By Drude Dahlerup

1. Introduction

In recent years we have been celebrating the centenaries of women’ssuffrage in the various Nordic countries. This has prompted us to reflect on where we stand today from a global perspective. The Nordic countries have, for a long time, been world famous for their relatively high proportions of women in politics. But people in other parts of the world often ask us how many years it actually took the Nordic countries to achieve such high levels of female representation.
 
It is precisely the instances of the many 100-year jubilees that force us to admit, as an answer to the above question, that it took us each about a century to reach the current level of between 38 and 47 per cent of female representatives in the parliaments and between 32 and 42 per cent in the local councils in the Nordic countries. Today there is, however, great impatience all over the world in respect of the under-representation of women. On an average, men have 81 per cent of the places in the world’s parliaments, and women only 19 per cent (www.ipu.org). Not least after the adoption of the UN Platform for Action at the 1995 World Conference on Women in Beijing, it is obvious that many of the new democracies in the world are not prepared to wait as long as 100 years to achieve gender balance within their political systems. Today the Nordic region, with its “step-by-step” model, is no longer the only possible alternative for the development of women’s representation in politics.
 
Two features have, for a long time, characterised the development of women’s representation in politics in the Nordic countries. Firstly, the Nordic countries have had absolutely the highest female representation in the world, which has made waves globally. The Nordic region has for a long time had a positive reputation in the world when it comes to gender equality.
 
Secondly, since World War II we have experienced a consistent increase in the female proportional numbers among the members of the political assemblies. This has been described as a continuous and advancing process (Christensen & Damkjær 1998). This success story has contributed to the notion that gender balance in politics, perhaps even equality between men and women in general, will “come in due time”. Is it not part of our faith since childhood that gender equality is advancing? The official gender equality policy has thus also been built on such optimism.
 
The Norwegian researchers Hege Skjeie and Mari Teigen speak of the “travel metaphor” in the notion of gender equality in the Nordic countries – we are on our way forward (2003). In this article these two characteristic features will be explored in more detail. Firstly, the developments in the Nordic countries are analysed from a global perspective. Why is the leading position of the Nordic region in terms of gender equality about to disappear? Secondly, the discussion will focus on how we should interpret the historical development of women’s representation in the Nordic countries: Can we expect a continually increasing female representation? Is there an invisible glass ceiling?
 
It will also be discussed why at present there is such a large difference between the Nordic countries in, for example, the use of quota systems. The third part of the article will focus on this empirical development in relation to four partly contradictory theories on the development of women’s representation: According to the theory of patriarchy, the male dominated society reproduces itself continuously, and when women finally get access to an institution, the power evaporates. An opposing approach to this is the time lag theory, which gives a more positive picture, according to which women will slowly but surely be integrated into the power elite. The saturation thesis, for its part, says that there will be certain weariness before gender equality is fully achieved, which kicks in at about the level of 30 per cent. Lastly, the significance of public debate and pressure from women’s movements will be discussed.
 

2. The Nordic countries are about to lose their leading position

For a very long period the Nordic countries, together with the Netherlands, were alone at the top of the ranking list of women’s representation in various countries. First, Finland was number one, then Norway and finally Sweden. Today, the Nordic countries are about to be caught up by a number of other countries, when it comes to women’s representation in their political assemblies. The Nordic leading position is challenged by other European countries, but mainly by countries from the global South.
 
This is a new and, in many ways, positive development. Table 1 shows that today, Rwanda is the leading country on the world ranking list. Generally, the increasing trend globally is modest. Ten years ago, the global average was 13 per cent; today it is 19 per cent. This can be seen as proof of progress. But is can also be regarded as an illustration of how very slow progress is. However, elections take place at certain intervals. If we look only at the countries with the most recent elections, a somewhat better development is discerned. At the same time, the differences between the regions of the world are about to be leveled (www.ipu.org).
 
Three interesting perspectives can be seen in Table 1. First of all, the table shows that not only Rwanda, but also other countries from the global South, i.e. Argentina, Mozambique and South Africa, now have a female representation of over 30 per cent in their parliaments, thus challenging the Nordic countries. Secondly, the table shows that most of the countries at the top have an electoral system based on proportional representation (PR). Thirdly, the table shows that several of the top countries use some types of quota systems.
 
 
            

Electoral system: PR: Proportional representation including party lists with multiple candidates. Mix: Mix of proportional representation and single-mandate constituencies. FPTP, first past the post: Single-mandate constituencies in which each party only nominates one candidate, and where the candidate receiving the majority of the votes wins the constituency.

Quota types: Legal quotas are included in the constitution or a law, typically an electoral law or a party law. Party quotes: Voluntary quota system passed by the party itself, e.g. a minimum of 40 per cent on that party's list of candidates must be women. A country is listed as having party quotas if only one party represented in Parliament has quotas. While many quota systems target the proportion of women/both genders on the parties’ lists of candidates at elections, the so-called reserved seats quotas which are always legally regulated reserve a number of seats for women. At local elections in India, for instance, a third of the elected candidates must be women. In Rwanda, two women must be elected for each polling district (Dahlerup & Freidenvall 2009). 

Sources: Interparliamentary Union (2009) www.ipu.org; International IDEA and Stockholm University (2009) www.quotaproject.org; official election statistics. The numbers indicate the proportion of women on Election Day. Some variation may occur compared to www.ipu.org, the latter including certain substitutions between elections.

A country indicated as having a quota law means that the quota rules are included in the constitution or electoral law or party law. Voluntary candidate quotas, on the other hand, mean that individual parties themselves have introduced quotas for their own candidate lists at elections. Research into quotas has shown that voluntary party quotas have typically been started by parties in the centre or to the left of the political spectrum, possibly with a contagious effect on other parties. In a number of cases, certain parties have introduced voluntary quotas, but later the quota system has been introduced through legislation for all parties; this has happened in, for example, Belgium and recently also in Spain and Portugal (Dahlerup & Freidenvall 2008; EUPARL 2008).
 

2.1 Quotas – a global trend

Today, it is impossible to talk about development of women’s representation in politics without involving the issue of quotas. Gender quotas is a new, exciting field of research, since quotas are associated with so many central areas of democracy theory and feminist theory. Empirical quota research has explored a number of themes – for example, the geographical diffusion of quotas and the various quota discourses in different parts of the world. The significance of various types of quota systems has been analysed, and studies have been conducted into the often troublesome implementation of quotas and on the impact of quotas in both quantitative (that is, numbers of nominated and elected women) and qualitative (i.e. in relation to the issue of women’s influence and power) terms (Caul 1999; Htun 2004; Dahlerup & Freidenvall 2005; Norris 2004; Krook 2004; Dahlerup 2006; Dahlerup & Freidenvall 2008; Teigen 2003; Dahlerup & Freidenvall 2009).
 
