Neil Wallace thinks about money a lot. That might be true of all economists, of course, but Wallace is widely considered the father of money’s “microfoundations,” which is to say, its very essence—how families and firms conceive of and use it. Without that base, many believe, macroeconomic theory and policy stand on unsteady ground. But once solid micro-level fundamentals are established, predictions about the impact of change at macroeconomic levels can be made with greater certainty. Wallace began working on monetary microfoundations in the early 1970s after reading a paper by former classmate Robert Lucas, an early proponent. The so-called Lucas critique, says Wallace in the following interview, “essentially broke down the barrier between microeconomic theory and macroeconomic theory.” Ever since, he has been building crucial monetary bridges from one to the other. In his early years at the University of Minnesota, often in partnership with the Minneapolis Fed, Wallace collaborated frequently with Tom Sargent. They, with Chris Sims and Ed Prescott, were soon anointed the “Four Horsemen” of Minnesota economics and viewed as leaders of a new wave in economic theory. Prescott, Sargent and Sims recently won Nobel awards for research begun at Minnesota. Wallace is far more modest about the value of his early work, pioneering though it was. “I learned all my serious economics very slowly at the University of Minnesota. And I’m still learning. I started at such a low base,” he jokes, “and I’ve been pretty lazy.” Few agree with that assessment, of course, and the following conversation gives a glimpse of the range and insights of his prodigious research, from analyses of government guarantees and risk-taking to the impact of quantitative easing and why foreign currency markets are different from all others.

SEPTEMBER 2013 PHOTOGRAPHS BY PETER TENZER 28 13 DECEMBERSEPTEMBER 2013 2013 THE INFLUENCE OF LUCAS ’72 Fed at that time. We were engaged in a number of projects to try to, you might That Lucas paper, “Expectations Region: In the early 1970s, you were say, bring some sort of rigor into the and the Neutrality of Money,” working with Tom Sargent at the Univer- making of monetary policy. had a major impact on our work. sity of Minnesota to build better models John’s vision involved building a mod- for deriving optimal policy rules under el of the macroeconomy that would be Even aside from its very varying economic assumptions. And useful to policymakers. Well, the Phillips important role as a counterex- then you got an early draft of Bob Lu- curve and how to build a relationship in ample to thinking of a Phillips cas’ 1972 paper on rational expectations. a model to account for it, and then to say Several years ago, you told me, “That pa- whether this relationship is exploitable curve correlation as invariant per cut away the underpinnings of what or not through policy—that was clearly to policy interventions and, we’d been doing.” at the heart of any such model. Tom and therefore, exploitable, it changed I were working on it, and doing nothing Wallace: Right. very different from what was standard at the standards for how we do the time. macroeconomics. It said you’ve Region: How did that paper influence And then I came across this working got to think about the econo- your work? In what fundamental ways paper by Bob Lucas. did it change your thinking about eco- And had I not known Lucas from hav- mies of dynamic systems. It also nomics and macro models? And perhaps ing been a classmate, I probably wouldn’t essentially broke down the barrier the course of your career since then. have read the paper. But knowing him between microeconomic theory and having a high opinion of him, I did Wallace: Bob Lucas and I had been fellow try to read the paper. Although it was and macroeconomic theory. grad students at the University of Chica- a pretty hard paper, I could quickly see go, and we both took Milton Friedman’s what the main message was, and I real- price theory course at the same time. This ized that it undercut what we were do- ing and that we ought to reorient our was two quarters of two, two-hour ses- move on to the next date. But it was re- research endeavor in a major way. sions a week, so it’s a lot. And Friedman ally just solving the model one date at a was well known for giving take-home time. Region: It was that convincing to you at exams, which were very open-ended That’s what Keynes did, and almost that point? questions—like, “This is a mystery; how everything up until that time, pret- do you think about it? How do you ex- ty much everything that grew out of plain it?” Those were the exams. Wallace: Yes, it was. Because I kind of Keynes’ general theory, was like that. Although this was microeconomics, believe that some of the best econom- And yet, all macroeconomic phenom- called “price theory” at Chicago, he gave ics consists of counterexamples. People ena inherently involved many dates. For us a question about the Phillips curve, a think A is true, and someone builds a lit- instance, people are making decisions negative association between inflation tle model whose ingredients don’t seem about how much to save and consume. and unemployment. He had briefly talk- particularly weird and it implies not A. Well, if they’re saving, it’s for the future, ed about this in class, but the take-home And so you’ve got to confront that. so they obviously have to be thinking exam question was basically, “How does about the future. this correlation come about?” I have no Region: These are theoretical counterex- Tom and I had been struggling with recollection of what I wrote. amples, not necessarily empirical? how to put these notions about the fu- ture into models. Tom had already done Region: This was the early ’60s? Wallace: Well, for me, yes, and certainly some of this in his Ph.D. dissertation at that paper was a theoretical counterex- Harvard several years earlier, but we were Wallace: This was the winter of 1961. Bob ample. And people had varying reac- struggling with trying to do this kind of Lucas was in this class, as I said, so may- tions to it, but Tom and I bought into it. thing in complete economy models. At be this was the start of his eight, nine, 10 And, in part, we bought into it because the time, my mathematical tools weren’t years of work on this question. Now, in macroeconomics at the time was emerg- good enough to do this. the early 1970s, Tom and I were work- ing from what we call static models. Such ing together under the general auspices models take conditions at a date and try Region: But as you and Tom developed of John Kareken, who was economic ad- to sort of figure out what is going to hap- the math, how did it focus your future viser to the president of the Minneapolis pen at that date and then, based on that, research?

