There’s only one James Tobin
Or else there are two Scott Sumners. Here’s Nick Rowe:
Not many people know this, but there were actually two James Tobins.
The first James Tobin (pdf) said that there cannot be an excess supply of money, because if anyone did have an excess supply of money he would immediately run to the bank to get rid of it.
The second James Tobin said that people hold positive average stocks of money for the same reason anyone holds positive average stocks of any inventory: because it is too costly to keep running back and forth to the bank to get rid of inventory immediately and then get it back the moment before you need it again.
Nick is forcing me to do something I thought I’d never do—defending Tobin. That’s because I also hold both views, and see no contradiction. Here’s what I believe:
1. At any given set of prices there can be an excess supply of money. That’s why there is a hot potato effect and that’s why adding more base money raises the price level.
2. However, in the short run asset prices adjust to prevent an excess supply of money. The higher asset prices make people willing to hold larger cash balances whenever the monetary base increases. Thus the money market remains in equilibrium. That’s really what Tobin is assuming when he talks about the ability to go to the bank. The banker is a person standing ready to buy base money (on demand) from the public, and supply financial assets like CDs in exchange. However very little base money actually stays in the banking system during normal times, so it’s actually asset markets more broadly that adjust to the monetary inflow.
So here’s what’s going on. The public determines how much cash they want to hold, and adjusts how often they go to ATM machines to restock. Tobin’s second point refers to the fact that the opportunity cost of holding cash affects the Cambridge k, or ratio of cash to income. Higher rates lead to a lower k ratio. But these adjustments occur almost instantly, as people restock every week or so. You make a mental note; “start going to the ATM every six days instead of every seven.” Done. Almost instantly. The lags (or “monetary disequilibrium”) before the first adjustment is less than a week, and of no macro significance. Here’s what is significant; the broader level of wages and prices take a long time to adjust. So in a sense there is macro disequilibrium, but it’s labor market disequilibrium, not money market disequilibrium. (It would be monetary disequilibrium at original asset prices.)
Now here’s where I part company with Keynesians who might have been with me so far. Although short term interest rates are one of those “asset prices” that cause the money market to achieve near instantaneous equilibrium, even as the goods and labor markets are in disequilibrium, they actually have very little role in moving NGDP and prices to the level necessary to restore long run macro equilibrium (and to move interest rates back to their original level.) In my view 60% of the heavy lifting is done by what Keynes called “confidence” and I call “expectations of NGDP growth” and Ford Motors economic forecasters call “expected nominal incomes in 2014 available to buy Ford cars.” Another 35% of the transmission is done by asset markets like stocks, forex, commodities, real estate prices, junk bond yield spreads, etc. And maybe 5% by risk-free short term rates. At most.
That’s why I call interest rates an “epiphenomenon.” I hope Nick and I can at least agree on that point.
PS. Recall that the hot potato effect is the only explanation for the long run change in NGDP. If expectations of higher NGDP growth do 60% of the heavy lifting, then the HPE provides 60% of the transmission mechanism even though the instantaneous adjustment of interest rates prevents any instantaneous HPE. That’s what Keynesians can’t get through their heads. How (expectations of) the HPE can even work at zero rates.
Tags:
29. August 2013 at 21:10
“At any given set of prices there can be an excess supply of money. That’s why there is a hot potato effect and that’s why adding more base money raises the price level.”
This seemingly trivial argument is actually flawed.
For every person who feels a hot potato in their hands, in order for there to be exchanges, there has to be someone else who wants a potato, which I guess would make it cold, or room temperature, to the individuals who receive money that is expended.
If the hot potato effect really existed, then why do individuals keep wanting to receive more money as it is “printed”? The hot potato effect really only applies when considering those individuals who want to spend more money, not those who receive more money. But even then it’s problematic, because in order for the average person to spend more and more over time, they have to receive more and more over time.
The hot potato effect presupposes a “cold oven mitt” effect. You can’t have one without the other.
Money should not be viewed as behaving as a “hot potato”, even if only as an analogy, unless the context is of hyperinflation, where almost everyone wants to get rid of money, because they have already received high quantities of money, and few if anyone wants to receive money going forward.
The reason why prices rise over time in a contect of a rise in the money supply is not because of a hot potato effect, but a combination of hot potato and cold oven mitt effects, which is to say, a lower marginal utility is attached to a given dollar due to the new, greater supply of money.
When individuals are receiving money, they would not be able to fetch a higher price, unless the current owner of money attaches a lower marginal utility to a unit of money they own. For then there would be a corresponding cold oven mitt for the hot potato.
In other words, the same logic of the “hot potato” effect, used by itself, could be used to show that the reason prices rise in a context of monetary inflation is due to a “cold oven mitt” effect. Everyone wants to receive more money when there is a greater supply of it, and that’s why prices of the things they sell rise over time in exchanges.
This comment:
“So here’s what’s going on. The public determines how much cash they want to hold, and adjusts how often they go to ATM machines to restock.”
