Money and inflation, Pt. 2 (Why does fiat money have value?)

The previous post described the long phase-out of the gold standard.  During the commodity money era we developed a dual medium of account (MOA), both gold and cash were MOAs.  For there to be two media of account, the price of one in terms of the other must be fixed.  Once gold prices started rising in 1968, only currency remained a medium of account; gold became a mere commodity.  (Silver was demonetized in the 1800s.)

Why does fiat money have any value at all?  This is actually two subtly different questions.  First, what is the value to society of having a medium of account?  And second, how can a fiat medium of account have value?  The MOA generally serves as a medium of exchange, and to some extent a store of value.  How valuable is an asset that fills those roles?   A medium of exchange is probably worth something like 1% of GDP, and as a store of value the MOA is worth another 1% to 10% of GDP.  How do we know this?  Because MOA traditionally didn’t pay interest, and hence the net present value of foregone interest from holding the entire stock of cash is simply the currency stock itself.  And currency stocks tend to be about 2% to 11% of GDP.

Thus if the nominal interest rate was 5%, and the currency stock was $1 trillion, then the (opportunity) cost of holding currency was $50 billion per year in foregone interest.  That could be viewed as the flow of liquidity services provided by currency (unfortunately including the flow of tax evasion services.)  If you divide this perpetual service flow by 5% to get its net present value, you get $1 trillion, which is the currency stock.

So what do people mean when they say that fiat money is “intrinsically worthless?”  They mean it has no value if it loses the role of being the MOA, whereas gold and silver had significant value in other areas of life.  (I.e. Confederate paper money after the Civil War.)  So this begs the deeper question:  Yes a monetary system is valuable, but what explains why this particular asset is accepted as money?  What gives it value?  There are several theories, which are not mutually exclusive:

1.  Its nominal price is fixed (unlike T-bills), which makes it convenient in transactions.  It comes in convenient small denominations (unlike T-bills).  Private small denomination currency issue may be banned.

2.  There is a sort of social contract/network effects/focal point thing going on.  People accept it as money because others accept it as money.

3.  There is an implied backing in an emergency.  For instance, if technological change was expected to make cash obsolete in the year 2047, the public may believe (correctly in my view) that the government would redeem it for some sort of real asset, rather than allowing hyperinflation.  One could think of this as the public having confidence that the government will do what is necessary to prevent hyperinflation in the future.

4.  The government allows one to pay taxes with currency.

5.  And in deference to Mike Sproul, because the cash is “backed” by assets on the central bank balance sheet.  (I don’t think this one matters, but it’s arguably related to point 3.)

Obviously all 5 reasons may be true, and they may work together.  It’s also worth discussing the peculiar evolution from gold-backed currency to fiat dollars.  I own a 1928 $20 bill, which looks almost exactly like a 1995 $20 bill.  Indeed pretty much like a just issued twenty, except Jackson’s portrait was smaller in the 20th century.  But the 1928 and 1995 twenties were radically different.  In 1928 the public viewed gold-backed currency as we would view personal checks;  ”an interesting piece of paper–now let’s see if we can take it to the Treasury (or bank) and get some real money.”  In 1928 “real money” was gold, whereas in 1990 we took a personal check to the bank to try to get “real money” in the form of cash.  So the monetary system evolved one level, what was considered debt in 1928, is no longer viewed as debt.  Cash is now the ultimate form of liquidity.  Cash can’t be cashed into anything more liquid.

Under the modern gold standard people actually used relatively little gold for transactions, mostly preferring cash and small coins.  So they got used to viewing cash as money.  To an economist studying inflation, the 1968 decision to end gold’s role as a medium of account seemed momentous.  But the public just yawned—they were already used to cash being the only MOA of relevance to their lives.  Appearances are very important—if something is regarded as money, it’s money.

From that point on “the social contract/network effects” story was probably enough, but clearly other factors like implied backing/stable monetary policy are lurking in the background.  If the public suddenly expected a policy of hyperinflation to be adopted in 2014, the value of cash would collapse right now.

This post has already run too long, so I’ll do the quantity theory of money in the next post.  We’ll see how the central bank can control the value of money (and NGDP), by changing the currency stock.


Tags:

 
 
 

62 Responses to “Money and inflation, Pt. 2 (Why does fiat money have value?)”

  1. Gravatar of 123 123
    20. March 2013 at 13:58

    It is much better to use another formulation of number five. Money has value because because central bank equity is subordinated to it. Otherwise we have a puzzle why the value of fiat money does not usually grow when central bank assets appreciate.

  2. Gravatar of Tyler Joyner Tyler Joyner
    20. March 2013 at 14:20

    http://www.smbc-comics.com/index.php?db=comics&id=2920

    Saw this comic and had to share it.

  3. Gravatar of Max Max
    20. March 2013 at 15:28

    Currency isn’t the medium of account. We didn’t go from a gold standard to a paper standard; we went from a gold standard to a price index standard. JPK explains it nicely:

    http://jpkoning.blogspot.com/2012/11/discussions-of-medium-of-account-could.html

  4. Gravatar of Rajat Rajat
    20. March 2013 at 15:28

    Isn’t the key reason fiat money has value is the nature of the fiat itself – ie that its supply is controlled by someone who the public trusts to maintain its relative scarcity (in relation to money’s other functions like making transactions or paying taxes)? We could use peppercorns as money if their supply was not so elastic.

  5. Gravatar of Neal Neal
    20. March 2013 at 15:29

    I really like explanation #2. When a primitive economy transitions from barter to money, a phase change occurs and the currency (MOE/MOA) assets centralize. Moving from gold to fiat is just a substitution of one MOE/MOA for another, which in the US occurred slowly between 1933 and 1968.

    You might argue that the dominant MOE is the MOA, but there are often other MOEs as well which could possibly become MOAs.

  6. Gravatar of Money and inflation, Pt. 2 (Why does fiat money have value?) | Fifth Estate Money and inflation, Pt. 2 (Why does fiat money have value?) | Fifth Estate
    20. March 2013 at 15:40

    [...] See full story on themoneyillusion.com [...]

  7. Gravatar of Negation of Ideology Negation of Ideology
    20. March 2013 at 17:18

    Good post.

    #2 is true, but doesn’t there have to be a “first mover”? I think #4 and #5 are the strongest reasons. They’re related, because often the assets on the Central Bank’s balance sheet are bonds backed by future taxes. So currency is still backed by the future tax base, or GDP.

