Journal of Comparative Economics 35 (2007) 689–710
Better off stateless:
Somalia before and after government collapse
Peter T. Leeson
George Mason University, MSN 3G4, Fairfax, VA 22030, USA
Received 22 May 2007; revised 30 September 2007
Available online 10 October 2007
Leeson, Peter T.—Better off stateless: Somalia before and after government collapse
Could anarchy be good for Somalia’s development? If state predation goes unchecked government may not only fail to add to social welfare, but can actually reduce welfare below its level under statelessness. Such was the case with Somalia’s government, which did more harm to its citizens than good. The gov- ernment’s collapse and subsequent emergence of statelessness opened the opportunity for Somali progress. This paper investigates the impact of anarchy on Somali development. The data suggest that while the state of this development remains low, on nearly all of 18 key indicators that allow pre- and post-stateless welfare comparisons, Somalis are better off under anarchy than they were under government. Renewed vibrancy in critical sectors of Somalia’s economy and public goods in the absence of a predatory state are responsible for this improvement. Journal of Comparative Economics 35 (4) (2007) 689–710. George Mason Univer- sity, MSN 3G4, Fairfax, VA 22030, USA.
© 2007 Association for Comparative Economic Studies. Published by Elsevier Inc. All rights reserved.
JEL classification: P59; O12
Keywords: Anarchy; Somalia; Predation
E-mail address: email@example.com.
0147-5967/$ – see front matter © 2007 Association for Comparative Economic Studies. Published by Elsevier Inc. All rights reserved.
[O]ppression by the government … has so much more baneful an effect on the springs of national prosperity, than almost any degree of lawlessness and turbulence under free institu- tions. Nations have acquired some wealth, and made some progress in improvement in states of social union so imperfect as to border on anarchy: but no countries in which the people were exposed without limit to arbitrary exactions from the officers of government ever yet continued to have industry and wealth.
John Stuart Mill (1848, pp. 882–883)
In the developed world, the relationship between state and society is fairly straightforward. Although rent-seeking, public corruption, and government abuse exist, to a large extent devel- oped economies have developed precisely because they have succeeded in overcoming these problems. While far from perfect in this respect, government in the United States, for example, does a good job of protecting citizens’ property rights and uses its monopoly on coercion to provide public goods that, at least in principle, stand to make society more productive.
In the developing world, however, the relationship between government and citizens can be quite different. Here, many political rulers routinely use government to benefit themselves and their supporters at the expense of citizens. These governments are unwilling or unable to protect their citizens’ property rights, but remain strong enough to prey on society. In the extreme, they devolve into little more than organized thuggery, seizing every opportunity to extort their citizens. Ultra-dysfunctional states not only fail to provide public goods and protect citizens’ property. They are in fact the primary threat to their citizens’ property rights and security.
It is common to think that most governments in world are the well-functioning variety. However, this conventional wisdom has it backwards. Well-functioning, highly-constrained gov- ernments that protect property rights and supply public goods are the exception, not the rule. According to the 2007 Failed States Index, nearly 16 percent of the world’s countries (32) have “failing states” (Foreign Policy/Fund for Peace, 2007). In them, governments are often ultra- predatory, dysfunctional, and threatening collapse. According to this index, another 49 percent of the world’s countries (97) are in “warning” mode. Although they have not yet reached the deterioration of those in “alert” mode, they are approaching it. If these measures are correct, in over half of the world, states are either critically or dangerously dysfunctional. The world’s “experiment” with government, then, has been a far more mixed one than most people think.
Since dysfunctional and predatory governments are disproportionately located in the poorest countries, this raises an important question about the link between state and economic devel- opment in the developing world. Is it possible some least-developed countries could actually perform better without any government at all?
Although a properly constrained government may be superior to statelessness, it may not be true that any government is superior to no government all. De Long and Shleifer (1993), for instance, find that in pre-industrial Europe, countries without unified governments performed better in some ways than those with absolutist autocracies. If a state is highly predatory and its behavior goes unchecked, government may not only fail to add to social welfare, but may actually reduce welfare below its level under anarchy.1
1 Moselle and Polak (2001) provide a theoretical model demonstrating when this is the case.
To investigate this question I examine the case of Somalia. In several respects, Somalia is typical of many least-developed countries (LDCs). Like most other LDCs, Somalia is located in Sub-Saharan Africa. Similar to other countries in this region, Somalia was a former European colony, achieved independence, and subsequently came under the rule of a brutal and highly- predatory political regime. Somalia is quite different from other LDCs in one important respect, however. It has no government.
In 1991 Somalia’s state collapsed, creating anarchy in its wake. Although, as I discuss below, there have been a handful of attempts to resurrect central government in Somalia, to date these have been unsuccessful, leaving the country effectively stateless. Somalia therefore provides an interesting natural experiment to explore the hypothesis that if government is predatory enough, anarchy may actually prove superior in terms of economic development.
There has been much hand-wringing over what to do about the situation of anarchy that has characterized Somalia since 1991. Reports from international organizations commonly express fear about the “chaos” of Somalia without a state. According to the International Relations and Security Network, for example, under anarchy Somalia has had “no functioning economy.” In- stead, “clan-based warfare and anarchy have dominated” the country (Wolfe, 2005). Shortly after Somalia’s government collapsed, the United Nations was similarly “Gravely alarmed at the rapid deterioration” of Somalia and expressed serious “concern with the situation prevailing in that country” (UN, 1992, p. 55). The popular press has tended to go even further in its condemnation of the “internal anarchy … [that] has consumed Somalia for the last 15 years” (Gettleman and Mazzetti, 2006). The view commonly presented by these observers is that Somalia “been mired in chaos since 1991” when statelessness emerged (Hassan, 2007).
To be sure, this concern is not without cause. In the year following the state’s collapse, civil war, exacerbated by severe drought, devastated the Sub-Saharan territory killing 300,000 Somalis (Prendergast, 1997). For a time it seemed that Somali statelessness would mean endless bloody conflict, starvation, and an eventual descent into total annihilation of the Somali people. Thus, conventional wisdom sees Somalia as a land of chaos, deterioration and war, and is certain that statelessness has been detrimental to Somali development.
The reason for this belief is twofold. On the one hand, popular opinion sees government as universally superior to anarchy. Government is considered necessary to prevent violent conflicts like those that erupted when Somalia’s state first crumbled, which disrupt economic activity. Government is also considered critical to supplying public goods such as roads, schools, and law and order, which are important to the process of development. From this perspective it is easy to conclude that Somalia, which has no central government, must have been better off when it did.
Second, there is a tendency upon observing problems in distressed regions of the world to see only on the “failure” of the current situation, ignoring the quite possibly even worse state of affairs that preceded it.2 This is especially easy to do for Somalia, which by international standards is far behind indeed. Educational enrollment is abysmally low—a mere seven percent for combined primary, secondary and tertiary schooling. Average Somali income is less than
$1000 (PPP), and preventable diseases like malaria are a genuine threat to Somalia’s inhabitants. These facts, however, say nothing about the status of Somalia before its state collapsed. Thus,
2 This remains a common problem in evaluations of Russia as well. The tendency here, like with Somalia, has been to focus on the significant defects that remain without an appreciation of the fact that, however severe these troubles are, they pale in comparison to the troubles of Russia under communist rule. Russia is undoubtedly better off today than it was under socialism. On this see, Shleifer and Treisman (2005).
forgetting Somalia’s experience under government, it is easy to imagine that nothing could be more damaging to Somali development than the current state of anarchy.
This paper investigates the impact of statelessness on Somali development. I compare the state of 18 key development indicators in Somalia before and after its government collapsed. These indicators are comprehensive in covering all angles of development for which data are available pre- and post-statelessness. While it is important to avoid romanticizing Somalia, the results suggest that statelessness has substantially improved Somali development. I find that on nearly all indicators Somalia is doing significantly better under anarchy than it was under government. This improvement has been made possible by renewed vibrancy in key sectors of the economy and public goods in the absence of state predation.
