Introduction

International financial development institutions (IFDIs) play a significant role in post-conflict state building, and the flow of finance they direct, as well as the economic restructuring they impose, may influence women’s politico-economic participation. IFDIs are understood to include multilateral development institutions such as the World Bank, the International Monetary Fund (IMF), the European Union (EU), and the Organisation for Economic Co-operation and Development (OECD). IFDIs often condition their funds on achieving goals that relate to ‘economic agency/efficiency’, ‘open markets’ and the like, which encourages economic restructuring (Viterna & Fallon 2008, p. 684).

The IMF and the World Bank are the most thoroughly researched IFDIs. The ‘Bretton Woods Institutions’ were set up at an international meeting in Bretton Woods, New Hampshire in 1944, with the aims of rebuilding the post-World War II worldwide economy and promoting international economic cooperation, under an international political project that entails deregulation, privatisation and structural adjustment (‘neoliberalism’) (Prugl 2014, p. 4). Their role has since evolved, through their involvement in financial crises and conflicts of the 1970s and 1990s (e.g. see Onimode 1999), to now be at the centre of post-conflict economic restructuring. Following war, the IMF and the World Bank will often serve as economic strategists and even principal drafters of commercial/banking legislation that govern post-conflict economic development (Ingves 2004, p. 11; Boon 2007, p. 514).

Throughout the 20th century, the IMF and World Bank mandated tight fiscal policies and structural adjustments in response to economic crises (Feldstein 2003, p. 15), which reduced the value of the national currency, increased interest rates, weakened labour laws (Feldstein 2003, p. 15), and implemented a market-based economy (Ghodsee 2004, p. 729). In the 1990s, these major IFDIs were encouraged by the United States to increase engagement in post-conflict/post-Soviet economic restructuring, recognising that they could provide “all the nuts and bolts of nation-building” (Boon 2007, p. 523). Subsequently, the IMF and World Bank played a significant role in shaping the social, politico-economic climates of Afghanistan, Africa’s Great Lakes region, the Balkans, Liberia, Nepal, Sierra Leone, Timor Leste, Sri Lanka, the West Bank and Gaza (Bello 2006, pp. 286-287).

IMF and World Bank policies in these countries during the 1990s were similar regardless of the context. These IFDIs would leverage the huge debt owed by nations (and the hope of possible relief from that debt) to institute broad-brush economic reforms (Barton 2005, p. 27). This policy was (and is) termed ‘conditionality’, where funding would be dependent on the target government adopting certain reforms. However, conditions on funding were not the only way the World Bank and IMF exerted political influence. In Bosnia, senior officials from the World Bank and European Union successfully pressured the Bosnian Serb President to attend post-peace ceremonies, by threatening the complete removal of aid; aid already drastically cut due to the perceived lack of engagement of Republic Srpska in implementing peace reforms (Boyce & Pastor 1998, p. 42).

Countries that were emerging from conflict in the 1990s, such as Nicaragua and El Salvador, had to accept harsh external stabilisation and structural adjustment programmes from IFDIs to access much needed foreign resources (Paris 2002, p. 44; Shaler-Sholk 1994, p. 2). Throughout this period, the implementation of IFDI conditionality and restructuring programs arguably disregarded the realities of civilian life under austerity (Feldstein 2003, p. 4). Especially disregarded was the evidence that IFDI reforms had deeply gendered effects (True, Mlinarević, Isaković, Chinkin, Rees and Svedberg 2017; Elson 1995; Nissan 2016). As the World Bank and IMF (primarily) transitioned post-conflict governments to a neoliberal philosophy underpinned by individual freedom, market forces, and minimal state intervention (True et al. 2017, p. 5; Prugl 2014, p. 7), hierarchies of gender and masculinised financial power were ignored (Hozic & True 2016, pp. 60, 65).

The transition to post-conflict market-based economics emphasises masculinised concepts of self-management, individualism, and self-enterprise (Marchand & Runyan 2011, p. 4). In this politico-economic system, the public sector is downsized, which excessively impacts women as they often hold most of State lower-level employment (Rama 2002), and fees for education and childcare are increased, both of which may lead to the withdrawal of girls from school (Haddad, Brown, Richter & Smith 1995). Earlier studies have argued that post-crisis/conflict economic reforms have negatively affected food affordability (Jolly 1991, p. 1812; see Onimode 1989), influenced the closure of schools and hospitals (Ghodsee 2006, p. 32), and may have further impoverished sections of society (Jansen 2006, p. 189). Indeed, social welfare and other like policies are generally reduced, which disproportionately affects women as they are more likely dependent on public resources (such as childcare and maternity leave) and are expected to fill gaps left by the State in caring labour (Peterson 2005). This entrenches women in ‘care labour’ that is situated in the private home, and is at the centre of women’s subordinate role in neoliberal societies (Barker 1995, p. 2198), as they will always struggle to ‘compete’ with men where the free-market economy demands masculinised competition (Elson 1989, p. 63).

Further, IFDIs have routinely been accused of turning a blind eye to a range of human rights abuses in target countries. For example, between 1991 and 1993, foreign aid to Rwanda rose by 50 per cent to $355 million, with no funding contingent on human rights concerns, despite substantial evidence of immediate catastrophic violence, and even though such pressure could have arguably deterred the 1994 genocide (see Uvin 1998). In Sri Lanka, following a ceasefire agreement between the government and the Tamil Tigers in 2001, IFDIs poured millions of dollars in aid and loans into the country; despite evidence that the ceasefire agreement was poorly implemented, civilians in rural areas were subject to human rights violations, and that there was poor development throughout Sri Lanka (Sriskandarajah 2003, p. 28).

Aggressive IFDI economic reforms and oppressive loan conditions have been applied indiscriminately throughout the 1990s and into the 2000s (see Uvin 1998). Such policies have sparked sporadic outrage, such as in Honduras following 1998 Hurricane Mitch, where economic reforms privatised most of the economy while the country was still ‘knee deep in rubble, corpses and mud’ (Bello 2006, p. 284). Indeed, this program of aggressive economic liberalism continued throughout the 2000s (see Griffiths 2006), and heightened with IMF and World Bank involvement in post-conflict reconstruction, such as in Iraq during the ‘War on Terror’ (Boon 2007, p. 530), and following the 2004 Asian Tsunami (see Dennis & Yunus 2008).

In more recent years, the IMF and World Bank have both acknowledged the perils of high income inequality and the adverse consequences of inadequate social protection policies that have resulted from its programs (IMF 2014a; IMF 2014b; IMF 2014c; IMF 2017; A SEED Europe 2008, p. 4), and have indicated a desire to reform their policies (Kentikelenis, Stubbs & King 2016, p. 545). Indeed, other IFDIs and multilateral funding institutions, such as the European Union, have also attempted to make a more visible effort to direct funding to equality projects in target countries; although, only 6 per cent of aid to fragile states targeted equality issues relating to gender (OECD 2015, p. 3). Further, recent conditions applied to massive loans by the IMF in the Ukraine have disproportionately affected women through reduction in public expenditure, decreased health and education investment, and aggressive decentralisation (WILPF 2017). Despite the acknowledgement that their policies may negatively impact lower socio-economic groups in target countries, there is inadequate evidence that IFDIs have made fundamental transformations to address these concerns (Kentikelenis, Stubbs & King 2016, p. 545).

IFDIs appear to have a significant impact on societies, as the key driver of economic restructuring following conflict or monetary collapse. However, their influence on women’s politico-economic rights is less clear, especially in the context of the IFDI’s new, purported concern for social equality in their programs. Given the scale of women’s exclusion in post-conflict society, the influence of IFDI-led economic restructuring programs on women’s politico-economic participation is under-examined and worthy of analysis.

Paper Aim

It is the aim of this paper to document whether IFDIs have had a negative impact on women’s politico-economic participation, following the collapse of Communism in the Soviet Union and the Socialist Federal Republic of Yugoslavia (SFRY). This paper aims to utilise an innovative method of fuzzy set qualitative comparative analysis to suggest how IFDI funding may be impacting women’s politico-economic participation; and then reconcile these fuzzy set results with a focus on two case studies.

Procedure

Analytical Techniques and Methods – Fuzzy Set QCA

This paper will adopt fuzzy set Qualitative Comparative Analysis (QCA) to account for women’s political and economic participation in post-Communist societies. This paper will then explore how IFDIs impact women’s political and economic participation in two selected case studies.

The goal of the fuzzy set QCA is to account for levels of women’s politico-economic participation in different causal contexts. Fuzzy set QCA is based on a comparison of cases, rather than a statistical analysis of a large ‘n’ number of cases, and focuses on the necessity and sufficiency of causal effects on the outcome (to be discussed) (Reynaert 2011, p. 409). Fuzzy set QCA strives for a context-specific assessment of cases, looking at the effects of a combination of conditions that account for an outcome, rather than focusing on the average effects of single variables (Buche, & Siewert 2016, p. 360).

QCA utilises Boolean algebra for logical comparison, where each selected case is a combination of causal and outcome conditions (see Ragin 2000), which can then be compared with each other. The combinations of causal and outcome conditions are typically presented in a ‘truth table’, which lists the logically possible combinations of causal and outcome conditions associated with each case (see Ragin 2008). QCA, through the construction and mathematical analysis of the truth table, is said to be a ‘mixed qualitative-quantitative method’; as it requires an in-depth knowledge of cases, and many iterations between theory, cases and truth table construction, whilst also operationalising this qualitative expertise into quantitative values that can be numerically measured and compared.

