THE RESOURCE CURSE

The academic literature on the resource curse is vast (Ross Reference Ross2015). Here, we examine three different dimensions of the resource curse which comprise a ‘political resource curse’ (Vicente Reference Vicente2010). First, oil-rich countries are unable to establish or consolidate democracy (Ross Reference Ross2001, Reference Ross2015; Jensen & Wantchekon Reference Jensen and Wantchekon2004; Ramsay Reference Ramsay2011; challenged by Haber & Menaldo Reference Haber and Menaldo2011; Brooks & Kurtz Reference Brooks and Kurtz2016). Second, oil wealth has been found to increase the risk of inter-state war (Colgan Reference Colgan2013) and civil war (Collier & Hoeffler Reference Collier and Hoeffler1998; challenged by Brunnschweiler & Bulte Reference Brunnschweiler and Bulte2008; Paine Reference Paine2016) as well as lengthen the duration of civil conflicts (Lujala Reference Lujala2010). Finally, resource wealth is believed to have corrosive effects on the quality of political institutions and leads to increased corruption and rent-seeking (Leite & Weidmann Reference Leite and Weidmann1999; Sala-i-Martin & Subramanian Reference Sala-i-Martin and Subramanian2003; Shaxson Reference Shaxson2007; Vicente Reference Vicente2010; challenged by Alexeev & Conrad Reference Alexeev and Conrad2009).

The so-called institutionalist turn in the resource curse literature (Mehlum et al. Reference Mehlum, Moene and Torvik2006; Smith Reference Smith2007; Morrison Reference Morrison2013) arguing that natural resource wealth has conditional effects depending upon a country's quality of institutions has undermined the view that the resource curse is inevitable. Increasingly, scholars believe, in Morrison's (Reference Morrison2013: 1122) phrasing, that ‘oil is less like a universal acid and more like a powerful tool, which can have good or bad effects depending on the circumstances’.

The resource curse literature is contested in all these areas. This article does not make a broader statement on the overall validity (or lack thereof) of the resource curse. Instead, it merely seeks to elucidate how the political resource curse might play out in Somaliland and why Somaliland is vulnerable to it.

Oil wealth is posited to impact democracy negatively through several channels but is most commonly attributed to a rentier logic where the ‘key mechanism linking authoritarian rule and resource dependence … is an incumbent's discretion over the distribution of natural resource rents’ (Jensen & Wantchekon Reference Jensen and Wantchekon2004: 821). As explained by Ross (Reference Ross2015: 246) ‘An abundant flow of oil revenues enables incumbents to both reduce taxes and increase patronage and public goods, making it possible for them to buy off a larger set of potential challengers and reduce dissent’. In sub-Saharan Africa, Jensen & Wantchekon (Reference Jensen and Wantchekon2004: 817) find that ‘executive discretion over resource rents leads to less political liberalization (transition of democracy) and a greater likelihood of democratic breakdown (consolidation of democracy)’. Importantly, for Somaliland, research suggests that ‘oil does greater damage to democracy in poor states than in rich ones’ (Ross Reference Ross2001: 356).

The argument that oil-rich countries are at greater risk of conflict follows two different sets of causal arguments – one which focuses on states and the other which focuses on insurgents (Lujala Reference Lujala2010; Ross Reference Ross2015; Paine Reference Paine2016). The state argument suggests that ‘because resource-rich rulers do not have to build strong societal ties to raise revenues, oil-rich governments tend to have weak bureaucratic institutions relative to their country's per capita income’ (Paine Reference Paine2016: 748). Weak state institutions lead to elevated levels of citizen dissatisfaction, simultaneously increasing the demand for change and decreasing the state's ability to neutralise it. The causal logic focusing on rebels essentially advances the two related arguments that natural resource rents ‘provide motivation and means for rebel uprisings’ (Lujala Reference Lujala2010: 15). In none of these arguments is oil the single explanation for violence. Rather, as Glaser (Reference Glaser2013: 125) emphasises, it ‘is one among a number of links in a chain that could lead to conflict’.

There are two main findings from this literature which suggest danger for Somaliland. First, although onshore and offshore oil production are both capable of generating significant revenues, only onshore oil production appears to affect the prospects for conflict. As explained by Ross (Reference Ross2015: 251), ‘When oil is found offshore, it has no robust effect on a country's conflict risk; when it is onshore, it appears to have a large effect’. Additionally, Lujala (Reference Lujala2010: 16) finds that the presence of ‘hydrocarbons in the conflict zone more than doubles the conflict duration. Resources located outside the conflict region do not have a prolonging effect on the duration.’ While Somaliland may have offshore oil reserves, the Nugaal oil basin in eastern Somaliland stretches across territory that is also claimed by Puntland and Khaatumo State. Oil increases the value of territory which ‘increases the probability that crises over territory will lead to war instead of negotiated compromises, as states are more willing to run the risks of fighting’ (Glaser Reference Glaser2013: 123). Most ominously for unrecognised Somaliland, ‘oil is especially likely to generate conflict when territorial boundaries are contested’ (Glaser Reference Glaser2013: 123). Such conflicts are also more difficult to resolve because they are not just monetary disputes but rather are sovereignty disputes. Revenues can be split. Yet, as Voller (Reference Voller2013: 79) observes, if ‘control over oil is associated with sovereignty … It may render any peaceful solution to the conflict impossible’.

Second, onshore oil can have these effects even if it is not actually being produced. As Colgan (Reference Colgan2013: 154) specifies, it ‘is perceived oil reserves that matter for resource wars, not actual oil reserves … resource wars can occur even when no oil exists’ (see also Lujala Reference Lujala2010: 23). Thus, Somaliland might not have commercially viable oil reserves, but it could still suffer increased risks of fighting in its eastern regions.

Early work on the resource curse found that capital-intensive or ‘point source’ natural resources like oil and minerals led to significant increases in corruption while food and agriculture did not (Leite & Weidmann Reference Leite and Weidmann1999; Sala-i-Martin & Subramanian Reference Sala-i-Martin and Subramanian2003). Later work has narrowed this finding to petroleum and excluded other mineral resources (Ross Reference Ross2015: 248). As with democracy and conflict, there is debate over the causal logics that could explain this. In one influential formulation, ‘The pernicious access to easy money’ lowers financial discipline, leads to reckless budgetary practices, ‘preempts efforts to mobilize domestic resources through taxation, reduces tolerance for austerity and produces a dangerous reliance upon the state for the resolution of all problems’ (Karl Reference Karl1999: 35). Other posited causal logics include politicians deliberately weakening institutions to increase their ability to misappropriate resource rents and revenues coming in faster than institutional capacity to manage them can be created (Ross Reference Ross2015: 249).

The institutionalist turn in the resource curse literature has established that ‘the effects of natural resources are largely determined by the institutional environment in which they are found’ (Morrison Reference Morrison2013: 1118). Put simply, one should expect different effects from an oil discovery in Norway than from an oil discovery in Chad. Smith (Reference Smith2007: 122) argues that resource booms do not occur in a vacuum but, rather, are incorporated into ‘pre-established institutions and patterns of decisionmaking’ with the result that ‘the windfall revenues merely magnified existing patterns’.

Oil wealth is mediated significantly by the institutional context it flows into but, in Mehlum et al.’s phrasing (Reference Mehlum, Moene and Torvik2006: 3), ‘natural resources put the institutional arrangements to the test’. As is the case with the risk of conflict being increased even if oil is not actually produced, the same point holds true for oil putting institutions to the test. Interesting comparative work has documented increased corruption in São Tomé and Príncipe even though oil has yet to be produced there (Vicente Reference Vicente2010; Frynas et al. Reference Frynas, Wood and Hinks2017). As discussed below, the nascent institutional arrangements in Somaliland are unlikely to pass this test.

OIL TO CASH

Oil to cash is an idea that has been more warmly embraced by scholars than by politicians. Scholars have advocated the idea generally (Moss et al. Reference Moss, Lambert and Majerowicz2015), for sub-Saharan Africa (Shaxson Reference Shaxson2007; Diamond & Mosbacher Reference Diamond and Mosbacher2013), and for Iraq (Birdsall & Subramanian Reference Birdsall and Subramanian2004) and Nigeria (Sala-i-Martin & Subramanian Reference Sala-i-Martin and Subramanian2003). It has yet to be implemented, although Alaska's oil dividend programme contains elements of the idea and other conditional cash transfer programmes such as Brazil's Bolsa Familia programme and Mexico's Oportunidades programme have operated for decades. There has, to date, been almost no discussion of the idea in Somaliland. It did not, for example, feature at all in the 2017 presidential election campaign.

Scholarly enthusiasm for oil to cash largely arises out of pessimism about the prospects for other proposed solutions to the resource curse. Perhaps the most widely promoted solution is greater transparency through the Extractive Industries Transparency Initiative (EITI). Common critiques of the EITI include that compliance is entirely voluntary (Hilson & Maconachie Reference Hilson and Maconachie2009: 91; Kolstad & Wiig Reference Kolstad and Wiig2009: 528; Gillies Reference Gillies2010: 119), that oil exploration and production involves a complex chain of events while the EITI focuses on just one step of the process – the payment of revenues by companies to host country governments (Kolstad & Wiig Reference Kolstad and Wiig2009: 527–8; Gillies Reference Gillies2010: 120), and that providing transparency on revenue payments is of limited utility if citizens cannot hold their governments accountable (Shaxson Reference Shaxson2007: 1134; Hilson & Maconachie Reference Hilson and Maconachie2009: 70; Kolstad & Wiig Reference Kolstad and Wiig2009: 529). The EITI is also probably not an option for unrecognised Somaliland. The EITI currently comprises 51 ‘implementing countries’ among its members but all are widely recognised sovereign states. Somaliland's lack of recognition has, to date, deterred investments from western oil majors. None of the four much smaller companies exploring in Somaliland – Ansan, DNO, Genel and RAK Gas – are among the 59 current EITI ‘supporting companies’.

