What is money? What is inflation?
The answers to these questions seem obvious to me, but I can tell from the recent comments that most people don’t agree. Saturos and Bill Woolsey have argued (in the comments) that money is the medium of exchange. I argue that money is the medium of account. What makes this issue so tricky is that money is almost always both, and the textbooks define it as having both characteristics. But which criterion is the “essence” of money?
Money is also that thing we put in monetary models of the price level and the business cycle. That begs raises the question of whether the price level is determined by shocks to the medium of exchange, or shocks to the medium of account. Once we answer that question, the business cycle problem will also be solved, as we all agree that unanticipated price level shocks can trigger business cycles.
So now we have to figure out what we mean by the term ‘inflation’. I’d like to propose 5 criteria:
1. Inflation is a general rise in the sticker price of goods.
2. Unanticipated inflation helps borrowers and hurts lenders.
3. When wages are sticky, deflation causes unemployment.
4. Inflation reduces the value of the medium of account.
5. Inflation reduces the value of the medium of exchange.
When the medium of exchange and the medium of account are identical, then all five of these statements are true. If they are split, statement 5 is true for the medium of exchange, and statements 1 through 4 are true for the medium of account. Let’s take an example to illustrate this confusing issue:
Imagine Zimbabwe uses gold as the medium of account. Then they have budget problems because their economy crashes when the government tries to take too much wealth from the top 1%. So they decide to print money. But the president (who is a madman) tells his treasury minister that he wants to stay on the gold standard, and will not tolerate any inflation. If the treasury minister violates this demand, he and his family will be executed. What’s a treasury minister to do?
Easy. Tell the public that they must continue to set all wages and prices and debt contracts in terms of grams of gold. At the cash register there will be a sheet of paper listing today’s exchange rate between paper Zimbabwe dollars and grams of gold. People will pay their bills with Zimbabwe dollars. (I’ve seen something similar in Canadian border towns, with US dollars as the MOE.)
In this example, gold is the medium off account and Zimbabwe dollar is the medium of exchange. What is the “true” rate of inflation? In terms of the definitions above, the criteria 1 through 4 apply to inflation in gold terms, and criterion 5 applies to inflation in terms of Zimbabwe dollars. I would add that the average person would view inflation in gold terms. Imagine you are a Zimbabwe diamond miner who works on a wage contract promising you 10 grams of gold per month. You could care less what the prices are in Zimbabwe dollars, you have a fixed wage in gold terms, and when you go shopping you will see price stickers on the goods in gold terms, not Z$. That’s your “cost of living.” When prices in terms of gold rise then you are worse off. Zimbabwe dollars are just a medium of exchange.
In any model of the price level you are essentially modeling the real value of the medium of account, and not (necessarily) the medium of exchange. And I’d add that the same is true of business cycles. Saturos and Bill Woolsey (and I think Nick Rowe) believe that recessions occur when there is a shortage of the medium of exchange. I disagree. I never have any problem getting cash, even during recession years. A lack of cash never causes me to spend less. A recession is caused by a decrease in supply or an increase in demand for the medium of account. Thus suppose that Zimbabwe was on the gold standard at the same time that soaring gold demand in Asia was pushing its value up, relative to other goods and services. Zimbabwe would experience deflation. If the wages of Zimbabwe gold miners were sticky in gold terms, then the diamond mine would have to lay some of them off. They would be unemployed, and would start buying fewer goods and services. Indeed in some cases market signals (such as higher interest rates) will encourage people to buy less even before the workers are laid off (the effect before the cause). But the fundamental problem is that there is too little NGDP, given the current (sticky) level of nominal hourly wages. The medium of account is becoming too valuable.
In the comments section I get the impression that people believe recessions occur because there is less media of exchange to spend. That’s not a useful way of thinking about the problem, for several reasons. Start with a gold standard, and think about the effect of worldwide deflation on a small economy like Canada. If Canada wants to maintain a fixed peg between gold and the Can$, they might have to reduce their currency stock. Is the lower currency stock “causing” the drop in spending? Obviously not. The global increase in gold demand that caused the global deflation in gold terms is the fundamental problem, and Canada’s declining currency stock is a symptom, just as the decline in lots of other nominal variables in Canada are symptoms.
In many cases the stock of currency does not decline during deflation. For instance, it actually rose in the US during the early 1930s. That’s because the negative effects of lower prices are more than offset by the increased demand for currency caused by the lower nominal interest rates (which are the opportunity cost of holding currency.) When this occurs it is especially easy to see why a lack of media of exchange is not a problem. During 1932 Americans held much more currency than in 1929. The lack of spending wasn’t because they had less currency in their wallets and piggy banks; they had plenty. Rather they had a higher demand for money, and simply chose not to spend it as often.
What makes this confusing is that today currency is the medium of account. Thus an increase in the demand for currency that causes deflation and depression would be the fundamental problem, not merely a symptom of a deeper problem in the ultimate medium of account (as in the early 1930s). And because these two roles (medium of account and medium of exchange) are now unified, both sides can look at the same set of facts and reach similar conclusions. It’s only when we separate out the roles of medium of account and medium of exchange that we can clearly see the real essence of “money.”
What’s the role of the financial system in all this? Financial systems are recent inventions; money and monetary policy and inflation have been around for millenia. Yes, financial shocks can impact currency demand, but at a fundamental level the financial system is nothing special, just one more factor (like drug dealing) that impacts the demand for currency. I don’t care if currency is only 1% of all financial assets. Give me control of the stock of currency, and can drive the nominal economy and also impact the business cycle. Cut the value of currency by 90% via OMOs, and you will raise the nominal GDP 10-fold, and prices 10-fold, and wages 10-fold, and the nominal stock of all assets by 10 fold.
PS. The equation of exchange (MV=PY) applies to the medium of account, but of course one could substitute any dollar-denominated asset for M, and it would still be an identity. After all, the definition of V is NGDP/M.
PPS. It’s true that the broader aggregates fell in the early 1930s, but you can find plenty of recessions where they didn’t, due to higher demand for liquidity.
PPPS. Totally off topic, but Matt Yglesias’s critique of the electoral college is a rare note on sanity in an otherwise horribly confused debate.
Tags:
29. October 2012 at 08:02
Matty continues to be a sod.
The PROBLEM we face right now is his party’s doing. And we’re not tossing the EC to fix the problem they caused.
The problem is the national power grab away from states. If the TAXES and POWER stays at state level, you get a political system far more responsive to any given individual voter.
Because rabid single issue voters can move with their feet to states far more likely to please them on their single issue.
Once that itch is scratched, for a better sorted populace (people are moving where they are most comfortable), there can be FAR MORE horse trading at the Federal level.
The current centralized power grab has ruined regional compromise, it has removed from DC most of the moderates that having secured the the one major factor for their region, had a vote to shop around.
The EC was set up to screw Matty with his pants on if power headed to DC.
Now that he is yelping about said screwing, the LOGICAL reaction, is to tell him to stop screwing himself.
The future of Progressives lies in states rights – it isn’t a guaranteed winning position for them, but it is the most strategic they have.
29. October 2012 at 08:08
Scott,
“Begs the question?” Not you too. Say it ain’t so, Scott. You’re helping ruin a good term for which we don’t yet have a good terse replacement.
29. October 2012 at 08:12
Great (hypothetical) example with Zimbabwe. I’ve read your writings about account/exchange before, but having a specific, concrete example really helps to clarify the issue.
29. October 2012 at 08:25
Speaking of Zimbabwe (and other monetary phenomena);
http://www.econtalk.org/archives/2012/10/hanke_on_hyperi.html
Hanke says the Zimbabwean economy ‘spontaneously dollarized’ a few years ago.
Also, that money is tight in the US and Europe thanks to Dodd-Frank, Basel III and the mania for recapitalization.
29. October 2012 at 08:43
Great example, your point is much clearer now. I have one question though. Would debt contracts also be in the medium of account, or the medium of exchange? Including government bonds? Because I think that’s the more important issue.
I gather from “Unanticipated inflation helps borrowers and hurts lenders.” being grouped at the top, you consider it part of the medium of account. But I’d imagine a government in the situation you describe would print currency to pay off its bonds and declare the currency good for private debt, and still use gold as the medium of account at the cash registers.
29. October 2012 at 08:43
David, I agree with your scolding, but I think we may have lost the war on that one. Whenever someone thinks of saying “beg the question” they should instead choose between “assume the conclusion” (the correct meaning of the phrase) or “raise the question”. That will end the ambiguity.
29. October 2012 at 08:58
David R. Henderson,
‘Petitio principii’ is shorter than ‘begging the question’.
29. October 2012 at 09:37
“I get the impression that people believe recessions occur because there is less media of exchange to spend.”
“This recourse to a rationing of credit caused renewed stringency in the money market in the spring of 1796 and evoked loud protests from the City (London).
It is not easy to reconcile these complaints about the continued scarcity of money during this period with the no less insistent complaints about high prices, and with the continued unfavorable course of the exchanges.” (Hayek, 1933, p. 40)
From a real-bills perspective, it is easy to reconcile the two sets of complaints. The fact that the bank was rationing its notes indicates that it was selling the notes too cheaply. Any issue of money on these terms would reduce the amount of backing per note. Thus the value of the pound would fall, even if the Bank were issuing fewer notes than usual (and thereby generating complaints of a shortage of money).
Economists often scoff at complaints of money shortages, since the quantity theory implies that a reduction in the quantity of money will simply cause a commensurate increase in the value of the remaining money. But on real bills principles, a reduction of the money supply–accompanied by an equal reduction of backing–leaves the value of money unchanged, while leaving the community without enough cash to conduct business conveniently.
29. October 2012 at 09:51
I think this was a really important post. I’ve been reading MI for 5 years now and understood the MoA vs MoE distinction. But I hadn’t really internalized that the primacy of money as MoA is what really distinguishes your perspective from most others. Once I did, a lot of things clicked into place.
So here’s a question for you. The “Hot Potato Effect”. I’ve always thought of this as people having more dollars in their pockets that they want to get rid of. They expect to conduct a certain number of exchanges in the near future and they budget their dollars. When they have more dollars, they’re willing to make exchanges farther down their priority list. This is due to money as MoE.
But what are you saying with money as MoA? Are you saying they estimate the “replacement cost” of the dollars in their pocket as lower?
29. October 2012 at 10:26
I didn’t see a definition of medium of account on Wikipedia, but did see your discussion with Andy Harless about this on December 11, 2010. Would it be fair to characterize medium of account as how we define our monetary expectations and the medium of exchange as how we go about meeting our obligations? You said “The medium of account is becoming too valuable” and to me, that implies it is supplanting too many value in use economic functions.
29. October 2012 at 10:41
That begs the question of whether the price level is determined by shocks to the medium of exchange, or shocks to the medium of account.
My answer:
In an economy with a medium of account, but no medium of exchange, the medium of account determines the price level, as the inverse of its market price. (Actually the inverse of its market price, times the exchange rate between the average price of all other commodities and whatever is on the other side of the MoA market)
In an economy with a medium of exchange, but no medium of account, there is no price level.
In an economy with both a medium of exchange and a medium of account, the price level equals M*V/Y.
Y is real output. V is the velocity of the medium of exchange (the reciprocal of the amount of MoE people wish to hold on average as a fraction of their nominal income). And M is the quantity of the medium of exchange, times the price of the medium of exchange in terms of the medium of account (the MoA value of the stock of MoE). M obviously becomes simpler to calculate when the MoE is the MoA, as the MoA price of the MoE is always 1.
I think my “theory” will describe the price level better than yours, in situations where it is possible to separate the two. In fact that is what I had been arguing on the previous page. Note that in the US gold standard, dollars were both MoE and MoA. But their supply was constrained by the need to keep the price of gold pegged at the fixed value. So shocks to the gold supply got transmitted to the circulating money stock. Often (though not always), we can do the above calculation by substituting the MoA stock for M. Then it’s possible to consider the impact of gold on the price level in your shorthand way. But not always.
Inflation reduces the value of the medium of account, but not necessarily the medium of exchange.
29. October 2012 at 10:42
Look, Scott, we agree that when gold is MoA, the price level is the real exchange rate between goods-in-general and gold. We also agree that if people want to spend twice as many of their dollars on gold, the price level falls in half. Even if the currency stock is halved, the real demand for gold is still double what it was, and prices fall in half.
We also agree that the real value of gold is not the same as the currency price of gold. If the currency stock is doubled, and gold prices in terms of currency double, the price of goods in terms of gold is unchanged (because the currency price of goods doubles as well), and there is no deflation. It’s the real value of gold, not the currency value (trying not to set up any strawmen here). So it looks like our theories are almost the same.
But there is a difference. Velocity/money demand is the velocity of MoE, not MoA. Let’s go back to our initial disagreement.
“If there is a big drop in apple output, then the value of apples will rise, producing severe deflation (and falling NGDP), regardless of what happens to the money stock.
Apples are MoA, currency is MoE. Suppose the apple crop falls. Also suppose people reduce their desired currency balances by an equivalent amount, without increasing their demand for apples. There will be no deflation. Yes the currency price of apples rises, pushing the price level down. But the rise in currency velocity offsets this exactly. So there is no deflation.
Now, if you wanted to you could call the rise in currency velocity “a decrease in the real value of gold”. I agree that it’s tautologically true that deflation is an increase in the real value of the medium of account (each unit of the medium of account exchanges for more consumption baskets), and vice versa. But none of the change occured in the market for gold itself, as gold is not traded directly for anything except currency, whereas currency is traded for everything.
“I never have any problem getting cash, even during recession years.”
Now you are commiting fallacies of composition. I remember you once told Arnold Kling that the Fed couldn’t determine how much money you wanted to carry. But it can; just not in real terms. People can’t get rid of money in the aggregate; P or Y adjusts until they are satisfied with the outstanding stock. Similarly when money demand rises in a recession, P and Y fall until people are satisfied with their MoE stocks. Of course you are satisfied; nominal income has fallen until you are! If you and many others became further unsatisfied and wanted more MoE, well the MoE stock is still fixed so NGDP will have to fall some more until you (people in aggregate) want no more in nominal terms.
The lack of spending wasn’t because they had less currency in their wallets and piggy banks; they had plenty. Rather they had a higher demand for money, and simply chose not to spend it as often.
A shortage is an excess of quantity demanded over quantity supplied. At the healthy level of NGDP, there was excess demand for money. So NGDP fell, until the demand for money fell to match the supply. Indeed even if the money supply is higher than normal, an even more abnormally high money demand will cause the flow of MoE spending to contract. This pushes down NGDP, which reduces the demand for money to match the supply.
You know all this, of course.
29. October 2012 at 10:42
“suppose that Zimbabwe was on the gold standard at the same time that soaring gold demand in Asia was pushing its value up, relative to other goods and services. Zimbabwe would experience deflation. If the wages of Zimbabwe gold miners were sticky in gold terms, then the diamond mine would have to lay some of them off. They would be unemployed, and would start buying fewer goods and services. Indeed in some cases market signals (such as higher interest rates) will encourage people to buy less even before the workers are laid off (the effect before the cause). But the fundamental problem is that there is too little NGDP, given the current (sticky) level of nominal hourly wages. The medium of account is becoming too valuable.”
OK, now I’m not sure if it was a strawman. Remember it’s the real value of gold that matters, not the nominal value. Let’s say that soaring gold demand doubles the $Z price of gold. But suppose we also doubled the $Z stock, and that the extra $Z did not further affect the price of gold. Then there would be no deflation. The increase in (nominal) demand for gold does not cause deflation/recession. It would have, but only because it was going to halve the nominal value of the stock of MoE – which was fixed by doubling the size of that stock (without further impact on its gold value), removing the MoE shortage.
Of course, you might astutely point out that doubling the $Z stock halves its purchasing power, and so holding the gold-$Z exchange rate constant will halve the real value of gold. But now we are back to the tautology.
Furthermore, deflation does not cause a recession if there is no MoE (so long as prices and wages adjust at the same rate). Any rise in MoA price corresponds to a fall in the price level, removing the excess supply of goods. If demand for the MoA rose, and the price level didn’t adjust, then neither would the MoA price. There would be an excess supply of goods, and a shortage of MoA. But this would not cause unemployment, as people would simply continue bartering goods for each other.
“The equation of exchange (MV=PY) applies to the medium of account”
Gee, you don’t think it might be a model of the flow of the medium of exchange?
29. October 2012 at 10:45
Kevin, I’d like to know as well. AFAIK the hot potato effect is all about the MoE, it doesn’t make sense otherwise. (And remember Scott’s views are not the only interpretation of market monetarism – there are other market monetarists as well, whose blogs are linked to in the sidebar.)
29. October 2012 at 10:48
Darn it Patrick, you got there first! I had also planned to draw Scott’s attention to the Hanke EconTalk, when it was released.
29. October 2012 at 10:52
Scott, my point is that although it’s strictly correct to say that “the price level is the inverse of the real value of the MoA”, this doesn’t mean that the currency price of the MoA is all that matters! You are probably going to miscalculate the “real value” (goods-purchasing power) of the MoA, unless you pay attention to the MoE stock and its velocity of circulation (because the MoE intermediates the purchasing power of the MoA). But if you do take that into account, then yes it’s a tautology.
29. October 2012 at 10:55
But it’s still necessary to have a shortage of MoE at the sticky-wage consistent level of NGDP to get a recession. Either that, or, deflation in prices without deflation in wages – a rise in the real wage. But that’s not a recession due to a lack of spending, and there won’t be a general glut.
29. October 2012 at 10:58
David, I changed it to raises the question. I’ll try never to make that slip again, but given my horrible memory, I almost certainly will.
Negation, Yes, debt is in gold terms. That’s actually in the post, you may have missed it.
Mike, Yes, there can be money shortages, when the medium of exchange and medium of account differ. My point is that this is not a common problem. Most US recessions do not involve money shortages, so there has to be some other mechanism reducing spending.
Thanks Kevin, Actually with modern currency it is both a MOE and MOA. The hot potato effect is another way of describing the process of getting rid of excess cash balances. So I can see how people interpret that from a MOE perspective. But what matters is the effect of this activity on NGDP and the price level, which comes from money’s role as a MOA. The Zimbabwe dollar in the example above was also a hot potato. But the effect of getting rid of excess cash balances was not to raise the price level, but rather to reduce the gold price of Zimbabwe dollars.
Becky, The MOA is the thing which all other goods are priced in terms of. It’s a good whose price is one, by definition.
29. October 2012 at 11:01
ssumner:
If the trickiness is due to money being “almost always both”, then why not consider cases when it is not both, but one or the other?
Can a money be a medium of account, but not a medium of exchange? I don’t think so. That which is used as a medium of account is borne out of it being a medium of exchange. The reason why people use dollars as a medium of account is because they are using them as a medium of exchange. Do you use apples or gasoline as a medium of account in your catallactic affairs? If not, why not? Is it because apples are not universally accepted for your goods and services, and so you think why bother calculating your gains and losses and tracking your expenditures/revenues in apples or gasoline?
Suppose gasoline really did become the universal medium of exchange (i.e. the state taxes income earners in gas, the courts issue rulings payable in gas, wage earners earn gas, businesses earn gas revenues, charity is given in gas, etc), do you think dollars would remain a medium of account? If you don’t think so, then doesn’t that imply the medium of exchange property is decisive and is the “essence” of money?
Since I hold that the “standard” of economic life is an individual private property order, and since the money of choice in such an order would almost certainly be precious metals based, such as gold, I like to define inflation this way:
“Inflation is an increase in the quantity of money greater than an increase in the production of gold.”
Of course, formally it would be “…greater than an increase in the production of free market money”, but to give it content for explanatory purposes, I use gold (all at the risk of being accused a “gold bug”, which is fine, because it enables me to point out who are likely to be “fiat bugs”).
Let’s be clear here. If Zimbabwe is using gold as a medium of account, then what of a medium of exchange? Is gold being used as a medium of exchange as well? If not, what are they using? If so, then please note you are giving an example of gold as the medium of exchange.
Interesting scenario. Completely opposite of the actual motivations of most Presidents, who tend to be madmen who want to continue to enforce fiat money (which greatly benefits spendthrift Presidents, especially when faced with “budget problems”). It is also interesting because a supposed small government blogger would, when considering “budget problems” for a spendthrift state, lean towards granting that spendthriftyness, and advocating for inflation so that the state does not have to cut back on its spending and hence on its size.
But I digress…
Actually gold would still be the medium of exchange in this example. The fact that daily exchanges are in the physical form of paper-based transferable claims to gold (gold which held by the state I am assuming), it doesn’t mean Zimbabwe dollars are the medium of exchange. What matters is the property to which the titles refer, not the titles themselves. The “Zimbabwean dollar” is a transferable claim to gold held by the Zimbabwe state, redeemable by I am assuming whatever the redemption rate is for 1 dollar. If one owns a Zimbabwe dollar, they own an absolute claim on a specific quantity of gold held by the state.
If a Zimbabwean has a paper based transferable claim that has written on it something to the effect of “The bearer of this note can redeem it to the Treasury for 1 ounce of gold”, then if the state inflates the number of paper based transferable claims, then they would experience a gold outflow. If they print too many, then they would default (Nixon?).
Here, since the additional transferable claims to gold remain circulating as part of the money supply, then the “true” rate of inflation, using my definition, would be the additional quantity of notes that are greater than the supply of note-gold redemption equivalents. For example, if there were originally a total of 1 billion ounces of transferable claims (and 1 billion ounces of gold), and then an additional 100 million transferable claims were printed with no corresponding increase in gold, then the 100 million “gold ounce claims” are the rate inflation (100 million on 1 billion is 10%).
I think I finally figured out your confusion. You don’t seem to understand that in a gold standard, the paper based transferable claims to gold would not be divorced from the gold. By this I mean in a gold standard, each Zimbabwe paper note would entitle the bearer to a specific quantity of gold held by the state, and this redemption rate would be either written on the notes themselves, or promised to the bearer of the note. It would not be like a fiat standard in which case the supply of them can increase to whatever degree, and nothing will go wrong for the gold standard, because heck, the exchange rate between dollars and gold can just go up.
That isn’t how gold standards work. If the Zimbabweans are on a gold standard, if they make exchanges payable in gold, then each note MUST carry with it a specific, rigid, unchanged quantity of redeemable gold. If there isn’t, then nobody actually OWNS any gold by owning those dollars!
Thus, if the Zimbabwe state inflates the quantity of transferable claims to gold, then they would end up doing what the US did prior to Nixon defaulting on the promises to redeem US held gold to foreign central banks who send back the US dollars!
You just described an example of a fiat money standard, with a weird sort of caveat where people track all their exchanges in terms of apples (in your case gold).
You have to be clear. If Zimbabwe has a gold standard, then there won’t be any “paper price of gold”. There will be a gold price for Zimbabwe paper. Either the owners of the paper claims are told that their notes are hereby revalued at a discount, say 10%, where the “old” notes do not exchange against the “new” notes at par, in which case people really DID experience a reduction in the cost of their living, or the state doesn’t tell the people about ALL of the notes that they inflate (usually the case), in which case the reduction in standard of living is felt mostly by the naive, and least so by the informed investor and insiders who receive the new paper dollars first.
If however the gold redemption value of each note remains the same as it was before, then the state cannot keep inflating those notes, because at some point, they will run out of gold as people make gold redemption requests, and if more people catch on to what the state is doing, then there may be a run on the Treasury for all its gold (since people are promised more gold than the promises contained in the notes), in which case the state will default, and then those Zimbabwe dollars would become wood kindling.
29. October 2012 at 11:12
Scott, you said yourself that the demand for money which determines the HPE comes from money being like wallets, helping to facilitate transactions. The HPE would make no sense otherwise, it only works for MoE, which is on every market, and which everybody holds inventories of to facilitate transactions. Nick Rowe explains this at length.
A drop in the nominal (MoA) value of the MoE can cause a shortage, as well as a drop in the actual quantity of MoE. And NGDP falls to clear that shortage.
“The Zimbabwe dollar in the example above was also a hot potato. But the effect of getting rid of excess cash balances was not to raise the price level, but rather to reduce the gold price of Zimbabwe dollars. “
No, it was foreign demand pushing up the $Z price of gold, remember? The supply and demand for $Z was in equilibrium, in your example. If the gold price doubled because the $Z stock had doubled, then there would be a hot potato effect on goods, which would cancel out the reduced gold-value of the $Z – and goods prices wouldn’t fall. But although the gold price would double, the “real value” of gold would not have.
“what matters is the effect of this activity on NGDP and the price level”
But if money isn’t a MoE, then a falling price-and-wage level is a rising MoA price. And that’s an economy successfully adjusting. If the price-and-wage level is stuck you get disequilbrium in the MoA market, but no problems elsewhere. Only a drop in the flow of MoE (which we call “spending”) can cause a spending-side recession.
29. October 2012 at 11:13
It is often difficult for people, even PhD economists, to think about how a non-fiat money standard works. This is probably due to the fact that most think in terms of history, which has been monopolized by fiat since 1913.
It cannot be stressed enough that there is a difference between:
1. A fiat standard where the “dollar” price of gold can rise to infinity with inflation
2. A gold standard where the gold price of “dollars” is fixed in accordance with the redemption rate of each “dollar”, where the supply of “dollars” cannot rise to infinity because the gold would be insufficient.
29. October 2012 at 11:15
MF, imagine a barter economy, such as those that have actually existed. Calculate the implicit exchange rate between one good, say apples, and everything else. These are the “nominal prices”, apples are the “numeraire”, and we still have no money or MoE in this economy.
29. October 2012 at 11:17
MF, we haven’t been talking about actual gold standards, just hypothetical scenarios to tease out the MoE/MoA distinction (and hierarchy). Keep up.
29. October 2012 at 11:20
Saturos, I’m having trouble following your examples. If apples are the MOA, and the apple crop falls by 99% so that only the ultra-wealthy can afford to buy apples, surely the prices of goods and services in apple terms will fall–regardless of what happens to the currency stock.
There is no “fallacy of composition” in my denial of the role of currency “shortages.” I certainly understand that when currency is a MOA (like today) a reduction in the supply of currency is deflationary. But that need not involve a “shortage,” indeed I think it rarely does. When 90% of the coffee crop is wiped out there is no shortage of coffee, just higher coffee prices. Don’t confuse the terms ‘supply’ and ‘shortage.’
I can’t see where the equation of exchange helps, it’s simply a definition of V, and holds for M being defined in terms of any nominal variable, including the nominal stock of bananas. And it’s not a “model” at all, it’s a definition of V, nothing more.
I think you misunderstood my gold example. Soaring Asian demand for gold raises the value of gold in terms of other commodities. Since I was assuming that Zimbabwe was on a gold standard, that means deflation in Zimbabwe regardless of what happens to the stock of Zim$s. They could rise to one quandrillion, or fall by 99%, and it would have no impact on Zimbabwe prices.
Regarding recessions, I see them as being about unemployment. The effect of recessions on inventories is actually not all that important. Recessions are big drops in output and employment—less purchases of goods and services tends to go along as a side effect.
You said;
“Kevin, I’d like to know as well. AFAIK the hot potato effect is all about the MoE, it doesn’t make sense otherwise. (And remember Scott’s views are not the only interpretation of market monetarism – there are other market monetarists as well, whose blogs are linked to in the sidebar.)”
See my reply to Kevin. Any market monetarist who disagrees with this post will be removed from my sidebar. 🙂
29. October 2012 at 11:25
‘Begs’, in the phrase ‘begs the question’, is one of those *problem words*, such as ‘disinterested’, ‘drastic’, ‘meld’, ‘reticent’, and ‘specious'””that so many people use “wrongly” (according to the stricter authorities) that their misuse has almost become the standard use: these terms can no longer be used as the old authorities prescribed without creating confusion. It’s probably best just to avoid them.
David R. Henderson says: “You’re helping ruin a good term . . . .” Too late, David””it’s already ruined. But W. Peden is there to save us; ‘petitio principii’ is surely safe from ruin!
29. October 2012 at 11:46
I see a problem here
“Imagine you are a Zimbabwe diamond miner who works on a wage contract promising you 10 grams of gold per month. You could care less what the prices are in Zimbabwe dollars, you have a fixed wage in gold terms, and when you go shopping you will see price stickers on the goods in gold terms, not Z$.”
Your Zimbabwe dollar is also a store of value, and a very bad one. Anyone holding these dollars loses when the value relative to gold falls – a covert tax. I would expect the result to be hyperinflation. The specification of dollars as the medium of account is a form of inflation proofing common for contracts in high inflation countries. But if everybody can play this game, then there’s no game.
In your model the price of goods in gold has to increase to compensate for the losses in nominal gold from time spent holding dollars. People will then buy less goods and/or demand increases in pay in gold terms to pay the prices.
What happens depends on the interface to the rest of the world, which you haven’t defined. Foreigners would be crazy to take Zimbabwe dollars. They would want real gold or a deep discount. The result is some combination of a gold shortage and foreign goods inflation depending on how government responds.
We’ve seen something like this with the communist countries. Some people get better foreign currency rates than others and there are restrictions on private holding and conversion of foreign currency and probably price controls as well. The stores are empty or full of shoddy goods. There are long lines for sales of anything good.
In the communist countries this caused a shortage of quality goods. People who had rubles (or zloty or lei)had nothing to buy with them. The way to get goods was either barter, home manufacture (gardens in the dachas) or use of influence.
now
” Zimbabwe would experience deflation. If the wages of Zimbabwe gold miners were sticky in gold terms, then the diamond mine would have to lay some of them off. They would be unemployed, and would start buying fewer goods and services.”