    I’m not sure what you mean by a technological change that makes cash obsolete? If you mean paper cash, stock certificates are obsolete but they still have the same value as electronic stock records. If you mean something like a cost free replicator and the end of scarcity, then yes – money would be worthless but I’m not sure why anyone should care.

    Ultimately, I don’t think it’s any more mysterious why the token the government uses for tax purposes has value than it is why a subway token has value.

  8. Gravatar of Niklas Blanchard Niklas Blanchard
    20. March 2013 at 17:44

    I think this post was a bit wandering. More-so than the first to at least.

    That said, I’m glad you brought up the point at the end that by the end of the gold standard most people became so detached from gold as MoA (and much earlier MoE) that they didn’t think twice about severing the tie.

    I have often made the case to people that in the future, they will be so far removed from the “money” they use, they’ll likely not even recognize it when they see it.

    Never thought about the parallel before today.

  9. Gravatar of ssumner ssumner
    20. March 2013 at 18:50

    123, Good point.

    Tyler, Thanks–very funny.

    Max, No, the Fed isn’t engaged in PLT. Also, the price level fluctuates, whereas the price of the MOA is constant.

    Rajat, There are probably many explanations for why cash is money.

    Neal, In the 1920s and early 1930s the dominant MOE was cash, but the dominant MOA was gold.

    Negation, Yes, it’s overdetermined as to why cash has value–but it’s still interesting to consider which reason is the key.

    I think the first mover problem is solved by the fact that currency was already a MOA when we had a dual gold/cash MOA. So dropping gold didn’t cause much stress, because the public saw continuity.

    If the demand for cash falls close to zero, and the supply is unchanged, you’d have hyperinflation. I’m saying the government would prevent that.

    Niklas, Thanks for the criticism, I think you are right that the post wanders around too much. I’ll think about re-working the ideas.

  10. Gravatar of nospam nospam
    20. March 2013 at 18:55

    Fiat money does have significant value – while it may not be backed by gold, it is still backed by lead.

    Consider the case of Canada – the vast majority of Canadians live within fifty miles of the United States, a country with an economy an order of magnitude larger. The American dollar and the Canadian dollar trade near parity.

    Why is it that Canadian businesses transact overwhelmingly in Canadian dollars, and the few times that the American dollar is used, it is for very small retail cash transactions?

    The answer is simple: The Canadian dollar is backed in lead. Once a year, the Canadian government will demand a certain number of Canadian dollars from you.

    Should you fail to make this exchange, the Canadian government will send men to your door to order your compliance. Continued non-compliance from you will result in you being full of lead.

    And that’s why the Canadian dollar has value in Canada – it is backed by lead.

  11. Gravatar of Doug M Doug M
    20. March 2013 at 19:20

    You can pay use currency to pay your taxes? I am sure it is true, but I just don’t know how I would do it!

    First I would go to the bank and say I need a lot of cash. And the bank would say, we don’t have that much cash on hand at the momement, but there will be an armored car delivery tomorro morning and you can collect your cash then.

    So, I would go back to the bank the next day, and request some cash, and they would say, you can have it, but first we are going to have to fill out this paperwork with the Treasury that says that I have asked for a lot of cash and to keep an eye on me because I may be up to nefarious activities.

    Then I have to find a window for the IRS. I would guess that there are IRS agengts who might take my cash in any federal building, and there are about 15 of those around the country. Anywhere else? I could be a couple of hundred miles from the closest federal building. And if I showed up at the window with my wad of cash, what is the chance that the agent gives me a strange look. I try to explain that each bill says “leagal tender for all debts public or private” and this is a debt I want to settle.

    It could be interstesting. But, not particularly practical

  12. Gravatar of Doug M Doug M
    20. March 2013 at 19:24

    So now we have defined gold as money and currency as money.

    But, I use neither for the bulk of my transactions. Most of my money exists as a string of ones and zeros in the computer of my bank. My bank and I happen to agree on the meaning of that string of ones and zeros. Where does that get its value? Is it solely becuase it could be exchanged for paper currency? Which as I went on a bit of a rant in my previous post, it may be theoretically true that I can demand currency from my bank, it may not be true in practice.

  13. Gravatar of Geoff Geoff
    20. March 2013 at 19:31

    There is just one thing, one word, that I would change in this post.

    “4. The government allows one to pay taxes with currency.”

    Should read

    “4. The government forces one to pay taxes with currency.”

    When the IRS demands to be paid, they don’t accept gold. They don’t accept silver. They don’t accept anything except fiat notes, or transferable claims to fiat notes.

    Saying the government “allows” one to pay taxes with currency suggests that the government will accept one among many different commodities, with fiat currency just so happening to be one of them.

    Imagine I sold hamburgers, and I only accepted gold. It wouldn’t be right to say “I allow payment in the form of gold.” It would be more accurate to say “I [require/demand/only accept/etc] payment in the form of gold”.

    The reason why the expression “Fiat money is intrinsically worthless” exists is due to the fact that the non-coerced valuation of fiat money would put the exchange value of it below the value of the paper it’s printed on. The “value” of fiat money is, ultimately, the value people put on their freedom from prison/rape/death/etc.

    Without the threat of government force in the form of taxation in fiat money and claims to fiat money (and after a transition period as more and more people come to learn that other commodities hold their value better than fiat notes), the value of fiat money would collapse. THIS is the ultimate source for why fiat money is exchanged. It’s pure naked aggression.

    Imagine tomorrow the government ceased taxing people in fiat money and transferable claims to fiat money. You would likely be attracted to earning a commodity that is most highly valued for storing purchasing power. Paper is more easily depreciated than other commodities, so it is almost certain that fiat money would fall in rank for the purpose of storing purchasing power.

    It is my position that with a free market in money, precious metals would become the money of choice, because they have historically, and probably will for the foreseeable future, be the best commodities for storing purchasing power. Heck, even in the existing fiat world, where gold isn’t money, central bankers around the world stockpile gold. Why? Because even central bankers know that gold is a better commodity for storing purchasing power than their own paper notes! Hahaha

  14. Gravatar of Mike Sproul Mike Sproul
    20. March 2013 at 19:33

    Theories #1 and 2 are not in conflict with the backing theory, and theories 3, 4, and 5 ARE the backing theory. The quantity theory requires us to believe that intrinsically worthless pieces of paper somehow hold themselves up by their bootstraps, while the backing theory only requires us to believe that money has value for the same reason that stocks, bonds, etc. have value.