Due to data limitations, my analysis of Somalia stops in 2005. Since then, the situation in Somalia has changed somewhat. Most significantly, in late 2006 Somali conflict renewed when the international community-backed Transitional Federal Government (TFG) attempted to oust the Supreme Council of Islamic Courts (SCIC), which controlled the southern Somali city of Mogadishu. This conflict disrupted the relatively long-lasting period of peace that preceded it, resurrecting the violence Somalia had largely under control leading up to this. As I discuss in greater detail below, in early 2007 the TFG succeeded in taking control of Mogadishu. However, like previous attempts to reinstate government in Somalia, the TFG’s authority is extremely weak and it is not yet clear whether it will succeed in establishing a new central government in Somalia.
Section 2 of this paper considers Somalia’s pre-stateless political economy. Section 3 com- pares Somalia’s welfare under government vs. anarchy. Section 4 investigates what accounts for Somalia’s improvement under statelessness. Section 5 concludes with a discussion about post-
2005 Somalia and the prospect of reestablishing central government in Somalia.
2. The grabbing hand: Somalia under government
In 1960 British Somaliland and Italian Somalia gained independence from their colonizers and joined together to form the Republic of Somalia. A bloodless coup in 1969 led by Major- General Mohamed Siad Barre overthrew the democratic government that ruled Somalia since independence. Barre went on to take power, and established an oppressive military dictatorship.3
He reigned for 21 years until 1991 when Somalia’s government collapsed and statelessness en- sued.
In 1970, under the influence of the Soviet Union, Barre transformed his military dictatorship into a socialist one. Full-scale central planning pursued under the government’s policy of “scien- tific socialism” brutalized the Somali people. The government slaughtered civilians who posed threats to the government’s plans or political power, used coercive intimidation to create artificial support for its activities, and forcibly relocated others to further the political or economic ends of Barre and his cronies. “Both the urban population and nomads living in the countryside [were] subjected to summary killings, arbitrary arrest, detention in squalid conditions, torture, rape, crippling constraints on freedom of movement and expression and a pattern of psychological intimidation” (Africa Watch Committee, 1990, p. 9).
The state ruthlessly suppressed free speech and controlled all forms of information reaching
Somalis. Newspapers (only one was officially permitted by the government), radio, and televi-
3 As is often the case in dictatorships, technically, the Somali “constitution” of 1979 guaranteed democratic elections for its “president.” In practice, however, this guarantee was worthless. The first “election” for Barre was in 1986 in which he received 99.9 percent of the votes (US Library of Congress, 2006).
sion were fully censored and dissent in any form squelched with force. Under Somalia’s National Security Law No. 54, “gossip” became a capital offense. Twenty other basic civil freedoms in- volving speech, association and organization also carried the death penalty.
The state invested aggressively in building its military. Besides weapons and troops for foreign defense, massive resources were devoted to military structures of domestic repression. Govern- ment created a secret police squad called the National Security Service and paramilitary unit called the “Victory Pioneers” for spying on and eliminating dissenters. Both had legal discretion to detain, invade, kill, and torture at the state’s behest (Africa Watch Committee, 1990).
This created a twofold dire effect for development in Somalia. On the one hand, it left few resources for investment in public goods, like education, health, or transportation infrastructure. This was especially so in pastoral areas where most Somalis lived (Little, 2003, p. 15). On the other hand, Barre’s military dictatorship eliminated any vestiges of restraint on the government’s predatory power. Law No. 1 repealed the constitution and all democratic checks. There were no elections for any political positions; all were appointed by Barre. Military suppression prevented popular uprising. Even dissent through free expression was eliminated. Government was let loose to plunder and abuse citizens for the ends of political rulers.
The state was notoriously corrupt and violent. Political actors and bureaucrats embezzled state funds, extorted and murdered weak portions of the population, and engaged in aggressive asset stripping of state-owned firms. As the UN Development Program characterized it, “The 21-year regime of Siyad Barre had one of the worst human rights records in Africa” (UNDP, 2001, p. 42). This is no small feat considering that during this period Africa was home to some of history’s most savage dictatorships, including the Democratic Republic of Congo’s Mobutu.
In 1975 all land was nationalized along with nearly all major industries and the financial sector. This facilitated government’s ability to expropriate citizens’ property for state projects, like massive state-operated farms, and for politicos’ personal use. Unpopular minority groups, such as the Gosha, were particularly easy prey. In the 1970s and 1980s Barre expropriated Gosha- occupied land to create state-owned irrigation schemes that benefited his allies. In other cases his minions expropriated land for their private use, making Gosha serfs on their own property (Menkhaus and Craven, 1996).
State control of industry in Somalia created inefficiencies like in the Soviet Union. Between
1984 and 1988, for instance, the government-owned Kismayo Meat Factory was open only three months per year. Government also owned tanneries. The “Hides and Skins Agency” paid herders less than half the market value of hides to process in these factories. These firms also utilized only a tiny fraction of their capacity. All told, capacity utilization of Somalia’s state manufacturing firms was less than 20 percent (Mubarak, 1997, p. 2028).
Incentives to be productive, keep costs down, or cater to consumer demands were virtually absent. Factory managers cared only about meeting quotas. This led them to pursue wasteful activities, such as purchasing inputs worth more as raw materials than the output they produced (Little, 2003, p. 39). Some state-owned enterprises were developed purely to benefit political rulers and their friends. For instance, government created the Water Development Program to subsidize private watering holes for the livestock of Barre’s allies.
In the late 1970s Barre abandoned full-blown socialism to attract aid from the IMF. However, government continued to rely on central planning. “Parastatal companies continued to receive subsidies, foreign aid was channeled through state institutions and the state remained sole arbiter in the allocation of profitable contracts. Private sector autonomy was further curtailed by political patronage, which was the easiest way to access resources controlled by the state” (UNDP, 2001, p. 140). In the 1980s there was only one bank in Somalia, state owned and operated. Government
used it to keep afloat failing public firms and to reward political supporters. Only state enterprises or politically well-connected Somalis were able to obtain loans (Mubarak, 1997).
Government also remained involved in most other important economic sectors. Livestock and pastoral-product exports have long been critical to Somalia’s economy. In the 1970s Barre nation- alized most of this trade and continued to control it throughout the 1980s. Government restricted imports and exports and introduced a bewildering array of regulations. Foreign exchange con- trols were also strict. They required exporters to exchange at least half their foreign exchange earnings at the state-set rate, which in 1988 overvalued the Somali shilling to the US$ by more than 120 percent (Little, 2003). This benefited wealthy political patrons who consumed imports but decimated Somalia’s export industry.
In the 1980s government turned to inflation to finance its corrupt and bankrupt projects. Be- tween 1983 and 1990, average annual depreciation of the Somali shilling against the US$ was over 100 percent. In some years depreciation exceeded 300 percent (Little, 2003). Hyperinfla- tion destroyed the savings of Somalis who managed to accrue modest sums over time. It also incapacitated the monetary unit as a means of economic calculation.
Government’s willful mismanagement of public resources prevented the state from being self- supporting. International development agencies, eager to woo Somalia from the influences of Eastern Europe, filled the shortfall with massive inflows of foreign aid. By the mid-1980s 100 percent of Somalia’s development budget and 50 percent of its recurrent budget was funded by foreign aid (UNDP, 2001, p. 118). In 1987 more than 70 percent of the state’s total operating budget was financed this way (Mubarak, 1996).
The early-1980s saw a temporary spike in government expenditures on items like education. But by the late 1980s the weight of nearly 20 years of rampant corruption, repression, and state control had reduced Somali welfare to horrifically low levels. Well prior to the government’s col- lapse the agricultural economy was in shambles, and malnutrition and starvation were common place (Samatar, 1987). In the 1980s Somalia had one of the lowest per capita caloric intakes in the world (UNDP, 2006). At the end of the decade government spent less than one percent of GDP on economic and social services, while military and administration consumed 90 percent of the state’s total recurrent expenditure (Mubarak, 1997).
