We know the two great certainties in life are death and taxes, but many thought there was a third: the inexorable rise in consumption of electricity. As the population grew and each of us got a little more prosperous each year, we'd use more power. The mighty electricity industry was built on that certainty.
Except that electricity consumption has been falling for the past four years. To say this has taken the industry by surprise is an understatement. For well over a century - even during the Great Depression - the quantity of electricity used in Australia each year was greater than the year before.
It took the industry and its regulators two or three years to accept the trend was more than just a hiccup on the ever-upward path, which delay probably added to the decline.
There are few aspects of the economy - global or national - where change is more significant, more diverse or more interesting than energy supply and demand - where energy covers coal, gas (conventional and unconventional), petroleum, wind, solar and other renewables. Expect to hear more from me on the topic.
But there are few questions more interesting than exactly why the unthinkable, a fall in electricity consumption, has come about. Short answer: a surprisingly large combination of reasons, although Tony Abbott's crusading against the carbon tax must get some of the credit.
The best attempt to quantify the various factors involved comes from a report prepared by Dr Hugh Saddler, an energy expert with the Pitt and Sherry consultancy, for the Australia Institute. Saddler's modelling covers the years to 2012-13, but we know from reporting this week by Origin Energy and AGL that the fall continued in 2013-14.
Saddler focuses on energy produced and consumed from the National Energy Market, which covers the five eastern states and the ACT, but the decline is occurring also in Western Australia. After peaking in 2008-09, consumption from the national market in 2012-13 was down by almost 8 terawatt hours, or 4.3 per cent.
But that's only half the story. Just as important as why demand has fallen is why it hasn't continued growing, as continued growth in the population and the economy would lead us to expect. Saddler estimates that had demand continued growing from 2004 at its average rate of growth over the previous 20 years (2.5 per cent a year) it would have been 37 terawatt hours more than it actually was in 2012-13.
This shortfall is equal to the output of almost 5000 megawatts of coal-fired generation capacity, the combined capacity of the Bayswater and Eraring power stations in NSW, or Loy Yang A and B and Hazelwood in Victoria.
''All of the decline in consumption has been at the expense of coal-fired generators, with the result that many are now barely profitable,'' Saddler says.
Greenhouse gas emissions fell by 9.2 megatonnes of carbon dioxide equivalent, about 2 per cent of Australia's total annual emissions.
So what has caused our power consumption to fall rather than rise? The biggest single reason is the introduction from the late 1990s of regulations to increase the energy efficiency of refrigerators, freezers and many other residential and commercial appliances, and to increase the energy efficiency of new buildings.
Saddler estimates this explains 37 per cent of the 37 terawatt-hour shortfall from what might have been.
The next biggest part of the explanation is structural change in the economy away from electricity-intensive industries. Over the year to September 2012, three major NSW industrial power users - Port Kembla steelworks, Kurri Kurri aluminium smelter and the Clyde oil refinery - were partly or completely shut down. This explains 10 per cent of the 37 terawatt-hour shortfall.
The evidence also suggests that power consumption by other major industrial users has been little changed over the three years to 2012-13. Saddler estimates this failure to grow explains a further 14 per cent of the shortfall, taking the total contribution from structural change to almost a quarter.
The next most important part of the explanation is the response of electricity users, particularly residential users, to the higher prices they were being charged. Saddler finds that after 2010 there was ''an abrupt change in consumer responsiveness to higher prices''.
This was the time when the possible effect of a carbon tax on electricity became a major political issue thanks to the efforts of Abbott and his ''sceptic'' mates. At the time, retail electricity prices were rising spectacularly, mainly because of a huge increase in spending on upgrading the transmission and distribution networks (poles and wires) to cope with an expected ever-rising peak demand on hot summer afternoons.
Saddler finds evidence to support his argument that all this carbon-tax-related fuss about the high cost of electricity caused many households to be a lot more conscious of what was happening to their power bills and to respond by finding ways to cut their usage - to the extent they ''have managed to completely offset the effect of higher prices on their household budgets by reducing consumption''.
This highly unusual jump in the short-run ''price elasticity'' of electricity explains 19 per cent of the shortfall, he estimates.
He further calculates that the growth in output from rooftop photovoltaic solar and other small, distributed generators accounts for about 13 per cent of the shortfall. This, of course, is a fall in the demand for electricity supplied by the major, mainly coal-fired generators, not a fall in the use of electricity as such.
Saddler notes that for the past three years the annual peak demand has been falling, not increasing, despite the huge investment to cope with ever-rising peaks. When will this additional capacity, which is now built and for which all electricity consumers are paying - and will continue to pay for some years to come - be required, if ever, he asks.
Ross Gittins is Fairfax Media's economics editor.