Andrew Carnegie’s 1889 essay ‘The Gospel of Wealth’, has done much to inspire modern philanthropists who want to ensure that their wealth contributes to making the world a better place. He was an early advocate of ‘effective philanthropy’, arguing that:

‘… one of the serious obstacles to the improvement of our race is indiscriminate charity. It were better for mankind that the millions of the rich were thrown into the sea than so spent as to encourage the slothful, the drunken, the unworthy. Of every thousand dollars spent in so called charity to-day, it is probable that $950 is unwisely spent; so spent, indeed as to produce the very evils which it proposes to mitigate or cure.’

We would all agree.  No one wants to see charitable money wasted, and Carnegie was clear that a sure fire way of that happening would be to give the poor themselves money.

He cites the example of someone who gave 25 cents to a beggar, arguing that while the donor might have felt better, the likelihood was that the money would probably be spent on alcohol.  According to Carnegie, then, doing good means doing things for the poor – giving them ‘ladders’ not ‘alms’, teaching them what a better life might look like, giving them things that we know they are good for them (food, shelter, healthcare).

In other words, it requires that there is an intermediary between the donor and the end recipient – some agent responsible for protecting the integrity of philanthropic intention, and making sure that it is not wasted.  Underpinning this approach is a fundamental assumption that the subjects of philanthropy (the poor) lack either the motivation or the skills to improve their own lives.  To be effective, charitable action requires more than money, it requires expertise to guide the poor as to what they need and how to get it.

But evidence is emerging that challenges these assumptions.  It suggests that if we trusted the poor more, we might get better value for the philanthropic pound.

GiveDirectly, is showing how greater trust in the poor can help to eradicate poverty in Africa.  Its work draws on a growing body of evidence that shows that giving people cash – rather than goods and services – delivers better results at lower cost.  Using the growing network of digital payments services, such as M-PESA, it  has sent cash grants of about $1,000 dollars from private donors to more than 60,000 households in East Africa.  And it has invested heavily in serious research to find out what happens from that point on.

A randomised controlled trial that followed the impact of its work in Kenya showed remarkable results.  These include over $270 dollars in earnings, $430 in assets, and $330 in nutrition spend, for each $1,000 grant. The number of days that children went to sleep hungry reduced by 42 per cent. The advent of digital money means that these results were delivered at exceptionally low cost, with about 91 per cent of donors’ contributions put directly into the hands of the extreme poor.

What this data reveals is that far from being either feckless or stupid, the poor often know how best to improve their own lives.  They understand their own capacities and skills.  They can read and interpret their local markets and identify opportunities for development.  And they can probably do that at least as well, if not better than many aid agencies; they can certainly do so more cheaply.

Behind these aggregate numbers are individual stories of change.  During a field visit to Kenya one story stood out for me.  We met someone who had received a grant from GiveDirectly about four years ago.

This person was far from the average aid poster-child.  Male, in his mid-30s and a devotee of reggae music, when he received his first payment his wife had left him because of his drinking, taking their two children with her.   But we trusted him with $1,000 dollars of philanthropic money.  That trust not only helped him invest his way out of poverty, but it helped to restore his dignity.  He now runs two small businesses.  His family is back living at home. His children are at school.  His visits to the local bar are infrequent and sociable, rather than self-destructive.

This story exemplifies the wider evidence provided by the World Bank and MIT that when you give the poor money, they don’t drink it away or become idle.  In fact the reverse: because they have hope and feel empowered they are likely to drink less.  Because they have more capital they are likely to work more.

Despite the pile of data testifying to its effectiveness, the aid community is proving slow to adopt this approach.  Less than 2 per cent of UK aid is delivered using cash.  So what’s stopping scale up of a proven, cost-effective approach to international philanthropy?

Could it be that Carnegie’s ghost, which has done so much to inspire philanthropy, is also keeping it locked in Victorian memes about the poor ?

A growing number of philanthropists big and small, including GiveWell, Google.org, Facebook co-founder Dustin Moskovitz, and others, are increasingly interested in supporting an alternative narrative which connects them more directly to the poor and allows them to be the agents of their own futures.

They are recognising that this dis-intermediated form of aid – which connects the donor and the poor much more directly, is not only the right thing to do – it delivers results.

Dr Joanna Macrae is Senior Partnerships Manager at GiveDirectly.