The use of gender quotas within politics has spread like wildfire over all the continents in the last decade. Even if quota regulations are often controversial, it is precisely this method that has spread – as a means to swiftly change an unwanted under-representation of women in public assemblies. Presently, quotas are used in about half of all the countries in the world, if we combine the approximately 50 countries where quotas have been introduced through legislation and/or a change of the constitution, and the approximately 50 other countries where parties represented in parliament have voluntarily implemented quotas for their own lists of candidates. This is a surprising development in view of how controversial quotas are as an instrument of gender equality policy (Dahlerup 2006; EUPARL 2008; see also the global overview on www.quotaproject.org).1
 

2.2 New explanations

The increasing use of gender quotas in politics challenge the prevalent theories as to why some countries are at the top when it comes to women’s representation. Differences in socio-economic development have previously explained most variations in women’s political representation, although the colour of governments and the contributions of various actors – not least the strategies and strength of women's movements – have been important as supplements to the structural explanatory factors. The nature of each electoral system is an additional important factor for the variations in the proportions of women. While female representation amounted to an average of 20 per cent in countries with proportional representation, it was only 11 per cent in countries with plurality/majority elections, such as the British electoral system, and, finally 14 per cent in countries with mixed electoral systems (Norris 2006). However, the influence of the electoral system seems to be smaller in non-Western countries than it is in Western countries (Norris 2004).
 
Since it is difficult to construct a quota system that suits elections in first-past-the-post systems (FPTP) – how can a requirement of 30 or 40 per cent female candidates work, when the parties only nominate one candidate in each constituency? – the differences between female representation in the two electoral systems will probably increase in future.2
 
Research has shown that while three quarters of the countries with proportional elections have quota systems, there are quotas only in one third of those countries that have plurality/majority systems (FPTP) constituencies represented by one person (Dahlerup 2007b). It would be a mistake to assume that quotas solve all the problems that women encounter in politics. Quota systems are not a miracle solution and cannot be used alone as a measure for gender equality. It should also be noted that from a global perspective, high female representation has also been achieved without quotas, as can be seen in the cases of Finland and Denmark in Table 1. It is important to underline that many different types of quota systems are used around the world. Research on quotas has shown that quotas can remain a purely symbolic gesture, unless the chosen type of quota system matches the electoral system of the country. On the other hand, quota systems have, given certain conditions, resulted in historical leaps in women’s re presentation, as when Cost Rica moved from 19 to 35 per cent of women in parliament in one single election in 2002, after the introduction of quotas through legislation. Several circumstances must be in place in order for a quota system to have the desired effects: 1) the type of quota system introduced matches the electoral system, 2) the system includes rules on the gendered ranking order of the lists (a requirement of 40 per cent women on the candidate lists does not result in more women being elected, if they are all placed at the end of the list!), and 3) sanctions in case of non-compliance with the quota rules. Legal sanctions are, naturally, restricted to quota systems regulated by law. But a party with voluntary quotas can put pressure on its local organisations, which, in most cases, are those who decide on the nominations. Quotas regulated by law or the constitution is the main form of the quota systems in Latin America, but it has now also reached Europe, where gender quotas for electoral lists have been introduced by law in Belgium, France, Spain, Portugal and several countries in the Balkan region (Dahlerup & Freidenvall 2008; EUPARL 2008).
 
The development has been particularly pronounced in a large numbers of post-conflict countries. In most countries today, there is an attempt to actively involve women in the efforts to further national reconciliation after catastrophes such as genocide or civil war. In Rwanda, number one on the global ranking list, various women’s organisations were active in the process of developing the constitution, and quota rules were included in the new constitution. International organisations have contributed, in Rwanda as in a number of other countries, to creating a pressure on political leaders for increased inclusion of women and other underrepresented groups. But without the local women’s organisations’ active work and mobilisation of women, no long-term changes will be the result. In countries such as Uganda and South Africa, there has also been some success in the attempts to include women in the reorganisation of the country (Tripp, Dior & Lowe-Morna 2006). The international community has strongly supported this development. The requirements for affirmative action and active measures included in the CEDAW Convention and the Platform for Action from the Beijing conference have contributed to giving increased legitimacy to the claims of national women’s organizations for gender balance in politics (Krook 2004; Dahlerup & Freidenvall 2005).
 
There is, however, also much discussion about the issue of what many see as an inclusion of women “from above”, that is, without sufficient mobilisation of women to create a permanent change. The quota systems in Afghanistan and Iraq are such controversial examples. Generally gender quotas are today used in elections in all types of political systems: democratic, semi-democratic as well as non-democratic (Dahlerup 2007b).
 
In the Nordic countries, we have never experienced such quick leaps of between 10 and 20 per cent units in women’s representation. The largest jump has been one of between 8 and 9, but usually the increases have been between 2 and 3 per cent units; a gradual, step-by-step-development which is typically Nordic. But while the proportions of women in the Finnish and Icelandic parliaments leapt upwards in the latest elections, both Denmark and the Netherlands – which can both be described as “glass ceiling countries” are steadily moving down on the global ranking list. Ten years ago Denmark was number two on the ranking list, today it holds the tenth place, see Table 1. The fact that the Nordic region is no longer alone at the top, and that some of the Nordic countries are about to be overtaken by a number of developing countries, is extremely interesting. This represents a challenge to our self-image in the Nordic countries.
 

3. A steadily progressive development?

The actual political development in the Nordic popularly elected assemblies has strongly contributed to the formation of a narrative about the Nordic region as a group of countries steadily progressing towards gender equality. Since World War II, continuous advancement in female representation has been the normal situation in the Nordic countries. Not least when it comes to female representation in politics, the Nordic region has become world famous in circles interested in gender equality.
 
 
Sources: Official election statistics from the five countries. From Raaum 1999:32. Updated. NA = not applicable. Danish numbers based on 179 MPs, i.e. including the Faroe Islands and Greenland. Only lower chamber.
 
