Some Back-of-the-envelope Calculations

Yes, an upper-middle-class American who gives aggressively to highly effective life-saving charities over a forty-year career can save roughly as many lives as Corrie ten Boom and her network rescued from the Holocaust.

Corrie ten Boom, her father Casper, her sister Betsie, and the network they led are estimated to have saved around 800 Jewish refugees during the Nazi occupation of the Netherlands. We use 800 as the target throughout, while noting that the precise figure is contested in the historical literature and that "saved" itself is doing some work (some of those they sheltered would have survived without them; their network's effectiveness varied across the war).

GiveWell, the most rigorous public charity evaluator we know of, currently estimates the cost per life saved at its top recommended charities at roughly $3,500 to $5,500. Across 2022 to 2024, Malaria Consortium's seasonal malaria chemoprevention program averaged $4,000 per life saved, New Incentives' vaccination program averaged $4,500 per life saved, and the Against Malaria Foundation averaged $5,500 per life saved (GiveWell, "Our Top Charities," updated 2024; GiveWell, "How Much Does It Cost to Save a Life?", 2024). We use $5,000 per life saved as a round, defensible figure throughout this calculation. The number represents children's lives, mostly under five, mostly in sub-Saharan Africa. Cost-effectiveness estimates carry wide uncertainty bands and are expected to rise over time as the most tractable interventions become saturated.

We model an "upper-middle-class American" as a household earning $200,000 per year in inflation-adjusted dollars across a 40-year career, which roughly tracks the top quintile of U.S. household income (U.S. Census Bureau, Income in the United States, 2024). Many readers will earn less; some will earn more; the calculation scales linearly. We assume the household gives "aggressively" to highly effective charities. The basic tithe, equivalent to the Giving What We Can pledge, is 10% of income. Some commit to substantially more: Toby Ord publicly committed to giving everything above a modest professorial salary, well over half of his income, through the Further Pledge.

At a 10% giving rate, our model household gives $20,000 per year, donates $800,000 over a 40-year career, and saves approximately 160 lives at $5,000 per life. At 20%, the household gives $40,000 per year, donates $1.6 million over a career, and saves approximately 320 lives. At 30%, the figures rise to $60,000 per year, $2.4 million lifetime, and approximately 480 lives. At 50% of pretax income, an unusual but not unprecedented level, the household gives $100,000 per year, donates $4 million over a career, and saves approximately 800 lives, matching Corrie's target exactly.

With ordinary middle-class income and a basic 10% tithe directed at the most effective interventions, the number is smaller, on the order of 160 lives, which is still extraordinary by any normal standard. To literally match her 800, our model household needs to give at roughly half of pretax income, which is the level at which someone has consciously decided to live well below their means in order to fund the saving of strangers' lives. Alternatively, a household earning $400,000 reaches 800 lives at a 25% giving rate, and a household earning $600,000 reaches it at the basic tithe.

Three further considerations make the comparison closer than the headline numbers suggest. First, income usually rises across a career, so a household whose lifetime average is $200,000 likely earns less in their twenties and more in their fifties; if they hold giving rate constant, total dollars donated rises with peak earnings. Second, donors who invest unspent income early and donate appreciated assets later can substantially increase lifetime donations, a strategy analyzed by 80,000 Hours and GiveWell. Third, cost per life may shift in either direction over time; we hold it constant at $5,000 for transparency.

Donations are only one route to impact, and arguably not the highest-expected-value one. Careers in policy, global health leadership, biosecurity, AI safety, or technical research can have expected impact orders of magnitude higher than donation-based impact, particularly when the work shapes policy or resource flows that affect millions of people. 80,000 Hours has argued that the most impactful careers may avert tens of thousands of deaths in expectation, sometimes more. The trade-off is uncertainty: a marginal donation to AMF has well-evidenced impact, while the counterfactual impact of one individual in a policy role is harder to estimate and often dominated by what would have happened otherwise. Readers who care about this question should treat the donation calculation above as a floor on what is achievable through giving alone, not a ceiling on impact across a full life.