Government consistently used state resources to privilege members of Barre’s clan at the expense of others. “The Barre regime awarded certain client groups preferential access to arable land and water … Indeed, the Somalia case is a good example of ethnic (and clan) favoritism where private land-grabbing in the Jubba and Shebelle Valleys favored the late president’s clan, the Marehan, while alienating other groups” (Little, 2003, p. 36). In 1988, for example, Barre supported Marehan herders’ unlawful appropriation of Ogaden water points in Southern Somalia.
Barre’s “ethnic favoritism” created tension between Somali clans. In the late 1980s exploited clans reacted by forming faction groups like the Somali Patriotic Movement (comprised largely of Ogaden), the Somali National Movement (comprised largely of Isaaq) and the United So- mali Congress (comprised largely of Hawiye). United against government’s predation on non- Marehans, they joined forces to oust Barre. Unfortunately, the inter-clan tensions Barre created did not immediately disappear with him. The seeds of clan conflict sewn by 20 years of “divide and rule” 4 policy erupted into violence when government crumbled.
4 The application of this terminology to Barre’s regime is from Little (2003).
3. The hidden hand: Somalia under anarchy
In 1988 civil war broke out in the northern part of the country (Somaliland), setting in motion the beginning of the end of government in Somalia. In January of 1991 a coup d’état toppled Barre’s regime, creating statelessness in its wake.5 Tellingly, the same year anarchy replaced government, 400,000 Somali refugees in Ethiopia returned to their homes in Somalia (UNDP,
2001, p. 59). For the next two years, rival factions fought to establish power. These were the days when Somali “warlords,” such as General Hussein Aideed of Mogadishu and Ali Mohamed Mahdi, battled to solidify their bases of strength. At the same time severe drought struck the country, creating famine in its aftermath. In 1992 the UN sent troops to Somalia to quell the conflict and ease suffering, but failed to establish authority, stability or peace in the region.6
Some fighting continued into the mid-1990s, but died down considerably since 1991. By the late 1990s peace prevailed over most of Somalia. Until 2006, when the attempted reestablishment of central government sparked new violence, conflict was isolated and sporadic, confined when it did occur to pockets of small-scale rivalry in a few areas (Menkhaus 1998, 2004; Nenova, 2004). Important to this expanding peace was expanding commerce, discussed below (Menkhaus, 2004; Nenova, 2004).
Most depictions of Somalia leading up to the 2006 period grossly exaggerate the extent of Somali violence. In reality, fewer people died from armed conflict in some parts of Somalia than did in neighboring countries that have governments. In these areas security was better than it was under government (UNDP, 2001). About the same number of annual deaths in Somalia during this period were due to childbirth as were attributable to war—roughly four percent of the total (World Bank/UNDP, 2003, p. 16). And these deaths were combatants, not civilians. “Atrocities against civilians … [were] almost of unheard of” (Menkhaus, 2004, p. 30). This is still too high, but far from cataclysmic. In fact, it’s not far from the percentage of deaths due to homicide in middle-income countries such as Mexico, which in 2001 was 3.6 (WHO, 2006).
In 2006 “a loose coalition of clerics, business leaders, and Islamic court militias known as the Supreme Council of Islamic Courts (SCIC)” gained increasing dominance over key areas of Somalia, including the capital, Mogadishu (CIA World Factbook, 2007). In response to this, the international community-backed Transitional Federal Government (TFG) attacked the SCIC, leading to a battle between them for power. It remains to be seen what long-run impact this conflict will have on Somali peace. But in the short run, at least, this conflict reversed the strides toward more peaceful anarchy that Somalia had largely succeeded in creating prior to the TFG- SCIC clash.
5 When Somalia’s government first collapsed, clans in the northwestern part of Somalia declared this territory an independent sovereignty called the Republic of Somaliland. Somaliland continues to exist, though unrecognized by the global community. In 1998 a number of clans in neighboring eastern portions of northern Somalia also declared themselves autonomous, forming Puntland. Unlike Somaliland, Puntland does not aim at independence from Somalia, but instead sees itself as an independent territory within Somalia. Although Puntland and Somaliland (to a lesser extent) both have “governments,” and thus more formal structure than the southern part of Somalia, these “states” remain weak at best. Neither “government,” for instance, has exhibited the ability to raise significant revenue through taxation. Somaliland and Puntland also dispute territory along their border creating confusion about which entity governs what, and contributing to the stateless or quasi-stateless atmosphere in both.
6 Interestingly, after the UN evacuated the country in 1995, rather than deteriorate, the Somali economy actually im- proved (Little, 2003, p. xvii).
3.1. Improved overall human development
There is no statistical office in Somalia to collect economic, demographic or other types of data that could be used for regression analyses.7 Even before 1991, government collected almost no such information. However, the UNDP, World Bank, CIA, and World Health Organization have collected sufficient data to conduct a study that allows us to compare Somali development before and after statelessness emerged. To do this I examine all development indicators in So- malia for which data are available pre- and post-statelessness, using figures for the most recent available year in each case. 18 key development indicators allow for comparison. I consider the last five years of government preceding the emergence of statelessness (1985–1990) and the most recent five years of Somali anarchy (2000–2005) for which data are available.
Before considering the results of this analysis, it is important underscore several features of the comparison. First, because there are not yet data for the period since the SCIC-TFG clash in late 2006, my comparison does not capture any change in Somali performance on these in- dicators post-the recent conflict. Second, this analysis compares Somalia under government to Somalia under anarchy circa 2000–2005, not to Somalia anarchy in the period of intense civil war immediately following government’s collapse circa 1991–1992. Of course, when state col- lapse coincides with high levels of armed conflict, economic development is not possible. Third, while highly suggestive, these data must be interpreted with caution. The correlation established here cannot establish causation. In addition to the possibility that state collapse is the reason for the improvements we observe in Somali development over this period, it is also possible that other factors may have contributed to this improvement. I will discuss what these are and their plausibility relative to the government collapse hypothesis below.
Data for the pre-1991 period come from the United Nations Development Program’s (UNDP) Human Development Report-Somalia 2001 and the World Bank/UNDP’s (2003) most recent Socio-Economic Survey in Somalia. Data for the post-stateless period are from the CIA World Factbook (2006), UNDP’s (2001, 2006) Human Development Report, the World Health Or- ganization’s WHO (2004) Somalia Annual Report 2003, and the World Bank/UNDP (2003) Socio-Economic Survey in Somalia. Table 1 contains all 18 indicators and the results of the pre- and post-statelessness comparison.
The data depict a country with severe problems, but one which is clearly doing better under statelessness than it was under government. Of the 18 development indicators, 14 show unam- biguous improvement under anarchy. Life expectancy is higher today than was in the last years of government’s existence; infant mortality has improved 24 percent; maternal mortality has fallen over 30 percent; infants with low birth weight has fallen more than 15 percentage points; ac- cess to health facilities has increased more than 25 percentage points; access to sanitation has risen eight percentage points; extreme poverty has plummeted nearly 20 percentage points; one year olds fully immunized for TB has grown nearly 20 percentage points, and for measles has in- creased ten; fatalities due to measles have dropped 30 percent; and the prevalence of TVs, radios, and telephones has jumped between 3 and 25 times.
Per capita GDP (PPP) is lower than its 1989–1990 level, but the data overstate the size of average income in the pre-1991 period, which is likely lower than in it is today. Three sources of bias inflate pre-1991 per capita GDP as a measurement of well-being. First, firm managers in planned economies have strong incentives to over-report output to meet quotas or obtain rewards
7 The “governments” in Somaliland and Puntland have been able to collect some statistics, but nothing substantial or that covers Somalia as a whole.
