Chapter 5: Do No Harm After a Disaster¶

Colombo, June 21, 2005. Six months after the tsunami of December 26, 2004, which killed more than 31,000 people in Sri Lanka and left nearly a million homeless, the Food and Agriculture Organization of the United Nations (FAO) and Sri Lanka's Ministry of Fisheries issue a joint warning. According to government estimates, 54% of the country's fishing fleet was destroyed or made unusable by the wave. Since then, boat donations have poured in. The statement lists the donors: local and international NGOs, aid agencies, companies, groups, individuals from the country and abroad, in unprecedented numbers.

Data from the coordination mechanism the FAO and the ministry set up suggest that in seven coastal districts, more boats are already being replaced than were lost. The statement does the math: if all the pledges of NGOs and donors are honored, coastal fishing could end up with more than 2,000 traditional craft and 2,200 fiberglass boats more than before the tsunami.

Yet, the director general of the Department of Fisheries points out, the fleet was already too large before the disaster. An FAO-supported study in 2003 showed that, outside the north and east of the country, coastal fishing output had gone from 130,000 tons in 1989 to 142,000 tons in 2002, while the number of fishermen nearly doubled. Fishermen were already using nets with ever finer mesh to catch ever scarcer fish.

The statement asks all actors to join the coordination mechanism. It notes that a few organizations have already sharply cut back their donations of new boats, and it names two: Sewa Lanka and the British Red Cross.

Nobody in this story meant to do harm. Every boat was given so that a fisherman could go back to sea.

What the scene teaches¶

What came next is documented by the FAO itself. In November 2006, in the review it presents to its Committee on Fisheries, it writes that "the avalanche of well-intentioned, but not always well targeted, assistance" from NGOs to replace boats and fishing gear sometimes increased fishing capacity, with too many small boats, often poorly built and unsafe. It points out that many skilled boat builders were among the tsunami's victims. And it adds that it is ironic that so many fishermen, having survived the tsunami, must go to sea in boats that put their lives at risk.

The figures that follow are of different kinds, and they must be read as such:

in participatory workshops reported by the FAO in 2007, fishermen in Galle estimate the boat surplus at between 77 and 315%, those in Hambantota at between 50 and 275%: these are perceptions put into numbers, not a census;
Sri Lanka's public reconstruction agency estimated in 2007, according to the humanitarian news agency IRIN, that more than 3,000 small craft had been donated in excess;
an FAO survey, cited in the same dispatch, found that 18.9% of new boats were not seaworthy, that 63% of replaced or repaired boats had no fishing gear, and that more than 7,100 of the 13,000 boats distributed by early 2006 were in the hands of people who did not own a boat before the tsunami.

The FAO is also self-critical. The emphasis it placed everywhere on physical assets, it writes, gave a visible sign of its presence, but at the expense of technical assistance. And it states the principle that would have avoided a good part of the problem: "fishers generally know what type of boat and gear they should use."

This story deserves a fair reading. The FAO acknowledges that it cannot clearly establish the link between this surplus and overfishing, in a fishery that was already overexploited. The synthesis of the Tsunami Evaluation Coalition (2006) notes that in Sri Lanka more than 80% of damaged fish markets, boats, and fishing equipment were quickly restored. But it draws the overall lesson bluntly: "The concentration on distribution of assets, especially boats, demonstrated a failure to understand and support diversified and sustainable livelihoods and communities." At best, it writes, the international response restored the situation that existed before; at worst, it strengthened the better-off, such as fishermen who owned boats, and marginalized those with few assets, notably women and the poorest.

The ministry eventually responded: its April 2006 strategy entrusted boat replacement to a single agency, on the basis of entitlement certificates issued to fishermen who had lost theirs, without exceeding the pre-tsunami number. Chapter 12 returns to this coordination mechanism.

Three lessons for your club.

Aid is measured against what existed before. A donated boat is useful only in relation to the number of boats before, the number of fishermen, the fish stock, the nets, the boat builders, the fish buyers. Nobody can judge a donation in isolation.

Affected people know what they need. The FAO's sentence about fishermen applies to shopkeepers, artisans, and farmers. A club that chooses the item to give in place of the people who will work with it runs the risk of choosing the wrong item.

The harm is not limited to the market. Poorly targeted aid also creates winners and losers. Boat-owning fishermen received boats; fish sellers, day laborers and women were forgotten.

The principle: all aid touches a market¶

What the standards say¶

The Minimum Economic Recovery Standards (MERS), published by the SEEP Network in their third edition in 2017 and recognized as a companion standard to the Sphere Handbook, set out, in the very first standard of their chapter on enterprise and markets, an instruction your club can post on the wall: "Recognize that all relief affects markets."