Table 2 shows the development of female representation in the Nordic parliaments. The table as such illustrates the perception of a steadily progressing – perhaps irreversible? – development in the form of achievements reached in the course of history. The table shows how slowly things happened in the beginning, after women had gained the right to vote. Only Finland from the very beginning reached 10 per cent of women in parliament, while women’s representation in the other Nordic countries remained at 2–4 per cent throughout the inter-war period. Denmark experienced a historical low point in the 1943 election, when only two women (1 %) were elected into the Danish Folketing. Only in the 1970s did the development actually gain momentum in the Nordic parliaments, with the exception of Iceland, which lagged somewhat behind the other four countries. It was also in the 1970s that the great breakthrough for women’s representation happened in Nordic local politics generally. After this, there was a constant increase in female representation in local governments simultaneously in all Nordic countries. However, Iceland has always lagged a little behind in its development, only reaching a breakthrough in the 1980s. On the other hand, Iceland broke another barrier with the establishment of the feminist political party the Women’s List, which in the early 1980s strongly contributed to a permanent change in the so far very male-dominated political life of Iceland, and in more or less all political parties. Throughout the 1980s and 1990s there was a steady and considerable increase in the proportion of women in the Nordic parliaments, which contributed to the Nordic region’s international reputation as the most gender equal group of countries in the world.
 
Today, however, we cannot count on a continuous increase in women’s representation, and in some Nordic elections we have even seen a decrease. The numbers showing the highest proportion of women ever in Table 2 have in the last few years not always been the figures from the most recent election. However, with the latest increase in female representation in Finland and Iceland, the highest numbers shown here are from the latest election. But generally it is a new phenomenon in the postwar history of the Nordic countries, that it cannot be expected that the female representation will always increase in each individual election. 
 
At the same time, the debate has developed in different directions in the Nordic countries, and the pressures put on the parties vary. When the women’s representation in the Swedish Riksdagen decreased in 1991 for the first time in decades, a great debate ensued, and the network Stödstrumporna (The Support Stockings) threatened to establish a Women’s Party, unless the political parties would nominate more women for the next election – which they did. Although the decrease in 1991 was primarily caused by the right-wing party Ny Demokrati (New Democracy) entering the parliament with only three women out of 25 places, the 1991 election gave rise to debate, activism and pressure groups (Freidenvall 2006). In Denmark, for its part, women’s representation has remained unchanged in recent elections without this attracting any attention outside of the women’s movements – except very recently.
 
In many other areas of society the Nordic countries are not a leading region internationally; for example, when it comes to the proportions of women in managerial positions in the corporate sector, or the numbers of female professors. The Nordic countries score under the EU average in terms of women in leading positions. Furthermore, the Nordic countries display some of the largest differences between the proportion of women on the labour market and the proportion of women in leading positions, that is, a difference of 20 and 30 per cent units, which is a significant imbalance (European Commission 2008:35).3 The perception in Sweden is that their country is the “most gender equal country in the world”. But presently, for example, The UNDP gender-related development index has Norway in first place (2009).
 

3.1 The responsibility of the political parties

It is interesting that for a long time female representation developed in parallel in all the Nordic countries, despite there being great differences between the countries when it comes to the voters’ opportunities to influence which candidates are actually elected. The extreme ends of this spectrum of difference are on the one hand the compulsory preferential voting systems in Finland and on the other the situation at parliamentary elections in Norway, where voters can only choose between parties (closed lists). There is no agreement within research as to what the significance of the personal voting has for the development of women’s representation. The answer is complicated by the fact that the net effect of the voters’ choices varies from party to party, but also depending on the current the debate. A lively debate on the under-representation of women can result in many personal votes for women candidates in the next election, which, in turn, affects the nominations of the parties. It is a widespread notion that it is the voters who decide which persons get elected. However, this is actually not true. There is agreement within election research that in most political systems, including those in the Nordic countries, it is the political parties that are the actual “gatekeepers” to membership of local councils or parliament. This pertains both to closed and open lists electoral systems The reason is that it is the political parties that have the monopoly on nominating the candidates to be presented to the voters. It is the parties that decide whether a candidate be nominated in a constituency where the party has relatively safe seat, or in one where the party traditionally has a weak position. Furthermore, it is the parties that decide the placement of the candidates on the lists. Norwegian researchers have coined the concepts “election places” (for certain winners), “competition places” (for those that may be elected, if the party improves its share of the vote) and “decoration places” (those expected to lose) (Hellevik & Skard 1985). The likelihood of being elected is, naturally, greatest for those who have the election places (the safe seats), followed by the competition places; generally speaking, being an old hand, which is to say, running for re-election is also an advantage (known as the incumbency factor).
 
The Danish Equal Status Council (Ligestillingsrådet), now dissolved, for many years wrote in its annual report as an explanation of the increasing proportion of women that “the voters’ inclination to vote for a woman has increased” (Ligestillingsrådets årsberetning 1987:144). However, the Council had no basis for this conclusion, since it would require detailed analyses of changes over time in the placement of women on the lists, combined with analyses of the effects of the voters’ choice of individual candidates. In general, the party’s ranking order is only seldom changed by the voters. It is therefore always best to be placed at the top of the list. Thus, when discussing women’s representation, the focus must be on the recruitment practices of the political parties. The parties’ actions are decisive, and they, for their part, act according to expectations of how the voters react to the composition of lists in various historical periods. Generally, the curves for the female proportion of the nominated candidates and of the elected ones develop in parallel to each other.
 

3.2 The use of quotas in Nordic politics

In the Nordic countries quotas are used for the popularly elected assemblies only in the form of voluntary quotas among candidates, not in the form of legislation. It was the parties on the left of the political spectrum that were the first to introduce voluntary party quotas: the Green parties, the socialist parties and the social democrats, and the centre party, Venstre, in Norway (Dahlerup 1988). There is, however, considerable variation between the five Nordic countries. In Sweden, most political parties today follow the principle of ”varannan damernas” (every second a woman), according to which female and male candidates are nominated alternately on the parties’ electoral lists; but formal, written quota rules exist only in the red and green parties (Freidenvall 2006). In Iceland voluntary quotas are used for the candidate lists by two parties, in Norway by five parties (40 %, Labour 50 %). No parties in Finland formally use quotas, but many do so in practice. In Denmark, the Socialist People’s Party had gender quotas for their parliamentary candidate lists for a few years, and this party and the social democrats used quotas for the election to the EU Parliament during a short period, but that quotas were later abandoned in the name of ‘gender equality’ (Borchorst & Christensen 2003). From a global perspective, it is very unusual that quota rules are withdrawn once they have been implemented. Many, nevertheless, hope that quotas are a temporary instrument, which at some point in future will no longer be needed.
 
