Capacity, Not Uptake

Evidence essay · Experimental badge

FAND Provisional Subnational Exhibit — Not for Policy Citation. County figures are ordinal/relational (rank correlations), carry large margins of error, and are shown to demonstrate a structural finding, not to rank places.

The largest line on a nation’s balance sheet is the one no conventional account books: the trust infrastructure. So we took that figure down to individual US counties and asked — does it line up with the things researchers already use to measure “social capital”?

The answer split cleanly in two, and the split is the finding. Measures of the opportunity to participate — how dense the nonprofit and associational fabric is — line up positively with the trust figure. Measures of realized participation — how much people actually turned out, clustered, joined — line up negatively.

At first that looks like a contradiction. It isn’t. The trust-infrastructure stock is a measure of capacity — the freedom and opportunity to act — not of uptake. A right is a liberty: the room to do a thing, whether or not it gets done. The measure behaves the way the concept says it should — which is the strongest thing an early-stage indicator can do.

The two families (Spearman ρ vs the FAND trust residual, N≈3,043 counties)

Supply-side — ALIGNED (positive): latent civic capacity / opportunity.

IndicatorρWhat it measures
Nonprofit density (NCCS 2014)+0.38organized civic infrastructure
MapAgora “take action”+0.20availability of channels to act
MapAgora “volunteer”+0.10volunteering opportunity
MapAgora events / membership+0.04–0.08associational opportunity

Behavioral / density — PARADOX (negative): realized participation.

IndicatorρWhat it measures
Social clustering (Chetty Atlas)−0.30how tightly people actually cluster
JEC Social Capital Index−0.23composite realized participation
Civic organizations (Chetty)−0.21realized organizational membership
Association density / voting (Goetz)neg.realized turnout & joining

This is the de-residualization path: findings like this are how the trust figure stops being a residual — by recognizing and directly scoring its components against the balance-sheet slot they occupy. Recognition precedes valuation.

Sources: Kim, de Vries & Han (2025) MapAgora; Rupasingha/Goetz/Freshwater NERCRD; Chetty et al. Social Capital Atlas (2022); JEC Social Capital Project (2018). FAND S323/S324. Experimental badge: subnational build deliberately gated.

← back to The Data