My guest in the latest Development Drums podcast is the moral philosopher Toby Ord.Toby has made a commitment to give away the majority of his lifetime income to charities working in the poorest countries. He is also the founder of Giving What We Can, a society of people who commit themselves to give away at least 10% of their income to wherever it will do the most to relieve suffering in the developing world.
In the podcast, Toby talks why he thinks it is important to identify and support the most cost effective programmes. The podcast also discusses the moral philosophy which guides Toby’s approach.
One of Toby’s insights which I found interesting was his observation that the cost per year of life saved varied enormously between different development programmes. You can buy an additional year of life for $3-$10 by investing in interventions such as zinc fortification, childhood vaccination and managing tropical diseases, compared to about $500 for antiretroviral therapy. Toby challenges the common assumption that we should spread money around to do a little of everything, or that we are entitled to choose the issue which most interests us. He suggests we have a clear duty to identify the most cost effective approaches and then focus our money on those.
In the final part of the podcast we discuss Toby’s decision to establish Giving What We Can, and the choice that Toby has made to give away most of his lifetime earnings.
You can listen to Development Drums directly on the website, download it to your MP3 player, or subscribe free of charge in iTunes.
[This blog post was updated on 27 April 2012 to reflect corrections to the original DCP2 estimates of the cost effectiveness of treatment of Soil Transmitted Helminths, which were found to be incorrect]
20 responses to “Giving what we can [Development Drums podcast]”
…and/or of course you can give your money (and your time) to bring the cost of ARVs down to the same as that for bed nets or parasitic infections; but as someone living with HIV I would say that wouldn’t I ?
This links to the results agenda. Choosing an ineffective programme is choosing not to save lives. In various sectors, often in health, where there is a tradition of measuring results, we can even calculate how many lives we choose to “terminate” by making bad choices.
In health provision in the rich countries, defending a non-results agenda would lead to prosecution.
I don’t think Julian Hows’s comment should be easily dismissed. Focusing so much on cost effectiveness implies that costs are constant and exogenous, and that is wrong. Costs change depending on what sorts of activities we engage in (there are scale economies, so more focus on bednets, for instance, might bring the cost of bednets) and costs can be altered through other interventions that lower prices (e.g. the introduction of generic competition lowers the costs of HIV/AIDS treatment). I’m not convinced by the argument that we should do less of things that are not cost effective and, to the contrary, find the inverse argument, that if there are good reasons for making certain interventions then we should try to make them more cost effective, quite compelling (or at least not easily discarded).
@julian @ken
Thanks.
I completely agree that in thinking about cost effectiveness we should take account of the future benefits of our actions today – and that those future benefits may include bringing down future costs. (There are other possible future benefits which need to be taken into account, such as new knowledge, public awareness, etc). There may also be future costs which should be included. A cost effectivess measure should include realistic estimates of these future effects of today’s decisions
Indeed, I have myself worked on programmes such as Advance Market Commitments whose cost effectiveness is significantly dependent on the expected impact on future prices.
None of that means we should not examine the cost effective of proposed interventions: it means we should be comprehensive and forward looking in our estimates of cost effectiveness to include our best estimates of all these future benefits and future costs.
Perhaps advocates for spending money now on ARVs believe that, properly measured to include the future benefits and costs, the cost per QALY of ARV programmes is substantially lower than the $500 estimated by DCPP. In that case, let’s see the analysis. I would be surprised if it changes the figures enough to change the conclusion, but I am certainly open to the possibility that it could do so.
In short: it is right to argue that cost effectiveness analysis should be done properly. What is not right is to say that because of these future benefits, cost effectiveness is not a good basis for taking these decisions.
Owen — Thanks for the reply, one of the things I appreciate and like about your blog is the willingness to engage with comments. Two brief remarks, if I may.
Your final paragraph: I don’t disagree with this and can’t imagine there are too many people that would. I hope my comment didn’t imply that “cost effectiveness is not a good basis” but rather that it needs to be done properly. But there is a lot of room for disagreement over what constitutes “done properly.”
ARVs: Consider the cost effectiveness of ARV treatment in 2000 vs 2010. It still may be low, but it’s gone down greatly. You have to ask why. There are lots of factors, of course, some having nothing to do with giving and philanthropy, but at least some role was played by decisions made by donors (individual and organizational) to promote the treatment ARV campaign notwithstanding low QALY-based cost effectiveness. Busby and Kapstein have forthcoming book about this process by which donors/activists built the generic ARV market, which in turn has increased the QALY of each dollar spent.
