Especially in a democracy, even trying to divorce foreign aid spending from "donor government’s strategic goals" is a mistake. First, it's a dereliction of the fiduciary duty of elected officials to look out, first and foremost, for the interests of their own voter constituents. Additionally, right out of the gate it reduces the size of the potential domestic coalition that might support a given item in the aid budget. For example, people who might not care much, per se, about sending their tax money to improve the lives of rural farmers in Central America or anywhere else might nonetheless support aid to Central America if they see it as supporting a national strategic goal of bringing Mexico and it's southern neighbors up the economic level of the US and Canada, whether for purposes of competing with China, of solving the immigration or drug smuggling problems on the border, or some other strategic purpose.
In short, it's just better all around, smarter politics and smarter policy, if the foreign aid budget is designed from the ground up to kill two or three birds with one stone.
Conscientiousness is not efficacy. The conscientious worker over-engineers work processes to avoid errors and scandals. The effective boss pushes back against this hard and implements production quotas that crush low impact refinements.
Poor children in the developing world have very simple needs. Whoever can get rice, beans, vaccines and de worming drugs to them at the lowest cost deserves an “unseemly” profit.
Agree with the objective here, but what would the mechanism be on the government side that would allow for (reasonably) quick and accurate verification that the development had been achieved? A mistaken payment would reap huge political backlash, but an overly cautious process that takes forever to pay out wouldn't necessarily get many takers, or result in the same players as today filling the new space.
Serious question, Matt: Do you see this at all different from the problem of domestic non-profits sucking up government funding and delivering nothing?
PEPFAR seems to have achieved major results with GiveWell charity-like cost effectiveness. What differentiated it from some of the other less effective aid programs?
This is all great and I'll resist the temptation to tell war stories from Afghanistan.
The only thing I'd add to Matt's comments is that progressives need to stop thinking of the needs of poorer people in poor countries as fundamentally different from the needs of poorer people in Europe or North America. The way to lift people out of poverty is to set up well-functioning national welfare states, not to send money across borders.
Elites in poor countries don't want to hear this because it would mean paying higher taxes. Western aid workers don't want to hear it because it implies they should be working themselves out of a job. But if cash transfers and health care are the two main things poor people in the Third World need, then the role of aid in anything but an emergency setting should be to help Third World governments organize effective systems of UBI and national health insurance. Rwanda's government understands this, which is why they have universal healthcare.
So I can't help but note that the "cost-plus" contracting formulation sounds a hell of a lot like the failed vision of technocratic education reform from yesterday and that "paying for results" sounds much more like an effective voucher program.
Worked for a big government aerospace and defense contractor in their schedule management department for 3 years during and right after college. I can confirm from experience that cost-plus contracts had little to no urgency to finish on time or in budget, because of the two factors MY mentioned - cost-plus means that these orgs are getting their profit no matter what and the barriers to entry for many government contracts are so high that organizations aren't worried about losing work if they consistently deliver late and over budget.
There's a spectrum from cost-plus to what MY is talking about (Firm Fixed Price or FFP), and sometimes project are more appropriate to operate under something of a hybrid than a FFP model, which can lead to projects being scrapped mid-way through if they become unprofitable, but cost-plus is truly never the right way to go. Honestly though, and this is anecdata from my experience in one company in one sector, the RFP process is more important to reform than the contract structures. The government contracting process is set up so thoroughly to discourage new entrants. The requirements for RFPs are unnecessarily onerous, the process involves a lot of personal relationships (again, at least on the defense/aerospace side), and the government does not emphasize past performance enough (for example, when a contractor has been late and over budget 10x in a row, they should stop giving them bids).
"Redirecting the money that’s currently being spent toward better programs could do a lot of good in the world. And my general inclination about a lot of categories of spending — from aid to mass transit and beyond — is that if the money was delivering more visible results, that would improve the politics."
This is probably correct but I'm not sure there's any foreign aid-related result that would be visible enough to get voters to really care about this. It's simply not a topic that's usually in the forefront of people's minds; they'll write their check to their favorite charity and then not think about it again for several months.
Interesting post but I don't see it happening anytime soon as the politics of "foreign aid" have always been more about domestic concerns and foreign policy than Give Well type goals. Like there's no rational reason from a human suffering or development standpoint for the US to still give massive amounts of aid to Israel...but it totally does make sense from a domestic politics standpoint. You could make similar points about Egypt, Columbia etc etc
It’s pretty frustrating to me that there are AFAIK no national governments funding GiveWell in significant size. They’ve attracted a fair amount of private sector money.
