Princeton Rules the World
I just added a FAQ section on the right side of the blog. But I don’t plan to answer comments over there.
Today I’ll start with a very brief bio of four uber-important economists who were together at Princeton back around the turn of the century. Since I can’t go five minutes without talking about myself, I’ll add a brief explanation of how their work relates to mine. Then I’ll use their careers to speculate about what an ideal FOMC would look like. Keep in mind I don’t know any of these people, and thus the portraits will be very sketchy.
1. Michael Woodford
Woodford’s claim to fame is his book Interest and Prices, which has become the standard new Keynesian reference work for upper level macroeconomics. The book’s strength is the emphasis on expectations. Woodford shows that what really matters is not changes in the current stance of monetary policy, but rather changes in the expected future path of policy. The weakness is in the title. It’s a book about monetary economics that pretty much ignores money. Woodford sees interest rates as both the instrument of policy, and the transmission mechanism. If prices are sticky in the short run it is possible to do monetary theory this way. But there is no room for error. Thus if you get in a liquidity trap then the solution is to promise to hold interest rates at zero for an extended period of time. But this only works if you have an explicit inflation target. If you don’t have such a target (and we don’t) then promising to hold rates at zero for an extended period of time is indistinguishable from a promise to drive us right into a Japanese-style secular deflation.
What about me? Woodford and his co-author Gauti Eggertsson did some interesting work on the importance of policy expectations, particularly during liquidity traps. Gauti later discovered that I had done some work on the Great Depression that used this insight, although of course I lacked their sophisticated model. Gauti and I discussed these ideas by email, and he was kind enough to cite three of my Depression papers in his 2008 AER piece. In a 1997 paper co-authored by Bernanke, Woodford pointed to a circularity problem in my 1995 futures targeting model. But they failed to notice that there were two versions of my model, and that an earlier paper contained a version that was immune to this criticism.
2. Paul Krugman
Krugman’s claim to fame is his role as a public intellectual. He is the conscience of the Democratic party on economic issues, and the intellectual leader of the (old) Keynesian revival. His strength is his brilliance and his uncanny ability to see the essence of a public policy issue. He saw that the real danger was recession and deflation before most other economists, and saw that modern central banks were ill-equipped to handle a liquidity trap (something I missed.) One weakness is that he shoots from the hip too much. He offered a theory that a high wage policy might have promoted recovery from the Depression without bothering to check the data, which massively refutes this theory. Another weakness is that he sees issues in Manichean terms; good and evil. If you read his academic writings you’d think the best way out of the recession would be a bold policy of inflation targeting. But his passionate embrace of the social welfare state has led him to throw his lot in with fiscal expansion, and he has generally avoided the sort of thunderous moral condemnation of conservative central bankers that he routinely directs against conservative politicians.
What about me? Not much of a link. I did a paper in 1993 arguing that temporary currency injections would be expected to have little or no impact on the price level. In a much more sophisticated paper written in 1998, Krugman used a similar idea to explain the Japanese liquidity trap. Oh yes, I often criticize Krugman. But I have also said a number of good things about him. Of course some people are more likely to remember the bad things . . .
3. Ben Bernanke
Bernanke’s (original) claim to fame was his research on the Great Depression. His most influential paper was a study reviving Fisher’s idea of debt-deflation, and providing a more rigorous theoretical foundation for the idea that distress in the financial system could provide an independent channel through which deflation could impact the economy. Pretty ironic, eh? In my view his strengths lay elsewhere. He understood how high wages could have slowed recovery from the Depression, and also how unconventional techniques could allow monetary policy to be highly effective in a liquidity trap. His greatest weakness came out of his most famous paper. He overestimated the extent to which the financial crisis was reducing AD, and underestimated the extent to which falling AD was worsening the financial crisis. (See this earlier post, and also this great Congdon article that I hope to do a post on.) And as I mentioned, he and Woodford underestimated the potential of futures targeting, due to their worry about the “circularity problem.”
What about me? The only one of the four I’ve met. He was my first choice for Fed chairman. He’s read my Depression work and cited it. He and I have very similar views on the Great Depression, and did very similar research on gold ratios and what Bernanke called “multiple monetary equilibria” during the Depression. That’s when the same money supply might be compatible with two very different price levels, depending on expectations. Want an example? Think about what would happen if the monetary base stayed at this level and inflation expectations started rising.
4. Lars E.O. Svensson
Longtime readers of my blog may have noticed that Svensson’s name cropped up often in the first few months. I don’t think all that many of you (excluding professional economists) even knew who he is. His claim to fame is forward-looking monetary rules, that is, targeting the forecast. His strength is that he is one of the few economists to understand that all previous monetary targeting rules are mere stepping stones toward forecast targeting. Once you start thinking about policy this way, Taylor Rules and monetary aggregate targeting make no sense, they are clearly inferior to a policy rule that is expected to hit the policy target. His weakness is that he focused solely on forecast targeting policies using internal central bank forecasts (I presume he accepted Bernanke and Woodford’s flawed critique of futures targeting.)
What about me? In 1989 I wrote a paper envisioning a policy regime where central banks would target NGDP futures contracts. So do I deserve credit for the idea? Not really, Thompson (1982) and Hall (1983) got there before I did. Svensson also discussed “foolproof” escapes from liquidity traps, another area I have done research in. He cited the example of FDR’s program of dollar depreciation, something I have also studied extensively. And finally, there is the negative interest on bank reserves policy recently adopted by Sweden. I am pretty sure that it was Svensson’s idea. And which blog has been promoting this idea relentlessly all year? Are you beginning to understand why I found Svensson so intriguing? This further strengthens my conviction that once you start looking at the world from the perspective of targeting the forecast, the field of monetary economics looks very, very different.
What got me thinking about these four is the recent story that Svensson dissented from the Riksbank’s recent policy statement. But not because the decision to cut rates to 0.25% was too bold a move, rather because it wasn’t bold enough. It seems Svensson wanted to reduce the lending rate to zero, and one report suggested he wanted a penalty rate of 0.5% on reserves. That’s my kind of central banker! I have a source that tells me Svensson is known as favoring a more aggressive policy that Bernanke. He speculates these two may keep in touch, but isn’t sure.
This made me wonder what qualities we should look for in a central banker. And the more I thought about it, the further I ended up from my normal populist instincts. Senator Hruska may have been right about the acceptability of mediocre justices on the Supreme Court, but we can’t afford mediocre FOMC or ECB members.
Although Svensson’s views are closest to my own, I’m not so delusional as to think that we should have 12 Scott Sumner clones on the FOMC. What if I am wrong? A diversity of opinion is essential to any well run committee; otherwise you might as well turn things over to a dictator. (Compare China’s economy under Mao to the current committee-run regime.)
So let’s imagine an FOMC with the above 4 members, plus a bunch of other distinguished monetary economists; say people like Mishkin, Mankiw, Rogoff, Hall, McCallum, James Hamilton, etc. I don’t want any inflation hawks or inflation doves; I want people who call for tight money when tight money is needed and easy money when easy money is needed. And most importantly, nobody who believes monetary policy is ineffective in a liquidity trap. (Yes, I’m talking about Janet Yellen.)
Here’s my hypothesis. A committee made up of these 10 people would have been far more likely to adopt a highly expansionary monetary policy once the scale of last fall’s crash became apparent. There’s enough intellectual firepower there to understand the threat of rapidly falling NGDP, and also the need for policy credibility. I think they would have been able to coalesce around something like the 3% inflation trajectory (level targeting) proposed by Mankiw in his blog.
Monetary policy is incredibly complex. You have to look at issues from a lot of different perspectives. My fear is that even fairly bright people may get stuck in an intellectual rut, looking at policy from just one perspective. Good isn’t good enough, bright isn’t bright enough, we need people who are extremely bright, but also have the kind of mind that allows them to see the problem from different angles.
And here’s another hypothesis. Monetary policy may be the only important policy area where this is true. In other areas like health care, we don’t let a bunch of “wise men” (and women) make important decisions, rather we let Congress and the President decide. (Go ahead, insert a joke here.) The Supreme Court might be the closest parallel, but a mistake by the Supreme Court generally won’t create a worldwide recession, or depression.
And here’s another hypothesis. The Great Depression was caused by our failure to have the best and the brightest in charge on monetary policy. The western world needed people like Fisher, Keynes, Cassel, Hawtrey, Robertson, Einzig, etc, on their monetary policy-making committees. The US did have a great central banker in the 1920s, Governor Strong of the New York Fed. By force of personality Strong was able to dominate a bunch of incompetent hacks who stayed out of his way and let him run the show.
I am reading Lords of Finance (recommended to those interested in monetary history), and recently ran across this comment about Governor Strong:
“The two main domestic indicators that Strong had come to rely on to guide his credit decisions—the trend in prices and the level of business activity—argued that the Fed should ease.”
Note that he is talking about inflation and RGDP growth. And what is the sum of inflation and RGDP growth? You guessed it, NGDP growth. Strong died in late 1928. When I first read Friedman and Schwartz argue that things might have been much different had Strong lived I was dubious. Then I later came across the same comment from two other economists that I greatly respect; Irving Fisher, and Ralph Hawtrey. Keynes also was a big fan of Strong. Then I came across this passage from Lords of Finance describing Fed policy in the summer of 1930:
In the early summer the Fed stopped easing. It proved to be a mistake. For just as it went on hold, the economy embarked on a second down leg, industrial production falling by almost 10 percent between June and October. There is some debate about Harrison’s reasons. [Harrison replaced Strong.] Some argue that he thought he had done enough. Having staved off catastrophe by pumping a large amount of money into the system and cutting rates to an unprecedented low level, he believed that he had been as aggressive as he could. Others argue that he was operating with what might be called a faulty speedometer for gauging monetary policy. The usual indicators that he relied upon suggested that conditions were very easy—short-term rates were truly low and banks flush with excess cash. The problem was that some of these measures were no giving off the wrong signals. For example, when banks overflowed with surplus cash, this was generally an index, in a more stable and settled economic environment, the Fed had pushed more than enough reserves into the system to restart it. In 1930, however, in the wake of the crash, banks had begun carrying larger cash balances as a precaution against further disasters, and excess bank reserves were more a symptom of how gun-shy banks had become and less how easy the Fed had been.
Thank God we no longer have central banks thinking that during a period of rapidly falling NGDP it is enough to reduce interest rates to very low levels and pump lots of cash into the system!
Oh wait . . .
Seriously, when I read this I was struck by how similar it was to what I have been saying. Not just one or two sentences, but all of it. Someone either understands that the only reliable speedometer is NGDP (or inflation) or they don’t. Governor Strong got it. So did the author of Lords of Finance (Liaquat Ahamed), and so did Friedman and Schwartz. Strong’s death really was a tragedy.
The elite bankers and financiers of Wall Street were pretty smart people. So were the central bankers of the US, Britain, and France. But they weren’t smart enough. After Strong died the best central banker was probably Montegu Norman, but the central bank he led (in England) lacked the gold reserves to decisively influence world conditions. Today the world’s best central banker might be Lars Svensson, but he can’t even muster a majority in Sweden.
So the wealthy conservatives of the interwar period who dominated central banking dug their own graves, and the graves of millions of others. Not through greed but through ignorance. The few people who were able to see the big picture; Keynes, Fisher, Hawtrey, etc, did not have a voice in policy.
I don’t care how much is costs, even if we have to pay FOMC members a billion dollars a year, we will save much more money in the long run if we can get “strong” central bankers (pun intended) who have the vision to see what needs to be done, and who understand that effective policies require explicit target paths for macro aggregates
And a message to my Congressman, Barney Frank. Stop telling Bernanke he can’t adopt an explicit inflation target. You are in way over your head, and have no idea how much damage is done by your interfering in monetary policy.
Tags:
10. July 2009 at 19:51
Macroeconomic ideas don’t matter for macroeconomic policy. It is D.C that rules the world, and when Princeton meets D.C, D.C comes out on top.
10. July 2009 at 20:10
“The Great Depression was caused by our failure to have the best and the brightest in charge on monetary policy.”
Even though I don’t have any particular axe to grind when it comes to central banking, this is I think one of the biggest flaws with it. There’s a central point of failure. As you yourself have observed before, we might have actually been better off with a gold standard than central banking. Obviously the gold standard is not ideal, but what about free banking? If we have a whole market of competing equivalents of the FOMC, wouldn’t that be as good as, if not better than a single FOMC trying to target the market? I would be interested to hear your thoughts on free banking vis a vis central banking.
11. July 2009 at 04:00
When I saw Lacker speak last spring, he said that some members of the FOMC believe that there are problems with particular credit markets and so favor policies targetted at fixing those problems.
He said that other members of the FOMC see the problem as being more “aggregate” and that the market can handle the allocation of credit between markets. With short term rates already near zero, the Fed needs to lower long term rates. And so, the Fed is purchasing long term assets.
During the question period, I asked him about sweep accounts and how much of measured savings accounts are really checking accounts. He said that it doesn’t matter. There are a variety of assets with a more or less monetary character and… well, then.. something about interest rates.
Putting these remarks together, I took Lacker’s view to be interest rate centered, but not interested in targetting credit to troubled sectors but with getting down interest rates–particular long term interest rates.
Usually, an expansive monetary policy lowers all real interest rates, (though it may raise some nominal interest rates through the fisher effect.) But, with short term rates at zero, then monetary policy can’t lower short term term nominal interest rates. It might lower short term eal interest rates through higher expected inflation. And it can lower long term nominal interest rates by purchasing long term assets, and lower long term real interest rates both through lower nominal rates and higher expected inflation.
And while I assume that Lacker favors something like that, there are other members of the FOMC who instead want the Fed to funnel money into “troubled” sectors. I presume that housing finance is high on the list, but perhaps all securitization markets are part of the issue too.
11. July 2009 at 06:36
TGGP, I’m not sure it’s that simple, although of course you might be right. But here is a counterexample: The idea of inflation targeting was adopted by many of the world’s central banks in the 1980s. That idea came out of academia, not DC.
Johnleemk, Just to be clear, we did have a gold standard in the interwar period. But your basic point is right. After WWI we might have been better off with a gold standard and no central banks, as compared to a gold standard with a central bank. I would ranks systems this way:
1. Fiat money with central bank.
2. Gold standard w/o central bank.
3. Gold standard with central bank.
Off course this doesn’t change my observation about Strong. Given we had a Fed, it was far better to have someone like Strong in charge, as opposed to those who followed him.
Bill, In my view the best central bankers are those who don’t get bogged down in the minutia of credit markets, loans, interest rates, monetary aggregates, etc, and focus on the big picture, which is the trajectory of prices and/or nominal GDP. Lots of people understand that today, but not nearly enough. Back in the 1920s people who thought the central bank should stabilize prices or NGDP were viewed as oddballs. It’s a miracle we had one in charge of the Fed.
I’m not sure I agree about expansionary monetary policy lowering real interest rates. It is certainly true sometimes, but not always. And it is certainly not always true for long term real rates. There are cases where long term real rates have increased in response to powerful expansionary monetary shocks. This is to be expected because a powerful expansion shock may raise real GDP growth expectations.
You said:
“And while I assume that Lacker favors something like that, there are other members of the FOMC who instead want the Fed to funnel money into “troubled” sectors. I presume that housing finance is high on the list, but perhaps all securitization markets are part of the issue too.”
This is a good example of people confusing cause and effect—thinking the financial crisis was causing the sharp drop in AD, whereas actually the sharp drop in AD was causing the financial crisis. And this supports my point that a lot of people in the major central banks are simply not up to the job. One needs a “macro” perspective. As soon as they start focusing on sectoral issues, all is lost.
11. July 2009 at 07:47
Interesting. Of all the “dream team” FOMC members you cite, one, and one only, does not propose some type of inflation targeting in this environment. That one hold out, of course, just happens to be the Chairman of the FOMC.
