The warlords never left Somalia. They were given budgets, airports and constitutional titles.
The rushed parliamentary vote in Baidoa on 28 March made that reality impossible to ignore. A small number of state parliamentarians — selected through the clan-elder and delegate system rather than any popular ballot — gathered and cast their votes in a single day. Laftagareen received 66, his rival 13. Federal forces then took control of Baidoa. Laftagareen resigned in a now-deleted social media post and left for Nairobi. What collapsed was not a political crisis but a symptom — one expression of an incentive logic that has shaped every level of Somali politics since the federal experiment began.
Laftagareen came to power in 2018 not because he commanded the broadest support among his people but because his main rival had been removed by Ethiopian officers on Somali soil. For seven years, Ethiopian military presence kept him in place. When that presence was withdrawn — Ethiopia placing its forces under Mogadishu’s command in February 2025 — he fell within hours. He did not lose power. He was discarded by the external patron that had installed him.
This is how Somalia’s federal system actually operates. Each federal member state has become its own mini-sovereignty — ports, airports, customs revenues, armed forces — answering first to the dominant clan network and to whatever external patron is currently underwriting it. The fragmentation perpetuates itself not because anyone designs it that way, but because the incentives reliably select for it.
The architecture of this system was not primarily a Somali political settlement. Its intellectual foundation was laid in a 1998 position paper drafted by the Ethiopian Ministry of Foreign Affairs, which proposed local administrative structures as “building blocks” for restoring Somali statehood. The donor coordination body SACB endorsed the concept the following year. By 2002, the peace conference in Kenya was dominated by warlords, shaped by the Ethiopian agenda, and designed to produce transitional institutions amenable to regional interests. When the US entered the picture after 2001, counter-terrorism funding flowed directly to Mogadishu-based warlords organised as the Alliance for the Restoration of Peace and Counter-Terrorism, explicitly to uproot the Islamic Courts Union. The federal architecture Somalis are now asked to work within was shaped, before it was ever constitutionally adopted, by a neighbouring government with its own strategic interests and a superpower running a parallel patronage operation under the cover of counter-terrorism.
The donor architecture reinforces this at every level. The IMF’s 2025 Article IV puts donor funding at roughly 67 percent of the federal budget. Neither the centre nor the states have any structural incentive to build a tax base when external money arrives regardless. The clearest measure of how thoroughly this has hollowed out Somali state-building is that the one armed actor with genuine domestic fiscal capacity is the insurgency everyone is paying to defeat. Al-Shabaab had to extract or die. The federal government and its member states faced no such pressure. So extraction was never seriously attempted.
Understanding why requires being precise about what every external actor got from this arrangement. Ethiopia’s returns were denominated in Somali fragmentation — a weak centre and dependent periphery served Addis Ababa’s regional interests. The Gulf states’ returns were denominated in compliant federal member state leaders who could be instrumentalised for competing agendas. The donor class’s returns were denominated in an insurgency that justified their continued presence and budget lines. Each actor extracted value from Somali weakness because that is what their incentive structure rewarded. The system produces no shared interest in resolving the conditions that sustain it — not because these actors conspire to keep Somalia broken, but because the incentives reliably select for outcomes that do. Uganda’s announcement of full withdrawal after nineteen years, followed this week by its army chief demanding a $1 billion ‘security dividend’ from Turkey, captures that logic precisely — the geopolitical equivalent of a pharmaceutical company demanding reparations the moment a cure threatens to eradicate the disease on which its profits depend.
Turkey represents a structural exception to this pattern. Its returns — on infrastructure, military investment, oil exploration and trade — require a Somalia that actually functions. You cannot extract a return from a rebuilt port if the port isn’t functioning. You cannot extract a return from military training if the army dissolves back into clan militias. You cannot extract an oil return if there is no sovereign counterparty capable of enforcing a contract. Turkey’s investment thesis requires Somalia to actually work — and that structural difference is what separates it from every other external actor in this story.
The evidence of that bet is visible in ways that thirty years of donor patronage never produced. Turkish companies rebuilt the Mogadishu port and airport. Turkish hospitals operate in the capital. Turkish-trained forces equipped with Bayraktar drones took Daynuunaay in under a day — a town al-Shabaab had held for eighteen years, left untouched because it served as a convenient buffer justifying aid and security funding while keeping the federal centre at arm’s length. The same system that preserved it as a buffer could not survive contact with a force that was no longer built to sustain the stalemate. And Turkey’s drillship began Somalia’s first offshore oil exploration this month, in ultradeep water, at a cost no Somali government could have financed independently. Paying those exploration costs upfront — in a country with no drilling history, an active insurgency, and thirty-five years of state failure — is not the behaviour of an actor planning to extract and exit. It is the behaviour of an actor that needs the country to still be there, stable and sovereign, when returns materialise.
None of this makes Turkey a disinterested partner. The oil deal was concluded through executive authority alone, bypassing parliamentary oversight — a governance failure that sits with the Somali administration as much as with Ankara. The 90 percent cost recovery ceiling is consistent with frontier market norms for high-risk deepwater exploration in conflict environments. The royalty cap is harder to defend — at 5 percent of output, Somalia is structurally locked out of meaningful early revenue regardless of how quickly costs are recovered, a figure that neither global standards nor frontier risk profiles justify. Context matters here. Somalia’s 70 percent profit share once costs are recovered compares favourably to Angola’s government take of around 60 percent, Nigeria’s 55 percent, and Ghana’s 58 percent — making it among the more generous profit splits on the continent for a host government. The legitimate grievance is not the profit share but the royalty cap and the absence of upfront bonuses, which deprive Somalia of early revenue regardless of how quickly exploration costs are recovered. Those are the terms that deserve renegotiation, not the overall architecture of the deal. What the deal needed was not simply parliamentary approval by the same political class that has sustained the spoils system, but full public transparency over its terms so that accountability runs to the Somali people rather than through their representatives. An actor pursuing elite capture has every incentive to preserve indirect selection systems — they are structurally easier to influence than a popular vote. Turkey’s backing of the push toward universal suffrage is inconsistent with that goal.
The Baidoa standoff also revealed something about the Somali political class that the external architecture tends to obscure. For months, the same opposition voices had accused President Hassan Sheikh Mohamud of electoral manipulation and power grabs, condemning the rushed constitutional amendments of March 2026 as a blatant attempt to extend his influence. Yet when a similar rushed, indirect process delivered a victory for a regional leader they favoured, many of the same figures quickly endorsed it. The principle that was so fiercely defended when it hurt their opponent was quietly set aside when it benefited their side. This behaviour is structurally rational, not merely personal hypocrisy. It is the rational behaviour of political actors who have learned that principle is a tactical instrument, not a constraint. The problem was never federalism as a constitutional principle — any institutional form gets colonised by the same spoils logic when the surrounding ecosystem of external patronage, insurgency fiscal resilience and clan-based selection remains unchanged.
The system this article has described is good at one thing above all others: absorbing external interventions and redirecting them into the spoils economy. It did it with Ethiopian troops, with Gulf money, with Western budget support. Turkey is the first external actor whose returns require breaking that pattern rather than sustaining it. Whether Somalia’s political class can hold that opening — or whether the system absorbs this bet too — is the question this moment is actually asking.