Key development indicators before and after statelessness
1985–1990a 2000–2005 Welfare change
GDP per capita (PPP constant $) 836b 600c,e ?
Life expectancy (years)
One year olds fully immunized against measles (%)
One year olds fully immunized against TB (%)
Physicians (per 100,000) 3.4 4h Improved
Infants with low birth weight (%)
Infant mortality rate (per 1000)
Maternal mortality rate (per 100,000)
Pop. with access to water (%)
Pop. with access to sanitation (%)
Pop. with access to at least one health facility (%)
Extreme poverty (% < $1 per day)
Radios (per 1000)
Telephones (per 1000)
TVs (per 1000)
Fatality due to measles
Adult literacy rate (%)
Combinedn school enrollment (%) 12.9b 7.5a,f Worse
a UNDP (2001); b 1989–1990; c CIA World Factbook (2006); d 1987–1990, World Bank/UNDP (2003); e 2005; f 2001; g 2006; h 2004, UNDP (2006); i 2000, UNDP (2006) j 2002, WHO (2004); k 2002, World Bank/UNDP (2003); l 1999, UNDP (2001); m 2003; n Refers to primary, secondary, and
tertiary gross enrollment.
(Shleifer and Treisman, 2005). Although Somalia officially abandoned socialism by 1980, the state continued to play a significant role in production until its collapse. In this environment, firm managers likely inflated reported output, leading to artificially high GDP figures. Second, under government a great deal of Somali production was military hardware that citizens did not consume. In fact, to the extent that this hardware was used to suppress the Somali population, this sizeable portion of pre-1991 GDP was actually negative value added from the perspective of citizens’ welfare. Finally, in the pre-stateless period Somalia was one of the largest per capita foreign aid recipients in the world (UNDP, 2001). In fact, “Pre-war Somalia was considered a classic case of an aid-dependent state” (UNDP, 2001, p. 118). By the mid-1980s foreign aid was
58 percent of Somali GNP (UNDP, 1998, p. 57) compared to only nine percent today (UNDP,
2001). In 1987 more than 70 percent of the state’s operating budget was financed by foreign aid (Mubarak, 1996). And before government collapsed, nearly 100 percent of Somali education was financed by foreign aid (UNDP, 2001, p. 120). This discrepancy inflates pre-1991 GDP per capita compared to per capita income today.
If it were possible, accounting for fictitious production under government, the negative value added of military expenditures, and the “foreign aid gap” would likely reduce Somalia’s pre-1991 average income level below its post-1991 level.8 The dramatic increase in post-1991 Somali con- sumption depicted in the data corroborates this fact. A substantial observed rise in consumption
8 The census information used to calculate pre-1991 per capita GDP in Somalia is also controversial (UNDP, 2001, p. 57) and, if understated, would further overstate per capita GDP under government compared to statelessness.
without an attendant rise in per capita GDP suggests an unmeasured increase in per capita income between the pre- and post-anarchy periods not reflected in the data.
Only two of the 18 development indicators in Table 1 show a clear welfare decline under stateless: adult literacy and combined gross school enrollment. Given that foreign aid was com- pletely financing education in Somalia pre-1991, it is not surprising that there has been some fall in school enrollment and literacy. This is less a statement about the Somali government’s abil- ity to generate welfare enhancing outcomes for its citizens than it is a reflection of foreign aid poured into Somali education by the international development community before government collapsed.9
Importantly, the indicators in Table 1 also do not measure the substantial increase in personal freedoms and civil liberties enjoyed by Somalis since the emergence of anarchy. The Somali government ruthlessly suppressed free speech, censoring newspapers, radio and television. Most forms of free expression were punishable by death and foreign travel was severely restricted. Today, in contrast, Somalis are free to travel as they please (restricted only by governments of other nations) and enjoy greater freedom of expression, both privately and publicly. 20 private newspapers, 12 radio and television stations, and several Internet sites now provide information to the Somali public (Freedom House, Reporters Sans Frontieres, 2003). Satellite-based televisions enable the transmission of international news services, including CNN (Little, 2003, pp. 170–
171). Authorities in Somaliland and Puntland have attempted to interfere with media providers in their territories, but freedom of expression remains improved compared to its status under government. This constitutes an additional important, though unmeasured, increase in Somali welfare under anarchy.
As a point of comparison, it is useful to consider Somalia’s development improvements from the 1985–1990 period to the 2000–2005 period relative to movements in the same development indicators in its neighboring countries, Djibouti, Ethiopia, and Kenya. Looking at these countries helps to interpret the findings in Table 1. In particular, it helps to establish if Somalia’s devel- opment improvements were the result of its predatory government’s collapse and substitution with anarchy, or if Somalia would have experienced the same improvements even if it had re- mained under government simply because ‘it was time’ for Somalia to improve. Similarly, this comparison helps to establish if, for instance, the rise of new information technology in this part of Africa is responsible for Somali improvement and would have occurred with or without gov- ernment collapse, or rather there is something unique about Somalia—namely the collapse of its predatory state—that accounts for Somalia’ progress.
In Table 2, I perform this comparison for all development indicators that data permit. I calcu- late the percent improvement (+) or decline (−) for each indicator in each of Somalia’s neighbors between the 1985–1990 period and the 2000–2005 period. The comparisons are unavoidably rough in the sense that they do not compare the precise years from Table 1 in all cases. Fur- ther, they fail to capture the fact that Somalia’s pre-anarchy government was considerably more predatory than the governments of Djibouti, Kenya, or Ethiopia. Thus, the experiences of these countries provide imperfect points of comparison that tend to understate the difference between Somalia’s strides under anarchy and the strides it would likely have made under Barre. Never- theless, they are sufficient to address the general question I am interested in.
9 Furthermore, according to one source at least, overall enrollment in Somalia may actually be higher than its peak in the 1980s (Nenova, 2004).
Somalia vs. its neighbors. % Improvement or decline in development indicators between 1990 and 2005
Djibouti Ethiopia Kenya Somaliad
GDP per capita (PPP)a – +15.5 −4.1 +? Life expectancy (years)b −15.4 +9 −15.6 +5.4
Adult literacy (%)c – – +3.7 −20
Infant mortality rate (per 1000)a +16 +28.5 +7.4 +24.4
Pop. with access to improved water (%)c +1.4 −4.3 +35.6 0
Pop. with access to improved sanitation (%)c +3.8 +333.3 +7.5 +44.4
Telephone mainlines (per 1000)c +40 – +28.6 +1150
Except for Somalia:
a 1990–2005, World Bank (2005) and CIA World Factbook (2006); b 1990–2006, World Bank (2005) and CIA World Factbook (2006); c 1990–2004, World Bank (2005) and UNDP (2006); d For sources and years, see notes in Table 1.
The data reject the hypothesis that Somalia would have improved equally whether it remained under government or not. Consistent with Table 1, Somalia performs worse on adult literacy compared to its neighbors between the periods. Still, on the majority of the indicators consid- ered here, Somalia improved more than its neighbors over the same period, suggesting that the collapse of government resulted in greater development improvements than would have occurred in its absence. In a number of cases, Somalia has been improving while its neighbors have been declining.
Although this analysis helps to exclude some alternative factors that might be driving Somali improvement apart from state collapse, with the data that are available only a tentative conclu- sion can be drawn. Further, the comparison in Table 2 does not help to exclude other possible sources of Somalia’s improvement unrelated to anarchy. For example, the period of Somalia’s state collapse coincides with the rise of a large Somali diaspora, which supports an enormous re- mittance economy that has undoubtedly been important to Somalia’s improvement. Similarly, in
1993–1994 UNOSOM intervened in Somalia and provided large quantities of humanitarian and other resources to Somali citizens, which might also have contributed to Somalia’s improvement without government.
