• Floods are important - they on average kill 5,000 people and cause $35B of damage each year.
    • Compared to other problems in global health, however, floods are relatively minor.
  • Flood prevention - and especially response - are less neglected than other global health problem areas.
    • This may be because floods harm rich people too, who are politically empowered and cause governments to invest in mitigation.
  • I estimate that to beat GiveWell’s top charities, we would need to prevent all flood damage using $400M per year.
    • This seems very difficult.
  • I investigate several interventions in more detail, which show some promise but are unlikely to be competitive with GiveWell’s top charities.
  • This report was produced with roughly 50 hours of research and writing, as part of the Cause Innovation Bootcamp fellowship program.

[EDIT: to clarify, I think there could well be cases where someone is unusually well placed to influence government flooding policy or develop scalable technological solutions, such as Google's Flood Forecasting Initative, and in these cases working on floods makes sense. My main claim is that floods should not receive significant attention from EA global health and development funders at this stage]


Floods are defined as land that is normally dry being submerged underwater. There are several quite distinct types of floods, the most important being those caused by:[1]

  • Rain: intense rainfall in a localised area can exceed the capacity of the ground to absorb water, and lead to a buildup of water in low-lying areas.
  • Rivers: distributed rainfall anywhere in a river’s catchment area, rapid snow-melt, or the collapse or quick release of dam water, can all cause a river to overflow its banks.
  • Storm surges: intense winds can blow ocean water onto land, and the lower air pressure of large storms causes the local sea level to rise by up to about 1m. Particularly if combined with a high tide this can inundate coastal areas.


Floods are a big deal. Of natural disasters occurring between 1998 and 2017, a UN report estimated that 43% of disasters were floods (next most common: storm), floods accounted for 45% of disaster-impacted people, at 2 billion (next highest: drought), 11% of deaths (earthquakes caused 56%), and 23% of the economic damage (storms were 46%).[2]

While the toll of floods is generally well recorded in terms of deaths and financial cost, at least for larger disasters, data is scarcer and less systematic for other harms.

Floods can have many negative impacts, ranging from direct to indirect and immediate to long-term:[3]

  • Deaths: the majority of flood deaths are drownings, with other causes including injuries from falling objects or debris in fast-moving water, and electrocution from fallen powerlines.[4] Floods caused an average of 5,000 deaths per year in 2001-2020, with the bulk of the deaths caused by many smaller events: the deadliest 104 floods accounted for half the deaths in this period.[5]

  • Injuries: as well as injuries during the flood, reconstruction and clean-up causes various injuries including sprains and strains, falling off roofs or ladders, and lacerations from sharp debris.[6] Floods caused an average of 15,000 injuries per year in 2001-2020, though injuries are probably significantly underreported.[7]

  • Water-borne diseases: particularly if sewage systems have been compromised, contact with floodwater can cause a variety of diseases notably diarrhoea, cholera and leptospirosis.

  • Mosquito-borne diseases: the abundance of stagnant water as floodwaters recede may cause a surge in mosquito populations and an associated spike in malaria, dengue and other vector-borne diseases.[8]

  • Malnutrition: lost crops and reduced purchasing power can lead to impaired nutrition in the years following a major flood event in a poor region. One study found wasting to be more than twice as common in children who were flooded twice in the last three years than in control households.[9]

  • Mental health: many studies have found increased incidences of PTSD, anxiety and depression in individuals who experience flooding, and this effect is sometimes noticeable after several years.[10]

  • Economic damage: the destruction of key physical infrastructure, including homes, transport routes, and factories, is a large part of flood damage. Floods caused an average of $35B in damage in 2001-20, with half the economic damage caused by the 31 most severe events.[11] The quoted financial toll of floods is often limited to the lost physical infrastructure. However, there is also significant financial loss from foregone productivity while workers and equipment are unavailable because of the flood, and this can last weeks as floodwaters subside and the cleanup progresses.[12] While dollar figures of damage are generally larger in rich countries, the moral importance of economic damage is greater in poor countries.[13] Overall the direct health effects, notably deaths, from floods are relatively small compared to other global problems, for instance the WHO estimates that 1.6 million people died of tuberculosis in 2021, and 0.6 million of malaria. Because flood deaths are relatively diffuse across the global population, any localised intervention will reduce few deaths in expectation. Therefore, it should be surprising if floods were cost-effective to work on just because of human deaths.

