rss
7

The RET is a high cost way to procure renewable energy

Print Friendly

In this note we outline a few simple reasons why the current structure of the renewable energy target produces high cost outcomes for producers and buyers of electricity. We then point to a competing “reverse auction” model that is already producing better outcomes for consumers and its sponsors.

The decarbonisation of Australian generation over the next 20 years will be expensive. Probably the gross cost will exceed $60 billion in total. It’s important to Australia that this investment be undertaken in a cost effective manner.

We think the RET target and method should be scrapped and replaced by the Clean Energy Finance Corporation setting out three-five year annual targets of new renewable supply it will let contracts for under a reverse auction process and then requiring retailers to buy their share of that output from the CEFC at the CEFCs’ weighted average cost.

We think this would provide much more certainty to the generation industry both thermal and renewable and produce significantly lower cost outcomes to consumers.

We start by looking at a list of renewable projects committed from around the time of the latest change to the RET target. On our numbers only 293 MW represent projects where a retailer has committed to a PPA, and this includes from Ergon, which has an additional State Government mandate, and we have previously discussed some peculiar features of the Origin deal with the Clare solar farm.

It’s true AGL has its big fund in the wings. There is a bit less than 600MW of new renewables, mostly wind, which will go towards the RET target but is being built on a “merchant” basis. The capital providers do not at this stage know who they will be selling the output to or at what price. Clearly that is a higher risk proposition, requiring a higher price to make it worthwhile. That means building renewable energy is risky, when it shouldn’t be. It’s as simple as that.

leitch updated updated

Renewable energy is mainly about capital cost and the cost of capital

The cost of different forms of electricity generation are usually compared within the industry using the long run marginal cost [LRMC], except that in this industry it’s commonly refered to as the levelised cost of electricity [LCOE].

The LRMC or LCOE is the real electricity price required to make the NPV of a project zero. In other words it takes the capital cost, the forecast opex each year, the maintenance capex, the interest and tax and finds the electricity price in $/MWh that will make the NPV zero, (revenues = costs).

In this way the differing capital costs and operating costs of say a nuclear or brown coal plant can be compared with a solar or wind plant. Usually a project is assessed over a 20-25 year life time, however this doesn’t have to be the case.

In general, large amounts of time are spent discussing the technical characteristics of the plant (capital cost, fuel cost, maintenance, outages, capacity factor). Generally speaking much less time is spent discussing what discount rate to use.

However, whereas the capital cost of solar and wind has continued to come down, it arguably hasn’t come down as much as interest rates. 10 year bond rates in Australia are about 2.2% today. In the USA the 10 year bond rate is 1.7%. If you run that through a WACC calculator you are going to end up with a WACC of 4%-6%.

We have used an equity beta for a renewable project of between 0.7 and 0.9. The beta is a measure of project risk. Books can and have been written about beta. We just content ourselves with saying that virtually every industrial company in Australia has an observed geared beta of less than 1 and resources companies more than 1. The observed beta for low risk gas and electricity network companies or toll roads is down around 0.5-0.7.

renew2

Just think about it. In the USA, if a windfarm owner can get a bank to provide 70% debt finance for a project, and with revenue certainty of a 15-20 year PPA that is likely, then the required discount rate is just around 3%. This is an incredibly low number compared to what we think the industry commonly uses as a discount rate.

For instance ,when ARENA call for proposals for its recent $200m solar PV round, the downloadable spreadsheet required a discount rate of 10%. In my opinion any estimate of the LCOE produced using a discount rate of 10% is going to be categorically wrong.

ARENA may get away with it because it was only comparing solar projects which were mostly similar, but even for comparing single axis solar with plain PV using a discount rate of 10% is likely to lead to error. It’s like saying the solar panels cost $2 watt instead of $1 watt. Yet, as far as I know, WACCS far higher than can be justified by any reference to market rates are routinely used in Australian project analysis. And, as a result, bad decisions get made.

So point one of the argument is that in our view most of the industry probably use a discount rate that is too high.

However, it’s Point 2 that really matters.

Point 2 is that the less confident you are about your revenue forecasts, the higher the WACC will be and therefore the higher the product price has to be to justify the project.

