State, public works and governance · Article published on LinkedIn on 24 March 2026
The Architecture of Delay and the Engineering of Rescue: An Audit of the Latency of Justice
Portuguese justice audited as a network: where the latency lies, the bypass of technical arbitration and the signs of rescue
Gaspar do Amaral · 24 March 2026
Translated from the Portuguese original. This version was issued on 29 October 2026.
On this site, this article has been

Time is not an abstract variable; in highly complex networks and systems, it is the mathematical boundary between operability and catastrophic failure. When the processing latency of a system exceeds the limit of operational tolerance, the network collapses. In Portugal, our most critical infrastructure for economic development — the judicial system — has been operating with destructive latency.
This is not a subjective perception, but facts validated by the European bodies. As the EU Justice Scoreboard relentlessly documents, the processing of a dispute in an administrative court takes, on average, 792 days. Worse still, the consolidation of data in the national Sistema de Informação de Estatísticas da Justiça (SIEJ, Justice Statistics Information System) exposes a suffocating bottleneck: the tax jurisdiction imposes a waiting time that often exceeds six years at first instance.
Recently, the manager and researcher André Antão published an excellent and incisive reflection on this very time frame, dissecting with surgical precision its lethal impact on trust, on planning and on the flight of foreign investment. André Antão’s article serves as a maximum alert for a chronic problem. However, if we accept his challenge and apply the analytical rigour of network engineering and process integration to these same metrics, we discover that collapse does not describe the whole picture. The system operates at two speeds, and the same official reports conceal signs that a silent re-engineering has already begun.
Here is the structural audit of our context economy (the cost of the surrounding environment in which businesses operate):
The worst: static noise and destructive asymmetry
1. The “black hole” of pending cases (the failure read in the SIEJ)
The six years of latency in the tax courts are not just a delay; they represent a massive static noise of complex cases held up for more than a decade. In a structural project, this captive capital stops being a calculable risk and becomes irrecoverable “packet loss” (packet loss). The dispute is not resolved; the money dies inside the system.
2. Oligopolisation by exhaustion (the reverse of the OECD data)
Our 9th-worst place in the OECD’s regulatory facilitation ranking hides a perverse anomaly: the bureaucratic barrier works as armour for the large monopolies. When the context cost becomes unbearable for micro and small enterprises, the inefficiency of the State eliminates competition at the base. Slow justice is an engine of market concentration by exhaustion.
The strategic bypass: the valve of technical arbitration
Faced with a congested network, systems engineering dictates that we cannot suspend operation while the main route recovers; it is imperative to create redundant routes. For the business community, and specifically in the management of complex contracts (such as the integration of transport infrastructure or submarine communications cables), the solution lies in the systematic introduction of technical arbitration mechanisms and expert committees.
Arbitration acts as a vital bypass. Instead of submitting a highly specialised dispute to the bottleneck of the tax or administrative jurisdiction, the parties delegate its resolution to a parallel processing node that is agile and technically literate. This mechanism not only takes load off the central system of the State, but also ensures that the schedule and the financial flow of the investment do not collapse while waiting for a judgment. It is risk management turned into contractual architecture.
The best: the signs of rescue in the central infrastructure
Despite the severe diagnosis, the same official dashboards send out signs of recovery that attest to the viability of the national system:
1. The mathematical inversion (Clearance Rate > 100%)
The recent SIEJ metrics show that the case resolution rate has broken through the 100% barrier. In queueing theory, the throughput (the capacity to clear cases) has finally exceeded the rate of arrival of new cases. The State has stopped accumulating delay and has begun actively to purge its chronic backlog.
2. The paradox of digital hardware (European top 5)
The same EU Justice Scoreboard that penalises us on time places Portugal in the European top 5 for digitalisation and electronic tools. The technological infrastructure (hardware/software) is excellent. The digital motorway is built and in operation; the failure lies exclusively in the procedural protocols inherited from the last century.
3. The high-speed prototype (the OECD’s praise)
The OECD criticises general licensing, but strongly praises Portuguese “Digital Government” and the schemes for start-ups. The State has already proved that it knows how to design world-class fast processing routes (fast-tracks), such as “Empresa na Hora” (a company set up on the spot). The laboratory worked; what remains is to scale up this architecture to the traditional economy.
4. The integrity of the processing unit (the CES data)
The 59% of Portuguese people who give up on the system do so because of its slowness. However, the studies confirm that public trust in the technical competence and impartiality of judges remains high. The central processing unit is not corrupted; it is the transmission bus that is obsolete. The integrity of the final decision remains guaranteed.
The analytical verdict
André Antão’s excellent alert underlines the urgency of the problem, and the data audit points to the way out. We do not inhabit a system in ruins. We have top-level digital infrastructure, validated agile protocols and technical rigour. Our challenge is not a fatalistic burden, but a rigorous and exciting process of network optimisation.
