With regard to the pledged amount, this comes from members’ predictions of their future annual salary, which we think are likely to be underestimates. We also use median wage as a stand-in if we’re missing future salary data, which (given our members are in general likely to earn more than median wage) we also think is conservative. Accordingly, it’s likely that the amount donated will be higher in reality.
We address this in more detail in our fundraising prospectus – see Appendix 2 for the full working.
From page 23 of the prospectus:
This methodology obviously relies on the accuracy of members’ predictions about their future income. In general we have found that these predictions seem conservative, as most people underestimate their future earning potential (1). If members do not estimate their future salary, we use the median salary for their country. We think that is a fairly pessimistic assumption, as our median member has an expected earning potential higher than the median wage (2).
And the footnotes to the above:
(1) For example, many members estimate their future income will be
the same as their current income, even though they are at the beginning of their careers—in reality, income typically increases throughout a person’s career
(2) For example, many members attend prestigious universities and/or are pursuing careers that have an average salary much higher than the median wage
The final $146m figure is arrived at by multiplying members’ estimates of future salary by the number of years they have left in their careers. It therefore doesn’t take into account any of this growth that you’d expect in reality. As such, it wouldn’t make sense to go back and try to model growth based on the $146 million figure (say, $1.7 million in year one, growing to around $5.6 million/year by year 40, rather than a flat $3.7m per year)*.
Instead, you’d need to apply your model (say, fast wage growth in the first 10-20 years of a career, then slower growth until retirement) to the member estimates first to derive the final figure, and use the yearly amounts in your calculation. Given our assumption that member estimates of future salary err on the low side, this means that both the final pledged amount, and the per-year amounts are likely to be higher, and therefore our effectiveness would in fact be higher, notwithstanding that the discounting/attrition rate would affect the final number more aggressively.
* I’ve tried this calculation, assuming a 4% growth rate for years 1-20 and a 2% growth rate in years 21-40. With a year one pledge of $1.7 million, this grows to $5.6 million by year 40, for a total of ~$147 million donated. This drops the effectiveness estimate down to 44-1 - a significant drop, but still excellent return for a donation to Giving What We Can. To reiterate, I think this would be a significant underestimate of peoples’ future incomes.
Thanks for the question Jon.
With regard to the pledged amount, this comes from members’ predictions of their future annual salary, which we think are likely to be underestimates. We also use median wage as a stand-in if we’re missing future salary data, which (given our members are in general likely to earn more than median wage) we also think is conservative. Accordingly, it’s likely that the amount donated will be higher in reality.
We address this in more detail in our fundraising prospectus – see Appendix 2 for the full working.
From page 23 of the prospectus:
And the footnotes to the above:
The final $146m figure is arrived at by multiplying members’ estimates of future salary by the number of years they have left in their careers. It therefore doesn’t take into account any of this growth that you’d expect in reality. As such, it wouldn’t make sense to go back and try to model growth based on the $146 million figure (say, $1.7 million in year one, growing to around $5.6 million/year by year 40, rather than a flat $3.7m per year)*.
Instead, you’d need to apply your model (say, fast wage growth in the first 10-20 years of a career, then slower growth until retirement) to the member estimates first to derive the final figure, and use the yearly amounts in your calculation. Given our assumption that member estimates of future salary err on the low side, this means that both the final pledged amount, and the per-year amounts are likely to be higher, and therefore our effectiveness would in fact be higher, notwithstanding that the discounting/attrition rate would affect the final number more aggressively.
* I’ve tried this calculation, assuming a 4% growth rate for years 1-20 and a 2% growth rate in years 21-40. With a year one pledge of $1.7 million, this grows to $5.6 million by year 40, for a total of ~$147 million donated. This drops the effectiveness estimate down to 44-1 - a significant drop, but still excellent return for a donation to Giving What We Can. To reiterate, I think this would be a significant underestimate of peoples’ future incomes.
Also, sorry if this reply doesn’t exactly address your rephrased question – I wrote it in response to your first comment :)
Here’s a copy of the spreadsheet with the calculations added in as above if you want to play around with it.
Thanks again for the question, let me know if there’s anything else you want clarified.
Thanks Sam! Very helpful, I hadn’t realized how much conservatism was built into the income estimates.