Sunday, February 19, 2023

Exploring Strengths in The Charter Challenge for Fair Voting and the Affidavit of Antony Hodgson

A Compilation of Key Excerpts (with Editor's Notes by Boyd Reimer)

(This post is duplicated in a Google doc at this link.)

Why a Charter Challenge:

Quote from the Charter Challenge for Fair Voting website: 

“The Charter guarantees our right to vote and our right to equal treatment. It's up to the Supreme Court to decide when this principle has been violated…. 

 “Each citizen is entitled to be represented in government.” – Justice Beverly McLachlin  

That’s each of us, not just some of us. That principle has been upheld in several cases throughout Canadian elections. 

The court has also said that voters should be treated equally, but when only half of us have an MP we support, the system is discriminatory and excludes those voices. [Countries with electoral systems based on proportional representation do much better in terms of voter equality (see evidence below).]

Constitutional lawyers have advised us that we have a worthwhile case to make against our current voting system on these grounds.

Why not change the system through political channels? 

Governments of various stripes have supported electoral reform while in opposition, and on the campaign trail only to back away from the promise of reform once elected.

The present government and parliament have decided not to proceed with electoral reform. It could be years, or decades, and several elections, before electoral reform becomes a possibility again. Democratic rights are human rights. It's the Supreme Court's job to ensure that those rights are protected when parliament neglects to protect them.”

End Quote from the Charter Challenge for Fair Voting website:

Editor's Note: How to make a strong case in a Charter Challenge

The Charter guarantees rights to individual citizens; not political parties. Therefore a successful Charter Challenge for Fair Voting must show how our current electoral system violates the Charter rights of individual citizens -- not political parties. (1)

“But,” you may be asking, “has the Supreme Court already made a previous ruling about fair voting and fair representation of individuals? In other words, do we have a pre-existing precedent and thereby have a much stronger foundation upon which to build our case?”

The answer is “yes!”

Here’s a quote from the Supreme Court of Canada (1991 Saskatchewan Election Boundaries Reference, p 183). (This ruling is about the sizes of riding populations):

“[The result of] a system which dilutes one citizen’s vote unduly as compared with another citizen’s … will be uneven and unfair representation.”

The above 1991 quote, among other things, provided Antony Hodgson with a strong foundation to submit an affidavit as evidence in the case.

That key affidavit is explored below.

Editor's Note: A remarkable finding of the Affidavit of Antony Hodgson

Before we explore that affidavit in detail, here’s one of its most remarkable findings:


Unfortunately, in recent decades our electoral system is still allowing for differences in voter power that are five times greater than those 1991 allowable differences in voter power due to the sizes of riding populations (see evidence below).

This means that, in terms of voter equality, one part of our democratic process (our electoral system) is five times worse than another part (differences allowed in sizes of riding populations).

If voter equality is like a chain then it’s only as strong as its weakest link. The weaker link in this case is our electoral system--as the below measurements and evidence show.

A weakness in voter equality is detrimental because the validity of our democracy depends on the strength of its voter equality.

If the validity of our democracy is like a chain then it’s only as strong as its weakest link. In Canada’s case, a weak link is a relatively low level of voter equality in our electoral system--as the below measurements and evidence show.

Photo credit: capacityllc.com

Digesting Excerpts from the Affidavit of Antony Hodgson to the Charter Challenge for Fair Voting case

The Affidavit PDF is available for download from this link (See “Fair Voting BC”): 

Quotes, paraphrases, and sources from that affidavit are noted in the below  “Excerpt Compilation Article” in parentheses (), and/or footnoted. Underlined emphasis is my own (ie Boyd Reimer).

