Question: Can Northern Ireland in the 1960s be considered to have been a democracy given the prevalence of gerrymandering?

Answer: This is an interesting question and the short (but imperfect) answer is ‘yes’. Based on the definitions of democracy that we have looked at, Northern Ireland would have qualified as such:

• Robert Dahl’s minimalist definition requires only that a country is high in terms of inclusion (i.e. that the bulk of the population can vote) and contestation (i.e. that voters have a meaningful choice over who they vote for), and both of those conditions were met in Northern Ireland.

• The Polity IV scale measures regimes in terms of:
  • the competitiveness of executive recruitment;
  • the openness of executive recruitment;
  • the constraints that exist on the executive;
  • the regulation of political participation;
  • the competitiveness of political participation.

In Northern Ireland in the 1960s, multiple parties competed in elections for the executive (both in local and national elections), anyone could (with minor nomination requirements) become a candidate with a prospect (albeit slim, though this is the case in most (all?) democracies) of becoming part of the executive, the executive was limited (in the sense that it was subject to the rule of law and could be voted out), elections were regulated and subject to the rule of law, and parties competed for votes. So, the criteria for democracy are broadly met by this definition too.

However, as always, things are not that simple. We will also cover the concept of electoral integrity, which is essentially about the fairness of elections. We can easily argue that gerrymandering in Northern Ireland was unfair and, indeed, that it impacted on the ability of the population to meaningfully choose the overall winning party in elections. This was because boundaries were drawn in a way that grouped large sections of the electorate together in some districts (restricting the number of representatives that they could elect) whilst ensuring majorities of smaller sections of the population in other districts (thus increasing the number of representatives that they could elect). Given that the case of Northern Ireland conforms to the above two definitions of democracy but clearly had significant problems of implementation, I would probably describe it as a flawed democracy at the time.

The above conclusion is however, at least in part, a judgement on my part. You could make the case that the gerrymandering was so severe that it meant Northern Ireland was not, in practice, a democracy (i.e. a majority had the value of their votes suppressed in some places whilst a minority in those same areas had the value of their votes exaggerated). Of course, if we go down this route then we need to think about where we would set the boundary of our definition. In other words, how flawed does the implementation of democracy have to be before we conclude that it is, in fact, not a democracy? Would we say that the UK, with a single member district plurality (i.e. non-proportional) electoral system, in which multiple governments have been elected against the expressed wishes of a majority of voters, is not a democracy? Ultimately, this is not very different from the judgement that went into saying that countries that score -6 or below on the Polity IV scale are dictatorships, those that score 6 or more are democracies, and those that score between 5 and -5 are anocracies. In other words, there is always an element of judgement in categorisation, and we should be ready to explain why we reach the conclusions that we do.

Question:  I read in the Huber and Gallardo (2004) article that the “alleged stability of majoritarian systems masks the substantial turnover within the cabinet itself”. I was wondering what specifically was meant by “alleged stability”, please? I was under the impression that majoritarian systems, by not providing tools for dealing with the problems it created (no vote of no confidence), was less stable than consensus democracies or PR systems.

Answer:  The Huber and Martinez-Gallardo article centres on challenging pre-existing definitions of stability (which focus on the time that passes between cabinet formation and collapse), and providing alternative ways of conceptualising stability (focusing on ministers’ experience of being in cabinet, and of holding their specific portfolios). The alleged instability of consensus systems is based on the more frequent collapse of cabinets due to one or more coalition partners leaving government (due to disputes or disagreements between parties). Of course, because consensus systems tend to have PR electoral systems, they are more likely to produce coalition governments (because no single party is likely to receive a majority of votes in elections), which then makes it more likely that there is a coalition partner to leave the government and cause the cabinet to collapse. As the article mentions, the French Third Republic, Weimar Germany, and Italy are taken as evidence of this tendency.

However, by presenting their alternative measures of (in)stability, they show that majoritarian systems such as the French Fifth Republic are not as stable as proposed. It is thought that they should be stable because they are more likely to produce majority government that lasts for the whole of its term (i.e. the cabinet will not collapse because there is no coalition partner who can quit). However, as Huber and Martinez-Gallardo show, majoritarian systems are not as stable as thought when we consider the experience of the ministers in cabinets (both in terms of the time they have spent in cabinet, and in charge of their specific portfolios). This is because, whilst cabinets may be less likely to collapse in majoritarian systems, they may well be reshuffled at the discretion of the prime minister (or equivalent), meaning that the turnover of cabinet ministers can be high.

