‘Public–Private Entanglement’: Entrepreneurship in Lebanon’s Hybrid Political Order

Abstract While the literature is clear that political influence and clientelism characterises the investment decisions of entrepreneurs and the performance of their firms when governance is weak, it is less understood how governance systems and entrepreneurs interact, particularly when governance is of a hybrid nature. We address this issue in this paper by studying how entrepreneurs obtain access to electricity in Lebanon, showing that the hybrid political order imposes a high cost on electricity. We furthermore find that a hybrid political order channels entrepreneurial talent into lobbying and bribery. The key constraint that emerges from the hybrid political order in this case is the corrupt organisation of governance of the electricity sector. This results in higher prices (because bribes for contracts have to be earned back) in entrenchment of oligopolies, because contracts often come with political protection.


Introduction
Entrepreneurial decision-making and firm performance are influenced by the institutional context within which entrepreneurs function (Naudé, 2010a(Naudé, , 2010b. One important dimension of any country's institutional context is the extent to which it is either a hybrid, open, or limited access order (Boege, Brown & Clements, 2009;North, Wallis & Weingast, 2009;North, Wallis, Webb & Weingast, 2007). While hybrid and limited access orders are argued to be detrimental for the development of entrepreneurship, there remains a gap in our understanding of how political hybridity and limited access de facto impact on entrepreneurship (Brück, Naudé & Verwimp, 2013;Kenyon & Naoi, 2010;Stel, 2013). In this paper we study this impact by exploring how entrepreneurs obtain access to electricity in Lebanon. Our analysis illustrates that Lebanon's hybrid political order imposes a high cost on electricity.
We focus on access to and cost of electricity because it offers a potentially useful angle from which to obtain a better understanding of how a hybrid political order shapes entrepreneurial behaviour. Problems with electricity supply and costs are often cited as the most serious obstacles to doing business faced by entrepreneurs across the world (Alby, Dethier, & Straub, 2013). Expensive and unreliable electricity delays production, raises prices, impacts negatively on firms' reputations, and makes the provision of certain goods and services uncompetitive. How such constraints affect entrepreneurs' behaviour will depend on the nature of the political order they operate in. In an open access order, for instance, entrepreneurs may use elections and political watchdog organisations to (indirectly) pressure elected politicians to invest public resources in electricity provision and ensure fair and reliable access. In limited access orders, by contrast, it is firms' direct access to the ruling elite and its patronage system that may determine the quantity and quality of the electricity they can obtain, from public electricity as well as from private generator electricity.
In hybrid political orders, entrepreneurs who are excluded may be permanently shackled by high electricity costs, resulting in their firms' sub-optimal performance and skewing their investment decisions. But since entrepreneurs are almost by definition individuals who overcome obstacles to create and manage firms (Leibenstein, 1968), we should expect entrepreneurs in hybrid political orders to employ various strategies to overcome their lack of access to electricity. In Lebanon, they invest in an electricity generator and resort to using political connections (Faccio, 2006). However, as we show in this paper, this means that entrepreneurs in Lebanon often have to pay for electricity twice, namely for public electricity and for installing a generator (Alby et al., 2013;Feng, Johansson, & Zhang, 2013;Reinikka & Svensson, 2002). Moreover, both public and generator prices are unreasonably high due to the conversion of interests between the politicians leading the ministry and the businessmen operating the generators. This, then, is the main obstacle that electricity presents to Lebanese entrepreneurs. It is a result of the oligopoly that follows from the hybrid nature of the political order. In this regard, our paper can be seen as part of a broader literature on state-business relations in developing countries which focuses on the phenomenon of government capture by businesses, and/or business capture by politicians; see, for example, Abega (2013) and Sen and Te Velde (2009).
