Brooks’ law in aid plans : is more always merrier? gain goes down the drain.

Definition

Brooks’ law on software development says : “adding manpower to a late software project makes it later”. Brooks adds to his law that “Nine women can not make a baby in one month”. Is there a similar law for development?

I could propose two options:

  1. If results are elusive, adding an extra partner to an existing  coordination mechanism only detracts attention more from getting results.
  2. Adding untied development money to an acute conflict area makes the conflict more chronic.

While option 2 is tempting, I fear it would be difficult to prove this option in a way it would stand scrutiny; Option 1 is an  open door I want to walk through.

Of course, as Brooks himself states, the law is an “outrageous simplification”, and further points to two main factors that explain why his law works all too often (wikipedia) :

  1. It takes some time for the people added to a project to become productive. Brooks calls this the “ramp up” time. Software projects are complex engineering endeavors, and new workers on the project must first become educated about the work that has preceded them; this education requires diverting resources already working on the project, temporarily diminishing their productivity while the new workers are not yet contributing meaningfully. Each new worker also needs to integrate with a team composed of multiple engineers who must educate the new worker in their area of expertise in the code base, day by day. In addition to reducing the contribution of experienced workers (because of the need to train), new workers may even have negative contributions – for example, if they introduce bugs that move the project further from completion.
  2. Communication overheads increase as the number of people increases. The number of different communication channels increases along with the square of the number of people; doubling the number of people results in four times as many different conversations. Everyone working on the same task needs to keep in sync, so as more people are added they spend more time trying to find out what everyone else is doing.

The picture is easy to recognize : the 2 or three main players in a development field know each other so well they don need to meet except for a beer. Indeed: it is a jungle out there, and the fight for funding is harsh: you must know your competitors. The odd outsider claims there is a need for coordination and what used to be a working, informal process must be reported on and formalised.
The Lilliput  partner brings his “important” expertise and money on board, but must be consulted in all important matters. Everybody gets bogged down in coordination meetings.

Lilliput partners at the table

Just look at a typical picture concerning humanitarian assistance in 2009 to a country in Africa, Niger (FTS).The total recorded humanitarian aid was  57.83 million USD:

  1. Central Emergency Response Fund: 20.23 %
  2. USA: 17.07 %
  3. ECHO + other Commission (Europe) 20.76 %
  4. Japan:  6.92 %
  5. Belgium:  5.27 %
  6. France:  3.61 %
  7. Less then 1 %: Germany, Sweden, Switserland, UK
  8. Unearmarked allocation WFP:  3.14 %
  9. Carry over : 19.76 %

There is a good case for involving donors 1 to 3 in a coordination effort from the start. But what is the added value for the humanitarian results to invite one of the group 4 to 6? Sure they might have some specialised insights, but shouldn’t they be invited more punctually? And why invite the donors grouped under 7? Their contributions are close to the margin of error in the calculations of the others.

On the other hand, donors 4 to 6 have a very strong case to want to be involved, as they can free ride on the work done by the strong partners: studies, needs assessments, feedback from the field. Pure gold for the free rider. We might hope the donors grouped under 7 wouldn’t even bother.
Taking into account all coordination costs, it is probably better to keep the contribution of the fringe donors outside than to have them inside.

The picture for data per “appealing organisation” is as interesting:

  1. World Food Programme: 41.93
  2. UNICEF 17.83
  3. MSF: France, Switserland, Luxemburg, etc.  9.50
  4. Save the Children – UK  7.14
  5. Mercy corps  4.97
  6. Food & Agriculture Organization of the UN  4.38
  7. WHO  4.15
  8. Cooperative League of the USA  2.40
  9. Concern Worldwide  1.33
  10. Action contre la faim  1.13
  11. Less then 1 %: Luxemburg red cross, Oxfam UK, Medicins du monde france, French Red Cross, care international, Africare,

In the area of direct food assistance, USD 24.1  million has been contributed, of which USD 22.8 million was going to WFP, some 94 %. It seems coordination on food assistance is just a waste of time. For the overall effectiveness of food aid, the internal management of WFP activities is much more relevant than the effects coordination could have. Why is WFP a reluctant UN-reformer? Perhaps because they want to have results with the group they are supposed to help, hungry people, and not just with any nobody that might show up in the field claiming they distribute food.

I am not saying that the small agencies or small NGOs are irrelevant: far from that: they are the ones innovating, reaching the victims most difficult to reach, challenging the effectiveness and efficiency standards of the juggernauts (and dinosaurs). A coordination effort draws them in the mush that general consensus is, with a risk of conforming to the mainstream. It might even be a prerequisite for coordination to comply with what is generally accepted.  To get a share of the pie distributed at the coordination table, you have to pay the dues.  They could end up selling their soul for being allowed to play with the big boys.

