The “Give Directly” Hypothesis


A man checks his phone to confirm that the charity GiveDirectly has transferred a cash grant to his account. (Nichole Sobecki for NPR)

In 2013 Daniel Handel, an economist with USAID—the U.S. government’s main agency for foreign assistance—had just moved to Rwanda when he heard about a charity that was testing a bold idea:

Instead of giving people in poor countries, say, livestock or job training to help improve their standard of living, why not just give them cash and let them decide how best to spend it?

Handel had been mulling this exact question. Aid programs were spending enormous sums per person to boost poor people’s income less than the cost of the program. At this rate, Handel thought, why not just hand over the money to people directly? This program called GiveDirectly was doing just that.

So Handel went to his bosses at USAID’s Rwanda office and proposed an experiment:

Take one of USAID’s typical programs and test it against cash aid. For the comparison, he selected a program to improve child and maternal health in Rwanda by teaching families about nutrition and hygiene.

A pool of families from nearly 250 villages was selected based on typical criteria and randomly assigned to one of four groups.

  • Those in the first were the “control” and received no help.
  • Those in the second group were visited by the nutrition and hygiene education teams.
  • Families in the third group were given small cash grants by GiveDirectly equivalent to the per-person cost of the nutrition and hygiene program, an average of $114.
  • In the final group, families got a much larger cash grant of around $500 – a figure chosen because this was the amount that GiveDirectly estimated was more likely to make an impact.

Following the experiment, the government released the results of the first study in the series.

The experiment found that the program met none of its main objectives. Teaching Rwandans about nutrition did not improve their nutrition or health. Neither did giving Rwandans the cash equivalent of the cost of the education program — about $114.

“Our hearts sank.”

The program’s focus on trying to change behaviors is one of the world’s major strategies for ending malnutrition. And, at least in this example, it had failed to achieve any of its primary goals.

A year on, the children who had been targeted by the nutrition and hygiene program were no more likely to eat a better or more diverse diet, and no less likely to be malnourished or anemic than children who had gotten no help at all. But providing a much larger cash grant of about $500 did make some difference.

Supporters of such “cash-benchmarking” exercises are heralding this particular one as a milestone. For years, anti-poverty advocates and researchers have complained that the U.S. government doesn’t do enough to make sure its aid programs actually work. “But when you talk about giving money to people straight up, with no conditions, people at USAID look at you kind of like you’re a crazy person. There’s ‘an inherent sense’ that they can’t be trusted to spend it wisely.” said Daniel Handel’s associate James Carbonell.

  • In this case, people who were given the cost-equivalent grants used much of the money to pay down their debts.
  • It remains unclear what, if any, material changes USAID is planning to its nutrition efforts based on the study’s findings.
  • At the time of this writing (FEB 2019), USAID remains reluctant to discuss the experiment and did not grant the authors of the NPR story permission to speak directly to Daniel Handel about the results.


  1. Did the authors of the study Fail?
  2. Would proving that cash-equivalent grants were as beneficial as the education program have qualified as Success?
  3. Or did the authors succeed by proving that simply handing recipients money without any stipulation was the wrong way to achieve a particular goal?
  4. Could the authors conclude that poor people really DON’T know “what to do with the money”?


Heavily edited from an original story by NPR.
Copyright 2018 NPR. To see more, visit

Link to the original:

Further Reading

The Planet Money story:

From Nonprofit Chronicles:

Brief Exercise

  1. In the Reply field below, briefly answer any or all of the Discussion Questioons, then discuss how you would respond to finding that your “My Hypothesis” proposal cannot be supported by the initial evidence.
    • (Assume in your Reply that you did not wait until the last week of the semester that the evidence did not support your Hypothesis. 🙂

About davidbdale

Inventor of and sole practitioner of 299-word Very Short Novels.
This entry was posted in davidbdale, Test a Hypothesis. Bookmark the permalink.

28 Responses to The “Give Directly” Hypothesis

  1. profs22 says:

    1.No, the authors of the study did not fail, they were actually successful. They were not able to prove their original theory upon doing research, but actually concluded that their theory was incorrect. They compiled evidence to disprove their original thinking and then shared the knowledge gained with the reader. If i discovered that my hypothesis was not supported by its initial evidence i would be frustrated but i would then use the new evidence against my hypothesis rather than supporting it to construct a meaningful essay.

  2. thecommoncase says:

    3. The authors did succeed in proving that just giving people money is the wrong way to help them have more nutritious diets for their children by saying that they mostly used the money to pay off debts and the children in the study did not have better diets.

