Intelligence analyst on a live MENA desk. Open-source collection and verification, intelligence cooperation and failure, counterterrorism, and the politics of resources across the Middle East and Mediterranean, with research on organised crime and state capture in Latin America.
The Mecca Agreement and the Fragmentation of Gulf Security
Written to standard analytic conventions: a stated judgement up front, estimative language throughout, and named indicators that would change the assessment.
Assessment
The Mecca Joint Defence Agreement, signed by Saudi Arabia, Türkiye and Pakistan on 7 August 2026, is best read as evidence that Gulf collective defence is fragmenting, not as the architecture that will replace it.1 The pact commits its signatories to treat an armed attack on any one of them as an attack on all, language modelled on NATO's Article 5, though the joint statement specifies no threshold, no command structure and no basing arrangements.2 It will likely fail to deter Iran and Iran-aligned actors, and its routing outside Gulf Cooperation Council (GCC) channels will accelerate the bloc's division into competing security arrangements.
The first test came within two days of the signing, when Iran-aligned Houthi forces struck Saudi Aramco's Jazan refinery on 9 August and no collective response followed.3 That silence points in the same direction as the September 2025 Saudi-Pakistan Strategic Mutual Defence Agreement, which was in force through the 2026 US-Israeli war with Iran and did not reduce the volume of attacks Riyadh absorbed.4
Strikes on Saudi territory have increased in frequency and reach since then. The Houthis claimed attacks on King Khalid International Airport in Riyadh on 5 October and ballistic missile strikes on Abha International Airport the following day, and Saudi authorities confirmed damage to two southern airports.5 A later strike on Riyadh's international airport killed at least twelve people, the first deaths in the kingdom since the fighting resumed, and led several airlines to suspend flights.6 On 4 October the Presidential Leadership Council announced Operation Dawn of Yemen, a Saudi-backed ground counteroffensive to retake Houthi-held territory, with fighting concentrated around Mocha and the Bab al-Mandeb approaches.5 None of it has produced a joint military response from the pact's signatories.
On 26 October the pact's three signatories sent their defence and intelligence chiefs to Tehran. Saudi Defence Minister Prince Khalid bin Salman, Pakistan's army chief General Asim Munir and Turkey's intelligence chief Ibrahim Kalin met President Pezeshkian, and the only agreed outcome was to continue talking.21 A Pakistani official said the purpose was to persuade Iran to use its influence over the Houthis to stop the attacks on Saudi Arabia.22 Three signatories of a collective defence agreement responded to sustained attacks on one of them by asking the attacker's patron for restraint.
A deterrent already tested
Tehran does not treat the Gulf as a single adversary, and the pact gives it no reason to start. Iranian strike distribution has been calibrated to each state's posture: through early 2026 the UAE absorbed more projectiles than all other GCC states combined, while Qatar and Oman retained working diplomatic channels with Tehran.7 Iran also maintains a deliberate separation from the pact's non-Arab members, protecting energy trade with Ankara and border management with Islamabad, and has avoided direct confrontation with either.8
That separation is what makes proxies the operative instrument. Israeli operations since October 2023 have substantially degraded Hezbollah and Hamas, leaving the Houthis the most capable element of Iran's network still able to strike Saudi territory. Their August strike, and the threats their leadership issued against all three signatories, indicate they read the pact as no constraint.9
That calibration has since moved. Houthi strikes on Saudi territory have increased.1011 Pressure on the UAE has changed form rather than ceased. Direct strikes on Emirati territory stopped after 4 May, while Iranian attacks on Emirati crude tankers have continued.12 Over the same period the UAE has reportedly delivered billions of dollars to Tehran in return for a halt to attacks, and has been recorded as the origin point for 27 million dollars of goods purchased by Iran’s military since the war began.24 Targeting the state that has just purchased a collective guarantee is where testing that guarantee yields the most information at the least cost.