Political economy of institutional coordination

It is not safe to assume that participants in a coordination mechanism share a common development goal. Indeed, sometimes one organisation is interested to push local ownership, while the other one pushes its own valuable and unique expertise. Coordination is more and more seen as an objective to aim for, not just as a means for better delivery

In our example, food assistance to Niger, the goal might be shared.
The different actors have some leeway to take decisions in country, but if push comes to shove, they are executing what their headquarters decide, sometimes in detail, sometimes only the general principles.

At the coordination table the organisations will do what they are supposed to do:

  • Pushing the mandate: Each NGO has a different political view on how society should work. This view will be different from the view of an international agency, guided by the General Assembly. Even within the UN, you might have a different focus between UNICEF and WFP on how to proceed. The goal is more to “trick” the others to your line of thought, as each partner has little authority to divert from their own mandate.
  • Dividing the loot: Coordination mechanisms are all about deciding what is important, and giving a place to each partner. In countries where there is an Emergency Response Fund or a pooled fund, the importance of getting the money becomes a matter of life and death for the organisation. If your specialised expertise is recognised as crucial to the strategy, it is like single sourcing a project.
  • There is no leadership that can enforce any compliance. You can say what you want, you don’t have to comply, the only pressure is peer pressure. This can be used to occupy a field before you are sure to have the capacity to deliver.

Moreover, the individual participants will have their own private agendas, priorities. Some might want to stay in the country for another year, others just want to leave. Alliances between institutions and friends are struck.

All these elements will be at play and deflect attention away from getting the results for the beneficiaries. In normal circumstances, the moral fibre of the humanitarian or development workers and institutions will limit the damage. With more partners, the risk to have somebody at the table who speaks only out of organisational self interest increases.

Cost of communication

In a typical humanitarian crisis, the UN classifies the aid in 11 sectors. Most organisations are active in more than one sector, so they must go to the different meetings. Agencies with a wide scope complain that up to 30 % of their in country management time is spent in coordination meetings (no hard data, no impact nor efficiency evaluations I know of). Smaller organisations could just stop working.  In an important humanitarian crisis, the number of international actors in a well funded crisis is important:  22 in Niger, more than 100 in the DRC. The number of permutations becomes infinite.
Jean-Michel Severino and Olivier Ray present this equation in a telling graph in their publication “The End of ODA (II): The Birth of Hypercollective Action”:

Conclusion

Should we coordinate?  Of course we must coordinate, but the objective must be more efficient aid, not coordination in itself. I would beg for more attention to levelling the playing field for all actors in the market by opening up all information, instead of trying to coordinate with everybody. Standardising information and have a transparent system for sharing is necessary for accountability purposes anyway, and could fulfil a lot of the needs that people wish to satisfy with coordination.  On the push for coordination at all costs, II would like to quote W.  Easterly in his immortal piece “Tower of Babble”:

“Coordination” and “partnership” are the equivalent in foreign aid of U.N. resolutions for world peace. Every different national donor, U.S. government bureaucracy, or private business has its own agenda, will not voluntarily sacrifice its own interests for some other organization, and there are no binding contracts to enforce any such sacrifices.

It seems Brooks’ law for development, including the caveat (“it is a gross simplification”) stands for development:

If developments results are elusive, adding an extra partner to an existing  coordination mechanism only detracts attention from getting results.

Moreover, the explanations stand in principle. In development there is also a third clause:

  1. It takes some time for the partner added to a coordination to become productive. This is is the “ramp up” time. Development programmes are complex endeavors, and new partners must first become educated about the work that has preceded them; this education requires diverting resources already working on the programme, temporarily diminishing their productivity while the new partners are not yet contributing meaningfully. Each new partner also needs to integrate with a team composed of multiple institutions who must educate the new partner  in their area of expertise in the sector, day by day. In addition to reducing the contribution of experienced partners (because of the need to train), new partners may even have negative contributions – for example, if they want to divert funding to their own organisations, or have a different viewpoint on development that move the programme further from completion.
  2. Communication overheads increase as the number of people increases. The number of different communication channels increases along with the square of the number of people; doubling the number of people results in four times as many different conversations. Everyone working on the same task needs to keep in sync, so as more people are added they spend more time trying to find out what everyone else is doing.
  3. The expected contribution of a partner to the totality – as much in expertise as in financial contributions, can be negligible, increasing the risk that point 1 and 2 outweigh the benefits of the extra partner.
This entry was posted in development, humanitarian and tagged , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *