The Commitment to Development Index: Ranking the Rich on Helping the Poor

Date July 3, 2007
Speaker David ROODMAN(Research Fellow, Center for Global Development)
Commentator SAWADA Yasuyuki(Faculty Fellow, RIETI / Associate Professor, Graduate School of Economics, Faculty of Economics, the University of Tokyo)
Moderator KIMURA Hidemi(Fellow, RIETI)


The Center for Global Development (CGD) is focused on how rich countries influence developing countries. We are interested in doing serious research and in getting our messages out. The Commitment to Development Index (CDI) was developed early in our history and embodies a lot of what we are about. In the first year we commissioned background papers on each of the areas we thought should be part of the Index, then had experts review our work. We circulated the initial the draft at conferences, got a lot of feedback, and published the Index in Foreign Policy magazine in 2003.

The Index is adjusted for size, so a small country can do well against a large country. We are attempting the difficult task of both getting across simple ideas about development as well as trying to remain rigorous in our analyses. Measuring exactly how the policies of rich countries affect poor countries really is impossible, yet it is worth pursuing as it does educate the public.

I would like to explain each of the following CDI components and how countries were rated in our results: foreign aid, trade barriers, policies that affect overseas investment, migration policy, environmental policy, military and security affairs, and technology policy.

Looking first at the aid index, aid comes in many forms: grants, loans, technical assistance, food aid, etc. Standard comparisons look only at quantity: net ODA as share of GDP. The aid component of the index uses that as a starting point and has three adjustments to reflect aid quality: a penalty if aid is tied; a weight for the selectivity of countries - with a higher score for aid that goes to poor but well-governed countries; and a penalty for project proliferation, due to the administrative burden that is put on recipient countries if the aid is in the form of many small projects rather than just granted in a large amount for general use. We also look at private giving to the extent that it can be credited to fiscal policy. The results show the Scandinavian countries at the top. Japan is at the bottom as the quantity of aid divided by GDP, with the interest received from recipient countries netted out, is low.

Secondly, trade: NGOs have made us more aware of the trade issue. In the trade component, we look at the complicated trade barriers that are in place - tariffs, quotas, quantitative limits, farm subsidies, etc. In place of these highly complex barriers, we ask, if there were a simple, across-the board tariff on all imports, how high would it have to be to have the equivalent impact on developing countries as the actual barriers. In addition to this there is some direct measurement of imports from developing countries.

I should explain that the results on each component in the Index are scaled so that a score of 5 is average, therefore meaning that negative scores are possible. On trade, if a country has no barriers, it gets a perfect 10. If it is average, it gets a 5. If it is twice as bad as average it gets a 0. And even higher barriers can get negative scores. In terms of trade, out of the 21 countries in the Index - all members of the Development Assistance Committee (DAC) - 14 are in the European Union and therefore have similar tariffs and subsidy policy under the Common Agricultural Policy (CAP); these countries are in the middle of the list. The countries ranked highly on trade are New Zealand, the U.S., Canada, and Australia, and at the bottom are Switzerland, Norway, and Japan, due to their high farm protection. For Japan, the issue is prohibitive tariffs on rice equivalent to 800%-900%.

Next is investment. We know that foreign investment can be very valuable for economic development. On the other hand, there can be negative impacts, as seen in the developed nations' herd mentality during the Asian financial crisis, labor rights abuses, environmental abuses in American or Japanese factories, etc. In order to quantify developed countries' investment policies we use a checklist that looks at their policies with regard to political risk insurance, prevention of double taxation of profits in sending and receiving countries, active participation in international efforts to prevent corruption, etc. The old financial centers of the UK and the Netherlands score highly. Ireland is at the bottom and Japan is below average, perhaps because it has a history of lobbying developing countries' governments to not enforce labor rules in Japanese factories.

Migration is also complicated as it may have both positive and negative effects: positive in terms of the transferal of skills and the fact that migrant workers send money home, negative in terms of brain-drain. We take the perspective that openness is generally a good thing, and the Index rewards immigration from developing countries, especially of unskilled labor; high percentages of foreign students from developing countries (which Japan scores highly on); and support for asylum seekers and refugees. Austria and Switzerland score highly in this Index. They accepted a lot of immigrants from the former Yugoslavia as it dissolved. Japan once again scores badly because it is difficult for workers from poorer countries to come and work here.

The environment component looks at what rich countries are doing to the global environment that will affect developing countries. Climate change plays a large role in this Index so countries are rewarded if they have signed the Kyoto Protocol, have low greenhouse gas emissions per person, the ratio of their emissions per GDP is being reduced quickly, and if gasoline is taxed highly. We penalize subsidies for the fishing industry, high imports of tropical timber, and more.

The countries at the top in this component, such as the UK and the Netherlands, have been aggressive in pushing renewable energy and raising gasoline taxes and have reduced their greenhouse gas emissions even as their economies have grown. The U.S. is in last place, mostly due to high greenhouse gas emissions and low tax on gasoline. Japan loses some points due to fishing subsidies and imports of tropical timber.

Security behavior in rich countries affects developing countries. In terms of the Index, rewards are given for contributions of money and personnel to both U.N. and other internationally sanctioned peacekeeping missions, as well as for protection of sea lanes for trade. Countries are penalized for arms exports if the arms are going to undemocratic regimes that are heavy military spenders. Norway is top of the list as it contributes both money and troops to U.N. peacekeeping; Australia is second in the list mainly for its role in East Timor - this list takes into account data going back a number of years. A significant criticism of this Index is that Iraq is not considered, so the U.S. appears in the middle of the list. The arms exporting countries of the UK and France appear at the bottom, as does Japan as it does not contribute much to U.N. peacekeeping, considering its size.

Technology has huge impacts on developing countries, but what is the role of rich countries' governments in the spread of technology? In this Index countries are rewarded for spending on research and development (R&D) - particularly for health, less so for military R&D - and for not pushing intellectual property rights beyond the WTO Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), as it is considered that the international property rights regime is already too strict from the point of view of developing countries and too far in favor of Hollywood and drug and software companies. France and Canada are at the top of this list, as their governments spend a lot on R&D. This is also the case with Japan.

Looking at the overall results, typically the small and more homogeneous countries where government is trusted and has a large role in the economy - the Netherlands, Denmark, Sweden, Norway - score highly in areas where government does well by intervening, i.e. aid, peacekeeping, environment, and appear at the top of the list in terms of the overall Index. Next are the former British colonies - New Zealand, Australia, the U.S., and Canada - that do well where government is not directly involved or is involved mainly in support of the private sector, i.e. trade and investment.

Japan is different from the other 20 countries in the list, not least in that it is the only Asian country. It appears at the bottom of this Index, superficially at least because it gives little in the way of aid relative to its size, has high trade and migration barriers, and does not do much in terms of security. Fundamentally, the premise embedded in this Index is that active engagement with developing countries is a good thing, therefore Japan's insularity is what causes it to be ranked last. However, when looking at the overall change in score for each country between 2003 and 2006, it can be seen that Japan is one of the most improved countries on the list.

More information is available on our website, Please remember that there is more to helping than aid, every country can do better, and there is a lot more to be done.

Yasuyuki Sawada:
Four years ago when David and I first met, the Center for Global Development had just published the first 2003 CDI. At that time I presented a paper listing eight reasons why the CDI was not an appropriate index, titled "Commitment to Development Index: Critical Comments." Since then there have been several important improvements to the Index, notably the penalty on arms exports to undemocratic nations. Hence, the title of my comments today is "Commitment to Development Index: Comments" dropping the term "critical." I have four points to add which I hope will facilitate some constructive interactions.

As a background to this I would like to mention the Millennium Development Goals (MDGs), an agreed upon set of goals for global poverty reduction based on the Millennium Declaration of the United Nations, adopted on September 8, 2000. I think that the CDI is important as it contributes to the monitoring process of goal number eight, "to develop global partnerships for development." The CDI can be regarded as a numerical targeting indicator for the 8th goal of MDGs.

The first point I would like to make is that the CDI is really an input-based index, not an outcome-based index. Two examples are aid and migration. Although aid is necessary for global poverty reduction, the aid-growth relationship of recipient countries has been shown to be a fragile one by academic research, therefore the CDI should be designed more to reflect outcomes. Also, migration may be harmful due to brain drain, but more information is needed regarding the real contribution made by migration. I suggest an outcome-based ranking, based on a cross-country growth analysis using aid, investment, trade, etc., as independent variables. Then with estimated coefficients, appropriate weights can be derived in order to construct the outcome-based CDI.

Secondly, it may not be appropriate to standardize the amount of aid by GDP for donor countries because there could be a nonlinear relationship between the amount of ODA and outcomes such as economic growth and/or poverty reduction. Accordingly, the scale of ODA itself provides important information about donors' contribution. Especially, regarding Japan's bottom ranking, I think it would be fairer to show both Japan's GDP-aid ratio-based ranking and its ranking in terms of absolute amount of aid, in which Japan is in fact fifth.

