11 October 2011

Energy Not Yet for All

The IEA has released a preview of its 2011 International Energy Outlook (here in PDF).  In it is describes the challenge of providing energy access to people around the world and how current policies are falling well short.
Modern energy services are crucial to human well‐being and to a country’s economic development; and yet globally over 1.3 billion people are without access to electricity and 2.7 billion people are without clean cooking facilities. More than 95% of these people are either in sub‐Saharan Africa or developing Asia and 84% are in rural areas.

In 2009, we estimate that $9.1 billion was invested globally in extending access to modern energy services. In the absence of significant new policies, we project that the investment to this end between 2010 and 2030 will average $14 billion per year, mostly devoted to new on‐grid electricity connections in urban areas. This level of investment will still leave 1.0 billion people without electricity and, despite progress, population growth means that 2.7 billion people will remain without clean cooking facilities in 2030. To provide universal modern energy access by 2030 annual average investment needs to average $48 billion per year, more than five‐times the level of 2009. The majority of this investment is required in sub‐Saharan Africa.  
Remarkably, the Millennium Development Goals of the United Nations do not even include energy access among their priorities.  Thus, it is no surprise that the IEA places making energy access a political prority at the top if its recommendations:
Adopt a clear and consistent statement that modern energy access is a political priority and that policies and funding will be reoriented accordingly. National governments need to adopt a specific energy access target, allocate funds to its achievement and define their strategy for delivering it.
In case you are curious, what does "energy access" actually mean?  The IEA defines energy access contextually, and it starts here:
The initial threshold level of electricity consumption for rural households is assumed to be 250 kilowatt‐hours (kWh) per year and for urban households it is 500 kWh per year. In rural areas, this level of consumption could, for example, provide for the use of a floor fan, a mobile telephone and two compact fluorescent light bulbs for about five hours per day. In urban areas, consumption might also include an efficient refrigerator, a second mobile telephone per household and another appliance, such as a small television or a computer.
I am sure that readers of this blog would hesitate to call such a level of consumption "energy access."  The average US household uses 20-40 times as much energy! Even if US households were to cut their consumption by half (unlikely) under aggressive assumptions about efficiency, it would still vastly exceed the initial threshold defined by the IEA.

The IEA observes that energy access is actually a process:
Once initial connection to electricity has been achieved, the level of consumption is assumed to rise gradually over time, attaining the average regional consumption level after five years. This definition of electricity access to include an initial period of growing consumption is a deliberate attempt to reflect the fact that eradication of energy poverty is a long‐term endeavour. In our analysis, the average level of electricity consumption per capita across all those households newly connected over the period is 800 kWh in 2030.
In anything, the IEA has underestimated future demand for energy, as 800 kWh per year is just the start. The world needs more energy -- much more energy.


  1. Prediction is hard, especially about the future;)

    in the 2004 World Energy Outlook the IEA projected growth of only 2.6% per annum in total energy use (total primary energy supply – TPES), and of 2.8% in CO2 emissions, in China over that period (IEA 2004).

    An authoritative study by the Energy
    Research Institute of the National Development Commission of China and the Lawrence Berkeley National Laboratory in the USA, released in October 2003, projected growth in energy demand in China of 3.8% per annum between 1998 and 2020

    In hindsight the past projections about future Chinese Energy demand don't appear to be any better then rolling dice or throwing darts blindfolded.

    IMHO Developing countries are like bottled champagne..you can't know how much champagne is going to spray out when you remove the cork just by looking at the bottle.

  2. The link to the 1997 US electricity use gives ~23 kWh per day as the median household electricity use. I would bet that newer statistics would be lower, partly because of efficiency increases since 1997 (e.g., fluorescent light bulbs, more efficient appliances, etc.). Our household (family of 4, 2300 sq. ft.), uses ~11.5 kWh per day, for example, although this is probably not typical because I am careful about electricity use.

  3. p.s. Of course, 11.5 Kwh per day is still ~8x greater than the IEA urban target, so Roger's point remains largely correct, although perhaps not as extreme as originally stated, especially given future anticipated efficiency increases.