So, their argument does not relate to the lack of a vote of no confidence in particular. Crucially, the presence of absence of a vote of no confidence is actually to do with whether the system is presidential (no vote of no confidence) or parliamentary (vote of no confidence), rather than strictly whether it is majoritarian or consensus orientated. Remember, of course, that a parliamentary system may be more majoritarian (e.g. the UK) or consensual (e.g. Germany) based to a notable extent on their electoral system. A presidential system will always have the additional consensus requirement that the president must approve bills passed by the legislature but may still be more or less consensus orientated depending on the electoral system used for the legislature.

Question: Why does median voter theory require that there be only two parties? Is it not always electorally efficient to align with the preferences of the median voter?

Answer: This is a good question and the answer centres on parties outflanking the party that positions itself at (or nearest) the median voter. So, let’s imagine a two-party system with a left party and a right party, in which the left party has positioned itself almost at the median voter whilst the right party is further away. Of course, the latter party picks up the votes of everyone in the electorate to the right of their position, as well as everyone to their left who is closer to them than the (almost-median) position of the left party. The left party does the same for all voters to their left and everyone between them and the median voter, as well as picking up the crucial median voter (because they are closer than the right party).

Now let’s imagine that a new left party enters the electoral arena to represent those voters on the left who feel alienated by the central position of the original left party. The new party will pick up all the votes to the left of their position, and all the votes of those to their right who are, nonetheless, closer to their position than that almost median position of the original left party. So, the left vote is split and the right party, which remains unchallenged on the right, can win without positioning themselves at the median voter. The right party can still count on everyone to their right (and those to their left who are closer to them than they are to the median) but the original left party cannot count on the votes of everyone to their left. In other words, they have been outflanked on the left as well as on the right, and there are not enough votes around the median to deliver victory. Of course, this depends on the precise positions of the parties because there are, in a normal distribution, more votes to be won around the median voter. So, if the new left and original right parties move to the extremes it still makes sense for the original left party to go to the median, where there are more votes.

Of course, all of this turns on other assumptions including, for instance, that voters turn out to vote. If parties moving to the median would alienate voters on the left or right to the extent that they stopped voting then that would also undermine the simple calculus involved in moving to the median voter (hence the idea of shoring up the base).

First question: In presidential systems like the US, does the president (as part of the executive and not the legislature) make any laws? E.g. how did the Trump administration enact the ‘travel ban’ and the 2017 tax cuts?

Answer: The short answer here is: no, the president does not make any laws. In practice, if they want legislation to be passed then they must find a friendly representative (usually from their own party) to introduce it into the legislature. Even then, though, it can be hard to pass legislation because party discipline is often weaker in presidential systems (because the president is outside the legislature and cannot promote representatives, or threaten an early election). This means that, often, the support coalition for legislation must be built on a case-by-case basis, and representatives can extract concessions for their support (meaning that the legislation often gets altered from the president’s ideal form before it is passed). With the preceding in mind, the ‘travel ban’ (I assume you mean in 2017 rather than in relation to coronavirus) and 2017 tax cuts might seem inexplicable. However, President Trump used an executive order to pass the former; executive orders are a constitutionally defined power of presidents to direct the practice of government departments, and do not constitute legislation. In fact, they can be overturned by legislation passed in the legislature (Congress, in the case of the US), and are subject to judicial review as well. As for the tax cuts, they were passed as legislation by the Republican Party in Congress, and President Trump had to rely on his allies in the party to introduce the legislation and ensure that it passed through both chambers.

Second question: In parliamentary systems like the UK, what powers do the Prime Minister and the Government have besides voting in parliament (as opposition MPs also do)?

Answer: Some very important ones: the power to introduce legislation, the power to call an election and, in the case of the PM, the discretion to promote or demote MPs to and from ministerial positions. In the first of those cases, most of the time legislation is introduced in a parliamentary system it is done so by the government, so they act as the legislative agenda-setter. In the second case, the PM and government usually have the ability to call an election (fixed parliamentary terms in the UK are a recent innovation and may be removed under this government), which of course can be used to ensure that their MPs vote in favour of their legislation (because many MPs would rather avoid fighting early elections). And finally, in the third case, the PM (and the party whips who support them and act as enforcers) can use promotion or demotion to ensure loyalty amongst MPs. So, all of these powers function through the parliament, which shows the nominal importance of the institution, and hence the reason for calling it a parliamentary system (though as Professor Hix pointed out, the power lies with the government!). Outside of parliament, the government, of course, also has some discretion over how government departments are run and how policy is implemented, or in other words the power to direct the day-to-day functioning of government.

