We should measure welfare's success by how many people leave welfare, not by how many are added.

The danger of counting departures

There's something intuitive about this metric that goes beyond politics. If you're running a fire department, you measure success by fewer fires, not more firefighters. If you're a hospital, you track patients getting better. So why shouldn't a safety net program measure itself by people becoming independent?

But here's where it gets tricky in real life: the metric can hide what's actually happening. Someone might "leave welfare" because they found a job paying minimum wage with no benefits—technically a success on the spreadsheet, but practically, they're just trading one precarious situation for another. Or they leave because they got discouraged navigating bureaucracy, not because they stabilized. The number going down looks clean until you ask what it really means.

The harder question this raises is about intention. Are we designing these programs to help people climb out, or primarily to reduce the rolls? Those sound similar but create different systems. One focuses on job training, childcare support, and gradual phase-outs. The other focuses on barriers to entry. Both can show declining numbers—one because people thrived, the other because fewer people qualified. The metric itself doesn't tell you which world you're building.