economic capital
/ ee-kuh-NOM-ik /
Suppose you ran the company yourself and asked, with no regulator in the room, the most basic survival question: 'how much spare money — over and above what I owe — do I need to keep on hand so that I would still be standing even after a genuinely bad year?' The honest answer to that question, computed from your own view of the risks, is economic capital. It is the cushion the business decides it truly needs to absorb shocks, not the cushion any rulebook happens to demand.
Concretely, economic capital is the amount of capital required to remain solvent over a chosen horizon (typically one year) at a chosen confidence level (often 99.5%, sometimes 99.95%), computed using the firm's own internal model of all its risks combined. The recipe: simulate the change in net worth (assets minus liabilities) across thousands of scenarios spanning market, credit, insurance and operational risk; find the loss at the chosen quantile or tail average; that loss, which you must be able to absorb without going under, is the economic capital. Because risks rarely all turn bad together, the combined figure is smaller than the sum of the standalone risk amounts — that gap is the diversification benefit, and economic capital is where it is quantified and credited.
Economic capital matters because it is the firm's own truth, the benchmark against which regulatory capital (RBC, the SCR) is judged adequate or not, and the engine of risk-based decisions: pricing that charges each product for the capital it consumes, deciding which lines earn their keep, and allocating capital across the business. The caveat is that economic capital is a model output, only as trustworthy as its assumptions — especially the correlations and the tail. It is genuinely useful as a disciplined, consistent lens, but it is an estimate dressed as a number, not a fact.
An insurer's standalone risks would need 60 (market), 50 (insurance) and 20 (operational) — summing to 130. But because a market crash and a claims spike are not the same event, its internal model finds the combined 99.5% one-year loss is only 100. Its economic capital is 100; the 30 saved is the diversification benefit.
Economic capital = the firm's own honest 'bad-year cushion,' net of diversification.
Economic capital is a model estimate, not a measured fact; its weakest links are the assumed correlations and tail behaviour, which are hardest to validate.