Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Tuesday, October 23, 2012

No more bubble talk (please!)

I am currently at the Institute of Advanced Studies in Vienna, which hosts one of the best little macro conference in Europe: The Vienna Macro Cafe. Excellent papers, lively discussions, wonderful camaraderie, and an unbeatable location. (Let me know if you'd like to be placed on our mailing list.) 
After this uplifting experience, I made the mistake of checking the econ blogosphere. Blah. 

I guess it all started with Paul Krugman (who else?), who goes off here in his usual assertive style: Bubble, Bubble, Conceptual Trouble. Steve Williamson takes issue with some of the claims that Krugman makes here: The State of the World. And then Noah Smith steps in to attack one part of Williamson's post here: Money Is Just Little Green Bits of Paper! Noah gets it all wrong, but that doesn't stop both Krugman and DeLong in congratulating him for an argument that even they apparently do not understand. And so it goes. 

Let me now explain why I think Noah gets the "bubble" issue wrong. Here is how Noah starts off. 
Have you ever heard people say that "money is just little green pieces of paper"? Well, that is exactly what Steve Williamson claims in this post.
Um, no...Steve never said that "money is just little green pieces of paper." So right away, we're off to a bad start.
To understand what Steve meant, we have to start with Krugman's own "definition" of a bubble:
Over and over again one hears that we can’t expect to return to 2007 levels of employment, because there was a bubble back then. But what is a bubble? It’s a situation in which some people are spending too much.
It's a situation in which some people are spending too much? Thanks for that, Paul. To which Steve replies:
What is a bubble? You certainly can't know it's a bubble by just looking at it. You need a model. (i) Write down a model that determines asset prices. (ii) Determine what the actual underlying payoffs are on each asset. (iii) Calculate each asset's "fundamental," which is the expected present value of these underlying payoffs, using the appropriate discount factors. (iv) The difference between the asset's actual price and the fundamental is the bubble. Money, for example, is a pure bubble, as its fundamental is zero. There is a bubble component to government debt, due to the fact that it is used in financial transactions (just as money is used in retail transactions) and as collateral. Thus bubbles can be a good thing. We would not compare an economy with money to one without money and argue that the people in the monetary economy are "spending too much," would we?
The only quibble I have with this reply is Steve's use of the word "bubble." Bubbles mean different things to different people. As Steve emphasizes, the definition should be made relative to a specific model. "Bubbles" are not something you can actually "see" in the data -- "bubble dynamics" are an interpretation of the data. 

In any case, what word might Steve have used instead of "bubble?" While less colorful, I think that the label "liquidity premium" is more accurate. It is the market price of an asset above it's "fundamental" value. The distinction here seems similar to the one that Marx made between "use value" and "exchange value;" see here.

Here is how I like to think about it. Imagine an economy with just one person, as in Robinson Crusoe. Crusoe likes to eat coconuts. So he values coconuts. And he values the trees that produce coconut dividends. One way to measure value here is to ask "how many coconuts would Crusoe be willing to give up to have one more coconut tree?" The answer to this question will provide a measure of the tree's "fundamental" value. Because there are no other people on the island (prior to Friday), there is no exchange value associated with tree ownership. There is no "bubble" in a Robinson Crusoe economy.

The situation can be quite different, however, in an economy consisting of more than one person wanting to trade intertemporally in credit markets. One friction that hampers intertemporal trade is what economists call a lack of commitment. Essentially, people cannot be relied upon to keep their promises. Monetary theorists have shown that in this type of world, various objects may be employed to enhance the volume of intertemporal trade. These objects are called exchange media.

Exchange media may take the form objects that are commonly viewed as "money"--objects that circulate widely from hand to hand, or from account to account. They may also take the form of collateral objects, like the senior tranches of MBS that (until recently) circulated widely in the repo market. Because U.S. treasuries are used widely to facilitate financial transactions, they too constitute an important medium of exchange.

Abstracting from risk, the market price of an exchange medium can be broken down into two components: fundamental value and market value. We can estimate the fundamental value of an asset by assessing its value under the assumption that it is illiquid (i.e., does not circulate as an exchange medium). The difference between market price and fundamental value is a measure of liquidity value. Because an asset may be priced above its fundamental value, there is a sense in which the asset price embeds a "bubble" component according to many popular definitions. But the word is more trouble that it's worth--our discussions might be clearer and more productive if we avoided the term entirely.

