That’s an interesting question and i don’t know! I had myself implicitly been assuming it was from volunteer labor being unpaid for. But now that you put the question to me explicitly I realize that it could easily be due other factors (such as fewer expenses in some other category such as infrastructure, or salaries per staff, hypothetically). I don’t think the answer to this question would change anything about our calculations, but if we happen to find out I’ll let you know.
In an earlier draft I was considering counting the volunteer labor as a cost equal to the size of the benefits from the cash that the volunteer might have earned otherwise, but I left it off because it would have been too small of an effect to make a notable difference.
Yeah, I was asking w/r/t that counterfactual cost of labour. Perhaps this is a dumb question, but if that cost would’ve been negligible, as you say, then what explains the gap? Is the counterfactual cost of volunteer labour really that much smaller than staff labour?
If you’re interested in the question of why I figure that doing voluntary unpaid labor for most people involves negligible harm or opportunity cost of the type that ought to change our effectiveness estimate, my napkin calculation goes something like: suppose I personally were to replace half my compensated work hours with volunteering. That would be several tens of thousands of dollars of value to me and would half my income, which would be a hardship to me. But it should only take ~$285.92 to “offset” that harm to me personally, by doubling someone else’s income—theoretically the harm i experience for losing one day’s labor can be fully offset by giving a givedirectly recipient 0.78 cents. Obviously my own labor is worth more than 0.78 cents a day on the market, but matters being how they are, 0.78 cents seems to be around what it takes to “offset” the humanitarian cost to me personally (and the same to any volunteer, with respect to them-personally) of losing a day’s labor.
Having established that, how much unpaid labor are we talking about?
5-12 people who meet for 90 minutes a week, over six sessions.
Assuming 8 hours of capacity-to-work per day, a volunteer does 5.6-13.5 days of labor per client treated. 1.5 hours × 6 sessions × (5 to 12 people) / 8 hour workday = 5.6-13.5 days of labor per client treated
So if we imagine that volunteers are actually missing work opportunities and sacrificing income in order to do this, $4.37-$10.53 donation to GiveDirectly per client treated might represent the cost to “offset” these potential “harms”. But hopefully volunteers aren’t actually choosing to forgo vital economic opportunities and taking on hardship in order to take on volunteer labor, and are doing it because they like it and want to and get something out of it, so maybe the true opportunity cost/disvalue to them is <20% of the equivalent labor-hours, if not zero?
anyway, this all seems kind of esoteric and theoretical—just something i had considered adding to the analysis (because at first I had an intuition, perhaps one which you had as well to motivate your question, that volunteer labor isn’t really free, just because it is unpaid), and then discarded because it seemed like the effect would be too negligible to be worth the complications it added. Like i said in the other comment, all of that is quite different from the other matter of what it might cost to pay a staff member or volunteer—staff and volunteers presumably have more earning potential as givedirectly recipients who are selected to be low income, and might be able command salaries much higher than that. And I don’t, actually, know the answer with respect to what the program might cost if all volunteers were instead paid, or exactly to what extent main costs are actually about the labor of people directly playing therapist roles (total staff salaries vs other costs might be easier to grab, but that’s answering a different quesiton than you asked, staff infrastructure won’t be all about leading therapy sessions).
To clarify, when i said it “would have been negligible” what I meant more precisely is that “doing voluntary unpaid labor does negligible harm (and probably actually confers benefits) to the volunteer, so I didn’t factor it into the cost estimate”. This is very different from the question of “how much would the cost change if the volunteers were to be paid”—which is a question that I do not know the answer to.
That’s an interesting question and i don’t know! I had myself implicitly been assuming it was from volunteer labor being unpaid for. But now that you put the question to me explicitly I realize that it could easily be due other factors (such as fewer expenses in some other category such as infrastructure, or salaries per staff, hypothetically). I don’t think the answer to this question would change anything about our calculations, but if we happen to find out I’ll let you know.
In an earlier draft I was considering counting the volunteer labor as a cost equal to the size of the benefits from the cash that the volunteer might have earned otherwise, but I left it off because it would have been too small of an effect to make a notable difference.
Yeah, I was asking w/r/t that counterfactual cost of labour. Perhaps this is a dumb question, but if that cost would’ve been negligible, as you say, then what explains the gap? Is the counterfactual cost of volunteer labour really that much smaller than staff labour?
If you’re interested in the question of why I figure that doing voluntary unpaid labor for most people involves negligible harm or opportunity cost of the type that ought to change our effectiveness estimate, my napkin calculation goes something like: suppose I personally were to replace half my compensated work hours with volunteering. That would be several tens of thousands of dollars of value to me and would half my income, which would be a hardship to me. But it should only take ~$285.92 to “offset” that harm to me personally, by doubling someone else’s income—theoretically the harm i experience for losing one day’s labor can be fully offset by giving a givedirectly recipient 0.78 cents. Obviously my own labor is worth more than 0.78 cents a day on the market, but matters being how they are, 0.78 cents seems to be around what it takes to “offset” the humanitarian cost to me personally (and the same to any volunteer, with respect to them-personally) of losing a day’s labor.
Having established that, how much unpaid labor are we talking about?
Assuming 8 hours of capacity-to-work per day, a volunteer does 5.6-13.5 days of labor per client treated. 1.5 hours × 6 sessions × (5 to 12 people) / 8 hour workday = 5.6-13.5 days of labor per client treated
So if we imagine that volunteers are actually missing work opportunities and sacrificing income in order to do this, $4.37-$10.53 donation to GiveDirectly per client treated might represent the cost to “offset” these potential “harms”. But hopefully volunteers aren’t actually choosing to forgo vital economic opportunities and taking on hardship in order to take on volunteer labor, and are doing it because they like it and want to and get something out of it, so maybe the true opportunity cost/disvalue to them is <20% of the equivalent labor-hours, if not zero?
anyway, this all seems kind of esoteric and theoretical—just something i had considered adding to the analysis (because at first I had an intuition, perhaps one which you had as well to motivate your question, that volunteer labor isn’t really free, just because it is unpaid), and then discarded because it seemed like the effect would be too negligible to be worth the complications it added. Like i said in the other comment, all of that is quite different from the other matter of what it might cost to pay a staff member or volunteer—staff and volunteers presumably have more earning potential as givedirectly recipients who are selected to be low income, and might be able command salaries much higher than that. And I don’t, actually, know the answer with respect to what the program might cost if all volunteers were instead paid, or exactly to what extent main costs are actually about the labor of people directly playing therapist roles (total staff salaries vs other costs might be easier to grab, but that’s answering a different quesiton than you asked, staff infrastructure won’t be all about leading therapy sessions).
To clarify, when i said it “would have been negligible” what I meant more precisely is that “doing voluntary unpaid labor does negligible harm (and probably actually confers benefits) to the volunteer, so I didn’t factor it into the cost estimate”. This is very different from the question of “how much would the cost change if the volunteers were to be paid”—which is a question that I do not know the answer to.