A reader who follows this calculation back to the book's footnote should resist a temptation we feel ourselves: to imagine that writing checks "makes us Corrie ten Boom." It does not. Corrie risked imprisonment, torture, and death; her sister Betsie died at Ravensbrück; their decision was made in the dark, without applause, with the Gestapo at the door. What the comparison does buy is a sense of scale. For most of human history, the opportunity to save hundreds of lives was reserved for the unusually placed. That opportunity is now broadly distributed.

Yes. The constraint on ending extreme poverty is not how much money American Christians have. It is which causes they choose to fund, how that funding reaches the poor, and whether the political and operational systems exist to absorb large flows efficiently.

The "poverty gap" is the total dollar amount by which everyone living below the international poverty line is, in aggregate, below that line. Using the World Bank's older $2.15-per-day line, the gap was estimated at roughly $100 billion per year, the figure Paul Niehaus of GiveDirectly has cited in public talks (Niehaus on the Christians for Impact podcast, 2024; Brookings Institution, "Concept Note: Ending Extreme Poverty through Cash Transfers," 2023). In June 2025, the World Bank revised the international poverty line to $3.00 per day, raising the count of people in extreme poverty to roughly 830 million and correspondingly enlarging the gap (World Bank, "June 2025 Update to Global Poverty Lines"). Closing the gap on paper is not the same as ending poverty in practice; GiveDirectly's own working estimate for actually reaching everyone in extreme poverty for a single year is $200 to $300 billion, accounting for targeting inefficiencies, delivery costs, and a margin for error (GiveDirectly, "A Global Cash-Transfer Fund Could End Extreme Poverty," 2024).

About 60 million American adults attend church regularly, splitting the difference between self-reported and behaviorally measured attendance (Pew Research Center, Religious Landscape Study). Using $60,000 as average individual income, slightly below the U.S. mean to reflect that churchgoers skew older and include retirees on fixed incomes, the aggregate annual income base is approximately $3.6 trillion. A full 10% tithe on that base equals $360 billion per year, well over three times the headline $100 billion poverty gap and comfortably within range of GiveDirectly's $200 to $300 billion realistic operational figure. Even under aggressively conservative assumptions, 40 million churchgoers at $50,000 average income tithing 10% yields $200 billion, still twice the headline gap.

Four honest caveats prevent the headline from being preached as written. American Christians do not currently give 10%; total U.S. religious giving in 2024 was $146.5 billion, implying an average rate closer to 4% of the income base above (Giving USA 2025, Lilly Family School of Philanthropy). Tithes currently go to the local congregation, religious schools, and missionary support, not to extreme-poverty interventions, and the hypothetical assumes 100% redirection, which is not a serious proposal in any tradition we know. Closing the gap once is not ending poverty; the gap is a stock figure for a given year, and closing it in year one does not close it in year two unless growth, institutions, and public goods also change. Finally, delivery infrastructure at this scale does not yet exist; GiveDirectly currently moves roughly $300 million per year, and absorbing hundreds of billions would require new mechanisms (digital ID, mobile money, government social protection systems) to scale faster than they currently are.

The most defensible version of the claim is therefore: American Christians collectively control more than enough resources to close the world's extreme-poverty gap many times over, and we do not currently direct those resources toward that end.

This calculation compares two ways an individual concerned about farm animal welfare might respond: switching their own diet to cage-free eggs, or donating to corporate animal welfare campaigns. The comparison favors donations by a wide margin. Donating effectively for animal welfare is likely many times more more impactful than dietary change.

For instance, a one-dollar donation to corporate cage-free campaigns affects roughly nine times as many chicken-years as a full year of personal dietary switching to egg-free. At the mean historical estimate of the effectiveness of corporate campaigns, the ratio is closer to forty to one. A modest annual donation to effective animal advocacy will, in expectation, do more for chickens than a lifetime of personal cage-free purchasing.