So the question to you, Scott, is what gives? Spend a moment, an empathetic moment, putting yourself in Bernanke’s board chair. What caused this “road to Damascus” conversion to “credit easing” and away from inflation targeting? When did it happen? Why? What, if anything, might make him change his mind? These are the important questions, at least as long as Mr. Bernanke is the Fed Chair.
Some speculation:
Maybe Bernanke came to believe in “credit easing” as he understood more and more the impact of securitization markets as monetary transmission mechanisms, and this didn’t really happen until Bear Stearns failed.
Maybe he then came to think that this crisis was quite specific: a crisis of the securitization markets, in the same way the 1930’s crisis was one of bank failures (one of the two causes, sparked by Fed policy, that he cites in “Essay’s” for the Depression’s length and duration — the other one being rising real wages).
And so, he fixed on the securitization markets as the target of policy. Get them working at any cost. Get spreads down, bail out associated institutions, buy securities outright. Anything to repair the damage to the “mechanism”.
That doesn’t answer the question of why not ALSO throw in an inflation target to fight deflation. Why didn’t he? Why does he seem, all of a sudden, so reticent? Could be he’s reluctant to be seen monetizing deficits, but I don’t buy that. After all, his reticence pre-dated the Obamian explosion in the deficit. Is he, instead, all of a sudden afraid of the long-term inflation cost? If he is, he’s keeping it a secret as the output gap appears prominently in his speeches and Fed statements.
So what do you think, Scott? Why only this Princeton economist? Why the author of the 2002 BOJ deflation speech? Why is Bernanke so opposed to what he once proposed?
Why?
11. July 2009 at 07:57
Monetary policy is complex in the same sense that the allocation of goods in an economy is complex: If a government agency tries to manage it, you will end up with some Rube Goldbergian institution doing a terrible job. But if the government will simply let free markets (i.e., free banking) operate, the “incredibly complex” system will run itself, as it has many times.
11. July 2009 at 11:19
If the Great Depression was caused by the want of talented (“the best and the brightest”) individuals at the Fed, then what were the early failures of the Vietnam War caused by? 😉
We need great monetary economists to be on the FOMC, but this hardly prevents misguided focus (such as Harrison’s focus on “speculation” in the stock market) or complacency. Especially in a world where the Fed is, implicitly, under the thumb of the Congress, even with the best economists, mistakes will more than likely happen.
Great economists have proposed ideas that attempt to target asset-price bubbles in order to prevent the ruinous consequences of a burst. The moderate camps have said it might be beneficial to “lean against” a bubble, which might potentially distract Fed policy.
A sampling of those proposing policy responses to perceived asset bubbles:
Michael Bordo & Olivier Jeanne, “Boom-busts in Asset Prices, Economic Instability, and Monetary Policy” (2002)
Claude Borio & Philip Lowe, “Asset Prices, Financial and Monetary Stability: Exploring the Nexus” (2002)
Stephen Cecchetti, Hans Genberg, John Lipskiy & Sushi Wadhwani, “Asset Prices and Central Bank Policy” (2000)
Stephen Cecchetti, Hans Genberg, & Sushi Wadhwani, “Asset Prices in a Flexible Inflation Targeting Framework” (2002)
William Dupor “Nominal Price versus Asset Price Stabilization” (2002)
IMF “Asset Prices and the Business Cycle”, World Economic Outlook (May 2000)
11. July 2009 at 11:46
David, It may be because of his focus on the credit view, as you say. But there is another possibility–political interference from Congress. Congressmen like Barney Frank like to remind him of the Fed’s dual mandate. Bernanke has suggested that inflation targeting is a good idea in testimony to Congress, only to be shot down. So maybe he would like an explicit inflation target, but doesn’t think Congress will allow it
11. July 2009 at 13:05
Mike, I favor market-oriented monetary rules, and have many posts on the subject. The best is “Spot the flaw in NGDP futures targeting,” from a couple months back. But the world we live in now has central banks using discretion. If we are going to live in this world, we need really, really good central bankers. There really was a huge quality drop-off when we went from Strong to Harrison.
Lance, you might recall in my post that I said this conclusion went against my normal populist instincts. I ofter cite the Vietnam example against those who argue that if we just get good people into government, we can get good results. Monetary policy is different, in my view. It is far more abstract that foreign policy. Common-sense may be useful in foreign policy, which is why intellectuals often screw up. But common sense is worthless in monetary policy, and indeed is very dangerous, as the common-sense view is often 180 degrees off from the real situation.
Regarding your list of “great economists,” I think they are very good economists, but I still think my list is better. I have a great interest in monetary history, and I am really struck by how many of the greatest interwar economists understood what went wrong in the Great Depression, whereas a lot of B-level economists didn’t. I’m not sure this is as true today, but my hunch is that a committee of the best and brightest would have been more likely to see though all the superficial turmoil in the financial markets, to the deeper problem of falling AD worldwide. But I’d be the first to admit it’s just a hypothesis, and I have no proof.
11. July 2009 at 14:43
Scott:
“If we are going to live in this world, we need really, really good central bankers.”
That’s like a Soviet saying they need really, really good agricultural commissars, when what they really needed was to get the government out of the farming business.
The best people to run the Fed would be the ones who would allow private banks to issue various kinds of money (including paper), so that in periods of tight money, people can get money from competing private banks, and not have to hope that the Fed has deemed it appropriate to issue cash to them. If the fed lost all its customers in that world, then those same people should be willing (fat chance!) to close the fed down.
I wonder, when you say you favor market-oriented monetary rules, if you mean that you would actually favor letting the invisible hand reign in the money market? Or do you think “the best and the brightest” government officials should decide when and how money should be issued.
11. July 2009 at 21:29
[…] is the original: TheMoneyIllusion » Princeton Rules the World var $j = jQuery.noConflict(); $j(function() { $j(“#get-article-code”).css({opacity: 0}).hide(); […]
11. July 2009 at 21:55
Baby Jesus cries whenever “ueber”/“über” is written “uber”.
12. July 2009 at 02:54
Sproul:
The notion that monetary policy is like central planning is wrong headed. Monetary policy is like nationalizing the car industry and trying to control the quantity of cars supplied so that the market for cars clears at a fixed price of cars well above the cost of prodution. I think that the competitive production and pricing of cars is a much better approach, but it isn’t central planning of the economy.
I favor allowing banks to issue hand-to-hand currency if they want and people are willing to accept it. I also favor allowing people to use alternative monetary systems if they want. I have little confidence that it would have much macroeconomic impact.
The reality is that privately issued money (deposits or currency) almost cretainly requires redeemability. And then, whatever serves as the medium of redemption has macroeconomic significance. The most likely scenario for the adoption of monetary freedom is that Federal Reserve liabilities continue to serve as base money, and the quantity of Federal Reserve liabilities and the demand to hold them will determine the price level and disequilibrium between the quantity and the demand to hold them will impact aggregate nominal expenditure.
Realistically, changing that situation will require a constructivist reform of monetary institutions. For example, a return to the gold standard. That is, paying off
Federal Reserve liabilites with gold at some fixed dollar price. And then, all the private dollar liabilities being redeemed with the implied quantity of gold. Now, that doesn’t mean that alternative monetary arrangments must be forbidden. It is just that those would likely be irrelevant and there would be a gold standard with money being privately-issued paper redeemable in gold.
And then, of course, the supply and demand for gold would determine the price level and disequilibrium would result in changes in aggregate expenditure in the economy.
Allowing private banks to offer monetary liabilities as they wish is a good idea (I think.) But the question of monetary policy will continue to exist unless and until the issue of settlement media is resolved.
Scott favors what I call index futures convertibility Rather than redeeming money in gold, money issuers (the Fed in Scott’s version,) keep money interconvertible with index futures contracts on nominal income.
This system is completely consistent with allowing banks to issue hand-to-hand currency. And it doesn’t require that alternative monetary arrangements be forbidden (or restricted in any way.)
I think the only realistic scenario would be that banks would contintue to keep their liabilities (monetary and otherwise) redeemable with Federal Reserve liabilities, which, in turn, would be interconvertible with the index futures contracts. According to Scott’s version, the quantity of Fed liabilities would change with market determined trades in the futures contracts, and so they would be market determined. However, the goal of monetary policy (mildly inflationary growth of nominal income) is centrally determined. I sometimes describe the proposal as harnessing market forces to the goal of targettting nominal income growth.
I believe that the Federal Reserve can be cut out of the picture with the interconvertibility maintained directly by private banks. I also favor nominal income growing at a slower rate consistent with price level stability on average.
I certainly favor letting banks issue the specific monetary liabilities they want. And, further, people can start using gold ounces or euros if they prefer rather than dollars interconvertible with index futures contracts on dollar nominal income growing at 3% a year. But I don’t think that private issue of currency or allowing alternative monetary institutions would have any material impact of the macroeconomic performance of the system.
12. July 2009 at 04:32
Well, this one may suggest a reponse! :
http://neweconomicperspectives.blogspot.com/2009/07/why-negative-nominal-interest-rates.html
12. July 2009 at 05:50
Scott:
Just to be clear, in the post above you are not suggesting inflation targeting is a better option than nominal income targeting? I agree that an explicit 3% inflation target would certainly be an improvement over current policy, but ultimately a nominal income targeting rule would be better.
Regarding your all-star list, none of them really foresaw this crisis coming. (Krugman claims he did though.) For that reason, I would go with the economists at BIS who did see it coming and gave many warnings: William White, Claudio Borio, Andrew Filardo, and Phillip Lowe.
12. July 2009 at 06:42
Mike, I favor letting the market determine the quantity of money and interest rates. This is discussed in my “spot the flaw in NGDP futures targeting” post. Is that market-oriented enough? I would continue to have the central bank set the nominal anchor, as would many Austrian proponents of free banking. If I am not mistaken some Austrian free banking types want their system linked to a gold standard, or a productivity norm, or some other such nominal anchor. I am very much in that tradition.
But I also strongly disagree with your Soviet analogy. It is obvious to most good economists that the Soviet system was a bad one. But it is not obvious to most good economists that the Fed should be abolished. Thus it makes sense to try to make central banking do as little harm as possible, as long as we have it. Again, when central banking does very poorly (like 1929-33) the free market almost always gets blamed. You may have noticed that this has happened again.
Thanks Daniel, When my computer keyboard includes the German letter you mention, I’ll start using it. I did know know “ue” was correct, and I’ll try to remember to use it–but I am horrible with languages, and verbal memory. (Oddly I have an almost a photographic memory for numbers and graphs, well not quite photographic, but quite good.
Bill, I mostly agree, but I’m not sure there is any reason to tie currency issue to banking. Those are two totally unrelated activities. In the early 1990s I published a paper in the JMCB entitled “Privatizing the Mint.” Maybe I’ll do a post on the paper. I proposed the government auction off the rights to produce currency and coins of various denominations. The goal was to prevent wasteful non-price competition, as price competition in currency is difficult.
anon, I posted three different comments in 2 different places to that post, and none of them worked. So here’s a 4th attempt. The proposal would not drive interbank loan rates significantly negative, as banks could always exchange ERs for T-bills. And T-bill yields could not go significantly negative because non-bank holders of T-bills can always hold cash. So he doesn’t seem to have read my specific proposal on April 22nd. I’m guessing he just read a summary somewhere. The idea was to move ERs into cash in circulation. It is not intended to significantly lower rates, but rather relies on the monetarist “excess cash balance” mechanism.
Another mistake is to assume it would hurt bank profits. It could, but need not if the Fed doesn’t want it to. They could simply pay positive interest on RRs to offset the negative interest on ERs. All this is explained in this post:
http://blogsandwikis.bentley.edu/themoneyillusion/?p=1032
If anyone knows how to put this reply in his comment section, I would be grateful. I’ve given up. I suppose I should thank him for at least showing an interest in my ideas. Any publicity is good publicity . . .
David, No, I still favor NGDP targeting. I was suggesting an idea that I think a majority of those 10 would support, in fact I’m almost sure a majority would support the idea as I’ve read similar ideas (as Mankiw’s) from quite a few of them. As far as I know, only McCallum supports NGDP targeting.
Your second question is very interesting, and I’ll throw out a very counter-intuitive answer. The names you mention would be the absolute worst choices I could imagine. Since I believe in the EMH, I pay no attention to whether economists can predict crises. I think crises are unpredictable, and anyone who is successful just got lucky. But there is more. The names you mention presumably saw the financial crisis coming. But I think the biggest mistake we have made is to assume the financial crisis and the macro crisis of falling AD are one and the same. In fact they are different crises calling for different solutions. I am afraid that someone who predicted the financial crisis would think that in order to “fix the problem” we needed to fix the financial system. That was our policy, and it was a huge mistake.
Now I don’t know the specific views of the names I mentioned, and I obviously may be being unfair. Please interpret my views here as a generalization about the type of person that I am looking for, not specific people.
I might add that I think belief in the EMH is helpful, as then people would pay more attention to warning signals from the TIPS market. But anyone who could predict crises would probably be less likely to believe in the EMH. Nobody wants to believe their success was due to luck.
12. July 2009 at 06:46
Bill Woolsey:
The Fed maintains the value of the dollar above its cost of production? You mean in the sense that it costs $.03 to print $1? General Electric can print stock certificates for $.03 and sell them for $60 each. That doesn’t mean GE made a profit of $59.97 per share. The GE share is a liability of GE, just like the dollar is a liability of the Fed. If the Fed, or any similar institution, kept the value of the dollar above its cost of production, then rival institutions could earn a free lunch by issuing rival moneys of their own.
We agree about the broad principles of free banking. Would you agree that if private banks came up with some new money, convertible, inconvertible, or whatever, and if that money left people not wanting any green paper dollars, then there would be no harm in shutting down the Fed?
As far as free banking not having material impact, think of the great Depression. The Fed was too tight in its issue of money. Private banks, if they had been allowed, could have issued paper money, which would have alleviated the tight money conditions.
Government interference with the money market can do a great deal of damage. For example, in 1797 the (then privately owned) Bank of England could not afford to maintain convertibility of its pounds at the established rate of (about) .25 oz./pound, and so a bank run started, complete with an accompanying recession. The bank directors pleaded with the government to be allowed to suspend convertibility. When permission was finally granted, the run ended and normal business conditions quickly returned. It would have been better if the Bank Directors had been free of government rules about convertibility in the first place.
12. July 2009 at 08:17
“The proposal would not drive interbank loan rates significantly negative, as banks could always exchange ERs for T-bills. And T-bill yields could not go significantly negative because non-bank holders of T-bills can always hold cash.”
The fact that non-bank holders of t-bills can hold cash wouldn’t prevent banks from bidding t-bill yields negative.
It’s just a reason for non-bank holders of t-bills to hold t-bills no longer.
It’s not a reason for bank holders of t-bills not to accept negative yields.
12. July 2009 at 09:30
Scott:
“But it is not obvious to most good economists that the Fed should be abolished. Thus it makes sense to try to make central banking do as little harm as possible, as long as we have it.”
1000 years ago, it was not obvious to most good astronomers that the earth orbited the sun.
I’m with you on limiting the harm done by the Fed, but what’s wrong with the Soviet analogy? The Soviet government took over the production of wheat, just as our government took over the production of paper money and coins. I have no problem with the government offering a (non-binding) definition of what constitutes a bushel of wheat, or a dollar, but the government that insists on being the only producer of wheat–or money–is making a mistake.
12. July 2009 at 12:23
Sproul:
I am not sure I understand your point about GE stock, and I wasn’t making any claim about Federal Reserve profits. My point is that monetary policy isn’t the same thing as central banking. It is about producing one particular good.
I don’t know whether there would be harm or not if the demand for Federal Reserve notes fell off to zero because people chose to use some alternative medium of exchange. I think they should be permitted to do it, however. And the Fed should reduce the quantity of its liabilities as the demand falls until both fall to zero. I don’t think this will happen.
I also don’t think that free banking will do much to help when there is an excess demand for base money. I don’t believe that the demand for base money rose (and the money multiplier fell) during the Great Depression because people needed more hand to hand currency.