Third question: Presidential systems are said to have more veto players than parliamentary systems. Is that only because the executive and the legislature are mutually independent, or are there other reasons (other veto players)?

Answer: Yes, the most important additional veto player is the president themself, who has a constitutionally defined veto; such a powerful additional block on legislation generally does not exist in parliamentary systems (as you say, this relates to the mutual independence of the executive and legislature in presidential systems). Additionally, presidential systems are often set up to have two powerful chambers in the legislature, whereas parliamentary systems may be unicameral or have a weaker upper chamber (as is the case in the UK). Additionally, there are then other things that may be associated with presidential systems, such as federalism, that introduce veto players. Of course, such things are not exclusive to presidential systems (e.g. Germany is a federal parliamentary system), so the association may emerge from historical or cultural context (e.g. countries in Latin American that adopted constitutions similar to, or influenced by, that of the US, thus implementing presidentialism and federalism) and thus are not integral to presidentialism.

Question: I was wondering if you could explain the recent political turmoil in several Southern American countries [at the time of writing: 11.11.2019] – where did come from and why, and how do you think this will play out for democracy?

Answer: Thanks a lot for the message; these are great (and big) questions!

I’ll take two examples of what you’re talking about as illustrative:

 – Evo Morales stepping down as president of Bolivia, in response to protests that emerged from a disputed presidential election: https://www.nytimes.com/2019/11/11/world/americas/evo-morales-renuncia.html

 – Chile’s president, Sebastián Piñera, agreeing to the drafting of a new constitution following protests sparked by rising public transport costs: https://www.nytimes.com/2019/11/11/world/americas/chile-protests-new-constitution.html

In each of the above cases, there is a specific issue identified as the cause of the unrest: in Bolivia, it was concerns over the president’s disregard for the constitution, and term limits in particular, and the threat that it posed to the country’s democracy; in Chile, it was frustration over the constitutionally defined limits on the state to intervene in the economy (stemming from the era of the Pinochet dictatorship), which has implications for inequality and the cost of living.

The above specific causes are important, and should not be discounted, but we are often also interested in common underlying factors that lead to phenomena (such as protest or government breakdown) in multiple cases. In this light, it is interesting to refer to the readings from our week on ‘Regimes: Presidents, Ministers, and Parliaments’, a number of which relate to impact of the choice between presidentialism and parliamentarism on government and regime stability. Drawing on his observation of regime instability in South America at the time, Linz argued that the problem was the adoption of presidential systems that caused deadlock between legislatures and executives. Horowitz argues against this, noting that Linz’s case selection is limited and that stable presidential systems can be designed if electoral systems are chosen wisely. As we will see in that week’s lecture, the combination of presidentialism with an electoral system leading to many parties in the legislature (as is the case in Chile) may be problematic because it creates too many veto players (causing gridlock).

There is also an interesting consideration about what causes protests to spread once they have begun. If there has been underlying discontent for some time, why do lots of people pick that moment to protest? There are lots of answers to this but, from our week on ‘Interest Groups and Social Movements: Why Mobilise?’, Kuran has argued that there could be a bandwagon effect (as more people protest, other people pass the threshold at which they become willing to do so as well) whilst Farrell examines the ways in which the internet (and social media) may play a part.

So, to summarise the answer to your first question, there may be long-term underlying causes such as the design of the constitution (and challenges that may pose for effective government) coupled with immediate political or policy concerns (a decreasingly democratic president, or inequality and the cost of living) that drive a protest bandwagon effect assisted by the ease of internet-based communications.

As for your second question, I am reasonably optimistic about the outcomes for democracy in the cases of Bolivia and Chile. The military had refused to defend the outgoing president in Bolivia but has not attempted to become involved in the government. In Chile, the protests are not anti-democratic and have forced compromise (though maybe not enough) from the president. In both cases, it seems like we are observing the important role of protest in democracy. There may be other countries however, such as Venezuela, where there is less cause for optimism.

I hope this answers your questions!

Question: Student groups were asked to find recent literature shedding light on whether democracy affects economic performance, vice versa, both, or neither.