The conventional Boolean values in a truth table (i.e. cases either being ‘in’ or ‘out’ of a set – i.e. ‘1’ denominating full politico-economic participation and ‘0’ being none) are dichotomous. A fuzzy set, by contrast, permits membership in the interval between 0 and 1 while retaining the two qualitative states of full membership and full non-membership. Thus, it would be possible to have a fuzzy set of cases that are ‘completely in’ the set (‘1.0’), those who are ‘almost fully in’ (membership ‘0.90’), some who are neither ‘more in’ or ‘more out’ of the set (‘0.5’, the ‘crossover point’), and so on down to those who are ‘fully out’ of the set (‘0’). The calibration of a fuzzy set is possible only through the use of theoretical and substantive knowledge, often assisted by statistical indicators, which is essential to the specification of the three qualitative markers (‘0.90’, ‘0.10’ (almost fully out), and the ‘crossover point’) (Ragin 2008, p. 30). Through this ‘philosophy of calibration’, fuzzy set QCA enables systematic cross-case comparisons that also considers within-case complexity (Stiller 2017, p. 84), a configurational comparative method that is gaining momentum in social/political sciences (Stoklasa, Luukka & Talasek 2017, p. 155).

Fuzzy set QCA offers numerous advantages over the classically employed ‘correlational thinking’ (i.e. regression modelling and econometric analysis). Firstly, fuzzy set analysis allows the researcher to utilise qualitative and quantitative sources of data to operationalise a theoretical framework (Mello 2017, p. 124), whilst it has been argued that the empirics informing regression analysis are based on ‘ad hoc’ specifications, with insufficiently rigorous theoretical foundations (Quibria 2014), which tend to ignore internal variation within cases (Suhrke, Villanger & Woodward 2005, p.2).

Secondly, fuzzy set QCA offers an ‘asymmetrical thinking’ approach, in juxtaposition to the ‘symmetrical thinking’ logic utilised by correlational analysis (Downey & Stanyer 2010, p. 334). To illustrate, if a researcher were to utilise correlational analysis to find a relationship between democracy and development, the researcher would find a weak correlation, as there are many less developed countries that are democracies. In comparison, with set theoretic thinking employed in fuzzy set QCA, the researcher would see a very strong set theoretic relationship, where developed countries are a subset of democratic countries (Downey & Stanyer 2010, p. 334). This would provide the researcher with a stimulus to examine what it is about developed countries that leads to democracy.

Finally, fuzzy set QCA allows the researcher to investigate the interplay between various causal conditions, and how they combine in a ‘recipe’ to produce an outcome; compared to correlational thinking that looks at independent variables as largely removed from one another, with each having a discrete additive net effect (Downey & Stanyer 2010, p. 335). This aspect of fuzzy set QCA is particularly useful in this paper, as many complicated, interacting factors combine to produce women’s political and economic participation. Therefore, fuzzy set QCA presents a unique opportunity to expand the feminist-political literature, to invigorate new scholarship that thoroughly investigates causal conditions producing changes in women’s politico-economic participation.

Truth Table Construction

The cases chosen for analysis in the truth table are that have emerged from self-described communist rule in the former Soviet Union or SFRY, and have sufficient data for causal and outcome conditions (Truth Table Cases). It is important to note here that many of the following analyses consider the causal condition data from world-wide countries. This is to provide a comparison in the construction of calibration points for the respective membership groups. Indicators (or ‘anchors’) were assigned to each condition, which sought to describe their level of membership in each condition as on the threshold of full membership (‘0.90’), the threshold of no membership (‘0.10’), or the cross-over point (‘0.5’), where their membership was the most ambiguous (Table 2). This set calibration allows the data to be truncated of unnecessary variation, with conceptual meaning attached to the set values through theoretical knowledge and statistical indicators (Buche, Buche & Siewert 2016, p. 362).

Hypotheses

Hypotheses are formulated as statements of conjunctive sufficiency in the fuzzy set QCA, as the analysis is based on a test of sufficiency (Stiller 2007, p. 87). The hypothesis which will be tested in this analysis are:

a.High amounts of loans or aid will produce a high degree of women’s politico-economic participation, if the country is democratic, with equal educational attainment, low infant mortality, and high income.