To roughly capture the full scale of harms caused by floods, I assumed that the total health impacts are ten times those just from deaths, as there is far less data on non-fatal health impacts.[14] I also calculated the economic harm from all floods in units of log change in GDP per capita, assuming that the true economic damage is double that which is reported (details in the accompanying model). The health impacts were 26% of the total harm, with the remainder being economic damage. In terms of GiveWell’s units of moral value, floods cause 20 million units of harm per year. Thus, if we had an intervention that magically removed all damage from floods, we should be willing to spend up to $400M per year to implement the intervention, given the current margin of cost-effectiveness in GiveWell top charities.


Completely solving flooding with $400M per year seems very unrealistic given how well-resourced flood prevention is compared to other problem areas. Twelve Asian flood-prone countries spend a total of $57B per year on flood protection, with $33B coming from China.[15] Another estimate found that worldwide spending on emergency responses to natural disasters (including but not limited to floods) averages $105B per year.[16] For comparison the world spent just $5.3B on tuberculosis in 2020 and $1.3B on malaria.[17]

So, very roughly, to be cost-effective our flooding intervention would need to spend two orders of magnitude less money per year than Asia currently does, while actually solving the entire problem worldwide for a year (or half as much to solve half the problem, etc).[18] In reality flood-mitigation interventions tend to last more than one year, so another framing is that we would need to spend Asia’s current annual flood-mitigation budget to solve flooding worldwide for the next century.

Note that even when lots of money is being spent on an area as a whole, specific interventions may still be very neglected. I do not have a breakdown of how flood mitigation money is currently spent, so cannot tell which interventions may still be neglected.

I speculate that this relatively non-neglected nature of floods is unsurprising. Issues generally have more resources devoted to them when they impact powerful people and interest groups. While many EA interest areas in global health and development almost exclusively target the very poor - malaria, malnutrition, parasitic worms, tuberculosis - natural disasters are far more egalitarian in their impacts. This is because it is far harder for the rich to defend against unpredictable natural disasters than it is to avoid most infectious diseases. Moreover, the economic damage of natural disasters is likely skewed towards the rich, both within and between countries, as they have more expensive assets, and will incur greater losses from missed business opportunities. The humanitarian toll, of course, still falls disproportionately on the poor. This means that rich and politically empowered groups have strong motivations to invest in disaster mitigation and adaptation measures (and encourage their governments to do so), while they can safely ignore diseases of poverty.


Despite reasons for pessimism about the cost-effectiveness potential of working on flooding, I briefly investigated various interventions. I tried to compile a comprehensive list of interventions from scanning the literature and then categorised and condensed these into the baskets presented below.[19] Within each intervention type I chose one more specific intervention to analyse in greater depth. This choice was based on a combination of which interventions I could find helpful data on, and which were subjectively the most promising to me. My cost-effectiveness analyses are available in a separate sheet, and I summarise some of the key findings and considerations below.

Intervention               Theory of Change Quality of evidence Cost-effectiveness Overall promisingness
Physical infrastructure Physical barrier to the movement of water → buildings less vulnerable Moderate 7x cash transfer Moderate
Land-use planning Avoid having essential infrastructure in flood-prone areas → community is better able to function despite a flood Moderate 1x cash transfer Low
Early warning systems Advanced knowledge of floods → fewer deaths, and moveable assets saved Strong 1.3x cash transfer Moderate
Emergency response Evacuate and treat affected people → fewer deaths and injuries Weak Not assessed Low
Long-term support Provide goods and services in the recovery phase → improve nutrition, mental health and financial stability Weak Not assessed Low

Physical infrastructure

These interventions construct physical barriers to the movement of water, provide less-harmful spaces for water to flow into, or otherwise make buildings less vulnerable to water. Options include:

  • Dams to control river flow
  • Dikes/levees to prevent a river from spilling its banks
  • Sea walls to prevent coastal flooding
  • Beach nourishment to bolster this natural barrier to coastal flooding
  • Mangroves, reefs, or artificial equivalents to slow water flow in a storm surge
  • Drainage systems to remove excess floodwater
  • Stockpiled sandbags or other temporary barriers ready to deploy at short notice

A Copenhagen Consensus report considered several physical infrastructure projects in the flood-prone city of Jakarta. The most cost-effective was found to be constructing a sea wall. I built my own CEA copying the (scant) data in the report, and found a cost-effectiveness 7x that of GiveWell’s estimate for GiveDirectly (‘cash’ hereafter). However, all things considered I think this is significantly too optimistic as:

  • The report used a figure of USD2B to build a sea wall, while various news articles suggest it could be closer to $40B.[20]

  • The report did not justify its (implied) claim that the sea wall would increase GDP per capita in Jakarta by 12.7%, and this number seems far too high to me.