But there can’t be general deflation if the cost of foreign inputs is fixed in terms of gold. You have a nontradeable goods deflation and a tradeable goods inflation. Tradeable goods will disappear because no one will be able to afford them. At this point the government will intervene to protect its interests, and the result will probably be like the communist countries. And we still have hyperinflation of dollars from the covert tax.
I suspect such a combination couldn’t happen because its inherent contradictions would destroy it.
29. October 2012 at 11:52
“Give me control of the stock of currency…”
Scott Sumner reveals his true plan all along.
29. October 2012 at 12:01
Saturos:
This cannot be “calculated” from one’s arm-chair. The exchange rate between apples and everything else is borne out of the market process of exchange between apples and everything else.
If there is no universal medium of exchange, then the exchange rate between apples and everything else would be something like this:
1 apple exchanges for 1 orange.
10 apples exchange for a pinapple.
100 apples exchange for a sheep.
etc.
There would be no “numeraire” of apples here, because each individual seller would be selling goods (which may or may not be composed of apples) for other goods (which also may or may not be composed of apples.
That is the very problem I was addressing. I am arguing that the Zimbabwe example as stated IS a gold standard (although because there is later on in the example a paper inflation beyond the gold, it goes into sort of a no man’s land of a transition between Gold and Fiat).
Teasing out the differences between this or that is impossible unless we first correctly identify what it is we are considering, and whether or not the hypothetical example even makes logical economic sense.
29. October 2012 at 12:09
But surely any medium of account has to be a medium of exchange? By saying that accounting is done in dollars, are we not implicitly saying that things can be exchanged for dollars?
29. October 2012 at 12:10
Saturos:
In a barter world, why would everyone be using apples as a “medium of account”? Why would horse breeders who sell horses for cows, and why would iron smiths who sell iron to shoemakers, bother with accounting for their transactions in “implict” apples as a numeraire? Why would the horse breeders for example go through the effort of tallying their profits and losses, revenues and expenses, in terms of apples, when they are neither selling their horses for apples, nor selling apples for horses?
If apples are not universally accepted as a medium of exchange, then it would make no sense why apples would become a universal medium of account for everyone, especially the horse breeders who don’t even like apples.
Money arises as a medium of account because the horse breeders can sell horses for that money, and then buy what they DO want, and they can calculate their gains and losses in terms of that commodity, and compare and contrast them with other plans.
29. October 2012 at 12:19
Felipe:
Yes.
It is impossible for there to be a market exchange rate between gold and dollars unless dollars are actually being exchanged for gold.
Sumner’s example drops this requirement, and he seems to believe that a daily “ticker tape” exchange rate can exist between dollars and gold anyway, even though as postulated in his example, everyone is only using dollars in their exchanges.
It is economically illogical.
29. October 2012 at 12:41
Murphy:
“Give me control of the stock of currency…”
Scott Sumner reveals his true plan all along.
I know you’re joking, well, maybe half joking half not joking, but this is serious. By “serious” I don’t mean that you shouldn’t joke about it, but rather I think you are referring to a very serious issue that has to be made explicit, however psychologically or emotionally uncomfortable it may be).
State control, in money for example, cannot last without court intellectuals who desire economic control either directly from themselves or vicariously through others.
Intellectuals bent with central planning urges are, conscious or not, wanting to control others via force. They may seriously believe they are helping others, but what they advocate always has winners and losers. They can’t justify themselves by focusing on the winners and ignoring the losers. But in central planning that is often easy, because the costs are so dispersed that empiricist minded intellectuals don’t even bother with them. It’s like if there is a single loud mouthed taker, and a billion slightly annoyed givers, then the taker trumps all because he garners the most attention.
It’s sad because the thing that made humans dominant and prosperous is a resistance to short term anything goes urges, i.e. might makes right, in lieu of prosperity through restrained cooperation, i.e. reason.
I would shudder if he actually believes central banking is an institution of cooperation, rather than old fashioned tribe-style obedience based on naked aggression, redressed in “modern” verbiage and rhetoric. Whew!
29. October 2012 at 14:32
The most liquid and divisible medium of exchange will BECOME a medium of account, naturally, while also becoming a store of value (even if it gradually drops in value during inflation, this could be thought of a a “premium” for liquidity). Liquidity, and therefore, medium of exchange is the essence of money, if an asset were more liquid than money, then that would replace the current thing designated with money as the new money.
29. October 2012 at 14:48
OK, let me see if I’ve got this Hot Potato interpretation, because I feel like I’m on the cusp of an internal breakthrough:
(1) I have a prioritized list of exchanges I want to make in the near future. My cash balance increases, so I go farther down my priority list. This is pure MoE.
(2) In effect, I’ve now claimed a larger share of NGDP.
(3) But everyone else does the same thing. So now there are more claims on NGDP than there is actual NGDP.
(4a) The MoA adjusts somewhat so that each unit of currency is now a smaller claim on NGDP.
(4b) Due to stickiness, the supply of output also increases somewhat to meet the perceived increase in demand.
The combination of (4a) and (4b) result in a new equilibrium. And the ratio of (4a) to (4b) is the slope of the SRAS curve.
So we’re essentially saying that the slope of the SRAS is the “coefficient of friction” in the nominal economy?
29. October 2012 at 14:50
Britonomist:
The most liquid and divisible medium of exchange will BECOME a medium of account, naturally, while also becoming a store of value
To pick at your brain, why do you suppose that is?
29. October 2012 at 14:59
Kevin Dick:
I don’t understand (3). If everyone has dollars, and everyone has a “claim” to the share of total spending, why can’t NGDP be the total of what all individuals intend to spend? Or are you assuming a centrally planned fixed NGDP? If so, then in order for the planned NGDP to be manifested, the state will have to reduce the cash balances of individuals, so that the sum of the remaining cash when spent does hit the target.
29. October 2012 at 15:00
In which case the “claims to NGDP”, to use your parlance, will decrease to match “actual NGDP”.
29. October 2012 at 16:31
I have one question. Does anyone know how to model the velocity of money dynamically. I find it surprising that I haven’t yet seen a differential equation to describe this. If someone could point me to an example/paper that would be much appreciated. There must be feedback mechanisms/nonlinearity that exist between the two.
The thing about the equation MV=PT is that changes in the velocity of money(V) effect the changes in the supply of money(M). Maybe my intuition is wrong on this, but I would certainly imagine that there are feedback mechanisms between M and V. Therefore, a 15% increase in the money supply right now might lead to a larger/smaller increase in the price level due to the nonlinearity.
29. October 2012 at 16:45
One more note, a better way to write MV=PT might be MV=sum(p(i)*q(i)) where p(i), q(i) are the i-th type of good. Also, why can’t increases in M or V flow directly into asset prices and have less/little/no impact on NGDP?
Again, does anybody know any papers or books that detail this issue? Any sort of feedback(even if the papers aren’t the greatest papers or there is some flaw in the work) would be much appreciated. I’m actually really interested in the nonlinear impacts of each of these variables on one another. Empirical work would be much appreciated as well.
29. October 2012 at 16:54
@MF. This doesn’t all happen simultaneously. I should have said, “I plan to claim a larger share of near future projected NGDP.”
So do other people. But different goods are at different positions in each person’s list. So you have a variety of “demand” signals arriving over time.
Some of the supply response is more output and some is changing the index of the MoA. At least, that’s what I’m asking Scott.
29. October 2012 at 17:05
Saturos, To get a recession you just need NGDP to fall much faster than wages. I’m not making any assumption about prices, which might also be sticky, or not.
When I compared currency to wallets I was envisioning a system where currency was the MOA, not gold. In that case the currency market determines the price level and NGDP, and the fact that currency is quite useful, like wallets, is important when modeling money. The demand for money is unit elastic (with the value of money on the vertical axis), as it’s only value is what it can buy.
You said;
“No, it was foreign demand pushing up the $Z price of gold, remember? The supply and demand for $Z was in equilibrium, in your example. If the gold price doubled because the $Z stock had doubled, then there would be a hot potato effect on goods, which would cancel out the reduced gold-value of the $Z – and goods prices wouldn’t fall. But although the gold price would double, the “real value” of gold would not have.”
I have no idea what you are saying here. Can you be more specific? What does the term “gold price” mean? Price of what? BTW, I didn’t mean to claim that foreign demand was pushing up the $Z dollar price of gold, I was claiming it was pushing up the value of gold in terms of other commodities—not just in Zimbabwe, but everywhere in the world.
You said;
“But if money isn’t a MoE, then a falling price-and-wage level is a rising MoA price. And that’s an economy successfully adjusting. If the price-and-wage level is stuck you get disequilbrium in the MoA market, but no problems elsewhere. Only a drop in the flow of MoE (which we call “spending”) can cause a spending-side recession.”
In any monetary model, deflation is a problem because prices are too high. It’s because deflation isn’t severe enough. If NGDP falls 50% and P only falls 25% because prices are sticky, then the other 25% shows up as falling output. You needed P to fall 50%, but it didn’t because wages and price are sticky.
Peter N, You are right that the miners suffer a loss of purchasing power as their $Zs depreciate. But your communism example is wrong. There are no shortages in Zimbabwe in my example, because there are no price controls.
Filipe, In the early 1800s the medium of account in the US was gold or silver (I forget) and the MOE were banknotes that traded at a discount, and hence were not media of account.
Britonmist, I agree that media of exchange usually become MOA as well, but the key role is the MOA.
Kevin, That sounds reasonable, but I’m not sure about the slope. The slope should be the change in P over Y, not the change in PY over Y.
29. October 2012 at 19:09
@Saturos/Scott
FWIW, Scott’s argument sounds the stronger of the two. I state this from a very specific perspective – his argument addresses the stock issue, while the other focuses exclusively on the flow issue.
If someone were to ask me what money is, my answer would be a unit of the societal expectation of near-future consumption, which is in a sense a unit of account.
It is precisely based on this view that I disagree with Saturos’ statement: “Furthermore, deflation does not cause a recession if there is no MoE (so long as prices and wages adjust at the same rate).”
This begs the question. Imagine that everyone desires a 2 year buffer in their savings account. A purely financial event happens. In this event, half of people see their money double, and the other half see it cut to zero. The second half will immediately strive to rebuild wealth, and their efforts are unlikely to be fully and immediately matched by additional consumption from the newly wealthy. Likewise, differential consumption patterns due to changes in wealth distribution will cause structural changes to demand (instead of 100 middle class houses, we’ll need 50 upper class and 50 lower class). [Yes, I just asserted there was a structure to demand, which is something I model every day – we call it consumer heterogeneity.]
In this scenario, we can observe deflation that is exactly matched by wage decreases, and STILL observe a recession. This recession was entirely monetary in nature, but it had nothing to do with a change in the value of the unit of exchange. It had to do ONLY with a change in the value of the unit of account.
Indeed, in this recession (which is only partly structural), it’s quite reasonable (though not certain) that wages will need to fall by MORE than the price fall to retain full employment.
29. October 2012 at 19:12
“There are no shortages in Zimbabwe in my example, because there are no price controls.”
The dollar is a covert tax. The mine owner needs 1.1 oz of gold in dollars at date t to pay the miner 1.0 oz equivalents on date t+2, which have a purchasing power of 0.9 oz on date t+4. The mine owner must bill for ore at a 1.2 oz rate at date t-2 so there will 1.1 left when payment is received. This is hyperinflation and it’s a negative sum game. You have no winners (at least initially), just greater and lesser losers.
Eventually, nobody will use dollars if it can be avoided by using other currency or barter. In theory the dollar could just inflate. The record for this is the Hungarian Pengo with a monthly inflation rate of 4.19 X 10E16, or 207% a day. It’s unlikely to be broken any time soon. The Zimbabwe dollar was a mere 7.96 X 10E10.
In reality, the process reaches it’s societal limit, and then the system breaks down.
Also, the problem of external tradeable goods will force some kind of currency control to protect the government establishment’s own supply of tradeables. Foreigners won’t take payment in Zimbabwe currency, so importers will need foreign exchange. To protect the dollar and prevent a gold drain, you need exchange controls. These also are a way to reward partisans with favorable conversion rates.
The shortagety of affordable foreign currency will cause the supply of foreign tradeables to dry up. Local goods could stay in reasonable availability, at least for barter, but at Hungarian rates of prices doubling every 15 hours, it’s unlikely anyone will want to sell for Zimbabwe dollars.
OTOH, the velocity of money should be phenomenal. Talk about a hot potato.
29. October 2012 at 19:22
Following up –
Any astute observer will note that the above argument is made without any reference to debt (because Scott is tired of me talking about debt). However, the artificial case above can be duplicated with a realistic case in which 70% of the population are net debtors (trying to repay debts in a fixed period of time) and 30% are net creditors.
A change in the unit of account alters things. Even if the economy were run by BARTER, a financial shock inducing deflation (or rather, any deviation from expected inflation) would cause huge distortions. Recovery would occur, however a financial shock causing an increase in debt could trigger an expectation of deflation which would propagate changes in the distribution of the demand for savings and hence in the structure of demand. [Yes, I just said it again.] This is all driven by perceived future wealth, not perceived future income (we have ample empirical evidence that households do use wealth to buffer consumption from income changes).
So, on the one hand, I agree 100% with Scott that money is primarily a unit of account. The only issue I have is how he can believe this, but at the same time deny the notion that there exists such a thing as a structure of demand, or deny the importance of debt (over wages) in deflation-drive recessions.
29. October 2012 at 19:32
Scott,
Great post – I’ll add one more example to Zimbabwe.
Imagine there are two island nations. They almost never conduct trade with one another, but they have a treaty that when one suffers a catastrophe, the other will provide a fixed amount of supplies to help its neighbor rebuild. Whenever this happens, the island receiving help gives the helper island a hundred moogles. The moogle is good to be redeemed at any amount of time for 1/100th of the fixed set of assets that is specified in the treaty (there is no interest).
Many decades can go by without a natural disaster, and trade is constantly conducted between the islands by barter. Occasionally, a trader’s wares rot on the journey, and the trader still picks up a new load of goods from the neighbor island. To specify what he owes in the future in an easy way, they will sometimes agree on a specific debt that is owed in goods, but at other times they will agree to pay a tenth of a moogle (or will actually exchange a physical embodiment of that tenth of a moogle if the trader is untrustworthy).
Is the moogle money or not? (I would say yes.) Is the moogle a medium of exchange? (I would say not primarily, but it could evolve into one since it’s convenient.) Does the moogle derive its value as a medium of exchange from its value as a medium of account, or vice versa? (I would say medium of account is primary.)
29. October 2012 at 19:52
ssumner:
Britonmist, I agree that media of exchange usually become MOA as well, but the key role is the MOA.
How can “the key role” of money be something that occurs after, logically and temporally, its medium of exchange use?