DECEMBER 2013 14 Wallace: Well, that Lucas paper, “Expec- to have. But economists are not satis- tations and the Neutrality of Money,”— fied with that easy acceptance and have When two people meet, it’s often which, by the way, I’ve heard Lucas say developed a number of theories about the case that one has something at a conference, speaking publicly, that its existence and utility, as a measure of that the other person wants, he regards as having been a waste of his value, medium of exchange and so on. time—had a major impact on our work. As a monetary theorist, you’ve studied but not vice versa. And without Even aside from its very important money far more deeply than most. Many money, nothing can happen. role as a counterexample to thinking of a economists consider you the intellectual Trade will not occur. Now, this Phillips curve correlation as invariant to father of money’s microfoundations. policy interventions and, therefore, ex- When you introduce this topic to story is incomplete. ploitable for policy, that paper did other students, how do you discuss the issues things. involved? Maybe it makes sense here to It changed the standards for how we quote from your 2008 discussion of fiat do macroeconomics. It not only said money in Palgrave, where you wrote, you’ve got to think about the economies “Money is helpful when there are ab- Aristotle have this absence-of-double- of dynamic systems and solve for the sence-of-double-coincidence difficul- coincidence notion? Probably not. But whole path of outcomes, not thinking ties that cannot be easily overcome he finds someone named Paulus, a Ro- about solving it just one date at a time, with credit; and a good money has de- man jurist in the second or third century but it also essentially broke down the sirable physical properties—recogniz- A.D., who said something like, when barrier between microeconomic theory ability, portability and divisibility.” two people meet, it’s often the case that and macroeconomic theory. [See Wallace 2008.] one has something that the other person I don’t know if that’s a good starting wants, but not vice versa. And without Region: Which is why establishing mi- point, but … money, nothing can happen. crofoundations became crucial. Wallace: It’s a reasonable starting point. Region: Trade will not occur. Wallace: Yes, right. Now, the idea of es- Every few years I’ve ended up writing tablishing microfoundations for macro- something about how we should think Wallace: Trade will not occur. economics had been in the air for a long, about money, and that paper was one long time. But people were often doing such piece. Region: So there’s one coincidence, not it—and you still see some of this in what two? In this example, at least, one guy I think of as bad textbooks—by taking Region: So, how do you introduce these wants what the other’s got, but not vice the equations that come out of Keynes- issues if you’re sitting down with under- versa. ian economics and trying to tell stories grads who, like most of us, simply take about those equations, one equation at a money, literally, at face value? Wallace: Right. Yes. My [University of time. But those separate stories don’t add Minnesota] colleague Leo Hurwicz al- up to a coherent view of an economy. Wallace: Well, one of the things I say is, ways objected to the “double coinci- Lucas became convinced that to build just because a thing is around doesn’t dence” terminology, even though it’s a model in which agents have a reason- mean we understand it. Cancer is commonplace. He said the word “coin- able view about what’s going on in the around. Does that mean we don’t want to cidence” itself means two. So “double co- world, you have to start with a coherent do research on it? And then I talk a little incidence,” he would say, is a redundant underlying view of an entire economy. bit about the history of thought on this term. But it’s standard usage somehow. Microeconomics had been doing that absence-of-double-coincidence notion. It’s very helpful to have this notion of for a long time. I think Bob’s paper in- single coincidence, meaning that I have fluenced many people in the direction Region: In a word or two, what is the ab- something that you want. That is a kind of thinking we have to do that in macro- sence of double coincidence? of coincidence. So Paulus said this. economics as well. I don’t know about the last edition, but Wallace: A professor at Texas A&M put one edition of [Frederic] Mishkin’s very EXISTENCE OF MONEY me onto a 1923 book called Monetary popular undergraduate money and bank- Theory Before Adam Smith. It was a ing textbooks has Mary or Alice as an Region: Let me ask you about money. Harvard Ph.D. dissertation on the his- economics professor. If she wants some Ordinary people don’t question its ex- tory of thought concerning money. And apples, without money, she’d have to find istence. It simply is. And it’s very useful its author, Arthur Eli Monroe, asks, did an apple grower who wants an econom-