Outcome condition

The outcome in this analysis is women’s political and economic participation. Ensuring women’s equal politico-economic participation is considered crucial for achieving economic development and strong governance (Neumayer, Eric & de Soysa 2007); and conversely, women’s politico-economic participation deteriorates during times of conflict (Caprioli 2005). The data for women’s politico-economic participation was extracted from the Global Gender Gap Index (GGGI), as it contained sub-indexes considered more consistent with women’s politico-economic participation, as compared to other similar databases.

Causal conditions

Democracy

The first causal condition selected was the degree of membership in democratic countries. It is generally considered that democratic governments have higher levels of respect for women’s rights than their non-democratic counterparts, as politicians in democratic systems are constrained by institutional mechanisms, public pressure, and systems of regulation (Beer 2009; Wyndow & Li 2013). Membership of the set ‘high levels of democracy’ was established with data from the Polity IV project. The ‘cross-over’ point in the data was considered the value that represented a basic democracy, where political participation is unrestricted, open, and fully competitive, members of the executive are elected, and constraints on the chief executive are substantial (Figure 1).

Aid

The influence that aid has on not only development, but post-conflict human rights, is controversial. Scholars such as Sachs (2005) and Stiglitz (2002) have argued that aid is beneficial to development, but the literature is not convincing (see Edwards 2014). Data in relation to aid amounts was extracted from the World Bank’s World Development Indicators 2017. The net cumulative values of official development assistance (ODA) between 2010-2015 were summed for the 142 countries with data available. This inclusion of total aid flows appears to be broadly favoured by aid scholars (Chauvet 2003), as it allows researchers to more effectively trace aid patterns in relation to outcomes (Demukaj 2011, p. 29).

The fuzzy set membership of the entirety of this dataset was analysed, with standard statistical indicators (e.g. interquartile range, median, and box-and-whisker plots) calculated to aid calibration of the set points, as there was no external theoretical advice that classified the ODA data into high/low values. The high threshold of membership (0.90) was chosen as the highest end of a box-and-whisker plot of the data, the cross-over (0.50) was appointed as the median value, and the low threshold of membership (0.10) was the lowest point of the first quartile.

Loans

Similar to the debate surrounding aid, the effect of loans on women’s politico-economic participation is not settled. It is argued that IMF programs may weaken the government’s ability to enforce human rights protection (Abouharb & Cingranelli 2009), and that IMF conditions and policies may unfavourably impact upon women’s working conditions and job security (Detraz & Peksen 2016, p. 87). Membership of the ‘high levels of loan’ group was established with data from a variety of sources, with the total amount of loans combined over the 2010-2015 period. As with the aid causal condition, no external standards were available to determine membership categories of cases. The ‘cross-over’ point was considered the median value, with the high threshold of membership chosen as the highest end of a box-and-whisker plot of the data, and the low threshold of membership was the lowest point of the first quartile.

Net Income per capita

Previous studies have suggested that women enjoy greater economic freedom, empowerment, and personal autonomy when the society is economically prosperous (Richards & Gelleny 2007; Elborgh-Woytek et al. 2013). Data from the World Bank’s World Development Indicators database (2015) was utilised in constructing the net income per capita membership group. Income group categories are taken from the World Bank, which classifies economies into four income categories based on income per capita (Table 2).

Infant Mortality

Infant mortality has been hypothesised to be a strong indicator of women’s general health, economic conditions, and access to facilities (Mishra & Newhouse 2007), and increasing infant mortality rates have been suggested to be an early effect of conflict (Hoeffler & Reynal-Querol 2003). The cross-over point here was chosen as the Millennium Development Goal for infant mortality, the high threshold was the utmost value of the whisker box-plot of the data, and the low threshold was set utilising the lower quartile of the data.

Education

High educational attainment for women has been associated with greater political and economic participation (see Losindilo, Mussa & Akarro 2010). Data to complete the membership set of women’s high educational attainment was extracted from GGGI. Due to the low range of data across world-wide cases for this causal condition, indicators were selected following analysis of a line graph of the data, displayed in Table 2.

***

The truth table (Table 1) was expanded utilising the direct method as described in Ragin (2008). Following the identification of indicators for each condition, the deviation of ‘raw’ data (such as number of infant deaths per year) from the cross-over point for each country was calculated, and then a scalar value was calculated that combined the natural log of the odds of full membership with the deviation value as operationalised by Ragin (2008, p. 88). The final step before calculating membership was to multiply the scalar value with the deviation score, to achieve values that reflect the log odds of membership in the set, which conforms to the values attached to the qualitative anchors (Ragin 2008, p. 91). A standard formula was then utilised to calculate the degree of membership values that appeared in the final truth table:

Degree of membership = exp(log odds) / [1 + exp(log odds)]

A sample of these transformations for the set of ‘high educational attainment’ is shown in Table 3.