While the importance of these factors cannot be definitively decided, there is some reason to be skeptical that they, rather than state collapse, are responsible for Somali development. For example, rather than an independent cause of Somali improvement under statelessness, the rise of Somali remittances after government’s collapse may in fact be a result of government’s collapse. In stateless Somalia remittances are handled through the hawilaad system, discussed below, a private and self-enforcing financial system for transferring remittances sent to Somalia from abroad. Under Barre’s government, however, the hawilaad system’s predecessor, the franco valuta system, which served a similar purpose, was eventually criminalized, making it more difficult to remit finances to Somalia. When the government collapsed this barrier was removed, leading to the growth of Somali remittances under anarchy.
Similarly, although UNOSOM’s intervention provided critical humanitarian aid to many So- malis, its affect on the situation in Somalia was not purely positive. UNOSOM’s presence led to surges in Somali violence, both against UNOSOM and between competing factions, which feared a shift in the balance of power that UNOSOM’s presence threatened to create. Thus, in ad- dition to providing resources, which likely helped Somali development, UNOSOM also spurred additional violence, which likely inhibited Somali development.
Another factor that complicates my analysis is Somaliland and Puntland, the two northern regions of Somalia, both of which, nominally at least, have some kind of government. Somaliland declared itself a fully independent sovereignty in 1991. Puntland, in contrast, identified itself as an independent territory within Somalia in 1998. Although Puntland and Somaliland both have “governments,” and thus more formal structure than the southern part of Somalia, these “states” remain weak at best. Neither, for example, is recognized as a state by the international community. Further, neither exhibits some of even the most basic characteristics we associate with governments. For example, the “governments” in Puntland and Somaliland do not have a monopoly on the law or its legitimate enforcement. Although some public laws and courts exist, in both regions, the legal system functions primarily on the basis of private, customary law and mechanisms of enforcement—the legal system that governs the totally stateless southern portion of Somalia—which I discuss below (van Notten, 2005).
Similarly, neither Somaliland nor Puntland has proved very successful in extracting taxes from their citizens. In Puntland, government’s “[r]evenue capacity is very limited” (UN, 2006, Chapter III, p. 6); likewise, Somaliland suffers from “weak revenue collection capacity” (UN,
2006, Chapter IV, p. 3). In addition to this, similar to fully-stateless Somalia, in both Puntland and Somaliland, the private sector delivers many, if not most, public goods (UN, 2006, Chapters III, IV). Calling Puntland or Somaliland “governments,” then, is misleading. It is more appropriate to think of these as ultra-minimal states, if they are states at all.
Unfortunately, there are little data that would allow for a disaggregated examination of So- mali improvement in the post-state collapse period. Nevertheless, in Table 3, I present all of the disaggregated data I could find that allow for a comparison of Puntland and Somaliland develop- ment to Somalia’s development overall to try to get some sense of how these somewhat different regions might be influencing the overall figures presented in Table 1.
Only five indicators allow even partial comparison. However, what they suggest is somewhat mixed. Somaliland has substantially better access to water and sanitation than Somalia overall. On the other hand, Somaliland actually fares worse on maternal morality than Somalia overall and has about the same GDP per capita and infant mortality rate. Puntland also has significantly better access to sanitation, but does worse than Somalia overall on access to water and infant mortality. Although disaggregated data that would allow for a thorough comparison are lacking, these figures suggest that although Puntland and Somaliland may be “pulling up” Somalia overall on certain indicators, they may be “pulling down” or not really influencing Somalia overall on several others. The UNDP (2001) reports that Puntland and Somaliland are doing better than southern Somalia, which may well be the case. However, it does not seem that these regions are the exclusive locations of post-Barre progress in the country. Still, the absence of additional data renders any judgments along these lines very tentative. It is therefore important to keep in mind
GDP per capita
Pop. with access
Pop. with access
rate (per 1000
to water (%)
to sanitation (%)
a See Table 1; b UN (2006, Chapter III); c UN (2006, Chapter IV).
that the indicators reported in Table 1 do not disaggregate the regions of Somalia, and thus reflect overall conditions that include both citizens in Puntland and Somaliland, as well as citizens in the fully-stateless, southern portion of Somalia.
4. The sources of Somalia’s progress
4.1. Economic advance
Much of the credit for Somalia’s improved development belongs to its economy, which has been allowed to grow in the absence of government predation. Although economic advance has been uneven, “in some areas, the local economy is thriving and is experiencing an unparalleled economic boom” (Mubarak, 1997, p. 2027). Somalia’s cross-border cattle trade with Kenya is particularly instructive of this progress. Livestock is the most important sector of the Somali economy. It constitutes an estimated 40 percent of Somalia’s GDP and 65 percent of its exports (CIA World Factbook, 2006). Examining changes in the cross-border cattle trade before and after statelessness is therefore a useful way of establishing changes in Somalia’s economy since anarchy emerged.
According to data from the Kenyan Ministry of Agriculture and Livestock Development col- lected by Peter Little (2003), Somalia’s export of cattle to Kenya more than doubled between
1991 and 2000. Figure 1 provides an event study investigating the effect of statelessness on So- malia’s cross-border cattle trade.
This figure examines changes in the Somali–Kenya cross-border cattle trade between 1989 and 2000. These data draw on the cattle trade at Garissa, the main Kenyan border district and a major livestock trading market in the Kenya–Somali borderlands. The dashed vertical line at
Fig. 1. Cross-border cattle trade.
1991 demarcates the emergence of anarchy. Since this time the value of the cattle trade has increased 143 percent, and its volume has increased 132 percent. Between 1989 and 2000 the value and volume of the cattle trade increased 250 and 218 percent respectively. Even during the most intense period of civil war between 1991 and 1992 the cattle trade grew substantially.
This growth tends to understate the true increase in cross-border cattle trade since the onset of statelessness. In 2000 severe drought struck Somalia and Kenya closed its border10 to Somali livestock for fear of importing animals infected with Rift Valley Fever. This depressed livestock trade in the final year for which data are available, shrinking its growth for the period 1991–2000. Between 1991 and 1998, for instance, the value of cattle traded at Garissa grew 400 percent, and between 1989 and 1998 this trade grew 600 percent. In terms of volume, annual sales grew from less than 25,000 cattle in 1989 to more than 100,000 by 1998.
Further, these data reflect only official cross-border cattle exports from Somalia to Kenya. They do not include the substantial cattle trade that occurs without the Kenyan government’s approval. In 1998 unofficial exports entering Kenya from the Lower Jubba region alone add an estimated 70,000 cattle to these data (Little, 2003, p. 38).
The frequency of larger-scale livestock traders has also grown under statelessness. In 1987–
1988, 80 percent of livestock traders had annual sales between one and 600 (small scale). Only
20 percent had annual sales above this level (large scale). By 1998 the percentage of large-scale traders had doubled. Consider Table 4.
Information about crime in stateless Somalia can also be gleaned from this sector. The cross- border livestock trade is facilitated by brokers (dilaal) who certify for buyers and sellers that traded livestock are not stolen. Dilaal incur liability if livestock they certify is illegitimate. In this capacity they act as insurance for cross-border traders. Data on brokers’ fees pre- and post- anarchy suggest that fees have not risen since government’s collapse. Between 1988 and 1998 dilaal fees remained the same (Little, 2003, p. 109). If thievery increased between 1988 and 1998 we would expect to dilaal fees to have risen. The fact that they have not suggests that, at least in the sizeable livestock sector, thievery has not increased under anarchy. In fact, dilaal fees are lower on the Somali side of the cross-border trade than they are on the Kenyan side, indicating that thievery is more problematic in Kenya, which has a government, than in Somalia.
The livestock sector’s expansion is not limited to cross-border trade with Kenya. During the
1990s Somalia accounted for more than 60 percent of all livestock exports in East Africa. In the northern part of Somalia (Somaliland and Puntland) production and annual exports of sheep and goats from the major ports of Berbera and Bossaso have surpassed their pre-1991 levels (Little,
The growth of large-scale livestock traders
Annual sales Percentage of traders
Notes. Data from Little (2003).
10 Saudi Arabia also banned livestock exports from Somalia during this period owing to the Rift Valley Fever.
2003, pp. 37–38). In 1999 these two ports alone were responsible for 95 percent of goat and 52 percent of sheep exports for all of eastern Africa (Little, 2001, p. 194).
Further, Somalia’s economic improvement under statelessness is not limited to its largest eco- nomic activity. Other sectors that have grown under anarchy include service and hospitality. A large part of this progress has been in telecommunications. Local providers have joined forces with multinationals like Sprint, ITT and Telenor to provide cheap, high quality, and extensive mo- bile phone coverage (UNDP, 2001, p. 107; The Economist Anon., 2005). Transportation is also a growing service industry in Somalia. In addition to local transportation services, Somali-owned airlines provide international service for Somalis. By 1997, 14 firms operating 62 aircraft were up and running, an improvement over this industry’s status under government (Nenova, 2004). In the hospitality sector, “unprecedented” construction is taking place in Mogadishu and other major urban centers (UNDP, 2001, p. 203), facilitating the growth of new restaurants and ho- tels. “In Hargeisa, Mogadishu, and Bosasso, investments in light manufacturing have expanded, indicating local investor confidence in the economy and local security” (UNDP, 2001, p. 39).
An improved monetary climate has also contributed to Somalia’s stateless economy. Infla- tion was a significant problem pre-1991 when government appealed to the printing presses to fund its corrupt activities. Skyrocketing inflation made it increasingly difficult to purchase con- sumables. It also created business uncertainty and distorted monetary calculations of economic participants. Although the monetary situation in Somalia is still problematic, under anarchy the Somali shilling (SoSh) has been more stable.
The SoSh was the official currency of pre-1991 Somalia. Post-1991 there was no government to mandate its usage; however the SoSh continued to trade on the world market. Today the SoSh, along with the US$, is the basis of Somalia’s private monetary system.11 There is no central bank or treasury in Somalia. This means that primarily old notes circulate, though in some cases discussed below private parties have printed new currency, adding to the SoSh supply. Figure 2 examines the SoSh/US$ exchange rate between 1986 and March of 2000.
The first dashed line in 1991 indicates the emergence of anarchy. Under Barre’s predatory regime the exchange rate soared. Steep depreciation drove the SoSh from SoSh 110 per $1 in
1986 to SoSh 5700 per $1 by 1991. Following the coup the exchange rate fell precipitously to SoSh 4200/US$ despite the fact that Somalia was in the throws of civil war. Under statelessness, the SoSh has shown significantly greater stability. It lost significant value against the dollar twice during this period—first around 1996, and then after March of 1999. These dates, indicated by the second and third dashed lines, mark two monetary increases. The first was instigated by the Mogadishu-based warlord, Hussein Aideed, who imported new shillings he had printed abroad to fund his faction’s activities. The second was instigated by the fledgling Transitional National Government in the spring of 1999. In an attempt to establish the TNG as a formal authority, its supporters imported SoSh 30 billion they had printed in Canada.
Since 2000, TNG supporters have further added to Somalia’s money supply leading to addi- tional depreciation against the dollar. Nevertheless, the average annual rate of depreciation under anarchy is still only a fraction of its size under government. In just the last four years under government (1986–1990), average annual depreciation of the SoSh was nearly 120 percent. In the first nine years of statelessness (1991–2000), average annual depreciation of the SoSh was just over six percent. The 2000 monetary injection of TNG supporters boosted the 1991–2001 average to around 14.7 percent, and more recent injections promise to depreciate the currency
11 Somaliland also has its own currency, the Somaliland shilling.
Fig. 2. SoSh/US$ exchange rate.
further.12 Still, money appears to be more stable under Somali anarchy that it was under the last years of government.
The SoSh’s improved stability is also reflected by the fact that, at least until several years ago, in parts of neighboring Ethiopia the SoSh was used more extensively used than Ethiopia’s own currency (Little, 2003, p. 144). In fact, prior to the large monetary injections in Somalia in March of 1999 and then in 2000, the SoSh showed greater stability than the national currencies of both Ethiopia and Kenya. From 1996 to February 1999 the SoSh depreciated against the US$ only
12.14 percent. Between 1996 and 1999 the Kenyan shilling lost 32.55 percent against the US$
and the Ethiopian birr depreciated against the dollar 26.58 percent.13
Somalia’s financial market has also improved under statelessness. Numerous remittance firms discussed above, called hawilaad, handle an estimated $500 million–$1 billion sent by members of the Somali diaspora to their friends and family in Somalia each year (UNDP, 2001). Haw- ilaad are instrumental in connecting Somalis with the resources they need to survive and expand their enterprises. At least one of the Mogadishu-based firms is multinational with branches in countries throughout the world (Little, 2003; UNDP, 2001). Remittance businesses are also mor- phing into private depository/lending institutions and will likely contribute to the development of a Somali banking sector. Some offer travelers checks and non-interest bearing deposits, make small loans, and perform other bank-related services (Nenova, 2004). Though still in its nascent stages, Somalia’s financial sector affords greater access to most Somalis today than it did un- der government when financial services (especially loans) went exclusively to public enterprises
12 Calculated using data from Little (2003) and UNDP (2001).
13 Calculated using data from IMF (2006) and Little (2003).
and political allies (Mubarak, 1997). As a result, under anarchy, “Somalis lend and borrow an extraordinary amount of money to one another” (Menkhaus, 2004, p. 34).
The financial sector is not alone to attract attention from multinational businesses. Others have also found stateless Somalia an attractive place to invest. Dole Fruit, for instance, has invested in Somalia. General Motors’ Kenya subsidiary has as well. Total Oil Company now operates in Somaliland, and the BBC has forged a formal affiliation with one of Somalia’s emergent media companies. Several international companies have expressed interest in investing in Somalia’s energy industry, and numerous fishing fleets from Europe and Asia have reached agreements for commercial fishing in Puntland (Little, 2003, pp. 166–167). The Somali economy has a long way to go, but in many ways has progressed since statelessness emerged.
4.2. Improvement in public goods
Supporting the growth of the Somali economy is an improved public goods sector. Public goods remain extremely low, especially in Somalia’s rural areas. However, like Somalia’s econ- omy, they show progress under anarchy. While factions are able to “tax” Somalis traveling on roadways they control, taxes and restrictions on Somalis’ movement and trading activities are substantially lower under statelessness than they were under government. “Taxes, payable to a tentative local authority or strongman, are seldom more than 5%, security is another 5% (more in Mogadishu), and customs duties are next to nothing. There is no need to pay for licenses, or to pay to put up masts” (The Economist Anon., 2005). Further, it does not seem that Somalis are any less likely to enjoy the benefits of fees paid to militia leaders than they were when they paid considerably higher taxes to government (Little, 2003, pp. 7–9).
Public goods come from a variety of sources in stateless Somalia, including the “taxes” charged by militia. Clan militias provide security to citizens in their territories, and militiamen for hire protect businesses, seaports, large markets, and trade convoys. In other cases shari’a, a form of religious law/courts discussed below, provide security by including guards in their court militia in return for payment from businessmen (UNDP, 2001, pp. 109–110). Clan leaders also work together to provide needed public goods in areas outside of Somalia’s big cities where very few exist.