15 DECEMBER 2013 lots of Amish communities around here. 2006.] The Fed publishes monthly re- “Money is memory” is a hugely When they’re isolated, the usual story ports, for example, on quantities of M1 powerful idea. It leads you to about an Amish community—or an iso- and M2, as they’re termed. think about various kinds of lated Israeli kibbutz—is that they didn’t use money. Wallace: Right. Now, I think the idea of payment instruments in terms adding up these quantities into some ag- of the kind of informational Region: Trust was their currency. gregate is a really bad idea. structure that supports them. “Money is memory” is a better idea. Wallace: Well, that’s a word that Douglas It leads you to think about various kinds When you use a credit card, Gale used, but it’s probably not the best of payment instruments in terms of the you’re issued a loan. Why are word. [See Gale 1978.] Think about this kind of informational structure that you able to receive one? Because Amish community. The vision is, if my supports them. The money that is the barn burns down, then everybody will best current counterpart to the “money there’s an informational network come and help me rebuild it. In econom- is memory” idea is currency. You don’t behind your card. Your bank is ics, we try to rationalize behavior without need much of an informational network actually guaranteeing your credit altruism, if we’re able to; so what makes for currency; in fact, you probably don’t that work without altruism? Everybody need any, except for worrying about payment because they know notices who shows up to help rebuild it. counterfeiting. something about you. The bank is When you use a credit card, you’re is- Region: keeping pretty good tabs on what A sort of credit accounting. sued a loan. Why are you able to receive one? Because there’s an informational you do. Wallace: Yes. And the guy who doesn’t network behind your card. Your bank is show up, if he does that repeatedly, will actually guaranteeing your credit pay- get kicked out eventually. This can work ment up to probably some large amount, without money because people remem- as large as you mostly use. And they’re ics lecture—and he didn’t go on and say ber what people have done in the past. doing that because they know something almost no one wants an economics lec- about you. ture, so she’s really in trouble [laughter]. Region: So, money is memory. In any case, that’s what the textbooks say, Region: So that’s a form of memory about and this has been repeated. Wallace: Yes, “money is memory” is a ca- a person’s past transactions. Now, this story is incomplete. That’s sual way to state that. Now, that’s a huge- what I tell students: It’s incomplete. ly powerful idea that I and other people Wallace: Right. Why is it incomplete? Well, think about have been working with. some isolated pioneer family. At times, Region: In a paper you just presented at someone is going to not feel so well, so Region: Kocherlakota [1998], Ostroy the Minneapolis Fed, you wrote, “Money he or she isn’t going to be able to chop [1973] and Townsend [1989], for in- is potentially useful in trade between down wood for a fire. There’ll be lots stance. strangers. It is not needed when every- of absence of double coincidences that one knows what everyone else has done arise in that family situation, many cases Wallace: Right. And I think a lot more in the past.” Now, if you’ve got a credit where one family member is called on to needs to be done. Let me mention one card, the bank is keeping … do a favor for another family member. thing sort of related to that. People ask, Well, are they going to get paid for doing what is money? Friedman and [Anna] Wallace: … pretty good tabs on what you this favor? Schwartz, in their monetary history, right do. at the beginning have a long discussion Region: Of course not, not in a family. about money. Of course, he’s written about A CASHLESS SOCIETY it extensively. So, what is money? They Wallace: Right. Now think about Robin- make a decision that they’re going to use, Region: That leads to another question son Crusoe, after he meets Friday. They if I recall correctly, what we would call M2. about technological advances and po- don’t need money, but again, there might You’ve heard of these subscript things? tentially a cashless society. We’ve got be plenty of absence of double coinci- electronic payments, debit and credit dences. Now think further. Here we are Region: Sure, categories of money in cards. Now people use smartphones for in the middle of Pennsylvania. There are the total stock of money. [See Bernanke retail purchases; soon we might use our

DECEMBER 2013 16 The idea of a cashless economy, both in theory and in those examples of an isolated Amish community or an Israeli kibbutz, shouldn’t trouble us. We know about that already, in some sense. And maybe we’re headed that way. And so, what is left for central banking in that kind of world? Well, that’s a matter of—I would say—some dispute in the profession. It almost goes back to Phillips curve issues.

