PR best route to electoral integrity, and other information from the Oxford Handbook of Electoral Systems (2018)

"Electoral systems and Electoral Integrity", chapter 24 of the Oxford Handbook of Electoral Systems (2018)

Pippa Norris, the author of the chapter, makes several important points.

Here are some extracts [I have changed her wording where I consider appropriate]

Norris says PR systems are generally believed to strengthen electoral Integrity more effectively than FPTP and other plurality systems.

Why is this true?

1. PR generates checks on the unconstrained power of the single-party governments and executives and thus reduces the danger tht the largest party in a legislative chamber will manipulate the elector rules in its own favour in later elections

2. By widening the range of winners, PR systems and PR election are seen to build trust in the electoral process among a variety of stakeholders and competing parties They thus reduces the urge for losing parties to disrupt the electoral process and to undermine the outcome by besmirching its legitimacy.

3. PR systems encourage parties to offer balanced group of candidates, and thus PR contests are more inclusive to women and minorities

4. Plurality system have in the past created an incentive for candidates to try to win through illegal, fraudulent, or corrupt actions. This is especially true in winter take all contests where a wafer-thin difference between the two most popular contenders, in such cases the shift a few dozen votes sometimes makes the difference between winning and losing. Sometimes both parties are well below a majority but still the winner is determined only through relative strength. (While PR is not zero-sum game. If a candidate takes a certain number of votes, he or she will be elected; if two candidates take that certain number of votes, they will both be elected.

   PR with a set quota, whether merely imposed through district magnitude or through a set electoral threshold, seems to be more concrete and evoke a win//lose library result, but actually because there can only be one winner in FPTP contests, and that winner is determined through relative strength (plurality), under FPTP it becomes more potentially beneficial to bribe a few dozen votes to shift their vote than under PR. relative strength can be guessed at through polls ND gut instinct but to win a seat or not under PR is actually less predictable. for the border line cases while it is quite certain that the most-popular candidates willl be sure to be elected.

5. The benefits of partisan gerrymandering are easily seen in FPTP single-member districts, while under PR the use of MMDs means less benefits for that kind of partisan electioneering. (if every district elects fairly , it hardly makes any difference how you draw the districts. Also fewer districts dividing the electorate means less opportunity for gerrymandering. Also MMDS with varying number of seats means a system might simply use pre-existing natural boundaries such as city corporate limits or counties preventing even possibility of partisan districting.

PR acts in positive way to raise the level of electoral integrity (p. 497)

PR typically produces a coalition multi-party cabinet that provides internal checks and balances on the power of any single party in government and thereby curbs the capacity of the largest party to manipulate the electoral rules or to nominate and appoint partisan powerful electoral officials in its pocket.

Strong and independent legislatures with powers of oversight, where opposition parties can hold electoral authorities to account, provide a safeguard to protect the integrity of electoral laws and procedures.

The choice of PR, mixed member [parallel?], or plurality-based electoral system used for the lower house of the national parliament predicts the general levels of electoral integrity in the political culture overall, with PR having the most positive effects. (p. 497)

PR increases the range of parties represented. Inclusive power-sharing arrangements build feelings of political trust, social tolerance, and legitimacy, long regarded a he foundation for a civil political culture and stable democracy (p. 497)

Other notes:

through some form of electoral magic ,even authoritarian regimes can use PR.

Cambodian People's Party has consistently maintained and dominated legislative elections that use PR. (OHES, p. 389)

During single-party rule in Kenya and Tanzania, elections were competitive.  (OHES, p. 389)

in addition to electoral rules (to convert votes to seats), autocrats sometimes use other mechanisms sometimes to control results -

candidate nomination, candidate campaigns, voter turnout thresholds below which the election is void.

simplest form of PR uses a single round of voting in which each voter casts a single "categorical" vote

no second voting required

no ranking of candidates or parties.

"categorical" vote sometimes called X voting.

Implies no tiers such as in MMP

====

The larger the DM, the better the chance for a small party to elect someone

=========================================

Seat Product Model determines (or predicts) how many parties have seats in the chamber, the effective number of seat-winning parties.

Derived by multiplying total seats by average DM and finding root to the power of 6.

thus a political entity with 625 seats and all single-seat Districts would have same SPM prediction as a place with PR and multi-seat Districts (MSDs) of 25, and the same as a place using PR with 5 MSDs and total of 25 seats. (p. 48)

number of pertinent parties is the SPM plus one party. (p. 53)

These formulas work good for most of the 49 countries that use simple voting systems.

but U.S. and UK and five others do not conform as they have fewer effective and pertinent parties that the formula would predict. (p. 55)

only India has more parties with representation than the formula predicts. (p. 56)

Canada has about what formula says despite being thought to e exceptional. In part this is likely due to its use of cube root to determine rep. Both UK and U.S. have much smaller chambers than the cube rule would indicate.

France with TRS and Ireland with STV also meet the formula prediction. (p. 56)

================================

UK local elections

London (England) used Block voting (p. 654)

(table 30.3) 2014 Barking and Dagenham district -- Labour won all the seats with 69 percent of the vote

Tower Hamlets district - Labour won almost a majority (22 seats) with 39 percent of votes;

Tower Hamlets First won 18 seats with 35 percent of vote,

Conservatives won five seats with 12 percent of the vote (about proportional)

small parties won no seats.

STV was brought in for Northern Ireland elections in 1970s to represent minorities more effectively than the FPTP and BV system that the Unionist gov't had reinstated for local councils in 1922 after a brief use of STV.