I guess as Matt says the public sector process is just too rigid and wasteful to notice this great opportunity. It makes me very skeptical of the left-wing idea that we should try and replace all philanthropy with taxes.
Another issue is that governments don’t care much about people who aren’t their citizens. Maybe this is correct, but I think we should try and spend at least a little of our resources in a country-agnostic way because between-country differences are so large (eg median Americans are literally 100x richer than the poorest people in the world).
One concern I had about Obamacare is that it essentially turns health insurance companies into cost plus providers by requiring carriers spend at least 80% of premiums on medical costs or return the excess. This seemed like it would turn carriers from a market force pushing against rising health care costs (that they are ultimately required to pay) into one that wants higher costs so their 20% goes up with them.
It does appear that the ACA bent the cost curve, so maybe my worry was misplaced, though I've never really looked into whether the cost curve bent because of or in spite of this rule.
Does just giving people money really scale that well? I would think that at some point you're just causing inflation rather than an increase in living standards
For additional context, the other contract type is called "lump sum" typically. "Lump sum" contracts entail a fixed cost to achieve the goal and I think that's what Matt is advocating for. Some public sector entities (like many cities and municipalities) do use lump sum contracts. Unfortunately, large public sector entities like major utility districts and DOTs use cost-plus contacts. As a consultant, I unabashedly prefer lump sum contacts because they give us much more flexibility.
The Republicans’ debt ceiling hostage taking should never be described as anything but a huge, completely dishonest graft. It won’t result in a new park getting built or any other good thing being done efficiently. It’ll just be ten-millions of dollars of profits for everyone who knows a day in advance which markets to short, to profit from whatever the rightist thieves decide to do. This includes the so-called “blind trusts” that politicians erect to maintain the pretense of not doing this.
Many of these same arguments could be applied to the community development finance industry in the US. A whole industry has sprung up over the past 40 years for investing more capital in underserved neighborhoods. This has taken the form of community development financial institutions (CDFIs) and various tax credit programs such as LIHTC , New Markets Tax Credits, and Opportunity Zones. The industry has, no doubt, directed a lot more capital to building affordable housing, health centers, charter schools, and other community facilities than otherwise would not have been built. But it's time to ask whether there now may be more efficient ways of investing in those communities (e.g. direct transfers.)
Especially in a democracy, even trying to divorce foreign aid spending from "donor government’s strategic goals" is a mistake. First, it's a dereliction of the fiduciary duty of elected officials to look out, first and foremost, for the interests of their own voter constituents. Additionally, right out of the gate it reduces the size of the potential domestic coalition that might support a given item in the aid budget. For example, people who might not care much, per se, about sending their tax money to improve the lives of rural farmers in Central America or anywhere else might nonetheless support aid to Central America if they see it as supporting a national strategic goal of bringing Mexico and it's southern neighbors up the economic level of the US and Canada, whether for purposes of competing with China, of solving the immigration or drug smuggling problems on the border, or some other strategic purpose.
In short, it's just better all around, smarter politics and smarter policy, if the foreign aid budget is designed from the ground up to kill two or three birds with one stone.
Conscientiousness is not efficacy. The conscientious worker over-engineers work processes to avoid errors and scandals. The effective boss pushes back against this hard and implements production quotas that crush low impact refinements.
Poor children in the developing world have very simple needs. Whoever can get rice, beans, vaccines and de worming drugs to them at the lowest cost deserves an “unseemly” profit.
Agree with the objective here, but what would the mechanism be on the government side that would allow for (reasonably) quick and accurate verification that the development had been achieved? A mistaken payment would reap huge political backlash, but an overly cautious process that takes forever to pay out wouldn't necessarily get many takers, or result in the same players as today filling the new space.
Serious question, Matt: Do you see this at all different from the problem of domestic non-profits sucking up government funding and delivering nothing?
PEPFAR seems to have achieved major results with GiveWell charity-like cost effectiveness. What differentiated it from some of the other less effective aid programs?
This is all great and I'll resist the temptation to tell war stories from Afghanistan.
The only thing I'd add to Matt's comments is that progressives need to stop thinking of the needs of poorer people in poor countries as fundamentally different from the needs of poorer people in Europe or North America. The way to lift people out of poverty is to set up well-functioning national welfare states, not to send money across borders.
Elites in poor countries don't want to hear this because it would mean paying higher taxes. Western aid workers don't want to hear it because it implies they should be working themselves out of a job. But if cash transfers and health care are the two main things poor people in the Third World need, then the role of aid in anything but an emergency setting should be to help Third World governments organize effective systems of UBI and national health insurance. Rwanda's government understands this, which is why they have universal healthcare.