Aside: Steve's point is that in a world of financial frictions, exchange media and their associated "bubble" prices may be useful for increasing the level of spending closer to socially optimal levels. If so, then how does Krugman's definition of a bubble--a situation where people are spending too much--make any sense? There may be bubbles that have this property. But then there may be bubbles that do not. Williamson is telling Krugman that he needs to be more careful. The message, unfortunately, seems hopelessly lost. 
Alright then, back to Noah, whose whole column is based on Steve's throwaway comment: 
Money, for example, is a pure bubble, as its fundamental is zero.
To which Noah replies:
Can this be true? Is money fundamentally worth nothing more than the paper it's printed on (or the bytes that keep track of it in a hard drive)? It's an interesting and deep question. But my answer is: No. 
First, consider the following: If money is a pure bubble, than nearly every financial asset is a pure bubble. Why? Simple: because most financial assets entitle you only to a stream of money. A bond entitles you to coupons and/or a redemption value, both of which are paid in money. Equity entitles you to dividends (money), and a share of the (money) proceeds from a sale of the company's assets. If money has a fundamental value of zero, and a bond or a share of stock does nothing but spit out money, the fundamental value of every bond or stock in existence is precisely zero.
While it may not have been clear to the average reader, I happen to know that Steve was referring to a special kind of monetary object: a pure "fiat" currency. "Fiat" in the modern sense of the word means "intrinsically useless" or "zero use value."The USD issued by the Fed may not fit this description exactly because, as others have pointed out, government money does have the power to discharge a real tax obligation. On the other hand, pure fiat money does seem exist; see my post: Fiat Money in Theory and in Somalia. The point is that even fiat money can have exchange value, and if it does, then its value is entirely a liquidity premium or "bubble" and that, moreover, it is probably a "good" bubble to the extent that fiat money facilitates trade. 

What of Noah's claim that if money is a bubble, then nearly every financial asset is a bubble? This just seems plain wrong to me. Financial assets are typically backed by physical assets. For example, the banknotes issued by private banks in the U.S. free-banking era (1836-63) were not only redeemable in specie, but they constituted senior claims against the bank's physical assets in the event of bankruptcy. Mortgages are backed by real estate, etc.

Of course, there is the problem of dividing up the physical assets, but at some level, someone ends up with property rights in the physical asset--and it is this property right that gives most assets a "fundamental" value.

Note: I see that Steve Williamson has his own reply here: Money and Bubbles


  1. Good voice-of-reason post.

    I'm broadly in agreement with you, but here are a few criticisms.

    1. Here is my commentary on the issue, Would Warren Buffet Buy Green Pieces of Paper. I don't think fiat is a pure bubble.

    2. What do you think Steve means by: "The safe market rate of interest is now too low, relative to where it should, or could, be."? This is the safe asset shortage argument, right? If so, we've had this argument before, and I don't buy it. There can be no unmet demand for safe assets, their prices will immediately rise to satisfy that demand.

    3. You say that financial assets are backed by real assets. Stocks are backed by real assets, bonds are not. Upon default, bonds provide a claim to a fixed nominal amount of the collateral asset. For instance, say a bank lends $400 so a company can buy a $500 factory, and then the company defaults and the bank forecloses. If the bank can sell the factory for $600 (there's been some inflation), then it can only keep the amount it originally owned - $400. The remaining $200 goes back to the company. Lenders never have more than a fixed nominal claim on underlying property (unless bonds are indexed).

    4. The uphsot of 3. is that as long as you think pure fiat is a pure bubble, then most bonds are also a variety of pure bubble. Sure, their fundamental value is derived from the property right, but if fiat is less valuable, then so is the property right. If fiat goes to 0, so do bonds. Stocks no. Indexed bonds no.

    1. "then it can only keep the amount it originally owned - $400"

      typo. lent, not owned.

    2. JP,

      [1] I don't think the value of USD is a pure bubble either. That's not the point. The question is whether USD trade above their fundamental value. The probably do.

      [2] Yes, it's the shortage of collateral objects in the face of binding borrowing constraints. One can derive this liquidity premium formally in mathematical models. Whether it exists in reality and, if so, how large it is, is a matter of debate.

      [3] By "bonds" do you mean (non-indexed) US treasuries? These have the same fundamental value of USD. But they trade a a lower market value (i.e., they are discounted --they generally earn interest--so they have a lower liquidity premium than USD).

      [4] I'm fine with that.

