Hybrid democracy : electoral rules and political competition in Afghanistan

Departing from the hybridity literature on peacebuilding and electoral studies, this paper treats democratization in Afghanistan as a hybridization process. As a result, to understand the durability of corruption, fragile political parties, and non-democratic practices in the country, one should look more closely at the interplay between the current electoral system and how elites and local people respond to it.


Introduction
W hy do a fragile party system and non-democratic campaign practices persist in Afghanistan?This question permeates Afghan reality, even after the last presidential election.On September 29, 2014, the first democratic transition took place, when former Minister of Finance Ashraf Ghani replaced Hamid Karzai, assuming office for an initial five-year term.Despite this achievement, the electoral process was marked by heavy accusations of fraud (International Crisis Group 2014) which led supporters of the defeated candidate, Abdullah Abdullah, to reject the initial results.It was only after international mediation that both candidates accepted a recount of the votes and the formation of a unity government, with the losing candidate assuming the role of chief executive officer.
The controversies surrounding the 2014 presidential elections are just another chapter in the fragile history of democracy in post-Taliban Afghanistan.Suspicions of fraud and corruption, as well as the intimidation of candidates and voters, are normal during election time (Coburn and Larson 2014).Furthermore, clientelism and patronage networks 1 are important political tools in the parliament: during his two terms, Karzai was able to maintain an extensive patronage system in the Wolesi Jirga with payments to MPs and other types of pressure in order to further his agenda (Sharan 2011).
Recent studies have tried to explain these problems.On the one hand, analysts and policymakers have continued to highlight the inherent fragility of Afghan institutions (especially the current party system) (United Nations Development Programme 2013) and the inability of the U.S. government and other international actors to devise a proper state-building strategy for the country (Ponzio 2011).On the other hand, recent studies have put more emphasis on another set of problems.In particular, non-state armed groups running in elections (Giustozzi 2012), elites fighting amongst themselves for control of the state (Sharan 2011), and a lack of attention to local politics on the part of international actors has created a situation in which elections and democracy have been perceived by the majority of the population to be nothing more than an arena for different groups to compete for power (Coburn and Larson 2014).
Although built upon all of these findings, this article takes a different route.Relying on insights drawn from the hybridity literature on peacebuilding and electoral rules, as well as electoral data gathered by think tanks working in Afghanistan during the 2005 and 2010 legislative elections, I look at democratization in Afghanistan as a hybridization process.Broadly speaking, hybridization should be understood as constantly mixing processes that challenge and transgress essentialist conceptions of identity by highlighting how subaltern agents resist, and even subvert, the power of elites.In peacebuilding contexts, hybridity is an analytical tool used to reveal how relationships between international actors, national elites, and local communities can create new forms of conflict and peace.
Based on these insights, I propose the following argument: In order to understand the resilience of non-democratic practices and the fragility of Afghan democracy, we must also bear in mind the interplay between an electoral system that provides institutional incentives for the fragmentation of political parties and the cultivation of a personalistic vote, and how elites, organizations, and citizens respond to them.Therefore, in a complex context in which transitional justice is absent, the formal justice system is still very fragile, and massive flows of financial resources distort social and political relations, new rules frame political competition, confronting and coalescing with other forms of politics; this results in a hybrid democracy.In other words, even though democratic procedures have framed political competition in the country, they are perceived as simply another set of tools for maintaining elites in power, rather than as a system that engages the majority of the population with the formal structures of government.
The article is divided into three sections.First, I outline my analytical framework, presenting a dialogue between the hybridity approach to peacebuilding and insights from electoral studies.Second, I analyze the hybridization process in Afghanistan.Finally, I summarize my findings.
1 Overall, clientelistic practices represent the exchange of a citizen's vote in return for direct payments or continuing access to employment, goods, and services (Kitschelt and Wilkinson 2007).Patronage can be defined more narrowly, related to the use of state resources to provide jobs and services for political clienteles (Van de Walle 2007).