12. July 2009 at 12:38
A bit off topic comment– no response needed. Regarding the EMH belief that those who successfully predict are just getting lucky. Does this amount to suggesting that all speculators are suckers, yet all those suckers, in aggregate, are wise?
Mike, if competing private banks issued their own currencies, would that mean that when the cashier gives me change at the store I would need to evaluate the quality of the currency issuer? Since the credit of all banks would not be equal, wouldn’t their be speculative market among all the outstanding currencies?
Wouldn’t one bank inevitably win a monopolistic position? Because like software formats, people want to conduct business with others using the same software they are. The most popular currency would have a positive feedback loop in popularity. So we’d end back up with one big private bank.
12. July 2009 at 13:22
Scott:
You say that currency has nothing to do with banking. What?
The currency we use today is nearly all Federal Reserve notes which are liabilities of Federal Reserve _banks_. All of the Federal Resereve notes are matched by assets on the balance sheets of the Federal Reserve _banks_. Federal Reserve banks also have deposit accounts. Shifts between Federal Resreve notes and deposits have no impact on Federal Reserve assets and just shift the form of its liabilites.
I believe that this activity should be handled by ordinary commercial banks.
If banks are free to issue currency, then changes in the currency deposit ratio, especially seaonal changes, have no impact on bank asset portfolios. Banks simply change the form of their liabilities from deposits to banknotes and back.
The Federal Resreve was supposed to deal with this issue by advancing currency to banks. However, this got balled up with the real bills doctrine. And then T-bills were standard collateral. And how we see what happens when the Fed starts using this and that secret collateral of varying qualities.
If commerical banks issued their own currency they would just shift the composition of their liabilities without there being any impact on their assets. It is very convenient for the banks and gives them an incentive to issue banknotes if it is permitted.
Also, note, that it means that changes in the currency deposit ratio have no impact on the demand for base money and require no adjustments in the quantity of base money to avoid creating an excess supply of demand for base money.
The only question is whether or not the depositors will accept banknotes from their own bank. Which, of course, depends on whether or not retailers will accept them. Which, of course, depends on whether the retailer’s banks with accept them for deposit. I think the answer to that question is, obviously they will. They accept checks. And, of course, they did accept banknotes from other banks for deposit.
So, you have to argue that this market should be outlawed. That bankers should be threatened with jail or fines to prohibit these contracts. The burden of proof is on you, in my opinion.
Restricting freedom of contract, especially to ban competion by creating a government monopoly is usually a bad idea.
Banks did issue banknotes until it was outlawed. First, the Federal government limited it to national banks, and had a 100% U.S. government bond collateral requirement. This meant that banknotes issue required a change in bank asset portfolios. Government regulations destroyed much of the benefit of private currency issue. (It was finally outlawed
completely in the Great Depression.)
Many of the currency problems of the national banking era were created by the government bond collateral requirement. Allow banks to issue checkable deposits on the same terms as deposits, and the problems go away.
By the way, the regulations were not done because economists convinced the government there was a natural monopoly and competition is wasteful. The regulations were aimed at increasing the demand for U.S. government bonds.
Personally, I don’t think having more ATM machines and no fees would be a “waste.” Furthermore, I think it would be easy to credit interest for outstanindg banknotes to the accounts of the person who withdrew them until the time they clear. While that would involve more clearing costs, banknote holders would earn more interest than the do today.
I don’t think monopolizing the issue of currency for the govenrment is a desirable source of government revenue. I favor explicit taxes whose immediate burden is clear to tazpayer-voters.
Using currency as a source of revenue creates an incentive towards inflation.
I favor mimizing the base of any inflation tax. I don’t want the govenrment to have any financial incentive to create inflation.
Of course, currently we have a situation where issuing currency is not profitable. With private currency issue, the banks would simply stop and we would still have deposit money at maybe negative interest rates. (Well, that requires that the government get out of the currency business altogether.. which it should.)
Anyway, I think currency was connected to banking until the govenrment intervenend. I think there are obvious economies of scope.
Everyone already gets their currency from banks. (Well, I get mine from the grocery store, but they get it from their bank.)
12. July 2009 at 14:04
Rob:
With competitive note issue, you do need to worry about the quality of the change you receive. If it is a significant sum and you plan to hold the funds, depositing it might be a good idea.
There is no speculative market in currencies because of interbank clearing. Think about checks. Is there a speculative market in checks drawn on BoA? No. They are deposited in other banks and cleared at par. If BoA closes, then no one accepts checks drawn on BoA at all.
Is there a tendency for monopoly? No. The tradition is for each bank to issue both notes and deposits. Checks and notes are accepted by retailers at par. They deposit them in their own banks. They are all cleared.
People only lend to the business with the best credit, and it produces everyting? Well.. maybe.. but there are diseconomies of scale.
If there is free entry in banking, then usually there are a good number of large branched banks issuing a variety of deposits as well as currency.
I believe the best way to think about currency with competitive note issue is like checks. While “uncashed” checks aren’t an important store of wealth, they are used to make payments–over 90% of them by dollar value.
Trying to think about today’s monetary system as if the checks are an investment vehicle is a mistake.
Of course, once all of this paper currency goes away and everyone uses debit cards, it won’t be an issue.
Still, it would seem to me that the question of receiving change would create an incentive to use small denomination bills to minimize change. (Using a $1,000,000 bill to buy a candybar and then having $999,999 of currency I might not trust is a bad idea. Maybe I should carry a $1 bill.)
Of course, I think the obvious approach is to pay interest to the deposit against which notes are withdrawn until they clear. Which means that everyone who passes on currency recieved in payment allow someone else to continue to earn interest.
If we think about change, there is actually an incentive for retailers to pay out the change they withdrawal from their bank rather than pass on the small bills received in payment. Those should be deposited. Interesting.
Most retail business obtain small denomination currency from their banks every day in order to make change. And, of course, they make daily deposits of the checks and currency they recieve. And, they also pass on some small denomination currency they received as change.
I have always thought that there would be more of the first and less of the second. Larger withdrawals of currency. Larger deposits of currency. And more gross clearings.
But maybe not. If people use more small denomination notes, then maybe there would be less change.
12. July 2009 at 14:35
Bill Woolsey:
My point about GE stock was that GE does not, in any meaningful sense, maintain the value of GE stock above its cost of production. Likewise, the Fed cannot maintain the value of paper dollars above their cost of production. GE’s stock price is determined by GE’s assets and liabilities, and the value of green paper dollars is determined by the Fed’s assets and liabilities. You made a related point yourself: “All of the Federal Resereve notes are matched by assets on the balance sheets of the Federal Reserve _banks_.”
“I also don’t think that free banking will do much to help when there is an excess demand for base money.”
How could free banking fail to help? For example, during Christmas season, people will come to their bank asking for more paper money. As long as those customers offer sufficient collateral in exchange, a free banker will eagerly oblige—in contrast to a Princeton-educated central banker, who might have some pet theory that leads him to restrict the issue of cash.
Good job of making the case for free banking in your post to Scott, by the way.
12. July 2009 at 14:40
Rob:
“Mike, if competing private banks issued their own currencies, would that mean that when the cashier gives me change at the store I would need to evaluate the quality of the currency issuer?”
Correct. We do that now when we accept checks, credit cards, Toys-r-us dollars, etc. We have lots of historical precedent for privately-issued paper money. There’s the US Free banking period (1836-61) and the Scottish free banking period (covered by Lawrence White, if I remember right).
12. July 2009 at 14:51
As a non-economist, I have been pondering and trying to understand the assertion here that the current problem is an insufficient money supply. The argument, as I understand it is:
1. Banks aren’t lending sufficiently because there is greater incentive to hold reserves and receive interest on them.
2. Meaning the argument by some that banks aren’t lending because companies don’t want to borrow is bogus?
3. Meaning companies would happily be borrowing more if banks would lend…?
4. Meaning business investment would increase, payrolls would increase, and consumers, having more money, would spend more….?
Is that correct? Or is it more complicated than that?
And I guess the argument from the fiscal stimulus camp is that the cycle works in reverse order:
Companies and consumers need more cash -> stimulating spending -> stimulating borrowing demand… ?
Does that mean the crux of the debate is mainly about the veracity of point 2 above?
12. July 2009 at 15:35
Rob:
The demand for money is the quantity of money that people want to hold. If the quantity of money that exists is less than the amount people want to hold, they spend less. The decrease in spending reduces nominal income. Lower nominal income reduces the amount of money people want to hold. And so, the amount of money people want to hold is equal to the existing quanity of money. However, nominal income should be higher. And so, there is a situation where if nominal income were at its higher “target” level, there is a shortage of money. The amount of money people would want to hold at that higher, desirable level of nominal income, is great than the current quantity of money. The quantity of money is too low.
None of this is directly related to the amount that businesses want to borrow. It is confusing because in business “the demand for money” is the amount of money people want to borrow and the “supply of money” is the amount available for loans.
But, in monetary economics, “the demand for money” is the amount of money people want to hold. And the “supply of money” or “quantity of money” is the amount that exists.
Nearly all the money that exists is created by banks or the Fed. The money is liabilities to them. They match those liabilities with assets, which are loans. And so, the quantity of money and the supply of loans are closely related.
However, the demand to hold money (how much of these monetary liabilites people want to hold) and the demand for loans are not closely related. People usually obtain the money they hold by selling things (like their labor,) not by borrowing. And when people borrow, it is usually in order to spend the borrowed money rather than hold it.
Nominal income is the dollar amount of all the wages, rents, interest and profits earned. It is also equal to the total spending on consumer goods and services, capital goods, and govenrment goods (measured in such a way that exports need to be added and imports subtracted.)
Nominal income is lower today than it was a year ago. With past rates of porductivity growth, it shold be about 3% higher without inflation. With past rates of inflation, it should be about 5% higher.
The quantity of money that exists today is too low relative to the amount that would be demanded if nominal income were at the level it was last year, and it would need to be higher still to be equal to the demand for money that would exist if nominal income were where it would be if it grew 3%. And too little by more if nominal income had continued on its trend of 5% growth– 3% growth in productivty and 2% inflation.
13. July 2009 at 05:03
Mike, In what sense is cash a liability of the Fed? I thought once we left he gold standard the Fed was no longer required to redeem dollars?
I don’t get the GE analogy. Printing more stock certificates doesn’t help GE, but printing more money brings real inflation tax revenue (seignorage) into the government coffers.
anon, Your argument implies the world’s entire stock of T-bills would be held by banks. I doubt that, but T-bill yields are also closely linked to the expected rate of return on short term T-notes. And certainly banks would not hold the entire stock of both assets.
I don’t want to make a big issue of this, as it certainly would not affect my argument even if T-bill yields went significantly negative. Indeed Keynesians like Mankiw would like my idea even more if rates went negative. I seem to recall that his concern was that my proposal would not make T-bill yields negative.
I do agree that they would go slightly negative, but I very much doubt they’d go to negative 2%, as there are too many other lucrative assets that banks could hold.
Mike#2, During the communist era there were lots of countries that did not go for central planning. On the other hand I know of no country that uses private, unregulated money. Let’s see how it works in a small country first, before we take a chance with the biggest economy in the world. What if it led to a Great Depression? I don’t see any reason why it couldn’t lead to such a result. Sure, it doesn’t seem likely, but when we set up the Fed in 1913 most people didn’t anticipate the Great Depression. I’ve read lots of free banking literature, and I have yet to see a persuasive argument that it would stabilize the economy and prevent depressions.
Rob, That’s a good question and I don’t know the answer. Here are two possible answers:
1. Yes.
2. No, speculators may have a bit of inside information because they see a bit of the economy that others don’t see. That gives them a slight advantage, which is enough for them to do slightly better than chance, but not enough to materially affect the EMH. In others words, the EMH is still approximately true (which is all I have ever claimed.)
Does anyone know how EMH proponents like Fama would answer Rob’s question?
Mike, When the US had free banking in the 1800s currency notes from different banks did have different values. Some notes did trade at a discount. It is possible that this problem would not occur today, I simply don’t know.
Bill, Despite the title “bank” I don’t consider Federal Reserve banks to be banks at all. Until recently they made very few loans. More than 99% of the base was injected in OMOs. That’s not banking, that’s legalized counterfeiting.
You raise two separate issues.
1. Should competition in currency be allowed?
2. Would banks win in that competitive environment.
I don’t have strong views either way on the first question. I can certainly see advantages to auctioning off seignorage rights for various denominations. But you have good arguments as well.
If we do go to competition in currency, then it is possible that banks might succeed. I certainly would have no objective in their issuing currency. But I would trust currency issued by Walmart, Microsoft or GE much more than I would trust currency issued by Bank of America or Citibank. I think the latter type of currency notes would trade at a steep discount to the dollar, as those banks are almost bankrupt.
But overall I don’t strongly disagree with anything you say, I think there is a strong argument to be made for competition in currency. (I assume we are both discussing this option within a framework where the governments determines and manages the nominal anchor for the system.)
Mike#3, I don’t see why you assume currency is backed by Fed assets. Suppose there is a trillion in currency and it is backed by 1 trillion in long term T-bonds. Now let interest rates go from 2% to 4% overnight, and assume the value of the bonds falls to $600 billion. Now the currency is no longer backed by assets. But what changes? Nothing, I would say.
I don’t think Bill is defending your position. He is discussing a free banking system where there is a central bank controlling the supply of bank reserves, unless I am mistaken.
Rob, To understand monetary policy you need to forget about banks and loans. If the price of apples is too high you need more apples, it has nothing to do with banks. And if the price of cash is too high you need more cash, it has nothing to do with banks. By definition, deflation is a period when the value of cash is rising. For a healthy US economy, right now we would need the value of cash to fall at about 2% a year. That is another way of saying we need 2% inflation. The argument is about supply and demand, not banks and loans. Check out my post from a few months back entitled a short course in monetary economics.
Bill, I agree. That is another way of making the same point to Rob.
13. July 2009 at 05:25
“The usual indicators that he relied upon suggested that conditions were very easy “” short-term rates were truly low and banks flush with excess cash. The problem was that some of these measures were no giving off the wrong signals.”
Precisely what does this say about Larry Summers’ and Tim Geithner’s argument that the key to restarting the economy was to “recapitalize the banks” – in other words, absorb the bank losses (one way or another), so they have adequate capital to support lending?
This mantra – by far the dominant press narrative of the time, was seen as the way to avoid a “lost decade” comparable to Japan’s. The incessant comparison to Japan drowned out everyone else’s voice.
The other side in the arguments contended that the majority of banks had adequate capital, but had a) tightened up lending standards (probably a good thing, to a degree) and b) encountered a scarcity of credit-worthy borrowers seeking credit. But no one believed this… No one seemed able to believe that the american consumer/small business/big business could actually wean itself away from credit-dependence. Indeed, no one WANTED the american consumer to do this… Not really, though they paid lip service to the notion.
But then our savings rate jumped EVEN WHILE the unemployment rate jumped (meaning that those people who were employed were saving EVEN MORE to compensate for all the unemployed people dissaving). So, the question is, have Geithner and Summers re-thought their dogma? And Bernanke?
13. July 2009 at 06:07
Scott:
#1: “In what sense is cash a liability of the Fed? I thought once we left he gold standard the Fed was no longer required to redeem dollars?”
The fed does not buy back its dollars with gold, but it does buy them back with bonds, and the Fed does have a huge stock of gold. If the time came to unwind the Fed, the gold could be used to buy back dollars. This is explained in my paper “There’s No Such Thing as Fiat Money”, which you can find by clicking my name above.
There’s also the fact that green paper dollars appear on the liability side of the Fed’s balance sheet.
You have to recognize that there are many kinds of convertibility. In the old days, private banks used to suspend convertibility of their notes into gold for 2 days every weekend. Nobody claimed that the notes suddenly became unbacked over the weekend, because the bank still had its gold and its bonds. The fed has suspended physical convertibility into gold for 76 years, but the gold and bonds are still in its vault. The difference between 2 days and 76 years is one of degree. The dollar is backed but physically inconvertible.