Lebanon is an interesting case to test some of the above speculations. It is a good example of a hybrid political order (see Section 4.1.). It is also renowned for its entrepreneurial acumen (Ahmed & Julian, 2012), and high utility (electricity) costs constitute a major constraint facing entrepreneurs (Dagher & Yacoubian., 2012;Kanaan, 2011;Stel, 2013;World Bank, 2008). After political instability, electricity costs are the most serious self-reported obstacle by Lebanese entrepreneurs in both the 2009 and 2013 World Bank Enterprise Surveys.
The remainder of the paper is structured as follows. In Section 2 we provide an overview of the relevant literature. In Section 3, we describe our research methodology. The Lebanese case context is introduced in Section 4. Regression results using World Bank data and interpreted through the lenses of our own qualitative survey conducted in 2012 are presented in Section 5 and further analysed and discussed in Section 6. Section 7 concludes. Our key findings are, firstly, that a hybrid political order imposes a 'tax' on entrepreneurship through channelling entrepreneurial talent into lobbying and bribery, by reinforcing family-owned dominance in business, and by distorting investment decisions. Secondly, hybrid political order does not primarily affect entrepreneurs' access to electricity but the price they pay for it (both the public and the private/generator electricity). As such, hybridity does not necessarily affect decision-making on the level of the individual entrepreneur or firm, but shapes decision-making on the governance level of both the public sector (the ministry) and the private sector (the generators) that become oligopolistic.

Literature Review
There is a substantial literature on the determinants of firm performance; see, for instance, Coad, Segarra and Teruel (2013), Coad and Tamvada (2012), Dollar, Hallward-Driemeier, and Mengistae (2005), and Sørensen and Chang (2006), and a growing literature on entrepreneurship in developing countries (Gollin, 2008;Naudé, 2010aNaudé, , 2013Quatraro & Vivarelli, 2013). However, in this paper our interest is not on firm performance per se, but on how firm performance is affected by a hybrid political order, and how entrepreneurs interact with such an order. As such, it suffices to mention that there are still relatively few studies of entrepreneurship in developing countries and countries in conflict, and particularly in countries characterised by politically hybrid governance structures such as Lebanon (Brück, Naudé, & Verwimp, 2013;Guglielmetti, 2010;Stel, 2013).

Hybrid Political Orders
A political order is the sum of institutionalised power relations that one can empirically grasp at a given time and place (Hagmann & Hoehne, 2009, p. 44). In a hybrid political order, political and economic power is divided along 'diverse and competing authority structures, sets of rules, logics of order, and claims to power [that] co-exist, overlap, and intertwine, combining elements of introduced Western models of governance and elements stemming from local indigenous traditions of governance' (Boege, Brown, & Clements, 2009, p. 17). Political hybridity has two core manifestations. The first is instability, often in the form of vulnerability to conflict or natural disaster (Naudé, Santos-Paulino, & McGillivray, 2011). Instability, however, also becomes apparent in the fluctuating power of political elites and the resultant deadlock in decision-making and policy implementation (Hasbani, 2011). The second manifestation of hybridity is informality. Informality, in the context of access to electricity, ties in with two other concepts: namely the dual game analogy and the notion of the open/ limited access order. The dual game logic (Cammett & Issar, 2010, p. 383) combines a relatively cosmetic 'electoral game', in which parties and groups aim to gain votes, with a more fundamental 'regime game' that often prevents a merit-based bureaucracy from emerging. Instead, it generates a public system that has the clientelist distribution of jobs as its core function (Briscoe, 2009, p. 16). The idea of the limited/open access order points at the degree to which access to decision-making by public institutions is either equal or discriminatory. In limited access orders, 'political elites divide up control of the economy, each getting some share of the rents' (North, Wallis, Webb, & Weingast, 2007, p. ii).
In essence, the concepts of instability, informality, dual game, and limited access order denote issues revolving around the institutional overlap between formal state institutions and informal authorities. This insight lies at the heart of the hybrid political order concept and explains an oligopolistic organisation of society, politics, and business. In such a setting, connections, networks, and relations are disproportionally important economic resources; they often are a prerequisite for establishing a business. Such connections are utilised through lobbying, bribery, and establishing supply ties to government. This has implications for the cost of utilities such as electricity.

