© 2024 WYPR
WYPR 88.1 FM Baltimore WYPF 88.1 FM Frederick WYPO 106.9 FM Ocean City
Play Live Radio
Next Up:
0:00
0:00
0:00 0:00
Available On Air Stations

Why Giving Tuesday Raises An Uncomfortable Moral Dilemma

Katherine Streeter for NPR

As holiday donations kick off with this , we're going to bring up an aspect of contributing to charity that makes a lot of us ... uncomfortable.

We're talking about the idea that every time we divvy up our money among good causes, we're making a moral judgment: Who is most deserving of our help and which outcomes are most valuable?

And we're not making this judgement in an abstract sense. On a conscious level we might be picking a charity based on some personal connections — in memory of a grandmother, we donate to a home for the elderly.

But through our choices we're effectively setting up a precise mathematical formula for say, how much a child's life is worth compared to an adult's. Or how much saving one person's life is worth compared to boosting another person's income.

How does this work?

Consider what you would do if you had $10,000 to give to charity and were presented with two options:

Charity A provides a low-cost yet remarkably effective job training program to people who are extremely poor. With $10,000 they could train 100 adults for jobs that will immediately lift them out of extreme poverty.

Charity B pays for sophisticated, life-saving medical procedures for sick kids from extremely poor families. With $10,000 they can save the life of one child.

So what will it be: Do you lift 100 people out of excruciating poverty or do you save one sick kid?

Let's say your answer is the job training program. Does your answer change if it turns out that the charity would lift only 50 people out of poverty? Well then, you're basically saying that to you, it's worth sacrificing one child's life to boost 51 or more adults out of poverty, but no fewer.

Acknowledging the hard-nosed calculations underlying our charitable instincts might feel unseemly – like playing God, even. But it's become a central preoccupation of a nonprofit called

That's because every year the organization comes up with an annual shortlist of charities that will give the most bang for your charity buck. In 2018 donors following GiveWell's suggestions contributed an estimated $160 million to charities.

To arrive at its recommendations GiveWell crunches reams of studies and other data. They've determined for instance, that if your goal is to save children's lives, one of the most cost effective ways is to distribute free bed nets to protect sleepers from malarial mosquitoes in poor countries. Or say you want to boost families out of poverty. GiveWell has found that dollar-per-dollar you can help more families by giving them cash grants versus using an equivalent amount of money to give them cows.

But things get thornier when GiveWell tries to compare programs with different aims. How do you decide between cash grants and bed nets? In other words, when it comes to weighing say, saving lives against boosting income, what is the right formula?

"Even if you don't want to think about it, there's no way to avoid that question because you have a fixed amount of resources that you're giving," says Josh Rosenberg, senior research manager for GiveWell.

This week GiveWell released a new study that tries to come up with a formula that can weigh saving lives against boosting incomes, based on the views of a group that's rarely consulted: the people on the receiving end of charity. Rosenberg says the findings surprised even him. NPR spoke with him to learn more. Here are the top takeaways.

Among governments and large donor organizations there's no standard formula for weighing life against income.

Rosenberg says GiveWell launched this research because it was unsatisfied with the models for weighing lives saved against income boosts used by governments and large donor groups. There's tremendous variation in the methodology. And all of them have drawbacks.

Take the U.S. government. Policymakers regularly confront choices between saving lives and boosting incomes, notes Rosenberg — for instance, when they have to decide whether to "implement this environmental regulation that's going to cost the economy say $10 billion, but it's going to save 10,000 lives." Is it worth it? In such cases, says Rosenberg, the government answers the question by applying a formula that saving one life is worth $9 million in economic costs

The U.S. has arrived at this figure by using studies of what's called "revealed preferences." In essence, says Rosenberg, that means "looking at people's actual behavior and trying to infer how they value life versus income." For example, such studies have analyzed how much more you have to pay people to take a dangerous job, like being a firefighter, versus an equivalent position that is less risky.

The trouble, says Rosenberg, is that "revealed preferences studies have a lot of issues – for instance, maybe people aren't really thinking through the risks when they take the firefighting job."

By contrast, European governments tend to use studies that ask people to state their preferences directly. Says Rosenberg, "They use questions like, how much would you be willing to pay for a medical treatment that would reduce your chances of dying by 5 in 1,000."

But here too, it's not clear that people are capable of processing such questions in a way that genuinely reflects their values.

Just as problematic: Studies of both types – those looking for revealed preferences and those asking people to state their preferences — are almost exclusively done in high-income countries.

People in poor countries appear to value life compared to income pretty similarly to those in wealthy countries.

Rosenberg says that in the absence of solid data, economists have tended to assume that people in extreme poverty would weigh a boost to their income slightly higher than those in rich ones. After all, "people are facing such difficult issues in poverty that having a little bit more money could be substantially more valuable to them," says Rosenberg.

The study commissioned by GiveWell – and conducted by the nonprofit research group IDinsight – surveyed 2,000 extremely poor people in Kenya and Ghana and asked them a series of questions about the best allocation of aid money. It essentially put them in the shoes of donors.

"People were willing to pay a lot for relatively small reductions in the risk of mortality – and they tended to choose programs that saved lives at high rates relative to programs that boosted incomes," says Rosenberg.

Rosenberg adds that some at GiveWell were also surprised by the weights that those surveyed gave to saving a child's life over an adult's life. After all, for people who are incredibly poor with little safety net, losing the family breadwinner is especially devastating.

"Some staff wondered whether communities would feel that adults are often caregivers for their families and are making substantial economic contributions to their communities, [so] a community would say that it's more important to avert the death of the adult," says Rosenberg.

Instead, he says, the survey found that "people tended to value averting deaths of children under 5 years old up to two times more than averting deaths of individuals over 5 years old."

What to do with this information

Rosenberg stresses that this research is very preliminary. Lots more must be done. But GiveWell has already tweaked its own weighting formulas. For instance, it now weighs saving lives of children versus those of adults equally – instead of giving slight precedence to adults.

As for the rest of us, Rosenberg says the larger takeaway is not that there is one correct weighting formula to use. (Indeed, GiveWell also provides would-be donors with the ability to adjust its recommendations based on their own weightings.) Instead, the point is to at least acknowledge the moral judgments we are making every time we give. And to make sure that whatever method we use, we're doing so with clear eyes.

Copyright 2021 NPR. To see more, visit https://www.npr.org.