The limits of Saudi escalation
Riyadh has escalated. It has struck Houthi targets in Yemen and it is backing the Presidential Leadership Council's ground counteroffensive. What it has not done is commit to the scale of war that defeating the Houthis would require, and it is fighting through a Yemeni partner rather than with its own ground forces. Turkey and Pakistan will not commit ground forces either.23
The campaign has been testing the pact, American engagement and regional engagement at once, and all three have returned the same answer. Saudi Arabia attempted a decisive campaign between 2015 and 2022 at a cost that bought nothing durable, and its exposure now sits in airports, energy infrastructure and the Vision 2030 programme rather than in territory. The Mecca pact was an attempt to buy deterrence without having to fight for it, and when the deterrent was tested and produced nothing, the only lever left was the one Riyadh had used before, which is talking to Tehran.
The nuclear dimension is weaker than it appears. Pakistan's declaratory doctrine is positioned against India and contains no stated extended-deterrence commitment to Saudi Arabia or Türkiye, which means the pact's most consequential capability may not travel beyond Pakistan's own borders.4 Islamabad's response to Houthi missile attacks on Saudi Arabia has been diplomatic condemnation and reaffirmed solidarity, and reporting on 7 October indicates that Pakistani troops are now present in the kingdom.13 What has not followed is joint action: no combined strikes, no shared command arrangement, and no statement framing the attacks as an attack on all three signatories. Deployment without joint operations is a reasonable preview of what collective defence means in practice.
Fragmentation, not supplementation
Routing an agreement of this scale around the GCC confirms how far the bloc's collective function has already fragmented. The Peninsula Shield Force lacks the integrated air defence, unified command and troop capacity to counter missile and drone saturation, and the GCC's consensus rule produces paralysis during exactly the crises it was built for.14 Gulf states have been substituting minilateral arrangements for bloc mechanisms for years, and the Mecca Agreement is the most consequential instance of a pattern already well established.15
The United Arab Emirates will likely read its exclusion as confirmation that its own non-GCC arrangements were the correct bet. Abu Dhabi already operates defence, intelligence and logistics networks decoupled from GCC consensus, including radar integration and real-time intelligence sharing with Israel and a permanent French base hosting over 700 troops.16 Rivalry with Riyadh over regional leadership, Yemen alignments and the handling of political Islam gives it ample reason to deepen those arrangements.17
Qatar will read the same agreement differently, given its existing defence relationship with Türkiye and the Turkish military presence on its territory. Doha and Tehran jointly manage the North Dome and South Pars gas field, which makes functional relations with Iran a permanent economic requirement regardless of what Riyadh signs.18 A pact that threatens Abu Dhabi reinforces Doha's position, which is itself the clearest illustration of what a fragmented bloc looks like: one agreement, three incompatible readings, and no shared threat assessment underneath them.
The receding American anchor
Riyadh's turn to non-Arab partners indicates that confidence in the United States as a security guarantor has already gone. The September 2025 Israeli strike on Doha, which drew no American action against an ally operating near the largest US base in the region, is the proximate cause most frequently cited.19 States do not build alternative architecture while they still expect the existing one to hold.
The defence-industrial dimension matters more than the security guarantee. Baykar has co-produced the Bayraktar Akinci inside Saudi Arabia since August 2023, with up to 70 percent of each aircraft built domestically, an arrangement that predates the pact and will likely expand under it.20 Saudi procurement of Turkish systems will likely reduce long-term dependence on US arms exports and the approval processes attached to them. Procurement dependence was the mechanism that converted American policy preferences into Saudi behaviour, and removing it leaves Washington's influence resting on channels it does not control.
Indicators and outlook
Iranian pressure through the Houthis will very likely continue at or above current levels into the first quarter of 2027. The relationship gives Tehran deniability and a low-cost means of testing the pact, and nothing in the agreement raises the cost of using it. Two developments would change that judgement: a negotiated settlement in Yemen, or a collapse in Houthi capacity under the current counteroffensive.
The clearest near-term indicator is whether the continuation of talks produces any measurable reduction in Houthi strikes on Saudi territory. A reduction would indicate that Iranian influence over the Houthis is both real and usable, and that Riyadh has bought with diplomacy what the pact did not deliver. Continued strikes at current levels would indicate that Tehran either cannot or will not deliver, which would leave Saudi Arabia without a remaining lever.
