August
22, 2007
Electoral
Boundaries by Region Show Error
"…we
have concluded that we should not treat these rural challenges
and contributions as overriding considerations when deciding
about deviating from parity of population among electoral
districts."
"Rather, our constitutional mandate (as discussed
in Part 4 - The Legal Framework) is to deviate from parity
only when necessary in order to ensure effective representation."
Cohen Commission, Part 8, p. 61
In
a sense this article could act as a "citizen's guide
to the 2007
BC Boundaries Commission", the Cohen Commission.
Their Preliminary Report is available for download from
its website; the entire document, available in parts, is
over 300 MB. Despite an enormous amount of information in
the report, plus additional
tools on the website, including a Google Earth view
of constituencies, the Preliminary Report does not contain
a convenient comparison of current and proposed constituencies.
This may result in difficulty seeing the forest for the
trees, or seeing the mistake that I believe the Commission
made in the Southern portion of Vancouver Island where it
proposes to increase the deviation from the provincial quotient
(the average population per MLA) . Discussion of the possible
error requires both a statistical description and a discussion
of the legal requirements for drawing constituency boundaries.
The
regional summary table shown below provides comparisons
between current and proposed constituencies. Clicking on
the links that are imbedded in the name of each region will
bring up details for the constituencies in each region and
maps of the current and proposed constituencies taken from
the report. The column headed "average deviation with
81", next to the current constituencies, is calculated
from what the Commission's report provides as the average
deviation with 79 MLAs. Change in deviations from the "provincial
quotient" is made in two steps: 1) by changing the
quotient as a result of increasing the size of the legislature,
and 2) by changing boundaries. Showing the deviation under
the assumption of 81 MLAs for the existing 79 constituencies
isolates part of the change (assuming that the addition
would not affect the constituency being examined). For example,
with 79 MLAs the current constituency of Surrey-Whalley
deviates from the provincial quotient of 52,069 by minus
4.2%, lowering the provincial quotient to 50,784 by increasing
the number of MLAs to 81 reduces Surrey-Whalley's deviation
to minus 1.8%. Taking the next step of changing Surrey-Whalley's
boundaries further reduces its deviation to +1.0%. The average
deviation in each region, as shown in the table, is calculated
as the average of the absolute values of the deviations
for each constituency in the region; in other words, sign
doesn't count, although being under means votes are overvalued
while being over means votes are undervalued.
A
careful observer will notice that the number of current
constituencies in the table totals 80, one more than the
actual number. That is because "Burquitlam" is
counted twice, once in Burnaby and once in the Tri-Cities.
That double counting is found in the Commission's Preliminary
Report which shows Burquitlam in both regions (and also
includes it in both regions in "Appendix L", pp.
410-413). Although Yale Lillooet also overlaps two regions,
it is treated differently in the Report. This summary does
not attempt to guess why the Commission treated Burquitlam
and Yale Lillooet differently in its analysis of regions.
The
table shows that the Commission's proposals would substantially
reduce deviations from the provincial quotient in all but
two regions: Richmond and Delta and Vancouver Island and
the South Coast. Changes to constituency boundaries combined
with the reduction in the provincial quotient makes no overall
change to the average regional deviation for Richmond and
Delta, but the changes increase the average deviation in
Vancouver Island and the South Coast by almost a full percentage
point. That movement, in the wrong direction, is worse when
the five
South Island constituencies are examined. The average
deviation for those constituencies was 2.3% with 79 MLAs
and 2.2% with 81, but changing the boundaries would increase
the average deviation to 10.0%. That's almost a fivefold
increase in the wrong direction. Population projections
to 2013, contained in the appendix to the Preliminary Report,
indicate that the Commission believes that the South Island
will not grow as fast as the rest of the province so two
elections from now the deviation would be worse than it
is now but not as bad as it would be in 2009 if the boundaries
were changed. Does the Commission have the legal authority
to recommend boundary changes that exacerbate deviations
from voter parity?
The
Commission argued as follows:
"For
each electoral district we propose, we must be satisfied
that its population is at or near the provincial electoral
quotient, or its positive or negative deviation is justified
by consideration of constituents' access to their MLA and
the MLA's ability to represent their interests.