Not least in Norway and Sweden the voluntary quota rules of the political parties have had a considerable influence on the proportions of nominated and elected women (Freidenvall et al. 2006, table 3.2, pp 71– 72). While many countries in the global South have introduced gender quotas in order to increase women’s representation above the first 10–15 per cent, voluntary party quotas were implemented in the Nordic countries only after women’s representation had reached a high level of 20–30 per cent. For several decades after women’s suffrage, the step-by-step model was prevalent in the Nordic countries, and still is in many areas, as opposed to what has been called fast track policies, where pro-active gender equality measures are used to enable historical leaps in women’s representation (Dahlerup & Freidenvall 2005). However, Table 3 shows that quotas are not unknown in Nordic politics, not even those regulated by law.
 
As Table 3 shows, Norway is the “quota country” in the Nordic region. It is also the one of all the Nordic countries where quotas have met with least opposition. Denmark has made least use of quotas. In Sweden quota systems are extremely controversial and trigger much debate and many emotions, but are, nevertheless, used in certain areas. 
 

3.3 Leadership positions

A member of the Danish parliament has said: “It was in the air back in the 1950s, that the women in Folketinget should not expect to get any of the leading posts in parliament”.4 This situation has changed markedly. Today the Nordic female members of parliament hold more or less a proportion of leading positions in the parliaments which mirrors the proportion of female members as a whole; see the country reports5 (vertical gendering). Many other countries in the world have perhaps increased their proportion of women in politics, but female politicians have only a small share of the leading posts as, for example, committee chairs or speaker of the house. However, in the Nordic countries there is still a horizontally gendered division of labour within politics; but, apart from the most prestigious posts, this present division in the Nordic region can most likely not be said to represent anything else than the personal priorities of the male and female politicians (Refsgaard 1990; Wängnerud 1998; 2000). It is, nevertheless, important to specify that numerical female representation is not identical with influence and power. A totally different set on analytical tools is needed to study this. 
 
Some might perhaps claim that simply since women’s representation is so high in the Nordic countries today, it is impossible to count on a continued constant increase. But how should this be understood? Firstly, several areas in some of the Nordic countries are far from a 50 per cent gender balance. Secondly, it must be asked whether there is a new boundary at 50 per cent? Despite men having held 100, 90 and 80 per cent of all places for so long, 50 per cent for women is perceived as a kind of upper limit – or is it? The concept of “gender balance” indicates this as the final target.
 
The assemblies furthest from the 50 per cent target are the Danish local councils with 32 per cent women, that is, as much as 18 per cent units away from a gender balance, if this is defined as 50-50. The highest proportion of women is found in the Finnish government with 60 per cent women (+10). None of the Nordic parliaments have ever, as has Rwanda, passed the 50 per cent limit, but the Swedish Riksdagen comes closest to a gender balance with 47 per cent female members. But in some areas the development seems to have come to a halt, and the constant increase of female representation so far has stagnated.
 

3.4 The glass ceiling

The glass ceiling is a metaphor for, or image of, the invisible mechanisms that seem to hamper the advancement of women within politics, business, academia, etc. According to the Danish power study Magtudredningen, there were altogether only 12 per cent women in the entire elite in Denmark, and according to the corresponding Swedish study Maktens kön (The Gender of Power) the proportion of women in the power elite in Sweden increased from 13 per cent in 1989 to 26 per cent in 2001 (Christiansen et al. 2002; Göransson 2007). The greatest progress has taken place within politics, not least after the take off phase in the 1970s. 
 
But within political life there seems to be a stagnation of the development in certain areas. Among the Nordic countries, this pertains particularly to Denmark. Up to the latest election it seemed as if Finland and Iceland, too, had turned into “glass ceiling countries” with a stagnation of female representation, but both countries have recently experienced a considerable increase in the proportion of women in politics. This proves that glass ceilings can indeed be broken! In Denmark women’s representation has been unchanged at a level of 37–38 per cent in the last four parliamentary elections, and at 28, 27, 27 and 27 per cent in the last four municipal elections, 1993-2005, until the most recent local election of 2009 in which the percentage finally increased somewhat, to 32 per cent. After a merger of municipalities to larger units, the proportion of female mayors in Denmark fell as low as 9 per cent. At the same time, female representation in the last two elections to the now abolished county councils (amtsråd) in Denmark fell from 31 to 29 and finally to 27 per cent. The proportion of women in the election to the new Regional Councils nevertheless amounted to 34 per cent; but then, too, the risk for a decrease because of the larger units was also publicly debated. In a comparison of regional elections in the EU/EEA countries, Denmark is only just over the median (European Commission 2008). The question is whether such a stagnation should only be seen as a temporary mitigation, or as a sign of the activation of the attitude “this far – but no further”.
 
A Nordic comparison is important for the understanding of the character of the stagnation. Denmark holds the lowest Nordic record, when it comes to the proportion of women on municipal councils, since Iceland, which tends to have the lowest level, has now overtaken Denmark. The lack of debate about the problem is without doubt an important factor explaining the stagnation. In Denmark there has been relatively little discussion on the low percentage of women, and only in connection with the celebration of 100 years of women’s right to vote in the municipal elections, introduced in 1908, has a debate started on the low and stagnating female proportion in Danish local elections (www.kvinfo.dk). In Sweden, too, the proportion of women stagnated in the last local elections, but at a far higher level than in Denmark: 42 per cent.. Looking at the parallel historical development in the Nordic countries after World War II, a difference of 10-15 per cent between Denmark and Sweden in this area must be taken as an abyss. 
 
However, Denmark is not alone in being stuck at around 30 per cent. This is the case also in the Netherlands. In the past both of these countries were in the world’s absolute super league, but today they have gone down to 10th and 13th place respectively; see Table 1. In the Netherlands, too, the question is presently asked as to why a stagnation has taken place (Leyenaar 2007). But what is a glass ceiling actually? The glass ceiling concept is used to describe invisible, structural barriers (non-legislative), which hamper the careers of women. In their attempts to advance and gain real influence, the women come up against an invisible ceiling. The glass ceiling is a metaphor for structural barriers or, in other words, structural discrimination. Even though the concept of a glass ceiling is primarily used in regard to career opportunities in the labour market, it is also used concerning political representation.
 
But the glass ceiling, as such, is not an explanation. It does not say anything about which mechanisms can be assumed to lie behind it, nor what or who is doing the hampering, nor how a glass ceiling can be broken. We must therefore dig a bit deeper. Within Nordic research on women in politics, there are several opposing theories when it comes to explaining the variations in the female political representation. Below I will discuss four such theories (Dahlerup 2009).
 

