What are we talking about?  Here's the published definition of fuel poverty:

The most widely accepted definition of a fuel poor household is one which needs to spend more than 10% of its income on all fuel used to heat its home to an adequate standard of warmth. This is generally defined as 21ºC in the living room and 18ºC in the other occupied rooms - the temperatures recommended by the World Health Organisation.
The UK Fuel Poverty Strategy

An important point to make is that the definition is based on needing to spend rather than actually spending more than 10% of income, recognising many poor households don't heat their homes to this comfortable level due to other essential needs.

The Energy Consultation states that in 1996 it is estimated there were around 6.5 million (out of ~22m) UK households in fuel poverty with 5.1 million of those falling into the category of vulnerable (homes with children, the elderly, sick or disabled) leaving 1.4 million non-vulnerable.  This total had fallen to 2 million by 2003 with 1 million described as vulnerable.  The reasons for this fall are interesting though, the consultation states that 60% of the decrease has come through increased incomes - increased through government initiatives targeting the vulnerable households with Pension Credit, Winter Fuel Payments and the Child Tax Credit.  The reaming 40% came from decreased fuel prices and efficiency improvements.  Given that fuel prices fell substantially in real terms from 1996 to 2003 I think it's safe to say efficiency improvements had little more than a marginal impact.

It's worth noting that that non-vulnerable level of fuel poverty had only fallen from 1.4m to 1m (29%) whilst the vulnerable level had fallen from 5.1m to 1m (80%). The difference is that the vulnerable group received the bulk of the extra government income where the non-vulnerable improvement was mostly due to falling price.

(The text of the consultation says 5.1m vulnerable households in fuel poverty 1996 yet the graph shows 4m - can't explain the inconsistency... Assuming the 6.5m total is correct considering the 4m vulnerable and therefore 2.5m non-vulnerable would change the percentages to 60% non and 75% vulnerable. I don't that that's right though.)

Progress to 2003 had come mainly from government buying fuel for households in fuel poverty and falling fuel prices.  Things started to change in 2003 as fuel prices started rising dramatically.

This recent graph from the current consultation illustrates the swing in 2003 as fuel prices increased.

Click to enlarge (source: Energy Consultation)
An important point to note about this graph is that it shows  a fuel poverty forecast after economic effects of income and fuel price movements.  The remaining 1.1 million in 2010 suggest that further policy intervention is required to meet the target of zero vulnerable households in fuel poverty in 2010.

The thing I find most surprising about this graph though is the forecast.  We have seen how increased income (through government policy change) and cheaper fuel resulted in the dramatic fall from 1996 and increasing fuel prices from 2003 caused a rise.  For fuel poverty to fall after economic effects without further policy change as indicated fuel prices must have to fall considerably - starting now.

Domestic fuel prices are not falling – in fact we have recently seen very large domestic price increases this year:

Npower: 32% gas, 29% electricity
Powergen: 24% gas, 18% electricity
British Gas: 22% gas, 22% electricity
Scottish & Southern: 14% gas, 12% electricity
EDF Energy: 15% gas, 5% electricity
Scottish Power: 15% gas, 8% electricity
(Source: BBC Online)

The variations just reflect when increases have ocured, Npower for example didn't increase prices at all in 2005 and have since increased twice this year in January and March.

My biggest criticism (and doesn't just apply to this graph but almost all graphs which show historical reality and future projection) is that the present, this year, is described as an inflection point. The trends are presented as changing right now.

I see no evidence to suggest that there is anything special about 2006 with respect to fuel poverty; I expect the levels of fuel poverty to increase. The government expects the opposite yet don't justify this extraordinary claim.  This is another example of government making future projections based on assumptions that don't consider peak oil.

Their approach of subsidising fuel ahead of enabling lower fuel consumption is also a questionable long term approach to fuel poverty.  It reminds me of the saying "Give a man a fish, and you feed him for a day. Teach a man to fish, and you feed him for life."