No, INDIVIDUALS determine how much cash they want to hold. There is no “public” desire for cash. For every dollar in existence (save that which is lost) is owned by someone. That ownership means that “the desire” to hold cash is exactly 100% equal to “the supply” of cash owned.
There is no such thing as “a greater public desire for cash tha nthe cash that exists” nor is there “a lesser public desire for cash than the cash that exists”. When you believe that “the public” wants more or less cash, what you are really referring to are only a particular group of individuals who want more or less cash. There is an exactly offsetting desire for less or more cash on the part of everyone else NOT in that group of particular individuals in question.
This is important to understand because by believing there is such a thing as “a greater (or lesser) public desire for cash”, one is then likely to be incorrectly led to believe that there must exist a group of people (not in “the public”) to provide “the public” with more or less cash. How many times have monetarists claimed that “the central bank must print more money, because “the public’s” demand for cash went up.” This is confused thinking.
Individuals are the foundation of everything economics related. There is no “public”. I disagree vehemently with Dr. Sumner when it comes to money. How can our disagreement possibly be reconciled with such statements as “the public wants this or that”? This collectivist approach to understanding the world is very, VERY problematic. It ignores individual tastes.
29. August 2013 at 21:23
When monetarists claim that “the public wants more money, so give it to them!” what they are actually claiming, whether they know it or not, is that they want some particular group of individuals to receive more money than they otherwise would have received in the absence of inflation. Please note that by “particular group of individuals”, it need not be only treasury debt holders. It can be treasury debt holders PLUS those particular individuals they trade money with.
Inflation works by raising the incomes of some particular individuals before everyone else. Inflation of the money supply does not raise everyone’s cash balances at the same rate. That would be impossible. The Fed sends checks to the banks, not me. My income only goes up AFTER the banks lend more or spend more money, and even then there are typically more individuals in between whose incomes grow before mine. There are others whose incomes go up because my income finally goes up and I spend more on their goods.
So monetarists are not wanting “the public” to have more money, they actually want some individuals to have more money NOW, and everyone else to have more money LATER ON, once the initial group of individuals spend out of their greater incomes.
In terms of real wealth, of goods, this implies a restribution of wealth. As more and more money is exchanged outward in an uneven wave-like manner, from the initial injection points, purchasing power and real wealth is exchanged in the exact opposite direction.
Monetarists do not want an even increase in money and wealth. They want a relative increase in money and wealth for some individuals and a relative decrease in money and wealth for all other individuals.
Here is the result:
http://i.imgur.com/cZaaXqz.jpg
You can thank unconstrained money printing capabilities. What the progressives wanted.
Progressives and conservatives are of the same cloth. They just differ in how they want mommy and daddy government to control everyone.
29. August 2013 at 22:21
Yeah, I tend to agree with you over Nick about this, although I think Nick is trying to make a somewhat different point…
29. August 2013 at 22:50
[…] Scott Sumner — There’s only one James Tobin […]
30. August 2013 at 01:01
Scott: “Thus the money market remains in equilibrium.”
Let’s start by cleaning up your vocabulary;-) Because sometimes a bad use of words can stop you seeing things.
Let there be n goods, including money. That means there are n-1 markets. In each of those markets money is traded against one of the other n-1 goods. There are n-1 money markets. You cannot talk about *the* money market. You can say that the apple market is in equilibrium, or the banana market is in equilibrium, or the carrot market is in equilibrium…but when you say “*the* money market is in equilibrium” what do you mean by that?
If apples are in excess supply, and bananas in excess demand, and the carrot market clears: that means there is an excess demand for money in the apple market, an excess supply of money in the banana market, and neither excess demand or excess supply of money in the carrot market.
If the bank redeems its money for gold, you presumably mean the gold market is in equilibrium. If the bank redeems its money for apples, you presumably mean the apple market is in equilibrium. If the price of bonds is perfectly flexible, you presumably mean the bond market is in equilibrium.
30. August 2013 at 01:19
Now let’s assume there are transactions costs, or at least lags, in going to the bond market. (As the second Tobin argued, we have to assume there are those transactions costs, otherwise people would never hold strictly positive inventories of money, because if they held a positive stock of money they would instantly sell it in the bond market, and then sell bonds the instant before they needed money to spend.)
Now suppose I find I have more money than I planned to have (say I totally unexpectedly find a $10 bill lying on the sidewalk, and it’s a good one!) What happens next?
First off, the answer may be “nothing for a while”, because the whole point of holding a buffer-stock/inventory is that it lets you handle unexpected events like that.
But if you do decide to get rid of it now, you might find the transactions costs of getting rid of it in the apple market are less than getting rid of it in the bond market. Because you were going to the apple market anyway, but would need to make a special trip to the bond market (or bank). Or maybe the apple market is closer to your house.
30. August 2013 at 01:30
Nick beat me to this comment, but I will do it anyway. It really is about this post of yours: “The higher asset prices make people willing to hold larger cash balances whenever the monetary base increases. Thus the money market remains in equilibrium.”