    3. 2). Sounds to me like more of a relative problem. Kind of like, my collateral has risen in value, yours has fallen. I can borrow more now, you can borrow less. Net-net, what's changed?

      Have fun in Vienna.

    4. Well, not exactly. Let me try again.

      Consider two worlds, A and B. In world A, you and I exchange services with each other on a credit arrangement. No need for collateral, if can both be trusted to make good on our promises. Suppose that our collateral assets generate a flow of services that we value.

      But suppose we cannot commit to honoring our promises. In this case, the credit market might be rescued by offering our assets as collateral when we want a favor. The asset held by the creditor is held hostage, but returned once the debt is repaid (this is like repo).

      In such a world, the collateral object will have a market value greater than in world A (at least, if debt constraints are not completely slack). The intuition is simple: the asset is valued now for it ability to facilitate intertemporal exchange.

    5. Hi David, I agree with all that. I like the concept of a liquidity premium. We are grokking one another. Perhaps its the macro implications that I'm skeptical of, and applications of these ideas to something like 2008-09.

      1) Backing up a bit. You said that you don't think USD is a pure bubble. Do you agree that money has some "fundamental value" for the same reason as me? Do you agree with Steve here?

    6. JP, there is room to debate the macro implications for sure.

      Yes, I have no disagreement with what you said. In addition, I think the ability to discharge tax obligations also adds some fundamental value (as I have defined it).

      What part of Steve's post do you disagree with?

    7. "What part of Steve's post do you disagree with?"

      When I asked if you agreed with Steve I was referring to this specific comment.

      It starts: "Hi Mike, I thought you might show up to complicate things.... Is that even feasible?"

      Myself, I agree with it. I'm just canvassing you to see if this a minority opinion or other "new monetarists" have the same opinion.

    8. Yes, I agree with what Steve says here. (Don't let Steve know, though... ;) )

  2. Karl Smith has a good reply to Noah. Kind of surprised he didn't quote Rowe, since Rowe constantly makes the point that if there are N goods and barter is prohibitively expensive, we exchange (N-1) for money rather than having (N-1)*(N-1) markets of goods for goods.

  3. What is the value of physical assets if you can't use them for exchange?

    Say a currency issuing bank defaults. I am given its 10 tables, 10 chairs and 3 computers in the bankruptcy. What is the value of these to me? Assume I already have the tables, chairs and computers I need, i.e. I am neither short nor long tables, chairs & computers.

    1. Ritwik,

      While you may not personally value these objects, you may dispose of them on the market in exchange for other objects that you do value. One has no such option with an unbacked money instrument.

    2. David

      But what is its fundamental non-exchange value then? It derives its value from exchange.

      The problem is the same. In the first case I need to find someone who will accept my tables & chairs. In the second, I need to find someone who will accept my 'unbacked' money instrument. Why is the second case necessarily, or even probabilistically, more difficult?

    3. In fact, the second case is arguably a LOT easier. It is much easier to find someone who will accept my unbacked money instrument than my hard assets.

      And that precisely is the liquidity premium. And it is no bubble. It is as fundamental as the chair or table that the defaulting bank promises me.

    4. Ritwik,

      True enough. But I guess the point is that the tables and chairs are less liquid -- you will dispose of them at a discount.

      I'm not going to disagree with your last point. It just boils down to how one defines "fundamental value." I gave my definition. Perhaps you do not like it. That's fine. (Btw -- I am not calling it a bubble either, so I'm not sure if we really disagree on anything here.)

    5. Thanks David, I'll take that!

      I was going to link to the 'Evil is the root of all money' paper by Kiyotaki & Moore but read some of your archives and saw that you have already covered it recently. If that paper agrees with how you think about money, then I think we have much to agree with!

  4. Never ever talk about financial matters again at least until you get a degree in finance. NEVER!
    A) There may be a bubble in Crusoe's economy, but it needs the only subject lacking a bit of correct information and then obtaining it. Like overestimating tree productivity or whatever.
    B) Bubbles aren't matter of liquidity. Period. It's purely expectational.
    C) Banks hold relatively small amount of real assets, dealing mainly with financial ones. Inventory is worth less than third of accounting value even in liquid novadays market. Fully illiquid assets are worth only their consumer value. Read previous post.

  5. You write, "The point is that even fiat money can have exchange value, and if it does, then its value is entirely a liquidity premium or "bubble" and that, moreover, it is probably a "good" bubble to the extent that fiat money facilitates trade."

