Skip links

Political economy of Lamu coal plant and why we don’t need it

“Facts are stubborn things; and whatever may be our wishes, our inclinations, or the dictates of our passions, they cannot alter the state of facts and evidence,” said John Adams, the second President of the United States.

Do we need a 1,000MW coal power plant? Our installed electricity generating capacity is 2,325MW. This excludes 310MW capacity of the Lake Turkana wind plant, which is complete but is not connected to the grid because the government is yet to complete the transmission line. The installed capacity, which has increased by 40 per cent over the last five years, consists of equal shares of hydro and diesel plants at 35 per cent, 28 per cent geothermal, and two per cent wind and sugarcane cogeneration plants.

That is, two thirds renewable sources, and one third fossil fuel. The geothermal share has increased more than double from 12 per cent. In fact, geothermal accounts for two thirds of the increase in installed capacity. When we include Turkana wind plant, renewable rises to 70 per cent.


But when it comes to generation, geothermal leads. Geothermal accounted for 45 per cent of the electricity generated last year, followed by hydro at 40 per cent for a combined share of 85 per cent. The diesel plants did 15 per cent and wind less than one per cent.

The sugar cogeneration plants did not feature, since the biggest one, Mumias Sugar, is not working. What this tells us is that geothermal is providing the “base load” power. This is reflected in the fact that geothermal plants operated at close to 80 per cent capacity, hydro at 55 per cent and diesel plants at 20 per cent.

As noted, the installed capacity has increased by 40 per cent, but the actual power generated has increased by 32 per cent, that is, capacity has grown faster than demand. Factoring in the idle Turkana power raises the capacity increase to 60 per cent, double the growth rate of demand. Earlier in the year, Kengen was reported to have broken ground on another 140MW of geothermal power, part of a plan to add 720MW to the grid by 2022, of which 680MW will be geothermal, and 90MW wind power.

These are by no means the only projects underway or in advanced stages of development, for instance. Others include Kipeto Energy’s 100MW wind farm in Kajiado and a 50MW solar plant being developed by the Rural Electrification Authority in Garissa. These plants combined with the Turkana one add up an additional 1,130MW, translating to another 50 per cent increase in generation capacity.

There is no compelling reason to anticipate that demand will accelerate. We can say without fear of contradiction that even without the Lamu coal plant, generation capacity will continue to outpace demand for the next five years.

There are two more important insights from this information. First, is that we are very close to being able to get all our electricity from renewable sources. When Turkana wind and the projects under construction are connected, we will most certainly be in a position to.

The second is a puzzle. We have seen that we are generating 80 per cent of our power from geothermal and hydro power, with the diesel share down to 15 per cent from 37 per cent five years ago. Since geothermal and hydro are our cheapest sources, this begs the question, why is electricity not becoming cheaper?

This is a question for the energy honchos in government. I will highlight two factors, and they are not the only ones. First, generating capacity is paid for whether it is dispatched or not, transmission and distribution inefficiency notwithstanding. Our transmission and distribution losses are in the order of 18 per cent of the power generated. It is not the worst in the world  some countries lose 30 per cent  but the global developing norm is 10 to 15 per cent and 5 to 10 per cent for industrialised countries. Money has been pumped into revamping the distribution system for decades to no avail.


On current generation, a one percentage point reduction translates to more than a billion shillings in revenue for Kenya Power. Second, generation capacity is paid for, whether it is used or not. So even though we are only using 20 per cent of the diesel plants capacity, they are also paid a “capacity charge” as long as they are in the system.

The long and short of it is that a lot more goes in the cost structure of power than the notional cost of alternative generating technologies. Just because coal power is supposed to be a cheap technology does not mean adding a coal plant will make our electricity cheaper.

In this case, the opposite is more likely to be the true. Why? First, coal plants are used for base load. As demonstrated above, by the time this plant comes on stream we will have more than enough geothermal base load capacity and possibly wind as well because the technical parameters of the Lake Turkana wind regime (which is in Marsabit) compare favourably with coal power in terms of reliability.