Introductory Synopsis:

 In the [1991] Saskatchewan Electoral Boundaries Reference [on the sizes of riding populations], the Supreme Court of Canada concluded (at p 183) that the first condition of effective representation is “relative parity of voting power.” The court argued that dilution of voting power is problematic because it reduces the legislative power of the citizen and access to and assistance from his or her representative. In their words:

 “A system which dilutes one citizen’s vote unduly as compared with another citizen’s vote runs the risk of providing inadequate representation to the citizen whose vote is diluted. The legislative power of the citizen whose vote is diluted will be reduced, as may be access to and assistance from his or her representative. The result will be uneven and unfair representation.” (Pg 29, par 63)

This previous Supreme Court ruling on citizens’ “legislative power” as it pertains to electoral boundaries can be used as a point of comparison to show that Canada’s electoral system is not as fair in the distribution of legislative power to individual citizens as when we compare it with the electoral systems of countries with proportional representation.

When we use something called the “Legislative Power Share score” to precisely and mathematically measure and compare the different degrees of “legislative power” that different citizens have within our electoral system... we find that “the [difference and] disparity in Legislative Power Share scores in recent decades has typically been over five times higher than the disparity allowed by the courts in the sizes of riding populations.” (Pg 56, par 101)

What exactly is a “Legislative Power Share score” and how does it precisely and mathematically measure and compare the different degrees of “legislative power” that different citizens have within our electoral system?

Let’s answer that by taking a deeper dive into the “Parity in Legislative Power” project carried out by Antony Hodgson when he prepared his Affidavit for the Charter Challenge for Fair Voting.

First, notice that parity of voting power is a phrase taken directly from the above-mentioned precedent ruling of the Supreme Court.

The the Affidavit of Antony Hodgson defines and summarizes three “indices,” which are “measurements,” or “metrics,” in the following way (pg 33, par 71):

 1. The Representation Metric (RM) - the percentage of voters who are represented by an MP for whom they voted;

 2. The Legislative Power Share (LPS) Score - the share of legislative voting power held by individual voters relative to their ‘fair share’ (parity)

 3. The Legislative Power Disparity Index (LPDI) - The distribution of legislative power amongst voters. (It summarizes the integrated impact and effect of disparities in legislative power that are felt and measured at the individual voter level by the Legislative Power Share score.)

[End of Synopsis]

&&&&&&&&&&&&&&&&&&&&&

A Deeper Dive…

… into the Affidavit of Antony Hodgson's measurement of the Legislative Power of Individuals:

(Below excerpts compiled by Boyd Reimer)(3)

1. The Representation Metric (or “measurement”) identifies the percentage of voters who are represented by an MP for whom they voted. If the votes of all voters counted towards electing an MP, the Representation Metric would be 100%. (pg 35, par 73) The below comparison chart (Figure 11) is taken from pg. 40:

Figure 11. Representation Metrics for Canada (2019 federal election) and two recent elections in each of three selected countries using proportional voting (to demonstrate consistency of findings). The solid colours represent those voters who are represented by a representative for whom they have explicitly voted (‘direct’ representation), while the shaded colours represent those voters who are represented by a representative who won a seat by virtue of votes cast for the representative’s party (‘indirect’ representation). Norway uses a List PR system with 19 compensatory seats out of 169 seats total, Ireland uses a Single Transferable Voting System, and New Zealand uses a Mixed Member Proportional system. The two systems that use multi-member regions (Norway and Ireland) have similarly high levels of direct representation, while New Zealand has a level of direct representation similar to that of Canada due to its use of single member districts. The two systems that have a national compensatory element (Norway and New Zealand) have similarly high levels of total representation.

2. Legislative Power Share (LPS) Score (pg 40): measures the share of legislative voting power that individual voters hold compared to other voters in Canada (their ‘fair share’ or 'parity'). (pg 41, par 78)

Did You Vote For Your MP Or Not?:

Pg 41, par 79: Each of Canada’s 338 MPs holds an equal share of legislative power in the Canadian Parliament. If a voter did not vote for their elected MP, this voter had no influence on the ability of the MP to exercise legislative power (and their MP did not rely on their vote). We therefore assign this voter a Legislative Power Share score of zero.

