ScreenLike every other market, the market for good has two sides. On one side you have do-gooders. And, on the other side you have those in need of some good. In an earlier post, we briefly outlined the three steps in creating a market for good. In this installment of Do-Goodernomics, we focus on step one: finding a poor person. Remember, you are a do-gooder. Your goal is to move as many people out of poverty as possible. You have $100. And, you must give it directly to one person.

How do you begin your search?

You can do what most people in search of other people do and publish a personal ad.

“Financially secure do-gooder in search of a poor person. If interested, please respond with your name, address and an answer to the following question: How would you use $100? Feel free to leave a brief comment.”

Let’s say you receive responses from around the world and create an excel sheet. Thinking that you may share your database with others, you substitute “Poor Person” for each respondent’s name and use a regional location for their address.

How do you begin to screen the respondents?

Your instinct may be to review the responses from your own community first.

Name Address How I would use $100. Comments
Poor Person #1 Southern California (United States) Job Training Program I am your brother.


In this case, you have a lot of information with which to make a decision. You know about your brother’s work ethic or lack thereof. You know about the privileges he has or does not have relative to others. You know of the opportunities he took advantage of or squandered. You may even feel comfortable passing judgment on how responsible you think he is for his current situation, whether or not he will land on his feet soon enough without your assistance, and whether or not your help will dull his incentive to seek out work.

Name Address How I would use $100. Comments
Poor Person #2 Corner of Two Streets (United States) Buy warm clothes. I am homeless.


In this case and all subsequent cases, you lack information. Will the actual use of your $100 differ from its declared use? Are the cardboard sign, milk crate and threadbare clothing props? You have heard of “professional pan-handling.” And, you are unable to distinguish the truly homeless from those who are mimicking the homeless.

Name Address How I would use $100. Comments
Poor Person #3 Appalachia (United States) Put food on the table. Single mother of three.


Even if you know everything you need to know to make a decision, there is one conclusion you cannot escape in all three cases. In your own relatively wealthy community, $100 does not go very far. In fact, it may not help any of the three out of poverty. And, even if it does, it may not move anyone else out of poverty. So, you turn your attention towards global poverty.

Your database is in the billions. Of course, not every poor person replied to your personal ad. And, some of those who did reply told you take your hand-out and go to hell. Even so, it’s in the billions. You have respondents living under $1.25 a day, $2.50 a day, and $5 a day. You decide to focus on those living under $2.50 a day. You also exclude the “Comments” column to make things more manageable.

Name Address How I would use $100.
Poor Person #1,265,983,042 Yoro (Honduras) Buy school uniforms
Poor Person #1,265,983,043 Central Province (Kenya) Host a festival
Poor Person #1,265,983,044 Badakhshan Province (Afghanistan) Expand food stand.
Poor Person #1,265,983,045 Capital District (Venezuela) Organize self-help group
Poor Person #1,265,983,046  Sonora (Mexico) Cross the border into the US


Now, who should get your $100?

You will need to add two additional columns to your excel sheet:

Does respondent move out of poverty? (Y/N) How many other people are moved out of poverty?


Of course, you will never know the answers to these two questions for sure. But, it’s our thought experiment. So, let’s make some more assumptions:

  • You have all the data you need.
  • Esther Duflo and Dean Karlan are your research assistants.

And, you understand the mechanism that links a respondent’s declared use of $100 to the number of people moved out of poverty. For example, the host of the festival earns entry into an exclusive network of his peers and gains access to economic opportunities. The food stand owner expands her operations and hires additional employees. The border crosser gets a job and sends home remittances to his family.

This is good news, right? Yeah, but…

What if the data suggests that you should focus your efforts abroad but you are adamant that all do-gooding “should begin at home”?

What if you had your heart set on buying uniforms for school kids but illegal immigration into the United States yields the highest return?

What if the local shop owner provides the surest escape out of poverty but uses child labor and only sells empty calories (fizzy drinks and chips) and alcohol?

What if the self-help group supports a left of center political candidate who rails against your home country in a national election?

Is it possible that your choice of who gets your $100 is guided by more than the objective to move as many people out of poverty as possible?  If so, which columns would you add to your excel file?

  • Sex
  • Religion
  • Race
  • Age
  • Disabled (Y/N)
  • Political Affiliation

What about the concerns you had when deciding whether or not to give $100 to your brother? These concerns also hold for those living on $2.50 a day, right? And, what if those answering your personal ad know your preferences? Don’t you face the same difficulties you faced when thinking about giving $100 to the guy on the corner?

And you thought this would be easy.

Stay tuned for future posts in the Do-Goodernomics Series.


  • Nate Delano
    8 years ago

    This comment kind of builds off of the last post in this series, but relates also to choosing which recipients receive aid. In my experience, the type of aid given, be it technical training, home improvements (aka stoves, composting latrines, etc), or money directly is inconsequential in terms of the overall effectiveness of the aid. Well, let me rephrase…If you blanket give something to a community, be it training, improvement, or money, there will be people who take advantage of that charity, and people who squander it. Now, if we assume the above is true, there are two options, either try to weed out the people who are going to use the aid ineffectively, or continue to blanket give to a community while trying to create the most efficient marketplace and minimize transaction costs, which would avoid other pitfalls created by the “weeding out” process.

    That second option is the tip I am on right now. The single example is tough, but I would probably go with the option that had the least transaction cost. If you assume that you have an equal chance of giving money to the right person if you pick someone at random from the list, then the best option is to give the money to the person that will get the most of it, perhaps the homeless guy on your corner.

    If you look at international aid in general and my experience with it in particular, the problem is a lack of data showing the effectiveness of any one project over another. Heifer International, for example (considered one of the “good charities”), can tell you that giving a cow will boost someone’s income 5x (for example), but can’t tell you that minimizing staff and maximizing giving (aka just putting money in a more poor people’s hands) wouldn’t be better for the poor in general.

    That is changing, however. Your work with La Ceiba and other charities are starting to maximize data collection and really trying to find out what is the BEST way to help the poor. This post is a round about way to send you a couple links, and to solicit your opinion on the idea of direct giving. Here they are: http://www.givedirectly.org/, and http://www.thisamericanlife.org/ (the story called “I was just trying to help). I’m considering applying for a D-prize to do something similar in Paraguay, and I want to get your opinion.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.