The “architecture of delay” is, therefore, made up of overlapping layers of unnecessary procedural complexity, a culture averse to performance management and an allocation of resources that rarely follows criteria of systemic efficiency. The administrative and tax courts work as saturated nodes in the network, where the rate of incoming cases vastly exceeds the processing capacity (throughput), generating permanent queues. Without mechanisms of dynamic load balancing and without a radical simplification of workflows, the system remains in a chronic state of operational timeout.
The engineering of rescue: systemic optimisation
In the face of this diagnosis, the “engineering of rescue” cannot be limited to palliative measures or to the indiscriminate increase of human resources, which, on their own, have little effect in systems whose processes are badly designed. Rescue requires deep and structural re-engineering.
The first phase of this engineering involves optimising processes through radical simplification. Procedural flows must be audited to eliminate acts that are redundant, bureaucratic and add no value to the final decision. Digital justice cannot be merely the digitisation of paper; it must be the implementation of intelligent workflows, with automation of administrative tasks and the use of data to anticipate and manage congestion.
Secondly, the aggressive promotion of alternative dispute resolution (ADR) is called for. In highly complex systems, diverting load from the main network to secondary routes is a fundamental traffic-management strategy. Mechanisms such as tax and administrative arbitration, mediation and conciliation must stop being exceptions and become organically integrated into the design of the system, freeing the courts for the cases of greatest complexity and social impact.
Finally, the engineering of rescue requires new governance based on data and performance metrics. What is not measured cannot be managed. It is imperative to implement real-time monitoring systems that make it possible to identify bottlenecks instantly and to reallocate resources (judges, court officials) dynamically to the areas under greatest pressure. Transparency about the real latency of each type of case is essential to restore the trust of citizens and businesses.
Conclusion: time as a national imperative
Slow justice is not only justice denied; it is a blocked economy and postponed development. If time is the mathematical boundary between operability and catastrophic failure, Portugal cannot afford to go on operating in the failure zone of its most critical infrastructure.
Recognising the “architecture of delay” is the first step towards implementing the “engineering of rescue”. This is not a corporate or purely legal task, but a national imperative of social and economic engineering. The country’s success in the global arena depends directly on our capacity to turn the judicial network from a node of congestion into an agile, predictable and efficient infrastructure, where time is managed as the scarce and valuable resource it really is.
Gaspar do Amaral
References and analytical usefulness
European Commission — The 2024 EU Justice Scoreboard
Summary of usefulness: A calibration instrument that compares judicial efficiency across the Member States. Crucial for isolating administrative latency (792 days) and, at the same time, for proving Portugal’s position of technological leadership in the communication infrastructure of the courts. Consult the EU Justice Scoreboard report
Ministério da Justiça (Ministry of Justice) — Sistema de Informação de Estatísticas da Justiça (SIEJ)
Summary of usefulness: The primary dashboard of the State’s operation. It allows systems analysts to project clearance curves, identifying the serious latency of the tax area, but also the vital metric of the trend reversal (a positive Clearance Rate). Consult the SIEJ dashboards (in Portuguese)
OECD — Indicators of Product Market Regulation (PMR)
Summary of usefulness: An international barometer of regulatory and bureaucratic friction. Fundamental for quantifying the lethal barrier that oppresses SMEs and for identifying the solutions already tested by the State itself in incubation environments and start-ups. Consult the OECD database
Observatório Permanente da Justiça (Permanent Observatory of Justice) — Centro de Estudos Sociais (CES) da UC (Centre for Social Studies, University of Coimbra)
Summary of usefulness: An audit report on perceived reliability. Essential for isolating the time variable from the corruption variable, showing that the system suffers from a failure of speed but preserves resilience in the competence and integrity of its judiciary (p. 189). Consult the 2013 perception study (in Portuguese)
Methodological note
What this page is. A translation of the article the author published on LinkedIn on 24 March 2026; the full Portuguese version is on this site. The article has no calculations of its own: the figures (792 days, more than six years, 9th-worst place, more than 100%, top 5, 59%, p. 189) are those of the sources it cites, exactly as in the original.
Names and terms. Portuguese bodies and schemes with no official English name (the SIEJ, the CES of the University of Coimbra, the Observatório Permanente da Justiça, “Empresa na Hora”) are kept in Portuguese, with an explanation in brackets. “Context economy” (economia de contexto) is the cost that the surrounding environment — bureaucracy, justice, licensing — imposes on businesses.
Changes of form, not of content. The section titles, which in the original are bold paragraphs, appear as headings and sub-headings. The link to André Antão’s publication, which in the original stands before the paragraph, is placed on the words that refer to it. The four references keep the original order; each “Summary of usefulness” is followed by the link of the original. The original illustration is reproduced with its texts in Portuguese and English and a translated caption.
The discussion is on LinkedIn (in Portuguese)
Comments on this text are made on LinkedIn, where the article was shared in Portuguese. You are welcome to write in English. Agree, disagree or add data: every well-founded objection receives a reply.
Gaspar do Amaral · Original in Portuguese of 24 March 2026, published on this site; translation issued on 29 October 2026 · Figures and tables © Aprecisa.com · AI-assisted