@ken Thanks. One of the main reasons I blog is because I learn a great deal from the conversation in the comments.
Hi, Owen, so this reply is actually a response to the larger dialogue that you, Matthew Greenall and I have had over the past few days, but which was spurred on by your podcast with Toby Ord.
First, I think Toby’s decision to give away the majority of his lifetime income to others less fortunate than him, through charities working in poor countries, is a noble gesture. Whether this will make a difference in the lives of the people he hopes to reach or their communities is an open question.
What concerned me in the podcast was the over-emphasis on a narrow conception of cost-effectiveness, not as a rubric for guiding Toby’s personal choices (which are his to make), but by extension as a framework for how others, including donor nations should be allocating aid.
I’ve pointed you to the critiques of simply tabulating the cost of interventions for a single disease like AIDS or between those for different conditions, for instance between HIV and helminth infections, and categorizing them as all crudely comparable in other ways (see Walensky and Kuritzkes from 2010 or Moatti from 2008). However, in the podcast, Toby runs down the cost of treatment for Kaposi’s sarcoma, for antiretroviral therapy, for condoms, for AIDS prevention education, in a stepwise fashion commenting on their relative and descending price tags, just in this way.
In our subsequent exchange, I think we both agreed that more sophisticated cost-effectiveness analyses are critical components of decision-making around health funding, lest we end up simply prioritizing the cheapest interventions for the poor.
As far as I know, at least where it concerns HIV and interventions for other diseases, none of the cost effectiveness analyses have hit this level of complexity, ignoring many of the spillover benefits of ARVs (e.g. on prevention of new infections; maternal survival and survival of uninfected children), let alone addressing the relative impact of these indirect effects in comparison to interventions for other diseases as you have suggested we require.
Meanwhile, these more rudimentary cost-effectiveness analysis are routinely trotted out and used in the public realm to suggest ART is not cost-effective, and Toby’s short analysis seems to fall into this trap. You seem to suggest the burden of proof is on AIDS advocates to do these analyses, but say nothing about the use of simplistic formula by others in the public sphere to make the case against AIDS funding.
It is also worth noting that even if we can say confidently that X is more cost-effective than Y, the political economy of global health doesn’t simply flow out of these kind of technical analyses. For instance, a certain political alignment happened in many countries and globally around AIDS that was very effective in driving institutions and the elites that control them to respond to the epidemic after years of neglect. Just because ARVs are more expensive than deworming medicines doesn’t affect that fact. It’s not as if you can push a button and redirect the complex history of AIDS activism towards another more cost-effective goal. In fact, sometimes I wonder if people prefer a generalized insufficiency where everyone suffers with little than to let progress happen by fits and starts, in ways that aren’t exactly how rational actors would have set the path forward.
You also seem to discount the role the AIDS movement has had in vastly expanding the pot of money for global health, saying at least in Europe, this activism has affected allocations for programs rather than the aggregate sum up front. What I would like to see though to support this is some data on global health spending pre- and post-2000 in the states of the EU, when there was a huge groundswell of AIDS activism which led to the establishment of the Global Fund to Fight AIDS, TB and Malaria, the US President’s Emergency Plan for AIDS Relief, and a series of other related initiatives. Is it indeed true that in Europe, governments were immune to the global advocacy around HIV, TB and malaria over the past decade or more?
As far the crowding out phenomenon from HIV funding, Shiffman’s paper from 2008 does indicate shifts in funding for AIDS vs. other areas of health, but also says the influx of funds overall may have mitigated possible displacement affects. Importantly, he notes that the counterfactual—what would have happened in a world without AIDS and an increase in funding—is missing from our analyses. The Lordan paper from last year sees a within-country crowding out of malaria funding and to a lesser extent of health funding overall. Furthermore, this crowding out disappears in countries with low HIV prevalence, which suggests that even if HIV is sucking money from other areas of health (i.e. malaria), it is happening in countries dealing with significant AIDS epidemics.
We need to maximize efficiency in spending, by donors, by national governments. Health spending needs to reflect local needs, with assessments made in a way that doesn’t reduce the complexity of what is happening on the ground to DALYs or QALYs alone. In the current economic and political climate, where funding for all global health is in jeopardy, we have to avoid fighting for the the crumbs from the table too, focusing on AIDS as the policy villain rather than larger issues around funding for health and other social programs that are being discussed in capitols across the globe (e.g. austerity-mania in Europe).
No one I know thinks AIDS should suck all the money out of global health and foreign assistance, in fact, most of us working on AIDS have branched out since the beginning to address larger health issues, even moving on to other development issues like education and sanitation, corruption and good governance.
We all need our assumptions questioned, constantly and vigorously. One of the wonderful things about the recent head-on collision between AIDS activists and economists around the Jim Kim nomination, is that people are being called out to defend their views, whether they are hot-shot economists or street activists. We have to keep this dialogue going—it is vital. Groupthink around how we deal with global health and development is our biggest enemy, something that does no good for anyone, particularly the world’s poor.
Owen, I agree with the reasoning in Greggs comment here.
In addition it would be useful to have a discussion of the cost, including the social and economic impact of not treating AIDS and not stopping HIV transmission. This would obviously be different depending on the prevalence of infection and burden of disease in each country, and underlines the complexity when comparing cost effectiveness of interventions. The argument for treatment as prevention would bring in yet another sett of factors to consider.
It is interesting to note that this global debate on cost effectiveness tends to focus on HIV/AIDS versus any other investment priority.
What then about investment in polio eradication? Nobody would say it is not worth it if transmission can be stopped and the wild polio virus eradicated. But the size of the investment necessary to complete the job continues to be huge. Does it make sense to compare this with the cost of additional year of life saved by treating parasitic infections?
Clearly there is a strong case for informing priority choice by cost effectiveness of each intervention. What is risky is however is if this is left to donors, making aid flows determine country priorities. These must be national decisions on how to spend overall resources available for health – where cost effectiveness analysis can contribute, but not be a stand alone basis for priority choice. We talk about political processes, and we talk about the need for broad inclusive processes with democratic accountability.
In terms of HIV/AIDS there is now a growing focus on national leadership for an agreed investment framework, country by country. This is a good basis for dialogue about better, smarter and quicker scale up to turn around the HIV epidemic both by direct investment in HIV specific interventions and in critical enablers to stop the transmission. HIV investments can no longer be done in isolation.
Given that health aid is still only 16 or 17% of sector allocable aid (DAC Development Cooperation Report 2011 Stats Table 19) shouldn’t we be focussing more on increasing the share of aid to health. Perhaps the focus should be on the cost effectiveness of health vs other types of aid.
@Garth Thanks. What do you think the share of aid to health should be? How do you know that 16-17% is too little? I’m interested that many of the comments here count as success increasing the share of aid going to global health.
Hi, all. The podcast was far broader than a discussion about HIV/AIDS, but as an example of the folly of some of the thinking around resource allocation, it’s a good example of what is happening now.
I agree with much of Sigrun’s analysis, particularly her observation about HIV/AIDS routinely hoisted up by many to be pilloried around the cost-effectiveness claims, while other massively expensive health campaigns (e.g. polio) are never brought up. There is something going on here that needs further explanation–what makes economists see red when they think of HIV/AIDS? Just as a side note–when Sigrun and I first met we clashed around some of these issues, though once we talked, we saw we had a lot of common ground and now consider her a dear colleague. So we don’t always have to end up on opposite sides of the table.
I do agree that we need to ensure decision making happens in-country, and even in country pushed down to the district level. However, we need to address the real crisis in governance in many countries, even those that are democratic, where health isn’t a priority for elites or in the case of AIDS in some countries marginalized populations are most affected and have little political power, where corruption and patronage capture sizeable chunks of public spending, where accountability mechanisms are weak and oversight lacking, etc.
Finally, Garth’s point is key. This is what I mean by fighting for crumbs from the table. We need a real analysis of aid overall to assess how it has affected people’s lives, which uses cost effectiveness analyses, but a wider range of tools as well. If we had more of the pie for health–if this turns out to be among the more effective choices for foreign assistance and domestic spending–some of our conversations would be a lot less fractious.
@Gregg
Thanks for your contributions to this discussion.
I admire your efforts to stand up for people with, or at risk of, HIV/AIDS. I also think it is unhelpful that antiretroviral therapies have become the most commonly cited example of interventions which are less cost effective: there are many other examples in development of spending which is poor value for money.
We agree, I think, that cost-effectiveness (properly defined and estimated) should be an important determinant of how aid is allocated, either by each of us and individual givers, or by us collectively acting as donor agencies.