N. Irish Assembly created by Good Friday (Belfast) Accord [in 1980?] and STV installed there

STV also installed for election of MEPs (and also installed in Scottish local Councils from 2007)

(p. 657)

1867 Reform Act (UK)

brief experiment with limited voting, one less than the number of seats in 3- or 4-seat districts.

but these were seen as failing to protect minorities and were abolished with the intro of primarily SSDs with the 1885 Redistribution of Seats Act. PM Gladstone said minority rep would be achieved well enough with the franchise extension. (OHES, p. 629)

[Gladstone was immediately proven wrong, or was eventually proven wrong once third parties got going,

Labour Party got going in 1900

While Radical members elected in the early-to-mid 19th century

[one Chartist (O'Connor) elected in 1847]

what Radical members?

and

Irish Nationalists operated in the late 19th century,

actually earlier -- Irish Repeal Association elected 42 in 1832

Irish Conservative in 1865 - 45 (often included in Conservative caucus)

Home Rule - 60 MPs elected in 1874

Labour Party got going in 1900

Keir Hardie is generally recognized as the first independent Labour Member of Parliament (MP) to be elected in the British House of Commons, free from affiliation with either the Liberal or Conservative parties, in the 1892 general election]

===

eventually Liberal party came to support STV but then it was too late - Conservative and Labour were lukewarm on PR.

PR meant STV in UK, and STV was put forward by reformers "who looked down on party politicians to such an extent that their system was designed to enhance the chances for independent candidates and allow voters to choose between candidates of the same party, something that most party leaders would like to avoid." (p. 629-630)

(See Jenifer Hart's book on 19th-Century electoral reformers)

"The president of the PR Society (now the ER Society) recommended STV as a way to reassure the Protestant minority in the event of Home Rule when he visited in 1911; the recommendation made its way into the 1922 constitution of the Irish Free State without controversy." [this was Leonard Courtney of Penwith, who was chair of the Society's executive committee from 1905 to 1918.]

A second wave of PR effort happened in 1970s after two gen. elections in 1974.

pluralistic system - Liberals coming up again in popularity, Labour, Conservatives

plus Welsh and Scottish nationalists.

in Northern Ireland, Westminster had suspended the FPTP-elected N.I. Assembly.

STV introduced for local elections and for the MEPs to reassure the minority (Catholics) that they would be equitably represented.

Welsh and Scottish nationalism rose as devolution was proposed. PR was discussed.

Conservative and Labour opposed PR, but Labour was friendly toward IRV.

Liberals and their successors, the Lib-Democrats, supported STV, as did Scottish Nationalist Party and Plaid Cymru.

During Conservatives' 18 years in power 1979-1997, Labour discussed PR.

Labour Party's Working Group on Electoral Systems, chaired by Raymond Plant,

its 1993 report called for IRV (what is called "majoritarian SV") for HofC, MMP for Scottish Parliament, regional list PR for elections of MEP and for a reformed House of Lords.

After the 1997 Labour victory, the number of U.K. electoral systems doubled, with three more added. serious consideration given to ER in HofC elections. (p. 632)

Roy Jenkins pushed for a form of IRV with just a few levelling seats (15 to 20 percent of the total) to be put to voters in referendum

no referendum held

2010 Conservative formed coalition gov't with Lib Democrats who had pushed for STV for decades.

referendum held but choices were FPTP or IRV , not PR. (p. 631)

68 percent (of the 42 percent who voted) voted for FPTP. (p. 632)

=======

in early 200s, UK using six Electoral Systems (p. 631- 640) --

FPTP -- Hof C, local council elections in England and Wales

AMS (MMP)

regionalized version -- Scottish Assembly (1990s-)

MMP with at-large levelling seats - later adopted for elections of London Assembly

regionalized version -- later adopted for elections of National Assembly of Wales

(Wales had 560 members and used 40 SSFs and four 5-seat regions).

(Wales has since switched to a closed list PR system using the D’Hondt method with 16 6-seat districts, as per the Senedd Cymru (members and elections) Act 2024.

number of members has grown to 96, all elected in MSDs and in one tier.

(A 2017 report of an expert commission led by Laura McAllister suggested a re-appraisal after the 2026 election when consideration should be given to a switch to STV or open list PR, and use of gender quotas.

(The commission's report is available online at https://www.gov.wales/sites/default/files/publications/2024-07/independent-commission-on-the-constitutional-future-of-wales-final-report.pdf.)

The Commision and other government officials saw MSDs (and list PR) as improvement on MMP.

"The proposed closed list method is an improvement on the current ‘mixed member system’, where in the constituency vote, voters choose between one candidate selected by each party and, in the regional vote, between closed lists selected by political parties. This will deliver greater proportionality but means that voters will no longer have a direct connection with their local MS. Voters will only be able to choose between lists put forward by political parties and individual independent candidates, should they stand for election.

They will not be able to vote in favour of, for example, a candidate who has been ranked in a lower position than another by their party.

We see a good case for alternatives such as a Single Transferable Vote or an open list system, where voters can choose between named individuals representing parties or independent candidates as well as between political parties. We recognise that this can lead to internal rivalries between candidates from the same party, but we encourage the committee to consider these factors along with voters’ perceptions of fairness in the system as part of its review [after the 2026 election]." (page 33 of commission report)

(AMS is not recommended as a term for MMP. The regional members are not additional, as in lesser than the constituency members. The term "Additional Member System" includes both MMP and MMM (mixed member majoritarian which means Parallel systems) as used in Italy and Japan). (p. 292).

STV -- N. Ireland's 3 MEPs (prior to Brexit); N. Ireland local council (since 1993); Scottish local councils (since 2007)

Block Voting (MNTV) -- some local elections in England and Wales, especially London borough councils (see above: Barking... and Towers...)

(BV was used to elect two members from most English Parliamentary districts from the medieval period to the 19th Century. (p. 634))

Supplementary Voting (two-round IRV) --

mayor elections in English cities and election of police ND crime commissioners

List PR (regional version) -- MEPs (starting in 1999, dropped when UK left EU).

