Energy and the Earth

Energy and the Earth

Yes minister, but …

In a piece published in the Guardian recently, Environment minister Greg Hunt was purported to have once again rejected claims that a price on carbon would benefit emissions reduction targets. This time because, to quote,
“emissions fell by six times the rate in the five years before the carbon tax than they did under the carbon tax”.

The claim was addressed in a FactCheck by Dr Peter Christoff published here on the Conversation, for which I was reviewer. The FactCheck focused on the legitimacy of the minister’s claim in terms of what emissions should be referenced when assessing the efficacy of the carbon tax.

Subsequent comments on the FactCheck reveal just how blinkered attitudes towards the carbon tax have become. As the issues are more complex than I was able to cover in my 100-word review, my purpose here is to flesh out some of the nuances in more detail.

The key point that seems to be missed, be it by proponent or critic, is that even though the carbon tax did achieve very substantial reductions in emissions across the sectors to which it applied, those reductions were to large extent a charade.

With each passing day it is becoming clearer that the emissions reductions that accompanied the carbon tax were unsustainable.

The background to the minister’s claim

Between mid-2006 and mid-2012, the rate of deforestation in Australia slowed dramatically. As a consequence, deforestation emissions reduced almost 60%, equivalent in total to almost 50 million tonnes CO2-e per year, according to the National Greenhouse Gas Inventory (NGGI) reported by the Department of the Environment [1].

NGGI deforestation CO2-e emissions in million tonnes aggregated by financial year. Horizontal coloured lines show means for 3 different periods: mid 2002 to mid 2006 (red), mid 2007 to mid 2012 (green) and mid 2012 to mid 2014 (blue). The red box shows the period carbon was taxed in some sectors of the economy, though not the forestry sector. Image by Mike Sandiford. Data source listed in note [1].

Across the five year period prior to the carbon tax, deforestation emissions reduced at an average rate of 15% per year, falling by around 20% in each of FY’s 2007-08 and 2008-09.

NGGI deforestation CO2-e emissions data as percentage change by financial year. See first figure caption for details. Image by Mike Sandiford.

In mid 2006, deforestation accounted for about 15% of total NGGI emissions, about the same as the transport sector. By mid 2012 it accounted for only 8%.

Changes in the rates of deforestation are reflected in the changes in the total emissions across all sectors of the economy, as neatly illustrated by the general correspondence between the figure above (showing rates of change of deforestation emissions) and the figure below (rates of change of emissions from all sources).

Total emissions data including LULUCF (land use, land-use change and forestry) as percentage change by financial year. Note that the rate of decline in total emissions in the five years before the carbon tax was about 6 times the rate during the two years of the carbon tax. See first figure caption for details.

Note that the correlation extends into the carbon tax era, when substantial increases in the rates of deforestation emissions are reflected in a reduction in the rate of decline of total emissions (from -1.68% to -0.26% per year).

It is this slow down in the rate of decline of total emissions that forms the basis of the minister’s claim.

An apology of sorts

And of course, the minister is technically correct - the numbers reported by his department do show that the emissions reduction was about 6 times faster in the five year period prior to mid 2012, compared to the period of carbon pricing commencing mid 2012, when summed across all sectors reported by NGGI (at least within the limits of their accuracy [2]).

Either because the way it was reported, or because the minister thought the technical details would be rather too much for the average punter, or because the minister did not want it, the Guardian report does not explore the relationship between deforestation and changes in total emissions.

Consequently, the impression is the minister would have us believe that taxing carbon was, and will forever be, an abject failure.

As I attempt to explain below, the carbon tax may have been a failure, but not because it didn’t reduce emissions fast enough.

Rather, if anything it failed because it reduced emissions too fast.

In reality, astonishing emissions reductions during the short tenure of the tax were achieved across the sectors directly impacted, and especially the electricity sector. But it is now clear that hydro power generators gamed a system vulnerable to the vicissitudes of extreme policy uncertainty. The emissions reductions achieved in the first two years of the carbon tax could not have been sustained into a third year, even if the carbon pricing experiment survived. By the second half of 2014, there simply was not enough hydro generation capacity left to do so.

Almost all sides of the debate seemed to have missed this point, including the minister. All along we have been blind to the veritable storm of new emissions coming our way that will dwarf any plausible headway we might have contemplated under our carbon pricing regime, or will plausibly do under direct action.

