Thanks for the transparency! I noticed two issues, which may be edge cases or non-issues for GWWC, but could be relevant elsewhere:
(1) One of my bright-line rules for nonprofit compensation is that an employee should not earn more compensation than the private sector would award for comparable work. In my view as a donor, breaking this rule makes the employee a beneficiary of the charity, which is both inconsistent with donor intent and risks the usual problems when a charity is run partially in the interests of its employees. The lowest pay band is close to the average salary in Oxford even at zero experience,[1] the average worker in Oxford might be more educated than the UK average, and at least the US benefit package looks fairly generous for median-wage work. That means that, for my proposed bright-line test to be satisfied, the lowest-paid GWWC employee needs to be performing work that the market would compensate at least at the median pay/benefits level in Oxford. That may well be the case at GWWC, but I submit that making sure the formula doesn’t produce above-market compensation with any inputs might be a worthwhile bounds-checking exercise.
(2) I’d consider a further case-by-case adjustment for graduate-school debt where the employee incurred debt for a graduate degree that is substantially related to their duties. [2]From a US perspective where debt can be catastrophic, the competitive skills bonus doesn’t really address this. GWWC probably isn’t hiring (e.g.) lawyers or physicians to practice their professions, so this is probably not needed at present. That really would be case-by-case, as some people get help from public-interest programs at their law schools, and others would get significant help from government programs.[3] But it might be worth a sentence in the plan so that, in the event such an adjustment needed to be made, it wouldn’t look arbitrary.
I’m confused insofar as the text says that the median is used to define pay grade 2B, but the spreadsheet identifies a median salary in Oxford of £34K, while band 1 is listed at £33,598 and band 2 ranges from £36,958 to £40,654. (Maybe these actual salaries instead of defining the bands?)
I am assuming that the bulk of employees have undergraduate degrees, and so the cost of undergraduate education is baked in to the standard salaries in a sense.
My recollection is that having a higher-earning spouse can spoil both of these sources of assistance, but things may have changed since the mid-2000s when I graduated from law school.
Can you say more about why it’s bad for employees to benefit from the charity? Does this philosophy apply to other procurement, or only labor?
As a donor myself, I care about results and I’m completely fine with a charity paying obscene bonuses if that’s what it takes to get results.
As Dan Pallotta noted in his TED talk, I think there’s a weird double standard in social recognition for charity work, where the person giving 10% of their income gets more accolades than someone making a 20% pay cut to do charitable work, if the resulting pay is still considered “high” by nonprofit sector standards.
I don’t think what I said was inconsistent with Pallotta’s insight. I think what I actually proposed was rather narrow, and I’m having a difficult time applying most of your questions to it.
On procurement, same standard as for employees: If market rate for widgets is $20, it’s not effective (or even generally appropriate) for a charity to pay $25 for widgets. Paying $20 is consistent with my bright-line rule. One would not speak of the widget-seller who sells to a charity at market rate as a beneficiary of the charity; nor would an employee who makes no more than market rate be such.
In contrast, there are organizations that employ individuals (e.g., individuals with disabilities) at more than the market rate for their services for the purpose of benefitting those individuals. Those employees are properly considered beneficiaries of the program. If you’re running a vocational rehabilitation charity, that’s fine—because the very purpose of your charity is to benefit your employees. But that model is not appropriate for most charities. While vendors and employees may incidentally benefit from the charity (by receiving fair compensation for their wares and labors), they should be giving something of at least equal value (their wares and labors) as determined by the market.
What I said would be perfectly consistent with someone taking a 20% pay cut from market rate to do charitable work, or even that person being paid full market rate. I would generally oppose someone being paid more than market rate to do the same work. And that can be a problem with lower grades in standardized pay scales, especially those in which compensation for the top grades is only 2-3 times higher than that for the lowest grades. Looking at the payroll of many government organizations reveals this potential issue with those kinds of pay algorithms.