    So why is the quantity theory in every textbook while the backing theory is in NO textbook? Simple. People see that the dollar is not convertible into gold, and they wrongly conclude that inconvertible=unbacked.

  15. Gravatar of Neal Neal
    20. March 2013 at 19:34

    Scott- So companies measured their revenue and income in “oz of gold” in the 1920s and 1930s?

  16. Gravatar of Neal Neal
    20. March 2013 at 19:47

    (Revenue and expenditures, rather.)

  17. Gravatar of ChrisA ChrisA
    20. March 2013 at 20:01

    Gold of course can also be considered a fiat currency as well, its just a shiny metal with few intrinsic uses. Just like paper currency, people accept gold as valuable because other people accept it as valuable. The only benefit versus paper currency is that it is harder to make gold, so the printing risk is less. But that may not be true if, say, asteroid mining became cheap and easy. Maybe this is unlikely in say 10 years, but in 50 years it could easily be possible. What can you say about gold as a long term store of value then? There were many episodes in the past where a new source of gold (Americas, South African mines) can online and reduced the relative price of gold versus other goods and services, so it can definitely happen.

    Thinking about your list, I think you missed the most important one. That is that when you see prices quoted in dollars all the time, it is perfectly rational for the individual to accept a dollar as valuable. In essence this makes a paper currency backed by assets and goods and services. This isn’t quite the same as what you describe as a social contract model, which is perhaps better described as a social inertia model. What I am describing is a path dependent agent model. Each actor being individually rational based on the surrounding actors behavior.

  18. Gravatar of Mike Sax Mike Sax
    20. March 2013 at 20:17

    Thanks a lot Scott. While I understand that some find these last few points quite basic and elementary, but I’ve found it quite interesting and helpful.

  19. Gravatar of Doug M Doug M
    20. March 2013 at 21:38

    Geoff,

    You can pay the IRS taxes by check or electronic trasfer. Checks are not currency. Nor are transfers.

  20. Gravatar of Suvy Suvy
    21. March 2013 at 00:46

    Prof. Sumner,

    I’d say fiat money has value for the same reason a dollar backed by gold does. The reason is faith and confidence; the same reason that everything happens in capitalism. What gives a money value is the same thing that has historically given a letter of credit or a bill of exchange value: faith, confidence, belief, and trust. I never really realized this until recently, but those 4 qualities are the lifeblood of capitalism. If those 4 things fall apart in an economy or currency, all hope is lost; the economy collapses into an abyss until the confidence is restored.

  21. Gravatar of Luis Pedro Coelho Luis Pedro Coelho
    21. March 2013 at 02:36

    “They mean it has no value if it loses the role of being the MOA, whereas gold and silver had significant value in other areas of life.”

    Curiously, this is very true nowadays: if silver was plentiful, we’d wire our houses with it as is conducts better than copper. Also, gold is pretty useful in a lot of electronics, but its use is minimized due to its high cost.

    However, historically, it was not true. There was no use in silver/gold except for its high value (yes, jewelry, but that just proves my point, doesn’t it?). It was just that it was hard to produce. So, gold was the bitcoin of the day.

  22. Gravatar of J.V. Dubois J.V. Dubois
    21. March 2013 at 02:47

    Scott: “Max, No, the Fed isn’t engaged in PLT. Also, the price level fluctuates, whereas the price of the MOA is constant.”

    I disagree. By the same logic if the price of gold was $20.67 in 1932 and then $32.32 in 1933 it meant that FED was not engaged in the gold standard. Price level fluctuated a lot and “price” of the MOA was anything but constant.

    And if we are at it I am now a little bit confused. What did you meant by “Price level fluctuates whole the price of MoA is constant”. Price of MoA terms of what? In terms of itseld – that it easy, it is always 1 and therefore it does not fluctuate. Or did you mean price of MoA is constant int terms of one hour of lowskill labor? Should it be 1kg of wheat? Or maybe you meant price in terms of green paper with picture of dead president and “1 dollar” written on it that happens to be also MoE? Shouldn’t MOA be that thing that serves to you as measure value of everything? Why should it matter if it has constant value in terms of something (even MoE)?

    And to conclude, I see it as pretty reasonable for anyone nowadays to consider CPI as medium of account. I saw several posts on MM blogs (like one here: http://worthwhile.typepad.com/worthwhile_canadian_initi/2013/01/the-bank-of-canadas-success-and-failure.html) about how Central Banks were actually quite sucesfull at targeting price level. It is valid for Canada, you yourself had several remark how incredible it is that BoJ managed to keep price level literally constant for almost 20 years.

  23. Gravatar of Geoff Geoff
    21. March 2013 at 03:38

    Doug M:

    “You can pay the IRS taxes by check or electronic trasfer. Checks are not currency. Nor are transfers.”

    Doug, I specifically said taxes must be paid in fiat money and transferable claims to fiat money.

    That would obviously include both checks and electronic transfers.

  24. Gravatar of Geoff Geoff
    21. March 2013 at 03:40

    ChrisA:

    “Gold of course can also be considered a fiat currency as well, its just a shiny metal with few intrinsic uses.”

    Fiat means “by decree.” The only way gold could be considered “fiat”, would be if gold were enforced by the state as money, through taxing people in gold only, regardless of their income. Like what they do now with state issued currency.

  25. Gravatar of Geoff Geoff
    21. March 2013 at 03:41

    Mike Sproul:

    “So why is the quantity theory in every textbook while the backing theory is in NO textbook? Simple. People see that the dollar is not convertible into gold, and they wrongly conclude that inconvertible=unbacked.”

    Unbacked means the fiat money has no redeemable nature to it for a commodity that is valued.

  26. Gravatar of ssumner ssumner
    21. March 2013 at 03:58

    Doug M, I have defined both of them as media of account under the gold standard, only currency since 1968.

    Mike, I think the QTM is in the textbooks because it’s seen as providing a better explanation for why OMPs are inflationary than the backing theory does. Next week I’ll do a post trying to explain why the QTM is so powerful.

    Neal, No, they were measured in 1/20.67 oz of gold in the 1920s, aka “dollars.” Bonds were payable in little pieces of gold, or paper backed by a fixed amount of little pieces of gold.

    Chris, Well that’s partly right, as once gold became an accepted MOA, some of the demand was motivated by the fact that others viewed it as precious, not the fact that it was 16 times prettier than silver. But it did have SOME intrinsic value.