The second indicator is whether the pact produces any joint action or joint statement at all. Eleven weeks after signature it has produced neither.
The third is American. A resumption of US strikes on Iran would collapse the mediation the signatories have invested in, and would force Riyadh to choose between the track it has chosen and the partner it no longer relies on.
The fourth is Emirati behaviour. Further UAE defence or intelligence agreements concluded outside GCC channels, particularly with Israel, would mark the transition from a strained bloc to settled competing blocs. A visible Saudi-Emirati reconciliation effort organised around the pact would point the other way.
Sources
"Saudi Arabia, Türkiye, and Pakistan Issue Joint Statement of the Makkah Al-Mukarramah Summit for Joint Defence," Saudi Press Agency, 7 August 2026.
"Fire Extinguished at Aramco Facility in Jizan, Saudi Arabia Says," Reuters, 9 August 2026.
"Understanding the Pakistan-Saudi Defense Agreement," Global Security Review.
"Yemeni Government Forces Launch Saudi-Backed Counteroffensive against Houthis," FDD's Long War Journal, October 2026.
"Has Yemen Returned to 'Full-Scale War'?" Al Jazeera, 10 October 2026.
"Mapping the Damage: Iranian Strikes on the GCC," International Institute for Strategic Studies, March 2026; "Battle Lines: Qatar," European Council on Foreign Relations.
"Beyond Maximum Pressure in US Policy on Iran," Middle East Institute, January 2025.
Senior Houthi official Mohammed al-Bukhaiti issued threats against all three signatories following the agreement, with the majority of his remarks directed at Ankara. Yeni Yemen.
Samia Nakhoul, "How the Houthis are helping Iran turn Saudi Arabia into a pressure point," Reuters, 15 October 2026.
Can Kasapoğlu, "Iran and Its Houthi Proxies Target Saudi Airports and Gulf Shipping," Hudson Institute, 13 October 2026.
"Iran’s hidden web of military agencies is slipping hardware past the US blockade," Reuters, 29 October 2026. The last recorded direct Iranian attack on Emirati territory was at Fujairah on 4 May 2026.
"Pakistan Confirms It Has Troops in Saudi as Conflict with Yemen's Houthi Rebels Escalates," CNN, 7 October 2026.
"Gulf Security: Looking beyond the Gulf Cooperation Council," Center for Strategic and International Studies.
"The Minilateral Moment in the Middle East: An Opportunity for US Regional Policy," Atlantic Council; "Can the Gulf Cooperation Council Transcend Its Divisions," Carnegie Endowment for International Peace, February 2026.
"The Abraham Accords after Gaza: A Change of Context," Carnegie Endowment for International Peace, April 2025; "France and the UAE: A Deepening Partnership in Uncertain Times," Brookings Institution.
"A New Rift in the Gulf, and Only the Gulf Can Solve It," Center for Strategic and International Studies.
"Iranian-Qatari Relations after al-Ula," Carnegie Endowment for International Peace, February 2021.
"Israel's Qatar Strikes and the Fallout for the Abraham Accords," Carnegie Endowment for International Peace, September 2025.
"Mecca Pact Locks Turkish Drone Production in Saudi Arabia as US Arms Lead Narrows," Tech Times, 7 August 2026. Local production agreements for the Saudi Akinci were signed between Baykar and SAMI in August 2023; Janes, 8 August 2023.
"Mecca defence alliance members in Tehran to defuse Iran-US, Yemen tensions," Al Jazeera, 26 October 2026; "Iran, Regional Countries Agree To Continue Talks After Tehran Meeting," RFE/RL, 26 October 2026.
"Saudi Arabia, Turkey, Pakistan send officials to Iran in push to end U.S. war," The Washington Post, 26 October 2026.
Ayesha Siddiqa, quoted in Al Jazeera, 26 October 2026.
"Iran’s hidden web of military agencies is slipping hardware past the US blockade," Reuters, 29 October 2026, for the procurement figure. The reported payments are from Reuters, June 2026, citing four sources; the arrangement has not been confirmed by the Emirati government and is not reflected in all public accounts of the war.