4. Theories on women’s representation

4.1 Theory of patriarchy

The glass ceiling approach is often connected with a notion that it is true that women in the Nordic countries have entered politics, but when they start coming close to power, they encounter an attitude of “this far, but no further”. The hypothesis is that men maintain their superior power over society through various mechanisms. Already when the proportion of women reaches 30 per cent, there are comments to the effect that “now women must not start dominating”. This perspective claims that men as a group maintain their hold on power because the institutions into which women have entered thereby lose their power (enter women – exit power). In a slightly different version it is claimed that women only manage to enter institutions which are about to lose their power (exit power – enter women). In the research both these versions are called the theory of shrinking institutions. It is, however, difficult to test these theories empirically, partly because they are often quite vague. Was the power of the Folketinget already diminishing or did it diminish when the proportion of women started to grow rapidly in the 1970s? Diminishing in relation to what? Is it, in that case, explained by the increasing proportion of women? Is the question of minority versus majority governments not more decisive than the proportion of women for the power of parliament?
 
This tradition also uses the concept of the law of increasing disproportionality, according to which the proportion of women decreases progressively the higher up in the hierarchy we look. There is also a great deal of statistical evidence to confirm that the proportion of men generally increases the higher we come in the hierarchy. The law of increasing disproportionality within politics in the Nordic countries is often ascribed to the book Unfinished Democracy. Women in Nordic Politics (Haavio-Mannila et al. 1983, in English 1985). Although the power perspective undoubtedly dominated this first co-Nordic book on women in politics, it also pointed out that there were exceptions already in the early 1980s, particularly when looking at politics within the individual sectors, e.g. the social sector versus the financial sector. Tine Kjær Bach gives these thoughts the joint name of the marginalisation hypothesis (2005). The hypothesis was dominant in the 1970s and 1980s, but is still heard today in the public debate. These theories are related to the more general theories on patriarchy and as it is called in Sweden, the gender power regime (‘könsmaktsordningen’ or ‘genussystemet’), which emphasize women’s structural subordination. Particularly in the 1970s and 1980s the progress of women was often interpreted within such a framework. “This only happens because...” These theories were criticised for being without nuance and too static; a criticism which the historian Yvonne Hirdman, mother of the concept “gender system”, has already responded to in the title of her book published in 2001: Genus – om det stabilas föränderliga former (Gender – On the Changeability of the Stable Form). 
 
If one uses a purely nominal definition of a patriarchy or male dominated society as a society where men hold the majority of all the leading positions in the fields of economy, politics, religion and social organisations, all societies known so far must be called patriarchal. But such a nominal definition does not, however, say anything about what mechanisms are at work, or why there are so many different types of patriarchies. In Sweden a report was published in 1986 on women and political power with the title Hit – men inte längre?(This far – but no further?). At that time, women’s representation in the Swedish Riksdagen was about 30 per cent. However, since then, the percentage has actually increased to the present 47. So, does the hypothesis of “this far, but no further” not apply in Sweden? The varying trends internationally challenge the thesis of the unambiguous reproduction of male dominated society, and shows the need for the application of nuance to the argument. This entire theory complex has been challenged by the time lag theory. 
 

4.2 The time lag theory

As a reaction to what was seen as too pessimistic a perspective, the time lag theory presented a more positive description: the issue is primarily a question of a historical time lag. The book Women in Nordic Politics. Closing the Gap (Karvonen & Selle 1995) presented an alternative thesis to Unfinished Democracy from 1983. The time lag hypothesis is based on the assumption that mobilization at a lower level must reach a certain level before mobilisation is possible at the next level up. Thus women, too, will reach the highest positions in society – but with a certain time lag. Women are slowly, but surely, in a process of being integrated into the leadership of Nordic society. Although women are still under-represented among mayors and managers, a gender balance will be achievedin due time. The powerlessness of women is a myth, which makes us overlook the new power of women. Therefore a time lag theory is to be preferred to the theory of constant reproduction of patriarchy (Raaum 1995). The theoretical basis for the time lag thesis originates in, among others, the Norwegian researcher Stein Rokkan’s ideas of democratic development as thresholds, which are gradually crossed through the mobilisation of new groups of voters. Nina Raaum uses this macro theory for proving also that the mobilisation of women has undergone various stages and has expanded, both as to its width and its depth (Raaum 1995). As is obvious, this theory focuses on the mobilization and activation of women, and on the improvement of women’s resources. The explanation for women’s under-representation is sought in a historical time lag. Before a company can have a female president, there must have been a female vice president. The time lag theory can be interpreted as assuming that there is a certain line of development in society, where overspill from one sector to another takes place, and where development is regarded as irreversible (Karvonen & Selle 1995).
 
Empirically, the time lag hypothesis is supported by the actual, strong increase in the proportion of women in a large number of areas, not least within politics. The presentation in Table 2, with thresholds which are crossed in a historically progressive process, is as such an illustration of the time lag theory. The time lag thesis is applied both to development in society at large and to development in specific areas. As in the case of the theory of patriarchy, this is a macro theory, which has become popular and is included in the general public debate. 
 
As said above, the time lag theory has gained support in the historical development. That is, until now. The very latest stagnation in women’s representation attaches a question mark to this theory. Will gender balance appear over time – all by itself? Women are now on the labour market, and young women receive as extensive an education – if not more extensive – as young men. And nevertheless 75–90 per cent of society’s elite are men.
 
It has been proved in the area of academia that the time lag theory has a significant limp, since the proportion of women among lecturers and professors is only slowly increasing, despite the proportion of female students, candidates and PhDs having increased for several years (Henningsen et al. 1998). Similarly, a cohort analysis on Swedish data also attaches a question mark to the time lag theory. Among a group of women and men who finished their doctorates in 1991, 8 per cent of the men, but only 4 per cent of the women from the same professions became professors within a 12-year period (Personal vid universitet och högskolor 2005).
 
So, the time lag theory is currently being challenged by actual developments in several areas. One example is the mentioned decrease in the number of female mayors in Denmark after the structural reform of 2005, which merged 276 municipalities into 98 large municipalities. Despite the growing proportion over many years of women who have been chairs of committees within local councils and held other important posts within local politics, the proportion of women among the Danish mayors was lower in 2005 than in the late 1980s. It is this increasing doubt as to whether gender equality will manifest over time which also has stimulated the discussion of quotas. Developments in eastern and central European countries after the fall of the wall has also contributed to dealing a blow to the belief in a constantly progressing gender equality.
 