  3. Nimadhury says:

    1. I wouldn’t say that the study failed because although his hypothesis remained unproved, this is information that future researchers can utilize to improve the effectiveness of US aid programs.
    2. Proving that grants of an equivalent amount made a positive impact on malnutrition would not be synonymous to being a success in my eyes. As mentioned, it’s not a failure to me to not be able to support your original hypothesis. Additionally, I would take this to mean that US aid programs didn’t invest enough money into their programs to make a positive impact to begin with and neglected to look deeper into what people actually need (ex: aid in paying down debts since that’s where most of the money seemed to go).
    3. No I don’t believe so.
    4. No, I believe they showed that these programs don’t spend enough to make difference. Additionally, based on what I learned in another course of mine it could also be that the people of Rwanda are reluctant to receive help from Westerners because of the cultural context of European colonization. While a fascinating study, I would love to see a project that includes such a context in adhering to the effectiveness of an aid program.

    If my hypothesis could not be proved with research I would change my hypothesis or change my way of thinking about my original one.

  4. runnerd4 says:

    The authors of the study failed to make any real change in the nutrition of the children in the families he gave money to. There was no real change in the types of food they ate. Only the group who was given $500 saw a small change. Proving that cash equivalent grants were as beneficial as the education program would have qualified as success for this situation, but it did not work so it was unsuccessful. The authors did succeed by proving that simply handing recipients money without any stipulation was the wrong way to achieve a particular goal, even though it was not their original hypothesis. The evidence in the study shows that giving them the money did not lead to bettering the nutrition of children. The authors cannot conclude that the poor people did not know what to do with the money because in the article it says that many of them spent the money trying to pay off their debt.

  5. cfriery says:

    The authors’ study did fail only because they didn’t go back and change their hypothesis and retest it after the experiment failed. If my hypothesis wasn’t supported by evidence then I would go back and adjust it accordingly.

  6. rowanstudent24 says:

    I don’t believe that the authors can conclude that poor people really don’t know what to do with money. I think it really depends on the type of person they are. There are many people that are middle class or even wealthy that don’t really know how to spend their money. Some people are just good at budgeting and deciding what’s really worth spending their money on. After realizing my hypothesis proposal can’t be supported, I would try to adjust my hypothesis and share with my readers what I have found out throughout the researching process.

  7. runnerd4 says:

    Part 2. If my hypothesis could not be proven, depending on the information I had found in my research, I would either change my hypothesis or prove exactly what about my hypothesis was wrong.

  8. clementine102 says:

    1.) The authors of the study failed because they came to the conclusion that the program met none of the main objectives. I think that the people of Rwandan did not want to accept the money because it was coming from somewhere that was not of their culture. However, you could say from these conclusions, that they came up with a new hypothesis. If they came up with a new hypothesis from this research, I don’t think they failed at all

    If my hypothesis proposal could not be supported by the initial evidence, I would try to make a new hypothesis based on the research I found. I don’t think it is necessarily a bad thing if this happens with your research and your hypothesis. In fact, your new hypothesis will probably be better than the hypothesis that failed.

  9. cardinal7218 says:

    1. I don’t think the study failed.
    2. I don’t think proving the hypothesis necessarily equates to success.
    3. I agree with this, I think disproving the hypothesis is a success. Now that the original hypothesis is proven wrong, the researchers can move forward by trying new ways to achieve a particular goal.
    4. The authors COULD conclude that, but I’m not sure I’d agree with them. Based on the article, it seems that when these people were given money, they had other expenses they had to use it on like debt before thinking about nutrition. They did “know how to spend it” they just had too much debt to cover and they couldn’t focus that money on nutrition. Maybe USAID has to focus on the debt they’re in if they want to move on to nutrition.

    If my research proved my hypothesis wrong, I would want to use that research to make a new hypothesis and start a new “experiment.”

  10. oaktree1234 says:

    1. They didn’t necessarily ‘fail’ they just didn’t prove their original hypothesis correct.
    2. This wouldn’t really prove their success. One outcome over another doesn’t contribute to success rather than how the experiment was conducted.
    3. There is some truth in this statement. By reaching this conclusion, they could adjust their experiment and find the outcome they desire.
    4. This is partially true. By providing these individuals with with financial guidance as well as the support, they may be more successful.
    -In this case, I would adjust my hypothesis.

  11. SmilingDogTheProfWants says:

    1. although we don’t get the full results we do get evidence that the money was used for important things like their debts which in the long run will help out the family and allow them to afford more in the future.
    2. Cash is king as they say, although knowledge is important if you don’t have the funds to use that knowledge it doesn’t matter and is forgotten, cash allows for debt to be paid off which is a long term effective goal that can have a large impact on the spending in the future.
    3. They didn’t archive their goal but they did have an impact which is more than providing the education can say
    4. It’s fair to say that some don’t but it does help to have extra money to spend on some basic things, but most importantly this idea can be expanded on and may lead to spending money for these people to best help them.