Hybridization and electoral systems
In order to take into account the complexity of peacebuilding activities, in which international actors attempt to introduce new forms of political and social organizations into societies with their own histories of social and institutional practices, analysts have begun to use the ideas of hybrid political orders and/or hybrid peace (Clements et al. 2007;Richmond 2011), starting what has been labeled the "local turn" in peacebuilding.Inspired primarily by the notions of hybridity and hybridization, 2 some of these authors claim that peacebuilding activities entail a collision between the strategies, institutions, and norms of international agents and the everyday lives of the local actors who are affected by conflicts.As a result, what is sometimes considered a failure of liberal peacebuilding is actually a sign of the success of local claims for autonomy or maybe a negotiated -or even subverted -form of peace, which are currently described as hybrid forms of peace.
As Richmond (2014) argues, the power relations that permeate these encounters, and the political structures that emerge as a consequence of them, may not necessarily lead to more positive and emancipatory forms of peace.A hybrid peace may lead the way towards more integrative and progressive forms of social organization, or, on the contrary, maintain, or even worsen, a scenario of structural violence and inequality.The point is to draw attention to the ways in which the interplay between international and local actors activate different forms of agency, which can vary from acceptance to resistance or even co-optation (Richmond and Mitchell 2010).
According to a hybridity perspective, all states and societies are, ontologically speaking, hybrids.In epistemological terms, any attempt to understand and produce knowledge about the dynamics of international interventions should avoid, for instance, problematic concepts such as "failed/fragile states" and focus on how actors confront and negotiate their actions, interests, and needs while embedded in specific socio-political and cultural contexts.Approaching peacebuilding through hybridity and hybridization perspectives gives us at least four advantages.First, it reminds us of the need to observe the interfaces of exchanges between actors.Second, it encourages researchers to recognize cultural fluidity within groups.Third, it moves us away from elite-level analyzes and forces us to take local actors seriously.Finally, it pushes us further away from state and institution-centric approaches (Mac Ginty and Sanghera 2012).
One of the main challenges so far in terms of methodology has been to devise concepts and analytical frameworks to approach hybridization processes, and at least three merit a mention.First, the Mac Ginty (2011) model is a conceptual proposal for mapping hybridization in peacebuilding contexts, emphasizing the interactions between international and local actors.Second, Millar's (2014) disaggregation of hybridity is an attempt to differentiate types of hybridizations.Third, the concept of friction (Millar et al. 2013) was designed to map the unintended and unexpected consequences of peacebuilding interactions.While the hybridity approach highlights outcomes, friction intends to draw our attention to the processes themselves.
2 Drawn from post-colonial studies, these concepts are associated with authors such as Homi Bhabha (1994) and Nestor Canclini (1995).For an overview of this discussion and its connections with peacebuilding, see Peterson (2012).
These frameworks, and the local turn itself, still face considerable challenges. 3It is important to emphasize that, although the above attempts have drawn our attention to these internationallocal interactions, one might argue that they are epistemological orientations which do not offer enough concepts to grasp the specificities of given empirical realities.To analyze hybridization in different contexts such as governance, security, and democratization, we must consider the particularities of each of these dimensions.Therefore, given that the goal is to analyze political competition in Afghanistan, we must take other concepts into account that can allow us to understand the empirical reality of the Afghan political system.I propose bringing insights about electoral institutions into this discussion.
Why electoral studies?There are empirical and theoretical reasons for this choice.Empirically, I follow Giustozzi (2012) and Perelli and Smith's (2014) diagnoses that, despite all its problems, the electoral process in Afghanistan was not a sham.The actors involved followed institutional procedures, mainly because they saw them as sources of political legitimacy.As a result, electoral rules should be taken into consideration, as they have imposed pressures and constraints, whether material or symbolic.
Theoretically, just as in representative governments, political institutions are intended to shape the rules of the game under which democracy is practiced, and electoral systems are of great importance because they mediate the relationship between would-be MPs and voters (Reynolds et al. 2005).At its most basic level, electoral systems translate votes cast in elections into seats won by parties and candidates.The key aspects are: (1) the electoral formula, or the way votes cast in a general election are counted and translated into seat allocations; (2) the ballot structure, which relates to whether the voter votes for a candidate or a party, makes a single choice, or ranks their preferences; and, (3) the district magnitude, which has to do with how many representatives a given district elects to the legislature.
It is also important to recognize that elections and peacebuilding are often associated with one another by studies that try to understand how to channel conflict in multi-ethnic societies into non-violent processes.Important questions in this literature include the timing of elections, the choice of the electoral system, and the effects of post-conflict elections on the development of civil society and political parties (Reilly 2004).Therefore, instead of following this path, I combine these two literatures in an original way in order to further understand hybridization processes concerning political competition.