    Luis, Teeth filllings? I’d say jewelry reflects intrinsic value, but see my response to Chris.

    JV, We went off the gold standard for 10 months in 1933. The price of the MOA is fixed in terms of the unit of account. For instance, if the unit of account is “francs” and the MOA is zinc, the the price of zinc (per pound) in francs never changes.

    Even if you were right, the CPI would not be the MOA, the MOA would be the basket of goods that comprises the CPI.

    I’ve proposed using NGDP futures contracts as the MOA, and then devaluing at 5% per year.

  27. Gravatar of Jake Jake
    21. March 2013 at 05:02

    Prof Sumner, I am enjoying these posts a lot. Keep up the good work!

  28. Gravatar of Mike Sproul Mike Sproul
    21. March 2013 at 06:44

    Geoff:
    “Unbacked means the fiat money has no redeemable nature to it for a commodity that is valued.”

    The word “redeemable” has a very broad range of possible meanings:
    1. instantly redeemable for 1 gram of gold, 24/7
    2. #1, but 9-5 weekdays only
    3. #1, but with a delay of 30 days
    4. #3, but 30 years
    5. #4, but for some unspecified period
    6. #1-5, but for bonds worth 1 gram
    7. #1-5, but for bonds worth a specified CPI basket

    This list could get very long, but it should be obvious that ‘redeemable’ can have many meanings, and money users actually don’t care very much about which kind of redeemability is offered. But they do care very much about the solvency of the money issuer. It’s backing that matters, no redeemability.

  29. Gravatar of mpowell mpowell
    21. March 2013 at 07:37

    Sproul – so your argument is that because you have to pay your taxes in dollar denominated assets that means cash is backed? This is an odd way of looking at it to me. I prefer to think of it as: dollars have value because their supply is limited, you have to pay your taxes in dollars and the gov takes the steps necessary to keep their value stable. Arguing whether this is some flexible definition of backed or not seems to be besides the point.

    Geoff – it is amazing to me that a person could prefer an economy with no reliable MOA or MOE just because they hate paying their taxes that much. But not surprising. With the internet I am now aware of many of the ridiculous theories of government and non-government floating around out there.

  30. Gravatar of Jim Glass Jim Glass
    21. March 2013 at 08:26

    I’ve really never understood the common insistence (obsession) that money must be “backed by” or be “convertible into” something else to have value.

    My refrigerator has a market price which is not determined by the “backing” of its component parts that it might be “converted into” (what would their value be backed by?) but by the value of the function it serves as a refrigerator, combined with the limited supply of refrigerators. The same with my car, my computer and everything else I own.

    Why would money be any different than everything else? Its price is determined by the value of the function it serves as money, combined with the limit of its supply.

    Bitcoin has a market price. It is literally nothing, is backed by nothing, and is convertible into nothing. There is no issuer in whose solvency people have confidence. It is pure quantified nothingness.

    Yet it has a market price that goes up and down, which derives from its usefulness to some people (for whatever) combined with its finite supply. It illustrates the price-value of pure quantified scarcity itself. How different is that from fiat money? Of course, bitcoin’s proponents say it *is* money, And if it is money, it surely is fiat money. (Though I say it is a commodity — the first fiat commodity.)

    As to the repeated claims that money gets its value from the government forcing people to pay their taxes with it, while the idea seems to have some sort of magnetic attraction to certain kinds of minds running from Austrian to MMTer, it is deeply anti-factual and anti-historical.

    Years ago Warren Mosler, when he was still calling himself a chartalist, told me that the US dollar derived its value from people having to pay US taxes in dollars. When I pointed out to him that this claim is not true and has never been true, he didn’t have an answer — except to keep repeating the bogus claim in different words. But from its first days the US govt has always accepted taxes in whatever money was at hand to pay them with. It still does today — if you earn your income in some other currency the Tax Code requires payment in dollars *IF* the currency is exchangeable for dollars, so the govt saves the exchange rate fee. But if you earn your income in some other currency that due to exchange controls (such as Malaysia a little while back) or for any other reason is not convertible into dollars, then the IRS will not only happily accept payment in that other currency, it will insist! You don’t imagine you then become tax free, do you? When the govt taxes your income, it doesn’t want your dollars, it wants your *income*.

    (And, of course, the dollar predates the US govt — instead of forcing it on the people, the people forced the dollar on the govt, since if it wanted to buy anything back in 1789 it had to pay with the dollars or it wouldn’t get it.)

    The historical record is full of nations that insisted on tax payment in their own money only to see their own monies collapse and be replaced by their populaces using some other money. And also full of regimes and societies that didn’t collect taxes (commodity financed) or with negligible taxes incurred on the general populace, which used non-gold/silver/”backed” money nonetheless.

    The stone money of Yap Island had no use value and was wasn’t backed by or convertible into anything — in fact, it was in substantial part located on the bottom of the sea (!), purely notional, existing only in people’s memory, the pre-electronic era bitcoin.

    Somebody tell me the people used it as money because the government forced them to pay their taxes with it.
    :-)

  31. Gravatar of JP Koning JP Koning
    21. March 2013 at 08:28

    Scott, your use of the term MOA is certainly unique.

    Here’s the progression of the British MOA since the 1200s or so.

    1. Silver
    2. Bimetallic (gold/silver)
    3. Gold
    4. US dollar
    5. Deutsche Mark
    6. CPI
    7. NGDP?

    … with gaps in between when the MOA was either private, undefined, or a mix and mash. But at no point was the pound note the MOA. That wouldn’t make any sense. See here for details and a good comment by Britmouse.

  32. Gravatar of J.V. Dubois J.V. Dubois
    21. March 2013 at 08:40

    Scott: Your answer is actually quite interesting. I admit that I do not know all the facts about the Great Depression, but I would really love to know answer on some questions:

    1. If USA really left gold standard for 10 months in 1933, what standard did they use? What was the medium of account then?

    2. I really have hard time imagining franc as “unit of account” for “gold”. If this were true, what part does the weight of the gold play all in this? This may be my science education speaking, but units have to always add up. If you have just number on one side of equation and kilograms on the other side, you did some invalid operation in some previous step.

    I can imagine gold being the medium of account and one ounce of gold being unit of account. If you imagine it in this way, does it even make sense to claim “price of gold is fixed in terms of ounces of gold”?

    Now I can imagine something regime where for instance people use gold coins with kings head on one side weighting of 0.9 ounces as medium of exchange. And making new coins unless you are a king is a capital offense. King may even promise that he will exchange such coin for another foreign coin of pure gold – if you show up in his palace on the day of king’s birthday. But this does not make this foreign coin unit of account.

    And now for some philospohical question. Since you deem fixed price of medium of account in terms of unit of account as crucial aspect of medium of account, what would be medium of account if FDR claimed that he will stay on gold standard, only he will commit to 2% gold inflation path? (price of gold in terms of dollars). Would you say that every time Central bank issued new unbacked dollars to carry out this policy is a day where old gold standard was suspended only to be returned a second later with new gold price?

    This line of thinking seems quite strange to me. But it is the only way you may claim that Japan (or Canada) use CPI basket goods as money.

    PS: Just to clarify. I am not claiming that Japan uses “CPI basket goods” money. They use Yen as money (and Yen happens to be tied to CPI basket goods). But Yen is important because it is MoE.

  33. Gravatar of Why Do Economists Obsess over Government Money? | Monetary Realism Why Do Economists Obsess over Government Money? | Monetary Realism
    21. March 2013 at 12:17

    [...] was reading this post by Scott Sumner in which he concluded with the [...]

  34. Gravatar of Mike Sproul Mike Sproul
    21. March 2013 at 13:17

    MPowell:

    The government has many assets, one of which is ‘taxes receivable’. Federal Reserve notes are one of the government’s liabilities. The government’s assets back the government’s liabilities.

    For example, a landlord receives rents of 50 oz. per year on his land. The present value of rents is 1000 oz. The landlord buys his groceries with his own IOU’s, each of which he accepts for 1 oz. of rent. He can easily issue up to 400 or so of those IOU’s, and they are more than adequately backed by his 1000 oz. of land. He may or may not offer to redeem those IOU’s for actual silver, but if he does not, you can bet that there will be people who think of his IOU’s as fiat money.

  35. Gravatar of Mike Sproul Mike Sproul
    21. March 2013 at 13:30

    Jim:

    As Schumpeter put it: You can’t ride a claim to a horse, but you can trade with a claim to money. He might have added that you can also trade with a claim to a horse.

    Or take it 1 step further: Convertibility into horses is suspended, but the issuer of the claim still has assets worth at least 1 horse. Which of the following is more likely?
    1) That claim is still valued because it is still backed by assets worth 1 horse, or
    2) That claim has value because its issuer has promised to limit its supply, even though it holds no assets with which to buy those claims back, should the demand for those claims fall.

  36. Gravatar of Max Max
    21. March 2013 at 15:43

    If you think about what it would take for a government to (attempt to) get an unbacked money accepted, the right term for it is financial repression. Because you would be trying to use regulations to coerce an artificial demand for something that, in the logic of finance, there should be no demand for.

    Whether that would work or not, it’s not the basis for real world central bank money, because central banks do have assets, which they can sell if needed.

  37. Gravatar of Fed Up Fed Up
    21. March 2013 at 17:35

    “During the commodity money era we developed a dual medium of account (MOA), both gold and cash were MOAs. For there to be two media of account, the price of one in terms of the other must be fixed.”

    How about a tri medium of account (MOA), currency, gold, and demand deposits? Demand deposits and currency are assumed to be 1 to 1 redeemable (fixed price).

  38. Gravatar of Fed Up Fed Up
    21. March 2013 at 17:47

    “The MOA generally serves as a medium of exchange, and to some extent a store of value.”

    So both currency and demand deposits?

  39. Gravatar of Fed Up Fed Up
    21. March 2013 at 23:02

    From the “why does money matter?” post:

    “Currency notes are the medium of account, the thing used to measure value. Abstract accounting units like the US dollar, the Canadian dollar, and the euro are examples of a unit of account.”

    So you are saying the total amount of currency is the MOA and $1 of currency is the UOA, right?

  40. Gravatar of mpowell mpowell
    22. March 2013 at 08:54

    Mike,

    Taxes receivable could be considered backing for the currency or a mechanism to limit the quantity / generate demand. But here’s the problem with your example: is anyone going to trade the landlord’s IOU’s as money without the expectation that they’re redeemable for something else? It doesn’t have to be silver, but you’d better be getting the rent somehow. Most users of a currency have no understanding of any of these details and couldn’t tell you anything about the convertability or backing of their dollars. But they will still respond to supply/demand issues. That’s where the appeal of the quantity theory comes from.

  41. Gravatar of Mike Sproul Mike Sproul
    22. March 2013 at 09:24

    mpowell:

    People initially accept the landlord’s IOU’s (‘dollars’) because they are redeemable for 1 oz. of rent, even if they are not directly redeemable for 1 oz. of silver. Once the landlord’s IOU’s are commonly used as money, the landlord can maintain the value of his IOU’s just by conducting open market operations with bonds or other assets. But this requires that he actually has enough assets to buy back all the IOU’s he has issued, should the demand for IOU’s fall.

    The average person might be clueless about the landlord’s assets and liabilities, and also about the supply of and demand for money. It’s only necessary to have a few smart arbitragers who would short dollars if they were inadequately backed, or buy dollars if they were under-valued relative to their backing.

  42. Gravatar of Becky Hargrove Becky Hargrove
    22. March 2013 at 09:27

    Scott,
    Thanks so much for these highly specific monetary posts. The social contract implied between government and citizens make your second point especially important in the present, as people who question that contract are also among the ones who question the fiat money concept, natural and contemporary though it seems. For me, the outcome of this particular ongoing discussion regarding social contract (and its effective transmission) is vital in the real “battle” for the future of NGDPLT.

  43. Gravatar of kebko kebko
    22. March 2013 at 11:33

    Off topic, here is the latest article seemingly immune to reality. A primary dealer looks at TIPS spreads that are roughly the same as they have been for 10 years, except for the intervening period of unusually low inflation, with real rates that have plunged, and decides that the problem is inflation fears:

    http://www.bloomberg.com/news/2013-03-21/treasury-sells-tips-at-negative-yields-as-buyers-doubt-bernanke.html

  44. Gravatar of Bill Ellis Bill Ellis
    22. March 2013 at 15:22

    Professor Sumner…I really enjoyed these money posts.

    Here is how I like to think of what money is… It is a promise we make to each other that we will try and make the world a better place.
    Super corny. I know.