17 DECEMBER 2013 fingerprints. They all reduce the need for ate moral hazard, that they encourage cash and checks, so an even less physical undue risk-taking and therefore ineffi- In general, you want to describe representation of that M2 or whatever. cient resource allocation. But at the same banking illiquidity as a balance Does that mean that money itself is time, government guarantees for depos- sheet which is unbalanced in becoming less important—even as a re- its are considered essential for banking. cordkeeping device? If cash is less im- These issues were central to some terms of maturities: short-term portant, how does that change monetary parts of the recent financial crisis, and liabilities and, on average, longer- theory? then Dodd-Frank legislation tried to ad- term assets. Some [economists] dress some of that. I know you are reluc- Wallace: Well, you asked me earlier how tant to weigh in on specific policy issues, have said this is a natural thing. I introduced money to undergraduates, so I won’t push it that way, but I wonder This is what banks are for. and there I talk about an Israeli kibbutz what general thoughts you’ve had about where there’s no money. The vision of the factors behind this particular finan- that is a small community, which makes cial crisis and then the evolving relation- it easy for us to remember. But the idea ships between financial institutions and of remembering actions, well, it doesn’t the government since then. a book called Economic Policy for a Free have to be a small community. It’s an ab- Society. And Friedman [1967] often stract idea and, with the technology we Wallace: In some circles, Kareken and credits him [with this idea]. Simons said now have, a lot can be remembered. I get some credit for pointing out this it’s not enough to force banks to hold risk-taking incentive of deposit insur- 100 percent reserves. He said we should Region: Too much? ance. But I don’t know, it was pretty well prevent limited liability institutions, like known in some sense. corporations, from issuing debt. (Now Wallace: It may be too much. But the that’s fairly ironic because what is the use idea of a cashless economy, both in the- Region: You formalized the thinking on it. of limited liability if you can’t borrow?) ory and in those examples of an isolated But he said you can tell people over Amish community or an Israeli kibbutz, Wallace: Maybe a bit, not all that well, and over again that a certain bond is shouldn’t trouble us. We know about that but a bit. But let me introduce a second risky, yet they’re not going to act like it’s already, in some sense. And maybe we’re literature that’s relevant to this question. risky. And then, when it fails to pay off, headed that way. Going way back, one of the issues that because they haven’t anticipated that that And so, what is left for central bank- had troubled economists and that they might happen, it’s chaos. So that’s what ing in that kind of world? Well, that’s a weighed in on is the issue of fractional he said, writing in the 1930s. matter of—I would say—some dispute reserve banking or, more generally, how So on one side is a bunch of people in the profession. It almost goes back do we think about an illiquid banking who are saying banking system illiquid- to Phillips curve issues, because it goes system? ity—and maybe illiquidity more gener- back to issues about whether prices are ally—is harmful; we ought to regulate it sticky or not, which was sort of at the Region: Where a bank holds cash on out of existence. And on the other side heart of Phillips curve theories. hand just a portion of what customers were people who vaguely said, “It’s natu- have deposited. ral, that’s the function of banks.” Region: And it still is central in a lot of In 1983, Doug Diamond and Phil Dy- New Keynesian work. Wallace: Sure, and you can think about bvig published what to me was an eye- this in terms of term-structure risk, so, opening paper, a very simple, stripped Wallace: Absolutely, it still is. in general, you want to describe bank- down model, but one whose elements all ing illiquidity as a balance sheet which seem quite reasonable. One element is DEPOSIT INSURANCE, TBTF, is unbalanced in terms of maturities: that people can’t fully plan the pattern of MORAL HAZARD short-term liabilities and, on average, their future expenditures, so they want longer-term assets. Now, economists something like a demand deposit to be Region: Let me ask you about a differ- have weighed in on this for a long time. able to spend at any time. A spending ent area of your research, something Some have said this is a natural thing. opportunity might arise that they hadn’t you’re very well known for. And again, This is what banks are for. anticipated, so they want the flexibility of it involved John Kareken. The two of you Others have said this is dangerous, being able to spend at any time. pioneered thinking about how govern- and we ought to regulate it out of exis- But giving them that flexibility in, say, ment guarantees for bank deposits cre- tence. Henry Simons, for example, wrote the form of a demand deposit, allowing