Another posited solution is the use of savings and stabilisation funds. Savings funds are designed to save a portion of the oil revenues for future generations. Stabilisation funds are designed to smooth the volatility of oil revenues as price and production levels fluctuate. Critics highlight the poor empirical track record of such funds in countries including Azerbaijan, Chad, Nigeria and Venezuela (Birdsall & Subramanian Reference Birdsall and Subramanian2004: 85; Moss et al. Reference Moss, Lambert and Majerowicz2015: 52). Another common critique of such funds is that their ability to function as intended presupposes the existence of strong institutions that can protect and safeguard them (Sala-i-Martin & Subramanian Reference Sala-i-Martin and Subramanian2003: 19). This is a subset of the larger critique against the institutionalist turn in the resource curse literature for, in effect, telling weak states to develop strong institutions or for asking Angola to become more like Canada (Morrison Reference Morrison2013). As Moss et al. (Reference Moss, Lambert and Majerowicz2015: 2–3) put it, ‘The advice is correct but hardly helpful’. Oil to cash is seen by its proponents as a more robust solution for addressing the resource curse.

Although the specific details can vary along numerous dimensions, oil to cash is essentially a three-step proposal. First, states need to create a dedicated fund to receive the revenues. Ideally, such a ring-fenced and transparent fund would receive all the revenues associated with oil production whether they came from signature bonuses, royalties, taxes or other sources and it would avoid the complexities seen in the revenue management system in Chad which made it difficult to understand and monitor (Pegg Reference Pegg2006: 11–12). A ‘future generations’ savings fund could complement this receiving fund but is optional. In contrast, ‘a medium-term receiving fund is a necessary component for Oil-to-Cash’ (Moss et al. Reference Moss, Lambert and Majerowicz2015: 58).

Second, distributions from the fund to citizens need to be made ‘in a regular, universal, and transparent payment based on a set of agreed-on fiscal rules’ (Moss et al. Reference Moss, Lambert and Majerowicz2015: 3). Distributions should be paid on a regular schedule with the regularity mattering more than the specificity of annual or quarterly payments. They should be calculated according to clear, transparent and simple rules, again avoiding the type of complexity seen in Chad's revenue management system. Distributions also need to be directly tied to oil revenues which can fluctuate with production volumes and prices. Thus, ‘If production or prices decline, the dividends should decline proportionally, so the fiscal purse is not left saddled with an unfunded cash transfer system’ (Moss et al. Reference Moss, Lambert and Majerowicz2015: 61–2). Obviously, a failure to communicate this feature clearly could result in a loss of citizen trust in the system when dividend payments decline.

Third, the state needs to collect taxes on the revenues distributed to its citizens. Taxation fundamentally distinguishes oil to cash from the lavish welfare spending seen in many oil-rich Arab countries. Such spending is an indirect distribution of resources controlled by government elites. Oil to cash is a direct distribution which sends the money to citizens who decide for themselves what to do with it. Taxation is essential for oil to cash because it goes directly at the problem of oil revenues appearing as non-tax revenues which governments do not have to raise from their own citizens. In this sense, it is an attempt to ensure that ‘governments will ultimately come to depend on tax revenues and be forced to build a tax administration, rather than bypass constituents by relying solely on rents’ (Moss et al. Reference Moss, Lambert and Majerowicz2015: 82). While distributing revenues directly to citizens and then taxing them might seem inefficient, the fundamental problem with Arab petroleum-fuelled welfare states is that when citizens are not asked to pay taxes, ‘the crucial bond of accountability never materialises’ (Diamond & Mosbacher Reference Diamond and Mosbacher2013: 95). Denying governments easy non-tax revenue and forcing them to raise taxes from their citizens is designed to ensure that ‘Relationships with government would tend towards ones of taxation and accountability, not of lobbying and corruption’ (Shaxson Reference Shaxson2007: 1135–6).

Ultimately, a functioning oil to cash programme only requires four things: an extensive public information campaign to explain how it works, a reliable means by which beneficiaries can be identified, a way to transfer funds to citizens electronically or via mobile payments and a way to tax back a portion of the revenue distributed to citizens (Moss et al. Reference Moss, Lambert and Majerowicz2015: 62).

ASSESSING SOMALILAND'S VULNERABILITY: QUALITY OF DEMOCRACY

One of the biggest impediments to implementing oil to cash is the opposition of political elites who do not want to lose control over resource rents. To implement an oil to cash programme in Somaliland one would thus need to make the case both that Somaliland is in danger of falling victim to the resource curse and that oil to cash is a viable solution to avoid that misfortune. This article assesses Somaliland's vulnerability to the resource curse by examining three different areas: the quality of its democracy, its hybrid mix of traditional and modern state institutions and its revenue management capacity.

Somaliland optimists would argue that any future oil revenues would arrive in an already consolidated democracy. The roots of Somaliland's democratisation date back to the guerrilla days of the Somali National Movement (SNM) when the SNM changed leadership five times (Duale Reference Duale2010, Int.). The significant impact of public opinion in Somaliland was demonstrated at the 1991 Burco Conference where the SNM leadership was not interested in discussing independence yet a groundswell of popular support for independence forced them to change course (Duale Reference Duale2010, Int.; Bradbury Reference Bradbury2008: 82–3; de Waal Reference de Waal2015: 131). The case for Somaliland as a consolidated democracy would also highlight the surprise death of President Egal in 2002, his replacement by a vice-president from a minority clan in accordance with the constitution, an extremely close presidential election in 2003 where the loser gracefully accepted defeat and a presidential election in 2010 where the incumbent lost, accepted defeat and peacefully transferred power to the opposition.