Why do you assume that “asset markets” are by default always in equilibrium? If there is an excess supply of money why should people go and spend it ALL on “assets” like bonds? What prevents them to invest the money on the equally good investment opportunities, such as expanding the business, insulating the house, buying a new car with better fuel efficiency etc.?
This is what I think Nick wants to say. Everything is money market. Excess supply of money will therefore have impact on everything, on asset prices but also (via hot-potato) on all goods.
30. August 2013 at 02:35
The Economist on who should lead the Fed: http://www.economist.com/news/leaders/21584340-larry-summers-and-janet-yellen-would-both-lead-fed-well-ms-yellen-safer?fsrc=scn/tw_ec/choosing_the_chairman
30. August 2013 at 04:22
I agree with the rest of Nick’s comment, but this seems dubious. Asset prices have to change before planned expenditures on real goods increases; extra money balances won’t make people suddenly indulge in unplanned real purchases, realistically.
30. August 2013 at 04:23
Where is dtoh when you need him?
30. August 2013 at 04:41
Saturos: “Asset prices have to change before planned expenditures on real goods increases; extra money balances won’t make people suddenly indulge in unplanned real purchases, realistically.”
And why is that? You deem it as realistic that when somebody unexpectedly finds $500 in a pocket of his winter coat that he will immediately run to some “asset” market with a plan to purchase bonds or pay-off the loan as opposed to running to nearest store with a plan to purchase something nice for the household?
30. August 2013 at 04:56
Scott,
Since you bring up the HPE at the zero bound again, this feels like a good time to pick up where we left off before you went on travels. I think we were really getting somewhere with the Apple Certificate of Deposits analogy.
Your last comment is here and I responded the following:
Scott,
“Now I see your mistake.”
Well, mine and Woodford’s and a whole bunch and other guys’, right? 🙂
“Cash is not like apple CDs, it’s like apples!”
Why is it like Apples? The central bank creates cash by exchanging it either against a repo or a short-term Treasury bond (in normal times). Both of these are “safe” assets who’s price is pretty 1:1 in terms of cash at all times (like Apples vs. Apple CDs).
Cash is functionally a fungible “certificate of deposit of safe nominal assets” much like the Apple CD is a “certificate of deposit of Apples”.
Bonds and Apples don’t flow through payment systems easily. Cash and Apple CDs do.
Cash can be redeemed into said nominal assets at any time. Generally that’s done in the market though functionally, it’s as if it was done at the Fed since the Fed is there in the market making sure those transactions happen at the interest rate that it has set.
“And even you agree that more apples will reduce the value of apples.”
Of course.
“Cash can no longer be redeemed into gold.”
Correct, cash can be redeemed into interest bearing nominal assets. When the Fed redeems cash, it doesn’t subtract from the net supply of nominal assets, it just swaps one for another. That’s why the price level is unaffected unless the expected real return of nominal assets is moved in the process (nominal rates minus expected inflation)
“Your 20 year remark (actually 15) is misleading, as Japan’s been at the zero bound that entire time.”
The price of Apple drops when the supply of Apple increases (ceteris paribus), even at the zero bound.. The fact that you have to qualify your quantity argument based on where nominal interest rates are is, I think, a hint that interest rates aren’t as irrelevant as the price of zinc (to use an example you had given me before).
“Monetary injections are only inflationary if they are permanent. And for all intents and purposes “permanent” means “until nominal rates rise above zero.””
I agree with that but I think it’s a convoluted way of looking at this. And I think it is strictly better for the Fed to say it will inject excess reserves when inflation picks up (or in their language, keep interest rates at 0%) than to inject reserves today which could be drained away instantly (like the punch bowl when the party gets started).
Do you not expect the Fed to drain reserves once inflation picks up? That would mean they would let the price level triple relatively rapidly if I understand your quantity-based model correctly?
30. August 2013 at 05:58
Scott, I believe that you are mistaken in holding that “higher asset prices make people willing to hold larger cash balances whenever the monetary base increases,” and in claiming that the money market is thereby kept “in equilibrium.” Admittedly there make s a sense in which equilibrium can be said to prevail. But it is a flow rather than a stock (real balance) equilibrium, with asset prices adjusting only enough to make it worthwhile for persons to borrow, but not to hold, more than before. The “hot potato” balances thus keep moving. But there is no stock equilibrium. Indeed, if the money market truly were in stock equilibrium there would be no eventual tendency for P (or at least factor prices) to eventually adjust, the quantity theory would have no merit at all, and instead those old-school Keynesians who treat the interest rate as the “price of money would be right. (Simple appeal to Walras’ Law, by the way, makes your claim that the money and asset markets can both be in equilibrium, while the goods and factor markets remains in disequilibrium–with either an excess supply of or an excess demand for both–problematic.) It follows that your claims that the “lag” during which monetary disequilibrium persists is very short is very misleading.
I believe that the same holds for your treatment of (not so) short-run asset price changes as “epiphenomena.” The Keynesians may be wrong in supposing that “hot potato” real balances can only be passed on through asset markets rather than being channeled directly into goods and labor markets via real-balance effects, themselves condition by expectations. But those asset price changes, and corresponding changes in real interest rates, which (as Wicksell taught us) last as long as it takes for monetary stock equilibrium to prevail, represent potentially important temporary relative price changes that presumably influence real spending, including investment, patters.