Let’s consider a wind farm that runs for 25 years. It has a capital cost of $2.4 m MW, a capacity factor of 35% and operating expenditures that average out to $20 MWh. In our view those are typical numbers for wind farms such as the Sapphire project in Glen Innes. Then let’s consider two discount rates 4% and 8%

If the discount rate is 4% then the real electricity price required is just about $70 MWh.

The one assumption I didn’t mention was inflation of 2.0%. The table below shows the calculations.

renew3

On the other hand, if the WACC is, heaven forbid, 8% then the wind farm needs a real price of $100 MWh. That extra $30 to some extent just a matter of opinion has to be paid for by consumers. Further more it may be the difference between a project going ahead or not going ahead.

The REC scheme might as well be called the high risk scheme

In my opinion the REC scheme, although it has some redeeming features, like bankable certificates is not well suited to the needs of the renewable industry. In some ways its typical of schemes designed by economists, and excuse the sneer. Another example would be phase 1 of the EEC carbon scheme. What’s wrong with the scheme?

1 It ends in 2030. Wind and solar PV take about two years from the time ground is broken to be up and running. Not long but it will be 2019 for any new project. That means just 11 years of REC revenue for projects with 25 year lives.

2 No one knows what the price of REC’s will be in the future, but history has shown the price can vary greatly, from about $35 per certificate up to about the current level of $83 per certificate. It is true that most business face price uncertainty but in my view the scheme has created unnessary price uncertainty. For instance when rooftop solar first became popular the assumed renewable energy resulted in REC creation, totally flooding the market. Rightly or wrongly, large utilities were successful in having the rules changed to remove future rooftop renewable energy included.

3 It takes no account of the low variable cost of renewable energy. In most industries its conventional for production in the short term to be driven by short run marginal cost. So for instance by and large a coal generators won’t produce if the revenue received is less than the fuel cost. This lets an industry build up a cost curve, and generally speaking means that supply of a product is curtailed gradually as the price moves below the highest short run cost, then the next highest etc.

But wind and solar have zero short run cost. So they will keep running and producing renewable certificates once they are built. So in theory if even 1 MW too much of renewable energy is built the rec price falls to zero and all the suppliers suddenly become losers. It probably wouldn’t happen exactly like that, because of contracts and other market frictions but there is a real risk of persistent very low prices if too much supply is built.

4 The biggest risk. The scheme probably will be changed again. The ALP has announced a 50% target. The current scheme won’t be anywhere near enough to achieve the required amount of decarbonisation to get to our COP 21 “promise”. Virtually every retailer we know expresses concerns about the “sovereignty” risk when considering PPAs.

5 Liable companies, principally large retailers, incur low penalties for non compliance. If they don’t buy enough certificates they pay a penalty price and just pass that straight on to their customers. Mostly, they pass it onto large business and industrial customers because those guys are lower margin and so the retailer doesn’t value their business as highly in the case that some other retailer does have enough certificates.

6 Existing players (those that have written PPAs in the past) or merchant wind farms, are today, for the first time in years, making good profits. Those are put at risk by new supply, so what’s the incentive?

Reverse auctions – a much better idea

Fortunately, the ACT Govt has shown everyone else how to do it. They use the “reverse auction” formerly known as a “tender” process. Under this scheme they invite participants to bid in their best price. The price is guaranteed to the project sponsor over 20 years by the financially secure, low risk ACT Govt.

Safe in that knowledge the banks are willing to lend lots of capital at their “best” rate, the equity holder takes a lower but safer return and the consumer gets a lower price. In essence, the ACT Govt has taken on the risk, but it’s achieved the lowest cost for doing so. There are risks with reverse auctions. If the project capital costs are too high, the project sponsor may walk away. Those risks are easier to manage.

Using the Clean Energy Finance Corporation [CEFC]

In some ways, although a bit less free market there is really no reason why the CEFC couldn’t fulfil the same role in the NEM as the ACT Govt does. The CEFC would set out a 3-5 year target of the amount of new renewable capacity to be bid for each year. Wind and solar PV suppliers would bid into the reverse auctions. Retailers would buy their certificate obligations from the CEFC at the weighted average cost. As decarbonisation increases over the next decade it’s a simple matter for the annual new renewable target to moved up.