For the purposes of this article, the starting point is what kinds of incentives the electoral system presents and how key actors respond to them.If this were the only objective, it would be fair to say that this analysis would be like rational-choice approaches, deploying theoretical simplifications (for example, politicians are vote-maximizers, and voters support candidates expecting particularistic governmental benefits).From a hybridity perspective, however, the incentives (and responses to them) of candidates and voters amount to more than just the expectation of winning an election or supporting a given politician to receive some form of benefit in exchange.The strategies for winning votes, the processes of winning over voters, and choices of candidates are all mediated by values and social norms.
In this sense, voting is more than just an individual act: it has a symbolic aspect.Negotiations between candidates and voters, for instance, do not just involve choosing who is going to be the next representative, but also how voters and candidates are situated in societies (Palmeira and Heredia 2010).As a result, the combination of electoral studies and hybridity provides two important inputs to analyze.There is still the need to consider the incentives and constraints offered by institutional designs.However, we must move beyond the rational-choice conceptual assumptions outlined above because the theoretical subject of hybridity is richer than the abstract rational citizen, and is involved in a network of social practices that are activated when in contact with these new electoral characteristics.These contacts create new spaces for agency that enable other positions to emerge, creating the possibility for new structures of authority and new political initiatives.
Looking at political competition in Afghanistan through this lens, we can gain three main analytical advantages.First, this analytical outlook offers proper tools for appreciating how the new spaces for political competition that are opened by electoral rules interact with other forms of politics, and how international actors, elites, and local actors can engage with, and even subvert, the institutional framework.Second, it prevents us from overemphasizing the weight of electoral rules when trying to understand the resilience of non-democratic practices, while also preventing us from romanticizing local actors as either the source of problems, or solutions for, Afghan democracy.
Third, we can also take into account negative forms of hybrid democracy, in which elections and party systems exist, but the main hybridization result is a situation where fraud, corruption, and other political practices distort political competition.Additionally, we can also imagine an ideal situation in which the main result of hybridization would be a state of affairs in which most people perceive political competition as free and fair, and there are ways for direct and indirect local engagement in political processes.As shown below, the Afghan case is an example of the first situation.
This analysis relies upon a series of Afghan official documents, reports prepared by the Afghanistan Research and Evaluation Unit (Coburn 2011;Coburn and Larson 2009;Lough 2011) and the Afghanistan Network Analysis (van Bijlert 2009;2011), among other sources.These documents and reports are the most comprehensive analysis of Afghan campaigns, with data gathered from both the 2005 and 2010 elections, which provides analysts with a considerable amount of information.I use three analytical axes to interpret these data: elections systems, electoral dynamics, and parliamentary politics.
The first axis' focus is on the institutional aspects of the electoral system that were put in place after 2001 and the incentives they provide.The second axis draws attention to the interplay between these institutional rules and how elites and local actors react to and negotiate them during elections.Finally, as electoral rules also affect how parties and politicians organize themselves within parliament, (i.e., through collective or individualistic terms (Carey 2005), the third axis aims to evaluate aspects of the political game within the Wolesi Jirga.

Political competition and hybridization in Afghanistan Electoral system
The process of choosing Afghanistan's new constitution and electoral laws goes back to the 2001 Bonn Agreement.Considering the unstable circumstances of the moment and the idea that Afghanistan could not afford to have a weak central authority, international supporters and local elites opted for a system that emphasized presidential authority (Grote 2004).In line with this purpose, the selection of representatives for the Wolesi Jirga is based on the single nontransferable vote (SNTV). 4 Under this system, Afghan voters cast ballots for only one individual candidate rather than political parties, and each province elects a number of candidates, some of whom have to be women. 5The candidates who get the most votes within each district are elected.As Reynolds (2006) argues, the advantage of the SNTV is that it is simple: it promotes the representation of independents within a nascent party system and can develop a sense of accountability, as voters know for whom his/her vote was cast.
