EASTERBROOK, Circuit Judge.

Farm Credit Services holds a first mortgage on the Koopmans' farm, securing a loan whose balance hovers near $110,000. The land is worth more, perhaps as much as $215,000, so, although junior liens on the farm exceed $500,000, Farm Credit Services is oversecured and entitled to interest in the Koopmans' bankruptcy. 11 U.S.C. § 506(b); United Savings Ass'n of Texas v. Timbers of Inwood Forest Associates, Ltd., 484 U.S. 365, 372 (1988). "At what rate?" is the question we must answer. The question arises at all only because the bankruptcy court has blocked Farm Credit Services from foreclosing and has approved a Chapter 12 reorganization over its objection. In such a cramdown the secured creditor is entitled to the "indubitable equivalence" (In re Murel Holding Corp., 75 F.2d 941, 942 (2d Cir. 1935) (L. Hand, J.)) of its property interest, 11 U.S.C. § 1225(a)(5)(B)(ii), which means a stream of payments including interest that adds up to the present value of its claim. Cf. In re Fortney, 36 F.3d 701 (7th Cir. 1994). At what rate of interest will Farm Credit Services be as well off in the reorganization as if it had been allowed to foreclose on and sell the farm? To put the question in this way is to supply the answer: the creditor must get the market rate of interest, at the time of the hypothetical foreclosure, for loans of equivalent duration and risk. The bankruptcy judge approximated this by starting with the prime rate of interest, which it found prevalent for new 20-year well-secured agricultural loans at the time, and adding 1.5 percent because it deemed this extension of credit more risky than the norm in light of the Koopmans' sorry repayment record. This produced a floating rate, 10.5 percent (9 percent + 1.5 percent) at the time the bankruptcy judge approved the plan. The district court affirmed, 196 B.R. 425 (N.D. Ind. 1996), surveying the cases and concluding that the bankruptcy judge' approach — often called the "coerced loan" method — represents the dominant though not exclusive view among the courts of appeals. E.g., General Motors Acceptance Corp. v. Jones, 999 F.2d 63 (3d Cir. 1993); United Carolina Bank v. Hall, 993 F.2d 1126 (4th Cir. 1993). See also John K. Pearson, Ending the Judicial Snipe Hunt: The Search for the Cramdown Interest Rate, 4 Am. Bankr. Inst. L. Rev. 35 (1996) (discussing possible approaches to the subject). Without implying that "prime-plus" is the only way to approximate the market rate of interest — for participants in the market may use other methods — we now hold that the creditor is entitled to the rate of interest it could have obtained had it foreclosed and reinvested the proceeds in loans of equivalent duration and risk. Nothing else gives the creditor the indubitable equivalent of its non-bankruptcy entitlement.

Consider a few of the potential alternatives. One, which some bankruptcy courts have used, is the rate of interest provided by the original loan — in this case, 8.75 percent for credit extended in 1976. This is a market rate of interest, but it does not exhaust the creditor's legal entitlements under state law. If a debtor does not pay, a creditor is entitled to declare a default, accelerate repayment of the principal, and cash out by foreclosure. In other words, the original contract provided for an 8.75 percent rate with an option for the lender to mark the rate to market if the debtor defaults. That is what happened. Note, too, that the original contract required repayment no later than March 2012, while the Koopmans' plan of reorganization extends the loan through December 2014. A debtor who wants to change the duration of a loan must refinance, which occurs at the current market rate. This may be lower than the one provided by contract. Just as the debtor cannot insist on the lower of the contract or current market rates, neither may the creditor obtain the higher of contract or current market. The market rate must be used consistently. Accord, General Motors Acceptance Corp., 999 F.3d at 71 n. 11.

What the standing Chapter 12 Trustee, representing the interests of unsecured creditors, favors is the rate the federal government pays to borrow money (the "T-Bill rate" for short). At times the Trustee appears to concede that this rate may be adjusted for risk; if so, that comes to the same thing as prime-plus, although the adjustment may be larger. The prime rate of interest is the benchmark rate for the banks' most credit-worthy customers, but even blue-chip debtors are more likely to default than is the United States government, so the prime rate exceeds the T-Bill rate. Because the prime rate includes some compensation for the risk of non-repayment, the add-on for less creditworthy customers is less than it would be if the court started with the risk-free T-Bill rate. See generally In re Oil Spill by the Amoco Cadiz, 954 F.2d 1279, 1331-33 (7th Cir. 1992); William F. Sharpe, Gordon J. Alexander Jeffrey V. Bailey, Investments 232-49 (5th ed. 1995). Because adjustments would make the final interest rate the same whether the bankruptcy court starts with the prime rate or the T-Bill rate, the choice of nomenclature is irrelevant — although it is best to stick with the market's approach to estimating the risk premium. On this record, the market's approach is primeplus.