(residents of England outside London didn't use PR for anything (after Brexit).

see ER Society blog:

https://electoral-reform.org.uk/where-is-single-transferable-vote-used-in-the-uk/

========================================

Routes to ethnic Minority Representation (OHES, p. 522-23)

a minority may elect rep. where FPTP is used, if it is sufficiently large and/or geographically concentrated.

it is also likely to elect representation if

-PR is used (many countries in the world)

-requirement that candidate slates include minority candidates (Singapore and Lebanon)

-non-geographically based seats are specifically reserved for minorities, by

-drawing districts that only minority candidates can compete in (India - scheduled castes)

-drawing districts where only minority voters can vote (NZ - Maori seats; perhaps Fiji "communal constituencies that guarantee rep for Indigenous Fijians, Indo-Fijians and Rotumans, plus "open" constituencies)

-drawing districts where minority voters are in majority (or perhaps at least have a plurality) (affirmative gerrymandering) (sometimes done in the U.S. - see next)

U.S. Voting Rights Act of 1965 outlawed cracking minority voting blocks.

Blacks, Hispanics, Asians and U.S. First Nations have rights for protection from that kind of disenfranchisement but only if the group:

-is sufficiently large and geographically concentrated to form a majority in a SSD,

-is politically cohesive,

-can demonstrate that the majority population votes as a block against the minority community's preferred candidates, who usually lose.

Because many minority groupings do not satisfy those conditions and do not have money to sue if a jurisdiction fails to create majority-minority district even when conditions are satisfied, minorities are still proportionally under-represented.

======

Gerrymandering (OHES, p. 500-4501)

large multi-seat districts under PR do reduce -- or sometime even eliminate -- gerrymandering.

esp. when districts are drawn according to boundaries of administrative regions

in about 50 countries, existing regional or administrative provincial boundaries are used to draw election districts in the lower house of the nat. legislature.

Critics though say that both PR elections and power-sharing arrangements in general have potential disadvantages (p. 501)

Donald Horowitz says PR rules and power-sharing arrangements may unintentionally serve to heighten latent ethnic identities, strengthen party extremism and fragmentation and freeze community boundaries, failing in the long term to generate the conditions of social tolerance and moderation, where democracy flourishes.

(and when PR does not produce these bad things, it is not clear that the credit should not go to the political culture, the free press, level of development, etc. (written perhaps to be devil's advocate) (OHES, p. 501)

Certainly, PR may increase minority solidarity and identity

is no recipe for all the parties co-operating;

the power-sharing (caused by minority government) is not transparent, nor foreseen by the voter;

but under STV-PR systems, some candidates of diff parties will work together (or at least not attack each other viciously), in hopes to get secondary preferences and voter transfers).

p. 500

compared with members elected through closed list PR contests in multi-seat districts, elected officials in SSDs -- esp. those elected by tight margins where a few votes can determine the winner - - may feel "strong incentives to engage in clientelism, patronage politics, rent seeking and other corrupt acts."

PR esp where large Multer Seattle districts are used are expected to prevent these temptations since many more wold need to be manipulated to change the outcome

with closed list PR, a candidate will not necessarily personally benefit from more votes for the party "malfeasance is more efficient under SSD rules" (p. 500)

Many argue that PR contests show less perceived electoral fraud and higher perceived fairness to minor parties compared to plurality-based systems. (p. 500)

========

Alberta and Manitoba dropped provincial-level PR in the 1950s. That was about the time or bit later than the wave of electoral rule changes in the post-WWI period made by authoritarian governments elsewhere.

(OHES p. 397) during the post WWII period, 10 percent of authoritarian regimes made major rule changes

28 percent of authoritarian regimes made minor rule changes.

In the post WWII period, 6 democracies made major changes, and 13 percent made minor rule changes. (names not provided)

This confirms the relative ease with which authoritarian regimes can change the rules.

major rule changes and frequent rule changes can make it difficult for competing parties to coordinate behaviour. (p. 397)

Behaviour arising from electoral reform includes splintering and merging of parties, running fuller or narrower slates, and election boycotts, etc.

under PR as each substantial party has chance to win some representation, boycotts are less likely.

Boycotts occur in about 40 percent of FPTP elections. (no dates given)

Political violence are less frequent in PR systems. TRS encourages the use of non-electoral means to win power. [likely during the period of time between rounds of voting.] the first round provides information about the geographic distribution of support for the various parties, and this info is used to channel violence against voters of opposing parties. (398)

=====

Electoral reforms (changes)

can be defined in two categories:

interparty - diverging from or approaching closer to party proportionality

intraparty "personalization" - voters and parties have more or less control over which individual candidate is elected. "the neglected dimension of the electoral system" according to Colomer 2011. (Colomer, "Personal Representation - the Neglected Dimension of Electoral System". Colchester, ECPR Press, 2011) (OHES, p. 128)

Some systems give parties lots of control over whom is elected; other systems give voters more control

Parties nominate their candidates, but some systems give independents more chance to be elected even without party label.

FPTP gives voter no choice over candidate if party is determined.

Electoral reform has generally been towards more proportionality.

Colomer (2005) determined that in 1874 there were 20 democracies with more than 1M population. They all used plurality-based systems.

1960 39 in total -- 16 used plurality, 23 used PR.

2002 82 in total -- 16 used plurality-based systems; 66 used PR.

2022 more than 100 used PR to elect all or some of their national legislators.

In early 1900s many European countries moved toward PR

after WWII newly democratized countries all adopted PR to some degree.

1993 NZ adopted PR, but Italy dropped PR in same years.

France used PR in the 1980s but then quickly went back to majoritarian [or plurality-based] system)

Europe between 1945 and 2012 in 35 European democracies 33 moved toward PR; 29 moved away.

===

Intraparty reforms -

between 1945 and 2009, in European democracies:

35 reforms increased personalization; only 12 reduced it.

elsewhere also showed mixed moves:

adoption of MMP in Japan and NZ reduced personalization

Colombia's adoption of list PR system in early 2000s reduced personalization

El Salvador and Iraq have both increased personalization by opening their closed-list PR systems.

Electoral change to or from personalization may be passed in conjunction with other reforms.

Reformers may achieve their aims through other than electoral reform [I don't know what he means]

Incremental nature of institutional change such as electoral reform - face opposition from entrenched interest, so reforms also maintain the status quo in some regards.

hence a range of patterns-

displacement where subordinate elements rise in importance

layering where aspects of he old are maintained as new arrangements are adopted

drift where institutions change when they are updated as the context changes around them.

