Putting VEPC to the test: Is shopping around really worth just $45?

When grand statements are made about the value of engaging in the retail electricity market, they are generally made by one of two kinds of people: Those who claim to know something about the market, and those that don’t. The latter category tends to be the domain of commentators aiming for effect. The former category tends to revolve around data.

The Australian Energy Market Commission (AEMC) publishes annual competition analysis that looks at the difference between standing and market offers, and the range within. It largely relies on market data, and some information retailers voluntarily provide to aid the analysis. The Australian Competition and Consumer Commission (ACCC) similarly publish market performance data, using their strong and broad powers to compel retailers to open their books, allowing an insight into each component of the cost stack that was previously opaque.

Yet a recent report released by the Victorian Energy Policy Centre (VEPC) entitled Do Victoria’s households leave less money on the table when they switch electricity retailers? utilises a unique dataset of almost 50,000 energy bills, described by the authors Bruce Mountain and Stephanie Rizio as the largest dataset of this nature ever analysed. As a result of this analysis, Dr Mountain told the ABC’s 7.30 program that he gives the market an “F” rating, because “it has been around for 10 years and hasn’t evolved to a standard that is delivering for customers”.

Given such grandiose claims, it is incumbent on us to take a deep dive into the results, to identify if the conclusions are warranted, and whether this research is in fact insightful, or produced by somebody aiming for effect.

What does the report set out to prove?

The main aim of the VEPC research is to ascertain whether customers who ‘switch’ get better outcomes than those who ‘remain’, aiming to challenge the standard contention that engaging with the market will deliver customers substantively better outcomes than if they were to disengage. A better outcome is defined as a cheaper deal.

To prove this hypothesis, the authors develop a concept they call “money left on the table”, or MLT. This is an important element of the report, distinguishing it from other analysis by the AEMC or ACCC who often scrutinise market data to better understand price dispersion. MLT is different from price dispersion, because it focuses on the difference between the absolute cheapest offer for that customer, and the offer they are on. Dispersion on the other hand compares the worst possible offer with the best.

The VEPC analysis aims to compare the MLT of customers who had “switched” in the past 12 months, and those who had not.

Here is where the rubber hits the road. With only one data point for each customer, the authors attempt to draw conclusions as to what has happened in the 12 months prior to the bill. They use that singular data point to infer whether the customer has engaged in the market, how much their annual bill might be, and how much they could have saved by switching to the best generally available offer on 31 August 2018[1].

What does the report claim to find?

The research found that customers who ‘switched’ had an MLT of $45 lower than those who ‘remained’. This saving equates to 3 per cent of the total bill for the typical customer. Customers as a whole typically have an MLT equivalent to 20 per cent of their bill.

Interestingly, the report shows that customers are most likely to switch to a tier 1 retailer or Red Energy, despite the offers they end up on typically being significantly higher than the cheapest offer available at that point.

The end result – the commonly accepted bifurcation between the engaged and disengaged barely exists. This bifurcation is a mere 3 per cent, and barely covers the efforts it takes to bother engaging.

Why? The report doesn’t aim to provide a silver bullet. Maybe it’s because the market is too complex. Maybe comparison sites aren’t suggesting the best deals. Maybe customers ‘lack the necessary PhD’ needed to engage. On the other hand, and admittedly less likely, customers might not value price as highly as we first thought.

So does it succeed?

In short, no.

The report does deliver a good snapshot of what was the state of the market in the second half of last year. What it clearly shows is that customers were not generally finding themselves on the absolute cheapest offers for their circumstances. This is a known problem, and one that the many reforms implemented post this dataset have sought to mitigate.

But when it comes to proving the hypothesis that customers who engage get materially better outcomes than those who don’t, it fails on a number of really important points.

1. We don’t know why the customers ‘switched’

The authors use the 12 month consumption history shown on a bill as a proxy to determine whether the customer is engaged. If the consumption history is shorter than 12 months, the customer is classed as a “switcher”, if 12 months history is shown, they are a “remainer”.

But this is too simplistic and the most important failure of the research. Yes, it shows that the customer was not with that retailer at that premises for 12 months, but it is unable to show what led to the new account being set up. The customer might have chosen to switch retailer, they might have moved house (and either stayed with their existing retailer, or switched to a new one at the same time), or they might have built a new house. All of these scenarios would present as the same on the bill. 