    What about the fact that "fiat money" also has a Coarsen value, which in no sense of the word is a bubble, as Delong explained very well not too long ago when he wrote:

    That is, I think, where von Mises is coming from.

    And, of course, this is wrong--so so so so so so so so so unbelievably wrong.

    It is simply not the case that we can cheaply and easily buy things with money because it is valuable. It is, instead, the case that money is valuable because we can cheaply and easily buy things with it.

    One way into the tangle of understanding why it is wrong is to ask each of us: Why are you happy accepting money in exchange when we sell useful commodities?

    Hint: It's not because we are looking forward to going down to the bank, exchanging our bank notes for the little disks of gold usually decorated with pictures of bearded men on one side and allegorical female figures on the other with lettering saying things like "Fecund Augustae" or "Concordia Militum" or "Fides Exercituum" on them, taking our little disks home, and feeling happy looking at them.

    That's not why we accept money.

    We accept money because if we don't have any money we have to buy commodities with other commodities, and when we do so we are unlikely to receive the cost of production for what we sell. Have you ever tried to buy a latte at Peets with a copy of Ludwig von Mises's Money and Credit? It does not go well.

    The fact is that your wealth is only worth its cost of production if you are liquid--if you can wait to sell until somebody willing to pay full cost of production comes along, which is not every minute. The use-value of money is that it allows you to time your other transactions so that you can realize the full exchange value of what you sell, rather than having to sell it at a discount.

  6. Second question:

    You like you coconut hypo, but this one seems more apt.

    Assume a world with no money, where "we have to buy commodities with other commodities." Can there be inflation in such a world? How?

    Can there be a bubble in such a world? How?

    I would submit that the answer to both is yes and that these realities point out that the way many people (Williamson and Friedman come to mind) talk about money is entirely mistaken.

    Money is not a cause of anything, it is more like an accelerate or perhaps like oxygen. In a frictionless world, excess money could never cause inflation, for example, because all future contracts would be hedged and the hedging premiums would soak up the excess money.

    In sum, the study of money is, a best, a study of its useful function, but answers to questions about money will not yield fundamental truths about the macro economy (the World today is awash in cash but remain in the Lesser Depression and the better argument is that it is getting worse again).

    1. This comment is beyond insane. Any time you print extra dollars, the real value of those dollars must decrease (holding fixed real factors) so that all dollars are held by someone; that is by definition inflation. Jeebus, are you people all this stupid?

    2. Anonymous

      This comment is beyond insane. Printing extra dollars only raises the nominal price. In a frictionless world of rational economic actors, the printing of extra dollars would be hedged by all economic actors and would be of no moment in real terms.

      This insight is crucial to understanding that the argument about printing money is not an economic argument but is political one over whose ox gets gored, given what we know about the actual frictions and lack of rationality of economic actors.

    3. Gak, you're just not too bright. If the nominal price level goes up, then M/p goes down. Please tell me you're not an economist.

  7. "The point is that even fiat money can have exchange value, and if it does, then its value is entirely a liquidity premium or "bubble" and that, moreover, it is probably a "good" bubble to the extent that fiat money facilitates trade."

    The Fed's money is backed by treasury bonds, which in turn are backed by taxes, and ultimately real (privately owned) assets. There's no bubble involved.

    If you want to see what a real money bubble looks like, look at bitcoins.

    Also, it's not totally clear that the Fed's money (most of it - the bank reserves) does have a liquidity premium. Why do treasury bills yield less than interest on reserves? The liquidity premium seems to be negative!

    1. Max,

      It would be more correct to say that Fed money is partially backed by the government's ability to tax. There are limits, after all, to how much a government can tax. The question is whether the value of Fed money exceeds this backing. If it does, then its value embeds a liquidity premium.

      Treasury bills may yield less than interest on reserves because not everyone has access to reserve deposit accounts?

    2. Well, it's really the public debt that needs tax backing more than the Fed's money. Suppose the public debt were zero and the Fed created money only by lending against liquid assets (so very low risk). How much tax backing do you need then? None really, just some start up capital.

  8. sebenernya saya belum begitu jelas mengenai apa yang dimaksut dalam postingan diatas
    tapi selaku sahabat blogger tetap semangat ya kawan

  9. saya suka dengan postingan mu sobat, makanya saya sering membaca ulng artikel artikel yang sudah anda buat
    terus berkarya sobar

  10. all it was a good system to be applied, depending on who implements and to whom or target market ckeck: Harga Samsung Galaxy | Harga BlackBerry Monaco