DECEMBER 2013 18 them to withdraw whenever they want, think about what it would mean to im- means they might also withdraw not just pose 100 percent reserves. You can do Others have said this is when they want to spend but because that in that model, and then you can dangerous, and we ought to they’re worried about the [safety of that trace back to the welfare of the people financial] institution. in the model who are really, essentially, regulate it out of existence. The just the depositors. You can trace back to literature on banking has always Region: And if they withdraw for that them the consequences of that policy in been—like that on money—a reason, lots of other depositors might as that model. well, creating a bank run. That’s a huge accomplishment. The troublesome literature. literature on banking has always been— I don’t know any research Wallace: Right. Now, one element in the like that on money—a troublesome liter- that answers the question, “Are model is people’s desire for flexibility. It’s ature. This goes back to economists’ feel- a little bit technical, but in the model, ings that the general competitive model, we seeing the right degree of the realization of the spending desire is often labeled the Arrow-Debreu model, illiquidity?” This has come up private information. So, as an example, is the main model in economics. It’s very repeatedly. Is that an instance of general. We don’t need to have a special when you go up to the bank window to the right kind of maturity trans- make a withdrawal, it’s not written on theory of production for bookcases and your forehead whether you genuinely a special theory for bottled water. formation that Diamond-Dybvig want to make a payment or whether But when people try to shove banking says has a social role? Or is it into this model, it’s hugely unsuccess- you’re worried about the solvency of the really just a combination of this bank. That information is private. Then ful. Why? Because anything that banks a second element in the model is that might be viewed as doing is redundant in risk-taking incentive—the moral the technology is such that longer-term that model. According to the Arrow-De- hazard aspect of government breu model, you face prices at which you investments have bigger payoffs than guarantees—and this overselling short-term investments. can costlessly trade anything for any- thing. More generally, no activity that we of the quality? Region: That’s the maturity transforma- see in the economy that has to do with tion, from short-term deposits to longer- transacting fits comfortably within that term lending with higher returns. That’s model. In particular, nothing in the GDP where the bank wants to put all those accounts that falls under the FIRE head- short-term deposits. ing—finance, insurance, real estate—fits Wallace: Now we have this Diamond- into that model. Dybvig model, which, if you buy into Wallace: That’s why it’s socially a good Diamond-Dybvig overcame that. It it, says some degree of illiquidity in the idea for those deposits to be used to fi- was at the cost of abstracting from de- financial system plays a desirable social nance this long-term investment. It’s like tails that some people think are quite im- role. But we’ve also got this other side of wine, if you leave it in, it’s going to turn portant. But still, it was a major break- things that says various loan guarantees good. If you withdraw it quickly, it’s go- through. It pointed out, for the first time, potentially lead to too much risk-taking. ing to be just the grape juice that you a social role for an entire illiquid finan- started out with. cial system. Not just one bank, not just Region: That’s the moral hazard. the banking system, it is a model of an Region: A great analogy. So who steps in entire economy. And it points out a so- Wallace: That risk-taking often takes the and makes sure that that transformation cial role for this economy to have a spe- form of illiquidity. You read about what is possible? cific property of illiquidity—namely, if Bear Stearns was doing, what Lehman all these depositors try to withdraw at was doing. They were borrowing very Wallace: In some sense, this is a con- once, it’s in trouble. short and financing long-term asset po- troversial model of banking. Certainly sitions. You read casual accounts about many, many people liked the Diamond- Region: So there’s a social role for this Wall Street people going to pension Dybvig model. kind of banking system, but it won’t ex- funds and selling like they’re used car Earlier I said that Lucas’ paper was ist unless that private information that salesmen. I mean, the way to overstate a microeconomically coherent model people might have, their worries about the quality of what you’re selling in this of the entire economy. Well, so is Dia- the bank’s solvency, unless it’s reassured area is to say it’s safe and liquid. mond-Dybvig, because it allows you to by government, by deposit insurance. Now, look at liquidity in a financial

19 DECEMBER 2013 system. On the one hand, this Diamond- Dybvig model says that some degree of system illiquidity is desirable. But we’ve also got this moral hazard aspect of vari- ous explicit and implicit forms of insur- ance and we’ve got this general desire to oversell, to overstate the quality of what you’re trying to peddle. When we look at the whole financial system, well, I don’t know any research that answers the question, “Are we see- ing the right degree of illiquidity?” This has come up repeatedly. In the Asian fi- nancial crisis, people were investing in these countries, but rather than making direct equity investments, they were giv- ing short-term loans. Is that an instance of the right kind of maturity transfor- mation that Diamond-Dybvig says has a social role? Or is it really just a com- bination of this risk-taking incentive— the moral hazard aspect of government guarantees—and this overselling of the quality of what you have?

Region: So legislation is hard-pressed to ensure the “right” degree of illiquidity. We have institutions—and lot Wallace: It is. of the world does—independent QUANTITATIVE EASING ment asset exchanges.” I think it’s fair to central banks and so on—that say that economists heard your plea. In seem to create a separation Region: Let me ask about another of a very substantive way, it has served as your well-known papers, published in the basis for much subsequent research between monetary and fiscal the American Economic Review in June on monetary policy, from Sargent and policy. But despite this separation 1981, “The Modigliani-Miller Theorem Smith in 1987 to Eggertson and Wood- into different institutions, you for Open-Market Operations.” In it you ford in 2003, who wrote specifically that established the “irrelevance proposition” their work “was in the spirit of the irrel- can’t really separate monetary akin to Modigliani-Miller’s regarding evance proposition for open-market op- and fiscal policy. People own the corporate liability structuring, which erations of Wallace (1981).” government’s portfolio and one said that whether a firm’s debt is held as They suggested that quantitative stocks or bonds has no effect on overall easing, as it’s known—large-scale asset way to talk about that ownership corporate value. purchases by central banks—is general- is that these different earning To vastly oversimplify, your proposi- ly ineffective as a means of stimulating streams are handed back in the tion was that under a specific set of con- the economy, inferior in their minds to ditions regarding fiscal policy, the size forward guidance. Other scholars have form of transfers of some sort. and structure of a government’s portfolio reached roughly the same conclusion, of debts and assets has no impact on the theoretically. Empirical work seems economy as a whole. to show moderate impact at best. The And in that paper, you made a “plea,” Fed continues to debate the policy, of that the proposition “should serve as the course; the FOMC is discussing it in starting point for analyses of govern- Washington right this moment. And