Most reforms maintain the basic structure of the status quo

That may explain the popularity of MMP - they are manifestations of layering - while moving toward a new system, reforms are designed to preserve elements of the old - [single-seat districts].

Reforms face opposition or support from the actors

actors include: politicians, judges, interest and pressure groups, general public, international actors, and also economic groups and classes.

(international actors are involved mostly during democratic transitions but not generally in established democracies)

Judges have made rulings on franchise, districting, campaign finance but major change via the courts is rare.

General public -- sometimes exerted through referendum and citizens assemblies

but usually it is seen informally as public sentiment shapes political agendas.

widespread public mobilization is rare

but two indirect mechanisms sometimes shift the situation:

1. politicians sometimes retreat from making reforms that voters might see as being self-benefitting.

2. politicians might support reforms that they might otherwise not do if they think voters might reward them. (Many cases in Europe of reforms toward personalization were inspired by this motive as politicians responded to voters' disillusionment with the political status quo. -- Renwick and Pilet, 2016, p. 210)

pressure groups -- democratic reform groups can expert influence if conditions are right.

economic interests and classes --

certain powerful classes calling for reform inspired changes in early 20th Century in many cases Labour party calling for PR, or Labour party having strength and business community pushing for PR to retain some vestige of representation.

the exact cause-and-effect is not well known:

sometimes economic interest force reform directly

and sometimes perhaps deputies respond to electoral pressures in their districts and those pressures are shaped by economic patterns. (OHES, p. 119-120)

Politicians are part of the reform process (they are the ones that have to pass electoral reform legislation) and also are people most affected by reforms

Bowler, Donovan and Karp. "Why politicians like electoral institutions - Self-interest, values or ideology?" Journal of Politics, 68, no. 2 (May 2006)

Politicians may look just at the next election -- does a reform make their re-election easier or more difficult?

but also they may see that the stand they take on reforms themselves might impact their re-election.

As well, if they or their party favours democracy, they likely want to be seen to be supporting reforms toward PR. (OHES, p. 120)

in the three major reforms accomplished in 1990s - NZ, Italy and Japan [author leaves out the Balkan states for some reason] - none were done in favour of political interest of ruling parties. (OHES, p. 121)

in early 20th Century, PR was often adopted largely by consensus and and this reflected general acceptance of the principle that "each vote should count the same"

From 1999-2002, politicians interviewed revealed that their attitudes to ER was shaped partly by their electoral interests but also by their ideological commitment and democratic values.

Reforms also moved forward when there were high levels of popular "democratic aspiration" (OHES, p. 121)

Such was the case in Colorado in 2004 when state voters voted in referendum on the electoral seats being distributed among parties based on PR, instead of general ticket method (OHES, p. 121)

knowledge

two main relevant forms:

what actors know about Electoral reforms

what actors know about their own position.

A politician whose party is falling in popularity may want PR to preserve its representation or save what it can, but if it does not know that PR will do that or doesn't believe its popularity is that low, it may not support appropriate reforms. (OHES, p. 122)

today's interconnected world means many have heard of PR but practical familiarity is not same as basic awareness.

STV is important in British Isles but ignored elsewhere.

even reformers do not understand effect of PR - often exaggerate its positive results. (OHES, p. 122)

reforms cause uncertainty, and perceptions of risk.

Some facing uncertainty will prefer more proportional systems to hedge their bets, according to Pilet and Bol 2011. (OHES, p. 122)

in Dennis Pilon's Wrestling with Democracy (2013), he tracks electoral system change in the 20th Century through four historical epochs. He sees the shifting character of the political Left as the heart of the respective situations. (OHES, p. 124)

perhaps the four stages:

When left is absent, FPTP is unchallenged.

When left is just starting to emerge, FPTP still limits its representation so serves purposes of the powers that be.

Then when left grows stronger, the power that be desire PR to retain their due share of seats.

If left is in power, it does not want PR as it limits the overwhelming dominance in the chamber that FPTP would give it.

=====

Systematic failure occurs when the electoral system is incapable of delivering the normatively expected connection between the vote and the formation of executive authority" (Shugart, in Blais, To Keep or Change FPTP (2008)

Shugart (2008) says these expectations are shaped by the the prevailing electoral system.

Another source says any "extreme" system [perhaps where expectations are high such as a PR one] is more likely to be seen as failing than a more balanced system.

[PR seems balanced to me!]

change become more likely as its failure is widely seen.

according to this theory, the broad pattern of ER is determined at the systemic level, while the level of actors and their preferences provides the timing and detailed dynamics.

the systemic failure and perception does not determine that reform will happen. That depends on the actors and their rational calculations that the failure incites. (OHES, p. 124)

When reforms are achieved, the effect often does not fulfill the promises that the reformers had made or expected themselves. (OHES, p. 124)

Voter turnout does not always rise after adoption of PR

NZ did not see higher turnout (at least not immediately).

Italy's adoption of PR in 1990s did not fulfill reformers' promises (OHES, p. 125 - examples not given here)

Personalization reforms did not reconnect voters with politics.

Bowler and Donovan (2013) say "institutional changes may not actually change very much"

[I can see this :

likely the same party that had power under FPTP will yield power or lead a government coalition after adoption of PR.

the same small parties will have small caucuses in the Chamber and be held out of power except as secondary partners in a coalition.

there will still be polarization between the two main contending parties

more votes will be effective votes, but the improvement will be in votes that go to less-popular parties and candidates. who will have voice (which in many cases they did not have before) but still little power.

still power will be wielded by a majority but leaving others out in the cold same as before.]

ER that opens up the system to smaller parties yields more proportional outcomes.

this improvement yields both mechanical and psychological changes:

mechanical change -- more votes translate into seats.

effect on voters depends on responses of (potential) candidates, parties and voters.

and the effect on modes of campaigning or satisfaction with democracy depends on broad range of inter-related mechanisms. (OHES, p. 125)

Promises made (and not fulfilled) arise from "pathological optimism of reformers"

Reformers often believe (or say they believe) that relatively limited institutional changes will have transformative effects on politics.

