Is HumeLink worthwhile? The $5 billion dilemma facing the energy regulator

Very soon the Australian Energy Regulator (AER) will determine whether to approve the $5 billion HumeLink transmission project, the most expensive in Australia’s history.

The AER has previously approved $0.6bn for early works expenditure and now has before it TransGrid’s Application to approve the remaining $4.4bn for construction.  

HumeLink is the proposed 365 kilometre, double-circuit 500 kV overhead transmission line from Wagga Wagga to Bannaby (near Goulburn) via Snowy 2.0.  Initially called SnowyLink North, HumeLink’s role is to provide transmission services for Snowy 2.0, future solar and wind generators in southwest-NSW, and new interstate links to Victoria (VNI-West) and South Australia (EnergyConnect).

There is considerable pressure on the AER to quickly approve the application given the project’s fervent support by both the federal and NSW governments and the Australian Energy Market Operator.

But can the AER, which according to its website “exists to ensure [energy] consumers are better off now and in the future”, be confident that HumeLink is worthwhile, as consumers will be bearing the cost? Also, can the AER be sure that HumeLink will lead to efficient outcomes consistent with the National Energy Objectives.

Let’s put ourselves in the shoes of the AER executives and look at some of the perplexing issues they face, especially in the context of HumeLink’s history of underestimated costs and exaggerated claims.

In 2020 TransGrid estimated that HumeLink would be completed by 2024-25 at a cost of $1.35 billion.  That was for two single-circuit lines, which is equivalent to about $1bn for a double-circuit line as now proposed.

Two years later the cost soared to $3.3bn, prompting a warning from AEMO that “the rising project costs … cannot materially increase from the current estimate of $3.3bn. Further work to drive down costs should be undertaken urgently.”  Regardless, the cost has since driven up, not down, another $1.7bn. 

Worse still, the latest blowout is effectively another 14% higher due to HumeLink’s capacity dropping from 2570 MW initially to 2200 MW now.  So the increase in cost/MW has been almost six-fold in four years!  How did TransGrid get its costings so wrong (on a par with Snowy 2.0’s guesstimations) and can the AER be confident of no further blowouts?

A contributing factor to the cost increases has been TransGrid’s obsession with attempting to complete HumeLink within an unachievable timeframe. 

The latest target of July 2026, though two years later than the initial aspiration, is still considered to be impossible given the incomplete route planning, lack of social licence from local communities for overhead lines, uncompleted easement agreements, material shortages and tight workforce resources.  

There is no need to frantically attempt to build HumeLink in time for an outdated Snowy 2.0 commissioning target.  Snowy 2.0’s latest completion forecast is December 2028, but the actual date is likely to be years later.   

And why is TransGrid’s completion target more than three years before AEMO’s optimal timing of 2029-30? 

Most alarmingly, TransGrid revealed in its Application that, in order to try to meet its target, it entered contractual arrangements that “require the AER’s Determination … by 29 March 2024 to avoid incurring significant delay penalties under the D&C contract.  Such delays would increase the delivery costs of HumeLink, which would be passed on to consumers”.  

Why did TransGrid sign up to such contracts prematurely and set another impossible deadline, this time for the AER?  And why should the ‘significant’ penalties that have already been accumulating for more than three months be passed on to consumers?

One of the critical and most confounding issues to be considered by the AER is the claimed value of the project.  In 2020 HumeLink was estimated to have a net benefit of $1.1bn. 

Two years later, when the cost soared to $3.3bn, the net benefit plummeted to $39m.  At the time energy experts criticised the highly optimistic assumptions and dubious results, assessing the net benefit to be substantially negative.  Nevertheless, the AER granted approval to proceed.

Now TransGrid claims the net benefit has rocketed to $4.2bn above the $5bn cost blowout – an utterly implausible improvement, especially considering the 14% reduction in capacity.  AEMO’s latest (also optimistic) net benefit estimate is a more modest $1.6bn, and only $0.3bn higher than its previous estimate.  But how can it be that when the cost of HumeLink blows out its estimated benefit soars even more?

