade in permanent operational loss. That is all pertinent to fixed costs, thus to the cash we need to build up and keep in place the organizational structure required for managing the value chain the way we want to manage it.

I am translating those loose remarks of mine into observable phenomena. Everything I have just mentioned is to be found in the annual financial reports. This is my first source of information. When I want to study business models in the market of electric vehicles, I need to look into financial and corporate reports of businesses active in the market. I need to look into the financial reports of Mercedes Benz, BMW, Renault, PSA, Volkswagen, Fiat, Volvo, and Opel – thus the European automotive makers – and see how it is going, and whether whatever is going on can be correlated with changes in the European market of electric vehicles. Then, it is useful to look into the financial reports of global players present in the European market, e.g. Tesla, Toyota, Honda and whatnot, just to see what changes in them as the European market of electric vehicles is changing.

If my intuition is correct, i.e. if business models are truly an emergent property of technologies used, the fact of engaging into the business of electric vehicles should be correlated with some sort of recurrent pattern in those companies.         

Good. This is about the big boys in the playground. Now, I turn toward the small ones, the start-up businesses. As I already said, it is not like we have a crowd of them in the European industry of electric vehicles. The intuitive axis of research which comes to my mind is to look at start-ups active in the U.S., study their business models, and see if there is any chance of something similar emerging in Europe. Somehow tangentially to that, I think it would be interesting to check whether the plan of Polish government regarding ‘Electromobility Poland’, that is the plan to develop it with public and semi-public money, and then sell it to private investors, has any grounds and under what conditions it can be a workable plan.

Good. I have rummaged a bit in my own mind, time to do the same to other people. I mean, I am passing to reviewing the literature. I type ‘electric vehicles Europe business model’ at the platform, and I look at what’s popping up. Here comes the paper by Pardo-Bosch, F., Pujadas, P., Morton, C., & Cervera, C. (2021). Sustainable deployment of an electric vehicle public charging infrastructure network from a city business model perspective. Sustainable Cities and Society, 71, 102957., . The abstract says: ‘The unprecedented growth of global cities together with increased population mobility and a heightened concern regarding climate change and energy independence have increased interest in electric vehicles (EVs) as one means to address these challenges. The development of a public charging infrastructure network is a key element for promoting EVs, and with them reducing greenhouse gas emissions attributable to the operation of conventional cars and improving the local environment through reductions in air pollution. This paper discusses the effectiveness, efficiency, and feasibility of city strategic plans for establishing a public charging infrastructure network to encourage the uptake and use of EVs. A holistic analysis based on the Value Creation Ecosystem (VCE) and the City Model Canvas (CMC) is used to visualise how such plans may offer public value with a long-term and sustainable approach. The charging infrastructure network implementation strategy of two major European cities, Nantes (France) and Hamburg (Germany), are analysed and the results indicate the need to involve a wide range of public and private stakeholders in the metropolitan areas. Additionally, relevant, and fundamental patterns and recommendations are provided, which may help other public managers effectively implement this service and scale-up its use and business model.

Well, I see there is a lot of work to do, as I read that abstract. I rarely find a paper where I have so much to argue with, just after having read the abstract. First of all, ‘the unprecedented growth of global cities’ thing. Actually, if you care to have a look at the World Bank data on urban land ( ), as well as that on urban population ( ), you will see that urbanization is an ambiguous phenomenon, strongly region-specific. The central thing is that cities become increasingly distinct from the countryside, as types of human settlements. The connection between electric vehicles and cities is partly clear, but just partly. Cities are the most obvious place to start with EVs, because of the relatively short distance to travel between charging points. Still, moving EVs outside the cities, and making them functional in rural areas, is the next big challenge.

Then comes the ‘The development of a public charging infrastructure network is a key element for promoting EVs’ part. As I studied the thing in Europe, the network of charging stations, as compared to the fleet of EVs in the streets is so dense that we have like 12 vehicles per charging station on average, across the European Union. There is no way a private investor can have it for their money, when financing a private charging station, with that average density. We face a paradox: there are so many publicly funded charging stations, in relation to the car fleet out there, that private investment gets discouraged. I agree that it could be an acceptable transitory state in the market, although it begs the question whether private charging stations are a viable business in Europe. Tesla has based a large part of its business model in the US precisely on the development of their own charging stations. Is it a viable solution in Europe?

Here comes another general remark, contingent to my hypothesis of business models being emergent on the basis of technologies. Automotive technologies in general, thus the technology of a vehicle moving by itself, regardless the method of propulsion (i.e. internal combustion vs electric) is a combination of two component technologies. Said method of propulsion is one of them, and the other one is the technology of distributing the power source across space. Electric vehicles can be viewed as cousins to tramways and electric trains, with just more pronounced a taste for independence: instead of drinking electricity from a permanent wiring, EVs carry their electricity around with them, in batteries.

As we talk about batteries, here comes another paper in my cursory rummaging across other people’s science: Albertsen, L., Richter, J. L., Peck, P., Dalhammar, C., & Plepys, A. (2021). Circular business models for electric vehicle lithium-ion batteries: An analysis of current practices of vehicle manufacturers and policies in the EU. Resources, Conservation and Recycling, 172, 105658., . Yes, indeed, the advent of electric vehicles creates a problem to solve, namely what to do with all those batteries. I mean two categories of batteries. Those which we need, and hope to acquire easily when the time comes for changing them in our vehicles, in the first place, and those we don’t need anymore and expect someone to take care of them swiftly and elegantly.