Reformers also see their great project watered down by politicians and other actors, and thus also it seems they achieved their goal when they didn't actually and that is why promised change doesn't happen.

Some effects of ER may take several election to emerge as actors gradually change their behaviour [such as running more candidates or not running as many as PR at first had led them to do].

So early reports of effect of PR may under-estimate the ultimate changes that are seen only after two or three elections.

turnout may reflect campaigning practices, which take time to change

or perhaps election results are not so dependent on electoral system,

perhaps the relationship may work the other way

or some third factor may intrude.

party systems may indicate electoral system used

inclusive PR is adopted when system is inclusive.

PR brought in without an inclusive political culture may not produce same effect as adoption of PR in an inclusive culture (OHES, p. 126)

electoral reform does often have effects on party systems, voting patterns, modes of competition, governing arrangements.

some of these are more predictable than others.

But ER rarely delivers all that reformers promise. (OHES, p. 126)

=======

Democracy versus authoritarianism (or anyways super-majorities)

authoritarians debating becoming democratic should remember that reforms have trade-offs.

FPTP may yield a seat bonus for leading party, but also gives incentives for opposition candidates and voters to coordinate, potentially increasing the opposing strength.

PR makes it more difficult for opposition parties to coordinate, but PR potentially reduces the ruling party's dominance. FPTP "depresses voter turnout and incentivizes oppositon parties to build a pre-electoral coalition." (OHES, p. 392)

 [exactly what we see in Edmonton/Calgary during use of STV -- although UFA is accused of splintering the opposing parties/representation, PR ensured that opposition parties got seats which was not always the case under either Block voting or FPTP before and later --- often they got no seats at all.]

a dictator may prefer to use PR as it divides the opposition, and also increases voter turnout which gives the regime an image of popularity and invincibility."  (OHES, p. 392)

dictators may try to get a balance between co-opting elites within the ruling coalition and giving representation for opposition to appease them, it will sometimes adopt PR to splinter the opposition representation and then use financial resources, if it has any ,to to buy voter support.  (OHES, p. 392)

smaller reforms do not have risks of change to PR:

-gerrymandering

-high electoral thresholds

thresholds may require a party or presidential candidate to have at least 25 percent support in more than half the country, which is difficult for a regional party or candidate. (Kenya 1992)

-dampen possibility of use of pre-election coalitions (Mexico said any coalition had to be in place both in presidential and legislative elections)

-changes in voter registration, candidate nomination, and campaign laws.

PR appears to be more fair than FPTP

but FPTP lowers voter turnout and incentivizes co-ordination between parties

PR with high electoral thresholds, short campaign periods, and low district magnitudes may give dictator a seat bonus with no, or hardly any, benefit to opposition or smaller parties.  (OHES, p. 393)

impact of electoral systems and changes thereof are difficult to predict.

Hungary and Poland Communists would have benefited from adoption of PR,

 but instead they went to mixed member system with about even district seats and top-up, or in Poland's case, majority runoff for the lower house.  (OHES, p. 394)

perhaps the ruling party is under constraints (see below), or mis-estimates its popularity and distribution there, or misunderstands effect of a change, so makes sub-optimal choices. (see "Effects of Rules" below) (OHES, p. 394)

=======

CONSTRAINTS ON GOVERNMENT'S CHOICE OF ELECTORAL RULES -- by OPPOSITION

Governments, including autocrats, sometimes do not make choices about electoral rules unilaterally

threat of rebellion pushed governments to expand the franchise in Western Europe in the late 19th/early 20th Centuries. (p. 394)

Some suggest that the threat of rebellion is what drove the change to PR.

labor uprisings and strikes "effectively threatened the entire nation" in Belgium, Finland, the Netherlands, Sweden and Switzerland.

the strikers explicitly demanded PR, and electoral reform and the adoption of PR was a direct result of these threats. (Oxford Handbook of Electoral Systems, p. 394)

also radicalization of the Socialist Party led to PR. Right-wing forces saw that Socialists were an electoral threat but also a threat to the existing social order. Socialists' radicalization made an outright Socialist victory more problematic and the right-wing forces saw that FPTP made Socialist victory more likely so adopted PR to prevent such a situation. (OHES, p. 394 citing Ahmed 2013, p. 24)

Brought to the bargaining table by the threat of unrest, autocrats may then be constrained by the actual bargaining protocol and the degree to which their opponents can mobilize against them.

Such occurred in 1989 when Communist Party brass gave in to growing political opposition and the economic crisis. (p. 394)

==================

2021 Sened (Welsh Assembly)

However, as in previous elections, the large number of constituency seats won by the Labour Party in the South Wales West, South Wales Central and South Wales East electoral regions, when set against the small proportion of available additional regional seats, means that the Labour Party is over-represented by a margin of four seats, when considered on a proportional basis.

[it is forecast that the 2027 Scottish election will see the Scottish National Party benefit hugely by such overhang seats.]

=========================================

THE PROBLEM OF UNCERTAINTY (OHES, p. 395)

Eastern Europe (1989-1990), initial electoral formation -- leaders in Poland other former Soviet block countries pulled for FPTP thinking it would benefit them as the dominant party or due to name recognition but it backfired.

Algeria -- 1990 government organized local and regional elections and formatted them to be partisan contests, but a new party, the Islamic Salvation Front, took most of the seats.

the government then formatted the first national election the following year in a different way.

both new parties and new rules can have unexpected results.

Authoritarian countries have high degree of "preference falsification" so that adds more uncertainty.

[Edmonton -- strategic voting that occurs under FPTP is a form of "preference falsification."]

also uncertainty about effect of specific systems or rules --

Chile military government and its civilian allies adopted two-seat districts with goal to reduce the number of parties and the factionalization that had been in the pre-Pinochet period.

but the system has produced two party blocks, but not reduced the number of parties that compete and stability is not achieved under the new system. (OHES, p. 395)

====

both Communists and opposition parties in Hungary were unsure how choice of systems would effect the outcome.