There are some problems that are inherent to this system, however.Its main flaw is that it discourages the development of disciplined parties for three main reasons (Reynolds et al. 2005).First, party leaders do not have control over the order in which candidates appear on the ballot or over who the party endorses.Second, votes are not pooled across parties, but cast for individual candidates.Therefore, the electoral fate of a politician is not tied to the performance of her entire party.Finally, the combination of SNTV and a multi-member district system creates intra-party competition among candidates from the same party.Simply put, as various candidates from the same district can run for the same seats, SNTV leads to incentives for politicians to boost their personal reputation at the risk of going against the directions of party leaders.
Why does Afghanistan use this electoral system?First, this choice must be understood in the context of the struggle between the former Northern Alliance group and Karzai's internationally backed network.As a constitutionally stronger presidency would clearly favor Karzai's team (Sharan 2011), they may have hoped that the electoral system would delay the emergence of new parties, fragment the existing opposition block, and weaken the ability of regional strongmen to install their followers in parliament (Reynolds and Carey 2012).Second, there was also the danger that political parties might crystallize ethnic divisions in Afghan politics (Suhrke 2008).Additionally, we must also bear in mind the U.S. influence in the country; U.S. Ambassador 4 The effects of the SNTV are stronger in Afghan legislative elections, as presidential elections are based on a majority voting system.The president is elected by majority vote through direct voting (see The Constitution of Afghanistan 2004, Article 61), and the whole country is considered a single electoral constituency (see Islamic Republic of Afghanistan, Decree of the Transitional Islamic State of Afghanistan on the adoption of the electoral law, 2004, Article 10).
5 Article 83 of the Afghan Constitution states that two female candidates must be elected to the Wolesi Jirga from each province (Islamic Republic of Afghanistan 2004).
Zalmay Khalilzad personally pressed UN officials for the SNTV.An electoral system capable of shielding the president from the parliament and weaken parties was consistent with the U.S.'s interests of reinforcing its relationship with the presidency (Suhrke 2008).
In light of the SNTV, some institutional effects were expected.It tends to be more difficult to build strong parties, which affects the way politicians campaign and the dynamics within the Wolesi Jirga.Regarding campaigning, it is common for candidates to look for other ways to get votes rather than relying on party platforms.As a result, it is common to use clientelism and religious, ethnic and/or patronage networks to win votes in countries under SNTV.SNTV also tends to discourage strong opposition against the president.The responses to these two institutional effects are the focus of the next subsections.

Electoral dynamics
This axis of analysis aims to focus our attention on the interplay between the SNTV and the ways in which elites and local actors react to, and negotiate them, during elections.To further explore this hybridization process, I propose three main areas of interest: electoral behavior, the coordination problem, and campaign strategies.
Regarding electoral behavior, the point is to understand what is most important for determining candidate and voter behavior during campaigns.In this regard, a well-known characteristic of Afghan politics is its localism.Historically, kings and leaders have relied on local networks to govern a country in which attempts to centralize power and authority have from time to time faced resistance from ethnic, religious, and other local groups (Schetter 2013).Combined with an electoral system that encourages personalistic voting, one of the current manifestations of localism is the reliance of candidates and voters on ethnic, religious and/or clientelist linkages to position themselves during and after elections.Following van Bijlert (2009), six principles, alone and overlapping, tend to frame voters' decisions and political alliances in Afghanistan and could be considered current manifestations of localism.
The first one is instruction.Voters adhere to leaders' decisions based on the assumption that they will further their group's interests.Second, loyalty also plays a significant role: Afghans show support to certain politicians due to ties of friendship, duty, obligation, and reciprocity.Another important principle is patronage, strongly influenced by the concept of "waseta," which permeates Afghan society."Waseta" is based on the assumption that one will always need a person "on the inside" to get what she needs (van Bijlert 2009).Consequently, candidates and voters tend to engage in this kind of relation to get what both of them are seeking: political support and benefits, respectively.
The fourth factor is pressure, whether through intimidation, threats of violence, or marginalization, particularly in rural areas, where one's vote is rarely secret (van Bijlert 2009). 6 fifth is negotiation: voters do not vote blindly, and sometimes they seek to evaluate (individually or as a community) the consequences of their decisions by assessing the risks and estimating the value of the deal they are being offered.Finally, even though elections tend to be marked by personalism and reciprocity, programmatic considerations do seem to have a measure of influence in voters' decisions as well.
