PlanetNL20: Island Load Forecast Still Wrong
To sanction the Muskrat Falls project, Nalcor relied on
their development of a steadily growing electricity load forecast on the Island. The assumptions used have proven so weak that
in 2016 Nalcor decreased the total Island energy load forecast out to 2040 by
25%. That’s a major error for any
self-respecting utility to make. By
putting forward the 2016 revision, Nalcor has quietly admitted their prime
reason for Muskrat sanction, needing a lot of power soon, was untrue.
This posting will challenge why the latest Nalcor load
forecast still includes 10% net load growth leading up to 2040 instead of declining.
Does Nalcor believe this province has
insatiable electricity demand at any price?
Do they remain shockingly ignorant of global trends and energy
alternatives? It appears their revised forecast was a
significant step toward reality but not all the way there. The analysis also begs the question,
shouldn’t a proper pre-sanction load forecast have also pointed to declining
power needs instead of increasing?
Nalcor
Data Review
Below is a chart from an exhibit document recently reviewed at the Commission of Inquiry showing Nalcor’s Isolated Island
pre-sanction load forecast. Steady load
growth was predicted to occur despite their simultaneous acknowledgement of substantial
electricity rate increases, no population growth and an already high level of
saturation for electric space and water heating existing in the market.
Pre-Sanction Isolated Island Load Growth
Forecast
Nalcor would rely on this forecast in their Isolated Island
option to justify a series of costly generation developments including the
refurbishment of Holyrood, the build of many new CCCT and CT diesel thermal
plants, three new small hydro sites, and some increase of wind power. The heavy reliance on thermal power sent
forecast fuel costs skyrocketing to the billions annually. Little wonder it finished second best in the
two horse CPW race with Muskrat - Nalcor designed the Isolated Island scenario to
be too fat to run.
Jump forward to 2016 and 2017 when Nalcor released revised
load forecasts up to 2040 as part of then-new CEO Stan Marshall’s annual
updates on the Muskrat project. The
chart below indicates the 2040 total Island load forecast at about 7500 GWh, a
28% reduction from the pre-sanction forecast of about 10,400 GWh. No apologies were noted from the utility for
having made such a grievous error – no staffing changes, and no new consultants
were announced as performing their forecasting either. All was done “in-house”.
 |
| Nalcor CEO Stan Marshall |
Nalcor
Island Load Forecast Revisions 2016 and 2017
While the decline in energy sales toward 2022 seems to be at
least going in the right direction, if not steep enough given the threat of
doubling rates (an expectation reconfirmed in a new Cost of Service study
submitted by NL Hydro to the PUB on November 15), what could possibly be the
justification for projecting 10% load growth – averaging just over 0.5%
annually – between 2022 and 2040? Why wouldn’t
the load decrease trend simply keep continuing as would be consistent with
well-known price-elasticity behaviour?
Reviewing
Load Forecasts Elsewhere – No Growth
Prior PlanetNL postings have frequently referenced the US
Energy Information Administration for statistics and forecasts. The EIA 2018 Annual Energy Outlook provides a long-term reference forecast that indicates total electricity
consumption out to 2040 at a growth rate that matches population growth
estimates for the same period. US
electricity growth forecasts therefore are flat on a per capita basis.
A detailed United Kingdom energy forecast out to 2040 was
published by Bloomberg on November 21.
As seen in the chart below, total 2040 UK
utility-supplied energy needs are the same as 2015. In the UK case, there is zero net growth
despite projected UK population increase of 11% over this period. On a per capita basis, UK electricity usage is
projected to decline by 10% to 2040. Simultaneously
released by Bloomberg was an identical study on Germany that yielded a similar
declining energy per capita trend.
UK
Load Forecast 2015-2040
Take note also that nearly 20% of all energy in 2040 is expected
to be used by the emerging electric vehicle (EV) segment on the assumption EV’s
capture 50% of market share by then. The
lowest line on the chart therefore shows that traditional electricity
consumption needs (other than for EVs) is forecast to decline about 20%,
suggesting per capita electricity usage rates declining by 30%.
In all these major markets as well, the projected long-term cost
of electricity is flat (other than ordinary inflation) because the industry has
found that new capacity additions are tending not to increase utility costs and
can actually sometimes reduce costs.
Take for example the results of a recent call for bids to replace
coal-burning power plants in Pueblo, Colorado where the median bid price for
wind power was USD 1.8 c/kWh and wind with storage was just USD 2.1 c/kWh (footnote
1). This is an example being repeated in
many jurisdictions where thermal plants are yielding to wind and solar
renewables because the cost of the new project is less costly than the fuel
cost of the thermal plant.
Trends in the US, UK and Germany, G7 economic powerhouses,
cannot be ignored. People and business
are generally not expected to use any more electricity than they do today and
may likely use less.
Why then should Nalcor still be forecasting increasing
electricity usage on the Island?
Leading up to 2040, this province expects to double the cost
of electricity, has a stagnant or declining population, modest to weak economic
prospects, and electric vehicles are likely to be adopted here at a far slower
rate than elsewhere. Besides that, the
high saturation of electric heating usage gives both domestic and commercial consumers
considerable opportunity to decrease their electricity consumption through
energy substitution or high efficiency electric-heating alternatives.
Nalcor’s latest load growth forecast continues to be entirely
unrealistic and should feature significant and steady decline.
Pre-Sanction
Load Forecasts Should Have Been Far Lower
The pre-sanction Isolated Island forecast is subject to much
the same assumptions as presented here for 2018. Yes, we all know more about power generation
alternatives and costs in 2018 than we did in 2012 and some non-hydro renewable
technologies have definitely become much more affordable. But to the true experts in the energy
industry, none of these developments have come entirely out of the blue. Utilities are expected to have expert staff
or hire the professionals capable of assessing such long-term trends with
reasonable accuracy. Predicting the
short-term trends, such as what might happen in the space of only a few years,
should especially be far easier. No
utility, including Nalcor, could reasonably say they didn’t see these changes coming
in just six short years.
This comment is made not to blame particular individuals in
Nalcor middle management who perform forecasting work. It was the executives responsible for a
multi-billion dollar project that needed to assess their risk and ensure they had
obtained the best quality advice possible before proceeding. Those executives could have mitigated their forecasting
errors with professional external input but they chose not to. As Nalcor’s own revised load forecast has eliminated
80% of the load growth essential to justifying sanction of Muskrat, there is simple
and clear evidence of their disastrously poor judgement.
It’s becoming plausible, however, that not only should the
Isolated Island load forecast have had a lot less growth, it might well have naturally
tended into a steady decline toward 2040 and beyond without even needing much
of a push. With a forecast of load
decline, not only is Muskrat infeed power not needed but neither is any of the
series of new small hydro developments, additional wind developments and
certainly not all the fuel-guzzling combustion turbines as Nalcor had included
in the Isolated Island scenario.
Next
Post
With Nalcor’s first reason for sanctioning Muskrat – need
for power – clearly proven untrue, their second objective can be evaluated: to
eliminate the emissions and costs associated with the Holyrood Thermal
Generating Station. This objective has
merit but could have been solved in a much more efficient way than the brute
force high-cost generation replacement alternatives proposed by Nalcor.
In the next post a model is
proposed in which Holyrood energy output could have been eliminated without
substitute energy requirements. The
analysis also predicts electricity rates would have remained stable and
low. Relatively simple policy changes
were available but completely ignored by Nalcor and Government.
Footnote 1 – Colorado data
sourced from a Nov.22 article on The Narwhal that will interest many Uncle
Gnarley Blog readers