Hungary used (or uses) three tiers but this is not due to intent to confuse, as might be supposed, but actually product of bargaining at the time of the transition to democracy in the late 1980s, coupled with inertia. (OHES, p. 32)

Like the system used in Italy 1993-2005, Hungary's electoral system (used in 2018) has both elements of parallel and elements of compensation for the district result. (OHES, p. 32)

=======

EFFECTS OF RULES

Electoral system affect a variety of political and economic phenomena, but mostly electoral rules have effect on party systems, principally the number of parties that run candidates or that are elected. (OHES, p. 396)

FPTP commonly held to create two-party systems

Electoral system rule changes in autocracies produce change in the number of parties, as voters and candidates learn over time about more effective ways to engage in elections.

Parties decrease in number due to both both mechanical and psychological effects (Duverger, 1954)

Mechanical effect - the two largest parties win most or all seats and thus are over-represented in the chamber.

Psychological effect - voters realize that if they vote for third party candidates they are very likely or certain to be wasting their vote, so they instead vote for one or other of the two largest parties. (OHES, p. 396)

=======

Some general rules about party systems under certain District Magnitude (OHES, p. 62)

based on average DM in districts (M) and number of seats in the chamber (S)

a likely seat share for largest party in the chamber is M times S - to the power of -1/8th,[But I can't get this to work]

number of parties in the chamber: M times S, to the power of 1/6th.

the vote share for the largest party: (M times S, to the power of 1/4th +1) to the power of minus 1/2th

number of parties that run candidates: (M times S, to the power of 1/4th +1) to the power of 2/3rds.

smaller rule changes (as opposed to systemic change) may show effects in the short term, and the effect of rule changes are much clearer.

so authoritarian government may feel free to tinker in small ways with elections and stay away from major change.  (OHES, p. 398)

the effect of rules may take a while to bite - as candidates and parties adjust their behaviour, learning from elections as they go,

but even this long process can be prolonged if parties do not have accurate information and if rules fluctuate.

autocracies change electoral rules more often than democracies--

in post-WWII period, 10 percent of autocracies engaged in major rule change, and 23 percent engaged in small rule change

in democracies: the figures are 6 percent, and 13 percent respectively. (from Barbera (2013))

Voters and candidate need to know (other) voters' sentiment, and that is difficult if voters engage in "preference falsification."

Fraudulent elections also create noise.

after change to FPTP, the number of parties declines with each successive election;

after change to PR, the opposite occurs over successive elections (p. 397)

Behaviour within districts is different from across-districts:

FPTP two candidates emerge with each district [barring new parties and perhaps change in voters and sentiment]

but a two-party system may not emerge in the chamber, due to :

-distribution of power across the levels of government,

-the number of chambers,

-appointed members in the chamber,

-whether opposition parties remain independent instead of joining together, or if Independents run and are elected.

many conditions (other than electoral rules) help determine party systems.

boycotts not so common under PR as even small parties have something to lose.

violence is encouraged under TRS as it encourages use of non-electoral means to exercise [take] power.

first round provides info on geographical distribution and leading up to second round of voting, voters in specific area can be targeted. (OHES, p. 398)

==========================================

(Oxford Handbook of Electoral Systems (2018) says there were at that time 16 countries with reserved seats (a combined total from three different sources- none of the sources having the same figure).

None of the reserved seats were for women. [in 2026, some systems do have reserved seats for women.]

Colombia (black communities)

Croatia (several ethnic groups)

India (scheduled tribes and castes)

Jordan (Christians and Circassians

Niger (Tuareg)

NZ (Maori)

Pakistan (non-Muslim minorities)

Palestine (Christians and Samaritans)

Samoa (non-indigenous minorities)

Slovenia (Hungarians and Italians)

Taiwan (aboriginal community).

Kiribati (Banabans)

Romania (19 "small minorities")

Venezuela (indigenous population)

Iraq (Christians, Sabeans, Shabaks and Yizidis).

different countries use different mechanisms of minority inclusion, not just reserved seats.

(intentional drawing of districts (U.S.),

over-rep. of minority regions (U.K.)

lower thresholds for minorities - Denmark, Italy, Poland [Denmark -- I don't know what they talking about]

alternation in power -- Ghana (OHES, p. 166)

federalism in Canada.

consociationalism as in Belgium

encouraging multi-ethnic parties (so achieving minority representation but not through separate ethnic party)

geography of the minority group helps dictate what type of minority inclusion works

but it is not clear why some countries use some mechanisms and not others.

some countries try to amass national unity so undercut ethnic or regionally based parties.

so institute ethnic/regional party bans (Bulgaria and Brazil) (OHES, p. 167

Spatial Ballot Access Requirements -- to be on ballot, parties must show evidence of support (signatures, members or votes) in more regions of the country than their normal base.

Spatial Ballot Access Req. are seen as objective barriers to any party, unlike ethnic party bans.

used to prevent emergence of indigenous parties or used to promote national parties,

Saideman (2002) said PR helped suppress ethnic conflict between 1985 and 1998

Always and Templeman (2012 looked at 106 regimes in 100 countries and found that "PR is positively associated with political violence," these instances point to countries where population was relatively heterogenous, not homogenous (OHES, p. 168)

the same electoral rules perform differently under diff. conditions.

PR helps to suppress ethnic conflict (OHES, p. 168)

(electoral system codes:

1 = majority [perhaps the authors (Schneider and Weiseohomeier 2008) uses the no. 1 and no. 2 terms in diff. way]

2 = plurality

3 = semi-proportional

4 = full PR) (OHES, p. 171)

it is far easier to address ethnic and racial division before, rather than after, society descends into ethnic violence. ("conflict management by electoral systems is likely easier at relatively low levels of conflict." (OHES, p. 170)

Electoral rules are not likely to be enough to head off ethnic violence unless they are assisted by other institutions. [security forces most importantly, IMO]

(other institutions include

formal things -

-decentralization (but in 30 democracies between 1985 and 2000 -decentralization was not effective at minimizing conflict. it produced regional parties that helped to promote conflict.)