It is important to keep in mind that under SNTV, the development of strong political parties tends to be hindered.Nevertheless, this situation is worse in Afghanistan because, for historical and institutional reasons, parties have been perceived as personalistic entities whose programmes consistently repeat the same catchwords such as independence, national unity, and democracy, with just minor differences between them (Ruttig 2006).For example, there is an emphasis on social justice in more leftist parties, and a focus on religion on the Islamist side.Even so, it is no wonder that, for many candidates, the party is seen more as an organization that should provide material benefits to its constituencies and less as a group through which politicians should accomplish their political and policy goals.
In terms of coordination problems, it is important for us to understand that under SNTV the vote margin separating the lowest winning candidate from the highest losing one is often small; as such, not wasting one's vote is an ever-present concern.As a result, local communities face a serious dilemma: how many candidates from the same party (or from other organizations) should run in the district?In light of this context, bloc voting plays a key role.
Bloc voting, or the support of a whole group to a candidate, is a regular feature in Afghan politics, although the size of the bloc varies.Even though bloc voting might discourage individual reasoning and choice in elections (Coburn and Larson 2009), this political behavior should be understood in the context of the importance of localism and as a response to the new rules of political competition under SNTV.Furthermore, this kind of political organization persists in Afghan elections mainly because, at the local level, political and socioeconomic concerns tend to be framed by concerns about community and patronage.Acting individually wastes political capital, and bloc voting might generate better pay offs in the form of gains in political patronage (Coburn and Larson 2009).
To do so, political groupings based on qawn, 7 ethnicity, lineage and religious affiliation, the most common sources of power in local communities, act mainly in two ways: determining the number of candidates from a given community who will participate in elections, and negotiating as to whom the bloc should give its support.Taking into consideration findings from the Kabul province (Coburn and Larson 2009), it is possible to point out the role of political brokers, and local shuras in particular, in regulating the number of candidates in the field.
Two examples are worth mentioning.On the one hand, a provincial council member in Dasht-i Barchi insisted that, before engaging in another election in 2009, the elders from the local qawm ensure her that no other candidate from the family group or location of origin would run a campaign (Coburn and Larson 2009).On the other hand, during the 2005 elections, the Polpozais were the best organized tribe for the election in Kandahar: ex-President Karzai's brother, Ahmed Walid Karzai was able to use his influence to reduce the number of Polpozai candidates and organize support for only two politicians (Wilder 2005).Although these actions are not an exclusive feature of Afghan elections,8 they are all vital concerns under an electoral system that punishes wasted votes.
One strategic tool for the negotiation of blocs of votes among candidates is ambiguity.Bloc negotiators are frequently ambiguous, not only about whom a given group should support, but also about the composition of the bloc, which can be seen as a way to gain more space to maneuver among candidates.Several elders recounted receiving phone calls and visits from multiple candidates, but they waited as long as possible to declare who would receive their support (Coburn and Larson 2009).In spite of the acknowledged importance of bloc voting, it might not be as cohesive as it appears: transformations brought by urbanization and a younger electorate are changing voting patterns, reducing the strength of political brokers, and consequently diminishing the power of some blocs.
Finally, vis-à-vis campaign strategies, I seek to explore what kinds of strategies SNTV has activated in Afghan society.My findings point to two important features: personalism and fraud.Furthermore, there was a significant (unexpected) effect related to the performance of women candidates, which can be seen as another example of the hybridization process.
Personalism is a central characteristic of Afghanistan, and it is further incentivized by SNTV.Throughout the 2005 and 2010, parliamentary elections, campaigns focused on individual candidates rather than political issues.This situation should not be a surprise, as no fewer than 5,800 individuals competed against one another for 249 seats in Afghanistan's first parliamentary election, relying on their identification with particular groups or trying to mobilize networks of voters in order to win (Wilder 2005).Typical tactics were advertising on posters, television and radio, and holding campaign rallies and political gatherings.In these particular political meetings, people usually gathered in someone's house to listen to information about the upcoming elections, and politicians explained how they would benefit the local population.Those present, however, were not only to demonstrate support for the candidate but also (perhaps even more importantly in this context) demonstrate support for the organizer of the meeting (Coburn and Larson 2009).