Convertibility can be physical (the bank buys back its notes with gold) financial (the bank buys back its notes with bonds), instant, delayed, certain, uncertain, at the customer’s option, at the bank’s option, etc. In 1933 the Fed suspended one kind of convertibility: Instant physical convertibility at the customer’s option. The other kinds of convertibility remain in place.
“Printing more stock certificates doesn’t help GE, but printing more money brings real inflation tax revenue (seignorage) into the government coffers”
Every dollar that the fed issues gives the Fed a dollar’s worth of liability (Same for GE when it issues stock). There is no such thing as seignorage. There is also no such thing as fiat money, since fiat money is money whose whole value is seignorage. If any institution did earn seignorage, rival institutions would issue competing moneys until the seignorage was driven to zero.
#2) Free banking, as you know, has a long history–a history that does not include the great depression. That’s more than I can say for the Fed. Besides, I don’t think you have any objection to allowing privately-issued money. If the Fed survives in fair competition then fine, but right now the Fed won’t allow competition.
#3)”Now let interest rates go from 2% to 4% overnight, and assume the value of the bonds falls to $600 billion. Now the currency is no longer backed by assets.”
When assets lose value, the currency loses value. If a bank has issued $100, backed by 100 oz. worth of bonds, and those bonds fall in value to 40 oz, then the dollar will fall from 1 oz/$ to .4 oz./$.
13. July 2009 at 06:09
David Pearson:
“Why is Bernanke so opposed to what he once proposed?”
Suggested Answer: Fear of dollar flight and commodity-driven inflation.
In chatting with a professional stock broker yesterday, he expressed the belief that commodities are driving everything right now. And commodities are driven by the dollar.
My perception is this: Bernanke is defending a peg. (It’s not an official peg, but it’s effectively a peg because he’s trying to avoid inflation >2%.) And keeping inflation that low requires very limited monetary expansion, since perception of monetary expansion could lead to dollar flight, then higher commodity costs, and commodity-driven inflation. Since much commodity demand is inelastic, this could drive up the trade deficit and prices simultaneously, and the perception is that this would kill any recovery (of course, this is somewhat short sighted – it really depends on whether driving up commodity prices helps with jobs, asset values, and home values, which depends on a lot of other factors).
What Bernanke is attempting right now (and I might agree with it, if it were on a larger scale) is TARGETED reflation. Specifically, housing prices (via direct mortgage lending support), keeping Federal borrowing-costs low, etc…
In a sense, this is a poor-man’s version of capital controls (which arguably worked in the south-east asian banking crisis in the 90s). I don’t think it will work here because it isn’t nearly robust enough (way too porous), and I suspect that the net effect will be to end up with slightly lower consumer/homeowner debt, and massively higher federal debt (largely due to tax revenue depletion) – leading, of course, to expectations of future default/monetization.
And it’s probably all doomed to fail because at some level we need to restore spending power (which was sustained only by an unsustainable rate of credit _expansion_). That means decreasing debt, and that means creditors are going to take a bath… but here’s the kicker:
Almost everything that Summers, Bernanke, and Geithner have done (except, possibly, GM’s bankruptcy) has had the effect of making sure that creditors get their money, and the value of that money stays intact. (The same, btw, is true of Germany right now.) Whether this is by accident or design, who knows. But generally speaking, creditors want to extract their money as fast as they can, sit on it, and make sure it doesn’t lose value.
Retrospectively, we can observe that the ONLY reason the stimulus passed (in the smallest possible form) was the perception that if it did not, then creditors would be even worse off. In a sense, creditors want us to take the minimum amount of fiscal/monetary action that will support the current system long enough to extract their money from it. This is the real meaning of zombie banks…
Regarding paying interest rates on deposits, I’ve speculated that this was done to help bank earnings to help offset loan losses (aka, slip them a bit of extra cash without the general public seeing). It’s also possible the Fed is doing this to encourage banks to park their money onshore. If the Fed established a negative penalty, banks might convert reserves into foreign currency en-masse, which could initiate the dollar-run that they most fear.
13. July 2009 at 07:12
Scott:
Federal Reserve notes are a liability of the Fed. As long as the Fed is committed to reducing the quantity of them if the demand for them falls, then they are liabilities.
For example, if there is nominal income targeting, then the Fed is obligated to reduce the quantity of base money if there is lower demand.
If, of course, the Fed is willing to let the price level rise whatever amount necessary so that the real quanity of base money falls to meet the real demand, then it isn’t a liability and they can just print money and spend it.
If the monetary authority is instead committed to monetary stability, (nominal income targetting, for example,) then the base money it issues is a form of borrowing. They have to stand ready to pay the money back.
Having an indpendent Fed, holding assets (like govenrment bonds) allows the Fed to sell off those assets (or collect on them as they come due) and so reduce the quantity of them if the demand falls.
If the central bank is just a branch of the treasury, then the treasury has to sell bonds or run a budget surplus reduce the quantity of base money. But that still means they respresent a form of borrowing–liabilities.
Scott, when you make statements like this, it makes me think that you don’t understand Krugman’s point about an irresponsible central bank.
13. July 2009 at 07:24
Scott:
If people were holding currency as a store of wealth, then I could understand people choosing Walmart over BoA. But people don’t hold currency as an investment vehicle.
Where do you get currency? From your bank deposit.
What do you do with it? Spend it.
What do the retailers do with it? Deposit it at their banks.
Payments by check make up 90% of payments by value. Do people hold checks received in payment as investment vehicles? Do people refuse to accept checks from risky banks?
Anyway, if Walmart issued currency, it would be borrowing by the issue of currency. It would be obligate to pay that money back. If it currency was larger than the amount it wanted to borrow to finance its own activities (I see that you don’t see any benefit to having your bank diversify,) then it would have to purchase other sorts of assets and operate as a financial intermediary. Walmart would be bank.
13. July 2009 at 07:26
Scott:
I don’t favor regulations separating banking and industry. If a conglomerate including retail operations and banking operations is sucessful–fine by me.
But the banking operations are banking operations.
13. July 2009 at 08:31
Scott:
Regarding your comment regarding economists predictions of crisis and the EMH.
Do you have to reject EMH to believe that these economists were correct in predicting imminent trouble? I think this might be getting tripped up in what critics often get tripped up in: The joint-hypothesis of the EMH (market equilibrium and market efficiency). You can reject the former, without succesfully rejecting the latter and vice-versa.
The important question is whether these economists were pointing towards market disequilibrium or market inefficiency.
Rob,
as to speculators–it depends. In less liquid markets (like housing), you probably can be a significantly succesful speculator if you are smart and lucky. But in liquid markets (like capital markets), the chance for speculation is much more prone to risk and failure. Pointing to succesful speculators is more evident of the survivor bias than any clink in the armor of EMH. Speculators may co-exist with the statement that riskless abritrage opportunities are very limited in liquid capital markets, as they take significant downside risk when they speculate.
13. July 2009 at 08:43
What if there were a run on the US dollar in the near term? How would that change the current monetary picture? Would it effectively increase the base?
13. July 2009 at 08:48
Stats Guy,
Its possible that Bernanke is worried about dollar flight, but also highly, highly ironic.
Bernanke’s criticism of the 1930’s Fed was essentially that they paid too much attention to the dollar. He argued that competitive devaluations, in aggregate, were preferable given their positive impact on global AD. In his “Essay’s” book, he specifically highlights examples of European countries that exited the Depression earlier due to their devaluations.
So now, the great expert on the Great Depression has changed his mind? Wow. If only he would come out and admit that his academic work was misguided. Unfortunately, it will never happen.
Bottom-line:
Bernanke’s reticence to adopt real monetary stimulus is evidence that things look different sitting in the Fed boardroom than they do sitting in a University office. The reason is tail risk. Its one thing for you to dismiss capital flight and a dollar crisis as virtually impossible; its another to roll the dice and take the risk with your own capital. In this case, the “capital” that matters is history’s view of Bernanke’s actions.
The problem is that, by worrying about inflation tail risk, Bernanke invites deflation tail risk. This is the product of Fed management of the economy. It is the end point of a process, not a discrete event.
13. July 2009 at 09:30
For what it’s worth I have argued against Mike Sproul’s “There is no such things as Fiat Money” position before. I’ll try just once more….
The value of money is that it is a hedge against uncertainty. It can be sold in exchange for a great many commodities. As a result an agent need not decide _which_ commodities immediately.
For this purpose I must, as a resident of Ireland, use Euros. I may not use dollars or pounds since these are not legal tender here. If there were competition in currency then I may use other sources of currency should other currencies become widespread.
The phrase “backed but physically inconvertible” is meaningless. If something is backed but inconvertable then in what way is it backed? It is like saying there is an angel inside a block of marble and it only awaits release.
In reality the Fed’s assets are not relevant. It could have none and it wouldn’t matter. What matters is firstly the force that the US government control to restrict other currencies should they arise. Secondly, the Fed’s control of the money supply means that there is little chance the dollar will become worthless due to too many of them being issued overnight.
What you are doing Mike is confusing what happens in a Free-banking regime to what happens in a State-backed Fiat-currency regime.
13. July 2009 at 10:43
The need for highly intelligent central bankers implies the need for discretion. What situations would you need central bankers to deviate from NGDP futures targeting? If all you need them to do is follow a rule, mediocre bankers would be sufficient. This post seems to be a shift from your earlier thinking, but maybe I’m just missing something.
13. July 2009 at 14:02
Current:
Legal tender laws can’t prevent currencies from crossing borders. I can go just about anywhere in South America and spend US dollars freely, and yet I suppose you’d claim that all of those South American currencies are fiat moneys–valued just because the government limits their supply and prohibits other currencies.
There’s nothing meaningless about the fact that central banks can and do use their bonds to buy back their currencies. The Bank of England suspended physical convertibility in 1797 and resumed it in 1821. Would you claim that the Bank’s assets suddenly became irrelevant in 1797, and just as suddenly became relevant in 1821? Or would you admit that the pound was backed but physically inconvertible for those 24 years? (It was in 1810, by the way, that David Ricardo popularized the notion that money can have value simply because the government only issues so much of it–a perfect circle of an argument.)
What you are doing is claiming that money, alone among all financial securities, is valued not because it is backed by assets, but simply because the government limits its supply and prohibits rival moneys.
13. July 2009 at 14:12
David:
“Bernanke’s criticism of the 1930’s Fed was essentially that they paid too much attention to the dollar.”
The US fiscal picture in the 30s was different than it is today, but even more importantly, the world fiscal picture was different. The US in the 30s had amassed a large amount of gold through long running current-account surpluses. The US today has long-running deficits and large liabilities (foreign owners of US debt). US debt load, even in the mid-30s, was lower. Dollar flight today is much more of a risk than it was in the 30s (it’s not a tail risk, it’s one of the most talked about events among trading circles).
Bill:
“Payments by check make up 90% of payments by value.”
Could you elaborate on what exactly you consider a payment? I know this sounds silly, but is electronic deposit of my monthly salary a payment? Is it a payment by “check”? In the purchase of a home, when I take out a mortgage and wire money to the bank, or
Regarding Wal-Mart currency:
For many intents and purposes, it does issue currency. It’s called stock. Stock can be exchanged at a level comparable to other currencies. Stock represents a claim on the earnings power of walmart (much like the dollar represents a claim on the earnings power of the US treasury). Stock returns dividends (sometimes), but currency returns interest. The two are somewhat linked.
As walmart stock goes up in value, total wealth for shareholders and spending power increases… For this reason, some people call stock an investment and currency NOT an investment.
This view is myopic. From the perspective of someone living in a small island nation, as the earnings of the US economy increase (assuming the stock of dollars is fixed) those dollars get more valuable too (subject to the quantity of stock)! So if I lived in a foreign nation, holding dollars IS an investment in the productivity of the US. Like the US treasury, Walmart can issue new stock and invest proceeds from that offering in the company (just like the US can invest in the country)… if the company does well (or is expected to make good use of the new money), the value of current stock increases.
Asset valuation is something that general equillibrium models have failed to really address well – namely, that global asset valuations for highly liquid assets behave like currency in many ways (a store of wealth being one). And currency behaves like an investment (in the productivity a country).
Perhaps the largest single difference between Walmart stock and US dollars is something that no general theories (including monetarists) address perfectly either – transactions are settled in the currency, and thus altering the value of the currency alters the value of all contracts linked to it. This is different from simply saying that there’s a demand for currency to sustain transactions.
Frankly, this is something that most monetarists seem to viscerally (ideologically) object to… But it’s actually crucial. One reason for the Fed to have a legal monopoly on money that is used to settle common transactions within the US is to prevent bank currency bubbles (e.g. the Wildcat banks) and enable a strong coordination function in monetary policy because (frankly) markets don’t work perfectly all the time… (If they did, we wouldn’t be in this mess – plain and simple; you can argue the real estate bubble was created by the Fed, but the fact of the matter is that no one FORCED banks to make lousy loans or people to buy at inflated prices… EMH only applies _most_ of the time, and only to a certain degree)
Arguably, the problem right now is that the Fed does _not_ have a legal monopoly, because the dollar competes with other currencies. Friedmanomicons LIKE this aspect of the current international financial system because they believe it disciplines (and limits) size of government.
But frankly, we’ve seen a massive episode of market failure (EMH had a hiccup). It was accompanied by a massive wealth shock that leaves people indebted because they borrowed based on expected future earnings/asset values, and the only way to get out of this is through mass liquidation (yah, that’s fun) or by shocking the system by devaluing debt. The plummet in asset values very much mirrors the plummet in currency wealth (e.g. money) due to the direct liquidity of those assets.
There are three ways out: earn our way out, inflation, or default. But I suspect we’ve passed the Event Horizon where debt payments exceed revenue for eternity. The best way out is to devalue debt, but it’s quite possible that 2-3% inflation is not an option (it’s a tipping point, so we either have deflation or a currency run that generates high inflation). I vote #2.
Austrians (and most monetarists) get hives when we talk about devaluing debt, and so far creditors have been able to use the political process (in combination with the leverage presented by world financial market integration, bond markets, and commodity speculation) to “discipline” the US government. But all it means is that we’ll end up slowly strangling the economy by stacking a default-risk on top of T-bills (and implicitly all dollar-denominated lending rates) while growing other economies (e.g. China) until either the other economies fail (possible) OR until the US finally sucks it up and monetizes enough of the debt so that it’s manageable and future default risk is low again.
But the harsh reality of this is something that virtually all economists object to on philosophical grounds: the root of money is power, not free exchange. That means that in a world where the dominant political power enforces debt that is passed down from one generation to the next without recourse (as can the national debt), free-exchange can leave one very much not free… That’s what we call serfdom.
14. July 2009 at 01:48
[…] rule the world By Amol Agrawal Scott Sumner’s Blog is becoming a big favorite. His recent post reviews work of 4 leading Princeton economists- Paul Krugman (no intro needed), Ben Bernanke (again […]
14. July 2009 at 02:14
Statsguy: “If they did, we wouldn’t be in this mess – plain and simple; you can argue the real estate bubble was created by the Fed, but the fact of the matter is that no one FORCED banks to make lousy loans or people to buy at inflated prices… EMH only applies _most_ of the time, and only to a certain degree”
I don’t think you understand the dynamic involved. Let’s suppose that we have a set of about twelve major banks, as we had in the UK. Most of these banks know that it is not sustainable to make home-loans at a rate close to the current central bank interest rate.
A couple of them, let’s hypothetically call them “Northern Rock” or “Bradford & Bingley”, as not so smart. What is the outcome? Well, in the short term the reckless banks who are willing to lend at rates close to the central bank interest rate will become more successful. Northern Rock could start of as a minnow and reach the position of making 25% of all new home loans.