So I can't help but note that the "cost-plus" contracting formulation sounds a hell of a lot like the failed vision of technocratic education reform from yesterday and that "paying for results" sounds much more like an effective voucher program.
Worked for a big government aerospace and defense contractor in their schedule management department for 3 years during and right after college. I can confirm from experience that cost-plus contracts had little to no urgency to finish on time or in budget, because of the two factors MY mentioned - cost-plus means that these orgs are getting their profit no matter what and the barriers to entry for many government contracts are so high that organizations aren't worried about losing work if they consistently deliver late and over budget.
There's a spectrum from cost-plus to what MY is talking about (Firm Fixed Price or FFP), and sometimes project are more appropriate to operate under something of a hybrid than a FFP model, which can lead to projects being scrapped mid-way through if they become unprofitable, but cost-plus is truly never the right way to go. Honestly though, and this is anecdata from my experience in one company in one sector, the RFP process is more important to reform than the contract structures. The government contracting process is set up so thoroughly to discourage new entrants. The requirements for RFPs are unnecessarily onerous, the process involves a lot of personal relationships (again, at least on the defense/aerospace side), and the government does not emphasize past performance enough (for example, when a contractor has been late and over budget 10x in a row, they should stop giving them bids).
"Redirecting the money that’s currently being spent toward better programs could do a lot of good in the world. And my general inclination about a lot of categories of spending — from aid to mass transit and beyond — is that if the money was delivering more visible results, that would improve the politics."
This is probably correct but I'm not sure there's any foreign aid-related result that would be visible enough to get voters to really care about this. It's simply not a topic that's usually in the forefront of people's minds; they'll write their check to their favorite charity and then not think about it again for several months.
Interesting post but I don't see it happening anytime soon as the politics of "foreign aid" have always been more about domestic concerns and foreign policy than Give Well type goals. Like there's no rational reason from a human suffering or development standpoint for the US to still give massive amounts of aid to Israel...but it totally does make sense from a domestic politics standpoint. You could make similar points about Egypt, Columbia etc etc
It’s pretty frustrating to me that there are AFAIK no national governments funding GiveWell in significant size. They’ve attracted a fair amount of private sector money.
I guess as Matt says the public sector process is just too rigid and wasteful to notice this great opportunity. It makes me very skeptical of the left-wing idea that we should try and replace all philanthropy with taxes.
Another issue is that governments don’t care much about people who aren’t their citizens. Maybe this is correct, but I think we should try and spend at least a little of our resources in a country-agnostic way because between-country differences are so large (eg median Americans are literally 100x richer than the poorest people in the world).
One concern I had about Obamacare is that it essentially turns health insurance companies into cost plus providers by requiring carriers spend at least 80% of premiums on medical costs or return the excess. This seemed like it would turn carriers from a market force pushing against rising health care costs (that they are ultimately required to pay) into one that wants higher costs so their 20% goes up with them.
It does appear that the ACA bent the cost curve, so maybe my worry was misplaced, though I've never really looked into whether the cost curve bent because of or in spite of this rule.
Does just giving people money really scale that well? I would think that at some point you're just causing inflation rather than an increase in living standards
For additional context, the other contract type is called "lump sum" typically. "Lump sum" contracts entail a fixed cost to achieve the goal and I think that's what Matt is advocating for. Some public sector entities (like many cities and municipalities) do use lump sum contracts. Unfortunately, large public sector entities like major utility districts and DOTs use cost-plus contacts. As a consultant, I unabashedly prefer lump sum contacts because they give us much more flexibility.
The Republicans’ debt ceiling hostage taking should never be described as anything but a huge, completely dishonest graft. It won’t result in a new park getting built or any other good thing being done efficiently. It’ll just be ten-millions of dollars of profits for everyone who knows a day in advance which markets to short, to profit from whatever the rightist thieves decide to do. This includes the so-called “blind trusts” that politicians erect to maintain the pretense of not doing this.
Many of these same arguments could be applied to the community development finance industry in the US. A whole industry has sprung up over the past 40 years for investing more capital in underserved neighborhoods. This has taken the form of community development financial institutions (CDFIs) and various tax credit programs such as LIHTC , New Markets Tax Credits, and Opportunity Zones. The industry has, no doubt, directed a lot more capital to building affordable housing, health centers, charter schools, and other community facilities than otherwise would not have been built. But it's time to ask whether there now may be more efficient ways of investing in those communities (e.g. direct transfers.)