But her story, once a symbol of hope, now frames a harder truth.

Sadly, Rosita’s life was cut short on 12 January 2026. She reportedly died of anaemia in a provincial health centre. This condition might have been treatable in a stronger, better-resourced health system.

Her death coincided with a new wave of severe flooding. Southern Mozambique was under water again in late January 2026. Weeks of heavy rain affected more than 600 000 people. Residual flooding persists in low-lying areas because of upstream inflows and high dam discharges. Towns such as Xai-Xai and Chókwè have faced repeated inundation as the Limpopo swells.

We work for Inclusive Growth, a longstanding research and capacity development initiative in Mozambique. Our work is designed to support evidence-based policymaking to foster inclusive and sustainable economic growth in the country.

Rosita’s story mirrors what our research on Mozambique’s socio-economic development shows at scale: vulnerability persists where poverty, weak public services and deep-rooted inequalities intersect.

The same communities that experience the highest levels of multidimensional poverty, stagnant progress and widening inequalities are also the ones repeatedly exposed to shocks, with limited access to the health, education and infrastructure needed to recover. When floods strike, these disadvantages compound: incomes collapse, assets are lost. Already poor households fall even further behind, making each shock harder to escape and reinforcing long-term deprivation.

As Mozambique faces more frequent and more severe disasters, cycles of vulnerability will be exacerbated. To break this vicious cycle – and prevent future stories like Rosita’s – Mozambique must invest in rapid, well targeted post-shock support, swift livelihood restoration, and sustained, equitable public investment that builds long term resilience.

The fault lines beneath the floodwaters

Research produced under the Inclusive Growth in Mozambique programme shows clearly that the geography of deprivation matters.

Our paper, Evolution of multidimensional poverty in crisis-ridden Mozambique, shows that progress in addressing multidimensional poverty stalled after 2015. Since then the absolute number of poor people has increased, especially in rural areas and in the country’s central provinces.

Inequality trends tell a similar story. Real consumption rose for all groups until 2014/15. But it rose much faster for richer households. Relative gaps widened further from 2015 onwards.

There has also been an increase in between‑group, or “horizontal”, inequalities. These include:

• a widening of the wealth gaps tied to province, ethnolinguistic identity, and the urban-rural divide between 1997 and 2017

• an increasing disconnect between how cities and rural areas develop: average living conditions in urban areas have improved much faster, while, relatively, improvements have stalled in rural areas

• limited internal migration, preventing convergence.

These widening spatial and socioeconomic divides mean that floods don’t strike evenly. They fall hardest on the communities already facing the steepest disadvantages, shaping who is exposed, who loses the most, and who struggles longest to rebuild.

At the household level, Baez, Caruso and Niu (2020) show how quickly welfare collapses when flooding strikes. Cyclones, floods and droughts reduced per capita food consumption by 25%-30%. Consequently, the percentage of people in poverty increased.

Our study on Mozambique’s vulnerability to natural shocks found that affected households saw short-term consumption losses between 11% and 17%. Rural poor households were hit hardest.

This corresponded, at the time, to a 6 percentage points increase in the poverty rate, as a result of the flooding.

These findings show that once incomes fall and assets are depleted, households slip further from access to adequate health, education, or nutrition – the very gaps that contributed to Rosita’s death.

Acting on the evidence

First, protecting consumption in the aftermath of a shock is essential to prevent structural poverty traps. Once emergency support has been provided, temporary, well-targeted, timely and predictable cash support for flood-affected households must be delivered. Recent evidence uncovered that extended delays in transfer payments have materially weakened households’ resilience.

Second, livelihood recovery depends on restoring earning capacity quickly. Evidence from Cyclone Idai in 2019 in Mozambique shows that small enterprises recover more rapidly when the most affected receive immediate liquidity tied loosely to damage severity and sector.

In flood-prone districts, the same logic applies. Families dependent on informal production or trade cannot wait for long bureaucratic procedures. Their resilience depends on rapid access to the means of restoring work.

Third, building resilience before the next flood requires confronting structural inequalities. Patterns of poverty and inequality show that the areas repeatedly suffering the most damage are also those where public investment in health, education, water and local infrastructure lags behind.

Rosita’s life began in a moment of national tragedy and solidarity. Her death reminds us that resilience cannot rely on courage alone. It must be built through sustained, inclusive development and public investment, so that when the Limpopo rises again, more Mozambicans stand ready – with secure livelihoods, functioning clinics, and the possibility of a different outcome.

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

Written by Ricardo Jorge Moreira Goulão Santos, Research Fellow, World Institute for Development Economics Research (UNU-WIDER), United Nations University and Elina Penttinen, Partnerships and Programme Officer, United Nations University

This article is republished from The Conversation under a Creative Commons license. Read the original article