I speculate about why Open Philanthropy donated to criminal justice in the first place, and why it continued donating.
OPP, formerly GiveWell Labs, recommends grants. Good Ventures and other organizations that use these recommendations award the grants. The hypothesis that a funder was passionate about this cause and that OPP recommended effective solutions within this cause area rather than comparing the cause area to others seems prima facie plausible. However, I think that the influence is not as one-sided. OPP’s leadership, represented by Holden Karnofsky, started considering (US) political advocacy early on. Since there are many ways to develop trusting relationships with political decisionmakers, following up on prominent conversations relating to justice and benevolence, criminal justice could have been actually a great idea.
You should also note that recently, OPP delegated criminal justice reform to Just Impact, led by Chloe Cockburn and Jesse Rothman, the former OPP’s grant advisors in this cause area. This grant should suffice for 3.5 years and perhaps the organization gains its own funding or solves the issue. So, perhaps this decision makes your argument outdated.
Maybe, are you criticizing that OPP commits to causes? There were some doubts[1] about this in 2014. Currently, OPP conducts actually only shallow and medium-depth investigations, so the concern here can seem quite the opposite: where are the in-depth investigations?
I think that providing uncertain and perhaps flawed criticism is still worth it, in expectation
This should be the case if you are introducing new information in a way that is welcome by the decisionmakers and their informants. While you could rephrase some of your statements neutrally,[2] the readers will probably go for content. You also agreed to not defame or cause rep loss of anyone involved,[3] so you wrote the entry with this intention. This could also imply that you would engage with any responses of these stakeholders constructively.
In the report, at the first glance, I would focus on the time on parole (p. 17). It can be that an increased time on parole increases criminality.[4] Thus, increasing parole time may be a suboptimal solution.
The experiments’ interpretations on pp. 22–26 seem horrifying, at the first glance.[5] I am reading that deterrence, such as incarcerating someone for a day to discourage them, is bad while incapacitation (such as incarceration?) good.
On the second glance, I am also worried that the table shows the consistency of the studies with the report,[6] not using the studies to inform conclusions. If this only shows that the writers had an extra challenge to fit external biases, that may be ok, but it can also be that the writers are biased toward incarceration and use the perceived readers’ weaknesses and their appeal to authority. Thus, I suggest that the connections of the contributors to the profitable US prison system are overviewed.
On the third glance, the report simply analyzes various criminal justice factors that impact different measurable aspects of criminality, such as re-offences and their frequency/timing, but does not really discuss why people would commit crimes and how this could be prevented in ways other than those directly pertinent to criminal legislature and offenses.
In terms of the cost-effectiveness, reducing crime rate in the US can free significant capital,[7] which can be used cost-effectively, including for deworming programs. When you consider that many US officials would love when they could attribute some crime reduction (people could elect them), then they could be attuned to certain budgetary recommendations from the crime reduction organization.
For example, if incarceration is reduced by 10% in the US,[8] then, 600 million could be realistically annually spent on EA-related programs including deworming. [9]
The last one struck me as being both particularly bad and relatively easy to evaluate: A letter costs $2.5, about the same as deworming several kids at $0.35 to $0.97 per deworming treatment. But sending a letter intuitively seems significantly less impactful.
Have you considered that deworming may be a perpetual need while influencing a decision that motivates a sustainable systemic change a permanent solution? This could justify spending on advocacy, in general.
Furthermore, for change to happen, perhaps various media should be engaged. For example, letters should be complemented by local group support, by pro bono legal advisory that uses case studies to inform legal changes, by public awareness/electorate support that includes movies, by making sure that any new legislation is not discriminatory on the basis of race or otherwise, that judges are bought in, that any evidence-based appeal is complemented by US families’ emotional advocacy, etc. So, it can be that grants awarded at any given time always complemented current activities to optimize for marginal cost-effectiveness.