These scenarios provide some insight as to why some customers might not have searched out the absolute cheapest deal in the market. In short, merely suggesting that on average, those who had commenced with a retailer for whatever reason, were only $45 better off than those who had been with their retailer for more than a year tells us nothing. If we could separate these customer groups into sub-categories for the event that led to the ‘switch’, we might be able to gain genuine insights into why customers do not end up on the cheapest deals.

2. We don’t know what offers the customer saw when they ‘switched’

As noted above, the authors only compare the bills with offers available on 31 August 2018. Again, we know that the end result is that we don’t end up with all customers on the cheapest offer, but we are no closer to understanding whether active engagement with the market delivers you markedly better outcomes than disengagement.

The market evolves daily. The best offer is often cyclical – in some cases, retailers only make particular offers available to new customers for a short period. This isn’t necessarily a bad thing. If for example, a retailer gains access to cheaper than market supply, they might offer a cheaper deal until it is exhausted. The authors do not suggest that they are able to provide insights into the information customers had on hand before making a decision, but do infer that customers pass over cheaper products for other more expensive ones. Choosing one particular date for the best offer might not change the overall picture of the research, but it does diminish the ability to ascertain whether a customer who tried to find the cheapest possible deal was successful.

3. The data is now out of date

In any ordinary circumstance, scrutinising data from the most recent completed financial year would be more than timely enough to provide meaningful results. 2019 is no ordinary year.

As the Victorian Energy Minister has repeatedly said, the reforms that took place on 1 July 2019 were the most significant reforms to hit the sector in a decade. The Victorian Default Offer has been introduced, retailers are now required to advertise on their bills how much customers could save by switching to their cheapest offer, and disclosure requirements have been enhanced significantly to require retailers to advise customers at point of signup if there might be a more appropriate deal for their particular circumstances.

As such, any inference drawn from this dataset only goes to reinforce historical problems in the market that we have already sought to solve. Unfortunately, the report fails to mention any subsequent reforms, unhelpfully preferring to tell consumers that engagement is a waste of time.       

Can we explain the small difference between ‘switchers’ and ‘remainers’?

Not definitively, but the structure of the retail market, and what we have learnt about consumer behaviour provides a few clues.

The report found that the majority of customers switched to one of the tier one retailers, or Red Energy, and in doing so had a very high MLT. We know that a significant proportion of ‘switches’ are the result of moving house - 2016 census data showed that 16 per cent of Victorians moved house in the preceding year. When dealing with the hassle of moving, busy consumers might opt to stick with their existing provider or use a utilities connection service such as Direct Connect or Connect Now, prioritising ease over price. Given these retailers have the largest customer bases, and in many cases have direct relationships with utility connection services, the high ratio of ‘switchers’ they win is not surprising.

We also know that customers value brand recognition, and often consider their retailer is giving them a good deal. When a customer does choose to actively engage, and uses a comparator site, they might skip through the companies they don’t recognise until they find a known provider they feel might be less ‘risky’. This results in a higher MLT, and represents a consistent challenge for smaller retailers. They must offer more than just the cheapest price to compete with the more established players in the market.

So what should we take out of this report?

In the last financial year, customers on average, paid more than they needed to, particularly if they failed to engage. Those who didn’t switch, could have lowered their costs by 20 per cent, if they had switched to the cheapest possible deal. That’s good to know – it highlights that work needed to be done to develop practical means to help customers obtain a good chunk of the available savings.

We can also take out that we do not currently have adequate data to truly understand customer experiences over time in the market, but we do know that not all customers end up on the cheapest deal. This tells us we need to encourage engagement, but also continue to simplify offers, to help customers find the right deal. As market reforms occur, Government comparison sites need to continually evolve, using the best available technology to ensure that customers get what they need out of these valuable resources. Less transparent commercial sites should not be incentivised to fill the gap.

At the same time, stakeholders such as the VEPC, the media, industry and government need to promote the value of engaging in the market, and help their customers do it effectively.

We certainly should not be telling customers to give up on saving money on their energy bills because it’s all too hard. The 1 July reforms are significant – let’s work together to ensure they deliver the cheaper bills customers deserve.

[1] This date, according to the authors, represents the median end date of the bills in the dataset. On 31 August, the governments comparison site was scraped, and the available offers at that point in time were used as a proxy for the offers the customers had available to them. In effect, the MLT is the difference between the best available offer on 31 August 2018, and whatever price the customer pays in the bill VEPC is analyzing.