Thirdly, the technology category is a new addition, and I think that contributed to the improvement of Japan's ranking over the last four years. CDI is interpreted as a trial to quantify and evaluate the international contribution of each developed country from a synthetic point of view. International contribution of developed countries, however, is not restricted to aid, trade, investment, migration, environment, security, and technology.  I can think of two further categories; the first being the donor's contribution to the health and education sectors in developing countries and the second being a category for contributions to heritage preservation in developing countries. Our result shows that when we add these two new categories, Japan's ranking jumps from the bottom to 13th. Therefore, we can see that there is room for adding further categories. Moreover, a contribution for entertainment and leisure - things that make children and grown-ups in developing countries happy - could also be considered.

My final comment is about the policy implications of the CDI. There are three typical responses in Japan being ranked the lowest in the CDI. The first is to ignore it; the second is to insist that there are problems with the CDI's calculation procedure; and the third is to argue that the CDI should be used to improve Japan's ODA provision. My concern about these responses is that people might abuse the CDI politically, as an excuse to either call for a decrease or increase in the aid budget. I wonder what the optimal way is to fully utilize the CDI results for further improvements in Japan's ODA. This leads me to the question: What are the policy implications of the CDI, especially for the Japanese government and people in Japan?

David Roodman:
Thank you, Dr. Sawada. I appreciate your ideas and the spirit in which you offer them. I do not, however, agree with your suggestions on doing growth regressions on migration, investment, aid, etc., to figure out the coefficients. The regressions done with just aid are not reliable. Therefore I would not want to do it with the others. The seven components are equally weighted because I wanted to keep it simple and, as we do not know what the "right" weights are, I thought that this would be minimally arbitrary.

Looking at absolute impact is an interesting idea. On our website we have a map where countries are scaled according to the actual amount of aid they give and Japan looks much bigger than it really is; which gives an idea as to their financial contribution. I do believe that the primary emphasis should be on how well countries are living up to their potential to help, which argues for looking at aid/GDP. I did not want to start looking only at specific types of aid, as you suggested for the health and education sectors, because for almost any kind of aid one can argue that it deserves special weight. Commitment to preserve heritage does not feel as important as migration and other factors; UNESCO funding may however already be counted as aid by the DAC, from which we acquired the definition of aid we use for our calculations.

With regard to government responses, you said that one response is to ignore it: that has been the U.S. government's response. In Europe, many countries have responded differently, using the CDI as a framework for internal discussions about "policy coherence." You ask a very good question at the end, and I was hoping that you could answer it: How can we use these results to bring about improvements in Japanese policy? Is there a way that we can make this into a positive relationship to do some good?

Questions and Answers

Q: As active engagement is not necessarily a positive thing, do you not think that the CDI should include an output-based element taking into account country-specific contexts, like the support for Poverty Reduction Strategy Papers (PRSPs), harmonization alignment, or supporting country ownership?

A: I agree and should make clear that active engagement is not always a good thing; what we want to see is engagement of the right sort, which could be seen as either active or passive. I agree with you that in many of these policies, the impacts are going to depend on local circumstances.

You suggested other aspects of aid quality to measure; this is challenging as, for example, PRSPs can often be of no value in judging true aid quality. Nancy Birdsall is currently working on bringing some of these ideas about aid quality into a separate index, which could eventually influence the CDI. In terms of being efficient, it seems that donors are not doing very well right now, especially regarding aid proliferation, and the worse they are, the easier it should be to show that.

Q: In terms of using this research effectively for the purpose of helping developing economies, would it not be better to show Japan compared to certain other countries in order to illustrate how specifically it can improve in its contributions, rather than simply show it as being the worst? Also, did you take into account Japan's contributions to the U.N., which are especially large?

A: This is in line with Prof. Sawada's suggestion that we also present Japan's contribution in terms of absolute impact, as this would make it easier for Japanese people to listen to, and that will help get the message across. Regarding the U.N. contribution, we do not consider the whole contribution, but only the contribution for U.N. aid agencies.

Q: This is not a question to Mr. Roodman, but a question to the audience. The discussion of ODA is, in a sense, wrong here. What you should be discussing is why Japan's tax policy towards developing countries is so appalling. Why is this policy change not being made?

A (audience member): This is the result of the heavy protection of the rice market and other agricultural products in Japan. The Minister of Foreign Affairs, however, has criticized this ranking because it says that the weights used by CGD are unfair to Japan. There are some technical issues involved here, but I do agree that Japan should be ranked at the low end in terms of trade and that this does need to be discussed.

A: (Mr. Roodman): The issue is rice. Poor people who live in countries like Vietnam, Malaysia, Thailand, Indonesia, are dependent on the price of rice for much of their livelihood. The lack of opportunity to sell that rice into this rich country makes them poorer: that is what this comes down to. The weighting is based on the total value of rice production in developing countries divided by the total value of goods production in developing countries.