Is it that in EA, the marginal cost-effectiveness[10] is generally considered while factors such as the president’s pay are included in the analysis? For example, generally, a lawyer can be paid this salary. Legal manager who makes state and federal changes can be perceived to take a pay cut.
I apologize for I skipped your hypotheses.
It may be that at this point, the best course of action is reminding Just Impact to enter any conversations with the government with the awareness of favors return, the extent that reduced crime can increase budget, and the question on where they could see the greatest impact with some of this funding?[11]
It seems to me that the most common model in philanthropy … is to have staff who specialize in a particular cause (for example, specializing in criminal justice policy). … I think this model makes a great deal of sense, … Yet I’m not convinced that this model is the right one for us. Depth comes at the price of breadth. … I’ve been asking myself the question of whether there’s a way to be involved in many more causes at a much lower level of depth (October 30, 2014)
For example, “and to what extent it was determined by other factors, such as the idiosyncratic preferences of Open Philanthropy’s funders, human fallibility and slowness, paying too much to avoid social awkwardness,” can be safely omitted since it is implied that if any reasoning is flawed, it is due to biases.
Also, by stating “I end up uncertain about to what extent this was a sincere play based on considerations around the value of information and learning,” you mean that the decision could have seemed valuable initially but with learning over time it showed as not so marginally cost-effective? Or that it should have been clear since the beginning that the org will learn little by focusing on criminal justice reform? Any hypothesis can be stated explicitly.
In particular, see the assessment of Abrams (2012) (p. 23) that states “Strongly compatible after reanalysis.” This could point out that an external standard was being met by the study interpretation.
The societal cost of an inmate-year of incarceration is put at $92,000, with loss of liberty the largest item ($50,000) and the cost of incarceration next ($26,000 after netting out $5,000 in service transfers) (p. 135).
While deworming was used as an example, other expanded moral circle considerations can work. For example, I estimated, using references shared by a GiveDirectly employee, that targeting individuals at risk of criminal involvement who go through otherwise effective programs but turn to crime only because they do not have access to microfinance could multiply the cost-effectiveness many times. This also considers that the criminal activity of a single individual can reduce the wellbeing of entire communities while cash transfers targeted based on poverty rates only modest spillovers.
I find your comment a bit hard to read; but a quick answer to one particular point:
While you could rephrase some of your statements neutrally,[2] the readers will probably go for content. You also agreed to not defame or cause rep loss of anyone involved,[3] so you wrote the entry with this intention
Footnote 3 links to this document, which contains the following clause:
disparages or injures the reputation or goodwill of Sponsor, Prize Provider, or any of their respective officers, directors, employees, products, or services
Have you considered that deworming may be a perpetual need while influencing a decision that motivates a sustainable systemic change a permanent solution? This could justify spending on advocacy, in general.
It’s an interesting hypothesis, but I don’t think deworming is a perpetual need? I don’t think I took deworming pills growing up, and I doubt most Forum readers did.
Framed another way, I don’t think we should have a strong prior belief that if we subsidize health interventions for X years, this means they’ll need to be continuously subsidized by the globally rich, while “systematic” policy changes in the West are successful as one-offs.
I don’t think deworming is a perpetual need? I don’t think I took deworming pills growing up, and I doubt most Forum readers did.
That is true, infrastructure can be build and infections eliminated. That is echoed by the WHO (only some schools are recommended treatment while “sanitation and access to safe water [and] hygiene” can reduce transmission).
I possibly underestimated the facility of reducing contamination and overestimated the inevitability of avoiding contaminants. According to the above sources, sanitation, hygiene, and refraining from using animal fertilizer can reduce contamination. Further, wearing shoes and refraining from spending time in possibly contaminated water reduces the risk of infection by avoiding contaminants. Thus, only people that cannot reasonably avoid spending time in water, such as water-intensive crop farmers, who are in areas with high infection prevalence are at risk while solutions in other situations are readily available. Since farming can be automated, risk can be practically eliminated entirely.
I don’t think we should have a strong prior belief that if we subsidize health interventions for X years, this means they’ll need to be continuously subsidized by the globally rich, while “systematic” policy changes in the West are successful as one-offs.