In short, while we both agree that the Keynesians err in treating asst-price adjustments as the whole story when it comes to explaining the consequences of disequilibrating monetary base increases, I think the monetarists go too far in playing down the significance of asset price movements–a significance that is quite distinct from what Keynesian liquidity-preference theory suggests, because it stems precisely from the fact that the movements in question tend to be self-reversing rather than sustainable. Wicksell, in my opinion, had made a plumb job of this before the Keynesians and then the (old) monetarists began battling to swing things towards one or the other incorrect extreme. And I for one would very much like to see you and other Market Monetarists avoid the mistake of their predecessors, who seem to have had a blind spot when it came to appreciating the lessons taught by the great Swedish economists. Perhaps Lars Christensen is just the fellow to lead the way!
30. August 2013 at 06:20
Nick, Good question. Here’s what I mean when I say the money market is in equilibrium:
1. If you walk around NYC and ask people if they are holding the amount of apartments they prefer to hold, they will say “No.” At going rents, they’d like to hold more apartments. Because of rent controls the apartment market is in disequilibrium. Ditto the jobs market in a recession, at going wage rates. If you ask people if they are holding the amount of cash they prefer to hold, they will say “Yes.” But as I mentioned in this post, there is one exception—the ultra short run. People tend to go to the ATM about once a week, and hence after a helicopter drop they will hold excess cash until their next trip to the ATM. Yes, that’s monetary disequilibrium, and yes, they might find it more convenient to spend that extra cash on apples than bonds. But I argue that monetary disequilibrium lasts for only a few days, and is of no macro significance. (I think that’s what Tobin must have meant. Even more true when you consider that real world monetary injections start in the banks, via OMOs.) Rather the macro significance from monetary injections comes elsewhere. They raise expected NGDP growth and they raise a wide range of asset prices. Both factors tend to boost current AD, and current NGDP.
JV. Yes, to a very small extent they do the things you say. But since asset prices are 100% flexible, those are the markets where most of the instantaneous adjustment occurs. But look, I also believe goods markets respond, however I tend to think they respond more to the asset prices and expected NGDP, rather than the new money itself. This gets back to my critique of Cantillon effects. For me an OMO is a roughly equal swap, and the seller of bonds to the Fed is not really richer, if the bonds are sold at market prices. Maybe a tiny bit richer, as price rises a tad. But if the OMOs make stocks soar, as in Japan, then the newly rich stockholders will go out and buy the new car EVEN IF THEY WERE NOT THE ONES WHO GOT THE NEW MONEY.
Saturos: You said;
“I agree with the rest of Nick’s comment, but this seems dubious. Asset prices have to change before planned expenditures on real goods increases; extra money balances won’t make people suddenly indulge in unplanned real purchases, realistically.”
Yup, that’s exactly my point, and pretty much my only point. And yes, where’s dtoh when I need him?
JV, In your reply to Saturos you consider a helicopter drop, which is also fiscal stimulus. With OMOs you are basically no richer–from direct effects.
DOB, Cash is like apples because it is a medium of account and other financial assets are not. If there is another financial asset that is identical, it is essentially cash, and should be counted as part of the base. Thus I agree that a swap of $20 bills for $100 bills does nothing, as they are perfect substitutes. But T-bills are not a MOA, as their price can change, even if only a tiny amount. It’s enough to make it so people don’t shop with T-bills. Thus they are not apples, they are pears, a close substitute.
The zero bound matters because cash is an infinitely lived asset, that must be modeled like capital. The value today depends strongly on the expected value in 20 years. That’s one difference from apples–so I should have compared cash to gold, not apples. The demand for gold rises at low rates too.
Let me answer your final question by considering a case where velocity is expected to return to normal in 3 years, when interest rates return to normal (say 4% T-bond yields). Then my claim is as follows:
1. If the monetary injection is expected to be temporary, then NGDP will be expected to show zero growth in 3 years.
2. If enough of the injection is expected to be permanent enough in 3 years to raise the base by 15%, then the expected level of NGDP three years out will be 15% higher.
3. If the entire monetary injection is expected to be permanent, then NGDP is expected to triple in 3 years.
If the Fed want’s to set a 5% NGDP target path (i.e 15% over three years) then it essentially opts for option 2. Now I admit that markets don’t fixate on the base, for very goods reasons—velocity is not constant. Thus the markets think in terms of NGDP growth paths. They think in terms of the central bank leaving enough base money around when rates normalize that NGDP will be 15% higher than 3 years earlier, given the actual change in V. So when the Fed promises a 5% per year growth path for NGDP, it is promising that in the long run, when rates normalize, it will make sure that velocity adjusted base money grows by 5% per year. Thus if V fall by 2% per year, the base will grow by 7%. That’s what I mean by permanent changes in the base.