DECEMBER 2013 20 markets these days react sharply to any Region: Which means that such asset- talk of “tapering” QE. purchase policies should be pretty inef- The models in which the invisible What are your general thoughts fective. hand works, they don’t have fiat about quantitative easing policies for large asset sales when we’re at the zero Wallace: Yes, they should be pretty in- currencies. They’ve got cars and bound? effective. Part of that involves thinking apples, where the value of these about the interaction between monetary things is grounded in what we and fiscal policy. We have institutions— Wallace: I’m kind of surprised at the re- can do with them and that we cent attention that this paper has got- and lot of the world does—independent ten. Chris Sims, for example, remarked central banks and so on—that seem to like to eat them and things like about it in his AEA presidential lecture. create a separation between monetary that. You can’t do that with this and fiscal policy. Some people think [See Sims 2013.] It’s right that the back- stuff: money. So you don’t have ground model of that paper has money that’s very important. holding its own in terms of rate of re- this usual appeal. The key idea is turn. That means that money has a high Region: The independence of central potential perfect substitutability banks. enough return that it is being held as an among different currencies, a investment. When interest rates are at the zero lower bound, that’s certainly an Wallace: Yes. I don’t really have a view on potential that I expect to be example of money holding its own, in that. I think it would take some rather realized unless there are legal deep political economy model to really terms of rate of return. restrictions that inhibit it. Again, when I use the word “mon- rationalize that, but maybe it’s good. I ey” in this context, we might as well be don’t know, but despite this separation thinking about currency. It’s very differ- into different institutions [the central ent from a situation where, say, T-bills bank, Congress and the administration], are yielding 8 percent nominal and cur- you can’t really separate monetary and rency is yielding nothing. fiscal policy. FOREIGN EXCHANGE MARKETS So, that paper assumed that these as- sets were holding their own in terms of Region: It’s just dividing the pie different Region: Just a month before you pub- rate of return, which, by the way, is what ways. lished your AER paper on open-mar- the Modigliani-Miller theorem assumes ket operations, your article with John about bonds and equities. It’s assuming Wallace: Yes. If interest rates are positive Kareken on exchange rates appeared in that bonds and equities are packages of and the Fed buys some Treasury bills and the Quarterly Journal of Economics. That claims, and there’s a complete contingent hands out zero-interest currency or zero- paper with Kareken was closely linked market in claims, and these things are interest bank reserves, the way it used to to your 1979 Minneapolis Fed Quarter- just bundles of claims. be, there are fiscal consequences because ly Review piece, which had a much less Merton Miller is a very nice, modest the Fed turns over its interest payments, technical title, which I love: “Why Mar- guy. He has said something like, “Should its earnings, to the Treasury. kets in Foreign Exchange Are Different I have really gotten a Nobel Prize for say- The standard assumption that most from Other Markets.” ing ‘no matter how you slice up a pie, it’s policy models made about this was still a pie?’” That’s the way he character- that when the Fed made this kind of Wallace: Does that title mean anything to izes the Modigliani-Miller theorem. open-market purchase, lump-sum tax- you? Do you know where it comes from? I was extending that conclusion to es fell to leave the effect on the bud- Well, I’m Jewish by background, and at include currency, when the economy is get unchanged. Well, you can make the Jewish holiday, Passover, there are at the zero lower bound. When it is, the that assumption in a model, but is that meals called Seders, and there are these Fed’s monopoly on currency becomes ir- what really goes on? things called the four questions. relevant, because there are short-term se- But this 1981 paper said, in effect, that I don’t know what they are, exactly, curities which are perfect substitutes for it people own the government’s portfolio but one of the kids at the table is sup- because they have the same rate of return. and one way to talk about that ownership posed to ask, “Why is this night differ- Other people can issue those claims, hold is that these different earning streams are ent from other nights? Why are we eat- them and so on, so there’s nothing special handed back in the form of transfers of ing unleavened bread?” So I thought of about the Fed’s monopoly on issuing cur- some sort, and so it really is like a Modi- that title, and I thought it was too good rency in those circumstances. gliani-Miller theorem. to waste!