Statsguy: “But the harsh reality of this is something that virtually all economists object to on philosophical grounds: the root of money is power, not free exchange. That means that in a world where the dominant political power enforces debt that is passed down from one generation to the next without recourse (as can the national debt), free-exchange can leave one very much not free… That’s what we call serfdom.”
Are you objecting to the way that national debt is passed down? It seem to me that this is the only sort that can be passed from generation to generation.
In my view that is an argument against nationalism and socialism, not an argument for it.
14. July 2009 at 03:35
Statsguy:
Yes, the electronic deposit of a paycheck is a payment by check. I am not sure what you meant about the mortgage, but my monthly payment of my mortgage is electronic, but also a payment by check. Yes, there are no checks. And perhaps I should be more clear–90% of payments by dollar value are by transfer of deposit balances either by paper check or electronic means. And 10% of payments by dollar value are by currency. Payments through the banking system are much more important quantitatively than payments by currency.
Stock is not money. Stock has a money price that varies.
If the price of Walmark stock rises, people cannot take it to the store and exchange it for more goods. First they sell it, and then buy stuff. But the person who bought the stock has less money to buy things.
The failure of economists to develop GE models of money shows the inadequacy of GE theory. But then, I have no problem with Patinkin’s solution of market vs. individual thought experiments. Though I think the Menger/Mises regression theorem is more realistic.
Anyway, thinking about currency like the electronic signals that are used to transfer deposit accounts is OK. Thinking about currency like Walmart stock is foolish.
The first step is to think of a system with no currency and only deposits. Sometimes it is called a “cashless” payments system. All money pays interest.
Well, gee, for some small hand-to-hand transactions, this is a pain. OK, add currency.
Don’t make the tail wag the dog.
14. July 2009 at 03:54
Stats Guy:
If you live in some small island nation (or a large continental one) that doesn’t use dollars, and you have Federal Reserve notes stuffed in your safe-deposit box, then this is an investment in Federal Reserve notes.
If you understand this as an investment in U.S. productivity, then you just have a confused understanding of the value of the dollar. The value of the dollar depends on supply and demand. Now, I believe, other things being equal, that more productivity in the U.S. raises the demand for dollars, and, given the supply, that will raise their value. But, this is mediated through the demand, and the demand could fall for reasons other than lower productivity.
And it could rise for reasons other than lower productivity.
Current:
I think that the legal restrictions on other currencies is irrelevant. Money is valuable because everyone else is using it. The notion that people are on the cusp of switching to something else is implausible.
Federal Reserve assets are relevant. Of course, if you want to see the Fed as just part of the U.S. government, then the treasury is borrowing by issuing FRN and its ability to sell other sorts of bonds (with variable prices and yields in terms of FRN) is what is important. Its “asset” of course, is its taxing ability.
I think part of the problem we have hear is that on foreign exchange markets, “currency” refers to assets denominated in a particular unit of account. For example, dollar denominated assets.
When I use the term “currency” I am thinking about a particular sort of finaical asset. In particular, dollar-denominated Federal Reserve notes and coins. (Quarters, dimes, nickels, and pennies.)
Most payments in the U.S. are made using “dollars,” but almost all of those are by transfer of dollar denominated deposits. These are almost all privately issued already.
A small part of the payments are with dollar denominated Federal Resreve notes and coins. These are issued by the Federal Reserve and the U.S. Mint. I believe that instead they should be issued by banks. Dollar denominated, but redeemable by the issuing bank just like deposits.
14. July 2009 at 04:28
Bill, Statsguy,
I agree with Bill about stocks. Stocks are merely liquid titles, they are not money.
Bill, Mike Sproul,
Bill: “Money is valuable because everyone else is using it. The notion that people are on the cusp of switching to something else is implausible.”
Yes, that’s why I wrote “If there were competition in currency then I may use other sources of currency should other currencies become widespread.”
However others are not widespread. They are unlikely to become widespread unless the Fed create a lot of inflation.
Bill: “Federal Reserve assets are relevant. Of course, if you want to see the Fed as just part of the U.S. government, then the treasury is borrowing by issuing FRN and its ability to sell other sorts of bonds (with variable prices and yields in terms of FRN) is what is important. Its “asset” of course, is its taxing ability.”
Yes, I’d agree with you there.
14. July 2009 at 04:34
Statsguy,
Bill is right about this business of “shares” in the US. This is a fairly uncontroversial point as Mises wrote:
“Neither has the wealth of a country any bearing on the valuation of its money. Nothing is more erroneous than the widespread habit of regarding the monetary standard as something in the nature of the shares of the State or the community. When the German mark was quoted at 10 centimes in Zurich, bankers said: ‘Now is the time to buy marks. The German community is indeed poorer nowadays than before the War, so that a low valuation of the mark is justified. Nevertheless, the wealth of Germany is certainly not reduced to a twelfth of what it was before the war; so the mark is bound to rise.’ And when the Polish mark had sunk to 5 centimes in Zurich, other bankers said: ‘This low level is inexplicable. Poland is a rich country; it has a flourishing agriculture, it has wood, coal and oil; so its rate of exchange ought to be incomparably higher.’ Such observers fail to recognize that the valuation of the monetary unit does not depend upon the wealth of the country, but upon the ratio between the quantity of money and the demand for it, so that even the richest country may have a bad currency and the poorest country a good one.”
14. July 2009 at 06:24
Statsguy, You said:
“Precisely what does this say about Larry Summers’ and Tim Geithner’s argument that the key to restarting the economy was to “recapitalize the banks” – in other words, absorb the bank losses (one way or another), so they have adequate capital to support lending?
This mantra – by far the dominant press narrative of the time, was seen as the way to avoid a “lost decade” comparable to Japan’s. The incessant comparison to Japan drowned out everyone else’s voice.”
This is a good point, and got me thinking. In the upper levels of macro, people like Bernanke saw the key policy failure in Japan as being monetary, the BOJ refused to take steps to credibly reflate the economy, and prices drifted steadily lower. Why can’t they see the same problem in their own country?
Mike, I disagree on all three points:
1. The Fed was required to by back dollars for gold. They are not required to buy back dollars for bonds.
2. I’m fine with competition in currency, as long as the Fed can copyright the name “dollars”. The other currencies must use a different name. I am confident that the Fed will blow away any competitors. Indeed competition is already legal if you don’t use the term “dollar”, unless I am mistaken. Suppose I have a paper warehouse receipt for 100 oz of gold. Can’t I go to the store and by a car with the receipt? So why don’t gold receipts dominate dollars?
3. In my example the value of bonds fell to 600 billion, but 1 trillion in cash stayed in circulation. So the cash was not backed. Your .40 oz example doesn’t apply. I am valuing both assets and liabilities in the same units (dollars) and one falls to 600b, while the other stays at 1 trillion in the same units. So they will also differ in any other units you come up with.
Statsguy, You said:
“Almost everything that Summers, Bernanke, and Geithner have done (except, possibly, GM’s bankruptcy) has had the effect of making sure that creditors get their money, and the value of that money stays intact.”
There’s a big flaw in this argument. All three called for massive fiscal stimulus. This stimulus is aimed at raising the inflation rate. It only makes sense if monetary policy has run out of ammo.
Bill: You said:
“If the monetary authority is instead committed to monetary stability, (nominal income targeting, for example,) then the base money it issues is a form of borrowing. They have to stand ready to pay the money back.”
I’m just thinking out load, but isn’t the baseline policy here one that reduces nominal rates to zero? Friedman’s optimal quantity of money? That would led to a deflation rate of X%. Then any policy that is more expansionary (including stable prices) will inflict an inflation tax on cash holders. If inflation is zero then the inflation rate is x%, if inflation is 2% then the inflation tax will be (2+x)% Cash will earn a lower rate than T-bonds, and this will transfer inflation tax revenue to the government. This is permanent tax revenue, not borrowing, as no interest must be paid. But at the same time I think we are arguing about semantics. I can see your point about a monetary target representing an implied promise to buy back whatever debt is required to hit that target. So if that is a “liability” then so be it. It’s similar to the argument over whether social security promises are part of the “government debt.”
You said:
“Scott, when you make statements like this, it makes me think that you don’t understand Krugman’s point about an irresponsible central bank.”
I thought Krugman was simply referring to a central bank committed to raise the inflation rate. How do you interpret his statement?
Bill, I went 10 years with no bank account (until a year ago.) I had no trouble getting cash from ATMs with my Fidelity card. I picked companies with lots of net worth, so they would have enough assets to back the currency. Yes, they have to pay back the currency someday, but it is still a nice interest free loan. I see banking as a very different business unrelated to currency. But if you define entities with a string of ATMs as “banks” then yes, banks would be involved in currency. But I think Walmart could have currency and be part of an ATM network, without any other banking aspects. Indeed when paper cash is phased out (and it will be during this century) we will only have debit cards as cash. I expect Walmart will issue this type of cash. But if you had instantaneous EFT, then cash in circulation would fall to zero, and the base would be entirely interbank reserves. So who knows?
Free banking types are fighting the wrong battle. We’ll get rid of currency entirely long before we allow free banking. They should be trying to reform the next system (I seem to recall that you and Yeager and Greenfield did work in this area.)
Lance, Some types of trouble can be predicted. One of the few good predictions I made was 10 years ago when I said Fannie and Freddie were time bombs waiting to go off. But I had no date in mind. If your prediction implies a huge and sudden change in stock prices in the near future, then it is inconsistent with the EMH. If the prediction does not imply such a change, then it is not necessarily inconsistent with the EMH.
Rob, A run on the dollar would boost AD. As long as it didn’t go too far it would be a good thing, as the dollar neede to lose some value. The problem is that it has been gaining purchasing power over the past 12 months.
David, I have the same reply as I gave Statsguy. Bernanke supported fiscal stimulus, which is intended to be inflationary. Both of your arguments are intended to explain why Bernanke is anti-stimulus. But he isn’t. He is opposed to the more efficient type of stimulus (money) and supports the less efficient (and hence even more inflationary) type of stimulus (fiscal.) That’s what’s so odd.
Current, I mostly agree. I do understand that fiat money is a strange entity, and there are many ways of looking at it. The question is which way is the most useful.
azmyth, Good question. I do not favor discretion. If we had NGDP futures targeting we would not need skilled central bankers. My point is that we do have discretion, and as long as we have it, we need really, really good central banks, as even somewhat good ones can cause trillion dollar mistakes. And did just last fall.
Statsguy,
I don’t worry about tail risk. The fed observes inflation expectations and the dollar’s forex value in real time. It has all the resources it needs to stop a run on the dollar. Right now we need a bit of a run, however.
Currency generally has a fixed nominal value, as that makes it more useful for transactions. That’s why stock is generally not used as currency. A share of Walmart stock might be worth $23 when you walk into the store, and $24.75 by the time you reach the cash register. That’s not very useful currency.
More to come . . .
14. July 2009 at 06:32
Current, Are you assuming that Northern Rock’s strategy was obviously foolish? If not, then they did the right thing. If so, the even as they gained 25% market share, their stock price would have gone down the toilet. That’s the problem with all explanations of this crisis, then either have no useful implications, or that are logically inconsistent.
Bill, Yes, I made a similar response.
14. July 2009 at 07:04
Scott:
1. “Required”?? The Fed is a big enough institution that it can stand on its own. If the Fed says it will pay out gold for dollars, people will value the dollar whether or not some external agency requires it to do so. If the Fed announces that it will suspend one kind of convertibility, while maintaining other kinds, the people will still value the dollar based on the Fed’s assets and liabilities.
2. Correct (probably). Right now the Fed has about as much to fear from private paper money as the post office has from private mail carriers. All the more reason that Bank of America should be allowed to issue paper dollars if it wants to. Bank-issued dollars might come to dominate fed-issued dollars. Who knows? Back in the depression bank-issued dollars would have made a difference. And no reason to copyright the name ‘dollar’. B of A should only have to print “B of A Dollar” on each of its dollars, and recognize them as its liability. Right now B of A issues checking account dollars on the same terms.
3. If a money-issuing bank has issued a total of $100, backed by 10 oz. of silver, plus bonds, denominated in dollars, worth $90, then the exchange value (E) of the dollar is found by setting assets (10 oz. plus $90 in bonds) equal to liabilities ($100), or 10+90E=100E, for a solution value of E=1 oz/$. If the bonds fall in value to $80, then the equation becomes 10+80E=100E, or E=.5 oz/$. It is incorrect to say that a fall in the value of the bonds leaves the money unbacked. It’s just that the money has LESS backing, and so is worth correspondingly less. Otherwise you’d end up saying crazy things, like if the value of the bonds fell by 1 cent, the whole trillion dollars loses backing and becomes worthless.
14. July 2009 at 07:48
Scott: “Are you assuming that Northern Rock’s strategy was obviously foolish? If not, then they did the right thing. If so, the even as they gained 25% market share, their stock price would have gone down the toilet. That’s the problem with all explanations of this crisis, then either have no useful implications, or that are logically inconsistent.”
I don’t think that what NR did was obviously foolish. My point is rather that when interest rates are at a rate that is lower than what can be sustained in the long run strange things happen.
From a recent paper by Antony Carilli and Gergory Dempster,
http://www.gmu.edu/rae/archives/VOL14_4_2001/4_carilli&dempster.pdf
“The decision on the level of excess reserves to hold is related to the decision of how many loans to make. The bank will lend excess reserves as long as the expected marginal benefit of lending exceeds the expected marginal cost. The most important marginal cost involved in the decision of whether to lend excess reserves is the cost of being faced with an inability to meet the obligations of deposit withdrawals, i.e., liquidity risk. Let us assume that all banks realize that the increase in credit demanded is not due to an increase in saving, so each bank knows the “best” strategy, in the aggregate, is not to lend as much as the prevailing (lower) rate allows. But each bank also knows that if it does not attempt to increase its loans at this rate, others will increase their lending and, thus, draw its own customers to themselves. Even if there were an agreement, explicit or tacit, amongst the banks not to lend, it would still benefit each individual bank to cheat. This is the prisoner’s dilemma is in the best interest of each bank to cheat (lend) so that it can increase its profit. If each banker “knows” that all the others will cheat, it will have to cheat. Thus, through the profit maximizing behavior of individual banks, the boom is started. Each bank will be enticed to cheat (make “unjustified” loans) partly because each does not have to pay the full cost of its bad decisions. If each bank issued its own notes, it would
have to bear the full cost of lending including the possibility of not being able to meet its demand deposit liabilities. In other words, if there were no monopolization of the currency each bank would have to bear all liquidity risk associated with of increasing its “money supply.” With a monopoly currency, however, part of the cost of lending can be passed on to the rest of the banks. This happens because the original loan becomes only a small part of the monopoly money supply.”
15. July 2009 at 07:32
Scott:
I wonder if part of the problem is that you are thinking about a very simple money supply rule. Must be something about time spent by lack Michigan. Is the ocean just too cold in Boston? Come down to Charleston. You will learn to enjoy the salt water.
The goverment issues base money and it increases the quantity at a constant rate. The price level adjusts so the real quantity of base money equals the demand. The government spends the newly created money and its real revenue is the rate of money increase times real balances. The quantity of base money _is_ the nominal anchor.
Next level of complication. Unlike Mints, we don’t favor 100% reserve banking, but allow fractional reserve banking. We have a money supply rule, but it is the growth rate of currency held by the public plus some kind of bank deposits (like checking accounts.) That aggregate (like M1) is the nominal anchor. Now, the monetary authority has to stand ready to vary the quantity of base money depending on the money multiplier. The nominal target it has now no longer allows it to just create new base money at a constant rate. It has to vary the quantity of base money according to how much currency and reserves are demanded.
It has to stand ready to reduce the quantity of base money issued. Now.. if we assume a constant money multiplier, then yes, the constant growth in base money generates the constant growth in the quantity of money. And maybe this remains true on average, but, increases in the demand for base money (a lower money multliplier) allow increases in base money that must later reduced. It is like a loan that needs to be paid back.