On the other hand, if a voter voted for their elected MP, the voter’s share of the MP’s legislative power depends on how many other voters in the riding also voted for the MP.

The Scores Of Those Who Did Vote For Their MP Are Calculated As Follows:

Pg 42, par 81: To determine the “legislative power share” of a voter who voted for their MP, we divide the average number of votes cast per seat in that voter’s riding by the number of votes received by the elected MP. 

We then apply a weighting factor to adjust the share of each voter’s legislative power depending on whether the total number of voters in the riding was greater or smaller than in other ridings (since voters in ridings with fewer voters have more impact on the election of an MP than voters in ridings with more voters).

An Amplified Voice Or A Diminished Voice Are The Effects Of Deviating From A Representation Of “1”:

Ideally, if each voter had equal influence on the makeup of the legislature, then every voter would have a Legislative Power Share score of 1. If a voter’s Legislative Power Share score is greater than 1, that voter effectively has their voice in Parliament amplified relative to the average voter. Conversely, if a voter has a Legislative Power Share score of less than 1, then that voter has a diminished or diluted voice in Parliament. The relative representation of different voters becomes more uneven and unfair the more the distribution of LPS scores deviates from a uniform value of 1.

Pg 44: Figure 12: “Legislative Power Share” (LPS) Score Range in the 2019 Canadian federal election

 


Figure 12. Histogram of Legislative Power Share scores in the 2019 Canadian federal election. Scores closer to the parity value of 1 are shown in green, while scores further away are shown in shades of yellow and red. The sum of all bars in the histogram is 100%.

Pg 48, Figure 14: Percentage of Voters with Various LPS Scores – comparing Canada to Countries with proportional representation

 

Figure 14. Distributions in Legislative Power Share scores in Canada’s 2019 federal election and in recent elections in three selected countries using proportional voting systems.

The next chart, Figure 15 is basically an excerpt from the above chart. (pg 50)

Figure 15. Percentage of voters with a Legislative Power Share score within +/-10% (dark green), +/- 25% (dark green + light green) and +/-50% (dark green + light green + yellow) of parity (colours correspond to previous chart). All comparator countries have at least two-thirds of voters with LPS scores within +/-50% of 1 (ranging from 67-92%), while Canada has fewer than 10% of voters within that margin.

3. Legislative Power Disparity Index: The distribution of legislative power amongst voters. (pg 50)

To understand that concept simply look at the below Figure 16 (pg 51), keeping in mind that the bars at either end (sides) of the plot show the shares we would expect if legislative power were distributed equitably. In other words, the black line across is the benchmark of uniform equality:


Figure 16. Distribution of Legislative Power Shares across different segments of the voters for all federal elections in Canada since Confederation. The share of legislative power held by the bottom 50% of voters is shown in grey, the next 40% in orange, and the top 10% in red. The bars at either end (sides) of the plot show the shares we would expect if legislative power were distributed equitably.

Pg 51, par 92: In a voting system with perfect legislative power parity, the legislative power shares would be distributed evenly across all voters (as in the separated bars at either side of the figure above). In Canadian elections, however, the 10% of voters with the greatest legislative power scores have consistently held close to 25% of the legislative voting power. The next 40% of voters have consistently held most of the remaining 75% of legislative voting power while the bottom 50% of voters have often held no legislative voting power at all in the past 60 years.

The disparity in legislative power between voters in Canadian elections illustrated above can be more precisely visualized and quantified using tools drawn from the political science and economics literature. In this literature, the Gini coefficient is a commonly-used index to measure disparity in the distribution of a property (e.g., income or wealth); here we use it to measure disparity in the distribution of Legislative Power Share scores. It is calculated by plotting the cumulative percentage of the property against the percentage of the population that holds that percentage of property, ordered from those holding the least of the property to the most (this is known as a Lorenz curve).

[(Ed. note: But since the charts --both above and below-- are visually self explanatory it’s not absolutely necessary to fully comprehend that they show the “Gini coefficient,” or “Lorenz curve.”)]