I agree. According to Dr. Gabby Liang, the SCI Foundation currently works on “capacity development within the [program country] ministries.” This suggest that international assistance can be eventually phased out. It can also be that most health programs develop capacity and thus make lasting changes, even if they are not specifically targeted at that. Policy changes that are a result of organized advocacy, on the other hand, may be temporary/fragile, also since they are not based on the institution’s reasoning or research. So, I could agree with the greater effectiveness of SCI than the cited letter writing (but still would need more information to have either perspective).
OPP, formerly GiveWell Labs, recommends grants. Good Ventures and other organizations that use these recommendations award the grants. The hypothesis that a funder was passionate about this cause and that OPP recommended effective solutions within this cause area rather than comparing the cause area to others seems prima facie plausible. However, I think that the influence is not as one-sided. OPP’s leadership, represented by Holden Karnofsky, started considering (US) political advocacy early on. Since there are many ways to develop trusting relationships with political decisionmakers, following up on prominent conversations relating to justice and benevolence, criminal justice could have been actually a great idea.
You should also note that recently, OPP delegated criminal justice reform to Just Impact, led by Chloe Cockburn and Jesse Rothman, the former OPP’s grant advisors in this cause area. This grant should suffice for 3.5 years and perhaps the organization gains its own funding or solves the issue. So, perhaps this decision makes your argument outdated.
Maybe, are you criticizing that OPP commits to causes? There were some doubts[1] about this in 2014. Currently, OPP conducts actually only shallow and medium-depth investigations, so the concern here can seem quite the opposite: where are the in-depth investigations?
This should be the case if you are introducing new information in a way that is welcome by the decisionmakers and their informants. While you could rephrase some of your statements neutrally,[2] the readers will probably go for content. You also agreed to not defame or cause rep loss of anyone involved,[3] so you wrote the entry with this intention. This could also imply that you would engage with any responses of these stakeholders constructively.
In the report, at the first glance, I would focus on the time on parole (p. 17). It can be that an increased time on parole increases criminality.[4] Thus, increasing parole time may be a suboptimal solution.
The experiments’ interpretations on pp. 22–26 seem horrifying, at the first glance.[5] I am reading that deterrence, such as incarcerating someone for a day to discourage them, is bad while incapacitation (such as incarceration?) good.
On the second glance, I am also worried that the table shows the consistency of the studies with the report,[6] not using the studies to inform conclusions. If this only shows that the writers had an extra challenge to fit external biases, that may be ok, but it can also be that the writers are biased toward incarceration and use the perceived readers’ weaknesses and their appeal to authority. Thus, I suggest that the connections of the contributors to the profitable US prison system are overviewed.
On the third glance, the report simply analyzes various criminal justice factors that impact different measurable aspects of criminality, such as re-offences and their frequency/timing, but does not really discuss why people would commit crimes and how this could be prevented in ways other than those directly pertinent to criminal legislature and offenses.
In terms of the cost-effectiveness, reducing crime rate in the US can free significant capital,[7] which can be used cost-effectively, including for deworming programs. When you consider that many US officials would love when they could attribute some crime reduction (people could elect them), then they could be attuned to certain budgetary recommendations from the crime reduction organization.
For example, if incarceration is reduced by 10% in the US,[8] then, 600 million could be realistically annually spent on EA-related programs including deworming. [9]
Have you considered that deworming may be a perpetual need while influencing a decision that motivates a sustainable systemic change a permanent solution? This could justify spending on advocacy, in general.
Furthermore, for change to happen, perhaps various media should be engaged. For example, letters should be complemented by local group support, by pro bono legal advisory that uses case studies to inform legal changes, by public awareness/electorate support that includes movies, by making sure that any new legislation is not discriminatory on the basis of race or otherwise, that judges are bought in, that any evidence-based appeal is complemented by US families’ emotional advocacy, etc. So, it can be that grants awarded at any given time always complemented current activities to optimize for marginal cost-effectiveness.