Table 1: Truth table with membership scores for women's politico-economic participation and causal conditions. (NIC = Net national income per capita)
Country Politico-economic participation Democracy Aid Loans NIC Infant Death Education
Latvia 0.80 0.82 0.00 0.05 0.90 0.07 0.98
Lithuania 0.74 0.99 0.00 0.00 0.93 0.05 0.97
Moldova 0.74 0.95 0.57 0.54 0.10 0.13 0.97
Belarus 0.71 0.01 0.50 0.48 0.57 0.05 0.98
Serbia 0.62 0.95 0.71 0.80 0.50 0.06 0.45
Kazakhstan 0.60 0.01 0.53 0.81 0.80 0.12 0.98
Croatia 0.47 0.95 0.50 0.00 0.87 0.05 0.96
Ukraine 0.46 0.18 0.68 0.98 0.09 0.08 0.97
Tajikistan 0.45 0.01 0.57 0.23 0.04 0.61 0.97
Azerbaijan 0.42 0.01 0.53 0.67 0.54 0.42 0.98
Russia 0.41 0.18 0.00 0.65 0.75 0.08 0.97
Kyrgyzstan 0.40 0.50 0.60 0.52 0.04 0.22 0.98
Bosnia 0.38 0.01 0.62 0.69 0.46 0.06 0.96
Georgia 0.37 0.82 0.64 0.68 0.30 0.10 0.98
Montenegro 0.36 0.95 0.50 0.20 0.64 0.05 0.97
Armenia 0.33 0.18 0.56 0.66 0.29 0.12 0.98
Table 2: Anchor points for the direct calibration of causal and outcome conditions
Country Politico-economic participation Democracy Aid Loans NIC Infant Death Education
Anchor Ave. GGGI economic + political indexes Polity IV democracy index (0-10) Net cumulative aid ($US) Combined Loans ($US) per capita ($US) per 100,000 live births GGGI educational index
0.90 0.64 9 $13.09 billion $30 billion $12,736 80.2 0.99
0.5 0.42 7 $1 billion $3.53 billion $4,125 31 0.95
0.10 0.21 3 $150 million $400 million $1,045 3 0.75
Table 3: Example of direct calibration in the set of ‘high levels of female educational attainment
Cases GGGI educational index Deviation from crossover Scalars Product Degree of Membership
Armenia 1 0.05 75 Бер.75 0.98
Belarus 1 0.05 75 Бер.75 0.98
Latvia 1 0.05 75 Бер.75 0.98
Lithuania 1 0.05 75 Бер.75 0.98
Ukraine 1 0.05 75 Бер.75 0.98
Slovenia 1 0.05 75 Бер.75 0.98
Czech Rep 1 0.05 75 Бер.75 0.98
Russia 0.997 0.047 75 3.525 0.97
Moldova 0.996 0.046 75 Бер.45 0.97
Bosnia 0.996 0.046 75 Бер.45 0.97
Serbia 0.996 0.046 75 Бер.45 0.97
Kazakhstan 0.995 0.045 75 3.375 0.97
Croatia 0.995 0.045 75 3.375 0.97
Bulgaria 0.993 0.043 75 3.225 0.96
Hungary 0.992 0.042 75 Бер.15 0.96
Romania 0.992 0.042 75 Бер.15 0.96
Kyrgyzstan 0.991 0.041 75 3.075 0.96
Georgia 0.989 0.039 75 2.925 0.95
Montenegro 0.989 0.039 75 2.925 0.95
Azerbaijan 0.988 0.038 75 Лют.85 0.95
Tajikstan 0.937 -0.013 15 -0.195 0.45

Causal patterns: Consistency and coverage

Once the truth table was constructed, the next step was to account for women’s politico-economic participation through the combination of causal conditions, operationalised as the consistency and coverage of the conditions (Ragin 2008, p. 24). Consistency assesses the degree to which cases share a combination (or ‘recipe’) of conditions in displaying an outcome (Reynaert 2011, p. 417), with the question being, is the outcome present if a certain combination of conditions is present (set-theoretical importance)? The coverage of causal combinations assesses the degree that they ‘account for’ instances of the outcome; asking how likely is the outcome to occur without the antecedent causal condition (empirical relevance)? (Buche, Buche & Siewert 2016, p. 362).

Calculations of consistency and coverage were undertaken using an online fuzzy set QCA software package, produced by Charles Ragin. In interpreting the output generated by the fuzzy set analysis, this paper utilises the ‘intermediate solution’ (outlined by Ragin 2008, p. 174), which makes the results more interpretable, as it doesn’t rely on untenable assumptions included in other approaches within this methodology (Buche, Buche & Siewert 2016, p. 369).