They draw three rules from it:

Work through existing markets. "Utilize existing market mechanisms, unless there is a compelling reason not to."
Do not take the place of local actors. The MERS define market distortions as any unintended effect that harms a market, from extreme price swings to the physical destruction of a market, and they specify that it is up to those who intervene to make sure their action does not replace local products and local actors.
Do not become a market actor. "Implementers should not take on market functions": whoever helps must not become the seller, the lender, or the supplier.

Their fourth core standard is simply titled "Do no harm": the design and implementation of economic recovery actions must avoid or reduce possible harm, and must not worsen economic inequality, conflict, or risks to people.

The Sphere Handbook (2018) restates the same requirement through the Core Humanitarian Standard. It asks that programs promote an early recovery that benefits the local economy, that they examine market conditions before choosing between cash, vouchers, and in-kind aid, that they "Buy goods and services locally when possible," and that they reduce their negative effects on the market as much as possible. For agricultural livelihoods, Sphere asks that cash or vouchers be preferred where markets function or can be helped to recover.

The 2016 note of the Global Cluster for Early Recovery says it in one sentence: "External assistance is never neutral, and becomes part of the context in which it is delivered."

Where the phrase "do no harm" comes from¶

The Do No Harm framework was designed for armed conflict by CDA Collaborative Learning Projects. Its 2004 handbook presents itself as a descriptive tool, not a prescription. It shows that aid acts through its transfers of resources, and it distinguishes several effects, including market effects: "Assistance affects prices, wages, and profits." The MERS carry this framework over to crises in general. Your club does not need to apply the full method, designed for war zones. It keeps the starting question: whom does my aid benefit, and whom does it harm without meaning to?

What Rotary says¶

The Rotary Foundation says the same thing in its own texts. The Global Grants terms in force in July 2026 require that projects "cause no harm." Its disaster policy, written into the Foundation's Code by a decision of August 2018, notes that the immediate need of affected districts is generally cash, and adds: "The Foundation discourages donations of goods, unless requested by a local committee."

The Foundation's community economic development guidelines, in their April 2024 version, go further. Distribution of items such as wheelchairs or sewing machines must remain a small share of a project's budget, and you must explain how the community took part in choosing the items. "Projects that only seek to donate equipment" are not eligible.

What in-kind aid does to markets¶

Port-au-Prince rice¶

After the earthquake of January 12, 2010, in Haiti, an Emergency Market Mapping and Analysis (EMMA) was carried out from February 5 to 12 by a team from several organizations, including Oxfam GB and the International Rescue Committee. It describes the rice chain before the earthquake: 6 importers in Port-au-Prince brought in about 70% of imported rice; they sold to 8 large wholesalers, who handled 1,000 bags a day, then to about 200 small wholesalers at 100 bags a day, who supplied 10,000 market women and small retailers in the capital.

The hardest-hit link is not the one you would expect. It is the small wholesalers: 80% of them lost their storage space and cannot get back into the market, for lack of capital or credit. The study adds, conditionally, that the announcement of a massive rice distribution by the agencies may have discouraged importers from resuming their trade.

Its recommendations changed the response. Whereas the World Food Programme (WFP) at first had to import rice massively, bypassing local channels, the analysis showed that after the initial shock it was possible to work through local importers, and that working with markets rather than alongside them made cash preferable to in-kind distributions. WFP removed beans from its ration after the local harvest, so that families would buy Haitian beans. The MERS cite another action in Haiti that points the same way: repairing small storage depots allowed wholesalers to buy in bulk again, lowered the price of staple foods for poor consumers, and raised the wholesalers' profits.

Remember the move: the link to help was not the most visible one. It was the small wholesaler without a warehouse.

What the standards tell¶

The MERS report several examples without naming a country, which should be read as typical cases:

a government widely distributes free seed after a crisis; seed prices collapse in the shops, the incomes of producers, importers, wholesalers, transporters, and retailers fall, and importers order less for the following season, so that new drought-resistant varieties become scarce;
after an earthquake, free agricultural inputs weaken the farmers and traders outside the affected area who used to sell them, because the market analysis had stopped at the edge of the area;
in contrast, a program buys hoes and wheelbarrows from a local wholesaler to replace tools lost in a flood: the order gives the wholesaler the cash flow needed to restock, and all the farmers in the area benefit.

The same item, bought locally or imported, can revive a market or smother it. The MERS sum it up: a wide distribution of items bought outside can weaken local suppliers of those same products.

One nuance, which the researchers themselves make: the effect of food aid on local prices is debated in the literature. A synthesis by the Overseas Development Institute (2007) notes that it can lower prices, which helps poor buyers and hurts producers. So do not say that all in-kind aid drives prices down. Say that it can, and check.