    But money does represent value, and what is of real value if it does not make the world a better place ?

  45. Gravatar of paul Einzig paul Einzig
    22. March 2013 at 18:52

    Hi Scott

    I am really loving these posts on money and the discussions that follow !

    Thanks so much for the time you put into this blog.

    To Jim Glass, I loved your Yap comment. The thought that some past Yappian regent would have forced vw bug sized stones in fifty fathoms of water onto a puzzled Yap public as their new medium of exchange is obviously goofy and very funny but if you read David Graebers recent book on money as debt you will see that he actually resolves the apparent Yap/chartalist conflict you raise.

    Jim, at the risk of exposing myself for the undereducated cretin that I am please read on :

    You said:
    ” As to the repeated claims that money gets its value from the government forcing people to pay their taxes with it, while the idea seems to have some sort of magnetic attraction to certain kinds of minds running from Austrian to MMTer, it is deeply anti-factual and anti-historical.”

    I’d like to share with you my own, (possibly) irrational, attraction to chartalism and perhaps this will explain some of what you call magnetism.

    I am a worrier by nature, and I used to lay in bed at night and worry about all of the horrible things that would happen to me, my family, the country etc. if or when ” The Fiat Currency Collapsed !!!!”

    One day I confided this nagging fear to a learned friend of mine who then introduced me to an old and unknown book called The State Theory Of Money by Georg Knapp and I simply fell in love with chartalism. The thought that federal reserve notes are real because the strongest , most stable government on earth says so…… that really works for me! It’s a very strong tonic to my weary soul, the words “profoundly comforting” come to mind. It’s very freudian I think.
    Do you remember the motown tune from the 70′s ” If loving you is wrong, then I don’t want to be right” That’s kinda how I feel about chartalism !

    Anyway, from then on I have never, ever lost sleep about….. ” intrinsically worthless paper money” ( Now I just lay in bed and worry about catastrophic miscalculation by the Fed. LOL)

    But seriously, why is it ( chartalism) ” anti-factual and anti historical”. No offence but Yap and Malaysia are really exceptional examples, kind of a thin rebuttal.

    You know Yap was peopled by Malaysians originally, so your anti chartalist position is very Malayo-centric!

  46. Gravatar of OhMy OhMy
    22. March 2013 at 19:17

    Jim Glass,

    However much you would like to soothe yourself into believing that MMT is wrong on taxes as backing for money, this theory is actually backed by evidence. Multiple times in history kings to wage wars would impose a tax obligation on the local population and then issued worthless tokens that could be used to free oneself from said tax obligation. Worthless tokens had value. Even historians who never heard of chartalism recount these stories. So Mosler definitely had an answer for you, but maybe not one that you understood.

    This post is so opaque and confused, better read MMT primer or Wray “What is money”.

  47. Gravatar of ssumner ssumner
    23. March 2013 at 07:14

    Thanks Jake.

    Jim Glass, I agree, money doesn’t have to be backed.

    JP, The pound note has been a MOA throughout all of modern British history. At times they had a dual MOA, both pounds and gold, or pounds and the US dollar.

    The CPI was never the MOA.

    JV, They used cash as a MOA in 1933.

    The definition of the franc specifies exactly how much gold comprises one franc. If it’s 1/10th ounce, then the price of gold is fixed at 10 francs per ounce.

    I agree that a continually changing price of the MOA is a strange thing, and our language hasn’t caught up. But it’s still true that a surge in gold demand would be highly deflationary, and a huge gold discovery would be highly inflationary, under a gold standard where the price of gold rose at a fixed 2% per year.

    Fed Up, Is the price of DDs fixed? What if the bank fails?

    No, I mean the abstract unit “one dollar” is the unit of account–not the paper note. Just as a certain length is the abstract meter as a unit of measurement, not a wooden stick one meter long.

    Thanks Becky.

    Kebko, Yes, that’s a common mistake.

    Bill, I would hope so.

    Thanks Paul.

  48. Gravatar of Bill Woolsey Bill Woolsey
    23. March 2013 at 07:20

    The traditional approach has been to say that money has various roles or funtions. It serves as medium of exchange, store of wealth, unit of account and so on.

    The concept of medium of account is really associated with an alternative approach. What is the unit of account? What is the media of exchange? The medium of account is what defines the unit of account. While usually it has been a medium of exchange (and perhaps the emphasis should be on “been,”) but it doesn’t have to be.

    I think most who have worked on the alternative approach do so by considering alternative monetary regimes. And lots have been focused on substantially private regimes.

    So, the dollar is the unit of account. Suppose it is defined as 1/20th of an ounce of gold. Gold is the medium of account.

    Suppose that the dollar is switched to be 1.25 ounces of silver? The dollar is still the unit of account, but the medium of account has shifted.

    Suppose that the dollar is redefined as 1 ounce of silver? The same media of account, but the definition has still changed.

    Traditionally, governments on commodity standards minted coins out of the metal. But there have been long periods of time when “foreign coins” were perfectly acceptable too.

    And here is a typical thought experiment of those considering different functions. Suppose that coinage is fully private. Different private mints create dollar coins. The governnent just says the dollar are 1/20th of an ounce and punishes fraud.

    The gold media of exchange does not define the unit of account. A dollar isn’t a ACME gold dollar coin or a Franklin Mints dollar gold coin.

    If the U.S. and Great Britain are both on the gold standard, they have the same medium of account, (gold) but one uses dollars and the other pounds. They have different units of account.

    Commodity coins are pretty inconvenient. Banks have been able to convince people to use paper notes and deposits as media of exchange. These deposits and notes are denominated in terms of the unit of account, but the notes or deposits of any one bank in no way define it. With a gold standard, the definition would still be gold.

    If there was a devaluation or a switch to a silver standard, all of the bank deposits and notes used as the media of exchange would still be denominated in dollars, the unit of account, but the medium of account has changed. Either the amount of the good or the good changes.

    Consideration of a regime like that suggest that it is possible to switch not just from gold to silver, but perhaps to steel or bricks. It is not necessary for the medium of account to be suitable as a medium of exchange. Banknotes and deposits can serve as media of exchange without being the medium of account.

  49. Gravatar of Geoff Geoff
    23. March 2013 at 08:48

    Mike Sproul:

    “It’s backing that matters, no redeemability.”