You are correct on one point. When I said interest rates are not important, I was thinking in terms of the transmission mechanism, where I think other factors dominate. But interest rates have much more of an impact on the demand for money and V than the price of zinc, so in that sense they are important.
30. August 2013 at 06:40
George, Regarding your first paragraph, I claim that even the stock supply and demand for money is in equilibrium once asset prices adjust, as nominal rates are the opportunity cost of holding cash.
So why do prices later adjust? Because at the new level of asset prices, and the new level of expected NGDP growth, the labor and sticky goods markets are now out of equilibrium. I suspect that some of the difference here may be semantic, but am not sure. Perhaps you call that situation “monetary disequilibrium.”
In your second paragraph you say:
“But those asset price changes, and corresponding changes in real interest rates, which (as Wicksell taught us) last as long as it takes for monetary stock equilibrium to prevail, represent potentially important temporary relative price changes that presumably influence real spending, including investment, patters.”
I think you misread my post, as I entirely agree! I claimed that movements in short term risk free interest rates are not very important, but I think asset prices are very important. Indeed some asset prices (commodities, real estate, etc) are actually real goods. They are not sticky price goods, which I put in another category. The 1933 dollar devaluation is a good example of an action that massively impacted asset prices, but didn’t move short term risk free interest rates. Now Keynesians will argue with me that I’m still talking about “interest rates,” just in broader terms. But the proof’s in the pudding. Keynesians worry about the zero bound. Since most asset markets have no zero bound, I don’t worry about that problem.
I agree with Austrians, for instance, that bad monetary policy produces wild swings in assets prices, and that these movements may tend to destabilize the business cycle. But I also believe those swings must be modeled in a way that is consistent with the EMH.
30. August 2013 at 07:15
“Higher rates lead to a lower k ratio”
Higher inflation leads to higher levels of currency used outside the banks – but smaller levels of non-interest bearing bank deposits.
30. August 2013 at 07:17
Off topic, but a Japanese commenter on Historinhas links to this superb speech by the Deputy Governor of the Bank of Japan
https://www.boj.or.jp/en/announcements/press/koen_2013/ko130828a.htm/
30. August 2013 at 07:33
Some observations from the King of Bonds Jeff Gundlach:
– The Fed will be reducing their bond purchases and we can expect some language to that effect.
– Nobody is making money anywhere in the market right now, stocks/bonds/gold are all doing poorly and that is because investors see Central Banks tightening up liquidity.
– If there was one thing he would buy today it would actually be long-term government bonds because there is no inflation anywhere.
– He expects gold to continue hitting new lows.
http://www.gurufocus.com/news/227310/bond-king-jeff-gundlach–he-cant-find-anything-to-buy
30. August 2013 at 07:46
Scott: All right, you post (and re-read) helped me to understand it better. So the key part of the post is that excess supply of money will be reflected in various asset markets like stocks, real estate, forex etc.
So if you are defending Tobin, where is the reflux? So I thinkt that you are more in agreement with Nick on this. Excess supply of money is important in its own way because it applies to large variety of things besides short term interest rate.
Anyways I still think that you underestimate that people can get rid of money just by buying stuff.
1) First about helicopter money vs OMO money. I do not think there is a difference in one important detail – how this money will affect the price level (hot potato and all that). And I actually know that Nick also thinks the same, for instance here: http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/09/all-money-is-helicopter-money.html
Imagine that instead of finding $500 in your pocket there comes someone who wants to buy something from you so that you end up with $500 more than you expected. That somebody can buy some of your gold or other asset, he may even bought some of your labor. The point is that you have unplanned income of $500 more.
What will you do with it? Buy something else. But crucially by doing that you passed this hot potato on to someone else. You as an individual do not expect that this hot potato will ever return to you, until next month where you again end up with $500 more income than you expected. Nick Rowe described it more succinctly in this post: http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/08/what-steve-keen-is-maybe-trying-to-say.html
2) Second, in a way you agree with Nick more than I thought. Even if excess supply would be first expressed solely via excess demand for broadly defined “assets” we do not know where to look. Should we look for price of bonds, gold, stocks or real estate to see this excess demand?
And even if some of these asset prices are flexible, why should it be advantageous for us spotting excess demand? Should it not be exactly the opposite? For instance if price of house insulation is sticky (mabye insulation producers were not visited by Calvo fairy when excess money supply happened?) – should this not cause runs on Home Depots for house insulation? If price of house insulation cannot go up while expected inflation goes up this should be tremendous opportunity for anybody to take whole it lasts.
30. August 2013 at 08:31
Scott, although I admit, as I indicated previously, that in a sense the issue can be regarded as “semantic,” in that one can always characterize any macroeconomic state as an “equilibrium,” just as one might say that a man, having jumped from the top of the Empire State building, is in “equilibrium” as he rushes past the 30th floor, I still think your treatment of the presence case problematic in light of Walras’ Law. Thus, suppose a collapse of M, like that of the early 30s. NGDP shrinks, and both goods and labor are in excess supply–a classic general glut. Unlike general prices, and the price of labor especially, asset prices are perfectly flexible. Now if one takes this flexibility to imply that both those asset markets themselves and the market for money balances are quickly restored to equilibrium, one begs the question: with both goods and labor in excess supply, in which market is there a corresponding excess demand? Unless we are prepared to define away Walras’ law, the question is of course not merely a matter of semantics.