02 / Briefing paperMay 2026Academic exercise
The Evolution of Terrorism
Homegrown extremism, digital radicalization, and the future European threat landscape.
Written to the format of a European law-enforcement strategic briefing for a graduate counterterrorism course. An academic exercise, not a document of any agency. Extracts below; the full paper runs to seven pages with a full reference list.
Executive summary
Terrorism in Europe is evolving faster than the continent's counterterrorism architecture is adapting. The EU is not institutionally absent: Europol, the 2017 Counter-Terrorism Directive, and the Digital Services Act reflect important institutional adaptation. Yet these instruments remain unevenly integrated against a threat environment defined by homegrown extremism, digital acceleration, lone-actor mobilization, and fragmented intelligence coordination.
This briefing argues that the centre of gravity of extremism is progressively shifting toward Europe itself, driven by three structural forces: a liberal democratic architecture that creates a permissive environment for digital radicalization; the failure to develop non-kinetic capacity to interrupt radicalization before it becomes operational; and the fragmentation of intelligence sharing between member states, which produces a tiered architecture that systematically weakens the EU as a unified counterterrorism actor.
Europol's TE-SAT 2025 reports 449 terrorism-related arrests across 20 member states in 2024, including 289 for jihadist terrorism, 47 for right-wing terrorism, and 28 for left-wing and anarchist terrorism. The historical record of the IRA, the LTTE, and Islamic State points to a consistent conclusion: kinetic suppression absent political and social investment relocates extremism rather than terminates it.
The collective action problem in EU intelligence sharing
Bilateral intelligence relationships, whether with MENA partner services, EU member states, or the United States, deliver more operationally valuable intelligence than anything Europol's pooled architecture offers, incentivizing member states to protect those channels and contribute residual intelligence to European mechanisms. Each bilateral pairing produces a rational outcome for the states involved and a collectively irrational outcome for the EU as a whole, as the best intelligence circulates through preferred national channels while Europol operates on the residual.
Even entirely European threats generate indicators across separate national systems with no systematic mechanism to connect them, a synthetic failure the pre-Paris 2015 attack pattern illustrated directly. Brexit compounded this by removing from EU coordination structures a member state whose domestic counterterrorism experience, signals intelligence capacity, and depth on jihadist networks operating across European cities represented capabilities without equivalent replacement within the remaining twenty-seven. A third weakness deepens both: the bilateral infrastructure was built for externally projected threats, while far-right radicalization, online pipelines, and prison recruitment are domestic law enforcement problems in an entirely different institutional lane.
Reforms proposed
A joint analytical cell with direct SIENA access and a mandate to synthesize cross-border indicators across all threat vectors in real time, paired with targeted capacity-building in lower-capacity member states.
A formal EU-UK counterterrorism intelligence accord, distinct from the Trade and Cooperation Agreement, to restore the signals and network intelligence capacity Brexit removed from EU coordination structures.
Collectively resourced prevention infrastructure — deradicalization programmes, prison intervention, community engagement — treated as a strategic function rather than an auxiliary one.
A minimum standards framework for structured intelligence exchange with MENA partners and Israel, scoped to digital radicalization networks and diaspora monitoring, with human rights conditionality embedded in its legal architecture.
A short horizon-scanning brief on the collection disciplines most exposed to encryption, disinformation and automation, written with two concrete capability proposals attached.
Collection
Information systems will have expanded and intensified technological independence, both in collection challenges for agencies and advantages for adversaries. Expanding domains that are starting to move us from centralized telecommunication infrastructure to satellite and peer-to-peer networks — quantum-encrypted communications, social media networks, decentralized protocols — will render traditional interception methods largely ineffective. These networks will force intelligence agencies not only to adapt their collection methods, but to operate within environments saturated with disinformation designed to manipulate or obscure truth, a trend already apparent in deep-fake imagery and bot-driven accounts within peer-to-peer networks.
Processing
5G and 6G networks and IoT ecosystems, spanning sensors, cameras, vehicles, wearables and industrial machines, will generate exabytes of global data at a magnitude never seen before, creating multi-system data flows far beyond current processing capacity. Agencies must build adaptive processing and innovative storage to survive the overload, shifting priorities from "collect everything" to "collect only what is worth analyzing."