Some explanatory details

First a point of order. The objective of the carbon tax was to produce sustained emissions reductions over the medium and long term. But it could and should only reasonably be judged by its impact in the sectors to which it applies.

Just as the minister should and would not be held to account for the successes or failures of the policies of his cabinet colleagues, neither should the carbon tax be held account for what happened to deforestation, since its jurisdiction did not cover the so called “land use, land use change and forestry” sector - or LULUCF. Nor do it apply to the agricultural or transport sectors.

If we subtract these sectors from the total NGGI emissions then we arrive at a very different picture from that insinuated by the minister.

NGGI CO2-e emissions data as tonnes per year by financial year for the sectors impacted by the carbon tax.(i.e. total less LULUCF, Transport and Agriculture).

As illustrated in the figure above, in the years of the carbon tax there was a significant step down in the emissions across the target sectors (namely the electricity and large industrial sectors as well as fugitive emissions).

For these sectors, the rate of decline in emissions went from negligible (-0.02% per year) in the five years prior to the carbon tax, to (-1.68% per year) in the carbon tax era. By the minister’s own logic, that would seem to imply an 80-fold increase in the rate of emissions reductions in the carbon-tax era compared to the previous five years.

NGGI CO2-e emissions data as percentage change by financial year for the sectors impacted by the carbon tax (i.e. total less LULUCF, Transport and Agriculture).

(Note it would be ridiculous to attribute any significance to the factor of “80”, since the uncertainty on the 0.02% value is much larger than 0.02.)

As I pointed out in my FactCheck review, the impact of the carbon tax was most dramatic in the electricity sector. The average annual rate of decline in the electricity sector over the carbon tax years, about 5%, points to a nominal 10-fold increase in the rate of emissions reduction compared to the prior five years (note that some of the reduction in the June Quarter of FY 2013-14 maybe due to changes in methodology [3]).

NGGI CO2-e emissions data as tonnes per year by financial year for the electricity sector.
NGGI CO2-e emissions data as percentage change by financial year for the electricity sector.

So why was the carbon tax a failure?

While such changes might suggest the carbon tax was particularly effective, the reductions were achieved substantially through unsustainable fuel switching in the electricity sector.

During the carbon tax era, emission intensive brown coal generation in Victoria, in particular, was substituted significantly by zero emission hydro power generation in Tasmania. With the electricity exchanges between Victoria and Tasmania effectively tracking the emission intensity of the national electricity sector, the flows of electricity on the Basslink Interconnector across Bass Straight provides the “smoking gun”.

Electricity flows on the Basslink Interconnector from 2013 through 2014, averaged by month. Note the change in flows in the second half of 2014, following axing of the carbon tax when the net flow switched from an net average of over 300 MW northwards to about 200 MW southwards. Analysis and image by Dylan McConnell (PhD student at University of Melbourne).

The levels of Tasmanian reservoirs clinch the case, demonstrating unequivocally that the emissions reductions achieved across the first two years of the carbon tax were unsustainable.

As first alluded to by Dr. Hugh Saddler, hydro reservoir capacity was being banked in the lead up to the carbon tax, as capacity was restored following breaking of the millenium drought in 2009. During the carbon tax regime, hydro capacity was substantially depleted as hydro power flowed northwards across Bass straight. Since repeal, power flows across Bass straight have reversed with dispatch from Tasmanian hydro generation reduced, to help restore depleted reservoir capacity.

Tasmanian Hydro reservoir levels as a percentage of total capacity. Reservoir levels peaked in FY 2011-12, and diminished over the two years of carbon pricing to levels not seen since the end of the millenium drought in late 2009. Yellow diamonds show the average levels across the relevant financial years. Red squares show the levels at the start of each year.

And the numbers are not insignificant. The differential in the energy sent out from Tasmanian generators between the second half of 2013 and the second half of 2014 amounts to around 500 megawatts.

Energy sent out from Tasmanian generators as weekly averages in grey and as average across the financial year in yellow diamonds. Tasmanian electrical power demand as red line. Note that sent out energy must also cover network losses. When demand exceeds sent out energy and associated network losses, extra electricity is imported across the Basslink Interconnector. Likewise, when sent out energy exceeds local demand and network losses, the additional energy is exported to Victoria. The dip in energy sent out in FY 2011-12, when reservoir capacity peaked reflects the banking of future emissions. The dramatic fall off in energy sent out in the first half of FY 2104-15 is our first payout on the emissions debt.