Can you envision a scenario where the MoA of a commodity precedes it as a MoE?
29. October 2012 at 20:55
The moogle is not money because it is not a universally accepted medium of exchange.
It is not a universal medium of account either, because it is not a final means of payment which can serve as a datum for calculating gains and losses, revenues and expenses.
It is a specialized debt claim on a supply of assets which are themselves traded according to barter.
29. October 2012 at 21:04
FALSE:
“Once we answer that question, the business cycle problem will also be solved, as we all agree that unanticipated price level shocks can trigger business cycles.”
Assuming to confirm what must be shown is called “Begging the Question”.
There are all sorts of reasons to laugh at the notion that “we all agree that unanticipated price level shocks can trigger business cycles.” Ie the claim is dubious in the extreme, and is inconsistent with the findings of Econ 101, ie it’s a still born anomaly-ridden mess of a claim out of the barn door.
29. October 2012 at 21:05
ssumner:
So you are proposing a scenario where every seller, or enough sellers, overestimate their expected revenues, which leads to a difference between expected aggregate sales and actual aggregate sales over a given time period?
29. October 2012 at 21:13
Sorry, that was for Kevin.
29. October 2012 at 21:13
You’ve mistaken a consequence for a cause:
“We all agree that unanticipated price level shocks can trigger business cycles.”
Can’t see that you can possibly establish that I’m wrong, if you acknowledge the logical facts of choice over heterogeneous production goods and the empirical facts of the causal interactions leading the massive expansion and contraction of ‘shadow money’.
29. October 2012 at 22:24
Don’t worry Greg, as long as there is sufficient re-inflation of base money from the Fed, to counter-act the shadow banking deflation, then we can all pretend that the resulting investments and resource allocations are sustainable.
The wild swings in shadow banking asset prices is clearly due to bipolar disorder of investors, and has nothing to do with past inflation misleading them.
29. October 2012 at 22:40
@MF. Not exactly. Again, this is just me trying to figure out how Scott’s primacy of money as the MoA affects the story of the Hot Potato Effect.
From before, all buyers consume farther down their wish lists than they planned (due to having higher cash balances).
This causes sales for each seller to _exceed_ their previous estimates. But due to the heterogeneity of the wish list ordering, some of the sellers see this demand earlier than others. There’s no common knowledge among sellers of a general demand increase.
Therefore, each seller responds in two ways, he increases production and raises prices. Once this has all trickled through the system, output has increased some and the average price level has increased some.
The ratio is determined by the shape of the SRAS curve. But the average price increase re-indexes the MoA.
29. October 2012 at 22:54
StatsGuy, the flow of MoE is determined by equilibrium/disequilibrium in stocks. I’m not too sure about the argument you’re making, but I think it runs into the “why would everyone work less if they are poorer” Sumner critique. Unless you are talking about mere sectoral unemployment – but that’s structural unemployment, not cyclical as we were discussing. That happens even in a barter economy; you deal with it by “wing-walking” or lowering price, or by moving elsewhere in the economy.
I think you need to clarify your model. I don’t know what a savings account is when there is no MoE. Is is some kind of claim to fixed real assets? (Or a bank liability – in barter, IOUs for commodities – backed by a claim to real assets?) So half the population is hit by a wealth shock, which makes them spend the rest of the period trying to exchange their newly-produced output for fixed assets held by another group – or are they producing more fixed assets? Why does everyone become unemployed now? What is “aggregate demand” here? People either go on bartering, or they have to “recalculate” a new pattern of barter, as Kling likes to put it. Yes recalculation is always necessary, but that’s not a general glut, not demand-side unemployment.
In your second example, why would there be mass unemployment? How would there be a general glut of goods, putting the workers who made them out of work?
In your final example, you said nothing about moogles being the numeraire for all goods. And they are not replacing the barter system of exchanging to meet the coincidence of wants (there is no focal commodity for exchange). So moogles are credit, not money. (They are also cute furry winged creatures from the Japanese Final Fantasy games, as you might be aware – http://finalfantasy.wikia.com/wiki/Moogle)
Kevin, I’d hold off on trying to contrive a reconciliation between Scott’s views and the Hot Potato Effect, if I were you…
(Which reminds me, Milton Friedman would almost certainly have disagreed with Scott here.)
29. October 2012 at 23:02
Scott, you said:
The demand for money is unit elastic (with the value of money on the vertical axis), as it’s only value is what it can buy.
You are talking about a MoE there. Obviously we could take a single market in a barter economy, call one of the commodities traded there the numeraire, and this arbitrarily selected commodity may or may not have a unit-elastic demand curve. If money is only a MoA with no MoE, halving the money stock could triple its price, and reduce the price level to 1/3. Because money is not a MoE here, it’s not being used to buy stuff generally, so there’s no reason why people demand to buy exactly half as much money when the price level halves. The price level (inverse of the price of the MoA) might have to fall threefold before the MoA market clears – it’s just an arbitrary market selected as the numeraire. There is no general medium of exchange.
Regarding $Z and gold, if my paragraph didn’t make sense to you then I’ll just have to draw you a diagram. I’ll send you one when I have time. “gold price” = currency price of gold. Amount of currency paid to buy 1 unit gold. If foreign demand pushes up global gold price, the supply available to Zimbabweans shrinks, so the $Z price does too, yes?
“You needed P to fall 50%, but it didn’t because wages and price are sticky.”
Yeah, and I’m saying that the remaining 25% it doesn’t fall shows up as excess supply of Y and excess demand for M. With no MoE, the “M-Y” market is now in disequilibrium, and the “Y” does not all get traded for the “M”. But since M is not a MoE, there’s no reason why the Y that doesn’t get traded on this market should be subtracted from GDP, so long as the Y outputs can still be bartered for each other. (So “NGDP” falls 25%, not 50%) Think Walras.
29. October 2012 at 23:03
Saturos, To get a recession you just need NGDP to fall much faster than wages. I’m not making any assumption about prices, which might also be sticky, or not.
Scott, I can no longer assume I know what you are talking about when you say “NGDP”. I use it to mean MV, the volume of flow of the medium of exchange (times its value in MoA terms, usually 1). I think you use it to mean PY, the nominal value of output bought by the flow of medium of exchange. So if the price level is stuck and NGDP falls, then Y by definition is also falling. You have another tautology. That says nothing about the relationship of unemployment to aggregate spending (and spending is the flow of MoE). With a fixed price level P how can money cause NGDP to fall? Money isn’t generally being traded for Y, if it isn’t a MoE.
If prices fully adjust but wages are stuck, a drop in the flow of spending lowers P and leaves Y unchanged. But now it will no longer be possible to pay the workers who made the Y, as real wages have risen too high (and they lose their jobs). But if P doesn’t fall, you can’t blame real wages. Now NGDP has to be the flow of MoE to cause unemployment. If money is only a MoA and there’s no MoE, then if P doesn’t fall then it is impossible for the value of money to rise, even if there’s a shock to its supply. There is disequilibrium in the money market, but Y producers go on bartering. And “NGDP” is unchanged, as P and Y are unchanged.
OTOH if the value of a MoA rises in a market where it is traded for a separate MoE, then it’s possible that prices are sticky, the value of the MoE falls in MoA terms, and the reduced value of MoE buys less Y at the fixed P. That’s a recession – caused by less (nominally) MoE available to buy stuff. (How else could a general glut possibly happen?)
So MoA can’t cause NGDP to fall if all other prices are fixed. But if there is one flexible price – the MoA price of a separate MoE – then it can cause NGDP to fall, by reducing the value of MoE, whose flow is equal to the value of PY that can be sold. So with fixed P, Y falls – a recession. But you need that MoE to exist for that to happen, otherwise a fixed P means a fixed value of money. Without MoE, since goods aren’t generally traded for the MoA, even via an MoE, the value of money has nothing to do with Y.
I think you need to go back and read some elementary Keynesian economics. The flow of spending, the demand for outputs, a general lack of demand for outputs, output generally going unsold because there is no “demand” to buy them. They are talking about the nominal flow of MoE (even if they don’t realize it). It doesn’t make sense to say that a rise in the value of a pure MoA can pull down the amount of Y that gets sold, unless there’s also an MoE somewhere required to sell the Y. And if it’s not a general lack of demand for Y, then the only other way to get general unemployment is if P falls everywhere whilst wages are fixed, so all real wages rise above equilibrium. Those are the only ways of getting demand-side (money-spending side) unemployment.
You don’t “spend” a pure MoA. There’s no HPE for a pure MoA, people don’t all spend it in order to get other real goods they really want. The MoA could have intrinsic value. The MoA could be blue Stilton cheese, which has no HPE. It has intrinsic value to a few, and numeraire-utility to the rest. People don’t all hold inventories of Blue Stilton Cheese to hot-potato around. Potatoes are not subject to HPE.
The Equation of Exchange is an equation describing the relation between real output and the intermediating commodity needed to sell/exchange it all, money. Money flows round several times a year, you multiply that velocity by the size and value of the stock to get total purchasing power. This purchasing power is the total means for buying Y. The ratio between MV, the flow of money, and Y- is P: the exchange rate between goods in general and money. The MoA controls the nominal value of the MoE stock (usually it is the MoE stock), which * by its velocity is total nominal purchasing power MV. Divide that by the total output it has to buy to get the equilibrium price level; divide MV by the output it actually does buy to get actual P.
I give up, any more and I’ll have a nervous breakdown. Maybe I’m not really cut out for this economics-punditry thing…
29. October 2012 at 23:30
Suvy, it’s not that complicated. Modeling V is what the HPE is for. V = 1/k. People want to hold a larger fraction of their nominal incomes as an inventory of money-balances (for making exchanges). But the money stock is fixed by the central bank, so people can’t hold more money in aggregate. Individuals reduce the outflow from their existing inventories, buying less Y (actually less everything, as Nick Rowe points out). But because everyone else does the same, their inflow falls too. Eventually they realize that the inflow drops represent fundamental drops in their nominal income. And eventually the nominal income drop is large enough that they are satisfied with their existing stocks of money-inventory. They still want more money as a fraction of nominal income, but the latter adjusts instead of the former. The higher ratio of desired money balances to nominal income is k. If k doubles, V halves. If V halves, it means each dollar changes hands half as many times in the year.
29. October 2012 at 23:33
From the post:
Give me control of the stock of currency, and can drive the nominal economy and also impact the business cycle.
You could, but in so far as you caused fluctuations in real output you wouldn’t understand why.
29. October 2012 at 23:35
Actually, New Classical fluctuations could still happen without a medium of exchange, just not Keynesian ones.
30. October 2012 at 00:29
Just thinking about Hurricane Sandy – the Keynesian multiplier effect from the repairs (which won’t occur, because of the Sumner Critique) is all about the flow of the medium of exchange. In a barter economy with an obscure market in “US Dollar bills” used indirectly to price everything else, there is no way you could ever have a multiplier effect on anything else.
30. October 2012 at 03:23
No, no, no. I hope that Nick will weight in, but I will try to make as good response as I can get on my own. Money is both – medium of exchange and medium of account.
So the real question is – which of these two definitions of the money is important for explaining recessions? Some people like Richard Koo emphasize the medium of account side: its impact on balance sheet etc.
And some people like Nick Rowe emphasize medium of exchange side of money and its importance impact of money on recession. Why? Because only the medium of exchange has one peculiar property – it is the only thing in economy that you can “save” (or hoard) not only by “buying it” but also by refusing to buy. In Zimbabwe you cannot get more “gold” by for example refusing to buy a car today. You can only get more Zimbabwe dollars for that.
If you want to hoard medium of account – gold in your case, you need to go out and buy it from goldsmith using your Zimbabwe dollars. It is quite possible that no goldsmith will have any gold on stock so your attempt to hoard medium of account will be thwarted. But since Medium of exchange is what ultimately you get as your income, you can always save more of it.
This medium of exchange aspect of money, impact of money hoarding on validity of Say’s law and everything else is repeated ad nausea in Market Monetarist blogs (it is for sure repeated by Nick).
30. October 2012 at 03:35
PS1: Money is is also a store of value, but it is just a residuum of money being medium of exchange. Medium of exchange needs to return some of its value so that you can carry it in the wallet until you spend it.
PS2: Also, the main point of my response is not that anything you write is wrong – as long as the purpose of this thread was to find out metaphysical meaning of “money”. It is much easier to stick to basics and take evaluate what aspect of money is good for explanation of some interesting phenomena – like recessions. And in that area I am with Nick completely. It is the medium of exchange role of money that is important, it is what lies in the core of the recession.
30. October 2012 at 04:02
Scott:
This is my area of expertise. I realize it is a bit arcane.
The medium of account is not necessarily money.
The equation of exchange does not apply to a nonmonetary medium of account.
The price level is the inverse of the relative price of the medium of account. The relative price of a nommonetary medium of account depends on its flow supply and flow demand.
Now, you could define velocity to be the ratio of current inventories of the medium of account and the nominal value of output. While I find the equation of exchange to be of doubtful value at any time, that would be truly pointless.
All of the thought experiements where the quantity of the medium of account falls in half are silly. Suppose the medium of account is lettuce. Suppose it is electricity.
Even if it is steel ingots, as suggested by Fama, it isn’t like the current inventory of steel is significant relative to the flow supply and demand.
Most of my thinking about this issue never involved a single good as medium of account, but rather a bundle of goods. The dollar is defined as a fixed amount of steel, electicity, wheat, aluminium, plywood, and more. The quantity of the medium of account falls in half? What?
I have given the Zimbabwe-type thought experimennt a good bit of thought. Your error is to focus too much on equilibrium.
So, Asians demand more gold. The price of gold is fixed in Zimbabwe by definition. All the gold leaves Zimbabwe. When those needing gold to make jewelry or for dental work seek gold,they cannot find any gold.
How does this cause anyone to lower their quoted gold prices for bread, gasoline, diamonds, or anything else?
There is a shortage of gold in the gold shops. So what?
Realistically, the gold shop keepers just raise their gold price of gold and no one else does anything with their gold prices. Gold-units become an abstract unit of account not connected to anything. Or, there is some institution that ties gold to the medium of exchange.
Saturos keeps on referring to a Walrasian Auctioneer and barter. Forget barter. If there is a Walrasian Auctioneer, then he responds to a shortage of gold by calling out lower prices for everything else. That is how the gold prices for everything else fall. The Walrasian auctioneer lowers them. If there is money, then the Walrasian auctioneer can adjust the gold price of the fiat currency so that its nominal quantity equals the nominal demand to hold it (in gold units.)
But there is no Walrasian auctioneer.
How exactly does the gold price of fiat currency change? Every cash register posts the price. Where does this price come from exactly?
If there is an excess supply of that currency, and people start spending it, then the demand for all of the goods and services in the shops rise. Why don’t those selling those goods raise their gold prices? That is what they usually do when sales are strong.
