Evidence from Scotland and Wales: representatives elected via party lists are less likely to reply to constituents – but we should be careful about the conclusions we draw

Websites such as WriteToThem.com make it simple for constituents to contact their elected representatives, but how responsive are politicians to these communications? And does the system by which they are elected shape their responses? Research by Alex Parsons and Rebecca Rumbul shows that under the UK’s Additional Member Systems, constituency representatives are more likely to reply than those elected via party list. This raises further research questions about the subjects they reply about and different priorities for representatives under this system.

Picture credit: mohamed_hassan / 3058 images, via Pixabay/Public Domain

WriteToThem.com is a mySociety project that makes it easier for people to write to their representatives at different levels of government. The basic problem it solves is that many people don’t know who their representatives are. WriteToThem uses a postcode search function to give users a list of all their elected representatives, and provides a simple mechanism to write an email to one (or several) of them. You search, write, and then it’s up to the representative in question whether they write back.

We were interested in how responsive MPs were to this constituent contact. We therefore surveyed people a few weeks after they had emailed to see if they had received a response. In the past we have used these numbers to make tables of individual responsiveness. Currently, we are trying to understand the systematic drivers of responsiveness: what factors in a representative’s position or background make them more likely to respond to communication from constituents? This is useful for making comparisons between individual representatives fairer, and it also helps us to understand more about different attitudes towards constituent communication. Do some kinds of representatives see it as a more important part of the role than others?

We started by looking at three devolved bodies in the UK: the Scottish Parliament, the National Assembly for Wales (NAfW), and the London Assembly. These all use the Additional Member System (AMS) to elect their members. Briefly this means that the majority of representatives are elected directly from constituencies but voters also get a second vote for a party rather than an individual. This second vote is a mechanism for increasing the representativeness of the assembly by party: the additional seats are distributed to make the total distribution of seats match as closely as possible how people voted in the second vote.

This means these bodies are home to two kinds of representative – those elected from constituencies and those elected from party lists. These have the same rights in their respective bodies, but seem to behave differently. In the Scottish Parliament for instance, list-elected representatives spend more time on committee work, while constituency representatives in the Scottish Parliament and National Assembly spend more time in their constituencies. The explanation for this is that their re-election depends on different things: one set need to keep their eye on keeping their constituents happy; the other set have their position on the list decided by the party – and so have more incentive to do work that benefits the party.

We wanted to know if these two kinds of representatives responded differently to emails through WriteToThem – was there a gap in the responsiveness rate between the two groups?

From 2005 to the end of November 2017, we had 20,388 survey responses for the Scottish Parliament, 5,399 for the NAfW and 5,247 for the London Assembly. Using these survey results we found in all three cases constituency representatives replied to a greater percentage of messages. For the Scottish Parliament and NAfW we had enough data from other sources to control for other explanations. These included if the representative currently belonged to a governing party, their gender, age, length of time in the role, and how many emails they were receiving from the platform.

The effects of these factors varied between the two bodies, but the Scottish Parliament and NAfW models were in agreement in that even after accounting for these factors constituency representatives were more responsive – with a 19.48% (95% confidence that this falls between 15.34 and 23.62) difference in the Scottish Parliament and a 18.62% (95% CI: 6.94, 30.30) difference in the NAfW.

For the Scottish Parliament we can go further. There have been a number of individuals who having been elected via the party lists have subsequently been elected in a constituency. When we limit our analysis just to these people in their different roles we still see a difference in responsiveness, with a rate 17.06% (95% CI: 9.74, 24.39) higher in their constituency role. The exact same people have a different responsiveness rate in different roles, demonstrating the strong effect the role has in shaping this behaviour. Interestingly, there was no such consistent difference for the (smaller) group of MSPs who start out representing constituencies and were re-elected via list when they lost their constituency seat.

Here is the nice, simple interpretation of all of this: constituency representatives have more incentive to reply to constituents, and so do so more often. Here is the complication: we don’t have the data to know what people are actually writing about. It could well be that these two kinds of representatives are getting completely different types of emails. If a constituency representative is getting a lot of emails about the personal circumstances of a constituent, they might be very well placed to write back with guidance. If list representatives are getting more partisan emails about policy (that are less straightforward to address to the respondent’s satisfaction), this might explain why these are getting fewer responses. We can’t at the moment know if they are responding to the same kind of messages in different ways, or are being treated differently by constituents in the kind of messages they receive.

Even assuming the results are exactly what they appear, it is also not clear whether this is a good or a bad thing. While a headline-grabbing reaction might be: ‘List-elected representatives don’t answer their emails!’, it is worth remembering that they are answering lots – just fewer in comparison to constituency representatives. A representative’s role has many aspects – is it bad if a minority see their committee and legislative work as more worthwhile than replying to 20% more of their emails?

What these results reflect is that elected representatives have different priorities, but also that these different priorities aren’t random but a response to the structure of the role. If we have opinions about what those priorities should be, we can shape the job by how we design the roles – and where there are multiple opinions on the subject, a little diversity in how representatives approach their job is no bad thing.

This article gives the views of the authors, not the position of Democratic Audit. For more details, see the accompanying journal article ‘Did They Write Back? A Mandate Divide in Response to Constituent Casework in Devolved Bodies’, published in Parliamentary Affairs. Other mySociety research can be found at: https://research.mysociety.org/

About the authors

Alex Parsons is a Research Associate at mySociety. He holds a MSc in Democracy and Democratisation and tweets at @alexparsons.

Rebecca Rumbul is Head of Research for mySociety. She holds a PhD in politics and governance, and has worked actively as a researcher for many years. Previously managing the work of the Wales Governance Centre and lecturing in Politics at Cardiff University, Rebecca was responsible for the development and establishment of mySociety’s research agenda, and has spoken widely on the impacts of civic technology both at practitioner and academic conferences.