The main point of the previous blog article was: we emit about 85% of CO2 and about half of methane because we use coal, oil and gas. Only about 15% of CO2 emissions are primarily non-technological in origin, such as deforestation. This means that climate change is primarily caused by inadequate technologies.
The question now is: can we solve the problem by changing technologies? Or are lifestyle changes necessary? Or even systemic changes? This is not obvious. In this blog article, you will get some categorical thinking to help you explore this question in your own field and evaluate it for yourself.
Whenever you come across a proposed mitigation measure, you may think of the following three categories:
- Sustainability. It means  replacing technology that we might call “environmentally unadapted” with technology that is consistent with a favourable ecology in the long term (see next blog for a deeper discussion). Of course, every technology and every living thing has some impact. However, indicators of sustainable technologies are: no material degradation (e.g. through recycling), no pollution on a scale that causes suffering, no degradation of biodiversity in the long run, and so on. A counter-example could be: replacing fossil fuels with a technology that depletes certain metals and causes significant environmental damage.
- Efficiency. Making things more efficient. But efficiency may also improved in a broader context, such as optimising economic efficiency with different constraints than now, such as including external costs or assigning a value to natural resources.
- Sufficiency. Many of us think of sufficiency as primarily quantitative, like consuming and producing less, and many of us associate sufficiency with a frugal lifestyle. However, sufficiency also has qualitative aspects: You have probably regulated your needs all your life to prevent you from overspending your budget. And think of all the regulations that are out there. I know, it’s not the nicest part of economics….
Note an important difference: efficiency and parts of sufficiency are primarily quantitative measures, while sustainability and parts of sufficiency are primarily a qualitative shift. These categories become more interesting and useful when you look at their interaction, like in this following example.
Cosider lighting in the UK over three centuries, where detailed data is available [2–6] from the pre-industrial year 1700  with candles and oil lamps to today with LED lamps. And let’s look at the data first before drawing any conclusions.
Since the industrial revolution, the efficiency of lighting has improved, which roughly means  that more light comes out per given energy input. Even candles have improved due to improved tallow candle manufacturing (wax candles were always a luxury item and not widely used). Note that efficiency is plotted on a logarithmic scale, with cycles of ten-fold improvements being called “an order of magnitude”. On the usual linear scale, you would not see the improvements made over more than four orders of magnitude, you would only see electric light bulbs, neon tubes and white LED bulbs, because only with electricity has the efficiency grown to a significant level.
This graph makes it clear that most progress has come not from the optimisation of specific technologies, but from the invention and development of new technologies, what we usually call technology cycles. In other words, the main driver of progress has been qualitative shift.
More efficient lighting has also reduced costs in the UK by more than four orders of magnitude:
Here, the price per amount of light is corrected over centuries with the retail price index  and converted to British pounds (GBP) of the year 2020. Indeed, one can imagine that running all the streetlights, indoor lighting, signal lights etc. at the same light intensity as we do now, but using candles, would be about 10’000 times more expensive. The data point at the end is the price you pay now for a white LED bulb and electricity. It has become incredibly cheap. We live in an incredible age.
With new technologies, more efficient light sources and falling prices, the UK has been producing more and more light per person:
Of course, this has a lot to do with affordability. For example, the ancient Romans who lived in the affluent Bay of Naples (in Pompeii and Herculaneum) were buried under the ashes of the volcano Vesuvius in the year 79. They left behind an arsenal of oil lamps, torches, fireplaces and candles that was brighter per inhabitant than in London in the 1820s .
Today we afford an ever-increasing amount of lighting. With better light bulbs, street lighting is increasing significantly, monitored by satellites , :
More street light has been installed although more lighting does not reduce street crime or car accidents (it may even tempt drivers to drive faster). On world average, 30% of people cannot see the milky way from their home — 60% in Europe, 80% in the US . No empirical evidence is found for a saturation in per-capita generation of light, even in rich nations .
Now, let’s apply the three categories of sustainability, efficiency and sufficiency to these data.
Does the current lighting technology of white LED lamps meet the criterions of sustainability? That is, does it have a limited impact on us, the environment and its resources, so that the ecology can maintain its favourable metabolism for years to come?
The tiny light source inside a typical 12-watt LED bulb consists of a semiconductor device that works in the opposite way to a solar cell: applying a voltage produces light. It is mainly made of gallium nitride and contains layers of gallium indium nitride (GaN, GaInN), so typically contains 4.2 mg Ga and 0.035 mg of indium . Tiny amounts. But with a few billion light bulbs produced every year, they consume a whole quarter of the world’s yearly gallium production and a mere 35 micrograms of indium already uses about 2.5 % of yearly indium production.
The semiconductor emits blue light and is then partially converted into yellow light on the wall of the bulb, giving the impression of white light from the outside. The wall typically contains 300 mg of yttrium (Y), which consumes about half of the world’s Y production, and typically contains 6 mg of the rare earth element cerium (Ce), with a small share of its world production.
These four elements are very dispersed in the earth’s crust and occur only as by-products in other metal ores. For example gallium is present in trace concentrations in bauxite for aluminium production, indium in zinc ores, and yttrium in various ores. Increasing the production of these four trace elements would lead to overproduction and a drop in the price of their host ores and is therefore not economically viable. In this context, you may be surprised to learn that light bulbs actually consume about half of the yttrium production, and you may assume that there will soon be a yttrium shortage. However, neon tubes contain a whole 1 – 5 g of it, instead of only 300 mg, and their production is declining.
Substituting these four elements is difficult and will take some time, such as using silicon carbide as a semiconductor, manganese ions in the tube walls, etc. , .
For the sustainability criterion, the use of rare elements is not per se a killer. A killer is the fact that none of these four elements are recycled, and recycling is not on the horizon  because the amounts per bulb seem too small.
Nevertheless, these bulbs are more sustainable than most lighting technologies to date  (not to mention lamps that used whale oil). Still, I think it is the responsibility of the lighting industry to turn their technology into a sustainable technology, and they may need support and pressure to give this task enough priority. We will look deeper into sustainability in the next blog. Meanwhile, for a foretaste, you are invited to make some playful guesses in the comments section at the end of this article.
Looking at the graphs above, you might wonder if more efficient lighting has saved any energy at all: Efficiency has increased by more than 4 orders of magnitude, and so has the amount of light generated per person. This phenomenon is well known and is called the rebound effect . One can think of the rebound on three different levels:
- Products become cheaper but are therefore sold in larger quantities;
- leftover money is spent on other CO2-intensive goods and services; and
- greater efficiency creates new applications, which increases economic growth, which in turn can lead to more CO2 emissions.
Additionally, think about compensation: in cold climates, gas lamps and the old-fashioned light bulbs helped heat up the room, now this is compensated by more space heating.
So, how big is the rebound in lighting? Quantifying rebound is difficult. Even at the first level, this cannot be done simply by comparing the efficiency gain with the production gain, as it is difficult to know today’s level of production without the efficiency gain that has occurred. For example, if we were still using gas luminaires as our main lighting today, we might install many more such luminaires than was done in the past and we would cause more CO2 emissions for lighting today. An approximate idea of this is given by the so-called consumer surplus: people are willing to pay more for lighting because it is so cheap. Most technology cycles make goods cheaper and, with this, usually increase consumer surplus. All this suggest that :
Reducing CO2 emissions through more efficient technology alone may soon be outpaced by economic growth.
This is not to say that it is unimportant to make technology more energy efficient. For example, the electric motor in electric vehicles is close to 100% efficient, while the combustion engine is usually less than 30% efficient due to the basic physics of converting heat into mechanical energy. Due to this fact, if you run all vehicles electrically, you only have to increase electricity production by about 20%, which makes the switch to electric vehicles much easier than if electric cars required the same amount of energy as combustion engines .
So how can we save CO2 emissions from lighting?
The aim of this blog is not to give solutions, but to give you tools for thinking in your own field. So what is an effective strategy to approach such a problem?
One common way is to check whether the emissions in question are related to gross domestic product (GDP). Just keep in mind that GDP is not the best or even appropriate measure for all tasks. And that emissions and energy consumption are per se unrelated to GDP: most EU countries have reduced their emissions and stagnated their energy consumption while increasing their GDP, even when taking into account emissions from imported goods minus emissions from exported goods.
With this in mind, let’s plot the amount of light per person over GDP. To account for cost decay over time, let’s divide GDP by the cost of lighting (CoL) per amount of light :
Indeed, populations have spent a similar proportion of their GDP on light, both over centuries and across different regions of the world today. Regions where there is no electricity grid today are in a similar situation to the UK around 1865, just before the start of the 2nd Industrial Revolution .
Why is this?
I have come across two suggestions. One is  that very poor people focus on food and cooking, poor people on shelter and heating/cooling, middle-income people on mobility, digital communications and lighting, and richer people on consumer and lifestyle goods. Economists measure this development with so-called price elasticities, i.e. how the quantity demanded of a good would change if its price changed: This trend from poor to rich makes all these main goods rather price inelastic, which means that price changes cause a small change in the quantity bought.
With price-inelastic goods, policies to tailor supply tend to meet little resistance.
Who is upset when street lighting is curtailed? A minority of people and a specialised street lighting industry. Therefore, when the EU abolished the iridescent light bulb , it also introduced a lighting principle called ARLA (stands for “as low as reasonably achievable”) to reduce the rebound effect . To put it in general terms :
For a new technology to save CO2 emissions, it needs both: innovative products and innovative practices such as recycling, possibly supply tailoring or other measures. As much creativity should go into innovative practices as into innovative products.
Innovative practices is a nice word for sufficiency. Sufficiency is not new. Even affluent nations have a long tradition of sufficiency, be it the abolition of child labour or the 7-day week. We take them for granted and have forgotten that they were fiercely contested measures against GDP growth. The next time a climate activist talks about sufficiency, you might think of these two past measures to get cool on that word….
The second guess as to why the amount of light per person keeps increasing has to do with our complex visual system: it is key to how we experience the world around us, and we love to enhance that experience, including through the use of artificial light (and large TV screens). Another aspect may be how we regulate our needs .
My personal answer to all this is:
To reduce CO2 emissions, find the root causes and work on them.
The root cause of CO2 emissions from lighting is not the light bulbs themselves. Only small amounts of CO2 are emitted during their production and for the provision of materials . The main cause of their CO2 emissions is that they use electricity from fossil fuel power plants.
In the previous article, we looked at the allocation of CO2 emissions by primary responsibility. For light bulbs, it lies with the power plants.
This brings us back to the question: Are renewable energies a sustainable technology? This is a crucial question that I will write about soon.
Until then, I wish you romantic dinners by candlelight.
Add your comments
Let’s make a few playful guesses as to how we may make today’s white LED lighting technology more sustainable:
- Going backwards. With the growing share of renewable energies, the higher power consumption of the classic light bulb is no longer a problem, and we saw above that more efficient light bulbs lead to more consumption anyway. The filament of classic light bulbs is usually made of tungsten, a rare element but – like all parts of the classic light bulb – can be recycled by mechanical disassembly. What is the problem? Good old days?
- Present improvement. The semiconductor part of computer chips is designed to work for about 250 years (you threw away your old computer for reasons other than chip failure). The blue light emitting semiconductor in the middle of the bulb may be manufactured for a long life and reused, which of course is more expensive. The materials at the bulb wall may be removed, refined and reused rather than recycled, which is also more expensive. Would this be a problem, given that lighting tends to be price inelastic because it has become so incredibly cheap?
- Present improvement. Issue certificates worldwide for mining ores. They may be sold at a higher price for each metal the fewer reserves are left, and the profit may be used to subsidise recycling to stop wasting rare elements. A solution to many problems, not just lighting?
- Future improvement. Accelerate the next technology cycle to replace the white LED bulb with a bulb having three organic LEDs emitting three different colours that make up white. This has already been realised in many mobile phone displays. But their light intensity is usually too low and production is expensive. In any case, every new technology is expensive before it becomes mainstream. Our white LED bulbs are just not the best step in lighting technology, are they?
- Future improvement. Let the free market sort it out. Some of the four elements in white LEDs will run into supply problems, so manufacturers will jump to the next technology. That’s always been the case, hasn’t it?
- Press the button above to suggest your option. Feel free to be playful, we all don’t know the future.
 J. Huber, “Towards industrial ecology: sustainable development as a concept of ecological modernization,” Journal of Environmental Policy & Planning, vol. 2, no. 4, pp. 269–285, Dec. 2000. https://doi.org/10.1080/714038561
 R. Fouquet and P. J. G. Pearson, “Seven Centuries of Energy Services: The Price and Use of Light in the United Kingdom (1300-2000),” The Energy Journal, vol. 27, no. 1, pp. 139–177, Jan. 2006. https://doi.org/10.5547/ISSN0195-6574-EJ-Vol27-No1-8
 R. Fouquet, “Long run trends in energy-related external costs,” Ecological Economics, vol. 70, no. 12, Oct. 2011. https://doi.org/10.1016/j.ecolecon.2011.07.020
 R. Fouquet, “Consumer Surplus from Energy Transitions,” The Energy Journal, vol. 39, no. 3, pp. 167–188, Jul. 2018. https://doi.org/10.5547/01956574.39.3.rfou
 J. Y. Tsao and P. Waide, “The World’s Appetite for Light: Empirical Data and Trends Spanning Three Centuries and Six Continents,” LEUKOS, vol. 6, no. 4, Apr. 2010. https://doi.org/10.1582/LEUKOS.2010.06.04001
 J. Brodrick, “Solid-State Lighting, 2017 Suggested Research Topics Supplement: Technology and Market Context,” Sep. 2017. https://www.energy.gov/sites/prod/files/2017/09/f37/ssl_supplement_suggested-topics_sep2017_0.pdf
 E. Hawkins et al., “Estimating Changes in Global Temperature since the Preindustrial Period,” Bulletin of the American Meteorological Society, vol. 98, no. 9, Sep. 2017. https://doi.org/10.1175/BAMS-D-16-0007.1
 T. Goh and B. W. Ang, “Four reasons why there is so much confusion about energy efficiency,” Energy Policy, vol. 146, Nov. 2020. https://doi.org/10.1016/j.enpol.2020.111832
 L. H. Officer, “Retail price index 1256-2002,” Economic History Services (UK). 2004. EH.net
 K. Whiting, L. G. Carmona, L. Brand-Correa, and E. Simpson, “Illumination as a material service: A comparison between Ancient Rome and early 19th century London,” Ecological Economics, vol. 169, Mar. 2020. https://doi.org/10.1016/j.ecolecon.2019.106502
 F. Falchi et al., “The new world atlas of artificial night sky brightness,” Science Advances, vol. 2, no. 6, Jun. 2016. https://doi.org/10.1126/sciadv.1600377
 C. C. M. Kyba et al., “Artificially lit surface of Earth at night increasing in radiance and extent,” Science Advances, vol. 3, no. 11, Nov. 2017. https://doi.org/10.1126/sciadv.1701528
 P. Gaffuri et al., “Potential substitutes for critical materials in white LEDs: Technological challenges and market opportunities,” Renewable and Sustainable Energy Reviews, vol. 143, Jun. 2021. https://doi.org/10.1016/j.rser.2021.110869
 M. Buchert et al., “Substitution als Strategie zur Minderung der Kritikalität von Rohstoffen für Umwelttechnologien Potentialermittlung für Second-Best-Lösungen,” 2018. https://www.bmu.de/fileadmin/Daten_BMU/Pools/Forschungsdatenbank/fkz_3714_93_316_kritikalitaet_umwelttechnologien_bf.pdf
 M. Franz and F. P. Wenzl, “Critical review on life cycle inventories and environmental assessments of LED-lamps,” Critical Reviews in Environmental Science and Technology, vol. 47, no. 21, Nov. 2017. https://doi.org/10.1080/10643389.2017.1370989
 A. D. Steve Sorrell, Birgitta Gatersleben, “Energy sufficiency and rebound effects, European council for an energy efficient economy,” 2018. https://www.energysufficiency.org/libraryresources/library/items/energy-sufficiency-and-rebound-effects-concept-paper/
 R. B. Jackson et al., “Global energy growth is outpacing decarbonization,” Environmental Research Letters, vol. 13, no. 12, Dec. 2018. https://doi.org/10.1088/1748-9326/aaf303
 F. Ram M., Bogdanov D., Aghahosseini A., Gulagi A., Oyewo A.S., Child M., Caldera U., Sadovskaia K., B. J., Barbosa LSNS., Fasihi M., Khalili S., Dalheimer B., Gruber G., Traber T., De Caluwe F., Fell H.-J., and C., “Global Energy System based on 100% Renewable Energy – Power, Heat, Transport and Desalination Sectors. Study by Lappeenranta University of Technology and Energy Watch Group, Lappeenranta, Berlin,” 2019. http://energywatchgroup.org/new-study-global-energy-system-based-100-renewable-energy
 Wikipedia, “Second Industrial Revolution,” 2021. https://en.wikipedia.org/wiki/Second_Industrial_Revolution
 European Commission, “Energy label and ecodesign,” 2019. https://ec.europa.eu/info/energy-climate-change-environment/standards-tools-and-labels/products-labelling-rules-and-requirements/energy-label-and-ecodesign/energy-efficient-products/lighting_en
 European Commission, “Green Public Procurement Criteria for Road Lighting and Traﬃc Signals,” 2019. https://doi.org/10.2760/372897
 N. Schulte-Römer, J. Meier, M. Söding, and E. Dannemann, “The LED Paradox: How Light Pollution Challenges Experts to Reconsider Sustainable Lighting,” Sustainability, vol. 11, no. 21, Nov. 2019. https://doi.org/10.3390/su11216160
 E. Klein, “Why Do We Work So Damn Much?,” New York Times Podcasts, 29-Jun-2021. https://www.nytimes.com/2021/06/29/opinion/ezra-klein-podcast-james-suzman.html?showTranscript=1
Data sources for charts
Most data are extracted from Refs.  – ; the world map of night sky quality from Refs. [11,12]; the chart on materials from Ref. .