Aid for countries after a disaster is rooted in our best impulses, but the way we provide it urgently needs to be reformed. A new CGD framing paper and working group are focused on fixing emergency aid using insurance financing models to save lives, time, and money. Setting out the problems facing emergency response in a recent paper, Payouts for Perils: Why Disaster Aid is Broken, and How Catastrophe Insurance Can Help to Fix It, my colleague Owen Barder and I find that:

  • We spend too little on resilience against future disasters
  • Aid shows up too late, raising costs and undermining incentives to do better
  • Calls for reform are met with “too bad” because the poorest bear the greatest costs
  • This is a problem that we can fix. Insurance financing will save lives, save money, and save time.

We wait for crises to develop before asking for funding from donors because humanitarian aid only pays out after things go wrong. This delay ramps up the human and financial costs of tackling disasters. When money does arrive, it is often delayed and trickles in piecemeal.

Take Haiti, a country that’s slipped below our newspaper folds. After the magnitude 7 earthquake struck in 2010, the international community responded with great sympathy. Then a pledging conference was delayed for three months. Agencies struggled to work with each other, or with the government. Only half the money pledged for 2010 and 2011 actually arrived.

By 2013, analysis by CGD concluded that nearly all the $6 billion spent to help Haitians was spent on NGOs and contractors, with little transparency. The Government of Haiti—left to manage when others scaled back or left—got just 1 percent of the humanitarian aid and a fifth of the longer-term relief aid.

Haiti is a specific case of a general problem, one that’s threatening our ability to respond to humanitarian emergencies globally. But this dysfunctional system endures because the costs of failure are borne by the poorest and most vulnerable.

How can we do better? Our paper argues that insurance and insurance-type contracts that create incentives for risk reduction and pay out quickly are the key missing market in disaster response. Because CGD’s focus is on taking ideas to action, we have convened a new high-level working group to help scale insurance financing for disaster aid, to build a watertight public policy case, and to catalyze actual transactions for vulnerable governments and frontline agencies.

There are thorny policy and economic questions to work through: can we guarantee governments get the right contracts, at fair prices? Can we combine better insurance with better support for frontline agencies? But the case for innovation is now unanswerable. The financial tools are now available. And the pressure on overstretched emergency budgets is growing incredibly quickly.

Disaster risk reduction: too little

A vanishingly small level of aid has been spent on reducing the costs of future disasters, tagged as ‘prevention and preparedness’: just 40 cents in every $100 (real USD).

Notes: Data on disaster-related aid (only available until 2010) is from AidData. CGD Analysis.


Of course, this is a bit specious. Donors can improve ‘resilience’ without labeling spending as such: building schools to tougher construction codes might be flagged as spending on education, rather than ‘prevention and preparedness.’ But even if this figure were off by a factor of 20—1,900 percent wrong—that would get us to just $8 in every $100 of aid. That’s still  far too little compared to the high returns from investing in resilience: enforcing construction codes carries an annual cost, but recovery from an earthquake that destroys infrastructure and housing stock takes a generation.


Humanitarian aid: too late

The chart below compares agencies’ requests for funding through humanitarian-response plans. Underinvestment in resilience and increasing costs due to late response show up as a rising deficit, as calls on donors’ humanitarian budgets go unmet. Since response plans are filed after crises develop, funding is late almost by design. And it arrives in the straitjacket of annual disbursements, despite the multi-year nature of many crises (last year, just 13 out of 35 response plans were for more than a year). 

Notes: Data from UN OCHA Financial Tracking Service (FTS). CGD Analysis.


Response plans are not the entirety of humanitarian aid. But the deficit they face is, like example of Haiti above, a specific example of a broader problem. Relying on ex-post aid creates three damaging distortions.

  • Donors decide how much to give, undermining planning and preparedness (Why set out who will do what in an emergency if you don’t know how much money you will have to pay for it?)
  • Since these appeals get just 50 percent funded on average, organisations have every rational incentive to inflate needs (If you received $1 for every $2 requested, how much would you ask for?)
  • Because aid only appears when disasters occur, it undermines incentives to invest in resilience.

The consequences: “too bad”

From a financial perspective, disasters appear to have been kind to developing countries. That makes sense: highways in Tokyo, for example, cost more than roads in Sri Lanka. But the costs in terms of human lives are dramatically higher in developing countries. That makes humanitarian emergencies and natural disasters highly regressive—their toll falls disproportionately on the poorest and most vulnerable. 

Notes: Data from Em-Dat. ‘Human capital loss’ estimated as number of deaths from Em-Dat times difference between life expectancy and median age in country times real income per capita. CGD analysis.


Insurance works when aid doesn’t

Not all the aid to Haiti was delayed, fragmented, or incomplete. In 2007, the World Bank worked with governments across the Caribbean to set up a novel insurance pool. Haiti’s $8 million dollar payout was available and deposited in government accounts less than 24 hours after the earthquake—a speedier response than the IMF, donors, or even the World Bank itself..If that liquidity were combined with better planning, it would have accelerated the first response, saving lives in the critical window of hours or days after the earthquake struck.

Where ex-post aid does not provide enough money, insurance pays out in full and on time. Where ex-post aid is mismatched—providing late funding in the straitjacket of annual disbursement cycles—insurance contracts can be matched to needs and provide flexible funding.

And while ex-post aid is distortionary, undermining incentives to invest in resilience, insurance contracts can be designed to increase the returns to those investments. For example, we might make donor support for premiums contingent on seeing investments in lower disaster costs by enforcing construction codes, or by setting up clear emergency plans.

From ideas to action

In addition to the Caribbean risk pool that paid out to Haiti, newer programmes in the Pacific and in sub-Saharan Africa show that insurance financing for disasters works well, that there is clear appetite from vulnerable governments, and, that donors want to scale up access to insurance and preparedness because they, too, bear the costs of half-baked response.

CGD’s focus is putting ideas into action. CGD’s new Payouts for Perils working group brings together senior representatives from frontline agencies and governments, donor agencies, and the insurance sector.Tackling disasters more effectively won’t be easy. But the existing system is overstretched, and prone to catastrophic failure: now’s the time for disruption and innovation. We’re excited to see how Payouts for Perils will help donors, frontline agencies, and governments do much more with much less.