21 DECEMBER 2013 Region: Many of your other titles, like What would the demand for it be? that one, are wonderful. “Float on a The point of that article is to say, in part Despite rules intended to prevent Note” is another example. Well, in that … Well, had I thought about it, the U.S. the support of government bond QR piece, you said fiat money is different has at several times tried to introduce a and “we should at least consider pursu- $1 coin. One of those attempts was the markets in member states, the ing an explicit policy directed toward Susan B. Anthony coin. So here’s another ECB [European Central Bank] cooperative and permanent exchange cute title, “Why Call the Susan B. a One?” seems to be doing just that. In market intervention or toward controls Why not let it float? These currencies are on asset holdings.” different. When we say, let the market the U.S., the Fed should maintain That’s a pretty controversial statement, operate … an orderly market in government then and now, although I should mention securities. It seems a bit far-fetched that Minneapolis Fed economists took Region: For cars or oranges. up that cause for a good decade. to reconcile orderliness with Wallace: Yes. What’s that a reference to? default, but maybe. Thus, while Wallace: Did they? That’s a reference to what economists call one can imagine arrangements the first welfare theorem, which is the in which control of the price level Region: Art Rolnick and Warren Weber invisible hand. But the models in which wrote their 1989 Annual Report essay, “A the invisible hand works, they don’t have is maintained in the presence of Case for Fixing Exchange Rates,” based fiat currencies in them. They’ve got cars large and unsustainable govern- on your work. [See Rolnick and Weber and apples in them, where the value of ment deficits, it rarely happens. 1990.] But I think it’s fair to say that that these things is grounded in what we proposition for market intervention in can do with them and that we like to eat foreign exchange has little traction among them and things like that. You can’t do economists generally, and policymakers, that with this stuff: money. So you don’t certainly. Why do you think that is, and have this usual appeal. do you still believe in the proposition, Now, if you read Milton Friedman value of the Canadian dollar. And vice “indeterminacy,” as you’ve termed it? carefully, you’ll sometimes find state- versa, if the border for currency usage ments that sound like he’s appealing to extends into the U.S. Wallace: I do. Suppose Greece left the the same thing he would appeal to for not Hence, the key idea is potential per- euro system and started up its own cur- regulating the prices of cars in terms of fect substitutability among different cur- rency. apples. But to be fair to him, he is mostly rencies, a potential that I expect to be re- thinking differently. He’s going back to alized unless there are legal restrictions Region: Let’s call it the drachma. sticky prices, saying, why have millions that inhibit it. of nominal prices change when we need Wallace: Sure. Greece returns to its old to have international adjustment, as we REMAINING QUESTIONS currency. Well, would it be easy to fore- would need to have if we had fixed ex- IN MONETARY THEORY cast what the demand for that currency change rates or a single currency? Why would be? I mean, are we going to use one not let this just one price change? Let the Region: Let me ask you a question about of Milton Friedman’s estimated demand drachma fall in terms of the euro. There’s remaining questions. Your research has functions for money to forecast that? I essentially no formalization of that idea, addressed central issues in monetary don’t think it’s so easy to do. You’d prob- but it is there. theory, and in a 1985 Minneapolis Fed ably try to put in some restrictions to pre- The issue is why are different cur- working paper (281), “Some Unsolved vent people from continuing to use euros. rencies other than perfect substitutes? Problems for Monetary Theory,” you Think about the border between the raised two key policy issues in monetary Region: But that’s intervention. U.S. and Canada. There is a legal border. theory that remained at that point to be But what is the border regarding usage answered; we’ve talked about them both Wallace: Right. When there was agitation of the U.S. and Canadian dollar? Why briefly today: Should interest be paid on for an independent Quebec, the party does it have to correspond to the legal money? And, should currency provision advocating independence actually said, border? The more deeply it extends be in the hands of the government, rath- “We’re going to still use the Canadian into Canada, the lower is the demand er than the market?” dollar.” But suppose it had said, “No, for Canadian currency, and with fixed Have they finally been resolved to we’re going to have our own currency.” supplies, the lower must be the relative your satisfaction?

DECEMBER 2013 22 Wallace: When I wrote those things in the Unpleasant Monetarist Arithmetic,” 1985 paper, and when I wrote that 1981 with Tom Sargent, showing that because One way that we advance in paper on the irrelevance proposition, I of that interdependence, price levels are economics is someone invents was not working within models in which determined by fiscal policy as well as a model for one purpose and money is substituting for memory. Or central bank policy. to put it somewhat differently, I wasn’t What implications does this have with then we see other purposes for working within models that I would now respect to current budget deficits and the it. A model was not originally defend as models in which money really steady growth of entitlement expendi- designed to address those issues. has a beneficial role in the economy. For tures? And do you have any thoughts on the past 15 years or so, I think I’ve been the notion that if a government were to And yet, the model allows us to doing that. permit sovereign default, theoretically, think about those things. the central bank still could control price Region: So we’re closer. levels, at the risk of output drops?

Wallace: Well, I think we’re closer. Wallace: Yes, we can imagine arrange- ments under which large fiscal deficits for important countries—Britain during FISCAL-MONETARY POLICY are not monetized. In particular, when the Napoleonic Wars and at other times; INTERDEPENDENCE countries were on a gold standard, some the U.S. during the Civil War. may have gone through debt defaults And despite rules intended to prevent Region: We’ve spoken about the interac- while remaining on the gold standard. the support of government bond mar- tion of fiscal and monetary policy. You However, temporary suspensions from kets in member states, the ECB [Euro- established that inherent link in “Some the gold standard were more common pean Central Bank] seems to be doing