Answer: Many thanks for your very productive response to the task in the classes on Monday, in which you were asked to find a recent article shedding light on the causal relationship between democracy and economic performance (in response to the Przeworski and Limongi (1993) article). Your sterling work means that we have the makings of a good, up-to-date, literature review on the topic, which is now a shared resource. In alphabetical order of first author surname, the sources suggested [some by multiple groups] were:

• Acemoglu (2014). ‘Does democracy boost economic growth?’
• Baklouti and Boujelbene (2017). ‘An econometric study of the role of the political stability on the relationship between democracy and economic growth’
• Baklouti and Boujelbene (2018). ‘The Nexus Between Democracy and Economic Growth: Evidence from Dynamic Simultaneous-Equations Models’
• Brezis and Verdier (2014). ‘Geography, Economics, and Political Systems: A Bird’s Eye View’
• Chatterjee (2011). ‘Democracy and economic transformation in India’ Chapter Understanding India’s New Political Economy, edited by Ruparelia, Reddy, Harriss, and Corbridge.
• Doucouliagos and Ulubaşoğlu (2008). ‘Democracy and Economic Growth: A Meta‐Analysis’
• Dizikes (2019). ‘Study: Democracy fosters economic growth’
• Gerring, Bond, Barndt, and Moreno (2011). ‘Democracy and Economic Growth: A Historical Perspective’
• Gerring, Kingstone, Lange, and Sinha (2011). ‘Democracy, History, and Economic Performance: A Case-Study Approach’
• Gründler and Krieger (2016). ‘Democracy and growth: Evidence from a machine learning indicator’
• Hanson and Gallagher (2012). ‘Winning Coalition Size and Economic Performance: The Selectorate Theory and East Asian Developmental States’
• Heo and Tan (2001). ‘Democracy and Economic Growth: A Causal Analysis’
• Ishtiaq, Majeed and Sohail (2016). ‘Financial Sector, Democracy and Economic Growth: A Panel Data Analysis’
• Knutsen (2012). ‘Democracy and economic growth: A survey of arguments and results’
• Patti (2009). ‘Globalization, democratization and economic growth’
• Piątek, Szarzec, and Pilc (2013). ‘Economic freedom, democracy and economic growth: a causal investigation in transition countries’
• Pozuelo, Slipowitz, and Vuletin (2016). ‘Democracy does not cause growth’
• Shabbir (2017). ‘Corruption, Democracy and Economic Growth: Does Conditionality Matter?’
• Wu (2012). ‘When Is Democracy Better for Economic Performance and when Is It Not: the Interaction Between Polity and Structural Factors’

This looks like an excellent list to me so now we all just need to find time to read them all! Then, perhaps, we will finally be able to tell people whether democracy leads to improved economic performance, vice versa, both, or neither.

First question: Selectorate theory assumes that any leader’s goal is to stay in office and maximise rents. Are there clear cases in the world today of autocrats that do not fit that description? I.e. ‘benevolent dictators’ who deliver more public goods than are needed to satisfy their winning coalition? If so, why might that be?

Answer: The examples that spring to mind in answer to this question are a number of regimes in the Middle East that have high levels of public good provision (at least in some senses) despite being dictatorships with small winning coalitions (e.g. Saudi Arabia). The explanations in this case may be related to the immense wealth that such regimes possess; even if a very large proportion of that wealth is retained by the ruling family, they can still afford to spend significant sums on public goods. More specifically, increased spending by such regimes was observed in response to the Arab Spring, and I think this points towards an explanation that lies outside selectorate theory. The starting assumption of the theory is that the disenfranchised population has little or no say in the regime’s survival, for which only the selectorate and, more so, the winning coalition are key. However, it is possible to imagine the disenfranchised section of the population in dictatorship mobilising to pose a threat to the regime (or even just that the threat of such mobilisation exists), and that the regime would respond to this, for instance by increasing public spending. Further to this, and more broadly, it is worth noting that the authors have responded to previous criticism but that selectorate theory has come in for additional criticism subsequently, for example in the following articles:

• Kennedy (2009). ‘Survival and Accountability: An Analysis of the Empirical Support for “Selectorate Theory”.’
• Arena and Nicoletti (2014). ‘Selectorate theory, the democratic peace, and public goods provision.’
• Gallagher and Hansen (2015). ‘Power Tool or Dull Blade? Selectorate Theory for Autocracies.’

Second question: How does selectorate theory fit with the fact that there exists applied research in e.g. development economics which recommends policies to alleviate poverty, effectively spend on public health, etc? When autocrats take such advice, are they just responding to pressure from their winning coalition which has grown?