-constitutional rights,

-rules on party commitment to democracy

informal arrangements -- consociationalism, and others.

Electoral rules are not likely to be enough due to:

-reformers and electoral engineers having no certainly how new rules might perform.

something like direct election of president (Israel) was expected to be unifying but in reality it led to fragmentation.

we know we can lock in minority representation by having 10 percent reserved seats but that may not keep options open to possibilities,

[ethnic voters may be forced to vote only for ethnic candidates, for example]

-parties and elites will seek to game any electoral arrangement.

-systems that allow voters and parties to have agency may make predicting results of ER more chancy.

due to these uncertain results of changes, electoral engineering becomes electoral art. (OHES, p. 169-170)

Polarization

PR seems to possibly have an effect on lowering polarization, but the effect is modest.

PR sometimes encourages new new parties

PR's electoral permissiveness may allow large parties to move to steal policy from small parties.

PR may have an effect on encouraging parties to clearly state their positions, (OHES, p. 255-56) [which may lead to polarization IMO].

[this seems likely as under FPTP a candidate tries to be all things to all people]

but some studies say electoral permissiveness means large parties take inconsistent stands across policy impressions. (OHES, p. 256)

Women's representation (p. 175-191)

in 2017 women held about 23 percent of seats, twice the share they held in 1997.

party list PR can aid women's election as voters cast vote for parties, not candidates.

PR systems have higher rate of turnover than plurality systems (p. 176)

but according to many, most importantly PR uses MSDs while FPTP uses SSDs

in SSD electing a woman means a man is not elected, while in MSD both can be elected, and parties are functionally encouraged or even required to have gender balance on its slate. (OHES, p. 176)

as small leftist parties encourage candidacy of women, larger centrist parties copy them die to pressures of PR (but those pressures not as strong under FPTP)

Quotas meant to secure rep by women have been adopted in 130 countries due to campaigning by women's groups and also due to international recognition of gender-balanced decision-making. (OHES, p. 177)

three main mechanisms:

reserved seats

party quota parties self-impose rule that so many nominations go to women

legislative quotas - require all parties to nominate a certain share of women candidates

quota of either type are more effective at securing women rep. when they are used in conjunction with closed list PR and high DM. (OHES, p. 178)

quota or reserved seats have secured women rep even when SSDs are used.

the idea that PR is superior to SSF may be due to intervening variables that are distinct from the electoral system - DM, legislative turnover, party rules for candidate selection, and strategies employed by parties at election time. (OHES, p. 179)

bu of course DM is tied to electoral system

and SSDs make it difficult for women

Rule (1987) studies 23 advanced industrial counties and found that those with larger DM elected more women than those with smaller DM

Engstrom (1987) found that in Ireland more women are elected in DM-4 and DM-5 district than in DM-3 districts.

Maitland and Brown 1992: in the US more women are elected in multi-seat district s than in SSDs. (OHES, p. 179)

But other studies say that party magnitude (how many seats a party wins) are more determinant of women rep. than DM.

PM for sure affects the effectiveness of gender-balanced lists, and explains why Tunisia which requires party gender balance on party lists, sees only 31 percent of seats go to women. Women will be less often elected unless the first name on the various lists is a women.

most parties elected just one and had placed a man at the top of the list.

presidential decrees in Argentina determine the quota for gender balance based on how many votes the party expects to win. (OHES, p. 180)

Ballots

closed,

open (parties provide a list of candidates, the names often arranged in alpha order, and voter votes for candidate, the most-popular being elected), or

flexible (voters can vote for a party or for a candidate).

[of course any of those can have reserved seats for women by using Edmonton's old southside guarantee method, although apparently not mentioned anywhere]

closed list PR may facilitate election of women where party leaders place women higher than voters would vote for women,

or may hinder election of women where party leaders place women on the list lower than voters' support.

the performance of closed list over open list where it occurs is likely due to the fact that closed list PR has rules that women must be listed in winnable positions on on the list. (OHES, p. 181)

In Brazil 30 percent of list spaces must be given to women but still women have small share of seats. voters say they support women but the women's small share may be due to parties allocating campaign funds and resources to men's campaigns. (OHES, p. 181)

These considerations show that PR will better produce women rep when gender quota, higher DM , higher PM, and appropriate open or closed ballot structures are also used.

"Contagion" (the spread of innovations of party platforms and organizational practices) is facilitated under PR where small parties can emerge, the political costs of adopting the innovation is less, and the threat of losing votes is more serious. (OHES, p. 184)

study shows that parties are reluctant to nominate women and do so only when rules require and only to the degree that rules require it.

(Finland research says 72 percent of men and 53 percent of women vote for candidates of same gender (OHES, p. 184))

More than 80 countries in the world - including democratic semi-democratic and non-democratic countries use gender quota.

these should be considered full electoral reforms as they have wide impacts :

1. they often are products of social movements

2. as social movements and other other outside actors are important in ER such as gender quotas, motivation other than seat maximization should be considered, such as principles, international pressure and empty gestures. (OHES, p. 183)

3. fight for gender quota means social movements have opening to critique and redefine political values.

4. the historical record of adoption of gender quota shows a range of actors, motivations and strategies. (Dahlerup and Freidenvall 2005)

==========================================

Women's Representation

Open-list PR is good to women in some circumstances

Brazil has brought in gender quota for candidates/nominations. This has actually lessened women's rep. This might suggest voter bias against women but actually parties have fill their female quota with women who are not serious about seeking office.

Spanish Senate elections arrange candidate names on ballot in alpha. order. parties nominate women with alpha-lower surnames and males with alpha-higher surnames - to ensure preferred male candidates are elected. (OHES, p. 184)

where open-list PR used, in three Polish elections, voters improved women's list positions through preference voting.

so more women were elected than parties' lists would have predicted.

same was seen in Czech elections, where open-list PR also used, in elections prior to 2014.