Analysis
The overwhelming volumes and the amplification of AI in everyday life will affect human analysis as much as processing. Adversaries will use AI to sabotage datasets, generate false signals and exploit algorithmic biases, threatening analytical integrity and the concept of truth. AI and machine learning have the potential to dominate analysis workflows in order to counter adversarial AI usage, which will cause human analysts to over-rely on AI recommendations, eroding critical thinking and human oversight. Developing safeguards is crucial to democracy, while recognising that adversaries may not face similar constraints.
Proposal 1 — Integrated autonomous collection systems
Self-learning systems capable of autonomous target identification, source prioritization and dynamic collection adjustment across both OSINT and SIGINT, handling open platforms and technical signals while enforcing legal, ethical and jurisdictional constraints. Built-in transparency mechanisms, human-oversight triggers for high-risk decisions, and adversarial-AI detection to identify synthetic content, fake signals and manipulation attempts.
Proposal 2 — Quantum infrastructure
Investment in quantum computing for cryptanalysis, pattern recognition and multimodal data fusion across massive open and technical datasets, developed through partnerships with academia and the private sector to preserve democratic collaboration and oversight. Focused on OSINT–SIGINT workflows including cross-domain correlation, linking intercepted signals with public narratives and sensor data, and prioritising collection-worthiness filters that rank and down-select data by relevance, credibility, uniqueness and operational value.
Research
04 / M.A. thesisIBEI, September 2026Supervisor: Prof. Antonio Díaz
Buying Quiet, Losing Sight
How concessions to non-state actors reshape intelligence cooperation between states. A case study of Israeli–Egyptian cooperation, 2018–2023.
Invited presentation at the Israeli Intelligence Studies Conference, "Intelligence Under Pressure," Israeli Forum for Intelligence Studies, Jerusalem, November 2026.
Abstract
States often manage militant groups through concessions rather than coercion to preserve calm. Research on concessions to militant groups has focused on state–group relations; their second-order effects on intelligence partnerships with other states have received less attention. This thesis asks how concession-based arrangements reshape intelligence cooperation between states, focusing on "buying quiet": a concession-based policy that treats material support to a militant group as deterrence. It argues that this policy can produce closure around an analytical paradigm in which the group is treated as contained, filtering how the intelligence-policy system values a partner's intelligence on that adversary.
The argument is examined through a within-dyad comparison of Israeli–Egyptian intelligence cooperation between 2018 and 2023, comparing Wilayat Sinai with Hamas in Gaza. In Sinai, intelligence sharing translated into sustained operational coordination while Wilayat Sinai was substantially degraded. In Gaza, Egyptian warnings of an impending escalation were transmitted but did not alter Israeli policy toward Hamas. Treating cooperation as a three-stage process of sharing, valuation, and response, the thesis finds that sharing remained active while valuation and response diverged across the two domains.
As a plausibility probe, it argues that the divergence stems in part from Israel's Qatar-mediated arrangement, which reinforced the assessment that Hamas was deterred and led Egyptian warnings to be discounted. Bringing the cooperation and failure literatures into conversation, the thesis shows that an active liaison channel does not by itself ensure effective cooperation. A state's policy toward an adversary can shape how it receives a partner's warning about that adversary. This weakness may become visible only when failure exposes it.
05 / Research essayJanuary 2026IBEI, Prof. Antonio Díaz
The Feedback Loop of Failure
Politicization, conceptzia, and October 7th.
The earlier work the thesis grew out of. Where the thesis follows the Egyptian warning channel, this essay examines the receiving system itself, and treats October 7 as a non-state actor's intelligence success as well as a state's failure. Extracts below; the full essay runs to seventeen pages.
Argument
Intelligence failures rarely stem solely from a lack of warning signs. They result from systemic vulnerabilities: politicization that subordinates analysis to political preference, cognitive biases that filter evidence through flawed preconceptions, and organizational rigidity that underestimates the adaptability of non-state adversaries. The essay argues that Israel's October 7 failure resulted from three interconnected factors: politicization under Netanyahu, cognitive dismissal rooted in a rigid conceptzia holding that Hamas prioritized economic stability over military conflict, and the exploitation of both by a capable adversary.