Another perturbing issue for the AER is the errors and biased assumptions, too numerous to list here, in favour of the preferred design option.  They are significant enough to put into question TransGrid’s conclusion that the preferred design option provides the greatest value (though all three options have negative benefits in my view). 

TransGrid continues to be dismissive of underground transmission, especially for a direct current (DC) backbone, when the application of this technology is preferred for many overseas grids for long-distance transmission. 

TransGrid has moderated its estimate for the extra capital cost down to two to three times that of overhead lines, but a detailed study by international experts found the extra cost for HumeLink to be about 1.5 times or less.  

And undergrounding would bring many offsetting benefits such as less environmental and agricultural impacts, no susceptibility to bushfires and weather-related outages and, most importantly, minimal opposition from landowners and local communities.  Gaining social licence for an underground HumeLink would avoid enormous angst and expenditure for both the communities and TransGrid.  

Undergrounding may not be appropriate in every location but it would in some, and should not have been dismissed for HumeLink (and every other proposed transmission line).  The delays in Snowy 2.0 give time for the underground option to be applied.

Another issue for the AER is that HumeLink’s claimed role is unachievable.  

HumeLink, with a capacity of 2200 MW, cannot provide almost 8000 MW of transmission services for Snowy 2.0 (2200 MW), plus future southwest-NSW renewable generators (3000 MW) and interstate transfers to Victoria (1900 MW) and South Australia (800 MW)? 

Obviously not all services will be required at the same time and in the same direction, but there will be periods of significant curtailment and inability to service all needs.  HumeLink will be maxed out whenever Snowy 2.0 is generating or pumping at full capacity, which according to its updated Business Case is estimated to average sixteen hours a day.  

Why hasn’t HumeLink been designed to service the expected transmission needs for the next decade or so?  The prospect of TransGrid attempting to gain social licence and seeking approval for a HumeLink 2.0 doesn’t bear thinking?

And of course Sydney Ring South is still needed to transmit HumeLink’s capacity from Bannaby to the Wollongong/Sydney/Newcastle load centre, adding another $1.5bn or so to the cost.

Most relevant for the AER, if HumeLink is approved it will result in NSW transmission tariffs increasing by around 50%, for just this one new transmission line (TransGrid’s current Regulatory Asset Base is $8.8bn).

Inexplicably, the major beneficiaries, Snowy 2.0 in particular, are not being required to contribute their fair share of the cost. 

This anomaly was raised at a recent NSW Parliamentary Inquiry, with the AER Deputy Chair Jim Cox responding, “I think there are worthwhile arguments to say that perhaps generators should be required to make a contribution to some of the costs of these links that may be of particular benefit to them, so that certainly is an idea in good standing.  But it’s not the way the system works at the moment”.

HumeLink may well be needed, if only to connect Snowy 2.0, but the best solution from a holistic engineering, social and environmental perspective is not being presented for approval.

HumeLink should:

– Have been designed with sufficient capacity to service all its foreseeable needs and avoid a HumeLink 2.0.

– Have been routed directly between Wagga Wagga and Bannaby, with a spur to Maragle (or alternately Lower Tumut Switching Station). This was one of the three credible options developed by TransGrid (termed Option 2C) and was estimated to have equivalent economic benefits to the proposed option.  Also, Option 2C is electrically superior, being without the proposed 100km dog-leg deviation to Maragle that reduces HumeLink’s capacity, increases losses, and increases susceptibility to bushfire and weather outages.

– Not be rushed for completion by 2026, at huge additional cost.

– Not have entered into contracts prematurely with impossible deadlines, incurring significant delay penalties.

– Possibly have been DC rather than AC, as the first link in a Brisbane-Sydney-Melbourne DC backbone.

– Have been undergrounded partly or fully to minimise impacts on local communities and the environment and achieve some measure of social licence.

– Not be paid for entirely by electricity consumers. The major beneficiaries, Snowy 2.0 and the new REZ generators, should be paying their fair share (the majority) of the cost.

It is unfortunate that that the AER has before it a sub-optimal application. The pressure is now on the AER to finalise its determination and ‘ensure consumers [and the wider community] are better off now and in the future.