β οΈ LGD Calculator: Loss Given Default
Expected credit losses from default probability, exposure, and recovery.
Loss given default
Estimates for planning only, not financial advice.
Loss given default answers the lender's second question. The first is whether the borrower will default; the second is how much disappears if they do. Divide the unrecovered amount by the total exposure: a $1,000,000 loan that recovers $400,000 has an LGD of 60 percent. Add a default probability and you get expected loss, the number that loan pricing and bank capital are built on. This calculator runs all three from your inputs.
The concept separates two risks people constantly conflate. A shaky borrower with great collateral can be a safer loan than a solid borrower with none. PD measures the borrower; LGD measures the structure around the loan. Price one without the other and you are guessing.
How to Use This Calculator
Enter the exposure at default, the amount you expect to recover through collateral, guarantees, and collections, and the annual probability of default. The calculator returns LGD, the loss amount, and expected loss.
Enter the exposure at default: the outstanding balance plus any committed but undrawn amounts the borrower could draw before defaulting.
Enter expected recovery in dollars: collateral sale proceeds net of costs, guarantor payments, and bankruptcy distributions.
Enter the annual probability of default to see expected loss. For the borrower's side of default risk, our debt to income ratio calculator shows how lenders judge repayment capacity.
The Formula
The expected loss formula is the workhorse of credit. It converts three uncertain inputs into one dollar figure a lender can price against. Multiply through: 2 percent default probability times 60 percent LGD times $1,000,000 exposure equals $12,000 of expected annual loss, which the loan's interest margin must cover before the lender earns anything.
Typical LGD by Loan Type
| Loan type | Typical LGD | Why |
|---|---|---|
| Residential mortgage | 10 to 30% | House collateral, first lien |
| Auto loan | 30 to 50% | Depreciating collateral |
| Senior secured corporate | 30 to 40% | Asset backing, priority claim |
| Credit card / unsecured | 70 to 90% | No collateral |
The pattern is obvious once stated: recovery follows collateral. This is the entire economic logic of secured lending, and it is why mortgage rates sit far below credit card rates for the same borrower. To see how lenders protect themselves upfront, our down payment calculator shows the equity cushion that keeps mortgage LGD low.
Worked Example
With the defaults: $1,000,000 exposure, $400,000 recovery. LGD is (1,000,000 - 400,000) / 1,000,000 x 100 = 60.0 percent, and the recovery rate is 40.0 percent. The loss if default occurs is $600,000. At a 2 percent annual default probability, expected loss is 0.02 x 0.60 x 1,000,000 = $12,000 per year, or 1.20 percent of exposure. If the collateral outlook worsened and recovery fell to $250,000, LGD would rise to 75 percent and expected loss to $15,000.
Where LGD Gets Misused
The most common error is using boom-time recoveries to estimate LGD, then discovering in a recession that the collateral everyone counted on sells for half. Basel rules require downturn LGD for exactly this reason. The second error is ignoring workout costs: three years of legal fees and the time value of money turn a 60 percent accounting LGD into a much worse economic one. The third is assuming seniority protects you when the whole capital structure is impaired.
My rule: stress the recovery, not just the default probability. Anyone can model PD from credit scores. The investors who survive credit cycles are the ones who asked, in writing, what the collateral is actually worth in a down market.
How Banks Actually Estimate LGD
Real LGD estimation is statistical, not arithmetic. Banks build models on years of defaulted-loan history, regressing realized losses against collateral type, loan-to-value at origination, industry, and time in default. Mortgages typically show LGD around 10 to 25 percent because houses retain value; unsecured credit cards run 80 to 95 percent because there is nothing to repossess. Commercial real estate sits in between and swings wildly with the property cycle.
The subtle driver is downturn LGD: regulators require banks to use loss rates observed during recessions, not averages, because collateral values and recovery rates collapse exactly when defaults spike. This correlation, called wrong-way risk at its extreme, is why LGD models failed less badly than default models in 2008 but still underestimated commercial losses. For the probability side of expected loss, pair this with default-rate data for the same asset class.
LGD and Loan Pricing
Lenders price expected loss into every loan's interest rate, which is why unsecured credit costs more than mortgages. When you see a rate quote, part of it is the bank's LGD estimate for borrowers like you. Improving your collateral position, a bigger down payment, lower LTV, directly cuts the rate because it cuts their LGD.
Estimates for planning only, not financial advice.Frequently Asked Questions
What is loss given default (LGD)?
LGD is the share of a loan or exposure a lender expects to lose if the borrower defaults, expressed as a percentage of the exposure at default. If a $1,000,000 loan recovers $400,000 through collateral sale and collections, the LGD is 60 percent. It is one of the three core inputs in credit risk modeling, alongside probability of default and exposure at default.
How do you calculate LGD?
Subtract the expected recovery from the exposure at default, then divide by the exposure: LGD = (EAD - Recovery) / EAD. A $1,000,000 exposure with $400,000 recovered gives ($1,000,000 - $400,000) / $1,000,000 = 60 percent. The recovery rate is simply the complement: 40 percent here.
What is expected loss in credit risk?
Expected loss equals probability of default times loss given default times exposure at default: EL = PD x LGD x EAD. With a 2 percent default probability, 60 percent LGD, and $1,000,000 exposure, expected loss is 0.02 x 0.60 x $1,000,000 = $12,000 per year. Lenders price this into interest rates and loan loss reserves.
What is a typical LGD for different loan types?
Secured mortgages often see LGD of 10 to 30 percent because the house backs the loan. Auto loans run 30 to 50 percent. Unsecured credit cards and personal loans run 70 to 90 percent since little is recovered. Senior secured corporate bonds historically recover around 60 to 70 cents on the dollar, implying LGD near 30 to 40 percent.
What affects the recovery rate?
Collateral quality and seniority matter most: first-lien secured lenders recover far more than subordinated ones. The economic cycle matters too, since collateral values fall in downturns exactly when defaults rise. Legal jurisdiction, bankruptcy regime, and how quickly the lender can seize and sell collateral also move the number.
What is the difference between LGD and PD?
Probability of default (PD) is the chance the borrower defaults at all, while LGD is how much is lost given that default happened. A loan can have high PD but low LGD, like a well-collateralized loan to a shaky borrower, or low PD but high LGD, like unsecured lending to a blue-chip company. Pricing needs both.
How do banks estimate LGD under Basel rules?
Large banks using the internal ratings-based approach estimate LGD from their own historical loss data, segmented by collateral type, seniority, and industry. Regulators impose floors, such as 10 percent for residential mortgages, and require downturn LGD estimates that reflect recession recoveries rather than boom-time ones.
What is downturn LGD?
Downturn LGD is the loss rate estimated under stressed economic conditions rather than average ones. Since defaults cluster in recessions when collateral values are depressed, downturn LGD exceeds long-run average LGD. Basel capital rules require banks to use the downturn figure so capital covers bad-times losses, not average-times ones.
Can LGD exceed 100 percent?
Effectively yes, once workout costs are included. Legal fees, years of collections effort, and the time value of money during a long bankruptcy can push economic losses above the original exposure. Regulatory LGD is typically capped for capital purposes, but economic LGD in a messy workout can surpass 100 percent.
How does LGD affect loan pricing?
Lenders add expected loss to the required return: a loan with 2 percent expected loss needs roughly 2 percentage points more yield than a risk-free loan to break even, before profit margin and costs. This is why unsecured borrowers pay far higher rates than mortgage borrowers with similar credit profiles.