The case that Somaliland is a consolidated democracy would also draw from Posner & Young (Reference Posner and Young2007). They identify five main ways leaders can leave office, which they group into two broad categories: ‘regular means (which include natural death, voluntary resignation, or losing an election) and … irregular means (coup or violent overthrow or assassination)’ (Posner & Young Reference Posner and Young2007: 128). Somaliland's first four presidents all left office through regular means – Abdirahman Ahmed Ali ‘Tuur’ (1991–1993, losing an election among clan elders), Mohamed Ibrahim Egal (1993–2002, natural death), Dahir Riyale Kahin (2002–2010, losing a popular election) and Ahmed Mohamed Mohamoud ‘Silanyo’ (2010–2017, not seeking re-election). Posner & Young (Reference Posner and Young2007: 130) conclude that the regular turnover of leaders ‘marks the most important step toward restraining executive power and institutionalizing political authority more broadly’.

Yet, the quality of Somaliland's democracy also raises serious doubts about the country's ability to manage oil revenues wisely. Presidential elections have been repeatedly postponed with one originally scheduled for 2008 not held until 2010 and another originally scheduled for 2015 not held until 2017. Lower house of parliament members who ostensibly serve five-year terms have not been elected since 2005. Upper house or Guurti members supposedly serve six-year terms but have not been elected or selected since 1997. The Guurti’s repeated self-extensions of its own term have significantly undermined popular confidence in the institution (Hoehne Reference Hoehne2013; Richards Reference Richards2014). Ó Beacháin et al.’s (Reference Ó Beacháin, Comai and Tsurtsumia-Zurabashvili2016: 445) characterisation that Eurasian de facto states like Abkhazia, Nagorno-Karabakh and Transnistria feature ‘presidential systems in which the executive eclipses both legislative and judicial arms of government’ applies well to Somaliland. In 2005, there was great excitement when two opposition parties won control of Somaliland's parliament. Yet, in the words of Mohammed Hassan Ibrahim (Reference Ibrahim2010, Int.), ‘When the opposition won parliament, they realised they had nothing’. Indeed, if you ask Somalilanders if they think President Riyale would still be president today if he had control over hundreds of millions of dollars of oil revenues in 2008, few venture any confidence that he would have put his legitimacy to the test of free elections. Its repeated electoral delays, its presidentially dominated system, and its utter lack of legislative branch legitimacy do not augur well for Somaliland's ability to manage oil revenues successfully.

ASSESSING SOMALILAND'S VULNERABILITY: HYBRID STATE INSTITUTIONS

Several scholars argue that Somaliland's lack of recognition has provided it with valuable space in which to tailor its state-building project to the specific needs of its society. Looking at the upper house of parliament or Guurti, Richards (Reference Richards2014: 150) argues that, for Somaliland, ‘institutionalised traditional authority is not a sacrifice of acceptable statehood, but rather a necessity for acceptable statehood to be achieved in the emerging state’. She goes on to note that the ‘scarcity of international involvement in Somaliland's state formation has allowed for the process to reflect Somaliland society’ in a way that simply would not have been possible otherwise (Richards Reference Richards2014: 176). Phillips (Reference Phillips2016: 639–40) similarly emphasises that the political settlement in Somaliland ‘evolved without explicit externally driven expectations, schedules, or technical indicators of success’ which gave Somalilanders the freedom ‘to experiment with what was likely to work in the local context’. Extrapolating from this view, Somaliland's ‘hybrid mix’ of traditional and modern state elements has its quirks and contradictions but is a close reflection of Somali society. An influx of oil revenues 20 years ago might have prevented such a state from emerging, but now that it is institutionalised, it will find a socially acceptable way to manage any windfall revenues it receives.

Yet, Somaliland's hybrid system of traditional and modern governance increasingly comes in for sharp criticism. While there is widespread agreement that reliance on clan elders helped maintain peace and facilitate Somaliland's initial development, there are growing fears that ‘the very tools that helped it adapt in the first two decades of postwar recovery are now increasingly an impediment to democracy and economic investment’ (Menkhaus Reference Menkhaus2014: 170). In Hoehne's (Reference Hoehne2013) view, the Guurti no longer provides a check on the executive and has instead been subsumed into it. Its members have grown increasingly comfortable living in Hargeisa and, correspondingly, are now detached from the rural constituents they are supposed to represent. The result ‘is not only an imbalanced but a “crippled” hybrid order that advances neither effective democracy nor strong traditional governance but undermines both the Western-oriented state and the home-grown traditional system’ (Hoehne Reference Hoehne2013: 213).

