EMI reports updated – Switching reports now include switch types (trader, move-in, half-hour)

  • Last post 25 July 2016
Matthew Keir posted this 30 May 2016

We’ve augmented the EMI website's switching reports to include a filter on switch types. This enhancement provides further insight into the switching activity that takes place in the retail electricity market. 

The three switch types are ‘trader’, ‘move-in’ and ‘half-hour’. Trader switches follow the switch process that applies when the customer of an electricity retailer (customer) has an existing contract with the losing trader at the ICP. Move-in switches follow the switch process that applies when the customer is new to an ICP. Half-hour switches are switches of ICPs with larger metering categories. More information is available in the report notes.

The augmented retail reports are: switching trends, switching breakdown, switching summary, and relative switching.

Looking at the switching trends report we observe the following:

  • Trader switches peaked in 2011 when the What’s My Number campaign began and show a strong seasonal correlation to the winter months, which is also when this campaign runs.
  • Move-in switches have steadily increased and are the main contributor to total switching increasing over time. These results align with our surveys which show that moving house prompts a residential customer  to consider their electricity supply choices. An increasingly competitive market structure underpins this increase, as do customers with both awareness and increased choice. Switches are a count of traders supplying ICPs changing, not customers changing traders. Some move-in switches will be loyal customers who decide to sign up with their previous supplier at their new ICP.

Examples of insights that can be drawn from the augmented reports are highlighted at the following links:


Order by: Standard | Newest | Votes
msouness posted this 15 June 2016

Worth noting that the EMI switch reports switches at the month of initialisation, rather than the month of completion.  

msouness posted this 19 June 2016

On another note, I wouldn't rely on the switch type as an indicator of moves vs straight switch events.

daringodber posted this 19 July 2016

I've been looking at the comparison on switch type trends (Meridian included this in their last monthly operating statistics report). I'm interested in the trend of 'move in' switches increasing over time and what are the drivers behind this. Some thoughts were:

- Increased number of retailers combined with increasing market share for new retailers means statistically move-in are likely to increase over time (although I would think this affect is relatively small as the big 5 still make up 91% of market share)

- Are people moving house more often (MBIE's bonds data seems to suggest this isn't the case)

- A classification issue - are all switches being correctly classified? There seems to be some large differences between retailers in terms of %age of switches which are move ins vs trader switches

- Similarly, some retailers show pretty large increases in move ins over last couple of years but steady levels of trader switches.


Anybody have any thoughts on this?

Matthew Keir posted this 20 July 2016

Here are a few ideas for starters:

  1. You might consider market structure regionally rather than nationally. While a large portion of each market might be held by the four largest retailers operating there, there will be a large difference for the number of move-in switches generated between a structure of 85%, 5%, 5%, 5% and equal 25% shares (as an example) with a certain level of house moves. It should be relatively straightforward to determine a mathematical relationship between house moves and HHI www.emi.ea.govt.nz/r/0thdj, market size, and move-in switch counts www.emi.ea.govt.nz/r/5tfjp.
  2. You might also consider people purchasing houses rather than just renters moving although bond data is a good idea. I read somewhere that the turnover of houses in Auckland had increased markedly – I think it was presented as an average length of time between sales of the same property. Does anyone know of some stats available somewhere (especially stats for all regions) that could shed some light on this?
  3. We know moving is a catalyst for switching and some retailers might be very successful in marketing to new home owners, movers, or perhaps students flatting at universities who “move- in” often. This may explain disproportionate statistics. However, retailer strategies can change all the time. The example in the original post clearly shows a correlation to the university calendar and move-in switch counts in Wellington and Dunedin.
  4. That said, customer details are not held in the market systems. There are differences in the Code around how winning and losing retailers agree final meter reads for each switch process and saves protection clearly won’t apply to move-in switches. A classification issue would be difficult for a regular participant audit of any single participant’s system to pick up. However, where a losing retailer presented evidence of a breach of the Code - our compliance team would investigate.

Let us know how you go.

Does anyone have any other ideas?

daringodber posted this 25 July 2016

Thanks Matthew, I was actually thinking the other day I may be underestimating the impact of market structure by not looking at it at a regional level. The move in switches and HHI do show a strong relationship.

The Dunedin move in switches is also very interesting and shows a clear pattern of increased switching over the January period.