How exactly does this translate in to a lower gold price for the fiat currency?
Those of us who argue that the problem is about the medium of exchange never say that the problem is necessarily a decrease in the quantity of currency.
What an odd claim.
The argument is that the problem is an excess demand for all forms of money, including both currency and checkable deposits.
Your ancedote is worthless. You personally never had trouble finding currency in a recession?
Currency has nothing to do with it. Further, unemployed people have a good bit of trouble obtaining money during a recession.
Because money is medium of exchange, it is easy to obtain more money. You spend less of it. (Or, if you are wealthy, you sell some of your other assets for it. That has been your personal situation for some time, right?) Those who would have received the money don’t have it. But because their income is lower, they don’t want to hold as much money.
So, some people have all the money they want to hold (you.) And other people are unemployed and don’t want to hold money. This is what an excess demand for money does.
If the unemployed people were working and earning income, then they would want to hold money, and when their demand for money is added to that of the already employed people (you), there isn’t enough money to go around.
If there were a shortage of Zimbabwe currency, then those wanting to hold more currency would just spend less. Sales of all the goods in the economy would fall. Those selling the products produce less. They lay off workers. The unemployed people don’t want to hold as much money, and so the amount demanded falls to match the supply.
Those setting gold prices for assorted goods would lower them. Yes, this raises the relative price of gold and will tend to clear the gold market. (Presumably, people are spending less currency on gold in the gold shops too.)
Now, a Walrasian Auctioneer could fix the problem by raising the gold price of the currency. Suddenly, the currency everyone has would buy more goods in the shops. The problem is solved.
But there is no Walrasian auctioneer. What exactly causes the gold price of the currency to rise?
There is a simple answer, of course. I explained it. In the gold shop, they set prices in terms of the fiat currency. If you want, you can say they set the gold price of the currency. But if people stop coming in for gold for dental work, they lower their prices in terms of the currency. And if lots of people want gold jewelry, they raise their prices in terms of the currency. That determines the gold price of currency that everyone else uses.
If there is a shortage of gold, the currency price of gold rises, everyone else in the economy effectively raises their prices in currency. There are now surpluses of those goods, and so they lower their gold prices.
Or, you could say that the nominal quantity of currency falls (in gold units) and so people buy less at current gold prices. And gold prices fall.
If there is a surplus of currency, spending rises in the economy and everyone raises their gold prices, except in the gold market where they raise currency prices. This lowers the gold price of currency, and so everyone else in the economy effectively raises their currency prices more. Or, if your prefer, the nominal quantity of currency falls (in gold units.) Prices are now too high (there is a shortage of currency given the higher gold prices and lower gold price of currency) and so, gold prices go back down.
Now, I don’t deny that people would carefully watch the gold shops in such a system, and strong sales of gold might cause people to cut spending on other goods immediately and lower the gold prices of assorted goods. And, of course, if the currency price of gold is expected to rise, then those setting prices gold shops have the usual incentive to raise their prices in terms of currency immediately.
But without a Walrasian auctioneer, it is the relatinoship between the medium of exchange and the medium of account that determines the price level.
30. October 2012 at 04:29
Also, I don’t agree that unanticipated changes in the price level cause business cycle problems. I think changes in spending on output cause business cycle problems. I think that it is the failure of prices to adjust enough in response to changes in spending on output that lead to business cycle problems. In my view, the “unanticipated change in the price level” approach is wrongheaded, suggesting that the problem is changes in the desire to work and produce due to a change in the price level. I don’t think people are confused by changes in the price level. I think they just don’t change prices enough. The unanticipated change in the price level approach assumes continuous market clearing. I think that is a bad assumption.
30. October 2012 at 04:43
Statsguy, Good post, but I share Saturos’s confusion about the moogles examples. Are they the MOA during normal times?
Saturos, A few comments:
1. V in the equation MV = PY is NOT the number of times a dollar is spent. It is PY/M, nothing more. You do eventually spell out a model for V and k, which is fine. But the equation itself is merely a definition, not a model.
2. The question of real wages W/P is unrelated to anything we are talking about. If W is sticky and P is completely flexible, then a 10% fall in NGDP caused by a 10% negative monetary shock will NOT reduce P by 10%, because the MC of production has fallen by less than 10% (due to sticky wages.) Thus P might fall 4% and Y might fall 6%, or some other combination, even if P is 100% flexible.
3. If P and W fall by exactly the same abmount, that does not mean there is no fall in employment in NK models, as they generally assume wages are acyclical. In a monopolistic competition model P and W often fall at the same rate, but because NGDP falls even faster, the level of Y falls as well. This is all perfectly consistent with what I am saying, which is why I think the variable W/P is so misleading. You seem to think it has important implications, I think it has none. W/NGDP is what drives the business cycles.
My entire argument boils down to two propositions, each of which seem uncontroversial to me:
1. A huge increase in the demand for the medium of account, or a huge decrease in the supply of the medium of account, whill cause NGDP to fall sharply. Is that controverisal? Surely not. That’s true regardless of whether the MOE is the same as the MOA, or there is a different MOE, or even there is no MOE. The MOA rulz—it drives the nominal universe and all its components.
2. A big fall in the NGDP will cause lots of unemployment if wages are sticky, indeed even if wages and prices are equally sticky, and W/P doesn’t change. Obviously the latter case assumes monopolistic comp., as prices can’t be sticky in perfect comp. This unemployment occurs because nominal income/output falls more sharply than nominal hourly wages. That leads to fewer hours worked. In perfectly competitive industries this process will show up as higher W/P, while in monopolistically competitive industries it will show was as a rise in W/(P*Q). Obviously if you produce less, then aggregate expenditure will also fall.
When I said the demand for money was unit elastic, I must have meant currency, as the demand for gold is not unit elastic–I don’t recall the context of that statement. If I applied it to all MOA, I was wrong.
Sorry to drive you nuts. 🙂
BTW, Your Sandy comment reminds me that after the Japanese tsumani Tyler Cowen linked to some monthly output data that suggested we were going to have a “real recession”, not caused by monetary shocks. It never happened.
30. October 2012 at 04:44
Scott –
“Negation, Yes, debt is in gold terms. That’s actually in the post, you may have missed it.”
My apologies, you’re right, I missed it. (And, I actually read the post twice!)
I believe the best example of your point is the Great Depression. If in 1933, the courts had ruled that prior contracts had to be paid in based on the exchange rate in gold at the time, then we would have had your example. In that case, gold would have been the medium of account, and dollars would have been the medium of exchange. All debtors would have had to pay 69% more in dollars(35/20.67), we wouldn’t have had the huge recovery of the next 4 months that you’ve descibed in other posts.
Instead, the courts correctly ruled for dollars, so dollars become the medium of account and the medium of exchange and gold was removed from both roles – in practice, if not officially.
So, yes, it is the medium of account, not the medium of exchange that is important.
30. October 2012 at 05:08
Bill, You said:
“The equation of exchange does not apply to a nonmonetary medium of account.”
It applies to all assets, it’s an identity. It applies to toasters, to stocks, to bonds, to money, to nominal stocks of zinc, or anything else you want to plug in. V is NOT the average number of times a dollar is spent. It applies to M1 and M2 and any other definition of M. V is NGDP/M, nothing more.
It’s just a definition of V, nothing more. The equation is much more useful as a pedagogical device in this from:
M = kPY
You said;
“So, Asians demand more gold. The price of gold is fixed in Zimbabwe by definition. All the gold leaves Zimbabwe. When those needing gold to make jewelry or for dental work seek gold,they cannot find any gold.”
Of course the price of gold is fixed, that’s the definition of a gold standard. But an increase in the real value of gold doesn’t cause gold to leave the country, it causes deflation. Indeed in the early 1930s when the real value of gold rose the flow of gold was INTO the gold standard zone, and away from the non-gold standard countries. There is no gold “shortage” as there are no price controls fixing its real value. Price controls only cause shortages when they fix real prices. the gold standard fixes a nominal price (of the MOA.)
You asked:
“How exactly does the gold price of fiat currency change? Every cash register posts the price. Where does this price come from exactly?”
From the Z$/gold exchange market in the capital of Harare. The results are sent by cell phone to all the outlying communities.
You said;
“If there is an excess supply of that currency, and people start spending it, then the demand for all of the goods and services in the shops rise. Why don’t those selling those goods raise their gold prices? That is what they usually do when sales are strong.”
You are confusing goods prices, which are sticky, and forex prices, which are flexible. The Z$/gold price is a forex price, and hence more Z$s does not lead to more real spending in the short run, just a lower value of the Z$s. In any case, you can assume PPP determines the price of the commodities that Zimbabweans buy, whereas nominal wages are sticky.
You said;
“Your ancedote is worthless. You personally never had trouble finding currency in a recession?
Currency has nothing to do with it.”
Currency has everything to do with it. An increase in the dmeand for currency or a decrease in the supply is what causes the recession. But it doesn’t cause the recession by creating a “shortage” so that people buy less, it causes the recession by creating deflation and/or falling NGDP. The currency market is in equilibrium, the economy is not. People can get whatever personal currency stock they want, whereas in a gas shortage people simply can’t find gasoline at all. It’s a change in the equilibrium value of currency that causes the recession (assuming currency is the MOA.)
You said;
“Further, unemployed people have a good bit of trouble obtaining money during a recession.”
No they don’t!! They have trouble obtaining wealth, because they have no job. But they have no trouble turning non-monetary assets into currency.
Bill, I agree with the criticism in your second post, I was just using the more familiar wording. You are right, it is unanticipated NGDP shocks, but my point still stands. I don’t think you have addressed the core of my argument. You need to address the two core arguments I gave Saturos in the preceding comment. Both seem obviously correct. You guys haven’t shown that either argument is wrong, and you’ll need to in order to dissuade me.
30. October 2012 at 05:11
ssumner:
Filipe, In the early 1800s the medium of account in the US was gold or silver (I forget) and the MOE were banknotes that traded at a discount, and hence were not media of account.
I asked about the inverse relationship. MoE need not be a MoA, but (it seems to me) MoA needs to be MoE. I think in that era there were multiple banks that issued private notes, each one trading at a different discount. Thus (it seems to me), gold (or silver) would also be a MoE: I have a hard time imagining people would refuse payments in gold.
30. October 2012 at 05:12
Bill, I don’t think I was necessarily assuming a Walrasian auctioneer (I just wanted Scott to think about markets in the Walrasian sense, with no medium of exchange). I was actually talking about an individual barter system, no auctioneer. Something, gold, apples, just one commodity in the network of barter exchanges, is arbitrarily selected as the numeraire. The price of everything else is now imputed in terms of implicit exchange rates with this commodity. I argue that this is not money.
“without a Walrasian auctioneer, it is the relatinoship between the medium of exchange and the medium of account that determines the price level.”
Yep, I completely agree. (And with everything else in your comment.) I have tried to explain this to Scott. I gave him a number of different scenarios, over the past couple of days. I also described your scenario to him. No auctioneer. Currency is exchanged for gold in one market, and gold is the numeraire (this market determines the “gold price of currency”, or the nominal value of the money stock). And all the other markets are a monetary exchange economy like our own. I made all the same points to him. You went even further and explained how the supply and demand for money stocks reaches equilibrium, how unemployment reduces aggregate income and lowers aggregate demand for holding money balances.
I really hope he gets it this time. (All this time, I blithely assumed that Scott knew all this stuff already – how else could he confidently talk about the “hot potato effect” determining the price level?)
J.V. Dubois, for once I agree with you completely.
30. October 2012 at 05:17
Negation, I have a whole book coming out next year on gold as the MOA in the early 1930s, then the switch to currency. (BTW, “dollar” was the medium of account in both cases.)
One other point, if gold had been ruled the MOA for debts, currency would still have been the MOA for purchases of goods or services.
The decision was 5-4, and (just like with ObamaCare) one conservative switched into the majority.
30. October 2012 at 05:19
Scott, responding to you bit by bit –
“In a monopolistic competition model P and W often fall at the same rate, but because NGDP falls even faster, the level of Y falls as well.”
Yes. But NGDP is just P * Y, according to you. Bill and I assert that there will be no connection between real Y and the MoA, unless there is also a MoE somewhere. The NK models are all implicitly assuming a MoE. In other words they are fundamentally confused. Nick has expounded on this at great length on his blog. (So has Bill.) I can post the links.
30. October 2012 at 05:26
Filipe, The MOA is usually a MOE, but gold was not in the US between 1934 and 1968, indeed it was illegal. But it was still the MOA.
Saturos, You still haven’t answered my question; which of my 2 key assertions is wrong? Does the MOA not detemine NGDP? Or do NGDP shocks not cause business cycles when wages and prices are sticky?
The HPE explains how the price level and NGDP find a new equilibrium when there is an increase in the stock of the MOA. It doesn’t just apply to money, it also applies to commodities like zinc, regardless of whether they are a MOA/MOE or not. But with zinc it merely reduces the nominal price of zinc, and has no big impact on the price level.
30. October 2012 at 05:35
Saturos, I don’t follow. You are saying NK models are confused because they assume a MOE?
It isn’t just me that assumes NGDP is P*Y, everyone does.
You say there is no connection between MOA and Y. So that means what? There is no connection between MOA and NGDP? That’s obviously crazy. There is no connection between NGDP shocks and Y when wages are sticky? That’s obviously wrong too.
30. October 2012 at 05:49
“You are saying NK models are confused because they assume a MOE”
NK models are confused, in the Nick Rovian view (who knows more about them than I do) because they try to explain the demand-side determination of PY without modeling the MoE.
There is a connection between MoA and P. Insofar as NGDP is P*Y, there is also a connection to NGDP. There is no connection between the MoA and (real) Y in the absence of a MoE.
“There is no connection between NGDP shocks and Y when wages are sticky?”
Well “NGDP shocks” (P*Y shocks) and “fluctuations in Y when wages and prices are sticky” are obviously identical. Obviously that sounds crazy to you; it is crazy to deny a tautology.
What I’m saying is that there is no connection between the MoA and the realY part of NGDP, without a MoE. The market for MoA either clears, determining Y, or it doesn’t. Even if it doesn’t, and there is a glut of Y, that glut doesn’t matter for NGDP. Whereas a glut of Y in its exchange with the MoE would be subtracted from GDP, and also NGDP.
Scott, I know you want to say dollars are the medium of account, embodied in both $1 bills and 1/20.67 oz bags of gold. But there was a market in which gold was traded for currency dollars. That market had to be forced into equilibrium at the pegged price, by adjusting stocks of currency and gold. It is better to say that currency was the medium of account, and 1/20.67 oz gold was worth $1 as long as it traded for $1 of currency. But it was the stock of currency dollars, multiplied by the average number of times that each dollar was exchanged for final output, that determined the nominal value of output. That “average number of times” is the velocity. It is the number of times the currency stock as a whole, on average, exchanges for Y. Since the currency stock in equilibrium is kPY, each dollar in the stock of M will change hands to purchase some of Y, 1/k times on average. This isn’t just my view; the textbooks explain it this way too.
