Originally published September 8, 2011 on the New Economic Perspectives blog.
This week we addressed denomination of “money things” in the state money of account—for example, the Dollar in the US. We began a discussion of “leveraging”—making one’s IOUs convertible (on demand or on some contingency) into another’s IOUs. Next week we will turn to the notion of a “debt pyramid” with the state’s own IOUs at the apex. For now, on to the questions and comments. As usual I will group them into themes; some commentators actually answered several of the questions (Thanks!) but I’ll briefly repeat some of what they said.
Q1: (Jeff) Can’t the Fed just control inflation by raising required reserve ratios? At the limit, to 100% reserves? And would that affect interest rates?
A: As discussed in a bit more detail below, and in this blog later, required reserve ratios do not control bank lending. To hit its interest rate target, the central bank must accommodate the demand for reserves—whether the ratio is 1% (about where it is now) or 10% (the ratio usually used in textbooks to simplify math). (Note: the required reserve ratio in Canada is a big zip, zero! That is actually the most advanced way to run the system. Hats off to our neighbors to the north.) Since it would not control lending there is little reason to believe raising ratios would affect inflation. Also note that raising the ratio does not affect the overnight rate (fed funds rate in the US)—since that is the policy variable.
Higher ratios do act like a tax on banks—they must hold a very low earning asset. If the ratio is 1% they hold 1% of their assets (more or less—close enough for this analysis) in an asset that earns a very low interest rate (the support rate paid by the central bank on reserves). They need to cover their costs and make profits by earning more than that on the rest of their assets (99%). Raising the ratio to 10% means they only have 90% of their assets potentially earning higher returns. And so on. Will that affect lending rates earned (what they charge borrowers) and deposit rates paid (what they pay depositors)? Well banks live on the spread between those two—that is how they cover costs and make profits. So, yes, raising ratios might cause them to raise loan rates and lower deposit rates—not a good thing for borrowers or depositors.
Finally, what about 100% reserves? There is a good book by Ronnie Phillips (Google it) on the Fisher-Simons-Friedman proposal to do just that. However, this is usually presented as a way to make banks “safe”—they’d hold only reserves or treasuries against their demand deposits, on the idea that with safe assets, the deposits are always safe (so you do not need deposit insurance, FDIC). Sounds OK so far as it goes. Someone else has to do the lending since the banks are not allowed to do it. A big topic.
Q2: (Jeff) How can China operate domestic policy with a fixed exchange rate?
A: Trillions of dollars of foreign exchange reserves! No George Soros is going to bet against China’s ability to peg its exchange rate. So, yes, there are exceptions to the rule that pegged exchange rates reduce policy space.
Q3: (Neil) What about IMF conversion clause? What makes banks special? (others also asked that)
A: Come on, you are sounding like Ramanan. I answered that already—yes you can tie your shoes together and try to run a marathon. First, this discussion was general. Second, as I showed, in practice the clause has no impact. What makes banks special? We’ll save that for coming blogs. But two characteristics that are very important are: access to central bank, and access to deposit insurance.
Q4: (Godefroy) What does the central bank lend against? How does a bank get cash?
A: Central bank lends against qualifying assets. It’s the boss and can decide. Yes, it lends against treasuries (IOUs of the Treasury); it can lend against “real bills” (short term commercial loans made by banks to good customers); it can lend against toxic waste MBSs (maybe a bad idea?). It can use collateral requirements as a way to supervise/regulate banks: encourage them to make only safe loans by narrowing what it accepts as collateral.
When you go to the ATM to withdraw cash, your bank has a bit on hand—that counts as part of its reserve base. If everyone goes tomorrow, obviously the bank runs out quickly. It orders more from the Fed—shipped in armored trucks—and the Fed debits the bank’s reserves, and when that is insufficient it lends the cash (a loan of reserves) against collateral. The Fed holds the bank’s IOU as an asset; it is of course a liability of the bank.
Q5: Do money center banks influence the FOMC?
A: Is Goldman Sachs a bloodsucking vampire squid that bought and paid for Timmy Geithner’s NY Fed as well as Treasury?
Q6: (Glenn; Jeff) Why does government need to borrow its own IOUs and pay interest? And why pay interest on “fiat money”?
A: Good question! Government cannot borrow its own IOUs. Neither can you! If you give an IOU to your neighbor for a cup of borrowed sugar, you do not go back and ask if you can borrow it. It is a senseless operation.
Instead, government offers Treasuries as a higher interest paying IOU, exchanged for reserves. When you go down to your bank, you can exchange your demand deposit for a saving deposit on which you earn higher interest. That is really all that a government bond sale is—a substitution of a demand deposit at the central bank for a time deposit.
Note that cash (“fiat money”) does not pay interest. A Chicago Mafioso loan shark might lend you cash at 140% interest. Why? You are desperate. He gets compensated for the risk that you will run with the money. Of course, there is a substantial penalty for nonpayment. But why would the Treasury pay interest on bonds, and why would the Fed pay interest on reserves? There is no necessity of doing that. We’d accept cash and banks would accept reserves without interest—there is no default risk (on sovereign government IOUs on a floating exchange rate), and we need them to pay taxes. But it is nice to get interest, isn’t it? Think of it as a government transfer payment, a form of charity. It might be a bad idea—a topic for later.
Q7: Does a lack of sufficient reserves constrain loans?
A: No. Don’t take my word for it. Here’s a comment from the Fed’s Senior Vice President, Federal Reserve Bank of New York, Alan Holmes. Holmes explained why the then faddish Monetarist policy of controlling inflation by controlling the growth of Base Money had failed, saying that it suffered from “a naive assumption” that the banking system only expands loans after the [Federal Reserve] System (or market factors) have put reserves in the banking system. In the real world, banks extend credit, creating deposits in the process, and look for the reserves later. The question then becomes one of whether and how the Federal Reserve will accommodate the demand for reserves. In the very short run, the Federal Reserve has little or no choice about accommodating that demand…
Q8: (unknown) How do banks work? What happens if a borrower goes bankrupt?
A: We’ll look in more detail at how banks “work”. They’ve got assets on one side of their balance sheet and liabilities plus capital on the other. When the assets go bad, the capital is reduced (shareholders lose); once the capital is wiped out, the losses come out of the other liabilities, so creditors lose. Since the FDIC insures depositors, if losses are big enough to hit deposits, Uncle Sam covers those.
Q9: (HadNuff) Don’t you pay taxes with demand deposits? Banks can be illiquid but not insolvent?
A: You write a check to the IRS but your bank pays the taxes for you using reserves, since the IRS sends the check on to the Fed, which debits the bank’s reserves (and increases the Treasury’s deposit). The central bank lends reserves to solve liquidity problems, lending against collateral. Banks do become insolvent, as discussed above. They then must be “resolved”—there are a variety of methods but it comes down to selling the assets, covering insured depositors first, and then other creditors and the shareholders take the loss.
