The analysis of energy poverty has attracted increasing interest in some countries, including the United Kingdom, Ireland, Austria and New Zealand. Thomson and Snell (2013) examine the EU case as a whole. These studies have provided empirical evidence suggesting that households with some members over 60 years of age, families with children, disabled or chronically ill persons are the most vulnerable groups (ITD, 2001, pp. 8-9, cited in Boardman 2012: p 23) when it comes to energy poverty. The reason for this lies in the fact that their energy costs are higher than other basic needs (O' Neill et al., 2006). Empirical evidence also suggests that energy expenditure is essential; In fact, households could be considered as a "captive demand" affected by market control decisions - pricing - and this has severe social effects.

The relevance of this problem is twofold. Firstly, because an adequate temperature in the home ensures well-being at any income level. Secondly, because high energy costs could reflect low energy efficiency in buildings, which aggravates poverty situations. Reducing the energy bill does not necessarily imply a cold environment when buildings are energy efficient, a condition that could guarantee both lower energy costs and an adequate temperature if this problem is addressed to eradicate it. The latter relates energy poverty to the energy efficiency of buildings - a key element of EU energy policy to ensure the medium-term sustainability of cities in the European Union. If solutions are found to reduce fuel poverty problems, a twofold objective would be achieved: (a) to reduce energy consumption through a more balanced energy consumption scheme in buildings; and (b) to improve the health and welfare levels of disadvantaged households by reducing energy cost payments based on lower consumption. Incentive policies for investment in rehabilitation are the most widely accepted as they improve energy efficiency and reduce energy poverty.

The literature does not contain evidence that measures the sensitivity of energy poverty on changes in poverty levels or that assess the impact of property rates on energy scarcity. Economic logic supports the idea that a sudden fall in income can reduce the purchasing power and could have different effects on energy poverty levels depending on the type of tenure. 

In this paper, an indicator is calculated that identifies energy poverty in households using the Household Condition Survey (EU-Silk) for Spanish region, combining differently available indicators that allow an approximation to this phenomenon. It takes into account the structure of housing tenure and the level of poverty to explain fuel poverty. 

The present paper adds empirical evidence of the existence of fuel poverty using Spanish statistics to test two hypotheses. Ho1 is whether and how (housing) deprivation signals are linked to fuel poverty; whereas Ho2 tests the role of fuel poverty as an element directly related to poverty. They both allow us to support the Boardman (2012) views about the group of households affected the most by fuel poverty. This paper hypothesizes that a household is poor in energy if it cannot pay the electricity bill with the available resources. This implies that households whose income is close to a precarious without necessarily reaching the poverty line are considered (as there is no information on the cost of the electricity bill per household, it is not possible to apply the 10% rule nor to analyze whether households fall below the poverty line after paying the electricity bill).

The situation of the energy poor is approached by parameterizing a set of variables that capture the existence of energy poverty. The methodology used is based on an extraction of factors that include all these variables and capture their explanatory capacity to explain the situation of energy poverty. The factorial calculation is that the method extracts the common behaviour of these variables and combines them (linearly) in such a way that a variable is constructed that is an explanation of individual facts related to the analyzed phenomenon. 

Results suggest that the extracted factors combined with housing tenure characteristics, determine the existence of three main groups of fuel poor: (1) energy poverty derived from the status of' poor household', (2) energy poverty derived from the inadequacy of housing to household and (3) energy poverty derived from the low quality of housing. Using these factors in their standardized form, households are classified into two groups: those who suffer and those who do not energy poverty.