If the market is paying “obscene bonuses,” then the charity paying similar bonuses to similarly-situated employees with similar levels of performance is acting consistently with a cap of market-rate compensation.[1]
I do think that too many nonprofits—not speaking of EA, just in general—are run too much in the interests of employees and not enough in the interests of the claimed beneficiaries. To pick two examples from different sides of the social spectrum, I specifically think this about many churches and universities. In contrast, in a medium+ sized for-profit, executives usually have strong financial incentivizes to not authorize salaries that do not further the corporation’s profit objectives. In smaller ones, the owner usually is involved enough that the principal-agent problem is mitigated.
There isn’t any similar mechanism available for most non-profits. Boards, donors, and other potential checks often aren’t that effective. Thus, I think a norm that non-profits shouldn’t pay more for the same work than for-profits would pay for that work makes sense. Given generally prevailing patterns of non-profit compensation, that boundary check will rarely be an issue . . . but I think it is an important check to have in place.
Moreover, this is neither the rationale that GWWC gave for its pay scale nor one that would make much sense for a grade-based pay scale such as the one under discussion. If a charity is handing out “obscene” salaries and bonuses, it should be done on employee-specific evaluative criteria linked to the employee’s specific contribution to meeting business objectives, not to an algorithm.
FWIW Regarding your “bright-line” rule: For ~all our current roles/hires (even the most junior ones) their earning potential in the private sector is higher than the output of the calculator. We’re not typically hiring people who could attract <average wages as the counterfactual because we have been hiring exceptional people for specialist roles. As we grow we might hire for different roles where this isn’t the case and may at that point reconsider. I certainly don’t expect our current calculator/settings to last more than 1 year (at the very least we’re updating for things like inflation and currency shifts, we will also use these comments in the review process too).
Thanks—that makes a lot of sense. One of the things I’m thinking about in the back of my head is the possibility that other orgs may adapt the algorithm for broader use. So I think it’s helpful to document considerations like this. Doing so mitigates the risk of others applying the algorithm without first considering whether important assumptions upon which the originator org relied also hold true for their org.
Thanks for the transparency! I noticed two issues, which may be edge cases or non-issues for GWWC, but could be relevant elsewhere:
(1) One of my bright-line rules for nonprofit compensation is that an employee should not earn more compensation than the private sector would award for comparable work. In my view as a donor, breaking this rule makes the employee a beneficiary of the charity, which is both inconsistent with donor intent and risks the usual problems when a charity is run partially in the interests of its employees. The lowest pay band is close to the average salary in Oxford even at zero experience,[1] the average worker in Oxford might be more educated than the UK average, and at least the US benefit package looks fairly generous for median-wage work. That means that, for my proposed bright-line test to be satisfied, the lowest-paid GWWC employee needs to be performing work that the market would compensate at least at the median pay/benefits level in Oxford. That may well be the case at GWWC, but I submit that making sure the formula doesn’t produce above-market compensation with any inputs might be a worthwhile bounds-checking exercise.
(2) I’d consider a further case-by-case adjustment for graduate-school debt where the employee incurred debt for a graduate degree that is substantially related to their duties. [2]From a US perspective where debt can be catastrophic, the competitive skills bonus doesn’t really address this. GWWC probably isn’t hiring (e.g.) lawyers or physicians to practice their professions, so this is probably not needed at present. That really would be case-by-case, as some people get help from public-interest programs at their law schools, and others would get significant help from government programs.[3] But it might be worth a sentence in the plan so that, in the event such an adjustment needed to be made, it wouldn’t look arbitrary.
I’m confused insofar as the text says that the median is used to define pay grade 2B, but the spreadsheet identifies a median salary in Oxford of £34K, while band 1 is listed at £33,598 and band 2 ranges from £36,958 to £40,654. (Maybe these actual salaries instead of defining the bands?)
I am assuming that the bulk of employees have undergraduate degrees, and so the cost of undergraduate education is baked in to the standard salaries in a sense.
My recollection is that having a higher-earning spouse can spoil both of these sources of assistance, but things may have changed since the mid-2000s when I graduated from law school.