We also agree that a proper measurement of cost-effectiveness should include estimates of future costs and benefits of the decisions we take today. To use Sigrun’s example, the benefits of vaccinating against polio should include not only the benefits to the person vaccinated today: it should also include the herd immunity effects for their immediate community, and the expected contribution these vaccinations may make to the possible eradication of polio, with big benefits for everyone.
We agree that cost effectiveness calculations should be done right. Where we part company is in how we respond to your claim that the existing cost-effectiveness calculations are ‘rudimentary’. You say that this means that there is a problem with ‘simply tabulating the cost of interventions’. I don’t agree: with the best available estimates, such a tabulation is a very good starting point for a discussion about the use of resources. (I do not claim that such decisions should be taken mechanistically.)
Since you focus in your comment on the cost effectiveness of antiretrovirals, let’s persist with that example. The cost per QALY of some childhood vaccinations is estimated to be less than $25 per QALY. According to DCPP, the cost of antiretroviral therapies in sub-Saharan Africa ranges from $350 to $1,494 per QALY. You are certainly entitled to say that the HIV studies overstate the cost per QALY because do not take account of some important benefits which should have been included, such as prevention of new infections, or effects on the future costs of medicines. But if you want me, or governments acting on my behalf, to spend a lot of my money on antiretroviral therapy instead of on childhood vaccinations, then you need to do more than cast doubt on the existing estimates. If we spend $1m on these vaccines, our best estimate is that this would buy 40,000 life years. If we spend the same money on antiretrovirals, the current estimates (with which you do not agree) imply that we would buy about 2,000 life years. We cannot, in good conscience, simply sacrifice the 38,000 additional life years which appear to be available by spending money on vaccines. So for those of us who are not advocates for any particular disease, it seems entirely reasonable to ask you to produce evidence (or at least a plausible calculation) which shows that, when all these other factors are included, spending $1m on anti-retrovirals would save a number of life years which is broadly comparable to spending that money on childhood vaccination. If you think spending $500 on antiretrovirals saves more than 1 QALY, how many do you think it saves, and why?
We agree that governments have not been immune to ‘the global advocacy around HIV, TB and malaria over the past decade or more’. AIDS advocates have been very successful in using their political power to shift resources to these issues. The question is whether this has been a good thing. I live in a country which has an explicit target for aid of 0.7% of GNI, and the aid budget is set to follow a trajectory to that level by 2013. In this case, it is highly implausible that AIDS advocacy adds to the total resources available for development: the effect is to shift resources from within a fixed development budget. This is the case in most of the donor countries, and that means there is a direct trade-off in these countries between AIDS spending and other development spending. I do accept that the situation is different in the US, where the foreign assistance budget is built up from a number of specific items. It may be that PEPFAR added, at least in the short run, to the amount of money which Congress was willing to spend on foreign assistance. But I suspect that there is nonetheless an implicit budget constraint on foreign assistance even in the US, and that there is therefore a long-run trade-off (though perhaps not 1:1) between these programmes and other forms of development spending. There really is a very significant trade-off between AIDS spending and other development priorities.
I strongly agree with Sigrun in her comment that it is for developing countries themselves to determine their priorities. It is reasonable to challenge the global AIDS movement that they have made this very difficult, by advocating the establishment of vertical funds such as GFATM and PEPFAR which – by design – make it difficult for developing countries to use funding in their own countries according to their own priorities. I wrote about the effects of this in a blog post in 2009.
I was interested by your remark that ‘most of us working on AIDS have branched out since the beginning to address larger health issues, even moving on to other development issues like education and sanitation, corruption and good governance’. I’m glad to know that AIDS activists are now recognizing that there is more to development than providing funding for specific diseases – that is what many of us have said all along.
In reply to your question Owen: The share required depends on the total volume of aid but at current aid volume levels I think it should be around 30% to reach a total of around $40 billion a year. This is the level suggested by the main recent estimates of health aid requirements (eg High Level Task Force on Innovative International Financing for Health Systems) and the sum of separate estimates for HIV, maternal and child health, family planning etc.
Given 1) the importance of health to human happiness 2) the large unmet needs (eg 8 million people needing ARVs, at least 5 million preventable child deaths per year etc) and 3) that health aid appears to be having a significant impact (not so easily said about some other aid sectors) then I think there is a very strong case to increase the share of aid to health.
@Garth – thanks. That sounds like a perfectly good case for spending more money on health. But it can only become a case for spending a greater share of (unfortunately limited) aid resources on health if you make an explicit comparison between the benefits of spending on health and the benefits of spending on other possible priorities.
