WHO OWNS COMPASSION? — 4-PART SERIES

The problem: Government as the sole provider of safety net functions is neither efficient nor compassionate — and it crowds out the family, fraternal, and religious institutions that once made it work.

The opportunity: We can make progress for people by progressing and improving our safety net institutions with more pluralism and competition.

← Part 2  |  Part 3 of 4  |  Part 4 →

Americans gave an estimated $617 billion to charity in 2025. That is 19 percent of what the federal government spent on just Social Security, Medicare, and Medicaid. Both numbers matter. The question is not which one wins. It is how pluralism puts them both to work — each doing what it does best, neither claiming the whole field.

The $617 Billion Reality: What Private Charity Can — and Cannot — Replace

This is the article where the pluralism argument has to face its hardest test. If you believe government should not be the sole provider of safety net functions — and this series argues it should not — then you have to reckon honestly with the numbers. Six hundred seventeen billion dollars in private giving sounds large. Against the scale of federal social spending, it is a fraction. That is not an argument for government monopoly. It is an argument for designing pluralism with eyes open.

THE PLURALISM PRINCIPLE

No single institution — government, church, charity, or market — is capable of providing compassion at scale without becoming something worse: a bureaucracy, a monopoly, or a moralizing gatekeeper. Pluralism is not a compromise between these failures. It is the architecture that prevents any one of them from taking over.

Progressives point to gaps in private giving and conclude that government must finance, design, and deliver nearly everything. Libertarians point to bureaucratic failure and conclude that private charity could take over if government would simply get out of the way. The numbers reject both conclusions. What they support is a third architecture: government as guarantor of last resort, everything else as a competitive market for service delivery, recipient sovereignty, and civic voice.

That $617.2 billion is not $617 billion for food banks, homeless shelters, medical rescue, and emergency cash. Religion received $151.6 billion. Education received $92.0 billion. Foundations received $79.1 billion. Arts and culture received $27.3 billion. Human-services organizations — the category most closely resembling a traditional safety net — received $99.5 billion, or 16 percent of all giving.

Those are not frivolous choices. A church, orchestra, university, cancer laboratory, and animal shelter all create value. Donor freedom means people may support purposes other than the one a central planner would rank first. That is not a bug in pluralism. It is the feature: pluralism distributes moral authority across millions of individual choices, rather than concentrating it in a single legislative act.

The Replacement Arithmetic

In fiscal 2025, the federal government spent $1.569 trillion on Social Security, $992 billion net on Medicare, and $668 billion on Medicaid — a combined $3.229 trillion. All American charitable giving equaled 19 percent of that sum. Human-services giving equaled 3 percent.

This is not an apples-to-apples program comparison. Social Security and Medicare are contributory social insurance. Medicaid includes medical care that costs far more than a food parcel or night in a shelter. That is precisely why the comparison is useful. It exposes the scale of the promise hidden inside the phrase “let charity handle it.” No serious institutional design should rest on that hope alone.

But the right conclusion is not “therefore government must do it all.” The right conclusion is: pluralism must be designed, not assumed. A public floor for catastrophic, uninsurable, nationally correlated risk. Competitive delivery for everything that can be delivered competitively. Cash and choice for recipients wherever they know their needs better than a program administrator does. Civic voice and donor pluralism for everything in between.

When Government Crowds Out Compassion

When government expands as the sole provider, others step away. Church charitable spending fell an estimated 30 percent between 1933 and 1939 as New Deal relief expanded. Fraternal societies, which had insured roughly half of adult American men, declined steadily after Social Security arrived. The pattern is not coincidence. It is how institutions work.

The crowding-out problem is not just historical. It operates today. When a government program provides a service — food, housing assistance, job training — through a mandatory administrative channel, donors redirect their giving to other causes. Faith communities that once organized their identity around feeding the hungry pivot to other missions. Families that once provided direct support for elderly relatives find the government system has made that support redundant — and gradually lose the habit and the skill.

This is not a criticism of any individual donor, congregation, or family. It is a description of how institutional ecology works. When a government occupies a field exclusively, it does not just fill that field. It empties the other institutions that were building there.

What Competition Actually Disciplines

Private charity has advantages government agencies struggle to reproduce. A donor can exit. A local organization can specialize. A new entrant can test an idea without securing a congressional appropriation. Faith communities and mutual-aid groups can offer relationship, moral support, and local knowledge along with money. Competing organizations must continually persuade someone that their work deserves another dollar.

These are real disciplines. But in an ordinary market, the buyer pays, chooses, and consumes. In charity, the donor usually pays, an organization chooses the intervention, and someone else receives it. Competition for donations disciplines fundraising appeal. It does not automatically discipline outcomes — unless accountability is deliberately built in.

Need Does Not Create Demand — Geography Exposes the Gap

A market in charity is also unlike a market in shoes. The person with the greatest need may have the least capacity to attract a donor. From 2005 through 2010, rural America had 19 percent of the population but received only 6 to 7 percent of the value of domestic foundation grants. Rural grant funding per person was less than half the metropolitan level, and 18 percent of nonmetropolitan counties had no recipient organization at all.

Tax policy magnifies the imbalance. Taxpayers in the top third of the income distribution received an estimated 98 percent of the benefits of the charitable deduction in 2024. The gifts remain voluntary, but the subsidy is public and donor influence is wealth-weighted. This is the design failure pluralism must correct — not by eliminating charitable choice, but by equalizing the public encouragement that underlies it.

What Technology Changes — and What It Does Not

New platforms are beginning to solve the measurement problem that has long plagued charitable giving. GiveDirectly delivers cash directly to recipients with real-time tracking. Innovations for Poverty Action runs randomized controlled trials on charitable interventions with pharmaceutical-grade rigor. Giving What We Can helps donors compare cost-per-life-saved across charities. These represent a genuine advance: for the first time, donors can demand evidence, not just stories.

Technology will not solve the scale problem or the geographic coverage problem. But it is beginning to solve the accountability problem — making it possible to have something approaching a real market with real price signals for human outcomes. That is the missing piece pluralism has always needed: not just many providers, but comparable, transparent results that let citizens, donors, and policymakers know which institutions are actually working.

The future is not a total government safety net or a wishful return to the parish poor box. It is a public floor, many providers, recipient choice, donor pluralism, and evidence visible to everyone. That is what pluralism looks like when it is designed rather than assumed.

← Part 2: The Year Relief Became a Right  |  Part 4: A Market for Mercy →