Risk-Adjusted Return on Capital

Risk-adjusted return on capital (RAROC) divides a business line’s profit by the economic capital allocated to it—the cushion of loss-absorbing funds required to meet a target default probability. It answers the question: how much profit does this unit earn per dollar of risk it carries? Banks use RAROC to decide whether a loan, trading desk, or product line generates enough return to justify the capital it ties up.

The gap between accounting and economic capital

Traditional return on equity (net income divided by shareholder equity on the balance sheet) tells an incomplete story in banking. A commercial bank might hold $100 million of equity and earn $10 million, yielding a 10% ROE. But if a particular loan—say, a $10 million credit to a stressed borrower—could wipe out 50% of the bank’s equity in a stress scenario, the true capital required to absorb that loan’s risk is much larger than its nominal size on the balance sheet.

Economic capital is a forward-looking estimate of the loss buffer a bank needs to stay solvent if losses hit the projected worst case (typically defined as a 1-in-1,000-year event or a target probability of default). A retail mortgage portfolio and a leveraged buyout portfolio, each producing identical accounting profits, might require vastly different economic capital because of their risk profiles.

How RAROC guides capital allocation

A bank’s capital is finite. Management must decide whether to deploy it in mortgages (lower risk, modest margins), investment-grade corporate bonds (moderate risk, moderate margins), or leveraged buyouts (high risk, high expected returns). RAROC creates a common scorecard.

Suppose the bank’s cost of equity is 10%. A mortgage desk earning 2% return on $50 million of economic capital yields RAROC of 2%—well below the hurdle rate, so that desk destroys value. A leveraged buyout desk earning 8% on $10 million of economic capital yields RAROC of 80%—well above the hurdle rate, so it should expand. All other things equal, the bank should shift capital from mortgages to leveraged deals, until marginal RAROC across units aligns with the cost of capital.

In practice, banks rarely reach perfect equilibrium because of regulatory capital requirements, customer relationships, and competitive dynamics. But the RAROC framework at least forces transparent trade-offs.

Calculating economic capital: the mechanics

Economic capital is typically computed using value at risk or expected shortfall models. A bank specifies a confidence level (e.g., 99.9%) and a time horizon (e.g., one year), then asks: what loss level would we exceed with only 0.1% probability?

For a $10 million loan with a 5% probability of default, 60% loss-given-default, and assuming default events are mostly independent, the expected loss is $10M × 5% × 60% = $300,000. But there is volatility around that expectation; in a bad year, losses could spike to $2 million. The economic capital is roughly the difference between the worst-case loss (at the chosen percentile) and the expected loss, sometimes plus a buffer. The exact formula varies by institution and regulatory regime.

Credit derivatives and securitization markets have made economic capital more explicit: the capital tranche of a mortgage-backed security is the first to absorb losses, and its market price reveals what capital costs in that context.

Adjusting profit for risk: the numerator

The numerator of RAROC is typically “risk-adjusted” profit, not raw accounting profit. For a loan, this might be:

Loan interest income — expected loss — cost of funds — operating expenses = risk-adjusted profit

The expected loss is subtracted even if no default has occurred yet, because it is an implicit cost of bearing credit risk. A loan generating $100,000 in interest but carrying a $30,000 expected loss has risk-adjusted profit of only $70,000—much lower than accounting income.

For a trading desk, risk adjustment might account for stressed loss estimates or value at risk limits breached during extreme moves.

RAROC vs. Dodd-Frank stress testing

US banks subject to Dodd-Frank Act supervision now run annual stress tests that forecast capital consumption under severe recession scenarios. These tests implicitly produce RAROC-like calculations at the portfolio level: the regulator asks whether the bank can absorb a hypothetical wave of defaults and market losses while maintaining minimum capital adequacy ratios.

The post-2008 regulatory framework has made RAROC more constraining. A business line that generates 12% RAROC might still be shut down if the stress test shows it would deplete capital too quickly in a crisis. Conversely, a low-RAROC but stable business (like deposit-taking) is often tolerated because it anchors the balance sheet.

The assumption trap

RAROC is as good as the economic capital model underlying it. If the model assumes defaults are independent but they are actually correlated (as they are in recessions), the calculated capital is too low, and RAROC is overstated. A portfolio of loans to real estate developers looked profitable with high RAROC in 2006; the 2008 crisis revealed that the true economic capital was much larger.

Many banks now use Monte Carlo simulation and scenario analysis to stress-test RAROC assumptions. A business line must show not only high RAROC under the base-case economic capital estimate but also acceptable RAROC under plausible alternative capital models.

See also