Note, it feels poor to present a CEA that I do not myself particularly believe in, but I decided it was better to use numbers from a reputable report and then critique them than to fully invent numbers I have very little way to estimate.

How I could be wrong: the cost of different types of physical infrastructure could vary wildly, and I investigated one of the most expensive options, a sea wall. Possibly some interventions, especially those harnessing nature such as nudging mangroves or coral reefs to prosper, could be very cheap while still giving significant benefits. I did not see any good data or analysis to borrow from in the literature, but with more time these cheaper interventions would be good to investigate.

Land-use planning

Perhaps cheaper than large infrastructure projects to redirect massive volumes of water would be to shape human land use to be less affected by a flood. These interventions do not attempt to reduce the extent of a flood, but rather to minimise the impact on critical systems by a flood of given severity:

  • Building essential services (eg hospitals) in locations accessible during a flood
  • Building a water supply and sewage system resilient to flooding
  • Statutory building codes on minimum disaster-readiness of dwellings
  • Appropriate zoning of flood-prone areas (perhaps as parks, recreation facilities)
  • Pre-emptive relocation of dwellings from the most flood-prone areas
  • Compulsory flood insurance for flood-prone areas
  • Subsidise renovations and construction that are flood-resilient, e.g. elevated housing

Most of these interventions resist ready quantification in a CEA. Perhaps the most amenable is relocating people, as there is a clear cost - that of building new houses and destroying old ones - and relatively clear benefits - complete elimination of flood risk, assuming the relocation site is sufficiently high. Such an analysis was undertaken in De Risi et al (2018) using the Tanzanian city of Dar Es Salaam as a case study. I built my own CEA using some of their data, and other sources and assumptions, and found a cost-effectiveness of 1x cash. Some of the key uncertainties (these are bolded in each CEA) include:

  • The average annual economic losses from floods in Dar Es Salaam

    • To what extent voluntary community labour and recycled building materials scavenged from damaged houses lower reconstruction costs[21]

    • How much of the damage is directly to buildings versus intangible lost economic opportunities[22]

  • The second most important harm in my CEA after economic damage was mental health impacts. A meta-analysis found that after at least 6 months on average 11% of flood victims were suffering from post-traumatic stress disorder, which seems surprisingly and suspiciously high to me.

De Risi et al also studied an intervention called ‘green urban development’ which includes elements of both physical infrastructure and land-use planning interventions. This resulted in a cost-effectiveness of 1.5x cash.

How I could be wrong: governments are key in this domain, and interventions with roughly zero monetary cost like passing new legislation or regulations could still have substantial benefits. The question would then become how expensive is it to research, formulate and lobby for a policy solution, which I did not address here but could be the subject of future work.

Monitoring, early warning and communication systems

Some harms from floods are very difficult to prevent at short notice (eg crops being washed away) however some mitigating measures can be taken in the days and hours before a major flood. Unlike previous categories, for this system to work, each of these interventions must be implemented in concert:

  • Hydrology data collection (satellites, groundwater levels, flow rates, river depths etc)
  • Computer modelling to forecast fluid dynamics and flood scenarios
  • Mass communication to disseminate flood warnings and updates

While some of this infrastructure and capability is unique to floods and water-related disasters, much of it is useful for many natural disasters. As such, it makes more sense to model the costs and benefits of early warning systems for all disasters together. Because this is further outside the scope of my research, I relied more heavily on a World Bank working paper to construct my CEA. The paper considered a sweeping proposal to build early warning systems across 80 major countries in the developing world. I find this has a cost-effectiveness of 1.3x cash. Overall I think the cost estimate for constructing these systems is too high, and I am very unsure about the size of the benefits:

  • The paper looks at several case studies of countries building and using early warning systems and uses a figure of $10M per country per year.
    • Building a worldwide system could provide significant savings through economies of scale and shared infrastructure for neighbouring countries.
    • Philanthropists could build such a system in only one especially poor and disaster-prone region. This marginal cost-effectiveness of the first system could be higher than the average cost-effectiveness of building such a system worldwide.
  • The paper models the economic gains from improvements in weather-dependent industries such as agriculture, tourism and construction as being ~10x greater than gains from avoided destruction in disasters. I have not corroborated this or thought much about how accurate it is.