The analysis is presented below (Table 4), and according to these results, either:

b. the combination of low infant mortality, as well as high national income and high education (Model A); or

c. the combination of low amounts of loans, high aid, and high education (Model B),

are sufficient to support high politico-economic participation for women; in other words, in 73 per cent of these cases, Models A and B suffice to bring about high politico-economic participation for women post-conflict, and covers 88 per cent of the membership scores in the outcome of politico-economic participation (Stoklasa, Luukka & Talasek 2017, p. 58). The consistency scores for Model A, Model B, and the combination of both (solution consistency) is moderately strong, suggesting that the relationship of these causal conditions to the outcome is similarly robust.

Table 4: Analysis of combinations of conditions, intermediate solution. Consistency cut off = 0.75
Parameters of fit Education* ~infant mortality* NIC (Model A) Combination of Models A and B Education* ~loans* aid (Model B)
Consistency 0.78   0.92
Raw coverage 0.73   0.52
Unique coverage 0.35   0.14
Cases Latvia, Lithuania   Croatia, Tajikistan, Montenegro
Solution consistency   0.79  
Solution coverage   0.88  

Ragin’s seminal work on QCA interpretation assists with the analysis of the coverage scores reported here (2008, pp. 64-66). The interpretation supported by the data is that:

  • 35 per cent of cases with higher women’s politico-economic participation are explained by Model A without Model B (unique coverage of Model A);
  • 14 per cent of cases with higher women’s politico-economic participation are explained by Model B without Model A (unique coverage of Model B); and
  • 39 per cent of cases with higher politico-economic participation are due to a combination of all the causal conditions (i.e. the solution coverage minus the combination of Model A and Model B).

The sum of the unique coverages is 88 per cent. Together, these results suggest that Models A, B, and the combination of Model A and Model B, can sufficiently explain a high proportion (i.e. ≥85 per cent) of women’s higher politico-economic participation in post-communist countries.

One of the core strengths of fuzzy set QCA as a methodological approach is that there are multiple avenues of research interest from the analysis above. Firstly, the cases found with the models above need to be investigated more thoroughly, such as Latvia, Lithuania and Croatia (Stiller 2017, p. 94). In addition, this approach also points to the cases not covered by the solution or those which have membership in solution paths but do not feature high politico-economic participation for women (e.g. Bosnia and Ukraine), which need further in-depth investigation to reveal the explanation for this. Bosnia and Ukraine both have a very high amount of loans in the fuzzy set data, with high levels of aid. With the fuzzy set results in mind, it would be interesting to study these two cases and investigate why high aid here hasn’t assisted women’s participation, as in the fuzzy set, and what impact high amounts of loans may have had.

Case Study: Bosnia

At the start of the 1990s, the communist parties of Yugoslavia, observing the inevitability of democratisation, ceded power peacefully through democratic elections. However, a constitutional crisis followed these elections, and the rapid rise of Serbian, Croatian, and Bosnian ethno-nationalism provided the impetus for war (Inglis, 1998, p. 66). Women were overwhelmingly affected by the drastic social and economic changes during the transition from the Yugoslavian communist state. Social services previously established by the socialist government were dismantled, women’s political representation decreased dramatically, and aggressive decentralisation arguably made any provision of welfare practically non-functional (Avdic-Küsmüş, 2016, p. 40). With the increasingly violent rise of ethno-nationalism, women became the subject of identity politics, where they were assigned with ‘honourable’ roles as ‘Mothers of the Nation’, who were the embodiment of cultural and traditional values, but were in danger and had to be protected (Korac, 2006, p. 513). As the violence escalated into all-out conflict between warring ethnicities (the Bosnian War), women were especially targeted for atrocious violence; the nationalist identity politics surrounding Bosnian women had conceptually turned their bodies into territories to be ‘seized and conquered’, and were regarded as ‘precious property’ of the ‘enemy’ (Korac, 2006, p. 513).

The Dayton Peace Accords (the Accords) were the first major peace agreement that was adopted after the Fourth World Conference on Women in Beijing (Lithander, 2000, p. 12). Although the Accords were successful in ending the extremely bloody Bosnian War, most of contemporary literature regard the Accords as having been unsatisfactory for all sides of the conflict (Avdic-Küsmüş, 2016, p. 40). Following the peace agreement, the Accords instituted an extreme decentralisation process that divided the country into two major entities (Federation of Bosnia and Herzegovina and the Republika Srpska (RS)), and many smaller self-governing provinces (Caplan, 2000, p. 216; Avdic-Küsmüş, 2016, p. 40). The Accords have created one of the world’s least efficient and most complicated political systems, with a central government that is effectively powerless and holds an almost ceremonial role to the Office of the High Representative (OHR).