Understanding this, the Trustee makes a bolder claim: that the right rate of interest is the T-Bill rate with no adjustments, or 6.9 percent at the time the bankruptcy judge confirmed the Koopmans' plan of reorganization. After all, the Trustee observes, Farm Credit Services is well secured, which as he sees it means that Farm Credit Services will be paid in full. When there is no risk of nonpayment, why should lenders be compensated for riskbearing? If the Trustee is right about risk, then banks should be eager to make agricultural loans a few basis points over the T-Bill rate. Yet they are not willing to do so, because even an apparent cushion of security does not reduce risk to zero. The value of land may fall by the time payment is due; the appraisal showing the cushion may be in error; debtors may refuse to pay (and declare bankruptcy) for strategic reasons even if they are able to pay; other creditors may try to jump the queue and divert the debtor's assets; the costs of collection from recalcitrant debtors, and even from willing ones, greatly exceed the costs of clipping coupons from T-Bills. Market rates of interest measure the real risks of nonpayment and the costs of collection (including the costs of foreclosure and bankruptcy proceedings); it is to these market rates, rather than lawyers' speculations about business operations, that judges must turn. The bankruptcy judge found a real market rate in Indiana for well-secured agricultural loans. Poorly-secured loans have their own market (there is, for example, a flourishing market in second and third mortgages, and even unsecured loans to farmers); when the borrower is a business rather than a family farm, the court would look to the interest rate on junk bonds. The Koopmans' argument that the "coerced loan" approach must be abandoned because there is no "market rate" for loans of the kind is wrong.

According to the Trustee, the difference between the T-Bill rate and the market rate for agricultural loans is not attributable to higher risks and collection costs. It is instead a measure of lenders' "profit." As the Trustee sees things, Timbers of Inwood holds that even a fully secured lender is not entitled to "profit" in bankruptcy, with the consequence that the T-Bill rate must be used. Timbers of Inwood held no such thing. It held that an undersecured lender is not entitled to interest, not because there is a general principle that suppliers to debtors in bankruptcy can't make profits — for if that were so, suppliers of seeds and lessors of tractors could not charge the market price for their goods — but because sec. 506(b) itself appears to distinguish fully secured from undersecured creditors. Having used that statutory difference as the basis of its decision in Timbers of Inwood, the Court could not logically conclude that even fully-secured creditors must accept in a cramdown less than the market value of their claims. A supplier of capital, no less than a supplier of seeds or combines, is entitled to the market price. The Koopmans' plan makes Farm Credit Services an involuntary supplier of capital through the year 2014, overriding the terms of a contract that allowed it to cash out; for this service the Koopmans must pay the market rate, just as they would for heifers or fertilizer secured on longterm contracts stretching into the next century.

The Trustee's understanding suffers from a further problem: in competition, a financial intermediary does not make a "profit." True, there may be accounting profits, but there are no economic profits in vigorous competition, one of Adam Smith's principal points in The Wealth of Nations (1776). See generally George J. Stigler, The Theory of Price 178-92 (4th ed. 1987). Normal returns to entrepreneurial and managerial skill may keep the wolf from the door, but they are not economic "profit". What appears on the books as accounting profit is just the opportunity cost of keeping the firm's assets in this business rather than the next-best alternative. Financial intermediation is today highly competitive. Money is fungible and flows across the globe; local lenders package and sell their loans on national and international markets in order to attract fresh capital; none of the resources used in the agricultural loan business is so scarce that economic rents would accrue. To say that the lender is limited to its "cost of capital" — another of the Trustee's sallies — is therefore to say that the lender is entitled to the market rate of interest, for that is what its cost of capital is: the price it must pay to its own lenders, plus the costs of making and administering loans, plus reserves for bad debts (that is, the anticipated rate of non-repayment). Far better to read this cost of capital from newspaper tables based on actual transactions in competitive markets than to reconstruct it in a factual vacuum as an intellectual exercise. Tables may err, but reconstructions are bound to do worse.

AFFIRMED.