The flow exchange of currency (and other moneys) for gold is what determined the gold price. That flow had to be adjusted to keep the gold price pegged. So when the real demand or supply for gold changed, you had to reduce (or increase) the flow of currency, which reduced spending on output. This led to some combination of deflation and recession. This is equivalent to looking at how much output each 1/20.67 oz gold could buy – the “real value of gold”.
But without currency, if 1/20.67 oz gold is the MoA, then gold is traded directly for some or all of the final output Y. The value of gold fluctuates by changing the demand for gold in terms of Y, or by changing the demand for Y in terms of gold. With flex-P a change in these demands would cause inflation or recession. With fix-P an increase in the demand to buy gold with some Y causes a glut of that Y, and a shortage of Y. With all output prices fixed, the price of gold in terms of that Y is also fixed. Less output is “sold” for gold. But that does not matter, because GDP is the exchange of Ys for each other, not Y for gold (the MoA with no actual value). So gold as a MoA controls nominal GDP by changing P (if it’s flexible). It cannot control GDP by affecting Y, because Y does not trade for gold to be counted in GDP. Gold can only control the value of Y being sold by changing the value of the medium of exchange needed to buy it at whatever P.
I’m still reading the rest of your comments.
30. October 2012 at 05:49
So in short:
1920’s dollar = medium of exchange
1920’s gold = medium of account
2010’s dollar = medium of exchange AND medium of account
2010’s gold = just another commodity
Right?
30. October 2012 at 05:51
The market for MoA determines the average price of real Y. But even if there is a glut in that market, no one loses their jobs! Unless there is a medium of exchange. A glut of goods trying to be sold for the MoE (or its value in MoA terms) does cause unemployment.
30. October 2012 at 05:53
OK, and I just want to put this out there – an increase in zinc output will reduce the price of zinc, ceteris paribus. But there is no hot zinc potato effect! What madness!
Hot Zinc potatoes. Honestly.
30. October 2012 at 05:58
“The market for MoA either clears, determining Y, or it doesn’t”
That should have been “determining P”. I’m beginning to crack here,
30. October 2012 at 06:01
“With flex-P a change in these demands would cause inflation or recession”
Inflation or deflation, damnit!
30. October 2012 at 06:05
With regard to your hypothetical example of Zimbabwe where the ZWD is the medium of exchange and gold the unit of account . . .
It sounds very much to me like the situation in Russia where I was living in the early 90s when the USD was the unit of account and the RUB was the medium of exchange. The difference from your example in that real world case (and I a willing to guess, in others as well) is that not everything is equally well indexed in terms of the unit of account. For example, in the Russian case, salaries of foreign firms and some large private Russian firms were explicitly indexed to the USD on a payday-to-payday basis. That is the pure case of your Zimbabwe hypothetical. Prices of imported goods were explicitly indexed to the USD. (It was illegal to display prices in dollars, so stores displayed prices in “UE,” which meant “agreed units,” with 1UE=1USD.)
However, other things were less perfectly indexed. Those included stickier wages of government workers and small-business employees and prices of nontraded goods and services. They would sort of move together with the exchange rate between the unit of account and the medium of exchange, but imperfectly and jerkily.
Still other goods, for example, mass transit fares, remained fixed in RUB for long periods despite vast changes in their real value. (At one time the subway fare fell to something like $0.0001, but you still couldn’t get on without buying a token.)
What is my point? It is that in such a mixed system, it seems to me that the real economy is exposed both to shocks to RUR money and USD money, so the kind of examples you are given are not so clean, and it is not an either-or that one kind of money matters and the other does not.
30. October 2012 at 06:06
@ssumner:
Your example doesn’t work. If Z$ is the medium of exchange, that is what I get paid right? Well, that means that I am getting hit by the inflation between my receiving my Z$ and my having spent them. Since there is a higher cost of my holding Z$, my wage (measured against gold) must go up to allow me to maintain my standard of living. The printing of Z$ has clearly reduced the value of the medium of exchange and the value of the medium of account.
30. October 2012 at 06:09
“No they don’t!! They have trouble obtaining wealth, because they have no job. But they have no trouble turning non-monetary assets into currency.”
Their employers have a lot of trouble getting the money (MoE) to pay their wages. That’s why there’s a recession. If they paid them in other assets, at some point those assets would have to be converted into worker consumption, which requires money to solve coincidence of wants. So the income that the fired workers are missing out on is a demand for the money to pay it. The loss of revenues to firms which makes them fire workers and leave output unsold, is a lack of money flowing in. Otherwise it would make no sense to say there was insufficient demand for everything. It’s insufficient demand in terms of money, insufficient money available to trade for firm outputs. And lowering the price level allows a smaller flow of money to purchase all the output that wants to be sold, hence cures the recession.
30. October 2012 at 06:12
Ed Dolan agrees with me too. RUR is the MoE, shocks to its quantity and velocity affect PY. (They also change the purchasing power or “real value” of the MoA, USD.) Shocks to USD change the nominal value of RUR, and hence will also affect PY.
That is what Bill Woolsey would say too, I think.
30. October 2012 at 06:14
On second thought, I think Ed’s example has two distinct media of account, USD and RUR. So that makes price level determination really complicated… In fact there are two price levels, with sticky prices in one currency having to be converted into the other. Two price levels, depending on how you look at it, each with two separately stuck sets of prices…
30. October 2012 at 06:25
Recessions are a disequilibrium in the (consolidated) market with money (MoE) on one side, and other things in general on the other. That’s what a general glut is. The MoA-MoE market determines the nominal amount of MoE, and hence affects NGDP. But MoA with no MoE affects P but not consumed real Y.
So going back to your two uncontrovertial points – they are both true. And yet your conclusion (“MoA can reduce NGDP in the form of reduced real Y, even without any MoE”) is false. Can you see that now?
“Obviously if you produce less, then aggregate expenditure will also fall”
Here’s the problem. You think about “expenditure” like a conventional Keynesian. You are imagining Y being bartered for Y (“demand for Y in terms of Y”), without considering the flow of MoE intermediating the exchange. People aren’t just “producing less”, they are unable to sell everything they want to, as there isn’t enough MoE flow “lubricating” all desired transactions. It’s the drop in the flow of spending (demanding goods with money) that holds sales of Y back from its potential.
That’s it, I’m done.
30. October 2012 at 06:33
I think the adjustment, if you do change your mind, will be painful. Either you have to accept Nick, Bill and my way of thinking about “NGDP” as shorthand for M*V – the aggregate flow of MoE. Or you have to stop reifying NGDP – you have to think of it seperately as P * Y, as that is how the MoA affects the economy, when there is no MoE. There is no “NGDP” which the MoA first determines, then you look at whether fixed P is going to require you to take a hit to Y. That’s only true for MoE. For MoA, sure you could look at the amount of Y being “sold” on the market where MoA is traded for Y, and call a reduction in that Y (with fixed P) a reduction in “NGDP”. But that would have no bearing on unemployment.
It’s the MoE that determines demand-side unemployment. There is no “demand” for Y in general without the MoE. What would such “demand” for Y be in terms of, otherwise? The MoA? But a lack of demand (an excess supply of Y) in that situation has nothing to do with general consumption and employment.
Maybe I need to start drawing pictures…
30. October 2012 at 06:36
Scott, this might be one of those times when you’re the one who’s crazy.
Just a friendly suggestion from the Universe.
30. October 2012 at 06:38
Scott: Let’s establish, that medium of exchange is something called “Sumner”. We know that “Sumner” is used by police to set fines and that this price is quite sticky. We also know that BSA uses Sumner to calculate how much software gigants lost to software piracy. Please establish the price level.
Quite hard don’t you think? Unless you have some data about market prices of goods in terms of whatever medium of exchange is used, I believe that you will have tough time doing that. People using medium of exchange is what establishes the price system from which you can calculate relative prices from which you can create whatever “medium of account” you wish to construct.
The main point here is that you cannot get rid of medium of exchange and pretend that everything is medium of account and medium of exchange automatically clears on Forex or whatever. Why? Because people can hoard medium of exchange. They may refuse to use it and this has real impact – in real economy. It can change relative prices and thus even the value of medium of account.
If for instance the supply of Zimbabwean dollars is constant and people decide to hoard it, it means that price level in terms of Zimbabwean dollars will fall. One month you bring home $1000 and another only $500 – because everybody hoards. The more people hoard the more incentive everyone has to hoard money. It gets harder to sell things and easier to buy things – this is also known as “recession”. Nobody can prevent you from hoarding by buying less. I cannot imagine how this could not translate into deflation in gold terms.
Now imagine that people of Zimbabwe would start to hoard gold. Imagine that everyone gets their paycheck at midnight 15th day of the month. The moment they receive their income, they collectively try to buy more gold. But in order to increase their stock of gold, they first have to find someone willing to decrease his stock of gold. And that person will do that only if he wants to buy something for it (or if he wants to hold more money). It is impossible to have the same vicious cycle as in previous example.
30. October 2012 at 06:39
Sorry, the first sentence should be “Let’s establish, that medium of ACCOUNT is something called “Sumner””
30. October 2012 at 06:45
@ssumner:
Just thought of a more succinct and hopefully clearer way to put my point:
Somebody has to be holding on to the medium of exchange. (even if that is very many somebodies for a very short period of time each) Whoever is holding on to the medium of exchange will take into account the inflation to come and therefore demand to be compensated for it in whatever transaction will result in them holding the medium of exchange.
So let’s say I’m a baker with a listed price of 100 Z$ per loaf of bread which is also worth 1 gram of gold. The above policy is announced. The central bank will double the number of Z$ in circulation every day until the end of times. I now list my new price in gold. If you are right, I will list my price at 1 gram of gold accept 100 Z$ per loaf today, 200 Z$ per loaf tomorrow, 400 Z$ the day after and so on and so forth right? But if I accept 100 Z$ per loaf today, that means that tomorrow morning when I have time to go buy gold, that 100 Z$ will only buy me 1/2 gram of gold. So I can’t list my new price at 1 gram of gold. I have to list my new price at 2 grams of gold. And bam, just like that, the value of the medium of account has fallen.
The only way to avoid this is to not have anybody hold on to the medium of exchange ever. The central bank offers a service where they buy and sell gold at the prevailing market rate. Every transaction goes like this: We agree on a price in gold. I hand the central bank the gold price, they hand me some Z$, I give the Z$ to you and you immediately hand them back to the central bank in exchange for gold again before inflation can occur. But then, I would argue your medium of exchange is gold. You’re just pretending it’s Z$.
30. October 2012 at 06:53
Saturos,
The problem is that I don’t think that the money stock is fixed by the central bank. I’m a believer in endogenous money which means that I think the central bank can’t “fix” the money supply and that the banking system plays a huge role in determining the supply of money. I view the money supply as the sum of base money plus credit money–the base money is determined by the central bank, but the amount of credit money is endogenous(this is not to say that the central bank doesn’t have a lot of power over credit money).
What I’m primarily interested in is how changes in the money supply affect changes in V. For example, if the money supply increases rapidly; would that not change expectations of changes in the price level and have second/third order effects on V? So wouldn’t the impact on P and T be very different.
Basically, I’m just looking for dynamic models so that I can try to understand the qualitative behavior of the system better.
30. October 2012 at 07:23
Suvy, see:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/04/the-supply-of-money-is-demand-determined.html
and
http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/04/monetary-policy-is-just-one-damn-interest-rate-after-another.html
Expected inflation raises interest rates (Fisher effect) which lowers money demand and raises velocity (Cagan effect.)
30. October 2012 at 07:25
Yes, we just settled the debate.
SCOTT SUMNER IS THE MEDIUM OF ACCOUNT.
For realz.
30. October 2012 at 07:25
@Suvy see this one too http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/09/all-money-is-helicopter-money.html
30. October 2012 at 07:32
I will finish with a quotation from scripture:
http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/03/the-return-of-monetarism.html
30. October 2012 at 07:42
Update: guys, I just asked Major_Freedom the following questions:
30. October 2012 at 07:42
And now I forgot to close blockquotes, just like MF. Guess it’s contagious.
30. October 2012 at 07:46
Scott, Bill,
US gold standard will not persuade anyone, as MoE had a hard peg to MoA.
Ed’s Russia example is much more interesting, and it shows that both MoA (dollars) and MoE (rubles) are important.
Dollar shortage due to Asian crisis in 1997 has contributed to a 1998 crisis in Russia.
However, Russian authorities have tightened MoE too. Unemployment would have been lower if they had followed easy MoE policy in 1997-1998.
In the end I see no big difference between Scott’s MoA approach and Bill’s MoE approach.
30. October 2012 at 07:57
Saturos:
Why? What exactly is wrong with what I said? It’s one thing to bloviate whilst saying I’m wrong. It is quite another to show it. And no, telling me to go read some article or book won’t do. I would like you to show me why I am wrong, in your own words.
Let’s start with one statement that I made in my justification that seems wrong to you, and show me why it is wrong.
30. October 2012 at 07:58
Scott, you’ll want to weigh in on this: http://www.slate.com/blogs/moneybox/2012/10/30/legatum_institute_prosperity_index_shows_success_of_northern_european_welfare.html
also here:
http://www.thedailybeast.com/articles/2012/10/30/who-s-the-least-socialist-of-them-all.html
30. October 2012 at 07:59
123:
If I expected $X revenues last year, but I only received $(X – 10%), which then leads me to lay off some people, and cut back on production, then am I justified in saying that the Fed failed to ensure that my revenues were adequate? If not, why not?
30. October 2012 at 08:00
MF, suppose I’m a brilliant counterfeiter who immediately counterfeits piles of Treasury bills, equal to 10% of the outstanding stock.
I then try to sell them on the open market.
Do you seriously think the market price will go up?
Don’t Austrians believe in supply and demand?
30. October 2012 at 08:05
Saturos,
And now I forgot to close blockquotes, just like MF. Guess it’s contagious.
You’re funny. It’s like it’s my fault or something.
30. October 2012 at 08:11
Saturos:
Come on Saturos, my theory of interest is a theory of market rates of interest. Counterfeiting treasuries is not a market activity.
It would be like me critiquing your theory of interest by saying “Oh ya? You think the prices would fall? Well imagine I am a perfect counterfeiter of dollars, and I double or triple the nominal demand for treasuries after you added 10% to the supply. You claimed prices would fall! You are barking raving stark mad crazy etc etc etc…
30. October 2012 at 08:15
I ask you again Saturos, WHAT did I say in my premises in my explanation that you think is wrong? You keep asking me rhetorical questions that have nothing to do with my theory.
30. October 2012 at 08:35
Mf, suppose the government issues 10% more Treasuries. Answer the question! Your theory is supposed to predict events, isn’t it? You do know what a theory is? Are you now going to go crying that “theories can’t be proven true or false! This is Austrian economics, we don’t subscribe to falsification!”