Provision of Utilities
The conceptual relation between political hybridity and the provision of utilities follows from the notion of oligarchy. In hybrid political orders, access to and costs of electricity do not result predominantly from formal relations but rather from informal relations with elite powerbrokers, who have political as well as economic clout (Bardhan, 2004;Berry, Forder, Sultan & Moreno-Torres, 2004;Foster & Steinbuks, 2009;OECD, 2008;Vaux & Visman, 2005). Services such as affordable and equitable electricity are then, to a large extent, not a public right, but a political favour. The exact expression of high electricity costs as a result of the oligarchic governance of electricity sectors in entrepreneurs' behaviour, however, remains under-researched (Aggestam, Guazzone, Lindholm Schulz, Paciello, & Pioppi, 2012). Alby et al. (2013), for instance, using the same World Bank Enterprise Survey (WBES) database that we do, analysed a sample of 46,606 firms across 87 countries over the period 2002-2006. They were particularly interested in the degree to which electricity supply constraints affect enterprise behaviour. They found that 15 per cent of all entrepreneurs interviewed indicated that electricity supply is a major or very severe constraint to doing business. They also found that 31 per cent of all firms surveyed across the sample owned or shared an electricity generator. In 2013, however, no less that 55 per cent of all entrepreneurs interviewed in Lebanon indicated that electricity supply is a major or very severe constraint to doing business, and 92 per cent owned or shared an electricity generator.
What the case of Lebanon illustrates, then, is that owning a generator (which is not so much an option but a basic requirement for doing business) does not mitigate electricity problems, but rather is a manifestation of electricity problems, as it leads to higher (indeed 'double') electricity costs.

Methodology
To investigate how hybridity affects the costs of electricity and how this determines entrepreneurs' behaviour, we use quantitative data as well as information obtained from our own qualitative survey of decision-makers in Lebanon.
Our quantitative data comes from the 2009 and 2013 WBES of, respectively, 382 and 561 establishments in Lebanon. Although the 2013 survey contains useful information on the extent of electricity problems, it was more focused on firms' innovation behaviour and did not contain variables relating to family ownership, the networking behaviour of the firm, or the proportion of sales to government, which are key variables in our conceptual understanding of the dynamics of governance and electricity costs. We use, therefore, the 2009 WBES data in our OLS regression analyses of sales growth and employment growth as measures of entrepreneurial performance. As explanatory variables in these regressions, we include perceptions of political instability and electrical problems, the number of electrical outages, and the share of electricity generated through an owned generator, as well as a number of standard control variables (see Table 1).
We augment these regression analyses by the results from qualitative interviews with Lebanese entrepreneurship experts, conducted in Lebanon in the summer of 2012. The interviews covered 32 local experts on entrepreneurship and business. Respondents were selected based on preliminary stakeholder mapping and subsequent snowball sampling. Considering the lack of previous work on political hybridity and entrepreneurship, and hence the necessarily explorative nature of our current study, and bearing in mind the multi-faceted links between business and politics in Lebanon, at this stage we accessed a broad and diverse group of experts (from civil society, academia, journalism, the private sector, government, and the international donor community) rather than merely businessmen and entrepreneurs. 2 Before sharing our findings, it is necessary to introduce Lebanon's hybrid political order, the nature of Lebanese entrepreneurship, and the obstacles, including electricity, that it faces.

Lebanon's Hybrid Political Order
Lebanese society is organised along 18 religious communities that each have their regional strongholds, political parties, social institutions, and armed militias. The central concept to understand Lebanese society, then, is sectarianism, which signifies this division of society into religious, 'sectarian', communities (Faour, 2007;Haddad, 2002). Sectarianism corresponds with the polarisation of social control between Lebanon's various communities. This fractionalisation breeds a structural elitism: because society is organised along sectarian lines, citizens have historically depended on sectarian leaders for protection and provision (Ziadeh, 2006).
