Monday, July 19, 2010

Boundary Review: A Chance to Restore Equity?

A key component of the proposal to reduce the number of MPs to 600 will be a fresh boundary review and maybe this is the time to seize the opportunity to restore some equity into the system.

At the moment there are two distinct sets of rules covering boundary reviews – one for local government in England (Wales, Scotland and Northern Ireland have similar but distinct rules) and one for parliamentary reviews. Again, the rules for Wales, Scotland and Northern Ireland are different as they have a guaranteed minimum number of seats regardless of population.

The local government rules require local councils to produce figures for the electorates of each polling district or parish based on a five-year forecast.

The parliamentary rules are based on actual electorates at the date that the review is ordered. The ‘new’ 2010 boundaries were, therefore, drawn up based on electorates as at 17th February 2000. Ten years of actual population movements are totally ignored.

Parliamentary boundaries use local government wards as their building blocks so it has been the custom to complete a local government review before embarking on a parliamentary review.

However, even though a local authority’s ‘new’ wards were created to take account (on a five-year forecast basis) of known housing developments, estimated patterns of depopulation and similar factors, these are all ignored as the parliamentary review uses actual electorates on a given date.

And this is where the equity in the system needs to be addressed.

Because parliamentary constituencies use a snapshot of the electorate on a specific date, and the whole process takes several years before implementation, the electorate in any given constituency can vary widely from the theoretical quota of electors per constituency.

The electoral quota for parliamentary constituencies in England is (currently) 69,934.

A rough analysis suggests that there are currently 44 English seats with an electorate of 62,941 or less (that is a maximum of 90% of the electoral quota) and 42 English seats (excluding the Isle of Wight) with a population of 76,927 or more (a minimum of 110% of the electoral quota).

So nearly one fifth of the ‘new’ 2010 English constituencies are already 10% or more away from the electoral quota and these variations will only get worse as time passes and more people move from inner-cities to suburbs and market towns and from north to south.

Surely it is time for the parliamentary boundaries to be based on fresh five-year forecasts for each ward so that when the new constituencies come into being there is a chance that the disparities between electorates are minimised.


Tim Roll-Pickering said...

Again, the rules for Wales, Scotland and Northern Ireland are different as they have a guaranteed minimum number of seats regardless of population.

Iain can you stop repeating this myth regarding Scotland. The minimum rule was struck out several years ago.

And the number of Welsh seats is less due to the minimum number and more down to the "ratchet" effect in the rules.

What's actually needed here is a formal "apportionment" process similar to the one in Australia whereby an absolute number of seats are assigned to each state/nation at the outset, and the state/nation level reviews can only produce that exact number.

Paul Halsall said...

I don't have any very strong opinion on this subject, except that I think it is important for constituency boundaries to match in some way other ward and local authority boundaries.

We should try to avoid the weirdly shaped districts which do not overlap at different levels of government and which mean the council members, US reps, state reps, state senators, etc, do represent any "natural" communities.

Libertarian said...

All of this doesn't bring democracy any closer.

Until we directly elect the Executive there is no democracy. What size the Boroughs are doesn't stop them being rotten boroughs

richard.blogger said...


When we have a government who is planning to get rid of the most accurate and detailed survey of the population how can you expect the Boundary Commission to make an accurate decision of the Westminster boundaries? (2011 census is 'safe' but 2021 is likely to be cancelled.)

Redrawing boundaries is a lengthy and difficult process because it involves a lot of consultation - which is right, people have to feel confident about the process. Redrawing the boundaries is likely to be far more expensive than the census. As a big-state-is-better advocate I welcome such an action, but what about you small-statists, how do you salve your conscience?

Demetrius said...

Does nobody realise that the concept of the "constituency" is hopelessly out of date in the 21st Century? Then when you try to equalise territory you come up with very odd results. If you want to produce representation that truly reflects the political map you have to have some form of proportional representation from quite large areas. That way we might get some real balance in the House of Commons instead of some of the gross distortions that have bedevilled politics and the making of key political decisions.

Unknown said...

Again, the rules for Wales, Scotland and Northern Ireland are different as they have a guaranteed minimum number of seats regardless of population.

Pace poster #1, regardless of the 'rules' the fact is that Scotland, Wales and N Ireland all have a greater number of parliamentary seats in proportion to their population than has England.

This iniquity may have had some logical basis many years ago, but today, when all have their own assemblies/parliaments/what have you there is no valid reason for it. I very much hope that this is put right as part of the boundary review.

Sceptical Steve said...

The existing system takes a snapshot of the population every ten years, in much the same way as it's been done for over a century.
The fact that we only do it once every ten years tells you how cumbersome the process is and, almost by definition, by the time the figures have been collated, they're too late to be any use (except to genealogists!).
Please cut the Government some slack on this point.

Lord Blagger said...

The real issue, are the issues and not which MP gets elected.

If we all get an equal vote on an issue, they the question of representatives becomes a side issue.

Tim Leunig said...

Why not weight an MPs vote in the HoC by the number of adult constituents they have? Then the Isle of Wight's MP has a bigger say than any other, because they represent more people.

Harry Hayfield said...

I am trying to figure out Wales (but am not helped by the lack of electorate data for 2010) but have so far determined that Wales should lose 11 seats

Total UK Electorate 2010 = 45,429,735
Average Electorate for 600 seats = 75,716
5% range = 71,930 to 79,502

County Breakdown
Clwyd 331,525 / 75,716 = 4.37 = 4 seats
Dyfed 279,972 / 75,716 = 3.69 = 4 seats
Gwent 348,532 / 75,716 = 4.60 = 5 seats
Gwynedd 180,376 / 75,716 = 2.38 = 2 seats
Mid Glamorgan 393,121 / 75,716 = 5.19 = 5 seats
Powys 102,601 / 75,716 = 1.36 = 1 seat
South Glamorgan 332,608 / 75,716 = 4.40 = 4 seats
West Glamorgan 293,034 / 75,716 = 3.87 = 4 seats
Total 29 seats

Lithgae Dave said...

The constituency size issue is a red herring. The reason that the system is weighted in favour of Labour is

a) the Labour vote is distributed more efficiently, being concentrated in particular areas.

b) Turn-out is higher in non-Labour safe seats than in Labour safe seats.

Look at the situation in Scotland. Labour got 70% of the seats in Scotland with 42% of the vote. This was achieved not because Labour having smaller constituencies but because the Labour vote in Scotland in concentrated in West Central Scotland and is weak in other parts of Scotland. By contrast Tories and the SNP do very badly under FPTP as their vote is more evenly spread.

Alister said...

I live in Scotland, and given that we now have our own parliament, surely any argument about minimum seats is null and void! If any thing we should have less seats as a lot of powers are devolved?
We need a full debate on this, it can be summed up in the "West Lothian Question".

You just have to look at the size of the smallest Scottish constituency of 34000 voters (basically the 3 smallest Scottish don't add to the Isle of Wight). Personally I'd like to number reduced further to 500 which would bring the English average to 90K meaning that the Isle of Wight wouldn't be so under represented. Plus you could probably half the MPs from Scotland

Anonymous said...

Dave isn't quite right. Constituency size wouldn't make much of a difference if the variation in size was random, but it isn't: the smaller seats tend to be Labour seats.

The best thing would be a rolling review, so that seats can be adjusted as and when necessary, rather than wasting time and money reviewing the whole lot every so many years.