    Backing implies redeemability. No backing, no redeemability. Redeemability, backing.

    mpowell:

    “Geoff – it is amazing to me that a person could prefer an economy with no reliable MOA or MOE just because they hate paying their taxes that much. But not surprising. With the internet I am now aware of many of the ridiculous theories of government and non-government floating around out there.”

    Governments are not needed for a “reliable” MOA/MOE to exist. The hidden premise in your argument is the same one Bastiat exposed in “The Law”:

    “Socialism, like the ancient ideas from which it springs, confuses the distinction between government and society. As a result of this, every time we object to a thing being done by government, the socialists conclude that we object to its being done at all. We disapprove of state education. Then the socialists say that we are opposed to any education. We object to a state religion. Then the socialists say that we want no religion at all. We object to a state-enforced equality. Then they say that we are against equality. And so on, and so on. It is as if the socialists were to accuse us of not wanting persons to eat because we do not want the state to raise grain.”

    Not sure if you’re being sarcastic or serious though…

  50. Gravatar of ssumner ssumner
    24. March 2013 at 06:16

    Bill, Very good explanation.

  51. Gravatar of J.V. Dubois J.V. Dubois
    25. March 2013 at 05:58

    Bill, Scott: Ok, another random idea. Let’s suppose that there are two things – there is a thing called medium of exchange. Ownership for MoE is traded one one side during each and every market transaction. Then there is an algorithm that says how amount of this medium of exchange changes over time depending on various things like velocity of circulation of MoE etc.

    And now comes the weird part (for me). As far as I understand, Scott says that if there exists a “medium” that has its price exactly (nominaly) fixed in terms of MoE during some time then this medium is (for that given time) called “Medium of Account”. If there is any other situation – such as if for example there is some medium that experiences perfectly stable rise of its nominal price in terms of MoE, then bad luck. This medium cannot be considered a proper MoA. In such a situation it is “cash” (that to me translates as medium of exchange) that is automatically also MoA.

    So to sum it up, there exists a lot of rules and algorithms that can be used to regulate supply of MoE. There exists a tiny subset of rules under which we just arbitrary claim that something else (not MoE) is also MoA. And for some reason this whole MoA business is in some way (not related to MoE) very important to monetary policy.

    I can perfectly understand why somebody can stand up and say: “The rule which regulates supply of MoE is called MoA rule and it is important for monetary policy”. But am I the only person who finds an extended definition like this weird? : “The rule which regulates supply of MoE is called MoA. And this rule that regulates MoE is the most important aspect of money. It has nothing to do with MoE role of money despite the fact that its definition is tied to it.”

  52. Gravatar of ssumner ssumner
    25. March 2013 at 06:49

    JV, No, I am not saying that the price of the MOA must be fixed to the MOE, I’m saying that’s usally the case.

  53. Gravatar of J.V. Dubois J.V. Dubois
    25. March 2013 at 08:52

    Scott: Ok, then I appologize, I just extrapolated that from your previous explanation that pre-FDR devaluation of dollar it was gold that was MoA, then there was a 10 month period of confusion where “cash” (MoE) was MoA only to be replaced later by a new regime where gold was MoA again – albeit at new exchange ratio in terms of gold.

    All I want to say is that I see this definition of MoA as confusing. I could imagine gold being MoA if for instance all contracts explicitly claimed that for instance “Payment: creditor promises to pay equivalent of 40 ounces of gold in dolars as would be its price on Chicago comodity exchange for the given day”. Then no matter what FDR would do, he would not be able to devalue dollar (other then refusing to back up such contracts with state power, but that is different debate) But no, most contracts probably did not require payment tied to gold, they required payment tied to US dollars. Dollar was medium of account for all purposes and that is the reason why it looks so non-problematic to go from gold standard to a different one.

  54. Gravatar of JP Koning JP Koning
    25. March 2013 at 12:36

    As usual when it comes to MOA, Bill Woolsey makes 100% sense but I am always puzzled by Scott’s definition.

    Bill, what do you think of this? It is 1925 and Britain is defining the pound as 113 grains of gold — we’ve got a gold standard, and gold is the MOA. Good so far? Then Britain decides to redefine a pound by choosing a totally different MOA. It decides to use US$, specifically one pound equals $4.03. Britain is now on a dollar standard with $US as the MOA… which is what actually happened in 1939. But even though the MOA has changed, Britain still has the same UOA throughout: the pound. My illustration of Britain’s changing MOA/UOA from 1925 to 1939 seems to proceed naturally from your definition of MOA.

  55. Gravatar of Fed Up Fed Up
    25. March 2013 at 19:35

    “Fed Up, Is the price of DDs fixed? What if the bank fails?

    No, I mean the abstract unit “one dollar” is the unit of account–not the paper note. Just as a certain length is the abstract meter as a unit of measurement, not a wooden stick one meter long.”

    Price of DD’s fixed? Yes, it is supposed to be 1 to 1 to currency. Bank fails? Deposit insurance. If the losses go above capital and deposit insurance, then it gets more interesting like Cyprus.

    I sort of get the abstract “one dollar”. I like to think in terms of what “one dollar” (UOA) will buy (relative pricing).

  56. Gravatar of Fed Up Fed Up
    25. March 2013 at 19:53

    Bill Woolsey said: “So, the dollar is the unit of account. Suppose it is defined as 1/20th of an ounce of gold. Gold is the medium of account.

    Suppose that the dollar is switched to be 1.25 ounces of silver? The dollar is still the unit of account, but the medium of account has shifted.

    Suppose that the dollar is redefined as 1 ounce of silver? The same media of account, but the definition has still changed.”

    I like this definition here in the comments:

    http://jpkoning.blogspot.com/2012/11/discussions-of-medium-of-account-could.html

    “A medium of account is the commodity defining the unit of account. A unit of account is a specific amount of the medium of account. For example, for the gold standard the medium of account is gold, while the unit of account might be one ounce or one pound of gold of specific purity.”

    Gold standard with prices quoted in both dollars and gold. I still like to think of the UOA as $1 and the MOA is total dollars. UOA is also 1 ounce of gold and MOA is total gold. The price of one in terms of the other must be fixed.

    Next, gold standard with prices only quoted in terms of dollars. UOA is $1, and MOA is total dollars. The fixed price of gold is now only there to limit the MOA.

    The dollar is also the MOE.

  57. Gravatar of Bill Woolsey Bill Woolsey
    26. March 2013 at 04:03

    Koning:

    It seems to me that the pound stays the unit of account and the medium of account switched to dollar. But the dollar is a unit of account which is defined in terms of gold. Doesn’t that make the medium of account gold?