How I could be wrong: any longer-term modelling of the benefits of early warning systems should take into account how climate change will impact each of the disasters that the system warns against, and this could substantially increase the estimate of future benefits. The World Bank working paper I use does not explicitly take climate change into account in its modelling, though possibly underlying sources it used do. Further research could redo some of this modelling with better climate assumptions.

Emergency response

Emergency assistance could include:

  • Efficiently evacuating/rescuing anyone who needs to be
  • Having well-resourced evacuation centres people can go to
  • Providing medical care for those injured or infected in floods

I found very little information on which to construct a CEA, and I think this intervention is decidedly unpromising so did not proceed further. Emergency responses already receive over 90% of disaster-related development aid, so are significantly less neglected than preventative disaster risk reduction work.[23] I share the conventional EA view that disaster relief is generally less effective than preventative work.[24] Some of the motivation for this intuition is that evacuations and emergency medicine are expensive and logistically difficult, and most of the burdens of a flood would likely still exist despite this intervention.

How I could be wrong: the crisis period during a flood is when most of the deaths occur, so plausibly some intervention here could cheaply save lives without worrying about economic damage. I did not see data on the life-saving efficacy of emergency response, but further research could focus on this.

Long-term support

In the months following a major flood, many affected people will be especially vulnerable and in need of ongoing support, which could include:

  • Counselling, therapy, and peer support for mental health
  • Provision of clean water (if the water supply is polluted)
  • Provision of food (if farmland is significantly impacted)
  • Housing for internally displaced persons, especially if clean-up is long or old dwellings are unrecoverable

My guess is that long-term support is more promising than the emergency response, as it could be scaled and planned better. However, there is nothing special about floods for these interventions: this support is also valuable for people who lack food, clean water, shelter and mental health for reasons other than floods. Assessing the cost-effectiveness of these interventions well would require thinking about not just floods, but all factors that contribute to the deprivation of these necessities. For instance, clean water interventions may be cost-effective, but they should be assessed in their own light rather than through the lens of floods, and so are outside the scope of this report.

How I could be wrong: if very needy people are unusually geographically clumped after a flood, this could increase the cost-effectiveness of these interventions over just finding regular poor people to help.

Other interventions

Two of the most important ways to reduce future flood damage are to mitigate climate change, which will be an increasingly important driver of flood disasters, and to promote general economic development and growth, as richer people are more likely to invest in flood preparedness and will be less vulnerable to damage. Making climate change less bad and making people richer are very good things in themselves, but lowering flood damage is a small part of the benefit, so this report is an inappropriate place to delve into these macro-interventions.


Overall I think floods aren’t a promising area for EAs to devote much time, money or effort to. I think it is great that governments already have significant disaster preparedness and response systems in place, and I would endorse this being expanded. So plausibly EAs that have significant sway over government decision-makers and can convince them to invest more in flood defences should do this. This would be the case for almost any good policy though: if it is low-effort to convince the government to do it, you should.

I think the most promising intervention is early warning systems, especially as they are useful for disasters other than floods as well, so my generic reasoning about the (relatively) small total importance of flooding applies less. I did not do a deep dive into early warning systems as they transcend flooding, so I think the most valuable follow-up research to do would be on a more holistic assessment of these systems and their costs and value. My reasoning about powerful people standing to gain significantly from disaster preparedness still stands here, so it would be surprising if it were easy to significantly change early warning policy.

Finally, I would like to note that floods can be terrible, and I do not wish to belittle the suffering caused by them.


I received guidance and a grant from the Cause Innovation Bootcamp. Special thanks to Leonie Falk for her advice, with comments and feedback from Akhil Bansal, Teryn Mattox, Sarah Weiler, and Tom Delaney too. Views, and errors, are my own.