Chapter 3 told how the clubs of Romagna swamped hospitals with donations those hospitals had not asked for. The mechanism is the same: giving what you imagine is useful before knowing what is missing.

Cash rather than goods¶

A change of scale¶

Where markets function, cash has become the default in humanitarian aid, no longer the exception. According to the CALP Network's report on the state of cash transfers worldwide, 2023 edition, cash and voucher assistance accounted in 2022 for about $10 billion of programming, of which 7.9 billion was transferred to affected people, or about 21% of international humanitarian aid (preliminary data at that date).

In 2015, the High Level Panel on Humanitarian Cash Transfers, convened by the Overseas Development Institute and the Center for Global Development, framed the question every organization should ask: "Why not cash? If not now, when?" The Panel concludes that in most contexts, local markets responded to cash injections without inflation, with positive effects on local economies. The Red Cross and Red Crescent guidelines (2007) already presented the unconditional cash transfer as the default option when goods are available in markets, the risk of inflation is low, and security allows it.

The strongest evidence comes from a development setting, not a disaster: in Kenya, a randomized experiment paid about $1,000 to more than 10,500 poor households in 653 villages, a shock larger than 15% of local output; measured average inflation was 0.1%. It shows that functioning markets absorb a lot of money without overheating. It does not show what happens when the warehouses are destroyed.

The limits of cash¶

The Panel itself acknowledges it: giving cash is not always the best option. When markets are too weak or supply cannot keep up, cash is not appropriate and can push prices up. It adds that these situations are rare and often temporary, and that they must not serve as an excuse to continue in-kind aid long after they have ended.

This is where the assessment from Chapter 3 comes in. The question put to shopkeepers about their ability to supply two or three times more if demand returned tells you whether the cash will find goods to buy.

Symmetrical risks¶

The Market Analysis Guidance of the International Committee of the Red Cross and the International Federation (MAG, 2014) gives the most useful rule in this chapter. The larger the intervention relative to the local economy, the more a cash transfer risks causing inflation, and the more in-kind aid risks causing deflation. Good local availability reduces the inflation risk of cash, but increases the deflation risk of in-kind aid. And the guidance warns that no threshold allows you to predict it.

In other words, cash is not risk-free: its risks are simply on the other side.

Poorly calibrated cash also does harm¶

Cash for work wages¶

The MERS report that after the Indian Ocean tsunami, many cash for work programs paid daily wages well above what farmers earned in their usual work. Many households stopped farming or producing to take advantage of them, at the expense of local production. After Cyclone Nargis in Myanmar, wages set above the market to speed up reconstruction prevented farmers from hiring for the harvest. In Sri Lanka, according to a 2007 synthesis, farmers reportedly bought machines to replace the workers who had left for these programs.

The MERS also give the remedy, with the example of Haiti in 2010: rather than each setting its own rate, the agencies established with the government a local daily wage that did not draw workers away from their usual jobs while still covering the needs of the hardest hit, then adjusted it by area. Chapter 12 will tell how, in the Philippines after Typhoon Haiyan, rates aligned across organizations protected regular employment, and what happened with those that paid much more.

Cut-rate credit¶

After the tsunami, a Sri Lankan microfinance institution in the districts of Galle, Matara, and Hambantota, recapitalized by donors, had to lend to victims at a subsidized rate of 6%, when its nominal annual rate was around 37%. An econometric study published in 2012 then observes an abnormal rise in defaults among borrowers who had not been affected. The authors put forward two possible explanations, strategic default or contagion, and the hypothesis that the subsidy may have pushed some borrowers to declare themselves victims to get better terms. It is one institution, and a hypothesis; but it shows that a generous rate decided from outside can damage the repayment discipline the lender will need after the donors leave.

The MERS draw two rules from this: subsidies are not sustainable, they must be time-bound and used selectively; and financial products must not create over-indebtedness. Chapter 6 details what your club can do, and not do, with a lender.

The gift that weakens people's own effort¶

The MERS add a subtler warning. If households and businesses are already investing on their own to replace their assets, it is better to complement their efforts than to provide assets directly, which could weaken their own coping strategies. People often restart before aid arrives (Chapter 1). Aid that comes afterward must build on that momentum, not replace it.

The recipient's contribution¶

An asset is worth more to the person who receives it when they feel like its owner rather than the recipient of a gift. The MERS therefore recommend asking for a co-contribution: a share of the cost, a contribution in kind or in labor, or repayment spread over time. A fifty-fifty split is a common starting point, but it must be adjusted to risk: the lower the risk the recipient bears, the higher their contribution can be. And the contribution must not commit them beyond their means.

