Blog
Quantifying the economic impact of conflict traps

As we look towards securing a brighter future for coming generations, we must address the complex challenges that threaten sustainable development and global stability. Among these challenges, conflict stands out as a particularly destructive one. In our latest research, we find that a country that falls into conflict is expected to lose 20% of its GDP per capita after 30 years, compared to if it had always been at peace. Even for small countries, this represents a substantial economic loss, potentially amounting to billions of US dollars, as if the country had been set back by decades in terms of economic progress. In the worst-case scenarios, losses exceed 50%. 

Stable peace is a cornerstone upon which societies build lasting progress. In environments where peace prevails, health systems flourish, educational institutions thrive, economies grow, and governments operate effectively. This stability allows for continuous investment in infrastructure, human capital, state capacity, and many other essential components for development. Conversely, armed conflict interrupts and often reverses this trajectory of progress. 

The severity of conflict's impact has long been recognized, yet measuring its effects remains a complex challenge, particularly for international policy analysis that relies on thorough cost-benefit analysis. A critical, but often overlooked, aspect is the dynamic nature of conflict. Beyond its immediate consequences, conflict erodes social cohesion and degrades material conditions, setting the stage for future conflicts. This initiates a vicious cycle of violence—a phenomenon known as a conflict trap. Even years after a peace agreement, the seeds of past conflicts can sprout new violence.

A model of the conflict trap

To better understand the long-term economic costs of conflict, we aim to quantify them, considering the dynamics of the conflict. In our study, we present a model that can simulate how conflicts influence a country’s development throughout various stages of conflict and peace.

Our model of the conflict trap is a discrete-time Markov process—a mathematical model used to describe systems that change states according to certain probabilities. The model includes a state of conflict, multiple states of post-conflict peace, and a state of stable peace. A country in a state of conflict has a certain probability of remaining in conflict, transitioning to one of several states of post-conflict peace, or transitioning to a stable peace.

We estimate that countries experiencing conflict will continue to do so in the following year at an 80% probability. In the first post-conflict peace year, countries will have a 25% risk of returning to a state of conflict. Approximately half of the countries that exit a state of conflict will experience a resurgence within eight years. This means that the post-conflict phase is extremely risky. 

Regarding the consequences for economic growth, we estimate that being in the state of conflict reduces annual GDP per capita growth by 3 percentage points relative to stable peace, which is within the expected range based on previous research. The post-conflict years show very small positive effects, indicating a slow recovery. 

Simulating growth paths following a year of conflict

By simulating our model many times, we can determine the distribution of the losses caused by conflict. We only input a starting year in conflict. Whether the following year remains in conflict or transitions to peace depends on the transition probabilities we estimate (80% chance of conflict, 20% chance post-conflict). If the state is conflict, GDP per capita is reduced by 3%. If the state is post-conflict peace, there is a very small recovery. Countries go back and forth between conflict and peace states over 30 years, and we track the evolution of GDP during this period. Due to the randomness involved, each simulation is different. We simulate the process 100,000 times to obtain key statistics on the distribution of GDP per capita. 

Our main finding is that a country that enters conflict is expected to have 20% less GDP per capita after 30 years compared to if it had always been at peace. Even for small countries, this represents a substantial economic loss, potentially amounting to billions of US dollars, as if the country had been set back by decades in terms of economic progress. In the worst-case scenarios, losses exceed 50%. 

In other words, doing better or worse inside the conflict trap can explain substantial changes in the long run. Importantly, in our simulations, we do not assume a specific duration for the conflict. Instead, we estimate the impact of just falling into conflict for 1 year, with the rest determined by the estimated conflict dynamics and the estimated impacts of each state.

Finally, our model allows the substitution of GDP per capita with other variables from other fields like health or education. More broadly, our approach holds promise for other applications where dynamic costs arise from trap dynamics, such as in health conditions like cancer or societal issues like crime.

Overall, our research underscores that aiding countries in exiting the conflict trap, and more importantly, preventing them from falling into it in the first place, are key strategies for development. In our view, the size of potential losses justifies even very expensive interventions.

Addressing the impacts of conflict is essential for paving the way toward a sustainable future. By incorporating data-driven models into policymaking, we can support the business case for, and better design, interventions to rescue nations from cycles of conflict, prevent their recurrence, and establish a durable peace. This is critical to a stable foundation for future prosperity. Together, we can forge pathways that secure the futures of generations to come.

Joan Margalef is a PhD student in economics at the Barcelona School of Economics (BSE-UAB), specializing in public finance and the impacts of conflict. His expertise encompasses data analysis, econometric modeling, and economic theory. Beyond academia, Joan has experience as a Research Consultant for public organizations such as Open Philanthropy and the German Federal Foreign Office.

The views expressed in this piece are those of the authors, and do not necessarily reflect the views of the Institute or the United Nations University, nor the programme/project donors.