Answer: I would say that advice stemming from such research is likely to emanate from international organisations and this indicates another factor that is external to selectorate theory: external pressure. It is difficult to conceptualise such pressure, which may come with important economic implications in terms of aid or trade, within the selectorate theory framework, though this does not necessarily mean that the two explanations are mutually exclusive. To the extent that autocrats respond to such advice when it emanates from within their countries, we might think that it would be a reflection of the propositions of selectorate theory. Of course, selectorate theory does not suggest that autocracies with small winning coalitions will never provide public goods, just that they are less likely to do so. So, autocrats with small winning coalitions may respond to research-led advice on multiple occasions but, on average, still do so less frequently (and effectively) than autocracies with large winning coalitions (or democracies).

Question: I was wondering if you would mind explaining the stats terms below briefly (as the online sources I have found on this topic resort to rather convoluted language in their explanations) in the context of the Przeworski and Limongi (1993) paper (‘Political Regimes and Economic Growth’)?

• Simultaneity: the causal relationship between democracy and economic development is unclear because it could go both ways, which makes researchers’ jobs very difficult?
• Attrition bias: there are existing differences between the original state of treatment and control groups that are impossible to avoid in selecting case studies?
• Selection bias: the selection criteria of choosing regimes based on their growth level, dependent on survival (why specifically was this criteria chosen?) is problematic, and it’s relevant to exogenous factors (ones that are taken as a given and not explained in a model) and endogenous factors (ones that are explained by variables in a model)?

Answer: Thanks a lot for the question, which is a good one and particularly useful because I think a number of students found these terms slightly unclear.
Simultaneity: your definition here is correct. We observe a correlation in which economic development is related to regime type (i.e. democracy or dictatorship) but it could be that that the former leads to the latter, or the latter leads to the former. Or, indeed, it may be that reaching a certain level of economic development prompts transition to democracy and also affects subsequent growth rates (that are then attributed to democracy, but actually stem (at least to a degree) from passing a threshold of economic development). As you say, these competing plausible explanations for the correlation are not easy to distinguish between empirically.

Attrition bias: this is different from the definition that you suggest. Let’s start with an observed correlation between regime type and economic growth. It is easy to assume that such a correlation is because the regime type affects the growth. However, it may actually be the case that different regimes have a greater or lesser propensity to survive when confronted with positive or negative circumstances. So, (drawing on the article) let’s imagine a scenario in which dictatorships are more likely than democracies to collapse when faced with poor economic growth. If we measured growth and regime type in this scenario it would look like democracies lead to worse growth than dictatorships. In fact, what is happening is that dictatorships with bad economic growth collapse, and the democracies that emerge from them inherit poor growth. At the same time, democracies continue through periods of poor growth and have those figures included in their averages whilst, of course, dictatorships only have good growth figures in their averages (because they collapse when they have poor growth). Thus, overall it looks like dictatorships promote higher average growth but, in fact, they have higher attrition rates (hence attrition bias) when faced with poor growth.

Selection bias: I interpret this to be the overarching problem (encompassing the above two problems) that Przeworski and Limongi are concerned with, based on the following quote:

• ‘If democratic regimes are more likely to occur at a higher level of development or if democracies and dictatorships have a different chance of survival under various economic conditions, then regimes are endogenously selected [hence selection bias].’ p. 62.

The point here is that we cannot select an unbiased set of cases (i.e. one that is not beset by problems such as simultaneity or attrition bias). We cannot observe each country as it would be if it was the same but with a different regime type, and if we try to find an equivalent country we see that the similarity between the two countries means that they share regime type and economic growth as well. So we cannot observe the impact of changing regime type in otherwise identical circumstances. Another way of thinking about this is that the factors affecting regime type and economic growth are not distributed randomly across the world so every time we look at countries with a particular regime type we automatically look at countries with other shared characteristics that undermine our attempts to investigate the effect of regime type on economic growth.

In one of the classes this week, I was asked whether there has been any research on the impact of polling on public opinion and behaviour. The answer is a resounding yes! I asked my former polling colleagues and current academic colleagues, and they suggested the following:

• Bursztyn, Cantoni, Funk, and Yuchtman (2019), ‘Polls, the Press, and Political Participation: The Effects of Anticipated Election Closeness on Voter Turnout’: https://home.uchicago.edu/bursztyn/Bursztyn_Cantoni_Funk_Yuchtman_20190208.pdf
• Utych and Cam (2014), ‘Viability, Information Seeking, and Vote Choice’: https://www.journals.uchicago.edu/doi/abs/10.1017/S0022381613001126?mobileUi=0&journalCode=jop
• Morton, Muller, Page, and Torgler (2015), ‘Exit polls, turnout, and bandwagon voting: Evidence from a natural experiment’: https://www.sciencedirect.com/science/article/pii/S0014292115000483