The ability of Somaliland's executive institutions to manage a sudden influx of oil revenues must also be questioned. First, Somaliland's government is extremely small. Its government spending is less than 10% of gross domestic product (GDP) and its revenue base is significantly smaller than in other sub-Saharan African countries (World Bank 2016). Second, Somaliland's government typically devotes 50% or more of its limited resources to security. This is broadly accepted by the population which does not expect much else from the government. As Mohammed Fadal (Reference Fadal2010, Int.) explains, ‘The government handles security, otherwise it faces few demands or expectations from the people’. Essentially, Somaliland's government has largely outsourced all its non-security development needs to other actors such as aid agencies and its diaspora. Its limited capacity outside the security realm does not bode well for its ability to convert oil revenues into roads, schools or health clinics. ‘Clusters of competence’ exist in various parts of the Somaliland executive but ‘the capacity of the civil service to implement and enforce policies is low’ (World Bank 2016: 79). Menkhaus (Reference Menkhaus2014: 171) goes so far as to characterise Somaliland as a ‘functional failed state’ where ‘state institutions remain weak, corrupt, and even paralyzed, but where society enjoys basic peace, security, and some form of law’. The combination of small and weak state institutions plus an increasingly dysfunctional mix of hybrid and modern elements does not suggest Somaliland is well placed to manage a significant influx of oil revenues.

ASSESSING SOMALILAND'S VULNERABILITY: REVENUE MANAGEMENT

The case not to fear a sudden influx of oil revenues in Somaliland would draw from Morrison's (Reference Morrison2009: 109) empirical finding that oil, foreign aid and other sources of non-tax revenue do not have either pro- or anti-democratic effects but rather have ‘stabilizing properties … These revenues enable a regime to stay in power by whatever means are best for that regime, and this is as true in democracies as it is in dictatorships’. If non-tax revenues ‘stabilize the regime in which they appear’ (Morrison Reference Morrison2009: 131) then oil revenues in Somaliland would stabilise its existing democracy.

Optimists would also note that Somaliland already has some valuable experience managing revenue fluctuations. Saudi Arabia imposed livestock export bans on Somaliland in 1998, in 2000 and again in 2016. Livestock production currently accounts for almost 30% of Somaliland's GDP and is the single largest export earner (World Bank 2016: 11). According to its former finance minister, Somaliland currently receives about $40 million in annual revenue from livestock exports, but about $30 million of that comes during the Hajj, the annual Islamic pilgrimage to Mecca when more than 1,000,000 foreign pilgrims travel to Saudi Arabia and its demand for lamb peaks. The government both plans for less revenue than it thinks will come in and adjusts its spending so there is less before the Hajj money arrives and then more after it arrives (Samale Reference Samale2016, Int.). Its ability to manage these spikes or drops in revenue bodes well for its ability to manage a volatile influx of oil revenues.

Yet, Eubank's (Reference Eubank2012) argument on revenue bargaining clearly suggests that Somaliland is in great danger of falling victim to the political resource curse. Eubank essentially investigates Somaliland as a test case of the taxation leads to representation argument. As explained by Piccolino (Reference Piccolino2015: 558), ‘a central tenet of fiscal sociology has been that there is a causal connection between the dependence of governments on broadly levied taxes and the existence of binding constraints on governments and institutionalised political representation’. In examining Somaliland, Eubank finds that, at critical junctures, the government's need for revenue forced it to bargain with private businessmen and clans who controlled a particularly valuable resource (first, Berbera's port; later, Hargeisa's airport). In return for their consent to provide additional revenue, such groups were able to secure concessions from the government that promoted democratic accountability. In his view, ‘revenue bargaining forced Somaliland's central government to accept institutional arrangements that provided safeguards against the possible rise of a predatory state’ (Eubank Reference Eubank2012: 477). In this account, the minimal foreign aid that Somaliland received due to its lack of sovereign recognition had the beneficial effect of forcing the government to negotiate access to local revenues which facilitated the emergence of more democratic and accountable institutions. Eubank (Reference Eubank2012: 475–6) estimates that were it widely recognised, Somaliland easily could have received $70 million of foreign aid in 2001 which would have been approximately double its entire government budget at that time. In his view, such a large volume of aid would have obviated the government's need to bargain for local revenue and greatly improved its capacity to crush domestic rivals.

We can extrapolate this argument to oil. Somaliland's entire government budget in 2016 was ~$295 million. In 2014, Ghana earned $978.9 million of oil revenues on production that averaged just over 102,000 b/d (Ghana Extractive Industries Transparency Initiative (GHEITI) 2015). With 550,000 b/d of oil production, the KRG earned $750 million every month from oil revenues in 2016 (Genel Energy 2017: 7). Eubank feared foreign aid at twice the level of the government's budget would undermine revenue bargaining and the accountability it generates in Somaliland. Oil revenues potentially generate a much larger impact.

Another reason to fear Somaliland's susceptibility to the resource curse is its poverty. The anti-democratic effects of oil are much stronger in poor countries than in rich countries (Ross Reference Ross2001: 343–4). The corrupting effect of natural resources is also stronger in poor countries (Leite & Weidmann Reference Leite and Weidmann1999: 31). The World Bank (2014) estimates Somaliland's 2012 GDP per capita at $347 making it ‘the fourth lowest in the world’. Somaliland's political economy is also fragile. In a drought-prone region, roughly half the population depend entirely on livestock production for their livelihoods. Somaliland also has one of the highest levels of import-dependency and is one of the most remittance-dependent economies in the world (World Bank 2016). One reason proffered in a recent optimistic assessment of Ghana's prospects for avoiding an oil curse was that ‘Economic diversification, combined with strong macro-economic performance … suggests that Ghana is less likely to succumb to the resource curse’ (Kopinksi et al. Reference Kopinksi, Polus and Tycholiz2013: 588). Such conditions are simply not present in Somaliland.