But to return to semantics, is it really helpful to resist speaking of a state of general recession as involving a shortage of money–that is, to resist the insight, which has been around at least since G. P. Scrope used it as a chapter header in his 1833 Treatise, that “a general glut of goods presupposes a general want of money”? I mean, here you are, having spent most of the last few years making the case that monetary expansion is what’s been needed to combat the recession, and endorsing Friedman’s similar claim concerning what was needed in the early 30s, yet insisting that there has not been any shortage of money except one that lasted perhaps a week or so after the initial collapse of spending? Semantically this seems to me both counter-intuitive and counterproductive. Surely it would be easier (and, I believe, also analytically quite proper) to say that both the recent downturn and the Great Depression involved a shortage of money balances, and that it is precisely owing to the fact that money was in short supply that creating more of it constituted an effective remedy? Your approach, in contrast, means having to say that more money was needed even though,/i> there was no shortage of money; that exchange activity was frustrated, by a lack of spending but not by any lack of money balances, and so on. Surely, even if this were only a matter of semantics (and I have already explained why I believe it is more than that), the more sensible semantic option would be to speak forthrightly of a shortage of real balances, of the necessary delay involved in waiting for it to be solves by means of deflation, and of the consequent advantage of creating more nominal money balances instead.
30. August 2013 at 08:53
I don’t care about this particular dispute, but I just pray that there aren’t two Scott Sumners.
30. August 2013 at 09:14
George: Very well put. When reading your comment I recalled a thought that was implanted by Nick Rowe some time ago: “Peanut Theory of Recession” http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/07/the-peanut-theory-of-recessions.html
PS: Sorry for spamming Nick’s articles, but reading him for past few years if there is a topic that always comes to the fore in one way or another it is the knot of topics that are in the center of this Scott’s post. So since Nick has already spend considerable time thinking things like Law of Reflux through and through it is well worth it going there and reading what he has to say. That plus I have yet to find somebody that can beat the power of Nick’s style of explaining his thoughts using little models [OK, Krugman back in the day come close with things like babysitting co-op]
30. August 2013 at 09:48
James, I agree, Marcus sent me that earlier.
JV, You said;
“First about helicopter money vs OMO money. I do not think there is a difference in one important detail – how this money will affect the price level (hot potato and all that).”
I agree, because I believe that monetary injections raise prices regardless of who gets the money.
You said;
“That somebody can buy some of your gold or other asset, he may even bought some of your labor. The point is that you have unplanned income of $500 more.”
No, this isn’t “income,” it’s just a swap of one asset for another. (Unless labor, then it is income. But that’s not how money is injected. It’s an equal swap.)
You said;
“Anyways I still think that you underestimate that people can get rid of money just by buying stuff.”
No, I fully understand that.
Regarding insulation, economists generally assume it is very costly to store sticky-priced goods. If it wasn’t, arbitrageurs would eliminate the sticky prices.
George, That’s a very good question, and maybe you can help me answer it. Let me start with an analogy. Start with an economy where everything is in equilibrium. Now have NYC impose rent controls. So there is an excess demand for apartments. But all other markets remain in equilibrium. That’s my way of looking at things. Now you point out that Walras law implies that if the apartment market has excess demand, then other markets (in aggregate) must have excess supply. So maybe I am wrong in suggesting that excess demand for apartments does not imply excess supply of (say) toasters. Maybe it does.
Taking my analogy one step further, let’s suppose in the apartment case we could agree that an excess demand for apartments is best thought of as creating an excess supply of all other goods. But also that individual goods remain in equilibrium relative to other individual goods. So there’s excess supply of AOGs but no specific excess supply of toasters. If so, then I am claiming that there is an excess demand for all flexible price goods when there is an excess supply of sticky price goods. But no particular excess demand for money, at least vis-a-vis other flexible price assets.
So I’m not sure whether my use of terms is non-standard.
I.e. in a rent control case a microeconomist might say “the toaster market is still in equilibrium, because there are no price controls on toasters”, whereas a macroeconomist might say “because of excess demand for apartments, there is excess supply of AOGs, and toasters are one of those goods.” (Toasters being like money in my example) Does that get a the distinction?
JV, I agree that Nick is the best.
30. August 2013 at 10:16
The way I see it, Scott vs George/Nick/JV are saying roughly the same thing but language is getting in the way.
When a central bank eases, it reduces the return on money relative to the both the pecuniary and non-pecuniary returns provided by all other goods and assets in the economy. Asset prices will quickly rise until they no longer provide a superior return relative to money. On the margin, agents will once again be indifferent between assets and money.
But goods prices have not yet adjusted and therefore still provide superior returns relative to both money & flex-price assets. So while agents are indifferent along the asset:money margin, they are not indifferent on the asset:good or money:good margin.
