Most economists describe three determinants of the interest rates on a borrower's debt: a real interest rate, which is a charge for using capital; an adjustment for expected inflation to insure that debt is repaid in dollars having the same purchasing power; and an adjustment for the borrower's credit risk, which is intended to compensate the lender for the possibility that the borrower will default. Certain companies book gains and losses on their own debt on the income statement due to a revaluation of debt to fair value. For example, if expectations for inflation were to rise, the appropriate interest rate to charge borrowers would rise above the contractual historical interest rate initially used to value and record the debt. If the debt is adjusted to fair value (i.e., rediscounted at the new higher interest rate), the debt book value falls, and because a liability has decreased, a gain is recorded on the income statement. The counterparty to the debt (e.g., the lender) records a loss. These adjustments can be extremely large for banks. For example, Citi's 2011 third quarter net income was $3.7 billion, which included a $1.9 billion gain from revaluing its own debt (before tax) in revenues and income. In Citi's case, the gain was driven from a widening in its credit-default swap spreads, which is an indication of its higher probability of default on its obligations and derivative contracts. How should an analyst view gains on the revaluation of a company's own debt due to changes in its credit risk? How would a debtor realize such gains (i.e., how does such a gain affect cash flows)? Is the gain a persistent component of earnings?