The Lebanese state is organised through a consociational political system centred on inter-sectarian power sharing that includes corresponding sectarian quota guiding the allocation of all public positions so that 'the two sides of the redistributive mechanism, economic and political, are permanently intertwined' (Nahas, 2012, p. 132). The resultant quest for inter-communitarian balance generates endemic patronage and clientelism, and a highly oligopolistic market structure (Cammett & Issar, 2010;Hamzeh, 2001;Nahas, 2012). 3

Lebanese Entrepreneurship
The World Economic Forum (WEF) (2011, p. 9) estimates that 15 per cent of the Lebanese adult population is engaged in entrepreneurial activity, but notes that this is mostly necessity entrepreneurship that revolves around self-employment. The vast majority (96%) of Lebanese firms surveyed by WBES are in private, domestic ownership. None of the firms had government ownership in the official sense, but the nature of Lebanon's hybrid political order stipulates that even if these firms do not have formal government ownership they are subject to omnipresent political influence. Nahas (2012, p. 152) shows that the management of Lebanon's economic system, characterised by 'quasi-natural monopolies', is 'fundamentally based on an interweaving of the public and private spheres'.

Lebanese Electricity
The state-owned Électricité du Liban (EDL) is responsible for electricity production, transmission, and distribution in Lebanon. It provides around 77 per cent of the country's electricity demand, with private generators providing the remainder (Hasbani, 2011). Power outages are typical in Lebanon, and it is not rare to have only four hours of electricity a day (Dagher & Ruble, 2011). According to the most recent (2013) WBES survey conducted in Lebanon, there were 50.5 electrical outages in a typical month, each lasting an average of 5.2 hours. This compares to 17.6 per month for the Middle East and North Africa, lasting on average 6.5 hours. Hence the frequency of power outages is a disproportionate problem in Lebanon. In line with this, the WBES (2013) reflects that Lebanese businesses seem better prepared to operate under these conditions, considering that many more business firms own or share an electricity generator than in other Middle Eastern countries: 92 per cent of firms in Lebanon owned an electricity generator in 2013 compared to an average of 41 per cent in the Middle East and North Africa region.
The wide ownership of electricity generators underlines our argument that it is not so much access to but rather costs of electricity that constitute the main obstacle for entrepreneurs in Lebanon's hybrid political order. Mehzer, El-Saouda, Nasrallah, and Al-Ajam (2008, p. 44) show that, in Lebanon, 'utility costs are extremely high and considered as indirect taxes'. A microfinance expert we interviewed explained that electricity failures not merely required his company to sign up to a generator in addition to the public electricity network, but also to buy batteries of USD 4000 apiece for each of his company's offices because the generator operator he is connected with only works at night. 4 Other experts from the private, public, and civil sectors confirmed this reading of electricity costs as a major obstacle to entrepreneurship. Without us explicitly asking for electricity-related issues, interviewees regularly brought this up as a core impediment for Lebanese entrepreneurs and provided various examples of businesses that had failed due to unbearable electricity costs. 5 A governmental strategy coordinator, for example, lamented that 'Electricity is available, but so expensive it undermines Lebanon's competition power. It is not viable for, for instance, an information and communication technology data centre to be set up in Lebanon due to the electricity costs and cuts. Generators are there, but they are too expensive'. 6 That electricity is a significant business constraint in Lebanon is confirmed in the database we analyse below.