Law and order is provided privately by xeer, Somali customary law, which establishes rules regarding marriage, war, resource use, and social contracts between clans. It is also supported by diya, which defines rules regarding the punishment of misconduct, such as murder or theft.14
Although some secular courts exist, shari’a courts perform an instrumental function in creating legal order. Private courts are funded by the donations of successful businessmen who benefit from the presence of this public good in urban centers. Under anarchy, dispute resolution is free and speedy by international standards (Nenova, 2004; Nenova and Harford, 2004). This constitutes an important improvement in the provision of law and order compared to before 1991. Under government, the legal system was often used as a tool for preying on Somali citizens and punishing the opposition (Africa Watch Committee, 1990; Menkhaus, 2004). “[H]arassment, arbitrary arrest and imprisonment, denial of a fair public trial, and invasion of the home were common features of the life of the Somali citizen” (Hashim, 1997, p. 90). Rampant corruption and political pressures rendered the police and judiciary useless for most Somalis. Because of the state’s collapse, private providers of law and order have been freed to step in. Somalia’s
14 On the private provision of law and order see, Benson (1990, 1991, 1995).
stateless legal system is far from perfect. The justice system is still subject to abuse and the climate in a number of areas remained insecure even before the renewed conflict in late 2006. Nevertheless, there has been improvement compared to the situation under government. “[I]n some parts of Somalia, local communities enjoy more responsive and participatory governance, and a more predictable, profitable, and safer commercial climate, than at any time in recent decades” (Menkhaus, 1998, p. 220).
Education has also benefited in important ways under anarchy. There are more primary schools in Somalia today than there were in the late 1980s under government (UNDP, 2001, p. 84), and this number is growing. The number of formal schools has increased from 600 in
1990 to 1172 under statelessness (UNICEF, 2005b). There are many Koranic schools as well. These focus mostly on the Koran, but students also learn Arabic. Higher education has simi- larly benefited by statelessness. There was only one university in Somalia prior to the emergence of anarchy. Under statelessness, universities have emerged in Borama, Hargeisa, Bossaso, and Mogadishu. These universities offer subjects from computer skills to accounting. According to UNICEF, although the state of education in Somalia remains poor, there is evidence of “gaining momentum in the education sector” (UNICEF, 2005a, p. 2) and improving children’s literacy and numeracy.
Somalia’s “private sector has [also] proved to be a relatively effective provider of key so- cial services, such as water or transport” (UNDP, 2001, p. 42). Transportation for freight and people connects even small villages in Somalia to major urban centers, and is relatively inexpen- sive (Nenova, 2004). A state-owned electricity provider opened in Hargeisa in 2003. However, most Somali electricity is privately provided. Water needs are also supplied by private firms. Private social insurance provides a safety net financed through remittances from abroad. These remittances average $4170 annually per household (Ahmed, 2000, p. 384). Expansive domes- tic clan-based social networks also provide social insurance. In hard times, private welfare can contribute as much as 25–60 percent of household income (UNDP, 2001, p. 68). Private health- care is also available. Although the state of medicine in Somalia remains extremely low, medical consultations are affordable ($0.50/visit) (UNDP, 2001, p. 108). Further, the percentage of So- malis with access to a medical facility has nearly doubled since 1989–1990 before statelessness emerged. Privately-provided public goods like “education and health care services … and utility companies such as electricity and water, are also providing new income generating and employ- ment opportunities” (UNDP, 2001, p. 39) that have further contributed to Somalia’s economic improvement.
5. Concluding remarks
Somalia remains a country with severe problems. But it appears to have fared better under re- cent statelessness than it did under government. A comprehensive view of the data that allow pre- and post-anarchy welfare comparisons suggest that anarchy has improved Somali development in important ways. Contrary to our typical intuition, in Somalia it seems that social welfare has improved because of, rather than despite, the absence of a central state. Somalia’s government was oppressive, exploitative, and brutal. The extent of this predation created a situation in which social welfare was depressed below the level it could achieve without any government at all. The emergence of anarchy in 1991 opened up opportunities for advancement not possible before government’s collapse. In particular, economic progress and improved public goods provision in critical areas flourished in the absence of a monopolistic and corrupt state.
Recognition of this is not to deny that Somalia could be doing much better. It clearly could. Nor is this to say that Somalia is better off stateless than it would be under any government. A constitutionally-constrained state with limited powers to do harm but strong enough to support the private sector may very well do more for Somalia than statelessness. Further, Somalia’s im- provement under anarchy does not tell us whether continual improvement is possible if Somalia remains stateless. It is possible that past some point, to enjoy further development, Somalia might require a central government capable of providing more widespread security and public goods. De Long and Shleifer (1993), for example, show that while pre-industrial European countries under “feudal anarchy” performed better in some ways than those under absolutist autocracies, countries under limited government performed better than both. But this was not the type of government that collapsed in Somalia 15 years ago.
The relevant question for Somalia’s future is thus whether or not a government, were a stable one to emerge, would be more like the constrained variety we observe in the West, or more like the purely predatory variety that systematically exploited Somalis between 1969 and the emer- gence of anarchy in 1991. In the latter case, even if Somalia’s ability to improve is constrained by statelessness, Somali development would still be better served under anarchy than it would be under government. If “good government” is not one of the options in Somalia’s institutional opportunity set, anarchy may be a constrained optimum. Among the options that are available, ultra-predatory government and statelessness, statelessness may be preferable.
In August of 2000, select Somali clan leaders gathered in Djibouti at the urging of the interna- tional community. At this meeting they established the Transitional National Government (TNG) in an attempt to reestablish formal government in Somalia. The TNG, while remaining in name for three years, failed to establish authority. It was crippled by a lack of popular support and an inability to raise tax revenues. The terms of the TNG expired in 2003. This gave rise in 2004 to the Transitional Federal Government (TFG), led by Abdullahi Yusuf Ahmed. The plan was for the TFG to go to Mogadishu and set up the center of the new central government. However, strong divisions within the members of the TFG initially prevented this. Instead of creating a new government, the TFG effectively fractured into two new rival faction groups that did not fundamentally differ from the “warlord”-led factions it sought to replace.
In May of 2006, the TFG and the Supreme Council of Islamic Courts (SCIC), which provided the basis of Somalia’s private legal system, entered a conflict over control of Mogadishu and other key areas in Somalia. With Ethiopia’s assistance, in early 2007 the TFG succeeded in taking control of the capital city where it now resides. The SCIC continues to mount small-scale resistance, but for the moment at least, is not in a position to regain control of Mogadishu. The renewed violence this most recent attempt to reestablish formal government in Somalia created has undermined the relative peace and stability that preceded it in the earlier period of Somali anarchy.
Despite the TFG’s victory over the SCIC and movement to Mogadishu, Somali statelessness persists. The TFG enjoys the support of the international community, but like the TNG, lacks the domestic support needed to establish genuine authority. Surprisingly, it seems that Somalia’s pri- vate sector and has not totally collapsed in the face of the new violence. As one Mogadishu-based electronics store owner commented, for example, even “After the fighting between the Islamists [the SCIC-backed militia] and the warlords [the TFG-backed militias], people are still buying computers. The security [situation] is very, very good” (quoted in Tek, 2006, p. 31). Further, while it is certain that the renewed conflict has been harmful to the progress Somali achieved leading up to this, what little updated data we have on Somalia suggests that this conflict has not totally reversed the strides toward improvement Somalia has made since 1991. The only two de-
velopment indicators from Table 1 available for 2007, infant mortality and life expectancy, both show improvement not only over their levels under Somali government, but also over their levels in 2006. The improvement has been minimal in only one year, but is present nevertheless. Infant mortality has fallen from 114.89 to 113.08 and life expectancy has risen from 48.47 to 48.84 (CIA World Factbook, 2007). Whether or not this improvement is part of a larger trend remains unclear. However, it provides at least some reason to be less pessimistic about the possible impact that recent Somali fighting has had on the progress Somalia achieved under anarchy before this fighting.
Harold Demsetz (1969) famously cautioned economists to avoid committing the “nirvana fallacy,” which compares an imperfect reality with a hypothetical ideal state. Instead we should compare the situation we confront with the relevant alternatives actually available to us. The plans for a path from here to there must be grounded in an assessment of how things were, how they are, and how they realistically could be. His caution is especially useful when considering reforms in the developing world and, as Coyne (2006) points out, for Somalia in particular.