Agreed Owen – however in reality I find that other issues often appear to be more important in the making of these allocative decisions: personal priorities of elites in donor and receiving countries, economic and strategic benefits to donors, the eternal hope of kick starting economic development.
If we are to get rational about this stuff then it appears to me that health aid has more runs on the effectiveness board than many other aid sectors (eg cuts in measles, malaria and AIDS deaths). However I must admit I found Sven Wilson’s analysis which could not find a link between health assistance and mortality challenging and confusing.
Owen,
As we discuss all of this, I can see we agree on lots of things.
Now for the question of why not spend your money on vaccines rather than ARVs?
First, I don’t want to pit AIDS against childhood vaccinations or other diseases. It’s clear we have to grow the pot of money to allow us to do more for health as Garth suggests.
Second, I think it is problematic to posit that AIDS activists have been dictating what happens on the national level from some global perch. It doesn’t match up with the history. The push for ARVs started in developing countries, first Brazil, then Thailand with other countries such as Uganda and South Africa having strong local efforts to put people on ARVs or challenging their leaders when their governments refused to do so. We can’t erase the real local push for ARVs and then say we’re for local autonomy and decision making.
Third, I would maintain that the birth of things like the GAVI Alliance came about as an indirect effect of the AIDS push in the late 1990s, which gave birth to the Global Fund, PEPFAR, etc. with similar donors, actors involved in this work. So even if the UK hasn’t changed its aggregate amount of health aid over the past 15 years in response to pressures from AIDS activists, there has been a knock-on effect from AIDS in what has happened in global health more broadly. Another example–the human resources for health work, the current work on a framework for an international convention on health, have been largely led by people from the HIV/AIDS arena and many AIDS people are involved.
Finally, you miss one of my biggest points.
I want national health care in the USA, selfishly for myself, but also because we have 45-55 uninsured Americans. I want comprehensive primary health care around the world, a vision that has been with us for more than 30 years since Alma-Ata.
What I am trying to get across to you is the fact that wishing doesn’t make it so.
While governments can allocate funding for developing countries, or from their national treasuries based on cost-effectiveness, DALYs or QALYs, alone, this is insufficient for developing the needed political and institutional framework for success.
We have great ideas on paper around primary health care scale-up, but momentum has been lacking to secure real and sustained commitments from elites in many countries (even in the UK where it looks like your government is trying to weaken state commitment to the NHS).
What I am trying to say is you don’t push a button and say: “now build a movement to push for the most cost-effective intervention for X or Y”–though economists want people to act as rational actors, they do not.
As the great South African AIDS activist Zackie Achmat said recently: people have been advocating for primary care for years, about taking on the state and corporations on health,etc. but in AIDS we found a way to actually do it. For him and for the rest of us, this is a foot in the door.
This foot in the door has allowed the transformation of citizens’ expectations around health in South Africa and many other places, instituted real mechanisms of accountability and incentives for change (from legal challenges to monitoring initiatives), broken corporate resistance to generic forms of on-patent medicines.
What I hear you saying is I am glad AIDS activists have now acknowledged that we need more than vertical initiatives. What I am saying is: WE KNEW THIS ALL ALONG. Health systems need breadth and focus. AIDS IS a disease of primary care, it requires strong fundamental commitments to basic health services. As I mentioned earlier, most people I know were working on national health insurance issues for a long time, long before there was even a seed of the idea of the Global Fund.
What I hear you saying is: my (Owen’s) political strategy is to start with the most cost-effective interventions and push for them one by one, or even as a set, and then move on much later to things as expensive as ARVs.
I say: we have a foot in the door, now let’s push it open further, not shut it and try again, because we went with the wrong thing.
We can try this, but what I fear is that the perfect becomes the enemy of the good, that we can’t replicate the kind of phenomena we need to achieve health goals in this way and we head back to the old days where health wasn’t nearly as important on national and international stages, where galvanizing the resources to challenge elites on these issues simply can’t be reconstituted in the short to medium term.
This is about the political economy of global health and it matters as much as does the cost of what we are trying to provide.
@Gregg
I think we are narrowing our point of disagreement to this. You don’t think that your AIDS advocacy pits AIDS against childhood vaccinations, or other global health interventions, or education or clean water or agriculture. In contrast, I think that there is a very real trade-off between spending money AIDS and spending money on these other things. If I am right that AIDS advocacy shifts money from more cost effective alternatives, then the consequence is that on current estimates it may cost hundreds of thousands of life years. Like Toby Ord, I think it reasonable to ask whether this is justified.
