It’s one of the oldest questions in the history of philanthropy. What’s the best way to help the poor? Is it through charity—giving cash directly—or through philanthropy, programs designed to give poor people the skills they need to advance in life? And if we practice charity, how do we make sure that the poor are not dependent on our aid?

Here is what Andrew Carnegie wrote in “The Gospel of Wealth”:

“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 on so-called charity to-day, it is probable that nine hundred and fifty dollars is unwisely spent—or spent, indeed, so as to produce the very evils which it hopes to mitigate or cure. A well-known writer of philanthropic books admitted the other day that he had given a quarter of a dollar to a man who approached him as he was coming to the home of his friend. He knew nothing of the habits of the beggar, knew not the use that he made of this money, although he had every reason to suspect it would be spent improperly. The man professed to be a disciple of Herbert Spencer; yet the quarter-dollar given that night will probably work more injury than all the money which will do good which its thoughtless donor will ever be able to give to true charity…

“…In bestowing charity, the main consideration should be to help those who cannot help themselves, to provide part of the means by which those who desire to improve to do so; to give those who desire to rise the aids by which they rise; to assist, but rarely or never to do all.”

I quote this at length because University of Chicago professor Emma Saunders-Williams, writing on the History of Philanthropy blog, radically misinterprets what Carnegie wrote. In her view, “Carnegie explicitly denies that the poor can be trusted with unconditional transfers” and that Carnegie believed that “charitable resources should remain under the control of the donor rather than the recipient.”

The first claim is pretentious. Carnegie did not write about “income transfers,” but the moral consequences of giving a beggar a quarter. Second, it is true that Carnegie liked creating organizations to carry out his passions—the Carnegie Endowment for International Peace because he was a peace advocate, the Carnegie Hero Fund because he wanted to celebrate people who died or were wounded trying to save others. But as a poverty-fighter, Carnegie wanted to help poor people who wanted to better themselves, and his solution—a good one--was to create libraries, which could give the poor the information they needed to advance without telling the poor how to live or what to read.

Since Carnegie’s day, we’ve created giant philanthropies with scores of program officers who fund proposals where newly minted doctorates go into the field, observe the poor, and then return with recommendations for how the poor should live. These efforts reward the cognitive elite and do little to help the poor.

So the spectrum shifts back to charity. Why not help the poor directly through groups such as GiveDirectly?

Tim Harford, who writes an economics explanation column for the Financial Times, explored the issue last month. (Harford, helping those who are irritated by the Financial Times’s long URLs, subsequently posts the columns on his website.) He summarizes several papers that report on what happens when you give poor people cash.

Foundations, of course, dare not talk about morality in our enlightened century, but many donors are worried that if you give money to the poor, they’ll then blow it all on beer and cigarettes. But in a 2014 paper, World Bank economists David Evans and Anna Popova examined 19 studies on giving to the Third World poor, and found that poor people used the grants productively and gave up on the booze and smokes.

Another World Bank economist, David McKenzie, aided the Nigerian government in a giant effort in entrepreneurial job creation (funded by the World Bank and Britain’s Department for International Development). Half the winners of the $50,000 grants were chosen by lottery and half successfully pitched their proposals before a group of successful business people, in a method like “Shark Tank” (or, since Harford is British, “Dragon’s Den”). Those who won the head-to-head competition were, three years after they had gotten their grants, twice as likely to have started a business and three times more likely to have a business that employed more than 10 people than those who got the money through a lottery.

Harford also discusses Columbia political scientist Chris Blattman, whose research I discussed here last year. Since I last wrote about him, Blattman has expressed his views in a New York Times op-ed and a podcast on the excellent series EconTalk Blattman discussed in his opinion piece about how he gave small sums of money to poor Liberians, who then mostly used it to better themselves.

Blattman said that all the evidence about the benefits of giving cash to the poor show that we “should be skeptical of stereotypes of those we purport to help.”

But his primary example of stereotyping is an ambiguous case. In 2014, Chinese recycling tycoon Chen Guangbiao bought a full-page ad in the New York Times to announce that he would give $300 and a free lunch to several hundred homeless people in New York City. The New York City Rescue Mission held the lunch, but declared Guangbiao’s gift a $90,000 donation.

I don’t know any details of the negotiations between the rescue mission and the tycoon, and I’m sure there were many places that would have taken Guangbiao’s money with no strings attached (including restaurants). But the New York City Rescue Mission was right not to compromise itself. Like other rescue missions, they take broken people and make them whole through intense moral and spiritual Christian teaching. They do far more than simply “process clients.”

It’s clear that in the debate between philanthropy versus charity, the pendulum has swung too far towards philanthropy. In dealing with the poor, we need more charity and less philanthropy. But the success of rescue missions shows that philanthropy can and should play an important role in helping the poor, and is not an obsolete 20th century relic.