Finally, there is little evidence of serious preparation for the arrival of oil revenues along the lines of what has been seen in other sub-Saharan countries like Chad, Ghana and São Tomé and Príncipe. In the absence of sovereign recognition, there has been nothing in Somaliland along the lines of the two prominent World Bank-led capacity building projects in Chad in anticipation of the start of its oil production (Pegg Reference Pegg2006). In the run-up to oil production in Ghana, there was widespread recognition of the potential dangers of an oil curse. A National Conference on Oil in 2007 established the principle of government-civil society cooperation and the Civil Society Platform on Oil and Gas was formed in 2010 (Kopinksi et al. Reference Kopinksi, Polus and Tycholiz2013). In São Tomé and Príncipe, Weszkalnys (Reference Weszkalnys2014: 217–18) found ‘a temporal politics of anticipation that is animated by a variety of programs, plans and measures’ which included attempts to resolve maritime boundary disputes, the passage of oil-related legislation and the creation of so many new dedicated entities that the ‘oil sector has become, without a doubt, one of the most developed parts of STP's public administration today’. The point is not that all these preparatory measures were successful. They largely failed in Chad and were never put to the test in São Tomé and Príncipe. Rather, the point is that Somaliland has not yet begun to engage in any such attempts to bolster its institutional and legislative framework in preparation for the arrival of oil. With oil production now anticipated to start in 2019 or 2020 (Oil industry executive 1 2017, Int.; Oil industry executive 2 2017, Int.), time is not on Somaliland's side.

OIL TO CASH IN SOMALILAND

There are several reasons why Somaliland might make a suitable candidate for oil to cash. As oil revenues have not yet arrived, ‘the barriers from entrenched interests are presumably lower’ (Moss et al. Reference Moss, Lambert and Majerowicz2015: 110). Somaliland is also a democracy where, at least at the presidential level, ‘competitive elections with uncertain outcomes determine who rules’ (Diamond & Mosbacher Reference Diamond and Mosbacher2013: 97). It is hard to think of a more electorally compelling strategy for either an incumbent who wished to remain in office or an opposition leader who wanted to become the new president than promising to distribute oil revenues directly to the population. Although not specifically focused on oil to cash distribution schemes, Asunka (Reference Asunka2017) finds some support for this proposition in his investigation of why some Members of Parliament in Ghana choose to limit their discretion in distributing Constituency Development Funds while others do not. Asunka (Reference Asunka2017: 33) finds that ‘Non-discretionary distributive rules enable politicians to extend benefits to voters outside their circle of loyal voters, particularly unattached or weakly attached voters, and potentially broaden their electoral support without alienating their loyal voters who may otherwise feel betrayed if they fail to receive favours’. He also emphasises that ‘non-discretionary distributive rules have a signalling effect: they allow incumbent politicians to credibly signal to voters that they are committed to unbiased and efficient use of public resources’ (Asunka Reference Asunka2017: 33). Beyond this, politicians are ‘more likely to adopt non-discretionary or self-binding resource allocation rules in contexts where voters evince weak attachment to political parties’ (Asunka Reference Asunka2017: 29). The sundry weaknesses of political parties in Somaliland are widely noted (Bradbury Reference Bradbury2008: 217; Verjee et al. Reference Verjee, Abokor, Yusuf, Warsame, Farah and Hersi2015: 36), thus making Somaliland a potentially receptive environment for the kind of non-discretionary resource allocation discussed here as oil to cash.

Somaliland appears well qualified on the four basic requirements for an oil to cash programme to work. First, as a country that has already conducted multiple electoral campaigns and has newspapers, TV, radio, social media and a vibrant oral tradition of debate, Somaliland would not find it difficult to mount the public information campaign needed to raise awareness. On one recent count (Walls et al. Reference Walls, Heine, Klingel, Goggin, Farag and Mwape2018: 23), Somaliland's media landscape comprises ‘eleven newspapers, fourteen television stations, around 60 important websites and one (state-run) radio station … with BBC Somali and Voice of America both particularly influential’. Common criticisms of Somaliland's media environment include repeated arrests and harassment of independent journalists and media outlets (Walls et al. Reference Walls, Heine, Klingel, Goggin, Farag and Mwape2018: 25), a failure to allow private radio stations (Stremlau Reference Stremlau2013), and uniformly pro-independence views which stifle debate on alternative options (Höhne Reference Höhne2008; Stremlau Reference Stremlau2013). Yet, it is also commonly noted that Somaliland's newspapers have influence that extends far beyond their limited circulation figures and a notable ability to facilitate debate on controversial topics (Höhne Reference Höhne2008; Stremlau Reference Stremlau2013). Pegg and Walls (Reference Pegg and Walls2018: 332) note that during the presidential election #SomalilandDebates and #SomalilandVotes both served as valuable conduits of information, as did the Twitter accounts of individual Somalilanders, some of which have more than 10,000 followers. It is not unreasonable to expect that oil to cash would be a topic of great interest and that Somaliland's media landscape would facilitate debate on it.