You make the point that the perfect should not be the enemy of the good. Agreed. In a world in which we do not have unlimited budgets for health and for development, should we behave as if budgets were elastic, and maximise the amount of money spent on AIDS? Or should we accept that in a constrained world, ‘good’ is spending finite resources in the most cost effective way? I fear your perfect world – which involves much larger spending on AIDS – is the enemy of my good world in which we do the best we can with the money we have got.
Owen–
So, the point I’d like you to address is about the ability to drive an agenda on cost-effectiveness alone–not whether it’s right or proper–but whether we should make another kind of trade-off, that is, assume we can build the political and institutional support for the least expensive intervention(s) and all that entails when with AIDS we’ve made strong gains around the political economy of global health even though the intervention(s) are more expensive.
As Acemoglu and Robinson note in their new book, Why Nations Fail, the ignorance hypothesis, that if countries knew the right thing to do they’d just do it, doesn’t pan out, yet as they say the idea “still rules supreme among most economists and in Western policy-making circles.” I think you have to address this conundrum. It’s not OK, just to say it’s the right thing to do and people should just do it because sadly, as Acemoglu and Robinson note, this isn’t the way the world works.
Gregg
@Gregg
Thanks. In think there are two points to make in answer to your very fair question.
First: I accept the possibility of the situation you describe in which an intervention contributes to political change, institutional development, public awareness, change in attitudes or some other long term social gains, and that those gains may not be available to the same extent from an alternative programme which is otherwise more cost effective. In this case, I agree that these possible broader benefits should be taken into account when making the choice. But I think you will agree that these social benefits should not be preferred at any price. We need to know to what extent this choice would result in a lower direct health impact, and compare the direct health benefits forgone with the institutional benefits we expect to gain. Would (say) greater political commitment to primary health care be worth an immediate sacrifice of 100 life years? 10,000 life years? A million? Different people may reach different conclusions about this. But it seems reasonable to say that this trade off should be described explicitly, and quantified to the extent possible. In an ideal world, we might try to express these benefits in terms of their expected impact on the long term health of the population, and then include them directly in the cost benefit calculation.
My second point is to challenge your assertion that we have made strong gains around the political economy and global governance of health. I think that is far from certain. Let us remember some of the negative effects. We have undermined country health systems by channeling money through vertical funds. We have undermined country processes for deciding their health priorities, and strengthened the role of rich country elites in determining health priorities. We have used monopsony power to drive down the economic returns to investment in pharmaceuticals whose main use is in developing countries, threatening the flow of future innovation and investment. We have focused too much on treatment and too little on prevention. I am sure there are offsetting benefits which may balance some of this, but it is far from obvious that the balance of effects comes out positive.
So I do accept that the trade off you describe is potentially important, but given that many life years are at stake, I also believe that the judgement must be made explicitly and with the best possible quantified estimates of the costs and benefits.
Dear Owen,
As you know, I think the results agenda is very important.
However, as I can read from the discussion above, you seem to think the ultimate choices will be made on basis of the results. What Gregg hints to is only the start of it.
In a society, you normally do your distributions on basis of political choices. These choices are seldom technocratic. Within these choices, allocation is ideally made through transparent processes to the best deal (bang for buck). This is where the results agenda stands. (he will inform the other step but not decide there)
This means that even in times of crisis, culture will be subsidized. Even knowing the cost is prohibitive, some child illnesses will receive treatment. This means the pump and circumstances of the royal anniversary will be maintained and the Olympic Games will be organized.
And it is good that way. Where results come in, is when we decide to allocate money to lift somebody from poverty, then we should have as our baseline the practice we know will have some guaranteed result (say handing out the money through a government scheme) and judge every alternative. Calculate the diminishing returns of handouts, and from micro-credit schemes. etc. The same goes for health. If your goal is realizing the health MDGs as a rights based approach, probably your aids-allocation will be importantly lower than it is now.
As I said before, we should define our goals first, secondly allocate money to these individual goals and take it from there. Step one and 2 are very political. Step 3 is where the development expert and the results agenda comes in.
As you will notice, the critique on the results agenda comes mainly from two sides: development is a vague holistic concept, so just leave me doing what I am good at, and detailed discussions within a sector.
A vertical approach with clearly defined common goals per sector is what is really needed to get the results for the poor.