  1. Sources use different categorisations of floods, but all amount to something similar, see eg ↩︎

  2. ↩︎

  3. I think the best review paper was I also link to other reviews and primary literature in further footnotes. ↩︎

  4. ↩︎

  5. This is my main source for flood damage data. I did not dig into how reliable the data are. Many of the paers and reports I read used this source, and the consensus seems to be it is the best available. That said, it likely still misses many impacts that are not reported anywhere, such as deaths from indirect causes, and economic damage from wasted time. ↩︎

  6. ↩︎

  7. ↩︎

  8. Interestingly, one study found that a flood swept away stagnant pools of water and hence reduced mosquito breeding grounds, but this was the exception. ↩︎

  9. ↩︎

  10. ↩︎

  11. ↩︎

  12. ↩︎

  13. Of the twenty most damaging floods in 2001-20 in terms of GiveWell’s preferred units (log GDP/capita lost times affected population), all were in Asia (10 India, 5 Pakistan, 3 China, 1 Bangladesh, 1 Thailand). In contrast, seven of the twenty most damaging floods in raw dollar terms were in OECD countries, with a further eight in China. Data from I cannot reshare my sheet publicly though due to the terms of service. ↩︎

  14. This 10x factor lacks much justification. When I tried modelling other known health harms - malnutrition, mental health, disease - I found they were similar or less bad than raw deaths in DALY terms. My guess is that the less severe a harm is the more it is undercounted (few people will report minor cuts) so I feel comfortable with 10x as a guess. Premature deaths make up two-thirds of the global burden of disease (, so our (very weak) prior should be that it is unlikely for a health harm to be vastly dominated by non-death DALYs. ↩︎

  15. I did not find good global data ↩︎

  16.,%241.05%20trillion%20between%202010%2D2019. ↩︎

  17. ↩︎

  18. Note that we do not know how many deaths would occur without existing flood prevention measures. It seems likely that most of the best opportunities have already been taken, even if there were at some stage very cost-effective interventions to be had. ↩︎

  19. I use a modified version of the typology given in ↩︎

  20. See eg ↩︎

  21. This was not modelled in the paper, but after talking to the correpsonding author about it, I did include a term dividing reconstruction costs by four for this. ↩︎

  22. Again this was not modeled in the paper, I chose with limited justification a value of one third of the damage being direct destruction of buildings ↩︎

  23. Risk reduction expenditure is estimated at 5% by the Australian government and just 2% by a Copenhagen Consensus Report and a journal article ↩︎

  24. I could not find agood source for this, here is a web archived GiveWell page suggesting relief spending in Haiti may is not be the best way to spend philanthropic money ↩︎


Sorted by Click to highlight new comments since: Today at 10:24 PM

Overall, given the fact (?) that EAs generally believe in seeking out cause X, and are generally averse to publication bias, I'm surprised I don't see more content on the EA Forum that says something like "X is not a promising cause area"


Epistemic status should be 'no pun intended'...

I think you're wrong on policy tractability of some approaches. In some cases, policy for "appropriate zoning of flood-prone areas (perhaps as parks, recreation facilities)," and  "compulsory flood insurance for flood-prone areas" are relatively easy policies, and if there are places where EA has policy influence, these seem like potentially big wins for at least preventing additional exposure to flood risk going forward. (For the second idea, I think the key caveat is to do this for all new construction, rather than taxing current occupants.)

If people are interested in the topic of policy for reducing flood risk, please reach out. It's been years, but I did a bunch of work on this in NYC at RAND, both here, and for the Rockefeller foundation, but unfortunately the latter report was never released.  And the applications for developing countries weren't explored, but I think a small part of the work transfers.

Hmm yes policy work is tricky - probably even harder to model in a CEA than the more physical interventions I was mainly thinking about. I suppose this is what I was gesturing at with "So plausibly EAs that have significant sway over government decision-makers and can convince them to invest more in flood defences should do this. This would be the case for almost any good policy though: if it is low-effort to convince the government to do it, you should." But yes perhaps I did undersell the value of policy here. I think I mostly stand by my claim that if you are able to influence policy a lot you should probably focus on other things first. If as you say the policies needed for flooding are unusually tractable then yes that would change things.