I also undertook a quick search myself and turned up these articles:

• McAllister and Studlar (1991), ‘Bandwagon, Underdog, or Projection? Opinion Polls and Electoral Choice in Britain, 1979-1987’: https://www.journals.uchicago.edu/doi/abs/10.2307/2131577
• Irwin and van Holsteyn (2002), ‘According to the Polls: The Influence of Opinion Polls on Expectations’: https://www.jstor.org/stable/3078698?seq=1#page_scan_tab_contents
• Restrepo, Rael, and Hyman (2009), ‘Modeling the influence of polls on elections: a population dynamics approach’: https://link.springer.com/article/10.1007/s11127-009-9427-x

Hopefully that’s enough to sate your interest, at least for the time being!

Question: If, historically at least, left-wing parties are likely to have a higher proportion of their elected representatives being women, why is it the case that O’Brien (2015) finds no influence of party family (other than Green (positive) and Communist (negative)) on the likelihood of parties selecting female leaders? In other words, if left-wing parties have more elected representatives who are women, why are they not more likely to select leaders who are women?

Answer: The first thing to say is that, based on a quick (and I really do mean quick) search via the LSE Library, this question does appear to expose a gap in the literature. In other words, this might be an opportunity for someone who is interested in the topic to write a research paper; it’s always good to start such a project with an interesting conundrum, which I think this constitutes. However, given what we do already know, I would suggest two possible answers to the question:

1. On a relatively mundane, technical front, we should remember that O’Brien specifically measures the situation in 11 OECD countries from 1965 to 2013. So, it could simply be case selection (i.e. the countries she focuses on) and the particular analytical approach adopted by O’Brien that points towards left-wing parties not being more likely to elect leaders who are women. So, perhaps the observation that left-wing parties have higher proportions of their elected representatives who are women was based on observations relating to more, fewer, or different countries, or a wider, narrower, or different time period. It would be interesting to get hold of O’Brien’s data, code an additional variable focusing on the proportion of elected representatives who are women, and see whether left-wing parties are different in this regard within the countries and time period that she focused on. That said, it should be pointed out that Tripp and Kang observe that ‘strength of Left-leaning parties is statistically significant [in relation to the proportion of elected representatives who are women] only when we omit quotas and regional controls,’ and argue that this suggests left-wing parties become less important in promoting women’s political representation as a wider array of parties adopt quotas. So, perhaps there’s no conundrum after all.
2. If there is, indeed, still a conundrum here then a possible answer may lie in the blog that I cited as a contemporary example in the classes, which highlights that there are barriers facing women at each step of their political careers: recruitment to parties, selection as candidates, election as representatives, and then fulfilling their roles as elected representatives. The blog also highlights the importance of gatekeepers (i.e. important local or national decision-makers) in political parties, who have the power to implement programmes (be they focused on training, quotas, or other support) that remove some of the barriers faced by women (and indeed, may have the power to promote women themselves). If we apply this explanation to our conundrum then it seems eminently plausible that, within the context of left-wing parties, even if women overcome the barriers to becoming elected representatives (which may be lower in said left-wing parties, at least historically) they may still face other barriers to becoming party leaders. There may be different gatekeepers (perhaps in the form of party selectorates) who have different priorities for leader selection than the gatekeepers who they faced when becoming elected representatives. Perhaps, adequate training, opportunities to gain relevant experiences, or support structures (e.g. mentoring) are not available. So, in short, it is possible that left-wing parties have been good at increasing the proportion of their elected representatives who are women whilst also failing to alter institutional and cultural barriers to those women moving up to leadership positions.

In line with the second of the above explanations, and for those who are interested in this topic, I have found a few articles on the topic:

• Cheng and Tavits (2011). ‘Informal Influence in Selecting Female Political Candidates.’ This focuses on the importance of informal gatekeepers in selecting women into party positions.
• Verge and de la Fuente (2014). ‘Playing with different cards: Party politics, gender quotas, and women’s empowerment.’ This focuses on whether increased representation of women leads to greater power for women within parties.
• Jalalzai and Krook (2010). ‘Beyond Hillary and Benazir: Women’s Political Leadership Worldwide.’ Giving a helpful overview of the topic.
• Jalalzai (2008). ‘Women political Leaders.’ Similar to the above.