Under a system where the difference in votes between the winning and losing candidates tends to be small, fraud has been a useful way to improve politicians' chances.The first signs of such practices were documented during the 2004 elections (van Bijlert 2009, 19) when a total of 10.5 million voter's cards were distributed, even though there were only approximately 9.8 million voters in the country.By 2010, at least 17 million cards were estimated to be circulating in Afghanistan, which surpassed the estimated total of 12.5 million registered voters in the country (van Bijlert 2011, 4).Another feature of the electoral process has been the manipulation of data during vote counting.For instance, during the 2005 elections, there were allegations of poll workers invalidating the ballots of rival candidates, tallies being changed when copied from one form to another, and figures in forms being changed during data entry (van Bijlert 2009, 21).In 2010, at least 1.5 million votes were controversially disqualified and 27 candidates removed from the election.Some specific forms of fraud have emerged thanks to the interaction between new electoral rules and the local context.As an example, the 2004 Constitution and Afghan Law have entitled women to political rights, such as being able to vote and stand for political office.Nevertheless, in some Pashtun areas, "proxy registration" of female voters has taken place: male members of a given family are allowed to register their female family members by collecting a number of cards to distribute among women across their communities.This situation has resulted in the emergence of ghost voters, as the number of women is often inflated.
The performance of women during elections also shows an important effect of the hybridization process.The adoption of the SNTV, combined with the quota, led some analysts to initially raise doubts about the participation of women in the parliament, especially due to the fact that female candidates could win their seats with fewer votes than losing male candidates, which possibly could have generated some sort of resentment against women legislators (Reynolds 2006;Carey 2005).What actually happened, however, was different.Campaigning in tough situations, in a country where women do not have the same liberties as male candidates to move around the territory, female politicians had to be creative in order to circumvent such challenges.As a result, younger women have emphasized their youth and beauty, and older ones their respectability in their communities.Many of them have relied on private gatherings in large houses with other women, and have hired cars with megaphones, amongst other strategies (Coburn and Larson 2014).
The result is that the majority of women do almost as well as their male counterparts.For instance, 19 women out of 69 in 2005, and 18 out of 69 in 2010, won enough votes to be elected without the help of the quota, 9 a phenomenon that was not restricted to a given geographical area (Reynolds and Carey 2012, 11).Nonetheless, although an important theme, gender issues were not fully addressed by the parliament.Explanations revolve around the exact effect of quotas, the problematic assumption that, with female legislators, women will necessarily be represented (Coburn and Larson 2014), and the effects of the SNTV in parliament, which discourages both the building of collective positions around gendered platforms.No less important, we must bear in mind the conservative profile of many female politicians in the country, which further weakens feminist agendas (Lough et al. 2012).
These events have not only shown the engagement of the actors who are involved with the new rules, but also how these same rules have enabled forms of political action other than those that were expected.Personalism was further incentivized, different sorts of fraud (old and new) have taken place, and women have emerged as a significant political voice.Consequently, the hybridization process indicates that we are dealing with more than a mere acceptance of the rules of the election game.These rules, in connection with the reality of Afghan society, are creating something new, if not necessarily the best outcome in regard to political competition.9 Nevertheless, all of them were awarded quota seats.

Parliamentary politics
Looking at the politics within the Wolesi Jirga is important for understanding the hybridization process in Afghanistan.This is because it is an essential institutional place in which the new democratic rules of political competition meet social practices.As such, two dimensions are worth exploring: how the relationship between representatives and voters is mediated, and how the legislature interacts with the executive branch.Concerning the first dimension, two questions are of interest: why one would want to be a representative, and what is expected from a representative.
Broadly speaking, the main incentives to run for a seat include considerable political influence, the ability to have a pivotal role at both the local and the national level, a good salary, the potential for patronage gains through delivering services to communities, and immunity from prosecution (Wafaey and Larson 2010).Beyond that, engagement with democratic politics also seems to be an essential source of legitimacy to some candidates in the new post-conflict environment.This is mostly the case for armed groups: voter support, whether genuine or not, came to be seen as the best way to prevent marginalization at the national level and secure entrenchment as local leaders (Giustozzi 2012).Finally, there are several cases reported of candidates running not to be elected, but just to raise their profile.Some of them wanted to be regarded as local leaders, while others planned to launch their candidacy in order to negotiate a settlement with other candidates and then step back in their favor (van Bijlert 2009).