The new conceptzia
The term conceptzia stems from analysis of the 1973 Yom Kippur War, when Israeli forces underestimated Egyptian and Syrian mobilization because intelligence believed Arab states were incapable of coordinating an attack together. October 7 replicated this with a new preconception: that Hamas sought economic stability, not war. Unit 8200's interception of Hamas's "Jericho Wall" operational plan in April 2022, almost eighteen months before the attack, detailed border breach tactics, drone use to disable surveillance, and mass infiltration by paraglider and motorcycle. Military intelligence labelled the plan aspirational rather than operational. Observation soldiers at the Gaza border filed repeated reports documenting Hamas drills simulating the eventual tactics, which were dismissed as routine training.
What connects these dismissals is not an absence of information but its incompatibility with the prevailing concept. Once policy had established Hamas as deterred, evidence was filtered through that lens, and warnings became noise rather than signal.
The loop
Politicization and cognitive bias were mutually reinforcing. The concessions policy created the preconception and the environment for further politicization of information about Hamas; dismissal of warnings was then justified by perceiving the agencies as politically motivated, during a year in which intelligence officers were among those protesting the judicial overhaul. Politicization therefore operated bidirectionally: leadership pressure shaped intelligence production, while the suspicion of political motive licensed the dismissal of what that production produced. The result was a break in epistemic trust, and a vulnerability an adversary could engineer against.
The adversary's side
October 7 must be analysed as a planned operation rooted in collection, operational security and strategic deception by a non-state actor, not only as a state's failure. A resource-constrained adversary manipulated the perceptions of a technologically superior opponent and denied it critical information, which overturns the assumption that such an actor cannot overcome a state's surveillance capabilities.
06 / Research paperMay 2026Excerpt: conceptual framework
Water Scarcity, Cooperation, and Political Violence in Jordan
How Israeli-Jordanian water deals shifted contention.
The full paper tests this mechanism against ACLED event data for Jordan (2016–2026), the Pacific Institute's Water Conflict Chronology, and climate and scarcity indicators.
Thesis
Israeli-Jordanian water cooperation transformed water from a site of direct military confrontation into a shared resource management problem, bringing direct interstate conflict to a close while producing a new and sustained form of domestic political contention directed at the terms of cooperation itself.
Conceptual framework
This paper distinguishes between two analytically distinct forms of water-related contention. Direct water-related violence covers incidents where water access or infrastructure serves as the proximate target or trigger: sabotage of pipelines, armed clashes over distribution, attacks on pumping stations (ACLED 2015; Geneva Water Hub 2020). Political contention around cooperation covers protests, riots, and state repression triggered by water and energy agreements between Jordan and Israel, where the terms of bilateral cooperation rather than immediate shortage constitute the primary grievance (MERIP 2025; INSS 2025).
The relationship between these two forms follows a four-step mechanism. Chronic scarcity generates structural pressure on the Jordanian state to secure external supply (Haddadin 2009; Jordan Strategy Forum 2020). Israeli control over upstream flows and extensive desalination capacity translates that pressure into institutionalized hydrological and political dependence (Israel-Jordan Peace Treaty 1994; INSS 2025). Dependence then incentivizes successive Jordanian governments to deepen cooperation as a mechanism for managing scarcity and avoiding open interstate confrontation (Harvard Environmental Law Review 2018; Brookings 2022). Because cooperation with Israel carries domestic costs tied to normalization politics and Palestinian solidarity, bilateral agreements shift the site of contention from water infrastructure to the politics of the relationship itself (Schwedler 2022; MERIP 2025; POMEPS 2019).
The result is displacement rather than resolution: one form of conflict recedes as another rises in its place. This dynamic draws on Tarrow's (2011) framework of repertoires of contention, which holds that political opportunities and constraints shape the forms through which grievances find expression. Cooperation with Israel restructured the political opportunity environment in Jordan: direct water confrontation became institutionally foreclosed while anti-normalization protest emerged as the available and socially resonant repertoire through which the costs of dependence could be contested (Tarrow 2011; McAdam, Tarrow, and Tilly 2001).