As noted earlier in this Part, the legislation under which
we operate specifies the factors we must take into account
when proposing an increase in the number of electoral districts.
To wander from these considerations would be the misapplication
of our discretion. Our starting point is representation
by population. Deviations are permitted only to make that
representation effective, not to change the balance of electoral
power to favour one area of the province over another. To
quote the Alberta Court of Appeal: "No argument for
effective representation of one group legitimizes under-representation
of another group."
"Appendix
C" to the Preliminary Report elaborates on the 1991
Alberta reference (Reference re Electoral Boundaries
Commission Act (Alta.) (1991), 86 D.L.R. (4th) 447 (Alta.
C.A.) in which the Court of Appeal upheld Alberta's legislation.
A five member boundary commission acting on that legislation
produced five separate reports, resulting in a committee
of the legislature producing a new boundary scheme which
was tested in another reference to the Court of Appeal (Reference
re Electoral Divisions Statutes Amendment Act, 1993
(Alta.) (1994), 119 D.L.R. (4th) 1 (Alta. C.A.)). In dealing
with the later reference, the Court did not rule on constitutional
issues.
The
2002/2003
Alberta Electoral Boundaries Commission submitted its
final report in February 2003. It's legal authority, the
Electoral Boundaries Commission Act (Revised Statutes
of Alberta, 2000, Chapter E-3 as amended) gave explicit
direction to consider municipal boundaries "wherever
possible". The Alberta Commission quoted paragraph
44 of the 1993 Reference:
"It
is one thing to say that the effective representation
of a specific community requires an electoral division
of a below-average population. That approach invites specific
reasons, and specific facts. The constitution of Canada
is sufficiently flexible to permit disparity to serve
geographical or demographic reality."
"It
is quite another to say that any electoral division, for
no specific reason, may be smaller than average. In the
1991 Reference, we affirmed the first, not the second.
We affirm again that there is no permissible variation
if there is no justification. And the onus to establish
justification lies with those who suggest the variation…."
The
fundamental issue in determining whether the Cohen Commission
has erred by proposing to increase the average deviation
from the provincial quotient for constituencies in the Capital
Region depends on what kind of justification is required
in order to exercise the discretion to vary a constituency
by plus or minus 25% from the provincial quotient. The 1995/1996
Alberta Commission developed a "matrix" which
included six factors which were scored on a scale of 1 to
10. The 2002/2003 Alberta Commission built on and refined
that matrix; in other words, two Alberta Commissions developed
elaborate justifications for using their discretionary power
to vary any constituency by any amount.
In
contrast to the elaborate justifications developed by Alberta's
Commission, the Cohen Commission (consistent with previous
BC Commissions) did not quantify criteria it used to justify
boundaries, irrespective of how much those boundaries caused
a constituency to deviate from the provincial quota.
A deviation of 2% appears to be treated the same as a deviation
of 20%. In particular, although the BC statute (unlike Alberta)
makes no reference to municipal boundaries, the Cohen Commission
repeatedly uses municipal boundaries to justify deviations.
In introducing its proposed constituencies for the Capital
Region (or sub-region), the Cohen Commission said:
"Starting
in the south, the Capital Regional District has 13 municipalities
and four unincorporated electoral areas (see map of Capital
Region municipalities, page 298), which presented us with
a real challenge in developing electoral districts that
followed, as much as possible, municipal boundaries."
It could
be argued that respecting municipal boundaries while minimizing
variation from the provincial quotient is desirable, but
it is entirely a different matter to increase variation
from the provincial quotient so as to respect municipal
boundaries. The Commission is conducting a second round
of hearings which might help persuade it that an error was
made in its proposals for the Capital Region; failing that,
as happened when the BC Civil Liberties Association challenged
the process for drawing boundaries (Dixon v. British
Columbia (A.G.) (1989), 59 D.L.R. (4th) 247 (B.C.S.C.)),
BC may spawn another court case that tests the process followed
by a boundaries commission.
|
| |
|
About
Me |
Mail Me | Links