As for financial incentives, the basic income of an MP is their salary, which is around US$ 2,000 per month, supplemented by other Afghan and international funds (Coburn 2010, 3).Nevertheless, other sources of income are common among MPs.First, some politicians have their personal sources of income, mainly coming from their families and communities, which allow them to run expensive campaigns; 10 under SNTV, candidates running personalistic campaigns need to differentiate themselves from competitors.Second, Afghan businesses, especially import/ export merchants, are considered an important source of revenue for politicians as they seek to gain access to government contracts and/or protection.Finally, although illegal, there are reports of foreign countries financing representatives' campaigns, 11 attempting to shape the outcomes of elections in their favor, especially after the withdrawal of the U.S. troops (Coburn 2011).
In spite of these diverse forms of funding, being an MP is an expensive responsibility, mainly because of what is expected from a representative.Under SNTV, a politician's fate is tied to his ability to deliver particular governmental benefits to his/her constituency and distribute gifts and personal favors (Cox and Thies 1998).In Afghanistan, this is no different: MPs are often viewed as direct advocates of their constituencies, whether they are tribes, solidarity groups, 10 According to Article 49, Paragraph 1 of Afghan Electoral Law, campaign spending should not surpass certain limits, such 10,000,000 Afghanis (approximately US$170,000) in the presidential campaign, and 1,000,000 Afghanis (approximately US$17,000) in legislative elections (Islamic Republic of Afghanistan, 2004).Nevertheless, as Coburn (2010, 4)  or other communities, and they are expected to divert funds, resources, or aid to these areas.This circumstance has fueled more fragmentation and individualism in the country's politics because communities compete to elect their own candidates (Reynolds and Carey 2012).Conversely, communities without a representative in the government tend to feel unrepresented.Finally, MPs are regarded by voters as connections to the central government and are judged by their ability to not only provide goods but also to provide access to ministers, embassies, and other officials.
Political networks, however, are not restricted to resource provision and patron-client relations, particularly in the countryside.They also have a symbolic dimension and are about social relationships, marriages, and religious obligations.Many voters value their MPs for fulfilling traditional roles, such as providing hospitality for members of the community, holding feasts, attending marriages and funerals, and being involved in the resolution of local disputes (Coburn and Larson 2014).These activities sometimes are not only at odds with the Western tradition of representative government but might also conflict with the authority of other local leaders.It is therefore little wonder that a representative needs considerable financial resources to fulfill all of these roles.Wealthier MPs can maintain local offices to assist their communities, while others must rely on their local groups.Otherwise, as an MP from Herat complained: "every time we return to our province we need 100,000 Afghanis (US$2,200) for airplanes and bodyguards" (quoted in Coburn 2011, 3).
With regards to executive-legislative relations, supporting or constraining the president is another important topic.Even though there is no formal separation between pro-government and opposition blocs in Afghanistan, representatives have usually been divided into three main groups since the 2005 elections: pro-government, pro-opposition, and non-aligned, each having approximately 80 MPs.However, this division is a weak basis for analysis given that political organization and bloc formation within the Wolesi Jirga is extremely fluid, primarily because of formal institutional incentives and informal power and patronage relations (Coburn and Larson 2014, 91).
Institutionally speaking, a combination of the effects of the SNTV, which generates considerable difficulties for party cohesion within the Wolesi Jirga, and a presidency with high constitutional powers vis-à-vis the legislative 12 has created a situation in which it is possible for the former to enact its policy agenda and eventually bypass the legislative even though it does not have partisan power (that is, the ruling party does not necessarily have a bigger share of seats in the legislature).This situation has framed executive-legislative relations in a certain way.Within an institutional electoral system that encourages party fragmentation, it is worth for representatives to support the executive agenda if the president is popular, as this support is expected to increase one's chances of reelection.When the president is unpopular among the population, it is expected that representatives will oppose the presidential agenda in order to improve their personal popularity with the electorate (Kasuya and Kendal 2013).
