The other day I spoke at an employee event at the Gates Foundation. We do these “unplugged” sessions periodically. I open with a few remarks and then spend an hour or so taking questions from the team. I see our senior staff quite often, but these sessions are a great way to connect with the larger group.

I told the staff: If you want to learn about why human welfare overall has gone up so much over time, you should read The Great Escape: Health, Wealth, and the Origins of Inequality, by Angus Deaton, a distinguished Princeton economist who has spent decades studying measures of global poverty. It’s quite accessible, and the first six chapters teach you a lot about economics. But the seventh and final chapter takes a strange turn: Deaton launches a sudden attack on foreign aid. It’s by far the weakest part of the book; if this is the only thing you read about aid, you will come away very confused about what aid does for people.

Deaton’s parting shots came as a surprise to me, because it’s clear from the rest of the book that we see the world in similar ways. His first paragraph is dead-on: “Life is better now than at almost any time in history. More people are richer and fewer people live in dire poverty. Lives are longer and parents no longer routinely watch a quarter of their children die.” (I would have cut “almost” from the first sentence, but otherwise I could not agree more.) Picking up on the title of the old Steve McQueen movie, he calls this improvement in human welfare the Great Escape.

I have long believed that innovation is a key to improving human lives. Advances in biology lead to lifesaving vaccines and drugs. Discoveries in computer science lead to new software and hardware that connect people in powerful new ways. Deaton does a good job of documenting how innovations helped spark the Great Escape.

But the Great Escape, Deaton says, “is far from complete.” Innovations reach only those who can afford to pay for them. And that has led to great inequality. The rest of the book examines where this inequality comes from, whether it matters, and what if anything can be done about it.

One of Deaton’s most helpful passages explains how you measure human welfare in the first place. He argues that there’s no single good measure, but health and GDP are the best two that we have. You probably have heard about gross domestic product, or GDP, but you might not know how it’s calculated, how it’s different from gross national product, or why on its own GDP is not a great measure of the quality of people’s lives. Deaton provides a lucid and succinct explanation.

He is also humble about the limits of economic analysis. In this entertaining passage, he goes after economists who make unfounded assertions about economic growth:

Economists, international organizations, and other commentators are fond of taking a few high-growth countries and looking for some common feature or policy, which is then held up as the “key to growth”—at least until it fails to open the door to growth somewhere else. The same goes for attempts to look at countries that have done badly (the “bottom billion”) and divine the causes of their failure. These attempts are much like trying to figure out the common characteristics of people who bet on the zero just before it came up on a roulette wheel; they do little but disguise our fundamental ignorance.

Unfortunately, this humility goes out the window once Deaton starts criticizing foreign aid. After examining the history of GDP in countries that get aid, he concludes that aid doesn’t cause growth. In fact he makes an even stronger claim—that aid keeps poor countries from growing—and says the most compassionate thing we can do is to stop giving it.

In other words, in one chapter, Deaton dismisses people who think they know why countries fail to grow. In the next, he asserts that countries fail to grow because of aid.

Deaton makes another common mistake among aid critics: He talks about foreign aid as if it’s one homogeneous lump. He judges money for buying vaccines by the same standard as money for buying military hardware.

It seems obvious that programs should be measured against their original purpose. Did the program have a good goal, and did it achieve the goal? Instead, Deaton and other aid critics look at, say, aid that was designed to prop up some American industry, see that it didn’t raise GDP in poor countries, and conclude that aid must be a failure.

That’s a shame, because programs that are actually designed to improve human welfare—especially health and agriculture—are some of the best spending that a rich country can do on behalf of the poor. They help the poor benefit from the very innovations Deaton writes about so glowingly. And their success rate is at least as good as the track record of venture capital firms. No one looks around Silicon Valley and says, “Look at all the startups that go under! Venture capital is such a waste. Let’s shut it down.” Why would you say that about aid?

Deaton spends a lot of time on the concern that foreign aid undermines the connection between governments and their people. The idea is that leaders of countries that get a lot of aid will worry more about what their donors think than what their citizens think.

This argument always strikes me as strange. For one thing, many countries—Botswana, Morocco, Brazil, Mexico, Chile, Costa Rica, Peru, Thailand, Mauritius, Singapore, and Malaysia, to name a few—have grown so much that they receive hardly any aid today. Deaton doesn’t cite any evidence that aid undermined their institutions or democratic values. And his over-emphasis on institutions leads him to make sweeping statements like “drugs and vaccines save lives, but the pernicious effects are always there.” I have to wonder what pernicious effects came from eradicating smallpox, and whether these effects could be worse than a disease that killed millions of people every year for centuries.

Deaton is right to point out some problems with aid. For example, too much goes to upper-middle-income countries rather than focusing on the poorest people. More aid can be directed to specific goals like vaccination. And aid shouldn’t be used to replace domestic funding. But I wish he had spent more time than he does exploring better ways to give aid. Most importantly, we need to help develop a system that identifies the goals of every type of aid and makes it easier to measure which approaches work best in various situations.

If you read The Great Escape—and there’s a lot to recommend in it—you should also read something else to balance out Deaton’s negative views on aid. I’ve highlighted a few recommendations in the Other Reviews box at the bottom of this page. They’ll give you a clearer sense of what’s working, what isn’t, and how we can make our efforts even more effective in the years ahead.