Second, for its 2017 presidential elections, Somaliland successfully implemented an iris-based biometric voter registration system (Schueller & Walls Reference Schueller and Walls2017). If oil to cash was based on voter registration, Somaliland already has a functioning biometric system in place to do this. If oil to cash was based on a civil or national registry, Somaliland's broadly positive experience with the iris-based biometric voter registration system would certainly facilitate its implementation of a broader civil registration scheme.

Third, Somaliland already has a world-leading system of electronic or mobile payments in place that could be used to implement oil to cash transfers. Estimates on the volume of remittances to Somaliland range from US$500–900 million per year, equivalent to 35–70% of GDP with over 40% of Somaliland households receiving remittances (World Bank 2016: 45). Transfers from the diaspora average about $100 and the costs for making remittances to Somaliland are below both African and global averages (World Bank 2016: 47). Telesom's Zaad service and Dahabshiil's e-Dahab service have led to the almost ubiquitous personal and business adoption of mobile money in Somaliland. Mobile money is widely accepted by even the smallest of merchants and some firms pay their staff salaries in mobile money (Stremlau & Osman Reference Stremlau and Osman2015; World Bank 2016: 50–1). One survey of heavy users of mobile money found that ‘few of them reported problems in using the service. In cases where there were problems, the majority of them were resolved quickly’ (Stremlau & Osman Reference Stremlau and Osman2015: 2–3). Finally, the same mobile money channels used to distribute oil revenues could also be used to tax a portion of them back. This could be done either as a withholding from the initial payment or taken back shortly after the initial payment is made so people had a better sense of how much they were contributing to the government in taxes.

There are at least five reasons why Somaliland's government might be interested in implementing an oil to cash scheme. First, such a scheme could have profound humanitarian benefits. Even if distributions were only in the $50–100 range, this would represent a significant increase in income for most Somaliland residents.

Second, oil to cash distributions would spread development across the country. Economic investment is highly concentrated along the Berbera–Hargeisa–Wajaale corridor. Huge swathes of Somaliland have not participated in the economic growth so visible in Hargeisa (World Bank 2016: 80). The direct distribution of oil revenues would spread resources across Somaliland and lead to economic growth in more peripheral regions.

Third, oil to cash would have a positive political impact on Somaliland's outlying regions. Somaliland has struggled to secure popular legitimacy in the eastern regions of Sool and Sanaag. The equal distribution of revenues among all Somaliland citizens would certainly go some way toward winning hearts and minds in these regions and giving them a central stake in the continued viability of Somaliland. An equal distribution of oil revenues to these regions removes inequitable distribution as a potential source of grievance, raises incomes and thus the opportunity cost of rebelling against the government and gives people in the eastern regions where some of the oil is located a direct stake in maintaining and not disrupting oil production (Moss et al. Reference Moss, Lambert and Majerowicz2015: 84). None of this is to suggest that oil to cash will solve all grievances in the eastern regions or immediately make everyone love Somaliland. Yet, the direct distribution of oil revenues ‘can help generate or repair a sense of national belonging … Cash transfers represent tangible benefits for remaining part of the state and accepting the new political order’ (Moss et al. Reference Moss, Lambert and Majerowicz2015: 128).

Fourth, some Somaliland officials see oil as a route to securing sovereign recognition. Contemporary de facto states like Somaliland often pursue a strategy of ‘earned sovereignty’ whereby their case for recognition is increasingly based upon demonstrating empirical legitimacy in terms of such things as democracy, the provision of security, economic viability and compliance with international norms (Caspersen Reference Caspersen2012; Voller Reference Voller2013; Richards Reference Richards2014). In the case of Somaliland, as Richards (Reference Richards2014: 117–18) emphasises, ‘the campaign for statehood has grown to encompass this acceptableness, and as such the style and appearance of the state … has become a vehicle through which recognition is sought’. What better way for Somaliland to demonstrate its democratic legitimacy and its commitment to the welfare of its citizens than distributing oil revenues directly to its population?

Finally, Somaliland might find an oil to cash scheme marginally beneficial in attracting foreign direct investment in its oil sector. Whatever the size of its potential oil reserves, Somaliland's lack of internationally recognised sovereignty hurts its ability to attract oil investment which fundamentally depends on the stability of contracts honoured over decades. Oil to cash cannot solve this problem entirely, but it does at least partially address political risk by dramatically increasing host country popular support for oil production.

Somaliland may not be a geological frontier region, but its lack of sovereignty makes it a frontier political risk region. The KRG suggests that oil production is possible under disputed sovereignty, but it does not suggest that companies can ignore major political risks in making their investment decisions – a problem brought into sharp relief by the hostile international reaction to the KRG's independence referendum in September 2017. In the case of Somaliland, Somalia disputes its right to exist, the Nugaal oil basin spreads across Somaliland and Puntland and western oil majors can potentially reassert their pre-1991 exploration rights if sizable reserves are found (Walls and Kibble Reference Walls and Kibble2012; Reitano and Shaw Reference Reitano and Shaw2013; Balthasar Reference Balthasar2014).