  4. -2-climateprof

    I'd guess that the values would be lower for 2011 (surprised that I can't find that info), also because there are more and smaller households. Not that you picked median and I picked mean. Thanks!

  5. Demand is best measured not as a quantity of consumption but as a functional relationship between price and consumption. If energy were free, I'd consume a lot more. If energy were more expensive, I'd consume less. Rising demand means that I'd be willing to pay more to consume the same amount I consume today.

    Thus, comparing the quantity of energy consumed in the poor parts of the world to energy consumed in the rich parts we ought to distinguish whether the limiting factor is the ability to provide energy or the ability to pay for it.

    From this perspective, I don't see that American levels of consumption are remotely relevant to defining an energy poverty threshold, which seems to be the focus of the EIA report, any more than American levels of other forms of consumption are relevant to defining an economic poverty threshold that reflects the ability to provide the basic necessities of life.

    If we take your 1997 RECS numbers with a median US houshold consumption of 8400 kWh per year, then even if there were an unlimited supply of electricity at $0.10 per kWh, the global median household, with an income of about $2800 per year, would be hard pressed to afford US levels of consumption. And if we look at the bottom quintile of the world---households earning less than $800 per year---the entire household income would not pay for US levels of electricity consumption. And this is before we even start talking about bringing them up to US levels of gasoline consumption (around 450 gallons per capita per year).

    @Climateprof, #2: See the DOE's Residential Energy Consumption Survey (RECS) for more recent data: http://www.eia.gov/consumption/residential/ gives a mean household electricity consumption in 2005 of 31 kWh per day. I can't find quantile analysis of the 2005 RECS data, so I don't know what the median is (it's typically significantly less than the mean, but nowhere near a factor of 2). See also, Art Rosenfeld's famous chart here: http://ag.ca.gov/globalwarming/images/Rosenfeld.jpg, which shows fairly constant per-household electricity consumption (Rosenfeld's chart says per-capita, but I believe the numbers are really per houshold) of 33 kWh per day from 1998-2005.

    One thing we have seen is that even while household appliances have become more energy efficient, the average house has become much larger and the number and size of appliances has grown (e.g., cubic foot for cubic foot, refrigerators are much more energy efficient than they were 30 years ago, but the average size of a refrigerator is significantly larger and the fraction of households with two or more refrigerators has skyrocketed).

  6. Bill Gates's comment, eleven years ago, is relevant:

    " '[D]o people have a clear view of what
    it means to live on $1 a day?' ... Gates[] asked. 'There's no
    electricity in that house. None.'

    "When a moderator brought up solar power,
    Mr. Gates shot back, 'No! You can't afford a
    solar power system for less than $1 a day.'
    And, 'You're just buying food, you're trying to
    stay alive.' "

  7. -5-Jonathan Gilligan

    I doubt that your or my energy consumption is very price sensitive except at the margin (at what price point will you turn off your refrigerator or your heat?). We've got all the energy we want. That is because energy is cheap compared to our incomes.

    So I see a different causality -- People are too poor because they don't have access to energy, not just vice versa (of course they are related, but I suppose that put more emphasis on cheap energy as a precondition of growth).

    Either way, the way to expand energy access either is that we need massive amounts of cheaper energy or to make poor people richer. My view is that there is a virtuous circle here.


  8. @Roger: "Either way, the way to expand energy access either is that we need massive amounts of cheaper energy or to make poor people richer. My view is that there is a virtuous circle here." I couldn't agree more.

  9. @Jonathon Gilligan, #5. Thanks for pointing out the Rosenfeld chart. To my knowledge, the values on this chart are for total electricity use divided by population -- in other words, consumption includes electricity delivered to commercial sites, businesses, government, etc. This is why the value is much higher than the median household use values we discussed before. I believe that the RECS numbers you quote are derived the same way, because they are equivalent to the 12,000 KwH per year per capita value on the chart. Ultimately, I guess the question is which value is relevant to IEA's analysis. I think in their case they were targeting household electricity use, so the 33 kWh per day per capita total US electricity use is not the correct metric for comparison.