To keep the coal plant running, it will have to displace geothermal capacity, which does not make business, economic or ecological sense. This means that the coal plant will most likely be idle. We will be renting a plant we are not using. Second, technical transmission and distribution losses are a function of distance. Huge plants far from demand increase the losses, as well as the capital cost of the whole system since the transmission lines, which are very expensive, also have to be paid for.

The ideal system is one where generating plants are close to the customer. Marsabit is far from current demand centres, but that is where the resource is, and it is free therefore it merits building a transmission line. There is no coal in Lamu, the only reason for putting a plant there is to be close to the landing point of imported coal. This saves money for the investors, but the consumers and the tax payers pick up the tab for the capital cost of the transmission line as well as the technical losses of transmitting the power over long distances.

This column argued two weeks ago that grid-scale solar is now the cheapest source of electricity worldwide, and wind is not far behind. The facts speak for themselves. To give proper perspective, the information we have is that the Lamu coal plant will cost Kenya Power 7.52 US cents per kilowatt hour (Kwh) (Sh7.80). Statistics from the International Renewable Energy Agency, an intergovernmental organisation, show that the average price of solar power prices was 5 US cents per kilowatt hour (Sh5.20) down from 25 US cents (Sh25.90) a decade ago.

The investment bank Lazard publishes a report on the comparative cost of electricity generation. Its 2016 report shows the five cheapest sources of grid power are wind (3.20 to 6.20 US cents per KWh), followed by solar (4.60 to 6.10), natural gas (4.80 to 7.80), coal (6.0 to 14.30) and geothermal (7.90 to 11.70).

Abu Dhabi has recently signed on a 350 MW solar plant at 2.40 cents per KWh, that is a third of the Lamu coal plants power purchase price, Dubai has signed up an 800 MW one at 2.99 and Chile has a 580 MW plant which sells power at 2.91 cents.

The defenders of the coal plant contend that solar and wind are uncontrollable and erratic. This is true. The sun sets, and you cannot switch wind on and off to meet demand surges. I have three observations on this. First, the shortcomings of solar and wind do not make a case for coal. The Lazard data shows that natural gas is the next cheapest source after wind and solar. It’s the cleanest fossil fuel and unlike coal, it is also good for peaking plants, making it a good choice for replacing the diesel plants. A gas plant even in Lamu is a more rational option as we’d be replacing more expensive and dirtier diesel.

The second point has to do with the implications of cheap grid-scale storage. A recent study by the consulting firm McKinsey The New Economics of Energy Storage concludes “It is already profitable to provide energy-storage solutions to a subset of commercial customers in each of the four most important applications — demand-charge management, grid-scale renewable power, small scale-solar-plus storage, and frequency regulation”.

According to the Lazard data, the cost of diesel plants, which are the most expensive source at 21 to 28 US cents per KWh, that is more than four times the cost of solar or wind power. At these prices storage of solar power to meet peak demand is a very viable option. In a country like ours where the sun shines everywhere every day, there is no geographical limitation to where these plants can be located.


There is an interesting local angle to the storage story. The most popular technology, though not the only one, is lithium-ion, the same one in the mobile phone. As it happens, lithium is one of the mineral by products of geothermal power – indeed Kengen was shopping for a consultant to advice on how to commercialise it. So, here we are with clean power right by the key raw material for one of the biggest emerging industries in the world, talking coal, instead of courting Tesla and other champions of green power to manufacture batteries here. Our leaders have their eyes firmly on the rear view mirror. The people of Lamu do not want a coal plant. We do not need it. Why are they insisting on building one? The answer is simple. Impunity. The people behind it would not build it in their backyard. They are building it in Lamu because they wield state power, and can ride roughshod over the natives, in the name of “development”, as their colonial predecessors did. Where I come from, we call these people ngaati (homeguards). This, gentlemen, is for you: “Perchance he for whom this bell tolls may be so ill as that he knows not it tolls for him and perchance I may think myself so much better than I am as that they who are about me, and see my state may have caused it to toll for me and I know not that. … and therefore never send to know for whom the bell tolls; it tolls for thee” (John Donne).

By David Ndii, Daily Nation