Under this method the entire industry would know the minimum amount of new generation capacity that could be added each year. This would allow the entire system to be planned out years in advance. Thermal generators would be able to see the new supply coming and make their own plans with more confidence.

Because of the implicit Federal Govt. guarantee the bidders into the auctions would be relatively confident of being paid the guaranteed price over 20 years and so the cost of capital is low. In turn the CEFC can borrow at just 1% above the bond rate. The CEFC is already supplying finance to utilities in one form or another, for instance providing a $100 m facility to ORG for its household PV business.

Consumers would benefit by being supplied renewable energy at low cost.



The renewable energy industry would benefit as the build up of regular supply year after year will enable Australia to move down the cost curve at a steady rate. Australia has already shown with rooftop PV that given its chance the industry can be competitive on a global scale. Our installation costs for rooftop PV are way less than in the USA.

David Leitch is principal of ITK. He was  formerly a Utility Analyst for leading investment banks over the past 30 years. The views expressed are his own. Please note our new section, Energy Markets, which will include analysis from Leitch on the energy markets and broader energy issues. And also note our live generation widget, and the APVI solar contribution.  

Share this:

  • James P

    There are two main issues I see with the reverse auction, 1) the costs incurred in preparing the bids are substantial and not recoverable with unsuccessful bids. And 2) the consumers may receive lower bills but the taxpayer is also paying for the guaranteed/subsided price per MWh.

  • alexander austin

    These WACC numbers seem very low compared to industry norms. One question I would have is the tax adjustment for interest costs. Given the large upfront costs of most projects – which give rise to high depreciation claims – most projects dont pay tax for a long period.

  • Ian

    Now that renewables development is back on the agenda a discussion on how best to pay for the deployment is very beneficial indeed.

  • Tommyk82 .

    We’ve certainly seen the LGC market can be too rocky to provide investor certainty, but STCs are great. Even with the LGC price double, businesses prefer the bird in their hand than a revenue stream likely to be cut short.

  • This proposal is smart, nimble and improves on the ret – lets get this on the policy agenda. Is there value in focusing on the cefc – or would it be better to let state governments just get on with it like the act has?

  • AllanO

    David

    You might want to correct the first table which shows “Clare Solar Farm” as a wind project and “Sapphire wind farm” as PV.

    One issue to think about for a reverse auction process with a centralised buyer is how the electricity output is valued. Simple LCOE / LRMC style metrics completely ignore the fact that differences in energy production profile and dispatchability of different technologies (and even of identical technologies in different locations) can lead to very different values in the electricity market for nominally the same total quantity of MWh produced. As I read your proposal, the centralised buyer would ignore this in tendering for renewable projects and the ultimate buyers – either retailers or consumers would have to wear the differences. Modifying the auction rules to try to account for this would not be at all straightforward. Under the current LRET scheme, either the PPA buyer or the merchant project proponent is forced to consider how to value this element of market risk.

    For example, it would be very hard to see how a solar+storage project offering some level of dispatchability would be competitive against a basic solar array style project, even if that dispatchability substantially enhances the value of the former project’s electricity output. Or how the buyer would reject a “low-cost” wind project in a NEM region flooded with similar projects already depressing spot prices in windy periods, in favour of a slightly higher cost project in a non-saturated region where price and wind output were not so negatively correlated.

    You could modify your proposal so that the centralised buyer purchases only the “green” component of the output (effectively LGCs) leaving project proponents to deal with valuing the “black” (electricity) output – but then you are back much closer to something looking like the LRET, and the risk premiums and discount rates attached to the overall project economics would certainly have to rise.

    • David leitch

      Allan

      You make a good point. I did focus on one simple metric, lowest bid in price and did not value other attributes. As well as despatchability I am increasingly thinking that capacity factors have hidden value. The higher the capacity factor the less backup or storage is required. For this reason single axis PV may be more useful than fixed PV even if the first pass LCOE is lower. In principle the auction rules could allow for these factors but at the cost of complexity.

      Many hands make light work.
      Cheers