Implication
Cooperation in asymmetric settings produces interstate stability and domestic mobilization as joint products, a dynamic that conventional conflict datasets, designed to capture organized violence rather than contentious politics, tend to miss. The analytical implication is that peacebuilding assessments relying solely on political violence indicators will systematically overstate the stabilizing effects of resource cooperation in cases where domestic legitimacy costs are high. Demonstration data, anti-normalization movement tracking, and qualitative evidence on state repression all belong in the evaluative toolkit alongside standard conflict event counts.
Limitations of the data
ACLED's Jordan coverage opens in 2016, so the paper treats this decade as the core empirical window and draws on historical scholarship for the pre-treaty and transition-period patterns. Demonstration events are not coded by motive, so the attribution of protest waves to specific bilateral agreements rests on secondary literature documenting the anti-normalization campaigns. The 2018 peak in recorded political violence reflects protest repression rather than organized armed conflict, a reading supported by monthly concentration in May 2018 and by corroborating secondary sources, though full event-level disaggregation would be required to confirm it.
07 / Research paper2026Co-authored
Guayaquil's Narco Turn
Corruption, cocaine transit, and state-crime relations in Ecuador, 2017–2024.
Co-authored with Mayling Aguilar and Sarah Benítez. Full paper available on request.
Question
To what extent did corruption enable Ecuador's transformation into a cocaine transit hub between 2017 and 2024, and how did state-crime relations evolve in Guayaquil over that period? The period is bounded by two decisive moments: the realignment of trafficking networks following Colombia's 2016 FARC peace agreement, and President Noboa's declaration of an internal armed conflict in January 2024, a constitutional break acknowledging the state's loss of substantive control over parts of its territory and security system.
Argument
Existing accounts emphasise structural vulnerability: proximity to Colombia and Peru, dollarization, porous borders. Those factors cannot explain the speed or depth of institutional collapse after 2017. The paper argues that corruption functioned not as a symptom of state weakness but as the constitutive mechanism through which criminal organisations progressively infiltrated and, in specific domains, captured state institutions. Corruption is treated as a spectrum rather than a binary: discrete bribery at one pole, institutional capture at the other, with collusion — ongoing coordinated arrangements between state agents and criminal groups — in between.
Evidence
Ecuador's homicide rate rose from roughly 5.8 per 100,000 in 2018 to between 39 and 45 by 2023. Guayaquil reached 83.8, placing it among the ten most violent cities in the world, and Durán, a warehousing hub across the river from the port, reached 147, the highest urban rate in Latin America. Extortion alerts in Guayaquil rose 476 percent year on year in the first quarter of 2024 and kidnappings by 2,512 percent. Inside the prison system, 717 inmates died between 2018 and 2024, with each of the twelve wings of Litoral Penitentiary controlled by a different criminal group, and shipments coordinated through the port from inside. The Metástasis prosecutions sentenced the former head of the National Judicial Council, police commanders, judges, prosecutors and the former prison agency director.
Finding
The geographic distribution of capture is analytically decisive. Highland cities including Quito, Cuenca and Ambato maintained homicide rates near the international average and experienced no comparable institutional penetration, while coastal institutions sat close to the capture end of the spectrum. This unevenness distinguishes Ecuador from total state failure and points to selective institutional capture concentrated along trafficking corridors. The broader theoretical claim follows from it: institutional capture does not require a weak state in the aggregate, only weakened institutions at the specific nodes where criminal profit concentrates. Treating capture as an all-or-nothing condition consistently misreads cases where the state remains functional in some domains while particular institutions have been structurally reoriented around criminal imperatives.
08 / Peer-reviewed articlePublished September 2026
Pathos for Profit
Affective recursivity in populism.
The argument: some political messaging works by provoking a reaction and then running on it. Outrage and attention are not side effects of the message, they are what keeps it going, so the audience ends up sustaining the thing that provoked them. What matters is therefore how the messaging is built rather than what it claims, which is also why coordinated influence operations and domestic populism can look so alike from the inside.