While this BOTEC considers this tradeoff for eggs, we suspect tradeoffs for other animal products similarly favor effective donations over humane diet.

The average American eats about 281 eggs per year (USDA WASDE Report, January 2024). When an individual stops buying caged eggs and switches to cage-free, the market does not respond one-for-one; egg producers adjust gradually. Economists F. Bailey Norwood and Jayson Lusk estimate that for every chicken egg an American stops buying, approximately 0.91 fewer eggs are produced over the long run, roughly within five years (Norwood and Lusk, Compassion, by the Pound: The Economics of Farm Animal Welfare, Oxford University Press, 2011). Applied to a full year of dietary switching, this means roughly 255 of the eggs an individual would have consumed shift from caged to cage-free production in the long run. Since a laying hen produces around 280 eggs per year, one person's full-year dietary switch from caged to cage-free eggs translates to approximately one hen-year of improved welfare per year of dietary discipline (the equivalent of one hen experiencing cage-free rather than caged conditions for a year).

A donation to corporate cage-free campaigns achieves substantially more per dollar. Saulius Šimčikas at Rethink Priorities estimates that historically, corporate campaigns have affected 9 to 120 chicken-years per dollar spent, with a mean around 41 chicken-years, accounting for campaign costs, follow-through rates, and the years the welfare improvement persists (Šimčikas, "Corporate Campaigns Affect 9 to 120 Years of Chicken Life per Dollar Spent," Rethink Priorities, 2019). A more recent and more conservative estimate from The Humane League itself, using only direct and accountability costs and counting hens as binary "spared from cages" rather than multiplying by years, finds approximately 2 hens spared per dollar across 2015 to 2024, which is consistent with the Rethink Priorities range when the years-of-impact multiplier is included (Mills, Suchy, and Sweeney, "Cost-Effectiveness of THL's Corporate Cage-free Campaigns," EA Forum, 2025). Animal Charity Evaluators independently estimates around 11 hens per dollar at The Humane League specifically (Animal Charity Evaluators, The Humane League Review).

Even at the most conservative end of the published estimates, a one-dollar donation to corporate cage-free campaigns affects roughly nine times as many chicken-years as a full year of personal dietary switching. At the mean estimate, the ratio is closer to forty to one. A modest annual donation to effective animal advocacy will, in expectation, do more for chickens than a lifetime of personal cage-free purchasing.

A donor giving $25 per year to corporate campaigns moves on the order of 200 to 3,000 chicken-years, which exceeds what an individual could plausibly achieve through their own diet across a typical lifespan.

This is not an argument against personal dietary change. Several considerations weigh in the other direction. Dietary choices have effects this calculation does not capture: social modeling, downstream consumption shifts in households and friend groups, possible reduction in total animal product consumption rather than category-switching, and the formation of moral character that translates into other actions, including donations themselves. The Norwood and Lusk elasticity is a long-run market figure with substantial uncertainty, and the chicken-year metric assumes that cage-free conditions are genuinely better than caged, which most welfare scientists affirm but with real caveats about implementation (Cotra, "How Will Hen Welfare Be Impacted by the Transition to Cage-Free Housing?", Open Philanthropy, 2017). Many of the most thoughtful people in this space, including most of the staff at the organizations cited above, both donate and follow vegetarian, vegan, or reducetarian diets. The two are not substitutes; they are complements.

What this back-of-envelope establishes is a clear ranking when forced to choose between marginal options: another year of personal dietary discipline versus another modest annual donation. For most people, on most years, the donation moves more chickens. That is a useful fact for anyone trying to translate concern for farm animals into action that scales.

A "back of the envelope" calculation is meant to be rough. It uses round numbers, names its assumptions out loud, and aims for the right order of magnitude rather than precision. We expect the calculations are wrong in some of the details. We are confident they are not wrong by a relevant rough order of magnitude or in the direction of the claim.

Claude Opus 4.7 helped us draft and pressure-test the arithmetic in these calculations. The judgments, framings, and any remaining errors are ours.