Can you say more about why it’s bad for employees to benefit from the charity? Does this philosophy apply to other procurement, or only labor?
As a donor myself, I care about results and I’m completely fine with a charity paying obscene bonuses if that’s what it takes to get results.
As Dan Pallotta noted in his TED talk, I think there’s a weird double standard in social recognition for charity work, where the person giving 10% of their income gets more accolades than someone making a 20% pay cut to do charitable work, if the resulting pay is still considered “high” by nonprofit sector standards.
I don’t think what I said was inconsistent with Pallotta’s insight. I think what I actually proposed was rather narrow, and I’m having a difficult time applying most of your questions to it.
On procurement, same standard as for employees: If market rate for widgets is $20, it’s not effective (or even generally appropriate) for a charity to pay $25 for widgets. Paying $20 is consistent with my bright-line rule. One would not speak of the widget-seller who sells to a charity at market rate as a beneficiary of the charity; nor would an employee who makes no more than market rate be such.
In contrast, there are organizations that employ individuals (e.g., individuals with disabilities) at more than the market rate for their services for the purpose of benefitting those individuals. Those employees are properly considered beneficiaries of the program. If you’re running a vocational rehabilitation charity, that’s fine—because the very purpose of your charity is to benefit your employees. But that model is not appropriate for most charities. While vendors and employees may incidentally benefit from the charity (by receiving fair compensation for their wares and labors), they should be giving something of at least equal value (their wares and labors) as determined by the market.
What I said would be perfectly consistent with someone taking a 20% pay cut from market rate to do charitable work, or even that person being paid full market rate. I would generally oppose someone being paid more than market rate to do the same work. And that can be a problem with lower grades in standardized pay scales, especially those in which compensation for the top grades is only 2-3 times higher than that for the lowest grades. Looking at the payroll of many government organizations reveals this potential issue with those kinds of pay algorithms.
If the market is paying “obscene bonuses,” then the charity paying similar bonuses to similarly-situated employees with similar levels of performance is acting consistently with a cap of market-rate compensation.[1]
I do think that too many nonprofits—not speaking of EA, just in general—are run too much in the interests of employees and not enough in the interests of the claimed beneficiaries. To pick two examples from different sides of the social spectrum, I specifically think this about many churches and universities. In contrast, in a medium+ sized for-profit, executives usually have strong financial incentivizes to not authorize salaries that do not further the corporation’s profit objectives. In smaller ones, the owner usually is involved enough that the principal-agent problem is mitigated.
There isn’t any similar mechanism available for most non-profits. Boards, donors, and other potential checks often aren’t that effective. Thus, I think a norm that non-profits shouldn’t pay more for the same work than for-profits would pay for that work makes sense. Given generally prevailing patterns of non-profit compensation, that boundary check will rarely be an issue . . . but I think it is an important check to have in place.
Moreover, this is neither the rationale that GWWC gave for its pay scale nor one that would make much sense for a grade-based pay scale such as the one under discussion. If a charity is handing out “obscene” salaries and bonuses, it should be done on employee-specific evaluative criteria linked to the employee’s specific contribution to meeting business objectives, not to an algorithm.
FWIW Regarding your “bright-line” rule: For ~all our current roles/hires (even the most junior ones) their earning potential in the private sector is higher than the output of the calculator. We’re not typically hiring people who could attract <average wages as the counterfactual because we have been hiring exceptional people for specialist roles. As we grow we might hire for different roles where this isn’t the case and may at that point reconsider. I certainly don’t expect our current calculator/settings to last more than 1 year (at the very least we’re updating for things like inflation and currency shifts, we will also use these comments in the review process too).
(Typing on my phone, excuse the brevity)
Thanks—that makes a lot of sense. One of the things I’m thinking about in the back of my head is the possibility that other orgs may adapt the algorithm for broader use. So I think it’s helpful to document considerations like this. Doing so mitigates the risk of others applying the algorithm without first considering whether important assumptions upon which the originator org relied also hold true for their org.
Agreed!