In previous installments of this primer I’ve tried to convince you, first, that monetary policy is ultimately about keeping the available quantity of money from differing substantially, if only temporarily, from the quantity demanded and, second, that doing this boils down in practice to having a money stock that adjusts so as to maintain a steadily-growing level of overall spending on goods and services.
If we’re to pick the right arrangements for achieving this goal, we’d better have a good understanding of the determinants of an economy’s money stock, and of how that stock can be made to expand or contract just enough to keep total spending stable. Although I eventually plan to talk about monetary arrangements that might make maintaining a steady flow of spending a lot easier than our present system does, for now I’m going to stick to discussing how the same goal might be achieved, at least in principle, in our present monetary system or, more precisely, in the system we had until the subprime crisis of 2008. (A later post will discuss how things have changed since the crisis.) This means talking about the Fed’s “instruments of monetary control,” which include devices for regulating the total quantity of bank reserves and circulating Federal Reserve notes, and also for regulating the quantity of bank deposits and other forms of privately-created money that will be supported by any given quantity of bank reserves.
Money Proper and Money Substitutes
In trying to explain how these instruments of monetary control work, I’m tempted, if only for the time being, to revert to some old-fashioned terminology that, whatever its other shortcomings, seems more useful than modern terms are for shedding light upon the nature of money creation. Nowadays economists use the term “money” to refer to anything that’s a generally-accepted medium of exchange. Hence the manifold measures of the U.S. money stock — M1, M2, M3, MZM, and so forth — all of which include various sorts of bank deposits. To refer specifically to the dollars that the Fed itself creates, including both bank reserves and Federal Reserve notes circulating outside of the banking system, they use the terms “high-powered money,” or “base money,” or “the monetary base.”
In the old days, in contrast, economists — or many of them, in any event [1] — liked to distinguish between what they considered money in the strict sense of the term, or “money proper,” and “money substitutes.” Both money proper and money substitutes serve as generally accepted means of exchange. The difference is that, while “money substitutes” consist of various kinds of instantly-redeemable IOUs or promises to pay, “money proper” refers to the stuff that the promises promise, that is, what a bank customer expects to get in exchange for the substitutes if he or she asks the bank to pay up.
A century ago, when the terms were still current, in most industrialized economies “money proper” consisted of gold coins, while paper banknotes and demand deposits that were redeemable in gold were mere money substitutes. Today the same terminology might be used to distinguish the irredeemable currency supplied directly by the Fed from the redeemable exchange media created by commercial banks and other private financial firms. According to it, and thanks to a few twists of fate, paper Federal Reserve notes are now “money proper,” while bank deposits, and checkable deposits especially, are “money substitutes.” Note that “money proper” in this context isn’t quite the same thing as what modern economists call “high-powered” or “base” money, because the last includes bank reserves, which aren’t actually “money” at all: they are, true enough, means of payment so far as banks themselves are concerned, but so far as the general public is concerned, it’s bank deposits, rather than the bank reserves that stand behind those deposits, that serve as money.
Real Money as “Raw Material” for Banks
Why drag-in the old-fashioned distinction between money proper and money substitutes? Because it serves to remind us that even today the “money” that commercial banks and other private-market financial firms produce is in an important respect not the real McCoy at all, but ersatz (if often more convenient) stuff that serves in place of it, and does so only because the firms that supply it, not only make it very convenient to use (e.g., by swiping a debit card), but at the same time offer its users something akin to money-back (which is to say, a “money proper”-back) guarantees. It’s owing to such guarantees — that is, to the fact that bank deposits are, or are supposed to be, readily redeemable in central bank notes — that bank deposits usually command the same value as the “money proper” for which they’re a stand-in. Today, of course, those guarantees are for most depositors further reinforced by the presence of deposit insurance, as well as by the knowledge that government authorities consider some banks “too big to fail.” But such government guarantees don’t allow banks to manage without reserves: they only reduce the likelihood that a bank’s panicking customers will all rush at once to exchange its money substitutes for “real” money.
The understanding that bank deposits and such derive their value at least partly from the fact that banks are prepared to convert them into “money proper” in turn helps us to appreciate how private financial institutions’ ability to create money substitutes depends on their access to “real” (that is, central-bank-created) money. It depends, in the first place, on the amount of such “real” money that these firms keep on hand, either in the shape of actual central bank currency (“vault cash”) or in that of deposit balances they maintain at the central bank that are themselves readily convertible into central bank notes, and, in the second place, on their ability to borrow “real” money either from other private firms or from the central bank itself should their own inventories of it run out. One might even go so far as to think of “real” money (central bank notes and deposit balances) as a crucial “raw material” from which money substitutes (various sorts of bank deposits) are made.
The importance of these insights for a proper understanding of central banks’ devices for monetary control becomes instantly apparent once one realizes that, by regulating the actual quantity of its outstanding notes and deposit balances, together with the terms upon which it is willing to make more of the last available on credit to private sector financial firms, a central bank is able to control, not just the quantity of circulating paper money, but the quantity of money substitutes created by the private sector. Indeed, since the quantity of circulating currency tends to grow along with the extent of commercial-bank deposit creation, that quantity itself ultimately depends on the quantity of reserves that central banks make available to private financial firms.