With that differential accommodated to large extent by a rise in Victorian brown coal, it will add an extra 5 million tonnes, or about 3%, to FY 2014-15 electricity sector emissions.

During carbon pricing years, hydro generators were betting the likelihood of either repeal, or that a move to a floating pricing would lower the carbon price. The rates of hydro power dispatch in the carbon tax years could not be sustained.

The implication is that the apparent large reduction in emissions was achieved in effect by “borrowing emissions from the future”, to quote Hugh Saddler. We are now repaying that debt, and would have had to even if the carbon tax had remained.

Data from AEMO broadly consistent with the NGGI data cited above, substantiates this point in revealing the shifts in electricity sector emissions on the National Electricity Market in the last half of 2014, since repeal. In the last half since repeal, emissions on the NEM have risen by a nominal 9%, although about ¼ of the rise may be attributed changes in the way AEMO calculates emissions [4].

Electricity sector emissions from the National Electricity Market averaged by half year (circles first half, triangles second half). Note that the nominal rise of 8.9% in the second half of 2014 should be reduced to around 6.4% to account for a change in methodology [4]. Data source from AEMO.

The elephant in the emissions room

While the merits of the carbon-tax continue to be debated ad nauseum, the real elephant in the Australian emissions room continues to be ignored. We have committed, or anticpate, an additional 70 million tonnes per year of emissions in the ongoing development of our LNG resources. For example, CO2 emissions expected from the INPEX Icthys LNG project alone will add an average of 7 million tonnes per year of CO2-e emissions over its 40 year life, or around 800 kilograms of CO2-e for every tonne of LNG shipped [5].

The staggering scale of these new emissions from the liquefaction and venting of co-produced CO2, all acknowledged in the impact statements for the various developments, make the ongoing debate around the pros and cons of the carbon tax almost irrelevant.

Could it be that both sides of politics have been content to obscure this impending storm by letting the debate rage at the margins?


Notes

[1] Data are sourced from Table1A: Quarterly Emissions by Sector since 2001-02, Unadjusted, in the Quarterly Update of Australia’s National Greenhouse Gas Inventory: June 2014, Australia’s National Greenhouse Accounts published by the Department of Environment, December 2014.

[2] Because “Processed satellite images are not yet available to support the (LULUCF) emissions estimates for 2013 and 2014 … the preliminary estimates for deforestation are subject to change and have a greater level of uncertainty than the other sectors in the national inventory.” See - Quarterly Emissions by Sector since 2001-02, Unadjusted, in the Quarterly Update of Australia’s National Greenhouse Gas Inventory: June 2014, Australia’s National Greenhouse Accounts published by the Department of Environment, December 2014.

[3] Note that for the June 2014 quarter the allocation of fuel consumed for electricity generation has been aligned with the International Energy Agency resulting “in an increase in emissions attributed to the electricity sector and a corresponding decrease to emissions reported in the stationary energy excluding electricity sector.” See - Quarterly Emissions by Sector since 2001-02, Unadjusted, in the Quarterly Update of Australia’s National Greenhouse Gas Inventory: June 2014, Australia’s National Greenhouse Accounts published by the Department of Environment, December 2014.

[4] Note that “From 1 June 2014, the emission data used in calculating the CDEII changed from estimated to actual emission data via the National Greenhouse and Energy Reporting (NGER) scheme. An impact assessment performed for 2012/13 financial year indicates that this new data source would have resulted in the CDEII result being on average 2.5% higher than the published CDEII. The change to data source should be taken into consideration when analysing the change to CDEII over time. Further detail on this change can be found below under” http://www.aemo.com.au/Electricity/Settlements/Carbon-Dioxide-Equivalent-Intensity-Index

[5] For the Icthys project this comprises 2.4 million tonnes per year from reservoir CO2 and 4.6 million tonnes from combustion. After 20 years when gas from the Brewster reservoir (~ 7% CO2 reservoir content) declines the Plover reservoir will operate at about 18% CO2. The emissions intensity for the project is likely to be about 830 kg of CO2-e per tonne of LNG produced. After 20 years it will reach 1000kg of CO2-e per tonne LNG. (this data compiled by Dimitri Lafleur -PhD student University of Melbourne).