Constraints on Lebanese Entrepreneurs
According to the Lebanese firms surveyed, their greatest single obstacle to doing business is political instability, 7 a feature established as a key symptom of hybrid political orders in Section 2.1. Other major obstacles are access to and reliability of (public) electricity as well as corruption and finance. Figure 1 lists the responses of the firms regarding the relative importance of a number of possible obstacles to doing business and compares this to the situation in other Middle Eastern and North African countries (as well as the most recent averages from the WBES overall survey). Electricity comes up as the second-most noted constraint on doing business, after political instability. When asked how much electricity costs would need to decrease for it not an obstacle anymore, the average response in the 2013 WBES was 18 per cent.
Apart from an emphasis on political instability on the one hand and electricity on the other, almost 7 per cent of firms indicated corruption as the worst business obstacle (up from 5 per cent in 2009). This was especially the case for larger firms, which may be because larger firms tend to be more interested in tendering for government contracts. In Lebanon, 97 per cent of all firms indicated in 2009 that they would be expected to pay a bribe for getting a government contract (compared to 47 per cent on average in other Middle Eastern countries). The average value of a bribe was reported to be around 8 per cent of the value of the contract, which makes Lebanese government contracts almost twice as 'expensive' as government contracts on average in the Middle East (see Leenders (2012) for an elaborate dissection of Lebanon's governmental contracting system). Furthermore, in 2013, firms that indicated that they pay no bribe to government reported, on average, a financial loss due to electrical outages of 17 per cent of annual turnover. Conversely, firms that indicated they pay between 10 and 20 per cent of annual turnover as bribes indicated average losses due to electricity outages of only 7 per cent of annual turnover. 8 Thus, with regard to our interest in the entrepreneurship-hybridity-electricity interplay, the issue is not so much whether entrepreneurs have to pay individual bribes. The issue at stake, rather, is the organisation on the governance level of the electricity sector that is corrupt, resulting in higher electricity prices because bribes for contracts have to be earned back, and oligopolistic, meaning contracts often come with political protection and prices will not be lowered through competition.

Regression Results
We are interested, first, in how a hybrid political order affects firm performance, as measured by sales and employment growth. Second, we look into how entrepreneurs respond to electricity constraints generated by hybrid political order, in particular whether firms mitigate electricity access and costs through relations with political elites. We expect that firm performance is negatively affected by unreliable electricity provision, as it makes entrepreneurs dependent on generators that double their firms' electricity expenditure, but that entrepreneurs who are able to access political elites may alleviate these crippling costs by negotiating better deals. Because the establishment of substitutes such as generators and the lobbying of government officials take time and effort, we hypothesise that, although these measures will help firms overcome electricity constraints and alleviate the consequences of political instability, they will still impact negatively on firm performance, in particular on sales and employment growth.
Variable abbreviations and descriptions are contained in Table 1. Data on these variables were obtained, as mentioned, from the WBES of 2009 (see WBES, 2009). The results of the regression analyses are contained in Tables 2 and 3, each reporting, respectively, on the regressions with sales growth and employment growth as dependent variables. Table 2 contains the regression results on the determinants of sales growth of firms over the period 2008-2009, expressed in local currency units. Table 3 reports on the determinants of employment growth, expressed in number of permanent employees. Four models have been estimated in each case, the results reported in columns 2-5 of the two tables.
Model 1 is a simple regression of the measure of performance on the basic variables of interest, namely the degree to which electricity problems and political instability have been reported as a major and/or serious constraint, and the extent to which a firm is able to generate its own electricity in the case of power outages (variable = out*gen). We expect that the more a firm is able to do so the better its performance in the face of electricity outages will be. And, indeed, the interaction variable out*gen is highly correlated with the number of outages experienced (correlation coefficient 0.64) and the percentage of own electricity generated (0.75). In Table 2 (sales growth) all coefficients in Model 1 are of the expected sign and statistically significant. Hence we can confirm that, ceteris paribus, firms who reported political instability and electricity problems as obstacles also experienced significantly lower sales growth during the year. However, Table 3 shows that only those firms who reported political instability as a major or serious obstacle had a decrease in employment, and that electricity supply problems do not seem to impact on employment growth. Furthermore, from Table 2 it seems Public-private entanglement 261 that firms who could use their own or a shared power generator in response to outages had higher sales growth.
In Model 2 we added as explanatory variable the amount that a firm spent during the year on bribes ('informal payments') to government officials to get business done. Once we control for the ability of firms to engage in such payments, the extent to which firms report political instability and electricity problems ceases to have a statistically significant impact on sales growth (Table 2). Also, where political instability was reported as a problem, it ceases to impact significantly on employment growth (Table 3). As our qualitative survey found that the main issue with electricity is its cost rather than access, we may posit that entrepreneurs bribe not so much to obtain an Notes: T-values are in brackets. Significance at the *10 per cent level, **5 per cent level, and ***1 per cent level. electricity connection, but rather to lower the price they pay for the electricity they consume. Nevertheless, paying bribes is associated with significantly lower sales growth, indicating that, while it may lower the costs of electricity, it reflects the broader limited access order that depresses firm performance. In Model 3 we added the 'standard' control variables that may influence firm performance. From Table 2 it can be seen that once we control for firm size, age, sector, innovation, and for the skills of the entrepreneur and labour force, only the extent to which a firm's sales were to government had a significant impact on sales growth during the year. This may mean that having access to government contracts and providing government with goods and services is a clear advantage for firm performance in Lebanon, even when we control for all the standard determinants of firm performance. Our qualitative data (see below) suggests this is because access to government contracts generates better 'deals', for instance on electricity prices. This is backed up by further quantitative evidence from the 2013 WBES survey, which asked entrepreneurs the extent of financial losses (as percentage of sales) of electricity outages. For instance, the average firm in Lebanon in 2013 lost an equivalent of 13 per cent of sales due to electrical outages. One firm even reported losing 80 per cent.
A difference between Models 3 and 4 is that in the latter we included a dummy variable for whether the firm is a family firm or not. Family ownership of firms is ubiquitous in the Middle East and Lebanon, and a growing literature is emerging on the role and impact of family firms and their internal and external networks (Fahed-Sreih, Pistrui, Huang, & Welsh, 2010;Welsh & Raven, 2006). Through such networks, family firms may have some advantages in terms of dealing with the governance problems of the hybrid political order in Lebanon. The results in Model 4 seem to bear this out: once we control for a firm being a family firm, the extent of paying bribes to government officials are no longer associated with negative sales growth (the coefficient on bribes becomes insignificant). Although this is not statistically significant, in combination with the literature and our qualitative survey, we suggest that, somehow, the family networks of firms in Lebanon seem to help overcome the institutional gaps that require bribes and result in poor performance for non-family firms.
Our conclusion from the results in Tables 2 and 3 is that firms who engage in bribery to ensure government contracts seem to alleviate high electricity costs. This may be, as we will explore below, because working with and for the government generates access to political elites.

Discussion and Interpretation
Based on the above, it seems that hybrid political order in Lebanon has resulted in an environment in which poor public provision of a utility like electricity has made place for private initiatives such as generators. One interpretation is that due to the fragmented governance structure in the country and the traditional potency of entrepreneurship in Lebanon, official central control over these private solutions to electricity could not be established.
From our interviews with local entrepreneurship experts, however, we conclude that the distinction between public and private is misleading here due to the oligopolistic nature of the sector and the clientelist handing out of contracts (where even privatisation does not lead to competition and lower prices). This goes for EDL electricity as well as the generators that are private, but not competitive. Each neighbourhood is allocated to one generator provider that has the backing from politicians and thus does not face competition and can set the prices it wants; at least for smaller firms (Mohsen, 2012). So, while there is no formal central control, neither are private alternatives free from informal, local political control. Hence, the issue with the relation between electricity provision and political hybridity is not so much about bribery, or about the individual firm-state connection, but about the more structural oligopolistic organisation of the electricity sector that results in dual and high prices.
The experts we interviewed consider the clientelist and oligopolistic nature of the utility sector as the fundamental reason for the poor state of electricity supply. The sectarian quota system effectively working as an official mechanism to 'divide the pie' of state jobs among sectarian communities through a clientelist award system was often mentioned as the core problem. Observers noted that 'politicians put their people in the administration and now even the public companies are divided'. 9 A civil servant explained that 'there is a paradox in that the economy is business-driven, while key facilitating infrastructural sectors are government owned'. 10 The World Bank (2008, p. 4) notes that electricity subsidies have reached 39 per cent of total government spending between 1997 and 2006. Yet the overwhelming perception seems to be that while the government might be losing money on the electricity file, politicians, as private economic actors, are carefully scooping up the same money through their affiliated companies and contacts. 11 Our interviewees stressed this entanglement of the public and private sectors as a core aspect of the hybrid political order that can help understand the functioning of Lebanon's utility sectors. A leading analyst concluded that while the private sector claims it wants to remain out of politics, and therefore refuses to lobby for better governance regulation, this is in fact 'preposterous', as they are all 'entangled'. 12 Such observations seem more than justified in light of existing research into the workings of patronage networks in the public and private sectors in Lebanon, in which several major families dominate business and politics (Leenders, 2004(Leenders, , 2012. This is largely due to Lebanon's vague public contracting system. The Lebanese Transparency Association (LTA) (2011, p. 85) recognises that 'very limited information exists around contracting and there is no information available about the size of the procurement market'.
In practice, public contracting and procurement is subject to patronage and clientelism, 'whereby all political leaders take advantage of their position to promote the interests of their own communities' (LTA, 2011, p. 87). Many of the experts we consulted seem to agree with Kaplan (2009, p. 7), that, in Lebanon, 'successful entrepreneurs are not those with the best ideas, but those with the best ties to ruling elites'. Spokespersons of a civil society watchdog declared that, 'if you know how to bribe well, you can do well' and signal 'immense conflicts of interest between politics and business'. 13 Lebanese political analysts confirm that the political class is 'embedded in the private sector', as 'most businessmen are former state officials and most politicians have some businesses'. 14 This is consistent with the literature on Lebanon's political economy, which Nahas (2012, p. 126) describes as being characterised by 'the interpenetration of public and private concerns' and an 'endemic inability to produce efficient government structures'.
While our interviews contained few direct references to instances in which entrepreneurs lowered their electricity bills through contacts with politicians, the overall context emerging from the interviews seems very conducive to such practices. A representative of the LTA, for instance, insisted that Lebanon was, in some respects, a 'mafia state; there are immense conflicts of interest between politics and business. All politicians are businessmen and only very few leave business when they enter politics. Everyone knows everyone'. 15 An economist affiliated with the American University of Beirut assured us that 'There is a lot of vested power in the status quo and lots of deal-making through alliances to satisfy either financial or ideological private interests. [. . .] Businessmen play off different parties against each other; if the one wants 10 per cent rents, you go with the one that demands only 8 per cent and let him deal with the other'. 16 Based on our qualitative interviews, we thus propose to see the quantitative regression results in light of the overarching 'public-private entanglement' structurally referred to by our interviewees. Entrepreneurs' responses to electricity constraints illustrate the significance of the blurred lines between the public and private realms -'business and politics' (Feng et al., 2013, p. 1)poignantly. Firstly, while bribing is a strategy to alleviate electricity costs, it not so much concerns direct bribing on the consumption level of the electricity sector, but rather bribing for contracts, an informal form of bridgingor distortingthe public and the private. This, secondly, makes supplying to government an unsurpassed strategy to mitigate high electricity costs for entrepreneurs (see also Hasbani, 2011;Sulahian, 2004;Verdeil, 2009), which further cements the impression that it is private actors' inlays into public sector institutions that, to a large extent, determines their ability to deal with rising electricity costs that follow from the country's hybrid political order. The utility of family connections for entrepreneurs to deal with electricity challengesa UN-based economist mused that 'the political class is very friendly to the private sector; or, better put, the family sector' 17thirdly, might further suggest the importance of the overlap between private business networks and public decision-making networks in understanding the relations between entrepreneurship and political order.