The MERS tell of an example in northern Uganda: six community warehouses awarded through a competition, with the group's contribution simply making the application "more interesting" to the judges. One winning group told the evaluators that most residents knew only about the NGO's technical support, not its funding: "We told them that we did it ourselves."

One Foundation rule to keep in mind: under the July 2026 terms, no funds should ever be collected, in a global grant, from the people who benefit from the funded activities, and no cash contribution can come from them. If your project plans a contribution from recipients, in the form of labor, in-kind input, or repayment to a partner, check with the Foundation's regional grants officer which form is allowed.

Track prices¶

Aid that touches a market is monitored in that market. The MERS call for regularly tracking the prices of key goods, not just those being distributed or purchased, so a program can be stopped or adjusted as soon as a sign of inflation appears.

The MARKit price monitoring guide, published in 2015, offers a method within reach of a club:

in each market, choose three to five traders, always the same ones, taking care not to wear them out;
record prices every week, or at least every month, on the same day and at the same time;
note the most common price, not the average, and weigh the local unit yourself;
look into it if the price moves by more than 7.5% from one week to the next, 15% over two weeks, or 30% from one month to the next.

These thresholds are starting points suggested by the guide, not standards. A weekly record of a few products, kept by a club member in a spreadsheet, is often enough to see the problem coming before it costs someone dearly.

What the club does: the five-question filter¶

Everything above comes down to a filter your club will apply to every project idea, before committing its money or its name. It is a proposal of this book, which takes the CED guide's filter for normal times and adapts it to recovery.

Question	What you check	Warning sign
Are we competing with a local seller, trainer, or lender?	who supplied it before, who supplies it today	the club gives what a shopkeeper sells
Are we creating debt people cannot carry?	amount, rate, repayment capacity	subsidized loan outside a professional lender
Who decides: affected people or the club?	who asked, who chose the item	the club chose before asking
What happens when we leave?	who maintains, repairs, takes over the activity	no one to take over, no end date
Who is excluded, and who could capture the aid?	women, informal sector, non-owners, the poorest	recipient list not published, no complaint channel

For each idea, the team concludes in one word: go, redesign, or drop. An idea that fails a question is not necessarily abandoned: often, it changes form. The boat donation becomes a purchase from local boat builders, based on the list of fishermen who lost one; the food distribution becomes vouchers at neighborhood grocers; the club's zero-interest loan becomes a credit line entrusted to an institution that knows its clients.

The reflexes that go with it¶

Before buying anything, ask who sold it here before the disaster. Then ask whether they can sell it again.
Buy locally, even at a slightly higher price, if the market can supply without prices rising for everyone else.
Go through those whose trade it is: a payment provider for cash, a microfinance institution for credit, shopkeepers for vouchers. The MERS ask that you not recreate a payment mechanism when an institution or a market actor already knows how to do it.
Align your amounts and wages with those agreed in local coordination (Chapter 12).
Make selection criteria visible and open a complaint channel, as the MERS require.
Check the Foundation's rules before promising money: a global grant does not fund unrestricted cash donations to a person or an organization (July 2026 terms). If your project involves cash transfers, entrust them to a partner whose trade it is, and check with the regional grants officer what the Foundation can fund.

The pitfalls¶

Pitfall	Why it happens	Remedy
Collecting goods in the club's country and shipping them	it is concrete and rallies people	buy locally or use cash
Giving everyone the same item	"equal" seems fair	start from each person's trade; the MERS note that equal is not always fair
Choosing the item for the recipients	the club thinks it knows	ask those who will work with it
Paying more than the local rate	wanting to be generous	align wages through coordination
Lending at a cut rate	wanting to help fast	go through a lender, fund what it cannot fund alone
Not tracking prices	thinking the aid is too small	three to five traders, every week
Forgetting suppliers outside the area	the assessment stops at the disaster's edge	follow the chain back to its suppliers
Counting only owners	they are visible and organized	ask market women, day laborers, the informal sector

What your club now has in hand¶

At the end of this work, your club has its "do no harm" filter: five questions, applied in writing to every project idea, with the decision that comes out of it (go, redesign, drop), and a price record for projects that inject cash or goods. This is the third piece of the recovery file. It conditions all the ones that follow: none of the levers in Chapters 6 to 9 should be activated without passing through it.

The boats of 2005 were given by people who wanted to do good, to communities that needed them. What was missing was one question asked before the donation: how many were there before, and what do the fishermen say? It remains to see how cash, the first of the four levers, can restart an economy without putting it in debt: that is the subject of the next chapter.