Now, let’s move away from money supply rules and move towards price rules. It is true, of course, that with a traditional gold standard, there is redeemability in gold.
Federal Reserve notes were promises to pay gold. Clearly, they were liabilities.
But suppose we just have the Fed target the price of gold. What is the difference? The quantity of base money must adjust to equal the demand for it with the price of gold remaining at the target level. If the demand for base money falls to zero, then the quantity of base money must fall to zero. Not likely, of course, but the notion that the demand will always rise at a constant rate isn’t all that likely either.
See, you are somehow going back to an institutional framework where base money grows at a constant rate, and the price level just adjusts. And the issuer gets a revenue based on its growth rate.
Just like under a more conventional gold standard, where they have to redeem for gold, the gold price peg makes currency into a kind of debt instrument. The issuer is “obligated” to pay it back, if they are committed to the rule.
This should be important to you because index futures targetting is like the gold standard. It is even stronger than targeting the price of gold (or targetting nominal GDP.) There is a redemption obligation.
You aren’t quite correct about the Greenfield/Yeager system. It was consistent with privately issue hand-to-hand currency. It really wasn’t about developing a system without currency. It was, however, about developing an alternative to money supply growth rules. (For Yeager. For me, it was about finding something better than gold to anchor a free banking system.)
By the way, when hand-to-hand currency disappears over the next 30 years, where does that leave the model of currency creating free revenue? As the demand for currency falls, then it will all be deposited at the Fed, creating excess reserves. If the Fed wants to stablize nominal income, the price level, the inflation rate, or even the growth rate of some measure of the money supply like M1, M2, or MZM, it will need to undertake open market sales.
The money it had “borrowed” in the form of curreny will be paid back. It will intead have to borrow by selling other sorts of bonds. If instead, we think of currency using the constant growth rate approach, then the value of money will fall to zero.
I think it is sensible to see the Fed as an independent agency. Federal Reserve notes are its liabilites. Its assets are goverment bonds (and lots of private assets now.)
It earns income on those assets that it transfers back to the treasury. The greater the demand for Federal Reseve liabilities, the more assets it holds and the more it earns and the more the Treasury gets. For a long time, this just meant that the Treasury didn’t have to borrow from someone else. But in the last year, the Fed is holding lots of private assets. The Treasury does have to borrow from other people. And if the Fed earns interest on those private assets, then it will transfer back any net earnings to the treasury. (Of course, it has that interest expense for deposits too.)
As for your story about the Fidelity card… what are you talking about? Banks already hold like 30% of their assets in the form of securities. I think your notion of “bank” is a bit narrow. Free banking includes insitutions like Fidelity. If instituions want to offer deposis (and/or hand to hand currency) and fund portfolios of securities.. fine.
With redeemable competitive currency, it is obviously a debt instrument. Walmart currency would be a debt instrument. It wouldn’t be Walmart fiat currency with their revenue being equal to its growth rate and its real value adjusting to the demand. And, what is Walmart doing with the borrowed funds? Funding their own operations? How much debt does Walmart issue?
Fidelity currency, on the other hand, would presumably fund some kind of bonds. Probably commercial paper.
Go ahead. Imagine. Walmart has ATM machines and you use your Fidelity card to withdraw funds from your account. What happens? Walmart has to get funds from Fidelity. What does Walmart do with the funds?
As for the Krugman responsiblity issue…
A responsible central bank holds to its nominal target–say 2% inflation. Even if it increases base money now, it must stand ready to reduce it against so that its purchasing power shrinks no more than 2% per year.
An irresponsible central bank creates base money and worries nothing about the purchasing power of money. It is like the first scenario above. The price level adjusts so that the real quantity meets the real demand. The price level, nominal income, and the like.. not a worry. They can increase the quantity and spend it on what they like. There is no obligation of any sort to ever reduce the nominal quantity of money.
I realize that you realize that base money needs to sometimes go down. But I sometimes wonder to what degree that this insitutional framework (the quantity rule for currency) isn’t behind the concern that the huge increases in base money are inherently inflationary.
Zimbabwe as an irresponsible central bank. They print up money and spend it. The purchasing power will be whatever it will be. And so, they have no problem creating inflation.
The Bank of Japan is responsible. To avoid too much inflation, they must stand ready to withdraw base money from circulation. And so, any increase in base money is aways tentative.
Now, let us suppose that market clearing requires a negative 5% real interest rate. The central bank is credibly commmitted to a 2% inflation rate, and so with zero nominal rates, the real rate can be no less than -2%. Raising base money will do no good because any increase will always be expected to be reversed so that the inflation rate is 2%. The real interest rate cannot be -5%.
Zimbabwe doesn’t have this problem. There is no expectation that base money will decrease ever, because there is no expectation that the goverment will try to avoid inflation.
An irresponsible central bank has no committment to any inflation rate at all. They just print money and spend it. The price level will rise whatever. It should be easy to raise base money enough to get the real interest rate down to -5% and clear markets.
Now, there are many responses to this. For example, you can have faith that the real interest rate necessary to clear markets can only be less than -2% if people expect deflation or low real ouput in the future because of bad monetary policy. It is the fact that the monetary base didn’t increase enough that causes the need for a real interest rate less than -2%. Maybe.
15. July 2009 at 07:57
Scott:
With a free banking system tied to gold, when people substitute debit cards for hand-to-hand currency, the quantity of hand-to-hand currency shrinks (all the way to zero if there is no demand) and the quantity of deposits rises. Nothing else happens.
With a free banking system with a frozen quantity of base money, and all that base money is reserves, when people substitute debit cards for hand-to-hand currency, the quantity of hand-to-hand currency shrinks (all the way to zero if there is no demand) and the quantity of deposits increases. Nothing else happens.
With a free banking system and indirectly redeemability with a bundle of goods and services with gold, if people substitute debit cards for hand-to-hand currency, the quantity of hand-to-hand currency shrinks (all the way to zero if there is no demand) and the quantity of deposits increases. Nothing else happens.
With a free banking system and index futures convertibility, if people substitute debit cards for hand-to-hand currency, the quantity of hand-to-hand currency shrinks (all the way to zero if there is no demand) and the quantity of deposits increases. Nothing else happens.
If those banks or other finanical institutions that issue currency lose market share to banks or other financial instituions that were deposit only, then the banks that have more deposits get excess reserves. The banks that had issued currency (as well as deposits, or maybe currency only banks.. sure,) have reserve deficiencies. The banks with excess reserves buy assets. The banks with reserve deficiencies sell them. Nothing else happens.
However, with the Fed using index futures redeemability, if people substitute debit cards for hand-to-hand currency, then currency is deposited into banks. This creates excess reserves. This increases the quantity of money and aggregate expenditures. Recognizing this process, speculators buy index futures from the Fed. The Fed sells bonds. It debits the banks’ reserve accounts. Reserve deficiencies develop, and the quantity money falls back to its initial level. And spending falls back to a level consistent with the target. Speculators, however, can preempt the process at any step. And, of course, you have advocated that the Fed should undertake open market operations before futures are traded to make these sorts of technical adjustments.
Anyway, if the demand for currency falls to zero, then quantity of currency will fall to zero. Base money will be lower too. But, presumably banks will need somewhat more reserves than before because of the added deposit transactions that are replacing what were currency transactions. Some combination of Fed experts adjusting base money and speuculation on the index futures exchange would deal with this problem.
The drop in base money, of course, means that the Fed sells government bonds (or whatever other assets it is holding at this time.) And so, the Treasury (or someone) will have to fund their activities some other way. Of course, deposits expand, and the banks may fund the government (or the private borrowers.)
Anyway, it is clear that the Federal Reserve notes that were used as currency were a type of loan that has now been repaid because no one is willing to lend by holding them any more. They prefer bank deposits because they can be transacted with debit cards.
17. July 2009 at 05:49
Mike, I’m not sure if you are seeing my point. I claim that if the market value of the Fed’s bonds suddenly falls from 1 trillion to 600 billion, then the value of currency stays at 1 trillion. Even if there is hyperinflation, 1 trillion dollars cash is always equal to 1 trillion dollars cash. Are you agreeing or disagreeing? I can’t tell. But I am sure that I am right, because that sort of thing does happen (obviously to a lesser extent.
Current, I don’t see a prisoner’s dilemma at all. It wasn’t in Northern Rock’s interest (ex post) to make those loans. The prisoner’s dilemma is not about uncertainty, your best option is the same regardless of the state of the world. But in this case the state of the world did affect NR’s results.
So what if the competitors lose a bit of business? In America the banks that didn’t make a lot of foolish loans are doing well today.
Bill, I don’t think we are far apart, it is more a question of definitions.
What I resist is people simply looking at the Fed’s balance sheet and dogmatically saying the currency is a liability. Well the balance sheet could look exactly the same but if the Fed is just randomly increasing the currency amount over time, and never ever withdrawing any, then it clearly is not a liability. Under the gold standard where there was a statutory requirement to redeem currency for gold, currency was a liability. But what about all the gray areas in between? I’d say the past 60 years look a bit like a central bank just randomly injecting base money, and almost never pulling any out of circulation. And there is no statutory requirement to peg or target anything. Now I do agree that if banks adopt inflation targeting then they are sort of making currency a liability, not a legal liability, but simply an informal promise to redeem currency for bonds should it be necessary to prevent inflation. The BOJ has done something like that. I suspect you and I agree on what is going on, but maybe have a slightly different interpretation of the term “liability.” I was thinking in a very legalistic sense, but it’s quite possible that wasn’t the best way of thinking about it.
The question of banks might also be one of semantics. If Walmart does all the things necessary to implement a system of currency, then perhaps it is a bank according to some definitions. But it need no make commercial loans, it could decide to only “lend” to the Treasury by buying T-bonds to back the currency.
I think I also agree with your second point. Sometimes when I respond to free banking-types I assume they don’t have in mind any regime (like index convertibility) that would stabilize the price level. So I may over-simplify slightly. I think we both agree that the free banking issue is separable from the issue of how to anchor the price level, and who does the anchoring.
18. July 2009 at 12:29
Scott:
A simple example of what I am saying would be a bank that holds 100 oz. of silver as backing for 100 currency units, which can be called dollars. Then each dollar is worth 1 ounce. If the bank loses 10 oz, then there are 100 dollars laying claim to 90 oz, so each dollar will then be worth .9 oz. The dollar clearly does not lose all of its value. If assets fall by 10%, the value of the money backed by those assets falls by 10%.
Now return to the case of 100 oz. backing $100. If the bank swaps 80 oz. for bonds, denominated in dollars, worth $80, then the bank becomes vulnerable to what I call ‘inflationary feedback’. For example, if the bank lost 10 oz, then the dollars will fall in value. But the bonds, being denominated in dollars, will lose value too. This causes a further fall in the dollar, which devalues the bonds more, etc. This process can be handled by the equation setting assets equal to liabilities: 20+80E=100E, where E is the exchange value of money, in this case E=1 oz./$. In this case, a loss of 10 oz. of silver changes the equation to 10+80E=100E, or E=.5 oz./$. The same loss of 10 oz that caused 10% inflation before, now causes 100% inflation because of the effect of inflationary feedback.
If we went to the extreme you suggested, where $100 is backed by $100 of bonds, then the equation becomes 100E=100E, which leaves E undetermined.
On the issue of liability: If the bank held 100 oz as backing for $100, then the $100 can be considered a liability either if the government REQUIRES the bank to buy back dollar for 1 oz, or if the bank PROMISES to buy back $1 for 1 oz. The promise of an unreliable bank will have a low value, and the promise of a reliable bank will have a high value, but those promises will have some value in any case. Now, if this bank issues another $100, without getting any new assets, then there are $200 laying claim to 100 oz, and each dollar will fall to .5 oz. Those dollars didn’t suddenly cease to be the bank’s liability; they just became less valuable liabilities. Likewise, when the Fed diligently uses its bonds to buy back its dollars whenever those dollars lose value, then people will value the dollar highly. If the fed carelessly neglects to buy back dollars, then people will place less value on those dollars. But the dollars are liabilities of the fed in either case. Now, if the Fed stopped buying back dollars altogether, then those dollars will lose all value. The dollars will still show up on the liability side of the fed’s balance sheet, but since the fed will have disavowed all responsibility for those dollars, they would become a liability with zero market value.
19. July 2009 at 18:04
I might agree with you on commodity backing but that wasn’t the claim I addressed. I don’t think fiat money needs to be backed. You still haven’t addressed what happens if the market value of the Fed’s bonds falls (in dollar terms) but the monetary base doesn’t fall in dollar terms.
19. July 2009 at 21:15
Scott:
The claim I am making is that all money is backed, and that there is no such thing as fiat money. The paper moneys that are commonly called fiat money are actually backed but physically inconvertible. (This is explained in several papers that can be accessed by clicking my name above.)
Fiat money, of course, is a strange animal, and I’m sure you’re aware of the many paradoxes that pop up once we assert that pieces of paper can have value even though we can’t present them to the Fed and demand gold for them. I find it much easier to believe that the dollar is not fiat money at all, but money that is backed but physically inconvertible.
In the old days of privately-issued paper dollars, convertible into gold, private banks would suspend physical convertibility every weekend. Those dollars did not suddenly become fiat money over the weekend, only to become backed and convertible on monday. Clearly the paper dollars were backed but inconvertible over the weekend, since the bank’s gold and bonds were still there in its vault. The difference between a suspension lasting 2 days and one lasting 76 years is only one of degree, and of course the Fed’s gold and bonds are still there in its vault.
Now, if the fed’s assets consist of 20 oz. of silver, plus $80 worth of bonds (bonds denominated in dollars), and if the monetary base is $100, then that equation of mine is 20+80E=100E, for E=1 oz./$. Then if the Fed’s bonds fall in value from $80 to $70, while the monetary base remains at $100, the equation becomes 20+70E=100E, or E=.67 oz./$.
(I just realized that my earlier example had been discussing what happened when the bank lost some of its SILVER, when you had been asking what happens when the bank’s BONDS lost value.)
20. July 2009 at 17:29
Mike, You said:
“Fiat money, of course, is a strange animal, and I’m sure you’re aware of the many paradoxes that pop up once we assert that pieces of paper can have value even though we can’t present them to the Fed and demand gold for them.”
I don’t find this strange at all. Nice leather wallets also have value, and they cannot be redeemed for gold either. I think cash and wallets serve the same role, they make shopping a bit more pleasant. So I am not surprised that either has value
20. July 2009 at 20:41
Scott:
First of all there’s this: “I do understand that fiat money is a strange entity, and there are many ways of looking at it.”–ssumner, 7/14/2009.
But leaving that aside, consider two paradoxes of fiat money:
1) The central bank of Mexico issues paper pesos, sells them for one ounce of silver each, and uses the silver to throw large wasteful parties, rather than backing the pesos. But the bank does limit the supply of pesos, and people demand them for liquidity, so (according to the quantity theory) they maintain their value of 1 ounce. The US central bank, seeing this, makes every effort to keep the dollar stable, so that Mexicans will use dollars in preference to pesos. Now the US authorities can get a piece of the Mexicans’ profits, while the demand for the peso drops. As long as the peso has positive value, the US can earn a profit this way, so there is no stable solution short of the peso falling to zero. We are left wondering why the US dollar has not driven the value of all the other so-called fiat moneys to zero.
2) As private firms issue checking accounts, credit cards, gift certificates, and other derivative moneys, the demand for the base money falls, and it loses value. Thus, on quantity theory principles, issuers of rival moneys are no better than counterfeiters. This is strange, because counterfeiters don’t put their names on the money they issue, don’t recognize that money as their liability, and don’t stand ready to use their assets to buy back that money. We are again left wondering how the base money maintains any value at all, especially when derivative moneys can be issued outside of the country, not subject to legal reserve requirements (e.g eurodollars).