Yeah, my only disagreement with you is that I think people who have policy influence should look at tractability and effectiveness across more domains than just ones you've identified as the most effective, since policy windows are fairly unpredictable, and in my experience, over-focusing on "your" issues often means you miss opportunities to get very impactful things done, while spinning your wheels on items that aren't currently possible.

Do you know Sela Nevo, working on Google's Flood Forecasting Initiative? I wonder if you'd change each other's minds on this

Thanks, no I had not come across this! I should say, I'm still excited to see this work and think it is great and very net positive. I just didn't find the quality of evidence and effect sizes I would want to see to make a good case to GiveWell or Charity Entrepreneurship or OpenPhil or anyone that they should really expand into this area. So in that sense I had a pretty high bar in mind. I think it is still compatible with my views on the relative promisingness of flooding work that someone particularly well suited to it would be best placed working on it. (And working at Google seems perfect for this - particularly given I am roughly indifferent to Google having more or less money so do not need to weigh the opportunity cost of their money being spent.) So if you're reading this Sela, good on you, I celebrate your work, feel free to criticise my piece here, I think I stand by my argumentation though should perhaps make clearer the audience I had in mind for this being 'not promising'.


Hey Oscar, I am indeed reading this! (albeit a bit late)

First, I really appreciate you looking into this and writing this. I'm excited to see people explore more cause areas and give rough estimates of how promising they are.

There are quite a few details in the cost-effectiveness analysis that I think are inaccurate, and I've mentioned a few at the bottom of the comment in case it is of interest. However, I still think this is a good and valuable shallow investigation. If there weren't discrepancies between the conclusions from 50 hours of investigation and 6 years of work by different people that would be quite surprising (and Google would have wasted quite a lot of money on external analyses, RCTs, etc.).

Details aside, I broadly agree with what you wrote in this comment. There's a big difference between statements about the promisingness of a field as a whole, vs. claiming that there aren't uniquely impactful opportunities in a field - and I think we have a uniquely impactful opportunity. I think flooding is an incredibly important problem and there are many organizations I would love to see investing more in it, but I would not make the case that this should be a top priority focus area for GiveWell / OpenPhil / CE.

It’s also worth noting explicitly that even if I had more meaningful disagreements with the conclusions of this investigation I’d still be very appreciative of it. These questions are complex, multiple perspectives are incredibly valuable, and even well-informed intelligent people can disagree.

Finally, just in case it is of interest and useful, I’ll note the largest drivers of where my estimates diverge from those in the report:

  • Total impact - the EMDAT is definitely a candidate for the most comprehensive source of recorded flood (and other natural disaster) deaths, but it’s not a great source for an estimate of total deaths. This is partially due to undercounting of deaths (and other estimates that use EMDAT alongside other sources find higher numbers[1]<sup><strong>,</strong></sup>[2]), but more significantly due to these types of counts only counting immediate deaths (due to reasons such as trauma) and not due to longer term health effects[3]. Tracking the relationship between floods and all-cause mortality (or other health indicators)[4] leads to vastly larger numbers. Both these methodologies are probably far from the true numbers, but in opposite directions. The World Bank report you linked to thinks warning systems can save 23,000 lives per year, and though this may be an overestimate of the number of lives that would be saved by that plan, I think it might indeed be a reasonable lower bound on the number of lives lost annually in total. My current best estimate of fatalities is about an order of magnitude higher than yours.

  • Comparison between top intervention to all harms vs. all investment - we both already commented on how unique opportunities might be more promising than the field as a whole. I want to make a related but broader point here, because I’m worried the methodology here will systematically bias us in favor of existing cause areas over new ones. At least in parts, your analysis asks how much funding it would take to address all flood harms globally. You then compare your estimate of the cost-effectiveness of that to GiveDirectly, one of the top interventions globally, which invests immensely in targeting[5] to maximize its cost-effectiveness. Due to diminishing returns, almost any domain you’d look into would be very cost-ineffective to try and completely solve the problem globally (for example, trying to solve all poverty via cash transfers would also be far less cost-effective than GiveDirectly’s current margin). Specifically in floods this can lead to a difference in orders of magnitude. While on average the cost-benefit ratio of early warning systems has been estimated as 1:9[6], but in highly-affected low income countries the cost-benefit ratio can rise to the hundreds[7]. Also, I won’t name names, but most $100M+ programs in this space that I’ve seen were never actually completed or used - so the average cost-effectiveness numbers in this space are very far from the effectiveness of a well-functioning organization doing good work. Concretely, I think that your statements might be more than an order of magnitude off if we’re considering investing in promising projects to mitigate flood harms (focusing on severely affected low and middle income countries, early warning, and organizations/solutions with a strong track record).

  • Note: my views are disproportionately influenced by work on early warning systems, which are only a part of the work you’ve aimed to review, and you noted towards monitoring, forecasting, and alerting, which are the areas I’ve been most involved in, while your report touched on other areas in flood management as well.

Finally, you might also be interested in a report from 2016 by CEA, which also includes a review of the cost-effectiveness of flood management. I think it misses different nuances but again provides another interesting perspective.

I have A LOT more things to say about the empirical statements, framework for evaluation, and assumptions that went into this - happy to chat if you’re interested.

  1. A digitized global flood inventory (1998–2008): compilation and preliminary results ↩︎

  2. The Human Impact of Floods: a Historical Review of Events 1980-2009 and Systematic Literature Review ↩︎

  3. Health Risks of Flood Disasters ↩︎

  4. Health Effects of Flooding in Rural Bangladesh ↩︎

  5. Study: AI targeting helped reach more of the poorest people in Togo ↩︎

  6. Global Commission on Adaptation's Adapt Now report ↩︎

  7. Background Paper on Assessment of the Economics of Early Warning Systems for Disaster Risk Reduction ↩︎

Hi Sela, thanks for the long and thoughtful comment, and for your kind words. That is reassuring that you also do not feel this is a key area for GiveWell/OP to expand into.

Really interesting re EMDAT possibly being off by ~10x, I was aware that longer-term harms are a lot harder to measure but wasn't expecting the effect to be that large.

Re my references to ending all flooding harms, that makes sense; I wasn't trying to suggest that the average cost-effectiveness would be the same as marginal cost-effectiveness. Perhaps a better thing to say would be that in order to be competitive with top charities, marginal targeted interventions would need to be far better than the average of existing interventions.

Hmm yes I was a bit surprised at how expensive EWS were made out to be, particularly when I would have thought a lot of costs could be saved by rolling out the same model and infrastructure across different countries.

Thanks for the offer, I am not currently working on this and don't expect to go back to it, so I don't think there would be much value in talking further - I'll let you know if I am coming back to this though. I hope you make great progress on your flood forecasting work!

This is a decent note that, by virtue of its "shallow" scope of inquiry, leaves out some interesting policy solutions that might be cost-effective.  This is especially true for policies involving flood prevention. It has been shown that different cities flood at different rates for the same amount of rainfall (recommendations for better studies and links welcome!). In developing countries, flooding is worsened by encroachments, loss of urban lakes, concretization of surfaces, etc. As a consequence, the open space that permits seepage of water into the earth reduces. If more open spaces were present, (a) more water would seep into the ground, reducing volume of flood water, and (b) water would slow down, causing less damage (speed of water is associated with flood damages).


Thus, potential EA partners might purchase large tracts of land in flood prone cities, greenify them, and make the surface more amenable to water seepage. This has strict parallels with provision of ecosystem services like planting trees or preserving wetlands. The CEA is relatively easy. Even when urban land is expensive, strategic placement of these tracts can potentially lower damages from flooding. Studies that causally link flooding and concretization of surfaces remain scarce, with natural variations being hard to find. So, it remains an active area of inquiry.


Quick Edit: Potentially more relevant for urban, rain-induced flooding.

I think floods often result in an increase in snakebites, due to both humans and  snakes seeking out the bits of land and houses that are above water. I'm linking this post about snakebites as a cause area, in case it's helpful.

Thanks, I hadn't read that post, very interesting! I did read in one of the review papers on flooding that there can be injuries from being confined with animals, but chose not to included it as I didn't see any data on it and my subjective impression was that this would not be as large an impact as the others (but this could be wrong, if anyone finds good info on the flood --> bites causal link I'd be keen to see it).

FYI, I added an edit at the end of the summary reflecting some of the discussion in the comments.

[comment deleted]5mo20