The time lag theory, like the theory of the reproduction of patriarchy, focuses on structural factors, but also on the mobilisation of and resources available to women. And naturally women’s entry onto the labour market and their increasing levels of education have had a great impact. But even after the women have entered the labour market with at least as long an education as men, old and established sectors remain male-dominated at the top and so also do newer industries, such as IT, which has mainly male managers. Will the time lag theorists’ response be, that a sufficient number of years have simply not yet passed to make judgement? Or must other theories be taken into account?
 

4.3 The saturation theory

A Danish researcher on local politics, Ulrik Kjær, has presented an alternative theory: “saturation without parity”. It is an interesting thesis, saying that the increase in female representation can stagnate before gender balance has been achieved, because a certain “saturation” has been reached. The saturation theory is based on the assumption that both the party organisations that nominate candidates, and the voters “simply have had enough of women, and are satisfied with the present level” (Kjær 2001: 70).
 
The time lag theory and the saturation theory, thus, are opposites (Kjær 1999:162). The saturation theory does, however, have certain similarities with the “this far, but no further” perspective, but it operates on the organisational or meso level: the analysis focuses on the recruitment practices of the political parties. On the basis of thorough analyses of the development of female representation on Danish local councils, Ulrik Kjær even claims to be able to point to the existence of a critical point: that of 31 per cent. This is supported by the fact that the proportion of women in the Danish local elections in 1997 increased most on those lists that had a low female proportion to begin with, while the women’s representation decreased for over half of the lists from which over approximately 30 per cent women had been elected in 1993. Survey data further support the hypothesis. Kjær’s study among the chairs of local party organisations shows that two thirds think that it is important or very important to have a more or less equal number of candidates of both genders. But the study also shows that an equal distribution to them does not necessarily mean 50-50. On the contrary, nine of ten local party chairs, whose lists include over 30 per cent women, are satisfied with that result (2001). According to Ulrik Kjær, the potential for increased female representation has been capitalised – and thus exhausted (1999; 2001). He claims that there is a difference between the “saturation points” of various parties, but the phenomenon is found within all of them (2001).
 
Ulrik Kjær has developed his saturation theory on the basis of the Danish stagnation. He adds himself that it must be tested in other countries, too. And it is precisely this point at which the saturation hypothesis reveals its shortcomings. If a saturation point of about 30 per cent can be identified in Denmark – and Kjær is obviously making a correct observation here – why have the rest of the Nordic countries succeeded in reaching between 36 and 42 per cent? The saturation theory cannot explain why it seems that of the Nordic countries only Denmark has come to a saturation point!
 
Lenita Freidenvall has studied the significance of gender in Sweden, where the proportion of women is already very high. Does this mean that gender plays a less important role now? In her extensive analysis of attitudes and practices in the nomination committees of Swedish political parties for the 2002 parliamentary election, Lenita Freidenvall shows that even if many criteria are considered in the Swedish context when the list is composed, and even if the proportion of women already was as high as 42 per cent, the majority of the nomination committee chairs still ranked an equal gender distribution as the most important factor (2006). Neither have countries such as Costa Rica and Belgium reached a saturation point at 30 per cent, but, on the contrary, they have introduced legislation on even higher proportions, 40 and 50 per cent, of women on the electoral lists, as well as strict rules on the ranking of candidates on the lists. 
 
In short, it seems implausible, that there would exist a general limit of 30 per cent. The feeling of saturation is constructed in a certain context, and is the result of a great number of circumstances. It is obvious, that other factors must be considered.
 

5. The significance of public debate and the pressure from women’s movements

If we disagree with the thesis that gender equality will appear as a historical necessity, and yet do not think that male dominance reproduces itself in all periods of time and in all areas, nor that there is an absolute saturation point at about 30 per cent, then other perspectives have to be included in the analysis and the political debate.
 
My own analyses of the historical development in the Nordic countries, and my comparisons of the astonishing differences between developments in Sweden and in Denmark since the 1990s, as well as my research on the use of quotas in politics globally, have pointed to the significance of two interlinked factors: 1) changes in the discursive framework about gender, and 2) the power of women’s movements.
 
There are great differences over time and between countries as to the perception of women’s political representation. The perceptions vary from the traditional view that politics is a male business or that gender is irrelevant, to specific theories on why women are under-represented. A marked new position is the notion that gender equality has more or less been achieved already. However, this is so only for the “natives”, not for immigrants, as it is put in totally unreasonable generalisations (Dahlerup 2004 and 2007a). It is noteworthy that the stagnation in women’s representation – and in the gender equality debate in general – happens in both Denmark and the Netherlands at a time, when the immigration debate is at its liveliest and xenophobia is increasing (Leyenaar 2007). This obviously calls for closer comparative analyses.
 
The general thesis is that the discursive framework, that is, the dominant perceptions of women’s position in society at large and in politics in particular, is crucial for the attitudes and actions of the political parties and the voters. This is where the women’s movements enter the scene. The women’s movements have continually challenged these dominant views by using various strategies. The progress or stagnation of women’s representation is thus also connected with the power of the women’s movements. The pressure from women’s movements has undoubtedly been crucial for the development of women’s representation in the Nordic countries; this pertains to pressure both from autonomous women’s movements and from women within the trade unions, and from the women’s organisations and groups within the political parties, which were previously so strong (Dahlerup 1998 and 2001). Also gender studies have probably been an additional and important factor behind the increasing women’s representation in the Nordic countries.
 
The perspective used here also includes an actor perspective, and the focus is then on the significance of the women’s movement for changing the discursive framework as well as for the various forms of pressure that have been directed at the political institutions. Inspiration for this type of analysis comes from the Australian political theorist Carol Bacchi, whose argument is that policy analyses have been far too problem-centred and focused on how these problems are to be solved. But each suggestion contains an explicit or implicit diagnosis of what is understood as the actual problem, and much of the political struggle is about this – a point which is not unfamiliar to any women’s movement, it might be added.
 
Thus, the central feature in Bacchi’s analysis is how the problem is constructed and presented in the first place. She calls her perspective the “What’s the problem approach” (Bacchi 1999: 1f). So, how has the diagnosis been made in the debate on women’s under-representation? The following are a few focal points: 
 
In the past, it was common to mainly blame the low representation of women on the women themselves. The women are the problem. And when women’s representation increased, it was said that the women had finally pulled themselves together! This was typical of the period up to the 1970s, but this perspective was highly problematic (Dahlerup 2001).
 