30. October 2012 at 08:57
Saturos:
Easy Saturos, or you’ll blow a gasket.
My theory of interest rates is not a theory of the yield on government treasuries. It is, again, a theory of market interest. Do you know the difference?
To answer your question, if the government issued 10% more treasuries, then ceteris paribus, I would expect the prices to fall, but again, rarely are things ceteris paribus. For one thing, there would almost certainly be inflation alongside it, which can increase the nominal demand for treasuries and may end up offsetting, or even reversing, the fall in price due to the supply increase. Or, the nominal demand for treasuries may end up rising for non-inflation reasons.
Again, I ask you for the THIRD time, what exactly is wrong with my explanation? You keep demanding that I answer your red herring questions, but you don’t answer my justified, simple, reasonable question. You said I am wrong, so I ask you where in my explanation I made a mistake.
As for the predictions, not all valid theories concern predictions. I may know the theory of supply and demand, but that does not mean that I can predict what the demand for sugar will be next year, using some scientific formulas that presuppose humans are robots who don’t learn and adapt.
Can you please answer my question of what I said that is wrong? What you are saying is not helping.
30. October 2012 at 09:02
Scott:
So your example only applies if there is a world gold standard, but people in one country use a currency that is not tied to gold, and has a flexible exchange rate.
This is a small open economy so that exports are sold for gold (and then the gold is sold for currency.) And imports must be paid for with gold, (which is purchased with currency.)
The currency to gold transactions are not being made because gold is useful for dental work and jewelry but rather because it is international money.
Anyway, this is entirely different from what I had in mind.
I am not all that interested in a small open economy framework where gold prices for imported good can just be set based upon world money prices and export prices necessarily are world money prices. (Well, I have just written four posts on Coffeeland, which is a small open economy on a gold standard, so I guess it depends.)
Let’s expand this to the entire world. Gold is just a medium of account for the entire world, and every country has a fiat currency. Foreign exchange transactions determine the exchange rates between the fiat currencies, not between a world gold money and a fiat currency.
Or consider a situation where Zimbabwe uses gold as the medium of account (or lettuce,) and foreign exchange markets are between zimbabwe dollars and U.S. dollars, Euros and whatnot.
Then you could use the U.S. dollar price of gold and the U.S. dollar-Zimbabwe dollar exchange rate to find gold price of Zimbabwe dollars.
But, then, you are limited to a small open economy example. Still, I would think that the way that higher world prices of gold would be translated into lower prices in Zimbabwe would be a drop in the nominal quantity of Zimbabwe dollars. The US dollar price of gold rises, the U.S.-Zimabwe exchange rate is the same, and so the gold price of Zimabwe dollars fall. The nominal quantity of Zimbabwe dollars falls in gold units. Spending falls on Zimbabwe goods.
But, still, I am more intersted in situations where we aren’t using foreign price of the medium of account and they are determined internally.
By the way, you can devalue or revalue the medium of account.
People quote prices in terms of dollars. The dollar is defined as an 1/20 of an ounce of gold. You make it 1/35 of an ounce of gold. It is devalued even if gold is not used as money.
Just use an example of something that is cannot reasonably be used as a medium of exchange. Use electricity.
The dollar is 5 kw of electricity. Then you make it 6.
30. October 2012 at 10:06
Scott: Wow! What a comment thread! (And great to see Ed Dolan’s Russian example).
As you guessed, I disagree. Rather than write a very long comment, I did a post, stating my view. http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/10/medium-of-account-vs-medium-of-exchange.html
30. October 2012 at 10:19
Why did you have to complicate things by suggesting gold as a unit of account in Zimbabwe? In actual fact, in the ZWD era, US dollar was effectively the unit of account. Now it is offically the currency.
Does US monitary policy influence Zimbabwe’s economy?
When a country as a separte unit of account from medium of exchange, is it possible to print ones way out of a debt problem? The Treasury prints a lot of money, lowering the wages and government expenses in real terms and increasing tax revenue in nominal terms. This may bring the budget from deficit to surplus. But the debts of the country are still in hard currency, and the cost of debt service rises.
30. October 2012 at 10:51
Ed Dolan’s Russian example fits with my personal experience in the 1998-2002 recession when barter soared [1]. It seems to me that the MoE function of money is quite important.
Or it may be an artifact of bi-monetary economies. Lots of prices in Argentina are quoted in USD
1- http://www.taoaproject.org/le-projet-taoa-es/barter-clubs-in-argentina/?lang=en
30. October 2012 at 11:56
Ed Dolan, I agree, but my thought experiment involved a clean dichotomy–it’s the only way to really get to the heart of which role is the key attribute of money.
30. October 2012 at 12:02
[…] mean by “M.” If M is the medium of account, but not the medium of exchange (as in my Zimbabwe example) then the model works fine. If M is the MOE but not the MOA, then it doesn’t work at […]
30. October 2012 at 16:10
Scott, you never explicitly define the medium of account. But you note that:
“Tell the public that they must continue to set all wages and prices and debt contracts in terms of grams of gold. At the cash register there will be a sheet of paper listing today’s exchange rate between paper Zimbabwe dollars and grams of gold”
… so I’m going to assume that the medium of account is whatever item that referred to when posting prices, debts, etc.
You also say:
“The MOA is usually a MOE, but gold was not in the US between 1934 and 1968, indeed it was illegal. But it was still the MOA.”
But surely prices in the US were not quoted in gold between 1934 and 1968. See this 1938 comic book… the price is in cents, not grams of gold. So gold couldn’t have been the MOA.
Anyways, this is tangential to your main argument. But since examples help us readers to drill down into the core of your argument, I’d like some clarity on the 1934-1968 period. Or on the definition of MOA.
30. October 2012 at 16:39
“See this 1938 comic book”
Forgot the link.
30. October 2012 at 19:04
Just like to add my $0.02. If we accept that a medium of exchange can be anything that two individuals wish to recognize as such, and we accept that a medium of account must have an arbitrator or legal backing, I think we can say that:
Inflation cannot reduce the value of the medium of account.
Inflation can reduce the value of the medium of exchange.
A legal authority that does not buy from or sell into markets can adjust the value of the medium of account to always be equal to the same basket of goods. Settlement of legal claims between buyer and seller is resolved using said medium of account.
In a legal claim, either the buyer or seller has been shorted on the value of what he / she has received. And so a legal authority has the ability to equalize the trade agreement between both parties. He does this by demanding payment in the medium of account from the party at fault to be delivered to the party that has been harmed. Net value has not been added or subtracted from the economy, it has simply been transferred from one party to another.
30. October 2012 at 19:26
Suvy,
“What I’m primarily interested in is how changes in the money supply affect changes in V. For example, if the money supply increases rapidly; would that not change expectations of changes in the price level and have second/third order effects on V? So wouldn’t the impact on P and T be very different.
Basically, I’m just looking for dynamic models so that I can try to understand the qualitative behavior of the system better.”
Here it is in a nutshell, accepting that all money begins as a debt we write:
DV = PQ
The first item is really all means of paying for goods represented by PQ.
DV = Income * (1 – Liquidity Preference) + Change In Debt w/ Respect to Time
DV = I * (1 – LP) + dD/dt
Income is the sum of income from the sale of goods + interest income from the holdings of debt
Income = PQ + Interest Rate * Debt
DV = (PQ + INT * D) * (1 – LP) + dD/dt
But we know that PQ = DV and so:
DV = D * ( V + INT ) * (1 – LP) + dD/dt
0 = -D * V * LP + D * INT * (1 – LP) + dD/dt
Letting debt be a function of credit demand f(t)
D = exp ( f(t) )
dD/dt = f'(t) * exp ( f(t) )
V * LP = INT * (1 – LP) + f'(t)
V = [ INT * (1 – LP) + f'(t) ] / LP
And so the velocity of money increases with higher interest rates, lower liquidity preference, and accelerating credit demand.
Notice that even if credit demand increases at a constant rate ( f(t) = C * t ), velocity will remain stable. It takes accelerating credit demand to increase the velocity of money.
“For example, if the money supply increases rapidly”
It is not a question of how quickly credit increases. It is a question of how quickly the increase in credit occurs (2nd order effect) that causes changes in velocity.
Just an FYI, the interest rate that I am referring to is the nominal interest rate.
30. October 2012 at 22:39
JP, because the exchange rate between gold and the $1 bill was fixed, Scott can claim that 1/20.67 oz gold was also the medium of account…
31. October 2012 at 04:51
So by that definition, the medium of account in China right now is USD?
31. October 2012 at 06:55
Scott:
“Saturos and Bill Woolsey (and I think Nick Rowe) believe that recessions occur when there is a shortage of the medium of exchange. I disagree. I never have any problem getting cash, even during recession years. A lack of cash never causes me to spend less.”
I could not agree more, and thanks for saying it. That notion has always seemed nonsensical to me, no matter how hard I think about it — as have many of the related “demand for money” arguments.
If this post represents your revealed preferences, you like I are feeling the need (preference) for a coherent or at least agreed-upon defintion of money. (Specification of three different uses for the term [or the “thing”] does not constitute a definition.) It’s rather remarkable that economists have not come to agreement on this seemingly rather crucial item as yet.
I’d like to propose:
Money: Exchange value as embodied in financial assets.
Financial assets (even dollar bills) are not “money.” (Vernacular nothwithstanding.) They are embodiments of exchange value. That exchange value, as so embodied in financial assets, is “money.” Money does not, cannot exist, except as embodied in financial assets. (Though exchange value can, of course, also be embodied in real assets. But when you have a bushel of apples that you could exchange if you wanted to, you don’t call it money.)
Financial assets: embodiments of exchange value that 1. cannot be consumed to provide human utility (or employed to produce things that can be consumed to provide human utility), and 2. have no (or miniscule) production inputs relative to the exchange value embodied. Also, they do not physically decay. (*Holding* financial assets can [does] provide human utility, but that’s a tangent I’ll address elsewhere.)
As soon as the match hits the end of a cigarette in a POW camp, that cigarette is no longer a financial asset (it’s a real asset), and it no longer embodies “money.”
Financial assets can have many different characteristics, can embody many different types and orders of claims or rights (and limitations on same). I won’t get into that here except one characteristic, below.
Medium of exchange: there are many, especially in the financial markets, but in exchanges for real goods the most common are physical pieces of currency and bank-account credits (denominated in some currency unit). The exchange value embodied in most other mediums of exchange must generally be exchanged into these mediums before exchanging it for real goods.
Medium of account: the currency *unit* (dollar, euro, etc.). Not to be confused, conceptually, with a piece of physical currency like a dollar bill — which is just a physical token representing a credit tally (i.e. a a positive balance in a bank account).
This makes the notion of “demand for the medium of account” sort of meaningless. It’s like saying there’s a demand for inches, or degrees Farenheit.
I’d like to suggest that “demand for money” (unit of account *or* exchange) is a misnomer, a conceptual nonstarter, or at least ambiguous (as you demonstrate here).
Rather, there’s a demand for financial assets that have a characteristic that I’ll call “nominal stability.” Their exchange value changes little (or none) *relative to the unit of account.* A dollar bill, obviously, has 100% nominal stability relative to its unit of account (the currency unit called a “dollar”).
I’m going to leave this here, except to suggest the notion (which I have yet to fully think through) that inflation is (or is caused by??) decreased demand for nominal stability.
I don’t know if this helps. But it’s conceptually much clearer and less self-contradictory/muddled to me.
31. October 2012 at 12:49
[…] Scott Sumner (who got the ball rolling) […]
1. November 2012 at 06:26
Steve Roth, It would be helpful if you’d use terms like medium of account and unit of account as others use them. Otherwise the discussion gets hopelessly confused.
1. November 2012 at 06:28
JP, If China had a fixed exchange rate regime (it doesn’t quite) then they would have two MOAs. China, HK and Argentina all had deflation in the late 1990s due to their use of US dollars as a MOA. Those countries that didn’t use the dollar typically did not (excluding Japan.)
1. November 2012 at 06:31
Lucas, I agree that the MOE is “quite important” in the sense that if you don’t have one, or have a shortage, you have big problems.
1. November 2012 at 06:51
Saturos, You said;
“Their employers have a lot of trouble getting the money (MoE) to pay their wages. That’s why there’s a recession.”
You are confusing money and wealth. Suppose Boeing wants to sell an airplane for $100 million. The company buying it has no “money.” So they offer Boeing $100 million in T-bonds or $100 million in Apple stock or $100 million in gold bars. I claim that Boeing would say “yes” and absorb the 1% cost of converting that asset into cash (just as stores accept losses on credit cards.) Then Boeing has no problem getting the cash by selling that asset. (Remember the banks of full of ERs, they’ll give Boeing some anytime Boeing asks for it.) So the key problem in a recession is not that companies can’t get enough “money” (as a MOE) to pay wages, but rather that they can’t get enough nominal wealth for their goods, and thus can’t employ as many workers at the going nominal wage rate.
Boeing’s problem is not just that it can’t sell goods for money, it’s that it can’t sell goods for highly liquid non-moneys that could easily be turned into money. And that’s because NGDP fell relative to wages, so Boeing can’t earn enough revenue in dollar terms to employ it’s normal workforce.
Of course normally currency is the MOA. So a reduction in the supply of currency does raise its value and reduce NGDP. So the (AD) problem certain does exit in the “money market” in the real world, which is why you and I and Nick never seem to disagree about any substantive issue, just how many angels can dance on the head of a pin.
1. November 2012 at 08:15
ssumner:
You are confusing money and wealth. Suppose Boeing wants to sell an airplane for $100 million. The company buying it has no “money.” So they offer Boeing $100 million in T-bonds or $100 million in Apple stock or $100 million in gold bars.
Why wouldn’t the buyer just convert the T-bonds, Apple stock, or gold bars to fed notes first? You say they are cash strapped, but they have assets that can be sold for cash. If Boeing can easily sell those assets for cash, then so can the buyer.
I think what Saturos is saying is that in a recession, cash in general is harder to come by, so the assumption that Boeing can easily sell the T-bonds, or Apple stock, or gold, for $100 billion in cash, is an assumption that contradicts what he is saying.
So the key problem in a recession is not that companies can’t get enough “money” (as a MOE) to pay wages, but rather that they can’t get enough nominal wealth for their goods, and thus can’t employ as many workers at the going nominal wage rate.
Nominal wealth? Do you mean money? Or real goods? “Nominal” usually means money, and “wealth” apart from money usually means non-money goods, like stocks, bonds, factories and machines.
The way you phrased it suggests that nominal wealth is indeed non-money goods, because you said “rather”. Well, that means you are saying the key problem in a recession is that companies cannot get enough non-money goods in exchange for their own goods. But that is a barter economy, not a monetary economy where sellers look to get enough MoE, so that they can satisfy their MoE expenses like wages and hopefully earn MoE profits.
