Solar tariffs and the merit order effect: AGL Lobbying Masquerading in Academic Robes

AGL Solar Feed-in Tariffs and Merit Order Paper: Lobbying Masquerading in Academic Robes

A recently released paper by AGL’s chief economist, its head of economic policy and sustainability, and an AGL energy market analyst continues to distort the truth about Australian solar power by masquerading lobbying activity as a formal academic paper.

Many of this website’s readers will recognise such actions to be consistent with those of the vested interests of a utility whose business model is threatened by a popular disruptive technology.

Tim Nelson and Paul Simshauser’s lobbying activities have succeeded before, to the point where the authors infer their work is linked with the scuttling of solar industry tariffs in other states. It’s not that their latest paper, “Queensland solar feed-in tariffs and the merit-order effect: economic benefit, or regressive taxation and wealth transfers?” doesn’t make any good arguments,  it’s just that its supposed academic qualities are repeatedly undermined by opinionated language, a lack of impartiality, and generally poor rigour of calculations.

This is not the first time Nelson and Simshauser have created a faux-academic paper to influence the less well informed. Their earlier paper (with Simon Kelley), “Australian residential solar feed-in tariffs: industry stimulus or regressive form of taxation?”, was criticised by many solar academics for being unscientific, opinionated, and counterfactual.

In personal correspondence, the authors noted the many shortcomings I pointed out, but published their paper without satisfactorily addressing its failings. (Thankfully it didn’t make it into in any highly prestigious journals, and even its eventual publisher labelled it controversial, but this didn’t diminish its impact on less well-informed readers).

In making the argument that PV policy support represented a regressive form of indirect taxation, the authors claimed (on the basis of unsubstantiated, internally-created private evidence) that mostly wealthy families were installing PV, despite evidence to the contrary – evidence both direct and indirect. For example, the graph below shows the correlation between PV uptake and affluence, based upon ORER and ABS data, and is supported by direct evidence: income surveys of solar households that formed part of government rebate applications.

Source ORER, ABS as presented in Solar Progress

Nelson and Simshauser’s evaluation of feed-in tariffs are crafted on their own terms (and I quote): “Based upon an assessment against our public policy criteria, it seems clear to us…” The authors unilaterally decide to measure justification for feed-in tariffs by the extent that solar power reduces network expenditure.

This ignores the multiplicity of government-communicated reasons for feed-in tariffs (most of which relate to building a solar industry and providing householders with a simple way of reducing their electricity bills). Instead, the authors decide that people working during the day and returning home to use electricity at night is sufficient reason to abandon feed-in tariffs as a policy measure. It’s easy to win an argument when you invent the rules as you go along.

Nelson and Simshauser declare all premium feed-in tariffs to be bad because they lead to industry growth (the objective of such policies) followed by industry contraction – i.e. boom and bust. However, industry contraction occurs not because of the inherent nature of feed-in tariffs, but due to political response to poorly designed policies that cause cost blowouts.

In lobbying for removal of feed-in tariffs, they wish to simply throw out the baby with the bathwater, rather than consider how to raise a child. To make matters worse, their latest paper has to resort to incorrect assumptions, inaccurate calculations, dubious methodology, and misleading tables in order to make a case that the poor are $10/year worse-off due to feed-in tariffs.

In fact, it can be shown that because it is predominantly lower-income households that have bought PV systems, the welfare of this demographic group has been improved by $40-$80/year per household on average, and those who remain are shielded from the $10 average cost of the policy by Queensland’s $230/year pensioner electricity discount. In summary, supporting PV installations helps lower income families to reduce bills, funded by the better off families that don’t mind paying a few dollars a year more.

AGL’s chief economist lends his weight to the existence of the merit order effect. However, it is argued that the reduction in electricity prices is temporary, as long-term prices reflect average costs of production rather than instantaneous costs (which are zero for solar power). The authors state as incontrovertible fact (cheekily referencing themselves as the source of this ‘fact’) that renewable resources are “unambiguously more expensive than thermal plant,” which they reason should thus increase average prices in the long term, using wind farms as an example.

The fallacy of this argument (even before health and environmental externalities are considered) is that grid parity has been reached: at the point of consumption, distributed PV is now cheaper than thermal plant. This means a PV-specific merit order effect will occur in tandem with sustained substitution of centralised generation by point of consumption PV. Even if it was only a hypothetical example in their minds, at least Nelson and Simshauser concur “If a merit order effect occurred due to natural market disruption events, such as the entry of a new renewable or low emission technology with lower marginal running costs and lower total costs than incumbent generators, it would produce an unambiguous improvement in welfare.” Until you can set up a 10kW coal-fired generator in your backyard, PV has lower total cost at the point of consumption.

This is not to say that Nelson and Simshauser don’t make some good, if somewhat moot points. The Queensland feed-in tariff should avoid a boom-bust scenario that has hurt NSW. PV doesn’t cost-effectively reduce distribution network peak demand (though PV combined with energy storage can easily be a more valuable solution than network upgrades). It’s just a shame that their arguments are tarnished by misleading language, assumptions, calculations, and appearance.

Though posing as an academic paper, their calculations lack scientific rigour. And their language is clearly not that of an impartial academic: “excessively generous”, “too successful”, “surely make for sobering reading”, “It has been surprisingly well documented in academic literature”, “unambiguously more expensive”, “an oversupply of higher cost renewable capacity was purposefully engineered”, “would represent courageous assumptions at best”.

One could conclude from reading this paper that employees of a utility whose business model is threatened by a popular disruptive technology are seeking only to minimise the damage through a poorly disguised lobbying exercise. The paper notes that the real wealth transfer resulting from the merit order effect occurs from producers (generators) to consumers. The threat to, and of, a major utility is barely hidden, claiming premium feed-in tariffs “raise tangential doubts about the entire sustainability of the [electricity] industry.”

Perhaps the work of Nelson and Simshauser can best be summed up in their own misleading language: “has a certain superficial appeal, despite the obviously erroneous nature”.

(Please note that the author has nothing personally against the authors of the papers mentioned, but instead is acting to defend the solar industry from misleading lobbying efforts).


This article originally appeared in RenewEconomy.

Some references:

Nelson, T., Simshauser, P. and Kelly, S. (2011) “Australian residential solar Feed-in Tariffs: industry stimulus or regressive form of taxation?”, Economic Analysis and Policy, Vol. 41 No. 2, 113-129

Assumes the of the ‘$72m’ worth of feed-in tariff payment, one-third goes to the 309,000 households in the <$31,000/year income bracket, and one third to the 592,000 households in the $31k-71k income bracket.