Women in Bosnia and the IFDI Impact

The supremacy that the three major ethnonationalist ruling parties enjoyed before the war, have been reinforced by the Dayton Agreement (Inglis, 1998, p. 84). The people of Bosnia and Herzegovina are hostage to the ethno-nationalist discourse that continues to dominate activities of political parties, the media, and civil society actors (Avdic-Küsmüş, 2016, p. 42). The rushed negotiations of the Accords were argued by warring parties, national power elites, and international diplomats, with a lack of non-governmental and non-male actors (Inglis, 1998, p. 77). No women were present at the negotiation table, despite the chronological proximity the peace accords had to the Fourth World Conference on Women, and the Dayton Accords consisted of a masculine dialogue with militaristic overtones (Lithander, 2000, p. 18).

Immediately after the conflict, the socio-politico-economic situation for women was dire. A significant number of the female population had experienced war crimes of rape and ethnic cleansing specifically perpetuated against them, and a majority of the internally displaced persons in Bosnia were women and children (Inglis, 1998, p. 102). The first post-war elections in 1996 were also a set-back for women’s political representation, with only 2per cent of Bosnia and Herzegovina’s House of Representatives being female (Bjorkdahl, 2012, p. 298). The situation has slowly improved since then, with now 21per cent of the House of Representatives being women, off the back of the OSCE ‘Women in Politics’ programme that changed Bosnian electoral rules and required 30per cent of the candidates on candidate lists to be women. However, women’s political participation is still frightfully low in 2015, as demonstrated in the preceding fuzzy set analysis.

With the aim of accession of Bosnia to the European Union, a neoliberal, market economy has been pursued in post-communist/SFRY Bosnia. This has included measures such as the freezing of public sector wages and a restrictive employment policy in various governments in the country, as well the cut of public servant salaries. The low national income can be seen in the fuzzy set truth table (NIC); and indeed, UNDP have demonstrated that there are generally very poor socio-economic circumstances for Bosnian households, with barely any disposable income available. There has been a strong push to structural market reforms by using both direct political interventions and conditional aid. For example, any delays to sign-off on legislation by the Bosnian government is often followed swiftly with threats by the IMF and World Bank to withdraw financial assistance. Indeed, the RS have been substantially burdened by this politicisation of aid, where they receive less than 5per cent of all international reconstruction aid allocated to Bosnia, despite being 49per cent of its territory, mostly due to the international community perceiving the entity is not conducive to Dayton-reforms.

The economic reforms implemented post-Dayton present a very narrow view of the economic system, as only focusing on the ‘productive economy’, rather than including the substantial human resource aggregates of the reproductive economy (True et al. 2017). Indeed, women’s participation in the formal economy is less that than men, but the significant informal economy is a huge amount of economic activity that women participate in, which is not visible in GDP or IFDI analysis. The Bosnian reform agenda has been completely gender blind, and there is no analysis on whether the progress of its’ implementation is gender sensitive, nor is there any consideration of what the impact of rampant sexual-violence and ongoing gender-related violence is on Bosnian households.

The continued privatisation of the economy is also problematic for Bosnia, as it presents more opportunities for male political and ethno-national elites to ‘grab’ public resources and extort political influence. Whilst the collapse of the public system has meant that women have had to pick up the slack of the state and the IFDIs through their increasing labour in the care economy, contributing to improvements in human development outcomes despite the extraordinary human injury and needs following a major war (True et al. 2017). There appears to be no end to the market-based reforms of the economy, as the IMF continues to direct loans to Bosnia that are heavily conditional on these structural reforms, adding to the vast number of conditions that have been applied to Bosnia (92 conditions as of 2014). This support has institutionalised the causes of conflict, disempowers women and those who would be the drivers of change, and freezes an intolerable status quo.

Case Study: Ukraine

The modern history of Ukraine through the 20th century has existed mainly within the Union of Soviet Socialist Republics (USSR) as the Ukrainian Soviet Socialist Republic from 1922. The role of women in soviet society was often an important concern to Communist Party leaders, and an issue that was often exploited for propaganda purposes. During the early history of the USSR, the ‘emancipation’ of women from an ‘illiterate, uncultured and counter-revolutionary’ heritage (Chatterjee, 2002) was celebrated by the central government. Women’s role in society was a highly important question to the Bolshevik social reformers, who conceived protective labour legislation, social equality policies, and progressive family law to encourage women to play an active role in the development of Soviet society (Ilic, 1996, p. 229).