An alternative diagnosis, which was important not least in the Nordic countries in the 1970s and 1980s, and which created a strong and effective pressure, puts the primary responsibility for women’s representation on the political parties. The diagnosis here is that it is the parties, with their traditional recruitment practices and the entire political cultural context, which constitutes the problem. It should be added, that in the present heated debate on gender quotas for the boards of joint-stock companies, following the Norwegian example, the old debate on women’s resources and qualifications has nevertheless appeared again in its entirety. 
 
In this alternative view, the parties are criticized for not recruiting women to a sufficient extent. The parties are the gatekeepers, and it is they who are not able to, or do not want to, recruit as many female as male candidates. This view can be named a discourse of exclusion. The same perspective can be recommended for analysis of the political underrepresentation of ethnic minorities and immigrants. As gatekeepers, the parties posses the power to change imbalances. This alternative discourse has appeared strongly in many countries today, also outside the Nordic countries, particularly perhaps in post-conflict countries. The point of departure here is a justice perspective. In a democracy women have the right to equal representation. Women form half of the population, and if women in a purely statistical sense do not get a corresponding proportion of the places, this must be blamed on various mechanisms of exclusion, primarily in the organisations themselves. 
 
These mechanisms affect both the demand and supply side of the nomination process, that is, both the recruitment practices of the political parties and the women’s willingness to be nominated. 
 
The UN Platform for Action, adopted by the Fourth UN World Conference on Women in Beijing in 1995, represents this alternative discourse. The focus of this document – and in many corresponding international documents today – is aimed at the institutional and cultural barriers and exclusion mechanisms that obstruct an equal gender distribution in decision-making processes. At the same time, the demand has been radicalized from the earlier “more women in politics” to “gender balance”.
 
Taken together, this diagnosis and this target result in the demand for active political measures (positive action, affirmative action, special measures). Although the controversial word ‘quota’ is not written directly into it, the Beijing declaration has been used by women’s movements all over the world as support for their demand for quotas in order to rapidly remedy the under-representation of women (Dahlerup 2006; see also www.quotaproject.org). However, there are, as has been pointed out, several possible forms of active political measures, of which quotas is only one. The crucial factor seems to be the level of pressure on the political parties and on political leaders in general.
 
Thus, the conclusion is that it is a lack of political pressure, in combination with a notion that gender equality will manifest automatically, or actually has already been more or less achieved, which explains the stagnation in women’s representation which is being seen within certain areas in the Nordic countries. But this conclusion also indicates that stagnation is not a permanent state; glass ceilings can be broken. The Nordic region has for a long time been at the top of the league in a global comparison when it comes to women’s political representation. Today the leading position of the Nordic countries is challenged by a number of countries both in the global South and in Europe, such as Belgium and Spain. This is a positive development, but seen from a Nordic perspective, this challenge should lead to further attempts to speed up the process of improving women’s representation in the Nordic countries, both within politics and in the many other areas of society where the proportion of women is significantly lower than in political assemblies.
 
Notes
1. In a co-operation between International IDEA and Stockholm University through Drude Dahlerup & Lenita Freidenvall a global website has been created, where the various quota systems in the many countries with quotas are listed (www.quotaproject.org).
2. There are, however, examples of quota systems which work in plurality/majority electoral systems, for example, the Scottish Labour Party's twinning-system, where two constituencies together were obliged to nominate a man and a woman, or the local elections in India, where the places reserved for female candidates rotate between the constituencies from one election to the next.
3. Managerial positions: ISCO categories 121 (directors and chief executives) and 13 (managers of smaller enterprises).
4. Said by Grethe Philip, member of parliament for the social-liberal party Det radikale Venstre (personal communication)
5. Niskanen & Nyberg ed. (2009) Kön och makt i Norden. Del I Landsrapporter .
 
