Saturday, June 19, 2010

The Concept of Energy Poverty

The existence of energy poverty today is quite well accepted around the world. In fact alleviating energy poverty is a goal of many development organizations that deal with energy issues for developing countries.

Migrant Workers Cooking-India: Photo WB/Curt Carnemark
When it comes to defining energy poverty these organizations assume the position that many take in appreciating good art--"they know it when they see it." There is much talk about energy poverty but not much action in terms of measuring it.

There is a good reason the people avoid defining energy poverty. It just is very thorny to define. There even was a time not too long ago that development specialists refrained from using the term. One can ask several different questions concerning the definition of energy poverty. Is energy poverty the same as income poverty? Is energy poverty based on access to energy services such as cooking, communications or lighting? Or is it based on quantities of energy that people use? These questions have generated several different approaches to measuring energy poverty.
We have been working on ways to do this for about two or three years and there will be some papers coming out shortly. However, in the meantime we review some of the issues surrounding the topic. There are several different approaches to define energy poverty and they can be classified as follows and are explained below.
  • Minimum amount of physical energy necessary for basic needs such as cooking and lighting;
  • Type and amount of energy that is used for those at the poverty line;
  • Households that spend more than a certain percent of their expenditure on energy;
  • The income point below which energy use and or expenditures remains the same, implying this is the bare minimum energy needs.
Each one of these approaches has strengths and weaknesses. The first way of defining energy poverty has its roots in defining poverty as a minimum amount of food intake necessary to sustain a health life. This is a very common approach often used by international food agencies. Translating this method to energy, there also must be a minimum amount of energy necessary to cook, light and heat someone’s home. Although this might be true, pinning down the exact minimum level of energy necessary based on basis needs is very difficult due to the significant country and regional differences in cooking practices and heating requirements. We know the caloric levels that are necessary to sustain a healthy life, but pinning down the minimum energy needs is much more difficult to accomplish.

The second approach of defining the energy poverty line as the energy being used by households below the known expenditure or income poverty line is much easier to grasp. The expenditure based poverty line is well defined in most countries, so based on a household energy survey you assess the average fuel use below this level. This is fairly attractive because it is not necessary to actually measure how much energy people are using. However, this method also has the drawback that you are defining energy poverty based on more general critera as opposed on an energy basket of goods and services. This means that such a poverty line would not be based on the energy policies in the country, but rather would reflect general economic and social policies. Tracking energy poverty with this method would be no more than tracking general poverty trends. This method is not so useful for those that might want to tack the impact of energy sector reform.

The third line of thought is that the energy poverty should be based on the percentage of income spent on energy. It is well established that households that are poor spend a higher percentage of their income on energy than households that are wealthier. Empirical studies including ones I have done indicate that such percentages can range from about 5% or less to close to 20% of cash income or expenditure. It seems that when energy is above 10% of income, then conceivably it will begin to have an impact on general household welfare. The idea is that when households are forced to spend as much as 10% of cash income on energy they are being deprived of other basic goods and services necessary to sustain life. The drawback to this approach is that 10% is a rather arbitrary figure. So it suffers to a certain extent from the same problems as the methods based on physical measures of energy.

Rural Bangladesh End Use HH Energy Per Capita Per Month
by Income Decile.  Source:  Rural Energy Survey 2005
The last method is based on the level of energy demand as it relates to household income. The sweet spot for poverty specialists is the level of income at which the use of energy or the level of energy expenditures does not vary significantly with income (see Figure). The converse of this is that after a certain level of income people begin to consume more and more energy. For poor people this means that even if their income increases, their use of energy does not because they are at the bare minimum amount necessary to sustain daily life as illustrated in the figure. Unfortunately, this approach is quite data intensive and requires the analysis of household surveys that have good energy questions. The attractiveness is that this definition of energy poverty is based on how people actually consume energy based on local resource conditions, energy prices and policies.

Here are a few publications on this topic that are available in PDF form.   Many of the thoughts in this post are the results of research conducted with Shahid Khandker and Hussain Samad.  There are other articles available, but most are copyrighted, so if the authors have a PDF that can be published, I will list them. 

Energy Services for the Millennium Development Goals. Link to PDF
Energy Prices, Energy Efficiency, and Fuel Poverty. Link to PDF
Energy Access, Efficiency, and Poverty: How Many Households Are Energy Poor in Bangladesh? Link to PDF

So what do you think? Please take the poll and give opinions below. Keep in mind that this is a current topic of discussion and there are no correct answers.


Unknown said...

Hi Doug

A great posting as always! We're current working on a fuel poverty project in a rather different context: low income households in Romania.

One of the issues emphasized here is that there is a big difference in the terms 'energy poverty' and 'fuel poverty'. Energy poverty is seen as essentially a question of energy access, whereas fuel poverty relates more to fuel costs and household incomes.

Why not just define households that do not use improved stoves (we can define that perhaps according to emissions levels), do not have electric lighting, and, perhaps, do not have energy for communications as living in energy poverty?

Such a definition would make the definition more like the 'consensual model' definition of 'fuel poverty', and would focus attention on energy issues, and less on poverty issues. Otherwise it seems to me that if energy poverty essentially is the same as poverty poverty, it best way to address it is through poverty alleviation, rather than energy initiatives. If we have an energy poverty indicator, wouldn't it be best if it mobilized action addressing poverty issues related to energy?


Elizabeth Cecelski said...

Hi guys, have just read this rather late. I like Grant's suggestion which does address energy poverty not just poverty. Electric connection and energy for communications are easily measureable too; the improved stoves indicator tho is more difficult. Is there a proxy for emissions levels that could be used for this purpose?