Is it that in EA, the marginal cost-effectiveness[10] is generally considered while factors such as the president’s pay are included in the analysis? For example, generally, a lawyer can be paid this salary. Legal manager who makes state and federal changes can be perceived to take a pay cut.
I apologize for I skipped your hypotheses.
It may be that at this point, the best course of action is reminding Just Impact to enter any conversations with the government with the awareness of favors return, the extent that reduced crime can increase budget, and the question on where they could see the greatest impact with some of this funding?[11]
For example, “and to what extent it was determined by other factors, such as the idiosyncratic preferences of Open Philanthropy’s funders, human fallibility and slowness, paying too much to avoid social awkwardness,” can be safely omitted since it is implied that if any reasoning is flawed, it is due to biases.
Also, by stating “I end up uncertain about to what extent this was a sincere play based on considerations around the value of information and learning,” you mean that the decision could have seemed valuable initially but with learning over time it showed as not so marginally cost-effective? Or that it should have been clear since the beginning that the org will learn little by focusing on criminal justice reform? Any hypothesis can be stated explicitly.
5. D. vii. (6)
This could show that any human bias was in favor of releasing criminals on parole and the report aimed to mitigate it.
It may be because I am reading them wrong.
In particular, see the assessment of Abrams (2012) (p. 23) that states “Strongly compatible after reanalysis.” This could point out that an external standard was being met by the study interpretation.
From 2.3 million to 2.1 million.
0.23million∗$26,000/year≈6billion/year. If the crime reduction organization influences 10%, additional 600 million can be spent.
alongside with opportunity/timing, leverage, and other factors
While deworming was used as an example, other expanded moral circle considerations can work. For example, I estimated, using references shared by a GiveDirectly employee, that targeting individuals at risk of criminal involvement who go through otherwise effective programs but turn to crime only because they do not have access to microfinance could multiply the cost-effectiveness many times. This also considers that the criminal activity of a single individual can reduce the wellbeing of entire communities while cash transfers targeted based on poverty rates only modest spillovers.
I find your comment a bit hard to read; but a quick answer to one particular point:
Footnote 3 links to this document, which contains the following clause:
But this document is for the cause exploration prize, not for the criticism price. Also, I would be very hesitant to agree to such a condition.
Oh, sorry, I thought you are responding to the Cause Exploration.
It’s an interesting hypothesis, but I don’t think deworming is a perpetual need? I don’t think I took deworming pills growing up, and I doubt most Forum readers did.
Framed another way, I don’t think we should have a strong prior belief that if we subsidize health interventions for X years, this means they’ll need to be continuously subsidized by the globally rich, while “systematic” policy changes in the West are successful as one-offs.
That is true, infrastructure can be build and infections eliminated. That is echoed by the WHO (only some schools are recommended treatment while “sanitation and access to safe water [and] hygiene” can reduce transmission).
I possibly underestimated the facility of reducing contamination and overestimated the inevitability of avoiding contaminants. According to the above sources, sanitation, hygiene, and refraining from using animal fertilizer can reduce contamination. Further, wearing shoes and refraining from spending time in possibly contaminated water reduces the risk of infection by avoiding contaminants. Thus, only people that cannot reasonably avoid spending time in water, such as water-intensive crop farmers, who are in areas with high infection prevalence are at risk while solutions in other situations are readily available. Since farming can be automated, risk can be practically eliminated entirely.
I agree. According to Dr. Gabby Liang, the SCI Foundation currently works on “capacity development within the [program country] ministries.” This suggest that international assistance can be eventually phased out. It can also be that most health programs develop capacity and thus make lasting changes, even if they are not specifically targeted at that. Policy changes that are a result of organized advocacy, on the other hand, may be temporary/fragile, also since they are not based on the institution’s reasoning or research. So, I could agree with the greater effectiveness of SCI than the cited letter writing (but still would need more information to have either perspective).