16 September 2010

Why should DFID Give Aid to India?

Amidst all the reports today about whether the UK will end aid to India and whether the Indian Government will end its request for aid, it is worth stepping back to think about the reasons for UK aid to India.

I can see at least 5 poverty-related arguments that feel strong to me.

1. DFID works in 3-4 of the poorest Indian states. These are the size of medium countries with populations in the 50-60 million range and if ranked as countries, they would have some of the highest poverty numbers in the world. But this argument only goes so far. These states are not countries, they are part of India. The central government may not be a terribly effective donor to the states, but it is one, nevertheless.

2. But often it is difficult for States to get money from the donor centre. States need to have very strong administrative systems and capacities to do so effectively. The poorest states have the weakest tax base and hence the lowest administrative systems. DFID helps these states to access these central resources to boost investments in child health, nutrition and education where they are most needed.

3. DFID can help state governments take risks and innovate. There is a great demand for innovation in programming but a sense that the risks are high and prohibitive. DFID can act as a support and a lightning rod for risk-taking.

4. DFID has to back up this work by programming resources. They have to be seen to be willing to invest in innovative ideas they support. These investments leverage much larger state investments in areas such as nutrition that have benefit cost ratios of up to 17:1.

5. It is easy to be seduced by the hype of “Incredible!ndia”. But despite the rapid growth, GDP per capita is still just over $1000. This is 6 times lower than its frequent comparator, Brazil and 3.5 times lower than China.

India is still a poor country. DFID still has a key role to play.


Roving Bandit said...

Plus, India is CHEAP! http://www.rovingbandit.com/2010/09/why-dfid-should-still-give-aid-to-india.html

Jiesheng said...

Why DFID shouldn't have suddenly cut off aid to China:


paul isenman said...

Another good post. What strikes me is that if we are concerned about global poverty, recognise how many of the absolute poor are in India, and think that India uses money fairly well, that raises an important moral issue about the value -- exchange rate to put it crudely – of an Indian compared to someone in other developing countries, Europe, or the UK. Focussing just on developing countries, why should the objective not be to maximise the present value of global poverty reduction (or, better yet) of poverty-weighted human development? So the question should not be only whether India, still a poor country, should make bigger transfers from the centre to poorer States. It should also be whether we should strive to get big increases in aid to India. The fact that India is ambivalent about aid -- which has always been the case -- is not a reason for doing otherwise. This sense of self-reliance, linked to India’s ability to expand successful programmes to a national scale, makes India, if anything, a more effective recipient of aid.

At this time of review of progress on MDGs, it is important to note that meeting MDGs should not be the basis of cutting aid to India – or for country allocations in general. The MDGs were rightly intended to be global, even though it is useful for both recipient countries and donors to use them to spur progress on a country by country basis. The issue is easiest seen on MDG 1, halving the proportion of the people whose income is under $1 per day. If India meets this target but on the margin can still bring more people out of poverty than other countries, then we run into the same problem noted above of the exchange rate between Indians and others. In addition, MDG targets were meant to stimulate progress and support for aid, which in my view they have done brilliantly. But they were never intended as, nor should they become, de facto ceilings in allocating resources either across or within countries. It is clearly even more important to bring an individual from $0.90 to $1 than from $1- $1.10. But on what grounds do we use thresholds (headcounts) that give zero weight to the latter? This is why degrees of poverty are better taken account of by high elasticities (‘exchange rates’) not knife-edge thresholds.

Similarly, taking account of expected rather than current per capita income, as recommended by Adrian Wood (“Looking ahead optimally in allocating aid” (http://www3.qeh.ox.ac.uk/RePEc/qeh/qehwps/qehwps137.pdf) does not change the conclusion that India should get more aid. Although most of his simulations show increases in the share of total aid going to Africa, all save the most extreme (zero discount rate and predicting incomes 100 years ahead) show massive increases in aid to India – now at only about 1% (of ODA).

It is ironic that the view of putting no value in aid allocations in progress beyond MDG targets should come from donor countries whose anti-poverty programmes apply well above their own poverty lines, let alone the global income poverty target of $1-$1.25 per capita. In sum, MDGs and global poverty headcounts are extremely important in helping to motivate change. And their underlying objectives are valid as criteria. But applying them as thresholds is inconsistent with equity of treatment of individuals across countries. The discussion here focusses on India, but it applies to allocations in general.

paul isenman said...
This comment has been removed by a blog administrator.