Q: While I agree that Japan is too protective, I think the CDI focuses only on the negatives. Do you not think a chart comparing the development results both with Japan's contribution and then also without Japan's contribution to these poor countries would be more accurate and give a fuller story?

A: There is also an issue about the potential distinction between the role of government and the private sector. If you take the long view, however, it is clear that there is a government role and perhaps we are not capturing that. You could say that what is more important than tariffs and migration policy is the macroeconomic policy in the rich countries. At CGD we had a discussion about this and asked, "Is there a way that we could start to look at these kinds of policies?" Eventually we decided that we could not, say, second-guess the Federal Reserve on interest rate policy. I think you are making a very important point there that there is something very important that we are missing.

Comment: Regarding the rice tariff, this 900% is an artificial figure. According to my calculations the current tariff rate is about ¥400 per kilogram, but the actual price gap is only ¥150.

This is a political matter. The Ministry of Agriculture, Forestry and Fisheries wants to pretend that they are very protective of that sector, so they artificially increased that figure to say that they are protecting the rice sector by 900%, but the actual protection rate is 150% or 200%. If you change that calculation, your results will change dramatically.

Q: On the trade index, is the issue not much more complex than how the index is currently calculated with regard to rice?

A: Perhaps what you are suggesting on the weighting of rice tariffs would be to use trade flows to weight, but that brings us to the fact that imports and exports are endogenous to the barriers. The goods that face the most protection would get the least weight, if you weight by trade flows. Agricultural protection is high in all the rich countries, and that is why we switched to production weights. There is no one perfect way to measure this.

Q: How did you weight governance vis-a-vis poverty? Which gets more weight?

A: The penalty for project proliferation - having lots of little aid projects as opposed to a few big ones - is closely related to the median project size according to the database collected by DAC, called the Creditor Reporting System, which admittedly has lots of problems. The selectivity weight is based on two multipliers which are scaled as being between 0 and 1: one based on GDP per capita and another based on the Kaufmann-Kraay overall governance score, which looks at corruption, government efficiency, rule of law, etc.

Q: New Zealand gives preference to migrants escaping from natural disasters: perhaps the environment factor could be somehow combined, or taken into account, with the migration factor?

A: New Zealand scores well on our indicator for immigrants from developing countries. I am not convinced that we should start looking at giving extra credit to countries because they are taking people from countries that are susceptible to sea level rises; there are many people who one could argue deserve special treatment, and that is not an argument I would want to get into.

Q: Regarding migration, what is your opinion on the advantages that English-speaking or French-speaking countries have on their migration scores, considering that it is difficult for Japan to accept people who speak no Japanese whatsoever?

A: Looking at the results, there may be some effect, but it may not be a large effect. After all, most of the immigrants to the U.S. speak Spanish. Or think of all the Turkish and Vietnamese in Germany.

Q: Regarding the index for environment, what do you think about the effects of recycling on the index, as it currently does not seem to say anything about it?

A: While recycling has some effects, it is better for the environment if we do not generate the waste in the first place. Reducing use is more effective than recycling.

One weakness in the index is that we are trying to measure policies, but in the environment index we are looking a lot more at outcomes. While many policies, including recycling, affect greenhouse gas emissions, I think that other policies like energy pricing may have larger effects.

Q: According to Japan's CDI score, Japan's weakness on investment is the imposing of restrictions on pension fund investment in emerging markets, but emerging markets are not poor countries. As a selectivity issue, I do not think it is really important whether or not there are restrictions on the emerging markets.

A: I think perhaps I should have not used that term. I had used "developing countries" repeatedly and was looking for a different term, so I said "emerging markets." For me, the term emerging markets includes Malaysia, Indonesia, perhaps India and China now; countries with a lot of poverty, so one would hope that the investment flow would help accelerate poverty reduction, but perhaps we should change those words.

Q: I do not think that any money goes to low-rated bonds in developing countries, like most African countries. BBB is the lowest rate for investment in pension funds, normally. In that sense most African countries, except Botswana, are excluded from investment.

A: I think that pension funds should buy a little of everything for portfolio diversification, in order to reduce risk. Problems occur when governments say that pension funds cannot invest at all in certain developing countries.

Q: Why is aid weighted positively in favor of aid to the poorest countries, while migration and investment are not?

A: A selectivity measure for investment policy or migration is an interesting idea, but we might have a problem in terms of data as migration data is very poor. There may indeed be an argument favoring immigration from and investment in the very poorest countries.

*This summary was compiled by RIETI Editorial staff.