In Somaliland, oil to cash potentially offers the most benefits in mitigating political risk in the eastern regions where the government's support has been weakest. Oil to cash can also play the role of ‘moral guarantor’ that the World Bank's involvement granted the Chad–Cameroon pipeline project. Oil companies were explicit that they would not accept the political risks in Chad without direct World Bank involvement providing them some form of political risk mitigation (Pegg Reference Pegg2006: 8). Oil companies that might otherwise skip Somaliland because of its inherent political risks might find it more attractive if they could market their investments as directly benefiting some of the poorest people in the world through an oil to cash programme. One oil company executive interviewed clearly indicated that there was nothing about oil to cash that his company would oppose and several aspects of it they would benefit from (Oil industry executive 2 2017, Int.). Oil companies remain geologically bound actors. If Somaliland's reserves are large enough, they will come. Conversely, disappointing early results would keep them away even if Somaliland's sovereignty was widely recognized. In between these two extremes, oil to cash can mitigate some of the political risk inherent in contested sovereignty and marginally improve Somaliland's attractiveness as a site for foreign direct investment.

There are any number of specific permutations that an oil to cash programme could take in Somaliland. Some big questions include whether or not the programme should target specific beneficiaries, whether or not it should be conditional (i.e. the recipients must send their children to school) and whether the government should transfer all or only some of the revenues (Moss et al. Reference Moss, Lambert and Majerowicz2015: 63–7). Such matters need to be decided by Somaliland's parliament. To stimulate thought, though, we put forward one specific option to illustrate how this idea might work in practice.

Recognising that there are many developmental needs in Somaliland which require government investments, our proposal suggests a 50:50 split in total cumulative oil revenues from all sources (taxes, royalties, signature bonuses, etc.) between Somaliland's government and its population. In other words, if $100 million of oil revenue was earned, $50 million would go directly to the government and $50 million would be earmarked for oil to cash distributions with the government then taxing, say, 10% of that back (by comparison, Somaliland currently has a 5% sales tax and a 6% income tax). Importantly, distribution payments must be linked robustly to oil revenues. The idea here is that ‘Citizens must receive a share of the revenues, not an arbitrarily set (and fiscally unsustainable) handout’ (Moss et al. Reference Moss, Lambert and Majerowicz2015: 61). Thus, production or price declines would result in distribution declines.

Given its extreme poverty and the administrative costs and capacity required to target specific beneficiaries or enforce conditionality requirements, our proposed model would target oil to cash distributions to all Somaliland residents age 16 and older (Somaliland's voting age) regardless of their wealth or income. It would also target them wherever they lived. Oil to cash cannot address the specific environmental costs suffered by those who live in oil-producing regions. Additional programmes funded from the government's share of the revenues would be needed to benefit those most adversely affected by oil production. That is not the job of oil to cash. A crowd-pleasing distribution format in Somaliland would be to make the oil to cash payments a week before Eid which is an important religious holiday marking the end of Ramadan (the Islamic holy month of fasting) and a time when families in Somaliland often exchange gifts and host banquets and hence are in greatest need of additional spending money (Jirde Reference Jirde2017, Int.).

Table I uses actual production and revenue figures from Chad, Ghana and the KRG to illustrate the kinds of distributions that could be generated by oil to cash in Somaliland. These figures give some sense of the significant impact on poverty reduction that an oil to cash programme could have in Somaliland. Two different population bases are used. The figure of one million is slightly higher than the 873,331 people who registered to vote in Somaliland's 2017 presidential election. The figure of 3.5 million is based on the United Nations Population Fund's (2014: 31) estimate that 3,508,180 people live in Somaliland.

Table I.

Oil production and revenue figures from Chad, Ghana and the Kurdistan Regional Government of Iraq (KRG) (Ghana 2011 – Moss et al. Reference Moss, Lambert and Majerowicz2015: 137; Ghana 2014 – GHEITI 2015; Chad 2016 – Esso Chad 2017; KRG 2016 – Genel Energy 2017: 7).

These figures, which come from two relatively new African oil producers and one region with similarly disputed sovereignty illustrate a vast range of possibilities for Somaliland. At the lowest end, the 2016 figures from Chad illustrate how little might be available if Somaliland's production volumes disappoint and oil prices are low. They might also suggest some sort of lower threshold of revenues that would need to be reached before oil to cash would be implemented. At the highest end, the 2016 KRG figures are still less than what might be generated if the great hopes of Somalia oil optimists are met. The 2011 and 2014 Ghana figures suggest that per person distributions around the level of Somaliland's GDP per person are certainly not implausible, particularly if they are limited to voting age adults. The Ghana figures also suggest that even the government's 50% share of the oil revenues could still potentially meet or exceed Somaliland's entire government budget today. One can modify the share allocated to the government or how much it taxes back from the distributions, but the clear implication is that an oil to cash programme in Somaliland could both generate significant direct benefits for an extremely poor population (and spread them far beyond Hargeisa) and provide the government with substantial additional revenue to devote to improving the welfare of its citizens further. Beyond these direct benefits, such a programme would also combat the political resource curse, reduce the prospects for conflict and improve democratic accountability via broad-based taxation which ‘would be particularly appropriate for spurring a dual process of statebuilding and reinforcement of state accountability’ (Piccolino Reference Piccolino2015: 559).