In the longer run, goods prices will rise enough such that indifference along all margins is restored.
30. August 2013 at 10:23
Scott, I see that you have forced me into a corner with your rent control argument, from which the only escape is to appeal to ol’ St. Nick yet again, who has covered the matter, and whose arguments force me to concede the drawback of my having made appeal to Walras’ Law rather than (following Clower and Yeager also) to money’s unique role in a monetary (that is, non-Walrasian) economy. Nick’s posts on the topic are below:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/04/says-law-and-monetary-policy.html
http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/04/walras-law-vs-monetary-disequilibrium-theory.html
In a monetary economy, as opposed to a genuinely “Walrasian” one, excess supplies of goods necessarily translate into excess demand for money–the medium of exchange–and not for other stuff. What about rent control? Well, I see that Nick addresses that on as well, with you taking part:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/09/can-rent-controls-cause-a-recession-a-response-to-scott-sumner-and-bill-woolsey.html
In these posts Nick makes me regret having tried to dispense what Clower and Yeager taught me, so as to make what seemed a simpler case, and thereby step smack-dab into your very well-oiled rent-control and toasters trap!
30. August 2013 at 10:46
Saturos: “Asset prices have to change before planned expenditures on real goods increases; extra money balances won’t make people suddenly indulge in unplanned real purchases, realistically.”
If I have too much money, why should I be more likely to “invest” in buying a bond rather than a bicycle or BMW or BComm?
Wow! This comment thread is really rolling! I now see JV and George Selgin are making my points, probably better than i could.
I’m going to stop writing and start reading more and thinking.
30. August 2013 at 13:05
Its something I can’t get through my head either, can you link to a previous post you’ve done on it or write one if you’ve not done so? It’d really help me get what you’re trying to say.
30. August 2013 at 14:31
JP, Exactly my view, and very well said. I hope others will look at your comment.
And I agree we are probably close, but describing a similar picture using different terminology.
George, I find this stuff really hard to think about, so I’m going to do a few examples, some of which support my view and some support your and Nick’s view. I’m hoping people will look at these examples and tell me how I am right or wrong.
30. August 2013 at 16:34
Let me try to say this succinctly (ignoring ER).
1. Economic players (organizations, businesses, individuals) hold the amount of money they need for transactions.
2. People don’t spend more because they have more money. (In Scott’s words, “Bill Gates doesn’t go out and buy a Ferrari because he has more money.” ) It’s the other way around. If people decide to spend more they acquire and hold more money in order to effect the spending.
3. Interest and inflation rates have some impact on the amount of money people hold (for any given level of spending) but it is insignificant when interest and inflation rates are near zero.
4. The Fed can not do OMP (again ignore ER) unless there is an ultimate counter-party who wants to hold more money.
5. The reason the counter-party to OMP enters into the trade with the Fed is because they have been induced to exchange financial assets NOT for MONEY but for real goods and services. (They merely need the money to effect the increased spending on the goods and services).
6. The first reason the counter-party has been induced to exchange financial assets for real goods and services is because real prices of financial assets have risen relative to the prices of goods and services.
7. The second reason is that expectations of higher NGDP cause a shift in the indifference curve between financial assets and real goods and services.
8. The marginal increase in the exchange of financial assets for goods and services caused by OMP is an increase in AD/PY/NGDP.
9. There is a multiplier effect. If a business buys a new machine, the machine manufacturer has higher income and goes out and hires more employees, etc., etc.
10. The increase in NGDP eventually absorbs the increased money generated by the OMP so you have a very close correlation between M and NGDP, but it is not causal. INCREASED M DOES NOT CAUSE NGDP TO GO UP. The mechanism is a) higher real prices of financial assets and b) expectations.
30. August 2013 at 18:07
[…] the recent post I did defending Tobin led to a comment section discussion with Nick Rowe, George Selgin, Saturos, […]
31. August 2013 at 06:24
youko, See my newest post. (MOA vs. MOE)
dtoh, 1. In aggregate the public holds exactly as much money (nominal) as the Fed tells them to hold, they have no choice.
3. It is still significant at zero rates.
4. The Fed can always find a counterparty at the market price.
9. No multiplier effect–monetary offset.
10. It is causal. Indeed, there is no other way to explain why an 8700% increase in the MB causes an 8700% increase in NGDP.
31. August 2013 at 09:51
[…] a recent post, Scott whimsically wondered whether there were two of himself. But actually, there are, in the […]
31. August 2013 at 13:38
Scott,
1. In aggregate the public holds exactly as much money (nominal) as the Fed tells them to hold, they have no choice.
Absolutely not true. The public holds exactly as much money as the Fed can induce them to hold. Under the current system, there is no way to increase the money supply unless there is a willing counter-party.
3. It is still significant at zero rates.
We’ve debated this before. We’ll have to agree to disagree until I have more time to find the data I once saw on this subject. All I can say is that if you look at the cash balances held by businesses and individuals, common sense will tell you they are not going to be significantly affected by a small change in inflation or interest rates.