Neither of these paradoxes is a problem for the backing theory of money. On that theory, the value of the peso is equal to the value of the assets backing it, and this is unaffected by competition from either US dollars or derivative Mexican moneys. Quantity theorists have bent over backwards trying to explain all the paradoxes raised by fiat money, but with the backing theory there’s nothing to explain. Money has value because of its backing, just like any other financial instrument.
22. July 2009 at 06:06
1. I don’t agree that Mexicans would want to use dollars just because the value was stable. Prices in Mexico are denominated in pesos, that makes peso currency more convenient than dollars.
2. The demand for cash still exists despite alternatives for several reasons. Cash is quicker than credit cards for small transactions, and more readily acceptable than gift cards. Cash is also more anonymous that checks or credit cards, and hence a good way of hiding wealth from the government.
When people decide to take a ten dollar bill instead of a credit card to the movie theatre, I don’t think their thought process is “I’ll take the ten dollar bill because it is backed by Fed assets.” Rather they take it because it is convenient.
22. July 2009 at 13:28
[…] the least known of the “big four” Princeton professors I mentioned in an earlier post here, he is my favorite. It’s good to hear that he is so influential: “He’s had a […]
22. July 2009 at 15:37
1. It’s common in Mexico for big items like houses to be paid for in dollars. But anyway, the main point is that currencies can cross borders, and they are in competition with each other, especially in international markets. Given this competition, any currency that provides a profit for its issuer will attract rival moneys, and that competition should drive the value of any currency down to the point where the value of the currency is equal to the value of the assets backing it.
2. I agree about the demand for cash. It is the supply of cash that I’m going on about. If a unit of paper money is backed by 1 oz of silver, but sells in the market for 1.01 oz., then the issuer will emit endless quantities of the currency. If the same currency unit sells for .99 oz., then none will be issued. Thus the supply of currency is a horizontal line at 1 oz, and that currency will be worth 1 oz regardless of the demand for it.
23. July 2009 at 05:54
Mike, I am not sure what more I can say. Developing countries have often maintained high inflation for many decades, without their currencies being replace by another. Cash denominated in your own country’s unit of account is very convenient, and governments can therefore get away with issuing more of it than they have backed, and they do.
23. July 2009 at 06:17
“Developing countries have often maintained high inflation for many decades, without their currencies being replace by another.”
Only the quantity theory implies that inflating currencies would be replaced by a rival money. The backing theory says that the inflation just reflects loss of backing. But since the value of each currency unit is equal to its backing, there is no profit to issuers of rival moneys, and the existing money will not be replaced by another.
“Cash denominated in your own country’s unit of account is very convenient, and governments can therefore get away with issuing more of it than they have backed, and they do.”
Naturally people will continue to use it. The point of the backing theory is that when the central bank’s assets total up to 100 oz. of silver, and the bank increases its issue of money from $100 to $200, without getting new assets, each dollar will fall from 1 oz. to .5 oz. Otherwise there would be arbitrage opportunities. The quantity theory, in contrast, says that the value of those dollars is determined by how many dollars are chasing how many goods.
24. July 2009 at 17:10
Mike, I understand what the backing theory says, I just don’t think it has much predictive power. Nor do I think it matches common sense. If you increase the monetary base 10-fold, prices will usually rise, even if the money is fully backed.
25. July 2009 at 06:02
Scott:
Common sense says that pieces of paper that have no backing will have no value. Believers in fiat money assert that money, alone among all financial instruments, can have value in spite of having no backing.
I don’t know of any historical cases where the money supply was suddenly increased tenfold, while backing also increased tenfold. Most of the cases I know of had the money supply increasing tenfold, while backing either fell or was unchanged. In these cases, quantity theorists said inflation resulted from the fact that there was ten times as much money chasing the same level of real output, while the backing theory said inflation resulted from the fact that there was ten times as much money, backed by the same assets as before. The old observational equivalence problem.
The historical cases that I know of are cited in the first paragraph of my paper entitled “There’s No Such Thing as Fiat Money”. Studies by Bruce Smith, Thomas Sargent, Charles Calomiris, Thomas Cunningham, Bomberger and Makinen, Pierre Siklos, etc. all found that the backing theory fit the data better than the quantity theory.
Sargent, in particular, in “The Ends of Four Big Inflations”, found that inflations ended before money growth was restrained. He attributed this to the fact that fiscal reforms had allowed the inflating countries to back their currencies better. (In a typical quantity theory misunderstanding, Mankiw’s Principles book mistakenly cites Sargent’s paper as providing support for the quantity theory, when Sargent had actually said that the data refuted the quantity theory.)
25. July 2009 at 06:39
Mike, The US money supply has increased more than 10 fold, and so has the asset side of the Fed’s balance sheet.
I don’t consider cash a “financial instrument” any more than I consider a wallet a “financial instrument.” It is a useful object for making exchanges. Financial instruments are different, because they have no utility beyond their future cash flow. The closest analogy to fiat money is a celebrity’s check than never gets cashed because it is collected. It also doesn’t need backing to have value.
It is the usefulness of cash that gives it value. Obviously if it was a financial instrument, not one would want it, because other financial instruments are just as safe and do pay interest. People want cash because it has other advantages.
I published a paper criticizing Bruce Smith and Sargent’s view of the historical evidence. In my view their theory doesn’t explain things better than a forward-looking QT.
The backing theory is applicable to some of the hyperinflation cases. But those are not relevant to the US. We don’t print money to pay our bills, if we did, we would have massive inflation. In the US the Fed determines the money supply, and Congress must set a path of taxes and spending that is consistent with the money supply path.
25. July 2009 at 10:46
Scott:
The backing theory has two explanations for that tenfold thing: (1) Inflationary feedback can cause a large inflation even from a small drop in assets relative to money issued (2) A bank can maintain the value of its currency at a level below what its assets can support. For example, a bank with 100 oz. worth of assets backing $100 can, if it chooses, maintain convertibility at .5 oz./$. This amounts to robbing the customers, but it can be done. Hence it is possible that the Fed’s ratio of assets to money has not changed in 100 years, but the fed has chosen to reduce the rate of (financial) convertibility to a fraction of the level that it could afford.
Various kinds of money can, and do, bear interest, just like any financial instrument. Even paper money has sometimes borne interest. People accept a low (or zero, or negative) yield on cash because it provides liquidity services to compensate. For paper money, the cost of issue more or less eats up the interest, so the net yield on paper is close to zero (or negative).
You have not addressed my point about arbitrage profits. If money has value greater than its backing, then issuers of rival moneys will go after that free lunch until the value of the money is equal to its backing. If, as you say, backing is unimportant, why do central banks bother holding assets?
If I remember right, your paper on Smith just said that his evidence supporting the backing theory could also be explained by the quantity theory. Observational equivalence again. Thomas Cunningham’s paper had a fairly clever way of getting around this, and he concluded that the backing theory (i.e., real bills doctrine) fit the data better.
Many of the early American colonies printed money to pay their bills, without causing significant inflation. As long as a money-issuer has positive net worth, printing and spending just burns up net worth. Once net worth is gone, then printing and wasteful spending causes inflation.
26. July 2009 at 05:39
Mike, I agree that either the backing theory or a forward-looking QT of Money can explain the colonial currency episodes. But it was Smith who claimed that they discredited the QT. I showed that they don’t.
I don’t understand your rival currency argument. There are huge network externalities from using dollars. If a company started produce “zollars” why in the world would anyone be interested in this money? Who would want money that no one else would accept in transactions? You pointed out that Mexicans occasionally use dollars, but think about how often they use pesos, despite the fact that the dollar has clearly been a superior currency for a long time. I do agree with the backing school in one sense. The hoarding demand for money might switch over to anonymous interest-bearing bearer bonds in small denominations if there were no legal restrictions. But I think the transactions demand for cash is totally different, and can’t be analyzed in a finance model–it is more like the demand for wallets.
Suppose I went to BestBuy with a pocket of silver bars, each stamped 1 oz. 2 oz. 5 oz, etc. Would they take them in exchange for a TV?. No. Not even if there was zero risk of them being counterfeit. And yet it would be legal for me to do so, and also note that the rate of return on silver tends to dominate paper money. So why don’t we all use silver bars?
I don’t agree with your argument that money creation is not profitable. Many countries have relied on it to pay their bills. I thought that during normal times the US government earned about $30 billion a year in seignorage. Is this wrong?
26. July 2009 at 15:34
Scott:
Data from the colonial era was too sketchy to provide much support for anything, but that still leaves the studies by Cunningham, Sargent, Siklos, Calomiris, and Bomberger & Makinen, all of which supported the backing theory. Unfortunately, economists seem to have tunnel vision on this issue, as evidenced by the fact that the words ‘backing theory’ and ‘real bills doctrine’ don’t appear in 99% of economics textbooks.
There are network externalities in everything from typewriters to threaded pipe, but those products don’t produce unusual profits. There is no reason the zero profit theorem wouldn’t work just as well for money as for threaded pipe. And of course the Fed’s $30 billion in accounting profit was earned on over 2 trillion in assets. That’s well within the range that we could call ‘zero economic profit’.
Best Buy won’t accept silver bars, but they accept checking account dollars, credit card dollars, and gift certificate dollars, all of which act as rival moneys to paper dollars. Furthermore, there have been many times and places where many moneys circulated side-by-side. They clearly competed against each other, and any seignorage earned by one kind of money would have created arbitrage opportunies for issuers of other moneys.
Suppose you’re a landowner, and your land is worth 1000 oz. of silver. You could buy a loaf of bread from the local baker just by writing “IOU 1 oz. of silver” on a paper. If you accept your IOU’s in payment of rent, then your IOU’s could circulate as money and be valued at 1 oz, even though they were never actually redeemed for 1 oz. by you. You could spend up to 1000 of these IOU’s before you had burned up your net worth. Someone might accuse you of using money creation to ‘pay your bills’, but in fact you are just spending your net worth. Of course, once your net worth is gone, you are in danger of causing inflation with further money issue.
Now suppose you buy another 2000 oz. worth of land, giving the seller your 2000 oz. IOU (call this a bond) and using the land as collateral to back the bond. After a few months, you print another 2000 small-denomination IOU’s and use them to buy back your 2000 oz. bond. There are now 3000 of your IOU’s circulating as money, backed by 3000 oz worth of your land. No inflation. Each IOU is still worth 1 oz. A few local economists might accuse you of having issued fiat money, but that is clearly no the case.
Finally, suppose that your land falls in value to 2700 oz. The 3000 IOU’s are now backed by 2700 oz. of land, so they must fall to .9 oz each. If you had ever maintained convertibility at 1 oz./ IOU, you would now be forced to devalue or suspend convertibility, and the IOU’s must fall to .9 oz. You continue to accept the IOU’s as rent payments, but at .9 oz. You also occasionally use your bond holdings to buy back some of your IOU’s, just as the Fed sometimes uses its bonds to buy back its dollars.
Change the words ‘landowner’ to ‘government’, and ‘rent’ to ‘taxes’, and you have a pretty accurate description of the issue of US dollars. The landowner’s IOU’s are clearly not fiat money, and the US dollar has no more claim to being called fiat money than the landowner’s IOU’s do.
27. July 2009 at 04:04
Mike, I don’t think anyone has refuted my criticism of their interpretation of colonial money. If anyone has, I’d be very interested in reading it. They basically attacked a straw man, arguing that the QT assumed that if the money supply was doubled in one period, and then returned to the original level the next, that the QT predictedforward massive inflation followed by deflation, even if all this was expected. That is not the prediction of any forward-looking QT. So I really don’t think their analysis of colonial money showed anything of interest. The backing theory either matches a forward-looking QT, and adds nothing, or it is wrong.
I’d say the $30 billion a year is pretty impressive when you consider that the investors didn’t have to put up 2 trillion of their own money, but rather bought 2 trillion in bonds by simply printing money. Not a bad deal.
I agree that there are near-moneys that somewhat affect the demand for cash, but they are not perfect substitutes.
Suppose we are the only country, and we are on the gold standard. Also suppose an earthquake swallows up our entire gold hoard, so the gold-backed paper notes are no longer backed. They become fiat money. I say the price level would not change unless the future expected path of the monetary base changes. What do you say?
28. July 2009 at 06:40
Scott:
There are several areas where the backing theory has different implications from the quantity theory, so if it is correct it certainly adds something. The backing theory implies, for instance, that a loss of assets by the central bank will be inflationary, while the QT says it would not be.
“investors didn’t have to put up 2 trillion of their own money, but rather bought 2 trillion in bonds by simply printing money.”
That’s all any bank does: buy interest-yielding assets and issue money. The biggest difference is that the money issued by the Fed exists as printed pieces of paper, while the money issued by Wells Fargo exists as checking account dollars. $30 billion earned on average assets of 2 trillion gives the Fed a Return On Average Assets (ROAA) of 1.5%. A little googling shows that Bank of America and Wells Fargo had ROAA’s of 1.45%. (Although I’d expect the Fed’s monopoly of note issue to give it a higher ROAA than private banks that have to compete.)
Of course that’s almost beside the point for a believer in fiat money, since they believe that when the Fed issues $1, that whole dollar is pure profit to the fed (less printing and handling).
We agree on money demand, but then there’s money supply: The arbitrage processes I’ve mentioned make the money supply (for the most part) horizontal, so money demand doesn’t affect money’s value.
I also believe that losing the gold hoard would either not affect the dollar’s value, or else decrease it slightly. My reason is that I think the Federal government would bail out the fed. Only when bailout resources are used up would we see inflation.
A better example would be the French playing card money issued in Quebec in 1685. It was backed by a shipment of silver coins (livres, I think) on its way from France, which were to be used to redeem the paper livres the colony had issued. If that boat had sunk, then I’d say the existing paper livres would have lost value. I’ve been told of a similar case in New South Wales, where there was also a boatload of coins on its way to redeem some local paper. In that case, news reports concerning the boat’s progress, and possible sinking, had definite effects on the value of the local paper money, even though the monetary base was unchanged. (Sorry for my dim memory. I heard it at an econ. history conference at UC Riverside, maybe 10 years ago.)
28. July 2009 at 06:54
Mike, I still think they is a big difference between printing money and borrowing money (which is what commercial banks do.)
You said:
“I also believe that losing the gold hoard would either not affect the dollar’s value, or else decrease it slightly. My reason is that I think the Federal government would bail out the fed. Only when bailout resources are used up would we see inflation.”
That doesn’t answer my question. The clear implication was the the government would not bailout the Fed. What then? I say the price level would still be unaffected unless there was a change in the expected future path of the money supply.
28. July 2009 at 07:04
Mike, how do you explain Somalian coins and notes?
Some of these continued to be used and circulated a year after the Somalian state collapsed.
29. July 2009 at 06:46
Current, That’s good example.
29. July 2009 at 17:23
Scott:
The whole point of the money multiplier story is that private banks issue money through their borrowing and lending. Example: The central bank gets 10 oz. of silver on deposit, and issues 10 paper dollars in exchange. Then the central bank prints 90 more paper dollars and uses them to buy a $90 bond. A private bank gets 10 of those paper dollars on deposit and issues 10 checking account dollars in exchange. Then the private bank issues (lends) 90 checking account dollars to a home buyer, getting his $90 IOU (mortgage bond) in exchange. Both banks issue money. But for a change in the law, we could have the central bank issuing only checking account dollars and the private bank issuing only paper dollars, and nothing important would change.
I did answer your question. I said “Only when bailout resources are used up would we see inflation.” When a money-issuing bank loses its assets (gold, bonds, or whatever) and those assets are not replaced, or even expected to be replaced, by other assets offered as a bailout, then that bank’s money will lose value in proportion to the loss of assets. If all the assets are lost, then all the money’s value will be lost. If 10% of assets are lost (and the quantity of money is unchanged) then the money will lose 10% of its value.