A consideration of the relevant alternatives based on realistically assessing Somalia’s past and present suggests it is unlikely a new central government, at least in the near future, would resemble anything like a constrained, supportive state. The history of Somalia’s experience under government, as well as the ongoing experiences of its neighbors, implies less optimism than is often projected by the advocates of recreating government in Somalia. The factional disagree- ments that led to civil war in the few years after government’s collapse remain strong. Any ruler to come to power from one of these groups would likely turn the state’s power against its rivals rather than to the good of the country, much as Barre’s regime did before it ended. The TFG has sparse domestic support precisely because of this and because faction leaders recognize the strong possibility that any one faction gaining too much power could mean the virtual annihila- tion of the others.
Indeed, thus far in the stateless period, the three greatest disruptions of relative stability and renewed social conflict have occurred precisely in the three times that a formal government was most forcefully attempted—first with the TNG, later with the TFG, and finally most recently when the TFG mobilized violently to oust the SCIC. In each case the specter of government disturbed the delicate equilibrium of power that exists between competing factions, and led to increased violence and deaths due to armed conflict (Menkhaus, 2004). At the moment at least, it seems that in upsetting this delicate balance of power the attempted reestablishment of govern- ment in Somalia will lead to more conflict and obstacles to progress rather than less.
I thank Peter Boettke, Tyler Cowen, Christopher Coyne, Andrei Shleifer, Russell S. Sobel, William Trumbull, the Editor, and two anonymous referees for helpful comments and sugges- tions. Matt Ryan provided valuable research assistance.
Africa Watch Committee, 1990. Somalia: A Government at War with its Own People. Africa Watch Committee, New
Ahmed, Ismail, 2000. Remittances and their economic impact in post-war Somaliland. Disasters 24, 380–389. Anon., 2005. Somalia calling: An unlikely success story. The Economist (December 20).
Benson, Bruce L., 1990. The Enterprise of Law: Justice without the State. Pacific Research Institute, San Francisco.
Benson, Bruce L., 1991. Law and order without state coercion. In: Veljanovski, Cento (Ed.), Regulation Regulators and the Market. IEA, London, pp. 159–179.
Benson, Bruce L., 1995. An exploration of the impact of modern arbitration statutes on the development of arbitration in the United States. Journal of Law Economics, and Organization 11, 479–501.
CIA World Factbook, 2006. Somalia. Available at: http://www.umsl.edu/services/govdocs/wofact2006/geos/so.html. CIA World Factbook, 2007. Somalia. Available at: https://www.cia.gov/library/publications/the-world-factbook/geos/
Coyne, Christopher J., 2006. Reconstructing weak and failed states: Foreign intervention and the Nirvana fallacy. Foreign
Policy Analysis 2, 343–361.
De Long, Bradford, Shleifer, Andrei, 1993. Princes and merchants: European city growth before the Industrial Revolu- tion. Journal of Law and Economics 36, 671–702.
Demsetz, Harold, 1969. Information and efficiency: Another viewpoint. Journal of Law and Economics 10, 1–21. Foreign Policy/Fund for Peace, 2007. Failed States Index 2007. Available from: http://www.fundforpeace.org/web/
Gettleman, Jeffery, Mazzetti, Mark, 2006. Somalia’s Islamists and Ethiopia gird for war. New York Times (December 14). Hashim, Alice, 1997. The Fallen State: Dissonance, Dictatorship and Death in Somalia. Univ. Press of America, Lanham. Hassan, Mohamed Olad, 2007. UNICEF: Somali kids urgently need food. Associated Press (September 12).
IMF, 2006. International Financial Statistics Online. Available at http://www.imfstatistics.org.
Little, Peter D., 2001. The global dimension of cross-border trade in the Somalia borderlands. In: Assefa, Taye, Rugu- mamu, Severine, Ahmed, A.G.M. (Eds.), Globalization Democracy and Development in Africa: Challenges and Prospects. OSSREA, Addis Ababa, Ethiopia, pp. 179–200.
Little, Peter D., 2003. Somalia: Economy without State. Indiana Univ. Press, Bloomington. Menkhaus, Ken, 1998. Somalia: Political order in a stateless society. Current History 97, 220–224.
Menkhaus, Ken, 2004. Somalia: State Collapse and the Threat of Global Terrorism. Oxford Univ. Press, New York. Menkhaus, Ken, Craven, Kathryn, 1996. Land alienation and the imposition of state farms in the lower Jubba Valley.
In: Besteman, Catherine, Cassanelli, Lee V. (Eds.), The Struggle for Land in Southern Somalia: The War Behind the
War. Westview Press, Boulder, pp. 155–178.
Mill, John Stuart, 1848. Principles of Political Economy. J.W. Parker, London.
Moselle, Boaz, Polak, Benjamin, 2001. A model of a predatory state. Journal of Law Economics and Organization 17,
Mubarak, Jamil, 1996. From Bad Policy to Chaos: How an Economy Fell Apart. Praeger, Westport.
Mubarak, Jamil, 1997. The ‘hidden hand’ behind the resilience of the stateless economy in Somalia. World Develop- ment 25, 2027–2041.
Nenova, Tatiana, 2004. Private sector response to the absence of government institutions in Somalia. The World Bank, Washington, DC.
Nenova, Tatiana, Harford, Tim, 2004. Anarchy and invention: How does Somalia cope without government? Public
Policy for the Private Sector Note, No. 280, 1–4.
Prendergast, John, 1997. Crisis Response: Humanitarian Band-Aids in Sudan and Somalia. Pluto Press, London. Reporters Sans Frontieres, 2003. Freedom of the Press throughout the World: 2003 Africa Annual Report. Available at:
Samatar, Abdi, 1987. Merchant capital, international livestock trade and pastoral development in Somalia. Canadian
Journal of African Studies 21, 355–374.
Shleifer, Andrei, Treisman, Daniel, 2005. A normal country: Russia after communism. Journal of Economic Perspec- tives 19, 151–174.
Tek, Nathan, 2006. The profits of Islam. The Yale Globalist 2, 31.
UN, 1992. Resolution 751 1992. Available at: http://daccessdds.un.org/doc/RESOLUTION/GEN/NR0/010/92/IMG/ NR001092.pdf?OpenElement.
UN, 2006. Somali Reconstruction and Development Framework, 4 vols. Available at http://www.somali-jna.org/
UNDP, 1998. Human Development Report 1998—Somalia. UNDP, New York. UNDP, 2001. Human Development Report 2001—Somalia. UNDP, New York. UNDP, 2006. Human Development Report 2006. UNDP, New York.
UNICEF, 2005a. Education. Available at http://www.unicef.org/somalia/SOM_EducationFNL.pdf.
UNICEF, 2005b. Communities Unite Around Education in Somalia. Available from: http://www.unicef.org/
US Library of Congress, 2006. Country Studies: Somalia. Available at: http://countrystudies.us/somalia/.
van Notten, Michael, 2005. The Law of the Somalis: A Stable Foundation for Economic Development in the Horn of
Africa. The Red Sea Press, Trenton, NJ.
WHO, 2004. WHO Somalia Annual Report 2003. WHO, Somalia.
WHO, 2006. WHO Mortality Database. Available at: http://www3.who.int/whosis/menu.cfm?path=whosis,mort.
World Bank, 2005. World Development Indicators. World Bank, Washington, DC.
World Bank, UNDP, 2003. Socio-Economic Survey 2002—Somalia. UNDP and the World Bank, Washington DC/New
Why, then, do rulers take unto themselves a power which is not theirs?… petty princes busy themselves with dabbling in matters they do not understand in order to satisfy their own or someone else’s prejudices, or in blindly following some minister’s advice.
They gather together a great many of their subjects in separate flocks and bestow favors on them at the cost of the others, and these favors they elevate into fixed privileges.
Government can not make man richer, but it can make him poorer.