The oligopolistic governance of the Lebanese electricity sector that generates serious problems for Lebanese entrepreneurs is widely recognised to stem from the country's hybrid political order. Entrepreneurs' responses to the electricity problems presented to them, however, equally reflect that Lebanon's hybrid political order and entrepreneurs' bribing, supplying to government, and utilisation of family connections confirm and even reinforce the oligopolistic logic of the hybrid political order. It thus seems that, ultimately, Lebanese entrepreneurs do not so much challenge the hybridity that is one of their key constraints, but seek to utilise it to navigate these same constraints.

Conclusions
We explored how entrepreneurial decisions and firm performance are affected by the institutional context of a hybrid political order by studying how entrepreneurs deal with disproportionate electricity costs in Lebanon. In Lebanon, economic control is divided following an inter-sectarian power-sharing formula that results in endemic clientelism and a highly oligopolistic market structure. As such, entrepreneurial success may be a function of ties to the elite and the impact of high electricity costs, a major business obstacle in most developing countries, may affect entrepreneurs differently depending on the extent of these ties.
Hence, we expected that unreliable electricity provision would necessitate Lebanese entrepreneurs to invest in their own generators and, subsequently, in ties with political elites in order to mitigate resultant dual electricity costs. We hypothesised that entrepreneurs with access to both a generator and political networks might experience electricity supply, and also the political instability generated by the hybrid political order, to be less serious business obstacles. Firm-level data from the 2009 and 2013 WBES of Lebanon augmented with our own qualitative interviews suggested that this is indeed the case.
We found that entrepreneurs who engage in bribery (to access government contracts) alleviate electricity problems and tend to perform better. These results reflect the fact that a hybrid political order imposes a 'tax' on entrepreneurship through channelling entrepreneurial talent into bribery and lobbying, through reinforcing family-owned dominance in business, and by distorting investment decisions, for instance into using access to finance to invest in a generator and bribe both government officials and generator operators to negotiate better prices.
Our results are consistent with the entrepreneurship literature, wherein a lack of strong institutions is seen as leading to sub-optimal entrepreneurial performance and even destructive forms of entrepreneurship (Baumol, 1990). They are also consistent with the findings of, for instance, Faccio (2006, p. 369) that political 'connections are particularly common in countries that are perceived as being highly corrupt'. Our insights resonate with the empirical findings of Feng et al. (2013), who study the ways in which entrepreneurs in China use political connections to improve their firms' performance. Feng et al. find that access to 'regulated', that is state-owned, enterprises and government contracts is an important feature of the 'mixing of business and politics' that characterises entrepreneurship in China. Our results, coming from a very different type of political order, suggest similar dynamics, which generates possibilities for comparative research into the entrepreneurship-political order connection.
Ultimately, our conclusions confirm Malik and Awadallah's (2013, p. 297) assessment that private sector development in the Middle East is 'not simply a matter of improving business climate [. . .] it is also a political problem'. Our results show that the political order matters for entrepreneurship and that institutional reform geared towards realising broader and more equal access to affordable public electricity could make a positive contribution to entrepreneurial performance. At the same time, our case serves as a reminder of Nahas' (2012, p. 139) warning that 'the insidious relation of overlapping and differentiation between the public and the private spheres gives a specific content to the debate about reforms, privatization and "downsizing the state"'. Our conclusions indeed highlight 'the increasing difficulty in distinguishing between the public and the private spheres and the consequent Public-private entanglement 265 fictitious nature of reforms aimed at the privatization of goods and services which were not truly public to begin with', as noted by Aggestam et al. (2012, pp. 326-327).