    I suppose if my hypothesis cannot be supported by several sources and evidence I would simply rework the hypothesis around what knowledge I gained to hopefully be able to still use at least a piece of my original idea and not waste all the time I put into research and peer discussions to better my ideas.

  12. shadowswife says:

    1. After reading this article, I don’t think the author’s study was a failure. The money was spent in a way that has proved his suggestion to be a bad idea, but in reality, paying debts is mandatory because there are severe consequences to it. In fact, that might explain why those parents had a hard time prioritizing their child’s health and hygiene. Unfortunately, food that’s healthier and hygiene products are expensive, and applying them to your everyday life will be costly. It’s a possibility that those parents were on a budget to pay off those debts so they had a hard time buying more hygiene products and were forced to eat cheap junk food or eat less. Thus, when they were given the cash grants, they saw it as an opportunity to get them off hundreds of dollars of debt so they can get back to having a normal life.

  13. mesrurerowan says:

    If an opposite hypothesis was proven, the study could be considered a failure but it didn’t occur. However, the authors didn’t work on another possibilities, which could disprove their hypothesis proposal. In this case, I would rather say that the study is incomplete instead of calling it a failure.

  14. corinnebuck1219 says:

    1) I wouldn’t because say they failed, they just never proved their original hypothesis was correct.
    2) No, I don’t agree that just proving a hypothesis or the cash theory to be correct automatically proves their success
    3) I agree with this, they did successfully prove that point.

  15. pardonmyfrench13 says:

    1. I don’t believe the study was a complete failure. I think that their hypothesis just wasn’t proven correct like they thought it would be. With these results that have proven the hypothesis untrue USAID can now communicate the results and tweak their hypothesis again in order to discover more and add value to the world.
    If i found out my hypothesis could not be supported by initial evidence I would either make changes to my hypothesis or still give the results and show what parts of it were true and what parts were not so the public could still learn from my shortfall.

  16. rowanstudent says:

    the authors were successful, not wrong, in this study by just giving people money. If I were to discover that my hypothesis could not be supported by the initial evidence, I would either attempt to take the hypothesis in question, in a different direction and see if I am able to make anything new from it or adjust my hypothesis accordingly.

  17. dayzur says:

    1. I don’t believe the authors of the study failed but instead just proved their hypothesis to be false. Their hypothesis got the opposite results as intended so they now know they will need to switch ideas on how they will go about this issue in the future.
    3. The authors did succeed in that now, they have proved that handing recipients money without any stipulation is the incorrect way to go about things. They did not reach their intended goal, but now know to go a different direction from their approaches on the matter.
    4. No, the authors cannot conclude that poor people don’t know what to do with the money because this was just one set of people who was tested. A different set of people somewhere else in the world can show much greater results, and proving the hypothesis of the authors. This also goes for other groups who could show even worse results compared to the set tested by the authors.

  18. bluntwriting88 says:

    The study failed (link to original) as the study published by NPR as trends, biases, models, and data were abstracted away from the reader. Many of the claims of the article are simply by fiat, rather than having warrants. In essence, the study claims to establish an empirical trend while their publishing is word salad rather than science. Therefore, regardless of what the outcome of the ‘study’ was, success is in showing clear and established models which are backed by data and which can be easily retested. The format in which the study was written does not accomplish that regardless of what the outcome would have been due to lack of scientific rigor (Even if published for the layman, lack of rigor encourages obfuscation of truth).
    To say that the authors can conclude that poor people do not know how to use their grants is a logical fallacy and jump which does not have any established parameters which can clearly lead to that conclusion. Poor presentation of the author’s results displays this as the necessary explicit modeling is not found in the study.
    Now suppose my own hypothesis inst supported by evidence. This is where explicit modeling comes into play – We may form theoretical expectations yet they must line up with reality. Clearly if I have derived an opposing conclusion, then what lies between truth and falsehood is the scientific method.

  19. sonnypetro29 says:

    The study was proved to be a failure, the reason why is because giving the money to these families did not help with malnutrition it only helped families pay off their debts. This proved that you can’t just hand out money with no guidelines on how to spend it. These families are very poor and the last thing there are worried about is spending the money on food. They will use what little money they have to pay their debts. Only the group that was given a large amount of 500 dollars made a small difference because they had more money to spend on not only just debt but some types of food. If I found out my hypothesis was not supported I would find a way that I could change it slightly so that none of my research would have gone to waste. You can always get something out of a test its just a matter of how you word it.

  20. gabythefujoshi18 says:

    Q 2&3: I wouldn’t say that USAID failed in their experiment and I don’t think it would have been successful if they proved that the cash-equivalent grants were beneficial. The objective of the experiment was to demonstrate their findings and develop a conclusion to their theory. This experiment is successful in the sense that a conclusion was drawn from it, which would be that handing money to people out of the blue doesn’t increase improvements in nutrition.
    If the evidence that I find cannot support my hypothesis, then I would try to maneuver the evidence to counter argue my claim or to prove that my theory was unsuccessful.