References
Bacchi, Carol Lee (1999), Women, Policy and Politics. London: Sage.
Bach, Tina Kjær (2005), Kvinder i kommunalpolitik. Rapport udarbejdet for
Ligestillingsafdelingen. Freia, Aalborg universitet, oktober 2005.
Bergqvist, Christina et al. (ed.) (1999), Likestilte demokratier? Kjønn og politikk i
Norden. Oslo: Universitetsforlaget/Nordisk Ministerråd.
Borchorst, Anette & Ann-Dorte Christensen (2003), ”Kønskvotering i SF og i
forskerstillinger – diskursiv praksis og forandring”, pp. 101–23 in Anette Borchorst
& Drude Dahlerup, Ligestillingspolitik som diskurs og praksis. København:
Samfundslitteratur.
Borchorst, Anette & Drude Dahlerup (2003), Ligestillingspolitik som diskurs og
praksis. København: Samfundslitteratur.
Caul, Miki (1999), ”Women’s Representation in Parliament: the role of political
parties”, pp. 790–98 in Party Politics, vol. 5, no. 1.
Christensen, Ann-Dorte & Poul Knopp Damkjær (1998), Kvinder og politisk
repræsentation i Danmark. GEP, Tekst nr. 7, Aalborg Universitet.
Christiansen, Peter Munk et al. (2002), ”Køn og eliter”, pp. 72–91 in Anette Borchorst
(ed.), Kønsmagt under forandring. Magtudredningen. København: Hans Reitzels
Forlag.
Dahlerup, Drude (1988), Vi har ventet længe nok. Håndbog i kvinderepræsentation.
Nordisk Ministerråd. Also published in Icelandic (1988), Norwegian (1989), Swedish
(1989) and Finnish (1990).
Dahlerup, Drude (1998), Rødstrømperne. Den danske Rødstrømpebevægelses
udvikling, nytænkning og gennemslag 1970–85. Bd.I-II. København: Gyldendal.
Dahlerup, Drude (2001), ”Men kvinderne vil jo ikke selv”. Diskursen omkring
betydningen af køn ved danske kommunevalg, GEP, Tekst nr. 2, Aalborg
Universitet.
Dahlerup, Drude (2004), ”Feministisk partipolitik? Om skillnader i dansk och svensk
jämställdhetsdebatt”, pp. 234–63 in Christina Florin & Christina Bergqvist (ed.),
Framtiden i samtiden. Könsrelationer i förandring i Sverige och omvärlden.
Stockholm: Institut för Framtidsstudier.
Dahlerup, Drude (ed.) (2006), Women, Quotas and Politics. London: Routledge.
Dahlerup, Drude (2007a), ”Ligestillingsforståelser i svenske og danske partier. Er en
borgerlige ligestillingspolitik under udvikling?”, pp. 6–16 in Tidsskriftet Politik,
årg. 10, nr. 2. Københavns Universitet.
Dahlerup, Drude (2007b), ”Electoral Gender Quotas: Between Equality of Opportunity
and Equality of Result”, pp. 73–92 in Representation, vol. 43, nr 2, juli 2007.
Also in Chinese, Kunming University 2008.
Dahlerup, Drude (2009), ”Glasloft-landet. Udviklingen i kvinderepræsentationen i
Danmark i et nordisk og globalt perspektiv”, in Festskrift for 100–året for kvinders
valgret. www.kvinfo.dk
Women in Nordic Politics – A Continuing Success Story? 
Dahlerup, Drude & Lenita Freidenvall (2005), ”Quotas as a ’Fast Track’ to Equal
Politial Representation for Women. Why Scandinavia is No Longer the Model”, pp.
26–48 in International Feminist Journal of Politics, vol. 7, nr 1, March 2005.
Dahlerup, Drude & Lenita Freidenvall (2008), Kvotering. Nr. 15, Pocketbiblioteket.
Stockholm: SNS Förlag.
Dahlerup, Drude & Lenita Freidenvall (2009) ”Gender Quotas in Politics: A Constitutional
Challenge” in Susan Williams (ed.), Constituting Equality. Gender Equality
and Comparative Constitutional Law. Cambridge University Press.
European Commission, Women in decision-making 2007. Analysis of the situation
and trends. January 2008.
EUPARL (2008), Electoral Gender Quota Systems and Their Implementation in
Europe. Report to The European Parliament, see www.statsvet.su.se/wip (Drude
Dahlerup & Lenita Freidenvall, in co-operation with International IDEA).
Freidenvall, Lenita (2006), Vägen till Varannan Damernas: om kvinnorepresentation,
kvotering och kandidaturval i svensk politik 1970–2002. Stockholm:
Statsvetenskapliga Institutionen, Stockholms Universitet.
Freidenvall, Lenita et al. (2006), ”The Nordic Countries. An Incremental Model”, pp.
55–82 in Drude Dahlerup (ed.) Women, Quotas and Politics. London: Routledge.
Haavio-Mannila, Elina et al. (1983), Det uferdige demokratiet: kvinner i nordisk
politikk. København: Nordisk Ministerråd.
Hellevik, Ottor & Torild Skard (1985), Kommunestyrer – plass for kvinner? Oslo:
Universitetsforlaget.
Henningsen, Inge et al. (1998), Reduktionisme og kønsfordeling i forskerverdenen.
Køn i den akademiske organisation, arbejdspapir nr. 5.
Hirdman, Yvonne (1990), ”Genussystemet”, pp. 73–116 in Demokrati och Makt i
Sverige. Maktutredningens huvudrapport, SOU 1990:4. Stockholm: Allmänna
Förlaget.
Hirdman, Yvonne (2001), Genus – om det stabilas föränderliga former. Malmö: Liber.
Hit – men inte längre? Delrapport från utredningen om kvinnorepresentation. Ds
1986:5. Stockholm.
Htun, Mala (2004), ”Is Gender Like Ethnicity? The Political Representation of Identity
Groups”, pp. 439–58 in Perspectives on Politics, vol. 2, nr 3.
Karvonen, Lauri & Per Selle (ed.) (1995), Women in Nordic Poltics. Closing the Gap.
Aldershot: Dartmouth.
Kjær, Ulrik (1999), ”Saturation Without Parity: The Stagnating Number of Female
Councillors in Denmark”, pp. 149–168 in Erik Beukel et al., Elites, Parties and Democracy.
Odense: Odense University Press.
Kjær, Ulrik (2001), ”Kvinder i kommunalpolitik – er grænsen nået?”, pp. 68–78 in
Kvinder, køn og forskning, årg. 10, nr 1.
Kjær, Ulrik (2007), ”Kommunalpolitikerne”, pp. 133–58 i Roger Buch & Jørgen
Elklit (ed.), Nye kommunalvalg? Kontinuitet og forandring ved valget i 2005.
Odense: Syddansk Universitetsforlag.
Krook, Mona Lena (2004), ”Promoting gender-balanced decision-making: the role of
international fora and transnational networks”, pp. 205–220 in Hilda Rømer Christensen,
Beatrice Halsaa & Aino Saarinen (ed.), Crossing Borders: Re-Mapping
Women’s Movements at the Turn of the 21st Century. Odense: Odense University
Press.
Gender and Power in the Nordic Countries
Leyenaar, Monique (2007), Challenges to Women’s Political Leadership in Europe,
Paper, Radboud University Nijmegen, Institute for Management Research.
Niskanen, Kirsti & Anita Nyberg (ed.) (2009), Kön och makt i Norden. Del I Landsrapporter.
TemaNord 2009:569. København: Nordisk Ministerråd.
Norris, Pippa (2004), Electoral Engineering: Voting Rules and Political Behavior.
New York: Cambridge University Press.
Norris, Pippa (2006), ”The Impact of Electoral Reform on Women’s Representation”,
pp. 197–213 in Acta Politica (Special Issue), vol. 41.
Personal vid universitet och högskolor 2005. Statistiska Centralbyrån &
Högskoleverket: Statistiska Meddelanden, UF 23 SM 0601. Stockholm.
Raaum, Nina Cecilie (1995), ”The political representation of women: a bird’s eye
view”, pp. 25–55 in Karvonen, Lauri & Per Selle (ed.), Women in Nordic Politics.
Closing the Gap. Aldershot: Dartmouth.
Refsgaard, Elisabeth (1990), ”Tæt ved toppen. Kvindernes placering i Folketingets
arbejds- og magtfordeling 1965–1990”, pp. 106–139 in Drude Dahlerup & Kristian
Hvidt (ed.), Kvinder på Tinge. København: Rosinante.
Skjeie, Hege & Mari Teigen (2003), Menn imellom. Mansdominans og likestillingspolitikk.
Oslo: Gyldendal.
Teigen, Mari (2003), Kvotering og kontrovers. Om likestilling som politikk. Oslo:
Unipax.
Tripp, Aili, Dior Konaté & Colleen Lowe-Morna (2006), ”Sub-Saharan Africa: on the
fast track to women’s political representation”, pp. 112–137 in Drude Dahlerup
(ed.) Women, Quotas and Politics. London: Routledge.
Wängnerud, Lena (1998), Politikens andra sida. Om kvinnorepresentation i Sveriges
riksdag. Göteborg: Göteborg studies in politics 53. Avhandling.
Wängnerud, Lena (2000), ”Representing Women”, in Esaiasson, Peter & Heidar,
Knut (ed.), Beyond Westminster and Congress. The Nordic Experience. Columbus:
Ohio State University Press.
 
Websites: