Atomization of Society

Tue 25 June 2013

The esteemed Dr. Anjum Altaf argues in an email (with a link to his op-ed),

"The difference between a village and a city is density – the number of residents per unit of land. With low population and low density it makes sense for each household to provide its own services. With high population and high density, that makes no sense at all. In fact, the advantage of high density is that public services can be provided collectively at much lower cost per capita."

He backs up his claim with case studies he has done in the area of water supply.

I disagree with this claim - that public services are necessarily cheaper if provided collectively. I would agree that in general the principle of economies of scale works. But it is only a rule of thumb. It is a very useful assumption to start your analysis with. Then you must follow up with quantitative analysis.

However, let us start with some abstract analysis. Consider a particular public service, say electricity. The cost of the system can be divided into two, the generation of electricity and its distribution. In a central system, such as those provided by the government, both of these costs can be substantion. At the other end of the scale are what Dr. Anjum calls atomized system, where individual households generate their own electricity. Here, there is no cost of distribution, though there are possibly additional generation costs related to storage. There are one other kind of networks that bear mentioning. These the distributed systems. In these even though each household is a generator, distributer and receiver of electricity. Distributed systems bear both generation and distribution costs.

In each of these systems, there are various ways of generating the electricity and distributing it. Coal and gas power plants, generators and wind turbines are some ways of generating power. There is primarily only one distribution technique used across the world, AC current over transmission lines.

To compare different systems, we must be able to calculate costs on an equal footing. We do this by computing the Levelized Cost of Energy (LCOE) for each system. This is the the total cost of constructing and running the plant, and distributing electricty over its useful lifetime divided by the total energy delivered during this period. The units of this come out to be Rs/KWh. If you recall, these are the same units in which our local electricity distributors charge us. The typical price for electricity for domestic usage by LESCO is 10.65 Rs/KWh (there are higher and lower rates depending on the number of units consumed). The LCOE for electricity produced by domestic generators can range from 30-100 Rs/KWh. Other electricity generation systems each have their own LCOE.

The question we want to answer is which of these systems can provide electricity at the lowest possible LCOE. Till now it has seemed that indeed centralized systems using coal, oil, gas, nuclear or hydel power provide the cheapest possible electricity. The average cost of electricity across the world using these methods is about 11.5 Rs/KWh.

However, atomized and distributed forms of electricity generation are quickly catching up. On the one hand, the prices of traditional forms of electricity (the centralized systems) is increasing because of increasing fuel prices. On the other hand, the prices of these atomized forms of electricity are falling as the technology develops, much in the same way the prices of electronics falls year in and year out.

The two best known examples of these atomized generation systems are wind and solar photovoltaic (PV) panels. Even though, wind is much cheaper than solar PV right now, it is not technically feasible in all places. I will discuss solar PV instead.

The cost of solar depends on the cost of panels used and the solar insolation (the energy from sunlight) available in an area. The solar insolation can vary by as much as a factor of 6 across the planet, and variation in price can be just as much. However, certain regions of the world are achieving grid parity. This is the point in time when in a certain market the LCOE of electricity from solar becomes the same as the price of conventional grid electricity.

Take a look at Global overview on grid-parity by Christian Breyer, Alexander Gerlach, which projects grid parity events for more than 150 countries. From this study (in the supplementary data) we have a table which indicates grid parity events for the residential sector.

Residential grid-parity events by year and world region.
yearEuropeThe AmericasAfricaAsia-Pacific
2010Cyprus, ItalyGrenada, Guyana, JamaicaBurkina Faso, Chad, Liberia, Madagascar, Uganda
2011Denmark, Malta, Portugal, SpainBrazil, Cuba, Dominican RepublicMaliAfghanistan, Cambodia
2012Austria, GermanyChile, Suriname, UruguayCentral African Republic, Cote d'Ivoire, SenegalFiji, Japan, Palestine (W.Bank/Gaza), Philippines
2013Belgium, Hungary, Israel, Luxembourg, Netherlands, TurkeyBelize, El Salvador, Nicaragua, Panama, Peru, Puerto RicoBenin, Gambia, Kenya, Morocco, NamibiaNew Zealand
2014Croatia, Greece, Ireland, Slovakia, Slovenia, SwedenGuatemala, MexicoNiger, Rwanda, TogoAustralia, Jordan, Lebanon, Syria
2015Czech Republic, Finland, France, Norway, Poland, Romania, Switzerland, United KingdomArgentina, United States EastCameroon, Mauritius, Mozambique, SudanBangladesh, Brunei Darussalam, Burma (Myanmar), India West, South Korea
2016Bulgaria, LatviaColombia, Ecuador, Haiti, United States, United States North WestGabon, Guinea, Tanzania, TunisiaHong Kong (China), India East
2017Estonia, LithuaniaBolivia, Honduras, United States South WestBotswanaIndia, Thailand
2018Canada, Costa Rica, ParaguayGhanaChina, China East, Malaysia, Pakistan, Sri Lanka, Taiwan
2019ArmeniaAlgeriaChina West, Indonesia, Vietnam
2020Belarus, SerbiaTrinidad and TobagoCongo R, EthiopiaChina Central, Iran, United Arab Emirates
2020Moldova, Russian Federation, UkranieVenezuelaAngola, Burundi, Congo DR, Egypt, Libya, Malawi, Nigeria, Seychelles, South Africa, Zambia, ZimbabweAzerbaijan, Iraq, Kazakhstan, Kuwait, Kyrgyzstan, Lao PDR, Qatar, Saudi Arabia, Tajikistan, Turkmenistan

This table does not incorporate subsidies and government incentives that make solar power even cheaper in some areas. Despite this, already some countries have achieved grid parity, and others are not far behind. Hence, we can see that atomization of public services, at least in the case of electricity, can be cheaper than centralization.

The above were some positive statements. Now for some policy recommendations for Pakistan.

Pakistan is still a few years away from achieving grid parity as per this report. However, it did not consider the electricity shortage problem in the country. With people using generators and UPS and what not, electricity costs for typical domestic users are far above 10 Rs/KWh. Some back of the envelope calculations suggest that the price of solar PV panel setups can range from Rs. 20 to 30 per KWh - prices in other Asian countries have been empirically observed to be lower. Still, these prices are less than the price of generation through diesel/gas generators. On the flip side, there is significant capital costs which must be borne.

As Dr. Anjum argues, people are often unable to trust or rely on centralized systems to provide services. In Pakistan, people are already forced to turn to alternate forms of electricity generation to fulfil their needs and wants, techniques which make the shortage situation worse in the country (explaining this will require a post of its own). In this situation, I would recommend that people turn to solar power for generation as a relatively cheap and sustainable alternative.

To protect our privacy, I have disabled my third party commenting system. You can contact me on twitter @abdullahkhalids, or email me.