This was after Czech "expanded the number of preference votes and lowered the threshold needed to move up the list"

these and other examples show that it is elite bias and not voter bias that produces women's under-rep. (OHES, p. 184)

[see OHES, page 184]

=====

District Magnitude

districts of 4 to 8 are in electoral sweet spot (says Carey and Hix, 2011)

"legislators in smaller DM often fail to align with the district median voter due to 'coordination failures'...

voters in larger magnitude districts with many seats suffer from cognitive overload that makes it difficult for them to identify congruent representatives and monitor legislative behaviour." (OHES, p. 224)

====

Plurality systems produce coordination failures and political elites may form too many parties.

Thus they see failure for elected rep to conform to median voter -- the median party may not be the most-popular party.

proportional systems elect fairly but then incongruence from median voter occurs when government is formed. the median party is centre of coalition but other coalition partners are parties to right or left, away from the median. (OHES, p. 226)

plurality systems over-represent right -wing parties

Generally speaking, left-wing voters are concentrated in urban settings, and seats are won with massive majorities, thus wasting many votes. (OHES, p. 226)

right-wing parties do better under plurality as rural seats are won with low numbers.

plurality produces two-party system. the middle class votes for the right-wing party of the two, as "the left-wing party cannot credibly commit not to redistribute wealth from the rich and the middle class."

under PR, the middle class will have its own party, which will be a median party. "Given its pivotal position, the centrist party will partner with the left but [be in position to ensure redistribution is] only from the rich." (OHES, p. 226)

under PR left-wing governments form more often and also redistribute more,

and right-wing governments when they do form, also redistribute more than under plurality systems. (Iversen and Soskice (2006) as per OHES, p. 226)

many studies say that level of congruence between median voters and government is same whether PR or plurality system is used. (OHES, p. 227)

but under both types of systems, over time [perhaps he means after government is in power for a while] congruence weakens -

under FPTP, governments wander away from "static" median voters;

in PR median voter wanders away from "static" government (OHES, p. 227)

Chapter 16 -- Roles in the Legislative Arena

Koop and John Kraemer's work referred to on page 328. it says they found that foci of representation depends on how elected, whether in SSD or at-large.

[apparently multi-seat ward not considered]

Royce Koop and John Kraemer, "Wards, At-large systems and the focus of representation in Canadian Cities", Canadian Journal of Political Science 49, no. 3 p. 433-448 (2016) looked at Canadian city councillors.

Abstract:

"Proponents of both ward and at-large systems agree that these systems of election play a role in shaping the representational foci (that is, who representatives seek to represent) of city councillors and, in so doing, affect the quality of local democracy. Canadian cities employ both ward and at-large systems of election, and therefore provide an opportunity to explore the relationship between elective systems and focus of representation. We draw on data derived from both 52 interviews with and a survey of councillors in Canadian cities to test the proposition that cities' systems of election influence the representational foci of councillors. We find strong evidence that ward systems are related to a representational focus on geographically defined neighbourhoods, whereas councillors in at-large systems report prioritizing representation of their cities as wholes."

[but even ward councillor's vote on all questions, even those not affecting their own district,

ward councillors cannot represent all within their own ward.

so IMO this analysis is not deep enough.]

as DM grows, legislators downplay their district emphasis.

but do not necessarily take wider perspective, but instead shift to partisan positions. (OHES, p. 329)

in MMP, list members tend to have stronger partisan orientation than district members.

this was seen in Germany and U.K., Hungary and Romania (p. 329)

[perhaps result of using FPTP in districts]

[looks like Denmark was not in the study group for this or did not show that behaviour]

due to variety of possibilities of mixed member systems, it is difficult to know in general terms what impact a given number of seats in each district or pooling would have on overall representation. (OHES, p. 37)

SSD - candidates strive for personal vote

the smaller the DM the more heterogenous it is and then reps strive to learn the district's opinion [as if district will have one opinion]

greater DM means candidate downplay district foci and adopt partisan stances, as seen in Germany [apparently]

greater the DM candidate do not need to strive to the be most popular (plurality winner)

but with PR in MSD, candidate can try to win just with support from geographical, functional or even partisan sub-part of the district."

OHES says under PR candidate needs much smaller proportion of the vote (p. 330)

[but actually the number of voters in a Droop quota for a DM-3 and for a DM-4 district is almost identical]

diff. electoral system attract diff kinds of candidates and reps.

politicians self-sort, in part based on electoral rules. (p. 332)

some voters want elected members to lead; other want them to listen and act upon instructions. (p. 333)

where voters wants diverge from elected menbers' behaviour, you have political dissatisfaction and disengagement.

voters not too interested in district rep.

as DM grows, members emphasis on district rep grows weaker and begins to line up with voters' expectation. (p. 333)

reps who won with great margins are relatively free from system's incentives

while those whose grip on their seats is weak are constrained to put district opinion highly if the re elected in a district. (p. 335)

growing trend toward individualization in society has dramatically undercut the existence of stable and homogenous social or ideological categories of citizens that underpin legislative roles. (p. 336)

political parties and social collectivities have withered

the pace of social economic and technological change renders the public agenda unpredictable.

new issues such as those involving identity such as gender, race and ethnicity, which are non-territorial, clamor for attention.

to incorporate these issues, political theorists were (in 2018) reviving the notions of surrogate representation ][whatever that is] or even turning to rep. by un-elected private individuals (see Michael Saward, "Authorisation and Authenticity - Representation and the Unelected" Journal of Political Philosophy, 17, no. 1 (2009), p. 1-22)

["rep. by un-elected private individuals" may be reference to citizens assemblies]

But the way in which political posts are captured is the same as they were 50 years ago [in many cases]. (p. 336)

==================

Regarding Plebiscites in Dictatorships (p. 374- )

about 50 plebiscites were held in 1970 and 1990 by dictatorships (countries who scored six or seven on the Freedom House index). (p. 375)

============================

housekeeping https://montopedia.wixsite.com/montopedia/post/pr-best-route-to-electoral-integrity-says-oxford-handbook-of-electoral-systems-2018