    What changed is that pound notes are no longer redeemable in gold, but rather in dollar notes (or deposits.)

    I we had a bundle of commodities medium of account, and the dollar-denominated notes were redeemable in gold equal in market value to the bundle, this doesn’t make a variable amount of gold into the medium of account.

    Just because pound notes are redeemable in dollar notes equal in value to the amount of gold defining the pound, doesn’t make the dollar notes into the medium of account.

    I think Zimbabwe had a pure fiat currency. The medium of account was the Zimbabwe dollar note whose value depending on supply and demand with the supply being based upon the desire of the rule to get money.

    During the seventies, the U.S. had a pure fiat currency. The medium of account was the U.S. dollar note whose value depended on supply and demand with the supply manipulated to try to control unemployment.

    As we have moved to inflation targeting, we are moving to having a commodity bundle defintion of the dollar, with continual devaluation, and changes int the bundle. The rate of devaluation is not constant but varies with the output gap.

    But, given the Fed’s insistence on discretion, were continue to have something like the pure fiat currency.

    To me, it is really a matter of framing. What is the best way of understanding a monetary order?

    Even in the U.S., during the gold standard of the 19th century, there were periodic suspensions of convertibility. What was the medium of account during those periods?

    With Great Britain, there was a long suspension in the early 19th century. What was the medium of account.

    Sumner has already spoken of the Roosevelt period when the dollar price of gold was adjusted on a daily basis.

    In my view, if the money serves as medium of account, it isn’t debt. But when there is some other medium of account, even something as odd as an inflation target, then currency notes are debt.

  58. Gravatar of JP Koning JP Koning
    26. March 2013 at 06:17

    Bill:

    “But the dollar is a unit of account which is defined in terms of gold. Doesn’t that make the medium of account gold? What changed is that pound notes are no longer redeemable in gold, but rather in dollar notes (or deposits.)”

    Yes, makes sense to me.

    “If we had a bundle of commodities medium of account, and the dollar-denominated notes were redeemable in gold equal in market value to the bundle, this doesn’t make a variable amount of gold into the medium of account. Just because pound notes are redeemable in dollar notes equal in value to the amount of gold defining the pound, doesn’t make the dollar notes into the medium of account.”

    Yes, all makes sense to me.

    “As we have moved to inflation targeting, we are moving to having a commodity bundle definition of the dollar, with continual devaluation, and changes int the bundle. The rate of devaluation is not constant but varies with the output gap.”

    I agree. It’s a CPI standard in which redemption media is not commodity bundles.

    “I think Zimbabwe had a pure fiat currency. The medium of account was the Zimbabwe dollar note…”

    Here’s where I don’t follow you. Why not just say that there was no medium of account during the hyperinflation? The unit of account was Zim$, but it was undefined.

    Or imagine that Zimbabwe did have a commodity bundle definition of the Zim$, but the Reserve Bank of Zimbabwe only gave that definition lip service. There would have existed some medium of account, but that definition wasn’t enforced. It’s like having speed limits that no one listens to because the cops don’t hand out tickets should those limits be exceeded.

    Same with periodic suspensions in the US and Great Britain in the 19th century. The MOA was still gold, but issuers temporarily ignored the definition.

    It would be the same today if a modern central bank missed its inflation targets. The MOA is still the CPI commodity bundle, but the issuer has decided to stop enforcing that definition.

  59. Gravatar of ssumner ssumner
    26. March 2013 at 06:18

    JV, Many important long term contracts did require payment in gold–if the creditor demanded it. Otherwise cash was more convenient.

    Fed Up, I left a comment in the newer post.

    JP and Bill, Obviously commodity exchanges generally had dual media of account, or even three. But it’s interesting to think about which one was dominant. In my view gold was dominant, since it was traded in a global market and thus set the global price level in gold terms, and currency was mostly endogenous, albeit big countries had some discretion.

  60. Gravatar of Why is Fiat Money Valuable | Free Radical Why is Fiat Money Valuable | Free Radical
    2. April 2013 at 22:23

    [...] developing a theory of money and banking for a while and I noticed that Scott Sumner did a post on why fiat money has value in which he listed five theories explaining this and mine wasn’t among them.  This made me [...]

  61. Gravatar of Mike F Mike F
    2. April 2013 at 22:52

    I guess I missed the discussion but if anyone is still around, I have an explanation which is not on the list and nobody is talking about. If anybody is interested I try to explain here.

    http://realfreeradical.com/2013/04/03/why-is-fiat-money-valuable/

  62. Gravatar of Bill Tsfa Bill Tsfa
    13. May 2013 at 14:51

    Lets consider the two ways that fiat currency obtains value. One way is that the government sets its value and forces you to accept it. The other way…. is that an artificial “need” is created. Lets use cow-manure (shit) as an example and see how it can be transformed into something precious and highly coveted.

    1-You need to fertilize a field and need 50 lbs of manure.

    2-Your cow can only produce 1 lbs a day, so you go to your neighbor and borrow 50 lbs of cow-manure.

    3-Your neighbor gives you the 50 lbs of cow-shit but tells you that you must pay him back with 100 lbs at the end of the month (interest).

    4-Your cow has only produced only 30 lbs of manure by the end of the month. You need an additional 70 lbs to pay back your debt.

    5- You go back to your neighbor and ask to borrow another 70 lbs of manure .

    6-You pay your neighbor back with the 30 lbs of cow-shit you actually have, plus the additional 70 lbs he lent you.

    7-You have paid back the debt of 100 lbs of cow-shit owed… but you had to borrow an additional 70 lbs of cow-shit to do it.

    8- For the 70 lbs of manure you borrowed from your neighbor, you must now pay back 140 lbs next month.

    9- Your cow still only produces 30 lbs per month, so you will have to borrow 110 lbs of manure at the end of next month to pay back your debt.

    10-You will never have enough manure to pay back all your debt. Cow-shit is now the most valuable thing in your life and you can never have enough of it. You will spend the rest of your life working for shit.

    I just created a virtual monetary system based on Cow-Shit. Not really… This system is really based on debt and you can substitute anything in the place of manure including gold or paper. The key to making this system work, is to set a rate of interest that exceeds that which can actually be paid back.

    http://libertythinkers.com/education/gold-standard-myths-and-facts-part-2/

Leave a Reply