The Soviet State attempted to equalise the roles of men and women in the labour force (True, 2003, p. 5), granted maternal welfare (Hrycak, 2001, p. 140), increased political representation of women in the Soviet Republic legislatures (Hrycak, 2005, p. 74), and instituted divorce laws that allowed women a reasonable escape from an unsatisfactory relationship (Funk, 2004, p. 712). However, the gender order was of most benefit to the State, where women were expected to perform their role as emancipated workers, as well as complete household chores and produce future generations of workers (Ashwin, 2002, p. 24).

Switching focus to contemporary Ukraine, the country has experienced a highly unstable political environment since independence. The decade following the ‘Orange Revolution’, the economic situation in the Ukraine worsened (Hrycak & Rewakowicz, 2006, p. 312-313). When Ukraine looked to reject European economic support for the Russian ‘option’, the ‘Euromaidan’ protest movement was sparked, which swept the country in late 2013 (Davis, 2016, p. 179). The subsequent political instability, along with insurgent armed actors, has resulted in the creation of the self-declared republics of Donetsk and Luhansk, and the annexation of Crimea by Russia (Davis, 2016, p. 179).

Women in Ukraine and the IFDI Impact

Following the fall of communism in the early 2000s, nationalist and conservative Ukrainian governments arguably pushed women back to their ‘purely womanly mission’, (Phillips, 2005, p. 258) in the home as a wife and a mother (Hrycak, 2005). The descaling of the Soviet State and the rapid decline in light-manufacturing enterprises meant that women were disproportionally left unemployed (Hrycak, 2001, p. 147). National gender mechanisms have been significantly weakened (UNECE, 2014, p. 11), and women’s political representation drastically declined to just 3% of women in the Ukrainian national legislature in the first election after the fall of communism (Hrycak, 2005, p. 74). Recently, this has slightly improved to 12.3 per cent after the 2014 elections, but as indicated by the fuzz set analysis, this is still objectively points to women’s poor political participation.

International actors (governments, the United Nations, and international monetary funds) promoted the dismantling of the social safety net in Ukraine, on a pretence of instilling ‘initiative’ in these ‘newly-liberated’ citizens (Phillips, 2005, p. 257). The new capitalist economy imposed a gender order over Ukrainian society. The aggressive, patriarchal masculinity that operated in the market, secured private property, and explored liberal opportunities, was celebrated (Zdravomyslova & Temkina, 2012, p. 56); whilst women, who had to accept the burdens of the destruction of the welfare system, increased costs of social services, and the commercialisation of education, were confined to their traditional roles as a housewife, as perestroika had envisioned (Zhurzhenko, 2001, p. 37).

The acceptance of austerity policies instituted by the IMF and EU in exchange for loans has resulted in further cuts of government spending towards human development and social policies in Ukraine (The World Bank, 2016, p. 10). These have been large loans, demonstrated by the fuzzy-set analysis; and indeed, Ukraine have recently been approved a $17.5 billion bailout program by the IMF, in exchange for the adoption of economic reforms and tackling corruption (RadioFreeEurope, 2017), which includes the cancellation of fuel subsidies, and to reduce spending on areas such as health, education and social wellbeing. These reforms of decentralisation may have negatively impacted women in two major ways; firstly, as women and their families tend to be the primary beneficiaries of pro-social spending, and, second, because women tend to be employed in the sectors where job cuts have taken place (Cleary, 2016, p. 14).

Indeed, 165,000 civil service jobs were cut during 2014-2015 in accordance with IMF demands, with further down-sizing planned. As women represent more than 75 per cent of the civil service, in non-managerial positions, they are disproportionately affected by these job losses. Around this same period, approximately 900 schools have been closed, budgetary responsibility for funding education, healthcare and science has been transferred to the poorly funded local government, and childcare assistance before the age of three has been abolished (The Women’s International League for Peace and Freedom produced a Joint Shadow Report for the CEDAW Committee’s 66th session). These reforms are proportionately more damaging to women in the Ukraine.

Conclusions

The fuzzy set analysis presents an interesting way to look at how certain conditions may impact upon women’s politico-economic participation, and here it has been useful in analysing what impact IFDI loan programs and aid have had. It appears there is reasonable evidence from this analysis that low amount of loans and high amounts of aid is beneficial for women’s politico-economic participation. The damaging effect of loans, and the conditions upon them, has been outlined in the case studies of Bosnia and Ukraine above. It appears that gender-blind, market-led neoliberal reform is potentially damaging for women’s political and economic participation, as well as their human rights. To further this fuzzy set QCA, and delve further into the theoretical case-findings here, field research in these countries is warranted, to determine at the household level what impact these reforms have had. It is vital, if IFDIs are going to have a continuing and domineering influence over post-crisis/conflict state/economy-building, that the gender impacts of neoliberal economic reform are understood, which will allow new conceptualisations and formulations for this assistance to emerge.

© 2015 ВГО Центр "Розвиток демократії"