A Quest for the Best

Fledgling grant maker posts online charity evaluations

When Holden Karnofsky, 26, started earning good money doing research on hedge funds for an investment company

several years ago, he began thinking seriously about giving to charity for the first time in his life.

So he started looking for information about the most effective organizations — a process he expected would be akin to checking CNET, a Web site that reviews technology products, to find out which cell phone to buy.

“I wanted to give, and I wanted to give a fair amount, and I wanted to be able to just find the best one” to give to, he says.

That seemingly simple quest ended up transforming Mr. Karnofsky’s life — and introducing a passionate and provocative new voice to the world of philanthropy. When Mr. Karnofsky discovered that the kind of data he was seeking was not easily available, he became consumed with tracking it down, so much so that last June he gave up his $275,000-a-year job to work on it full time.

Today, he is executive director of GiveWell, a fledgling organization he created with a former co-worker, Elie Hassenfeld, to evaluate charities publicly and provide grants to those they believe are doing the best jobs.

But Mr. Karnofsky, who works out of his apartment in Brooklyn, has set an ambitious goal that goes beyond making GiveWell a success: “I want to influence the way people give money,” he says.

‘Completely Transparent’ Foundation

When Mr. Karnofsky delved into the world of philanthropy, he was stunned by some of what he found: that many charities he contacted had difficulty providing reliable evidence that their programs worked; that foundations did not explain publicly why they made grants to one charity over another; and that individuals provided the bulk of charitable donations, but had no neutral source to tell them whether an organization was actually saving or improving any lives.

He and Mr. Hassenfeld set out to shake things up. They raised $325,000 — $10,000 from Mr. Karnofsky and most of the rest from workers at the investment company — and created the Clear Fund, an arm of GiveWell that they dubbed the “world’s first completely transparent charitable grant maker.” Their goal: to scrutinize the data provided by grant applicants and to post their assessments of those charities, both positive and negative, on their Web site — giving charities the right to respond online to anything they wrote.

They vowed to conduct all business in public, posting detailed information about their operations online at http://www.givewell.net/clearfund-records — including all budget expenses and the minutes and audio of their board meetings.

“Unlike a for-profit company, we have no good reason to hide anything we know about the best methods for improving the world — and nobody else does either,” their business plan states.

The duo also started one of philanthropy’s liveliest blogs, the GiveWell Blog, which gets about 600 visitors a day and documents the group’s birth pangs, asks for advice, and takes on the charitable world’s sacred cows. In recent postings, Mr. Hassenfeld questioned the cost-effectiveness of using bed nets to fight malaria in Africa, while Mr. Karnofsky took foundations to task for favoring “high-risk” projects over “proven” ones. “My evidence is not as concrete as I’d like it to be,” he added, “because foundations are too busy cowering behind locked doors to tell anyone anything about what they do.”

Beyond Numbers

GiveWell is emerging at a time when a growing number of people, especially young donors, are experimenting with new ways to evaluate the impact of charities.

Like Mr. Karnofsky, many are critical of services like Charity Navigator, which awards up to four stars to charities based on how financially healthy they are. Critics argue that such ratings tell donors nothing about whether an organization is any good at what it does.

But finding an alternative way to evaluate charities is no easy task — so philanthropy experts are keeping an eye on GiveWell’s effort.

“It’s a window into a process that is becoming more and more common,” says Timothy Ogden, chief knowledge officer at Geneva Global, a group in suburban Philadelphia that advises major donors on international giving. “They’re doing it in public so other people can watch the process they’re going through.”

Mr. Ogden contacted Mr. Karnofsky after coming across the GiveWell Blog and offered to share what his firm had learned about how to measure charity effectiveness. He now sits on GiveWell’s board.

“The concept is good,” says Paul Shoemaker, executive director of Social Venture Partners Seattle, a grant maker that works with its grant recipients to measure results. “Their goals are admirable, and I think they’re very sincere about it.”

But, Mr. Shoemaker adds, “like anybody, they’re going to get a ways into it and they’re going to realize it’s harder than they thought it was.”

GiveWell, which gets help from about 20 active volunteers, including some Harvard MBA students, has just published its first assessments — of charities that provide job assistance to disadvantaged adults in New York. Mr. Karnofsky freely admits that there have been some snags. For example, he says, he and Mr. Hassenfeld underestimated the time it would take to adequately review charities — partly because it was hard to get some of the information they wanted — so they will not be able to meet their original goal of posting assessments by the end of the year of groups that focus on five different causes. Instead, they will cover just one additional cause this year — saving lives in Africa.

The group also angered one charity it evaluated, Vocational Foundation Inc., by publishing inaccurate information in its draft report, something Mr. Karnofsky himself calls “inexcusable.”

But, he says, GiveWell’s philosophy of explaining publicly how it makes its decisions allows it to learn from its mistakes.

Mr. Karnofsky’s approach to GiveWell reflects in part lessons he learned in the business world. After graduating from Harvard College in 2003 with a degree in social studies, he was recruited into the hedge-fund world, spending three years doing research for Bridgewater Associates, an investment company in Westport, Conn. — where he was strongly influenced by a culture of constant questioning.

“What was really reinforced there was how much there was to be gained by surrounding yourself with what I call ‘no men,’ who are very smart, very critical, and will just constantly criticize you and tell you what you’re doing wrong,” he says. “I had always believed in this, but I got more into the idea that there’s a ton to be gained by open discussion, admitting your weaknesses and letting other people point out your weaknesses.”