Pg 52, par 94: The chart below (Figure 17) illustrates Lorenz curves for four elections – one from Canada and one for each of the three example countries using proportional voting systems. In each plot, the blue line indicates an equitable distribution of legislative power - i.e., each voter holding equal shares. The red line is the Lorenz curve, which shows the cumulative shares of legislative power held by the percentage of voters on the horizontal axis.

In an election with an equitable distribution of legislative power (i.e., where each voter equally contributed to the election of an MP), the two lines would coincide. In contrast, increasing deviations between the red and blue lines indicate increasing inequities in the distribution of legislative power in the election.

Figure 17 (Pg 53) (LPS = Legislative Power Share):

Pg 54, par 97: The following figure (Figure 18) shows the Legislative Power Disparity Index for all Canadian federal elections since Confederation.


Figure 18. The Legislative Power Disparity Index for all Canadian federal elections since Confederation. The LPDI has averaged 57.4% in the period from 1970 until the present. 

Pg 55, par 100: As a point of comparison, the Gini score expressing the disparity in riding sizes (shown in Figure 20) averaged approximately 20% from Confederation until 1965, but has averaged only 11.1% since 1968, when the first federal Election Boundaries Commissions were established, and only 9.3% since 1993, which was the first election following the Supreme Court ruling in the Saskatchewan Electoral Boundaries Reference which limited permitted deviations to +/-25% in most circumstances.

Figure 20. Riding Size Disparity Index (Gini scores) for all Canadian federal elections since 1867. Since the Electoral Boundaries Commissions were established prior to the 1968 election, the index has decreased from roughly 20% in the first century after Confederation to under 10% in recent decades.

Pg 56, par 101: The disparity in Legislative Power Share scores in recent decades has typically been over five times higher than the disparity allowed by the courts in the sizes of riding populations.

[To illustrate that five to one difference the below chart shows the Riding Sized Disparity Index (Fig 20) superimposed on the Legislative Power Disparity Index (Fig 18):]

Are countries with proportional representation doing better than Canada? The answer is below:

Figure 21 (pg 57)

 

Figure 21. Legislative Power Disparity Indices calculated for Canada in 2019, along with those for two recent elections held in the three comparator countries. All comparator countries show LPDI values less than half that of Canada.

&&&&&&&&&&&&&&&&&&&&

Many thanks to Antony Hodgson and Fair Voting BC for the above excerpts from the Affidavit for the Charter Challenge for Fair Voting:  The affidavit PDF is available for download from this link (See “Fair Voting BC”): 

Footnotes:

Footnote 1: You may recall 2015 when Trudeau promised to change our electoral system. As a consequence, a Canadian House of Commons Special Committee on Electoral Reform was formed to investigate reforms to the Canadian electoral system. Their final report recommended, among other things, that the government would design the system with the goal of any proposed system scoring a 5 or less on the Gallagher index. The Antony Hodgson / Fair Voting BC Affidavit refers to that finding, but instead of using a measurement for parties, it develops a measurement for individuals. This is in keeping with the fact that the Charter of Rights is about individual rights instead of rights for political parties.

Footnote 2: The affidavit provides hyper links to supporting evidence in Byron Weber Becker’s online “Election modelling.” See links:

Election modelling 

Legislative Power Analyses

Parity in Legislative Power

Quote from Byron Weber Becker’s Election Modelling website:

“[As a counterpart to the Gallagher Index] this paper proposes a complementary voter-centric measure, the Legislative Power Disparity Index, which measures the fairness of an election as a whole from the voter’s perspective. It is built on a Legislative Power Score (LPScore), which is assigned to each voter in the election. In a perfect world each voter would have an LPScore of 1 and the election as a whole would have an Legislative Power Disparity Index (LPDI) of 0.”

Footnote 3:  Compiling Editor of the above excerpts: Boyd Reimer (Feb 19, 2023)

(This post is duplicated in a Google doc at this link.) 

 

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.