12 The following powers are representative of the strength of the Afghan presidency.First, the package veto, which allows it to veto bills that the legislature has passed (The Constitution of Afghanistan 2004, Article 94,).Second, based on article 79 of the constitution, the president can issue a decree that can become law during recesses of the Wolesi Jirga except in matters related to budget and finance.Third, the president can call a national referendum for issues of national importance (Article 65).Fourth, regarding the budgets, the Wolesi Jirga cannot delay the approval of budget proposals for more than one month after receiving it (Article 98), which indicates the power of the presidency to limit the legislature's ability to revise the budget (The Constitution of Afghanistan 2004).During Karzai's first term, the Wolesi Jirga tended to support the executive's agenda, but when his popularity started to fall on the eve of the 2010 parliamentary elections, representatives began to distance themselves from the president.Most MPs blamed Karzai for the country's poor economic performance, complaining about the lack of institutional capacity for monitoring government offices and criticizing the extensive use of presidential decrees (Weinbaum 2012).As a result, opposition to him started to increase.For example, executive budget proposals were rejected five times in a row (for the 2010, 2011, 2012, 2013, and 2014 fiscal years) by the Wolesi Jirga mainly due to disagreements about the allocation of resources among provinces and privileges for the security sector, and there were also threats to impeach some of Karzai's ministers for failing to execute the budget.
Nonetheless, this situation did not necessarily lead to gridlock in executive-legislative relations.Beyond relying on his presidential powers to circumvent situations like these, Karzai could also make use of the patronage networks within the Wolesi Jirga.As Sharan (2011) and Coburn (2011) have shown, Karzai maintained an extensive patronage system in the parliament throughout his term based on monetary payments and intimidation.For instance, during the 2011 election of the new speaker of the Wolesi Jirga, MPs openly stated that their vote would cost around US$10,000 (Coburn 2011, 8) while a pro-Karzai MP said that he was supporting the government because "I have been accused of killing 41 people so if I do not support him, what guarantee is there that they would not kill me?" (quoted in Sharan 2011Sharan , 1123)).
In sum, the institutional design of the chamber empowered specific forms of political action.Local actors in particular started to view parliamentary politics as a new source for political legitimacy and have engaged with these new rules.Beyond the mere acceptance of these rules, new and old forms of agency have been activated, mainly related to local perceptions about what an MP should do.Finally, regarding executive-legislative relations, the president, either through formal or informal means, has remained almost unchecked by the parliament.

Final remarks
This article has attempted to provide an understanding of why weak institutions and non-democratic practices persist in Afghan politics.As I have shown, the dynamic between institutional incentives and social practices has played its part, embedded in a broader context characterized by complex power relations between foreign agents and local actors.Although local actors in Afghanistan have abided by democratic political competition, these new spaces for agency have produced institutional incentives that contribute to the resilience of practices of fraud and corruption during election time.With respect to parliamentary politics, a combination of SNTV rules, institutional traits, and other types of processes has created opportunities for the practice of corruption and coercion, resulting in a scenario in which the executive has great space to maneuver, political parties are fragile, and people's access to policy processes are still problematic.
These findings can contribute to ongoing debates in at least two ways.Theoretically, this article has pursued the idea that we need to supplement the hybridity approach to peacebuilding with other concepts that allow us to understand concrete empirical realities.The goal of this article was to understand political competition and thereby dialogue with electoral studies.In addition, the proposed analytical framework is a novel way of linking the electoral studies and peacebuilding literatures, which could assist researchers who are interested in mapping other hybridization processes involving elections in post-conflict settings.
Empirically, the findings contest essentialist analyses that have overemphasized so-called intrinsic problems of Afghan society.These analyses usually propose a unidimensional causality for the problems in the country, mainly related to state fragility, but without calling into question the role of international actors or even the political economy of political competition in the country.In this regard, the hybridity-electoral rules approach have shown us that the Afghan political system itself, and the SNTV in particular, put into place by an alliance of international and local actors with their own interests, has contributed to fragmenting political parties and incentivizing personalistic campaigns.Moreover, the contact between these institutional incentives and the country's reality, which went beyond the mere acceptance of new rules, has activated forms of agency that have played a significant role in explaining the resilience of practices of reciprocity and patronage networks in Afghanistan.
Finally, I do not want to discard other vital features that could explain the failures of democratization in Afghanistan.The main claim in this article is to point out that there exists a complex scenario in which elites and local actors interact with institutional incentives, and that this kind of phenomenon deserves an in-depth analysis.
estimates, an average parliamentary campaign can spend approximately US$100,000.11 Article 49, paragraph 2 of the Afghan Electoral Law states that no candidate should accept technical or financial assistance from foreign citizens or states (Islamic Republic of Afghanistan 2004)