Open-Market Operations
It follows from this that the most obvious way in which a modern central bank can regulate an economy’s total money stock is by adjusting the available quantity of bank reserves and circulating currency. Central banks can most readily do that by adjusting the total size of their balance sheets, which they do by either acquiring or selling assets. For example, if the Fed wants to increase the stock of bank reserves by, say, $100 billion (admittedly a mere trifle, these days), it has only to purchase $100-billion worth of Treasury securities or other assets from dealers in the secondary or “open” market.[2] To pay for the securities, the Fed wires funds into the sellers’ bank accounts, instantly increasing the total quantity of bank reserves by the same amount. The banks that receive the new reserves will then have more “raw material” on hand to support their own and, eventually, other financial firms’ creation of various kinds of money substitutes. Just how this happens–and especially how it is that each dollar in fresh reserves can ultimately inspire the creation of several dollars-worth of substitutes, will be among the subjects of our next installment.
To shrink the money supply, on the other hand, the Fed has only to sell-off some of its securities, or to let them “roll” off its balance sheet as they mature, instead of replacing them. When the Fed sells $100 billion in securities, the sellers have their banks wire funds to the Fed for the amounts they purchase, essentially instructing the Fed to deduct the wired amounts from their banks’ reserve balances with it.
Although changes in the size of the Fed’s balance sheet — that is, in its total assets and liabilities — often involve like changes in the quantity of high-powered or base money (currency and bank reserves), and corresponding changes in the total money stock, this isn’t always so. Although banks’ reserve balances and outstanding Federal Reserve notes make up the bulk of the Federal Reserve System’s total liabilities, those liabilities also include deposit balances of the U.S. Treasury, of foreign central banks, and of some GSEs. Because these other Fed customers are, unlike banks, not in the business of creating money substitutes, their share of the Fed’s total liabilities doesn’t contribute, as the banks’ share does, to the creation of such substitutes. It’s possible, therefore, for the quantity of base money, and of various monetary aggregates, to change independently of any overall change in the size of the Fed’s balance sheet. An increase in the share of Federal Reserve deposit balances belonging to ordinary U.S. banks, rather than to the Treasury, foreign central banks, or GSEs, will, for example, lead to an increase in the total money stock, other things unchanged, while a decline in that share will reduce it.
Repurchase Agreements
Outright Fed security purchases or sales are only one of two sorts of “open-market operations” the Fed resorts to to change the total size of its balance sheet. The other involves so-called “repurchase agreements” — “repos” for short. A security repurchase agreement is literally a sale of a security that’s coupled with an agreement to buy the security back for a specific price and at a specific time. (A “reverse” repo is thus a purchase combined with an agreement to resell.) However in practice repos (and reverse repos) are practically equivalent to securitized loans, where the security that’s temporarily “sold” serves as collateral to secure a loan from the purchaser to the buyer of an amount equal to the purchase price, and the difference between that price and the later, “repurchase” price is the interest on the loan. The self-reversing nature of the Fed’s repos and reverse repos, many of which are “overnight” rather than “term” agreements (that is, ones providing for repurchase a day after the original purchase) has caused the Fed to prefer them as a means for achieving temporary adjustments to the money stock, while treating outright security purchases as a way of providing for permanent monetary expansion, and especially for secular growth in the demand for Federal Reserve notes. Since the crisis, however, the Fed has come to treat repos, and particularly overnight reverse repos (ON RRPs) with Money Market Mutual Funds and GSEs, as a means for securing long-term monetary control.[3]
As I’ve said, by altering the size of its own balance sheet, and especially by altering the available quantity of bank reserves, the Fed is able, not only to influence its own direct contribution to the money stock, consisting of the quantity of Federal Reserve notes circulating within U.S. borders, but to influence the availability of “raw material” that banks must have in order to “manufacture” readily transferable deposits and other “substitutes” for cash. But while comparing a bank to a factory is helpful up to a point, we mustn’t take the comparison too seriously. For while it’s true that banks can only create and manage deposits provided they have access to reserves, including vault cash, the connection between reserve “input” and deposit “output” is rather different from what goes on in any factory. Indeed, it’s different because it depends, not just on what any single bank can “make” out of a fresh increment of reserves, but on what the banking industry as a whole can make from it, which turns out to be something else again.
Explaining the relation between the Fed’s creation (or destruction) of bank reserves and banks’ creation (or destruction) of deposits takes a little effort, not in the least because doing so means confronting the different ways in which economists on one hand and bankers and banking consultants on the other look at the process, and deciding whether the difference is due to substantive disagreement, or mere semantics. Since that’s going to take more than one or two paragraphs, and this post is already long, I’ll take it up next time.
Next: The Reserve-Deposit “Multiplier”
________________________________
[1] Ludwig von Mises and Irving Fisher are two of the more prominent economists who employed this terminology, which can be traced to the early-to-mid 19th century writings of members of the British Banking School.
[2] Unlike some other central banks, the Fed is prohibited from purchasing Treasury securities from the government. Instead, it purchases securities already outstanding from a group of designated “primary dealers.” The financial losses of two such dealers — Bear Stearns and Lehman Brothers — figured prominently in the recent financial crisis.
[3] In particular, the Fed has used ON RPPs to encourage MMMFs and GSE’s to lend to (or park funds with) it, and to thereby reduce the quantity of Federal Reserve dollars available to banks. The Fed is thus able to reward non-banks for holding (instead of lending or investing) cash, despite the fact that they can’t keep interest-bearing balances with it. By simultaneously raising both the interest rate it pays on bank reserves and the ON RPP rate, as it did in mid-December 2015, the Fed is able to engage in monetary “tightening” without having to reduce the overall size of its balance sheet. I will have more to say about this and other post-crisis changes in the way the Fed conducts monetary policy in a later post.