By opening GiveWell’s doings to public scrutiny, Mr. Karnofsky expects that plenty of “no men” will weigh in. “I can guarantee you that we are going to get skewered and that a lot of the criticisms are going to be valid because giving is one of the hardest decisions you can make,” he says.

Seeking Evidence

Mr. Karnofsky’s path to philanthropy began when he started examining possibilities for his own giving. First, he searched Google using the term “charity” and came up with Charity Navigator. He decided to give only to four-star charities, but says that still left “tons of charities” to choose from.

He started contacting groups that were working on a cause that interested him — helping inner-city youths — and asked them for data to show they were effective. “They would usually respond something like, Yes, 99 percent of our money goes to programs,” he recalls. “I said, No, no, that’s not what I asked. I want to know if you help people.”

One or two provided “evidence” that he found unconvincing — for example, that test scores of young people in their programs had risen.

In the finance world, Mr. Karnofsky learned to test multiple hypotheses to see which one was the most likely cause of a particular event, so he was unhappy that the charities had not ruled out factors beyond their own efforts that might have caused an improvement.

Eventually, he joined with seven co-workers at Bridgewater, including Mr. Hassenfeld, to start a charity club that laid the groundwork for GiveWell.

The members started conducting research on different causes so they could decide where to give, and to share what they found with co-workers and others.

But he says they continued to have problems getting helpful information from charities and foundations. (The e-mail correspondence with some of the groups that the club contacted are reprinted in the Clear Fund business plan, which is posted on the group’s Web site.)

Finding himself “obsessed” with his charity research, Mr. Karnofsky had a “light bulb” moment one morning: While he liked his job, he says, “I would never stay up all night studying the bond market and I just stayed up all night writing about cleft-palate surgery.”

Mr. Hassenfeld, also 26, felt a similar calling. He was spending nights and weekends learning about techniques to help save lives in Africa, wanting to figure out, for example, why more charities don’t promote oral rehydration therapy, a cheap method of treating life-threatening diarrhea. He finally realized, “If this is the kind of commitment it takes, and I want this to exist, there’s no way I can do it part time.”

Last March, the pair took two weeks of vacation to write the business plan for GiveWell and the Clear Fund — and planned their departure from the finance world. With much-reduced salaries of $65,000 each (Mr. Hassenfeld, who lives near Boston, serves as program officer), they are now immersed in evaluating applications for the $140,000 in grants they plan to award this year — four for $25,000 and one for $40,000.

The duo picked five causes for 2007 that they believed could aid disadvantaged people: In addition to helping New Yorkers get job skills and preventing deaths in Africa, they wanted to improve early-childhood development and elementary and high-school education in New York and enable needy people in Africa to become self-supporting.

To find the right applicants, they scanned tax records of about 3,500 charities, asked more than 1,000 of the groups to fill out a survey, and then invited 246 out of the 310 charities that responded to apply for grants.

More than 130 organizations applied.

After evaluating charities that applied for a $25,000 grant to provide employment help to disadvantaged adults in New York, GiveWell announced it had recommended four of 19 applicants — Year Up New York City, St. Nicholas Neighborhood Preservation Corporation, Highbridge Community Life Center, and the Hope Program. The board will approve a grant for one of those groups in mid-December.

GiveWell plans also to recommend Vocational Foundation Inc., but has been revising its evaluation due to the complaints about inaccuracies.

While he is now working 60 to 80 hours a week evaluating grant applications, Mr. Karnofsky dreams of the day when GiveWell can grow so that it can scrutinize additional causes and become a philanthropic force.

While its influence is now limited by the relatively small pool of money at its disposal, it hopes that it will eventually attract other donors — including possibly foundations — who believe in its mission, Mr. Karnofsky says, adding that the group has had informal discussions with the William and Flora Hewlett Foundation, in Menlo Park, Calif.

“They’re a very interesting startup, and we’ll be keeping a close eye on them,” says Eric Brown, a Hewlett spokesman.

Lucy Bernholz, a philanthropy consultant in San Francisco who joined the GiveWell board, praises Mr. Karnofsky’s energy and says his quest reminds her of Frederick Goff’s efforts to shake up traditional philanthropy by founding the country’s first community foundation in 1914 in Cleveland.

But Mr. Karnofsky is undeniably brash, something that even his fans occasionally try to temper.

Mr. Karnofsky recently reprinted in the GiveWell Blog a testy response he wrote on a survey by the Bill & Melinda Gates Foundation that asked for reactions to its Web site (“Not enough information. Not even close to enough information.”). That prompted Sean Stannard-Stockton, an investment manager who writes Tactical Philanthropy, an influential philanthropy blog, to urge him to be more diplomatic.

“I get the impression that you are very authentic in your desire to make the world a better place,” he wrote in the blog’s comments area. “But my take is that a submission like the one you wrote will be dismissed out of hand because you used your normal writing style of sharp criticism.”

Mr. Karnofsky responded that it was not natural for him to “nice up” but conceded “there are times to tone it down.”

Trent Stamp, president of Charity Navigator — whose group GiveWell criticizes — says Mr. Karnofsky seems to “like to take swipes at people like my group.” He adds, “I’m distrustful of anybody who shows up and says they’re smarter than anyone else in the sector and they’re going to change the donor’s giving habits by shining a path for them to go down.”

Mr. Karnofsky says he knows some people think he and Mr. Hassenfeld are arrogant “because we run into this sector that we’ve never been in before and we say, ‘This sucks, this sucks, this sucks.’

“It’s not arrogance,” he says. “It’s just a belief that if you see something that doesn’t look right, you shouldn’t assume that it is right. You should say it isn’t right, and if you’re wrong, somebody will correct you.”