More About Neil Wallace

Current Positions Publications Professor of Economics, Pennsylvania State University, since 1997 Prolific author of research articles published in refereed journals on an array of subjects related to monetary economics; co-author (with Consultant, Federal Reserve Bank of Minneapolis, since 1969 George McCandless) of Introduction to Dynamic Economic Theory: An Previous Positions Overlapping Generations Approach, Harvard University Press, 1991; co-editor (with John Kareken) of Models of Monetary Economies and Professor, University of Miami, 1994-97 The Overlapping Generations Model of Fiat Money, both published by Professor, University of Minnesota, 1974-94; the Federal Reserve Bank of Minneapolis in 1980 Associate Professor, 1969-74; Education Assistant Professor, 1964-69 University of Chicago, Ph.D., economics, 1964 Honors Columbia University, B.A., economics, 1960 Distinguished Fellow, American Economic Association, since 2012 Distinction in the Social Sciences, College of Liberal Arts, Penn State, since 2005 Fellow, American Academy of Arts and Sciences, since 2005 Fellow, Econometric Society, since 1981 Honorary Degree, Universidad del Pacifico, Lima, Peru, 1995

23 DECEMBER 2013 just that. In the U.S., as I recall, the Fed- I think that’s possible for me because Region: Well, thank you for talking with eral Reserve Act says that the Fed should I started at such a low base [laughs] and us about all these things. I’ve really en- maintain an orderly market in govern- I’ve been pretty lazy, so it’s easy for me to joyed this. ment securities. It seems a bit far-fetched improve because that’s just the way it is. —Douglas Clement to reconcile orderliness with default, but Sept. 18, 2013 maybe. Region: Well, I don’t think many Thus, while one can imagine arrange- economists would agree with you. So, ments in which control of the price level which papers? The one you just pre- is maintained in the presence of large sented at the Minneapolis Fed? That and unsustainable government deficits, was a draft, a working paper, “Alter- it rarely happens. native Neo-Keynesian Models for the Study of Optimal Monetary Policy.” FAVORITE RESEARCH Wallace: Yeah, that’s a draft. You Region: You’re well known for a number shouldn’t use that title. I’ve actually of key contributions to economics. We’ve changed the title of that thing, and I talked about some of that work today: don’t know whether I’m going to submit “Unpleasant Monetarist Arithmetic,” the it anywhere. But the substance, yeah, I irrelevance proposition, your deposit in- kind of liked that. surance paper and the foreign exchange research, among other work. But other Region: Other papers come to mind? papers—some still unpublished—are far less known. Wallace: I’m fond of a little paper I did Looking back, which of your papers on the denomination structure of mon- do you most like and which do you con- ey. [See Wallace 2005.] sider less valuable? Is this like picking among children? I don’t want to put you Region: Why do you like that one in par- in that position. ticular?

Wallace: No. It’s not. It’s kind of easy for Wallace: Well, one way that we advance me to answer. in economics—and in other fields I think, too, but certainly in economics—is some- Region: This is the only question where one invents a model for one purpose and you haven’t paused at considerable then we see other purposes for it. length before giving your answer! Both that denomination structure pa- per and “Float on a Note” use a model Wallace: As I sort of indicated earlier, I that’s been around for a while. They may went to good schools, went to Columbia not be dealing with the most important as an undergraduate and it was a good issues, but they use a model that was not experience. I mean, it was intellectually originally designed to address those is- eye opening in a certain way. I went to sues. And yet, the model allows us to the Bronx High School of Science, which think about those things. That’s a fun is a good high school, but that wasn’t and interesting thing. intellectually eye opening at all. Then I went to the University of Chicago. It was Region: And it shows the utility of the an exciting place. model. But I learned all my serious econom- ics very slowly at the University of Min- Wallace: That’s right. It shows the value nesota. And I’m still learning. And I say of the model because these were things this because I think my recent work is that, you know, could not previously be my best work. talked about.

DECEMBER 2013 24 References Bernanke, Ben S. 2006. “Monetary Aggregates and Monetary Policy at the Federal Reserve: A Historical Perspective.” Speech at the Fourth ECB Central Banking Conference, Frankfurt, Germany, Nov. 10.

Friedman, Milton. 1967. “The Monetary Theory and Policy of Henry Simons.” Journal of Law and Economics 10: 1-13.

Gale, Douglas. 1978. “The Core of a Monetary Economy without Trust.” Journal of Economic Theory 19: 456-91.

Kocherlakota, Narayana. 1998. “Money Is Memo- r y.” Journal of Economic Theory 81: 232-51.

Ostroy, Joseph M. 1973. “The Informational Efficiency of Monetary Exchange.” American Economic Review 63: 597-610.

Rolnick, Arthur J., and Warren E. Weber. 1990. “A Case for Fixing Exchange Rates.” 1989 Annual Report. Federal Reserve Bank of Minneapolis. minneapolisfed.org/publications_papers/pub_ display.cfm?id=677

Sims, Christopher A. 2013. “Paper Money.” American Economic Review 103: 563-84.

Townsend, Robert M. 1989. “Currency and Credit in a Private Information Economy.” Journal of Political Economy 97: 1323-44.

Wallace, Neil. 2008. “Fiat Money.” In The New Palgrave Dictionary of Economics, 2nd ed., edited by Steven N. Durlauf and Lawrence E. Blume.

Wallace, Neil (author), Manjong Lee and Tao Zhu. 2005. “Modeling Denomination Structures.” Econometrica 73: 949-60.

25 DECEMBER 2013