Price discrimination and the modes of failure in deregulated retail electricity markets
Preprint
Repository URI
Repository DOI
Change log
Authors
Abstract
In Australia, as with Great Britain, governments have shown rising concern with the health of competitive residential electricity markets. A core concern is the practice of price discrimination and the rising dispersion of prices. The State of Queensland implemented Full Retail Contestability in 2007, but held a regulated price cap in place until 2016, when it finally deregulated its residential electricity market. Almost simultaneously, the two jurisdictions that pioneered retail price deregulation, Great Britain and Victoria, were questioning their prior policy decision. Queensland makes for a fascinating case study because Southeast Queensland comprises a fully deregulated retail market while Regional Queensland is a regulated monopoly – with common input costs across both zones. Consequently, a regulated monopoly with a uniform tariff and 640,000 customers forms a very large control group, which can be directly compared to the competitive market of more than 1.3 million customers – making such analysis globally unique. Analysis of Queensland market conditions concludes the policy is welfare enhancing. To be clear, rising electricity prices are a problem, but price discrimination is not. The deregulated competitive market is, perhaps unsurprisingly, better at regulating the overall average tariff and consumer welfare has been enhanced by $184 million per annum – with some consumer segments very materially better off. However, certain modes of failure remain, viz. an inter-consumer misallocation problem and lack of transparency vis-à-vis the anchoring of discounts – known as the “discounts off what?” problem. Resolving the inter-consumer misallocation problem is relatively straight forward via ensuring energy retailers (voluntarily) move vulnerable customers onto a Benchmark-equivalent or suitably discounted tariff. Due to the non-linearity of tariffs and the rising mix of discrete metered loads, the latter can be best solved by producing a weighted average of Standing Offers, and using this as the benchmark.