  21. gooferious says:

    1. The authors of the study did in a way fail the children of Rwanda due to not giving the correct amount of money to the head of the household. A mere $114 could not do much for families, most of the money used was spent to pay off debts that families had. Research showed that families that were given $114 had no significant difference from families who received no help at all. When proposed to give more money to the people of Rwanda, James Carbonell associate of Daniel Halden said that these families could not be trusted to handle theses funds correctly. When the head of the household received $500 worth of grants, some difference was recorded. It seems $500 was a good amount to make some form of change within Rwanda.

    If the evidence I discovered were to not support my hypothesis I would do either one of two things. The first thing I would consider is to simply rework my hypothesis to go along with the evidence I found. The second thing I would consider is to use the evidence I found and use it to help further disprove my original hypothesis, this option however is more risky.

  22. wafflesrgud22 says:

    2. Would proving that cash-equivalent grants were as beneficial as the education program have qualified as Success?
    I believe if it was proven that individuals could be responsible and that people could meet their needs (and even wants) with the cash grants then it would prove that it was a success because it would help people in poverty equally. If my hypothesis did not support my research and claim im trying to make then i would sway to the side that has more evidence.

  23. comicdub says:

    4.) No, the authors could not conclude that poor people don’t know what to do with their money because yes, the group who was payed the cost of the program may have not showed changes in nutrition or health, but this is because it was speculated that they most likely used the money to pay off debt which would for most would have to come first.

    If I found out that my hypothesis cannot be supported by my research, I would respond by either coming up with a different hypothesis or reworking my current hypothesis with the knowledge I gained from the research I conducted.

  24. l8tersk8ter says:

    I don’t think the authors of the study failed because while the results were not what they were hoping for, it still provided them with the answer that “giving directly” is not the route to take. I don’t think the results give the authors the right to conclude that poor people don’t know what to do with money because the article said most used the money to pay debts, and this shows they just have their priorities of what money needs to be spent on first. If I found that my proposal cannot be supported by initial evidence I would rework “My Hypothesis” and try to look for other angles or questions to ask.

  25. sunshine2818 says:

    2) Would proving that cash-equivalent grants were as beneficial as the education program have qualified as Success?
    Anytime that the families were are to show improvement is considered a success. even though most of the families used the money to cover their debts, now they have more opportunities to focus on their child malnutrition. Theses a classic stigma that people don’t know what to do with their money, or that they will waste it. the money might not go directly toward malnutrition for their kids, but the money was able to improve living situations. So the kids may have been help indirectly. this resect was definite not a waste of time, information is always important.

  26. BigBarry2 says:

    1. I do not believe that the study failed, even though they were not able to meet the objectives of the hypothesis. They were still able to see what group was able to benefit the most and it happened to be the one that they were the most concerned about, which is the group that just got the straight 500 cash.
    2. Yes, i believe this would qualify as a success because it was the most successful group out of the experiment. the people in the educational groups did not learn anything where as the people who received the cash were able to put it mostly to good use.
    3. I wouldn’t say they were able to succeed or fail, however they were able to test out new ways of relieving the poor and come closer which ways would be the most effective to help them out in the future. we were able to see that the straight cash plan worked out better than they thought it was going to.
    4. They can not conclude that at all, it seems that the majority was able to put the money they received to debts they owed and bills they have not yet paid.

  27. mhmokaysure says:

    1. Although the hypothesis ended up being disproven, with results showing that neither of the attempted strategies managed to actually make a change that lasted, the authors of the study did not fail. They were able to test out their hypothesis, and see that it was false as the majority of those who received the cash grant instead spent the money on other things that mattered to them, which goes to question the effectiveness of aid programs as a whole.

    2. I feel like proving the cash-equivalent grants to be as beneficial as the education program would be considered a failure, showing that instead of spending all of this time trying to help these people, you can reach the same effect by just throwing some cash their way.

    3. Although not a success, the authors did show that just giving out money to these people without fully specifying what it was for, or enforcing it in some manner was essentially giving them a bonus for no reason at all.

    4. I wouldn’t say that poor people don’t know what to do with the money, instead I would say that if they were given money for a specific reason, it should have been more closely monitored or enforced in order to combine the teaching with the money to reach the goal.

  28. l8tersk8ter says:

    1. I don’t think the authors of the study failed because while the results were not what they were hoping for, it still provided them with the answer that “giving directly” is not the route to take.
    4. I don’t think the results give the authors the right to conclude that poor people don’t know what to do with money because the article said most used the money to pay debts, and this shows they just have their own priorities of what money needs to be spent on first.
    If I found that my proposal cannot be supported by initial evidence I would rework “My Hypothesis” and try to look for other angles or questions to ask.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s