4. The Fed can always find a counterparty at the market price.
Exactly my point, they find a counterparty by agreeing to the terms sought by the counterparty.
9. No multiplier effect-monetary offset.
You’re thinking fiscal multiplier. I’m talking about the multiplier effect of a business or individual spending more on goods and services because the Fed has pushed asset prices up or has altered expectations. If Fed action causes me to buy a new car, there will a be a multiplier effect (unless a dealer just fills my order out of inventory and does nothing to replenish inventory).
10. It is causal. Indeed, there is no other way to explain why an 8700% increase in the MB causes an 8700% increase in NGDP.
Scott, re-read your response. It’s a totally circular statement. You really need to stop and spend an hour seriously thinking about this. This is the one point, where the “serious” people who agree with you on most everything else disagree.
Nobody disagrees that MB is closely correlated with NGDP. It’s a tautology…..MV=PY.
But…OMP is not simply the issuance of money….it’s an exchange of money for financial assets….like yin and yang. It’s a just as true to say that OMP results in a marginal decrease in net asset holdings by the public (non-financial sector) as it to say it results in an increase in money held by the public.
The public doesn’t do the exchange because they want to hold more money. They do it because they want to spend more and they need the money to effect the spending. The causal mechanism is the marginally increased exchange of assets for goods and services. The money simply facilitates the exchange. Higher real asset prices and higher expected NGDP is what causes the marginally increased exchange of financial assets for real goods and services.
31. August 2013 at 14:34
dtoh,
“Nobody disagrees that MB is closely correlated with NGDP. It’s a tautology…..MV=PY.”
If V was sufficiently variable, the MB wouldn’t be correlated with NGDP. MV=PY is a tautology, but that M is correlated with PY is an empirical claim. A tautology is true as a matter of logic, whereas an empirical claim is something we can know only by experience.
31. August 2013 at 14:45
Peden,
Yes, but…
1. If we subtract ER from MB, V is not that variable for the interest and inflation rates extant in the U.S. over the last 10 years.
2. I’m not disputing that it’s also empirically observed.
3. What I am disputing with Scott is that correlation equals causality.
31. August 2013 at 15:27
Scott you write:
“In my view 60% of the heavy lifting is done by what Keynes called “confidence” and I call “expectations of NGDP growth””
Hmmm, my calculations give me 43% for this. I used the IJPIOOMA method. I probably made an error somewhere! Would you mind providing a reference for your calculations on this? I know there’s not enough room here in the blog, but I’d love to pour over your detailed calculations and see where I went wrong. Thanks!
31. August 2013 at 17:54
W. Peden, dtoh, re: tautologies:
meet Scott Sumner:
“Except that MV = PY is a tautology,…” – S. Sumner, 8/30/’13
http://www.themoneyillusion.com/?p=23186
Yes, I’m being a dick today. 😀
31. August 2013 at 17:58
… also meet Wikipedia (where all the great economic minds get their information):
“This equation is a rearrangement of the definition of velocity: V = PQ / M. As such, without the introduction of any assumptions, it is a tautology.”
http://en.wikipedia.org/wiki/Equation_of_exchange
31. August 2013 at 18:04
W. Peden, My apologies… my attempt at being a dick backfired and instead just made me look like a dope. Now that I re-read what you wrote I SEE what your were complaining about… duh!
MV = PY may be a tautology, but the statement:
“M is proportional to PY”
is not. Correct? Duly shamed. I’ll shut up now. )c:
1. September 2013 at 08:07
dtoh, Even currency velocity has fallen sharply in recent years.
Tom Brown, I think you missed the joke.
1. September 2013 at 08:13
dtoh, 1. Wrong. The Fed simply buys bonds at the market price. The seller of the bond may not want the money, but he knows he can put it in the bank.
4. I think we are talking past each other here.
9. If there is no G multiplier effect, there is no I multiplier effect.
You said;
“The public doesn’t do the exchange because they want to hold more money. They do it because they want to spend more and they need the money to effect the spending. ”
I disagree. They do it because they want to sell a bond to the Fed, not because they want to spend more. But in aggregate they are forced to spend more because of the HPE. It’s all about the fallacy of composition, what’s happening at the individual level is completely different from what’s happening at the aggregate level. In aggregate the Fed is forcing the public to spend more (nominally.)
2. September 2013 at 01:28
JP Konig: Yes I perfectly agree with your post. I will therefore brake it up thusly
1) Change in money supply makes flexible prices (such as financial assets) to adjust and inflexible prices to adjust later.
2) What is the mechanism of adjusting these prices? Demand. Money injection means people want to get rid of it at existing prices (hot potato effect).
3) One cannot look at asset markets, observe that they are in equilibrium and say that therefore there cannot be excess supply of money (Tobin) in the same way one cannot look at flexible prices (like Gas or even beteer – Nick’s peanuts – because they do not matter for the economy by assumption), observe that they are in equilibrium and say that all is good and well because if people really wanted to change the ammount of money they hold they could just buy/sell Gas (peanuts, bonds or whatever).