Can’t you see how your insistence that money is primarily a MoA, rather than a MoE, is leading you astray?
1. November 2012 at 11:03
Scott, I’ll restate what I said. Of course employers can always liquidate their assets to pay wages. That’s not what I meant.
I want you to visualize the circular flow of income. When I say money, I mean an inflow of money. I mean money income.
The circular flow increases wealth/income at every stage. (It’s called the circular flow of income for a reason.)
At no point in the circular flow of income do employees sell assets to pay wages. (That’s what happens when inventory goes unsold, not when workers are actually earning their wages via bosses that employ them, by producing things that do get sold in exchange for money.)
When I say employers are having trouble getting money to pay wages, I mean they are having trouble earning money to pay wages. And it is money that they have to earn.
In a barter economy, firms could earn apples in exchange for cars, or financial claims on apples. They could exchange that for whatever asset the workers demanded to be paid in.
But we don’t live in a barter economy. We live in a monetary economy. There is a coincidence of wants problem, which money exists to solve. And when it fails to do so, we have recessions.
(Cue Randian diatribe: You have said that money was the root of all evil… [but gold would not have “fettered” anybody, if it had not been their lifeblood. Currency/moneyliquid assets were needed to exchange outputs for each other. It was bloodloss, not chains, that paralyzed the economy. Money was not a villain, it was an absence of a good thing that hurt us.)
People pathetically try to make up for the shortage of monetary flows by bartering. But barter sucks. So it doesn’t work very much.
When I say that employers don’t have money to pay workers with, I mean that workers are selling a service (labor) and to solve the coincidence of wants they want to be paid in MoE. But MoE doesn’t solve the coincidence of wants, unless the people you get it from also had to earn it, by producing something themselves, and selling that to somebody else. And when they did that, they faced a coincidence of wants problem themselves.
Money just has to keep on flowing, continually rescuing agents from coincidence of wants problems as it finds them.
Workers can’t barter with their employers, any more than employers can barter with consumers. Money is needed to pay workers. And that money has to be earnt by employers taking the labor they were given to produce stuff. And sell it. For money.
As I said, the root problem is that you don’t visualize as much as Nick and I do. (And when you do, your visualizations don’t correspond clearly enough with reality.)
If the picture of the economy in your head were a good one, it would consist largely of a million different markets for output. In each market there is a flow of newly produced output on one side, and a flow of money on the other. Flow of money, flow of income, flow of funds. You can almost literally go out there and observe this in reality.
Goods aren’t bartered for each other. They are sold for a medium of exchange, when they are sold to earn new income. That medium of exchange is used to purchase goods from somebody else. Which earns them new income.
1. November 2012 at 11:12
MF and Scott:
All markets are money markets.
Few barter transactions exist, compared to money transactions.
When you want to buy something, and you have something to sell for it, you sell your stuff for money. Then you pay for the thing you want to buy with money.
The money doesn’t sit around. It keeps on flowing. It’s entirely possible that a total money stock of $10000 could be enough to sell everything in a trillion-dollar economy. If it kept on flowing. (You’d need more than that realistically to actually solve coincidence of wants, though.)
Workers want money for their labor, not food stamps. They want to be paid in MoE so they can buy whatever they want. They sell labor for money.
Firms want money for the goods they sell, not apples or toilet paper. They want to ultimately give money to the workers and shareholders invested in them, in exchange for labor and capital received.
To keep on receiving new flows of money, workers have to keep on providing new flows of labor. To keep on receiving new flows of labor, firms keep on having to provide new flows of money.
To keep on receiving new flows of money, firms keep on having to provide new flows of goods. To keep on receiving new flows of goods, consumers keep on having to provide new flows of money.
Was there not a picture of the circular flow of income on your high-school classroom wall? They are worth a thousand words, you know (though I must have used several thousand by now.)
1. November 2012 at 11:18
Suppose Boeing wants to sell an airplane for $100 million. The company buying it has no “money.” So they offer Boeing $100 million in T-bonds or $100 million in Apple stock or $100 million in gold bars.
If a certain set of highly liquid non-moneys could always be used to intermediate a transaction, they wouldn’t be non-moneys. They would be money.
And if people were always willing to accept whatever asset in exchange for the outputs they sold, then there wouldn’t be a coincidence of wants problem. Or there would be a coincidence of wants problem, but people would be solving it by bartering instead of money.
I can’t believe I’m hearing this from the economist who famously said: “When I go shopping at Walmart I don’t take T-bills with me”
So as long as there are enough liquid non-monetary assets around, people have enough resources to demand as much output as their production-income allows them to? So given no fall in underlying productive capacity, there will never be a recession, provided that there are enough T-bills around? But of course, now you will retort:
And that’s because NGDP fell relative to wages, so Boeing can’t earn enough revenue in dollar terms to employ it’s normal workforce.”
Scott, you’ve already agreed to set aside problems caused by prices falling more than wages (i.e. rising real wages). How can NGDP possibly fall by more than the wage-and-price level? How can the MoA go beyond affecting NGDP via P, and affect it via Y? Since I now know you in fact do think like Tyler Cowen, so that NGDP literally means P * Y, not the volume of MoE flows M*V (of course they are always equal, but they mean different things), I have to ask you his question: how does NGDP have any causal force on output? Of course output falls when NGDP falls; once prices have adjusted all they can, NGDP is output. It’s a tautology. How does NGDP cause anything? And how does the MoA determine NGDP, beyond determining prices.
In fact, I’ll press this point further. How is a MoA supposed to determine P * Y ? What would even make you think that it did? Sure, I can understand why it determines P – because it is the medium of account, so the price level is measured as the rate at which everything else exchanges with it.
But how does it determine P and Y? How does it determine P times Y?
I can see how, if there is also a MoE in the economy. Then the MoA determines the nominal value of the MoE. And the MoE can determine Y, because Y only gets sold if it can be exchanged for the MoE.
How can there possibly be a general glut of all newly produced output without a shortage of MoE flows? Why wouldn’t they just barter?
So the (AD) problem certain does exit in the “money market” in the real world, which is why you and I and Nick never seem to disagree about any substantive issue, just how many angels can dance on the head of a pin.”
Scott, I do disagree with you on a very substantive issue. There is no literal “money market”. I mean literally money, not short term credit. The medium of exchange appears on every market. That’s why it’s the medium of exchange. Read the Nick Rowe posts I linked to.
Here’s another way of putting my disagreement. Suppose there is no MoA. I say there will still be recessions, if the exchange rate between the MoE and goods-in-general is sticky, when the MoE stock falls. Even if that exchange rate is not the “price level”, because there is no common unit or commodity for measuring all prices. People could go on quoting relative exchange rates between the component items of GDP, like the foreign exchange market. But if all exchanges were intermediated by a single good, money; the stock of that money fell; and the average exchange rate between that intermediating good and all other goods remained fixed – then there would be a general glut of all non-money goods: a recession.
If it’s this hard even to convince Scott Sumner of this stuff, I can’t imagine what it would be like to try and persuade a conventional Keynesian. (Oh, wait, I can…)
1. November 2012 at 11:18
Scott, I will simply conclude by noting that you are still failing to adequately distinguish between monetary stocks and flows. Just like every other unenlightened economist out there.
Worse still, I think you have been commiting a fallacy which is not entirely unrelated to the one you regularly lambast conventional economists for. Just like you scold them for looking at the symptoms of changing monetary policy, like interest rate changes, instead of the supply and demand for money, and instead of the actual nominal variables which policy is all about – so too are you making such a mistake.
Your focus on NGDP has been the mere symptom of a mismatch between the flow of MoE (M*V), and the average price at which GDP is traded for MoE (P). Such a disconnect will cause Y to fall.
But because the MoE is also the MoA, or nominally valued by the MoA (like the French and the Sun), – MoE fluctuations correspond to MoA fluctuations. And the change in the exchange rate between the MoE and Y-in-general, is called a change in the “nominal price level” (because it so happens that prices are denominated in the MoE). And once the price level gets stuck, further falls in the MoE stock (or increases in the demand to hold it), which reduce the amount of output sold, also correspond to falls in the MoA stock.
And so you say, “The MoA controls NGDP. Recessions happen when NGDP, controlled by MoA, falls – and prices do not absorb the whole decline. So output has to.”
Can you see that it is you who are looking through “the wrong end of the telescope” this time?
Again, what is the mechanism by which a change in the MoA stock affects real gdp? You say, it reduces NGDP, which is P*Y, so Y falls when P doesn’t.
But that’s just rearranging the question. Why does Y fall? How does the MoA affect the Y part of NGDP? Why should people lose their jobs when the demand for MoA rises? So what if prices stop falling, and there is still excess demand for MoA. How does that remaining excess demand for MoA put people out of work? How does that remaining excess demand affect Y, when it no longer affects P?
What else do I need to do to convince you?
But I’m still hoping for a Kuhnian breakthrough with you here. After all, even Nick Rowe needed to have one, once upon a time…
http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/04/living-in-a-demand-side-world.html
1. November 2012 at 11:34
“2. Unanticipated inflation helps borrowers and hurts lenders.”
Nope! What if prices go up and wages don’t?
1. November 2012 at 12:52
Saturos:
Your understanding of money is rather crude. You seem to believe that the importance of money is that it provides a “flow” sort of benefit, where as long as money “flows” in a necessarily subjective way, then that which is bringing about this particular subjective “flow”, then “production” will somehow take care of itself (assuming “good regulations”).
But money is not significant as a globular flow of stuff. It is significant as a tool of economic calculation. Relative prices and its influence on relative resource and labor allocation are where division of labor economies are made or broken.
You have to look at money as a calculation tool, not as a flow of blood, or gasoline, or oil that “greases” the machine.
It helps to think of money as stationary, and the relative ownership titles to the money “flowing” instead. If the relative ownership titles results in relative productions that cause problems, than altering the supply of stationary money won’t solve the issue. It would be like trying to solve the problem of distances by changing the yardstick hatch marks.
————–
Your reference to Kuhn is a little off, because his theory is of scientific revolutions, not the changing of one’s person’s mind. If you do mean to refer to some general revolution in science, then how do you know what you are saying is even well characterized by an existing scientific revolution? We’re still in a positivist, empirical, hypothesis testing structure, where the only relevant economic information is historically observable data.
At any rate, Kuhn’s theory isn’t as knock down as it first seems. He showed that when scientific theories are divorced from action constraints, and roam unbounded in verbal conventions, where one theory cannot conclusively refute another, and has thus given the impression that throughout the ages, one dominant scientific paradigm has given way to another, then the old doctrine of relativism seems to ground even science.
But whatever Kuhn has shown about the relativistic tendencies of scientific paradigms, what he has not shown is that one paradigm is just as practically valid as any other. While scientific theories may be incommensurable when they remain merely verbal conventions, practically they can never be.
————–
NGDP is not output. NGDP is the total money expenditures that purchased output over a given period of time.
1. November 2012 at 17:41
Picture an economy with two mediums:
1. Medium of exchange (M1) – Medium that is exchanged for goods and services.
2. Medium of account (M2) – Medium that is issued as an asset to counterbalance liabilities (tax, debt, etc.).
Where a medium of exchange has an implicit velocity:
M1 * V = P * Q
A medium of account has a rate of issuance that is determined by the issuing authority. For instance that authority could issue said medium at a rate equal to:
dM2 / dt = P * Q
Solving for V:
We let M * V = Income * (1 – Liquidity preference) + Change in money supply with respect to time
Income is simply income from sales (P * Q), and so:
M * V = M * V * (1 – LP) + dM/dt
LP * M * V = dM/dt
Letting M = exp ( f(t) ) where f(t) is the demand function for the medium of exchange we rewrite:
LP * V = f'(t)
V = f'(t) / LP
And so the medium of account can be written as:
dM2 / dt = M1 * f'(t) / LP
Integrating both sides we get:
M2 = M1 / LP
M2 / M1 = 1 / LP
And so an issuing authority can try to fight a recession by changing the value of the medium of account relative to the medium of exchange. This will lower the liquidity preference.
But that will only work if the rate of demand for the medium of exchange does not fall. If LP and f'(t) fall by the same amount, then this will have no effect.
1. November 2012 at 21:13
MF:
It was a very crude reference to Kuhn.
NGDP is not output. NGDP is the total money expenditures that purchased output over a given period of time.
It is both (in nominal values). But at least you seem to agree with me more than Scott does.
1. November 2012 at 21:16
MF, recessions can happen without any change in relative prices. And Scott, there need not even be any such thing as a “price level”. Only a disequilibrium in the trade between final output and its medium of exchange. There is no other reason why there should be a general glut of all output.
2. November 2012 at 04:23
C’mon Morgan! You guys loved the EC in 2000. If there is a split-and actually that is still unlikely-I don’t want to hear any grousing about it.
3. November 2012 at 05:40
Your example assumes that the exchange rate between gold and Zim$ is flexible and determined by the market. That doesn’t quite match up with your madman government hypothesis. I work in Nigeria on a fixed US$ contract. You claim I couldn’t (not could) care less about inflation. But I do. The exchange rate between US$ and Naira is fixed and hasn’t changed in the year I’ve been here. That means my real purchasing power has declined 12.5% in the last year.
An interesting data point on this part of reality is that the central bank head around here is worried about the informal dollarization of the economy and is threatening to clamp down on the dollars they will issue. Dollars are not being used to buy stuff from Americans, but moving from one part of the capital to the other – in part because the N1000 bill is only worth $6 and the legislature has stymied his attempt to introduce a N5000/$30 bill.
3. November 2012 at 06:03
Oh, and shouldn’t point #3 be something closer to “When wages are sticky, less inflation than expected causes unemployment”?
3. November 2012 at 07:55
[…] Saturos recently said this in response to an earlier post: Scott, I’ll restate what I said. Of course […]
12. November 2012 at 09:20
[…] “demand for the medium of exchange” or “demand for the medium of account.” Scott Sumner and Nick Rowe are, as so often, at the center of this discussion. But before I do, I need to work […]
25. November 2012 at 20:59
[…] here, and here), Scott Sumner started another one of his periodic internet dustups (continued here, here, and here) this one about whether the medium of account or the medium of exchange is the […]
1. February 2013 at 13:16
[…] reading: Wikipedia’s article on money Sumner on the Unit of Account Share this:TwitterFacebookLike this:LikeBe the first to like this. from → Livejournal […]
10. April 2015 at 05:48
[…] year Scott Sumner introduced the econ blogosphere to what he likes to call the medium-of-account function of money, or MOA, […]
10. April 2015 at 06:53
[…] Rowe, Scott Sumner, and most recently, David Glasner, all say that the US’s current medium of account is the […]
10. April 2015 at 06:56
[…] Woolsey, Scott Sumner (here and here), and Nick Rowe and a debate that was fun to follow. It seems to me that they more or less […]