If an old-fashioned paper money issuing bank lost its assets, then nobody would deny that its money (paper and otherwise) would lose value. I expect that this happened hundreds of times in the US during the 1800’s alone. If it didn’t happen, then there would be arbitrage opportunities. As soon as bankers found that they could issue $1 notes, sell them for 1 oz of silver, blow all the silver on fast living, and yet still have their notes trade in the market for 1 oz., then there would be an endless supply of eager bankers issuing limitless quantities of their own paper dollars.
Believers in fiat money assert that somehow, if the institution issuing those paper dollars is called a government, then that financial impossibility of unbacked money suddenly becomes not only possible, but an everyday event. They assert this even when there are rival governments, which are also perfectly capable of issuing competing moneys.
Current:
I explain the Somalian money by saying that people must have expected that it eventually would be redeemed. That expectation might have been finally proven wrong, but while it lasted the money would have value.
A more widely discussed case, covered by Greg Mankiw in his macro principles text, concerned the Iraqi Swiss Dinar, which had value even though the issuing government had fallen. Mankiw cited this case as an example of a currency that had value even though it was completely unbacked. But Mankiw then explains that as people started to think that the new government would honor the old dinars, they began to gain value, until they finally were redeemed in the new currency. Mankiw’s supposed example of unbacked money was, by his own words, an example of backed money.
30. July 2009 at 00:35
Mike Sproul: “I explain the Somalian money by saying that people must have expected that it eventually would be redeemed.”
Such currency can never be redeemed anyway, it is not backed by gold. It can be used to buy bonds, but that it not redemption. You are the only person who defines buying bonds with such money as “redemption”.
You may explain things this way, put where is the evidence?
Your explanation for all issues concerning the value of money is that people only hold it if it can be “redeemed”. I see no evidence for that.
Why seeks a complex explanation when a simple one is on offer. Namely that the notes remained scarce and therefore remained acceptable currency.
30. July 2009 at 04:55
Mike, The problem here is that it is hard to distinguish between views. If I can always say “a change in the quantity of money didn’t cause inflation because it was offset by an unobservable change in the future path of money” and you can always say “a change in backing didn’t effect things because there was an unobservable change in expected future backing” then how can this ever be resolved?
I agree that “near monies” like demand deposits are backed, but I don’t consider them money.
Current, I agree. Mike wants to view cash as a financial asset, but I just don’t think people look at it that way. The amount of real balances that people carry in their wallets when interest rates are 0% isn’t much different from the amount they carry at 5%. If they viewed the cash in their wallets as some sort of investment, wouldn’t demand fall off sharply when nominal rates rose to 5%?
30. July 2009 at 09:38
Current:
You find monetary theory simple? Scott finds it ‘endlessly perplexing’. So did I, back when I believed in the ‘something from nothing’ story of fiat money. But I am not at all perplexed by the idea that money has value equal to its backing. Everyone already recognizes the truth of that for financial securities, and applying that same idea to money is simplicity itself. Economists have bent over backwards trying to explain how pieces of paper can have value when they have no backing, (Remember ‘overlapping generations’?) but once you recognize that those pieces of paper actually are backed (by gold, bonds, land, taxes receivable, etc.), there’s hardly anything to explain.
Where’s my evidence? For starters, it’s in the work of Smith, Cunningham, Sargent, et. al., that I already mentioned. But where’s your evidence? Quantity theorists say they have centuries of evidence that prices rise when the quantity of money rises relative to real output. I say they have evidence that prices rise when the quantity of money rises relative to the issuer’s assets. Observational equivalence yet again. Sometimes we are forced to resort to logic.
If a bank issues 100 paper dollars, in exchange for 10 oz. of silver plus bonds worth 90 oz., then 90 of those dollars can be redeemed for the 90 oz. worth of bonds. Only the last $10 need be redeemed for silver. After all, 90 of those paper dollars were issued in exchange for bonds. It’s only natural that they can be redeemed with those same bonds.
Scott:
Yes. It’s hard to find evidence that distinguishes the two views. Given this, one would think that econ. textbooks might spend a sentence or two mentioning that the backing theory gives a plausible explanation for the data. Show me one that does, and I’ll show you a hundred that don’t.
If demand deposits are backed but paper dollars are not, then as demand deposits become popular, the demand for paper dollars falls, and we have inflation. What a tangled web that is! But the backing theory says both the demand deposits and the paper dollars are backed. Since the spread of demand deposits does not affect the amount of backing per unit of paper money, it would cause no inflation. I don’t remember any widespread inflation starting in the 1840’s, when demand deposits started becoming popular.
The average American carries $75 in paper in his wallet. So at an interest rate of 5%, he sacrifices $3.75 per year, in exchange for the convenience of having paper money. No surprise that demand for cash doesn’t fall off when interest rates rise. Besides, the ‘demand for cash’ only affects money’s value on quantity theory principles.
30. July 2009 at 10:06
Scott: “Mike wants to view cash as a financial asset, but I just don’t think people look at it that way”
Yes. Money is a present good. Other financial assets are future goods.
Money is held as a hedge against uncertainty. Other financial assets are held because of their yield.
30. July 2009 at 10:11
I’ll write a little more about this soon, I’m busy right now.
31. July 2009 at 13:26
Mike, Have you read any anti-fiscal theory stuff by distinguished economists? You sometimes seem to suggest that other economists just ignore that group, but I recall that people like Bennett McCallum wrote some powerful critiques. It’s been a few years so I can’t cite a title, but it shouldn’t be hard to find on the internet.
The $3.75 number is exactly my point. The cost of holding cash is very low, so why should anyone care if it is backed? I don’t care, as long as stores except the stuff. And stores will accept the stuff as long as others will.
Current, I agree
3. August 2009 at 07:42
Scott:
McCallum only addresses Smith’s paper, and the only thing approaching a conclusion was that the colonial data was too sparse to distinguish between the two views. I cited McCallum and several related papers in a UCLA working paper titled “The Quantity Theory vs. the Real Bills Doctrine in Colonial America”.
http://www.econ.ucla.edu/workingpapers/wp775B.pdf
“The cost of holding cash is very low, so why should anyone care if it is backed?”
That is a non-sequitur. The cost of holding a share of GE stock is also very low, but I definitely care whether it is backed. You could also make the argument that people only hold GE stock because they know other people accept it, and that may well be true for some people. But nobody takes a circular argument like that seriously as an explanation of why GE stock has value. A simple arbitrage explanation will convince anyone that the value of GE shares must be equal to the value of the firm itself. A similar arbitrage argument says that the value of money must be equal to the value of the assets backing it.
3. August 2009 at 13:33
Mike, I thought McCallum also attacked the theory in places, but it has been a while and maybe my memory is faulty.
I see your GE example as a non-sequitor. Of course GE stock is only valued if backed, it has not use as a medium of exchange. but cash is very useful as a medium of exchange, so it doesn’t need to be backed. You keep asking why cash is the only asset that supposedly doesn’t need to be backed (according to me) and I keep answering that it is the only widely accepted medium of exchange. People are willing to forgo backing, and also $3.75 a year in interest, for that convenience. Why is that so odd?
5. August 2009 at 06:46
Scott:
The thing I keep saying is that the existence of rival moneys makes it impossible for any money to trade for a value above its backing, regardless of whether that money is useful as a medium of exchange. Everyone recognizes that if any ordinary financial security traded for a price that exceeded its backing, then arbitrage would drive its price back down. For example, if a firm is worth $60, and yet the stock in that firm trades at $70, then traders would (1) start new companies with a $60 investment, sell the stock for $70, and profit $10, or (2) short the stock over and over, until the number of hypothecated shares grew so large that there were no buyers left for the original stock. The stock price would fall, at which time the trader covers his short position and profits $10. (There are a few other ways as well.)
It works the same for money. If a dollar is backed by 1 oz. worth of assets, and yet sells for 1.1 oz., then traders (including other countries) could either (1) issue their own dollars, sell them for 1.1 oz., keep the 1 oz as backing and profit .1 oz., or (2) Short the dollar, thereby creating hypothecated dollars (aka checking account dollars, credit card dollars, gift certificates, etc), which would reduce the value of the original dollars, ultimately to 1 oz. They then cover their short position and profit .1 oz. This arbitrage process would happen whenever the value of money exceeded its backing. In the extreme case of fiat money, where there is no backing at all, the value of the money must be driven to zero.
5. August 2009 at 11:06
Mike, You keep telling me that money is like an ordinary security, and I keep insisting it isn’t. I don’t see you addressing my argument as to why it isn’t. You just keep insisting it is. Obviously if you are right than the backing theory is right. But I don’t think it is an ordinary security. I think it is a real good that greases the wheels of transactions, like oil is to a car. What you are saying is like saying oil has no value to a car, because it doesn’t make the car go forward. That’s right, but it lubricates the parts that do make the car go forward. So it has real value to the car. So does cash, even unbacked cash.
5. August 2009 at 11:59
Mike: money is backed by hookers.
What gives gold its inherent value other than it is pretty and hard to mine? Women dig it.
5. August 2009 at 17:38
Scott:
We all agree that people have a demand for cash, and that cash greases the gears of trade. But we also know that when the supply curve is horizontal, demand is irrelevant to price. Furthermore, the more substitutes available, the more the demand curve becomes horizontal. In the limiting case, the supply and demand for dollars would both be horizontal at (say) 1 oz./$.
Backing determines the heights of these horizontal demand and supply curves. If the value of a dollar ever rose above 1 oz, rival banks would issue infinite amounts of dollars, while the quantity demanded would fall to zero. Vice versa if the dollars ever fell below their backing value of 1 0z/$.
Rob:
No; money is accepted by hookers. They only accept it because it is backed.
Gold is an actual commodity, with a positively sloped supply curve and a negatively sloped demand curve. Money can be pieces of paper or computer blips, which can be instantly created or destroyed. Also, as Schumpeter said, you can’t ride a claim to a horse, but you can trade with a claim to money. (I would add that you can also trade with a claim to a horse.)
5. August 2009 at 18:37
Mike: But what makes gold a commodity? Its industrial and jewelry demand doesnt explain its overall demand. Its demand is just like that for a fiat currency. It is universaly recognized: that is all. I do believe that the ev psych reason gold has power is due to its bling to women, but modern day fiat has the same bling. A note from the US Government has bling appeal in the same way gold does. and it greases wheels, as has been said, in a similiar yet more efficient way. for me, answer the question why gold has inherent value and fiat money does not, from your perspective.
6. August 2009 at 07:16
Mike, You said;
“But we also know that when the supply curve is horizontal, demand is irrelevant to price.”
I don’t know this. Supply curves exist in perfectly competitive industries. Because of network effects the dollar is a natural monopoly. So I don’t believe there is a supply curve. Yes, I could produce a competing “tollar” at constant marginal cost, and the demand for this competing currency would be zero, because of network effects. Only if the dollar was mismanaged badly (high inflation) do competing currencies make significant inroads. That is not perfect competition.
6. August 2009 at 09:18
rob:
If gold were not used as money, then 1 oz. might be worth 100 loaves of bread. If people start to use gold as money, the extra demand will push up its value to 140 loaves. That 40 loaf premium is what we’d call the purely monetary portion of gold’s value. But as people begin to use gold-pegged paper money, monetary demand for gold falls, and its value ultimately reaches 100 loaves once it is no longer used as money. Once gold has reached 100 loaves, new issues of paper money cannot drive gold’s value down any further, and as long as the new paper money is backed by new assets of adequate value, the paper money will still buy just as much gold as before.
That paper money, once in use, does not have to be instantly physically convertible into gold in order to have value, but inconvertibility will give it a false appearance of being unbacked fiat money, even though the backing is still being held by the issuing institution. That is why, for example, quantity theorists have trouble explaining why all the world’s central banks hold gold, bonds, etc. against the money they have issued, since on quantity theory principles the gold and bonds are unnecessary for the money to have value.
If we try to imagine that paper money could have any purely monetary value in the way that gold does, we are immediately faced with the fact that it is virtually costless to issue infinite amounts of rival moneys, and this would prevent the paper money from getting any monetary premium.
7. August 2009 at 10:41
Scott:
You don’t have to produce a tollar. Credit card companies create dollars all the time, as do issuers of eurodollars, private banks, etc. Every store that issues gift cards or scrip creates dollars. The natural monopoly model is fine for water and electricity, but you can’t drink a claim to water, while you can trade with a claim to money. Any increase in demand for dollars can be met by an issue of derivative dollars, with no effect on the value of the dollar.
Lots of small, weak countries issue so-called fiat money, but a moment’s reflection reveals that all those so-called fiat moneys are in direct competition with moneys issued by nearby countries. The only way they could have value is to be backed.
8. August 2009 at 05:54
Mike, That won’t work, because credits cards, etc, aren’t perfect substitutes for cash. My credit card gives me a 2% discount on everything I buy. Obviously if it was a perfect substitute for cash, I’d never use cash. Hence you don’t have perfect competition, and hence there is no supply curve. Cash has value because it is more liquid than all the near monies, not because it is backed.
8. August 2009 at 09:03
Scott, Mike,
I’ve thought a little about this discussion and I think a different approach is possibly needed. Scott, something you probably don’t know is that I’ve had this discussion with Mike several times before on the Mises economics blog. Neither of us have changed our opinion.
Let’s go back to a situation of Free-Banking. Suppose that we are in Scotland in ~1750. Banks are businesses. The state issue gold and silver coins that people can use. Banks use fractional-reserves, quite small reserves actually, perhaps even less than 5%. There is no protection in the form of insurance from bankrupt banks, none for the saver or the bank itself. Banks are free to issue banknotes and create accounts. If we talk about this we don’t get into the further question of disputes over fiat money.
Now, here is my view of that situation. I want money because I want a fund that I can dispose of as I wish. I want it because I can’t make the decisions on what to buy in the immediate future right now in the present. I can estimate the amount of money I need on the basis of experience. That part of my wealth I don’t need as money I can keep as other assets. I will keep it as other assets since they will give me interest or other services.
In Mises terminology I want a present good that can be exchanged for almost any other present and future goods. That is what money provides.
Mike Sproul, would you agree with me so far?
8. August 2009 at 12:21
Current:
Agreed so far. I’d then add that if the interest rate is 5%, then a costlessly issued IOU promising 100 silver shillings in 1 year would be worth about 95 silver shillings today. If the costs of issuing that IOU amounted to 3 shillings per year, then the same IOU would sell for 98 shillings today and would therefore yield 2%, rather than 5%. If the cost of issue were 5 shillings/year, then the IOU would sell for 100 silver shillings all year long, yielding 0% to the holder. If that IOU has some redeeming feature, like usefulness as money, people will hold it in spite of earning no interest. But the existence of that IOU will give people the false impression that money somehow follows different rules than ordinary IOU’s, when in fact it doesn’t.
On a related note, checking accounts bear interest, but for most purposes are just as much ‘currency’ as paper money or coins.
9. August 2009 at 08:15
Mike, You said,
“On a related note, checking accounts bear interest, but for most purposes are just as much ‘currency’ as paper money or coins.”
I don’t understand. If checks are just as good as cash, and checking balances earn interest, why does the average American hold more than $1000 in cash at any given time. BTW, I did my dissertation on currency hoarding, so you don’t want to get into a debate with me about whether the average American holds over $1000 in cash.
9. August 2009 at 18:48
Scott:
“That won’t work, because credits cards, etc, aren’t perfect substitutes for cash. My credit card gives me a 2% discount on everything I buy. Obviously if it was a perfect substitute for cash, I’d never use cash.”
That just means people use different kinds of money for different purposes, but that is a very slim reed to hang a theory of money on. Some new innovation in credit cards could come along tomorrow and cut our demand for paper money in half, or all the way to zero. On backing theory principles, the only result would be that the law of the reflux would come into play, and the fed would use its assets to buy back the paper dollars at par. If the idea of fiat money were correct, then in principle, the central bank could very well have no assets, and nothing with which to buy back the paper dollars. In that case the paper dollars would lose some or all of their value. In that scenario, the value of paper money is too fragile to be believable.
“Hence you don’t have perfect competition, and hence there is no supply curve.”
Actually, the whole concept of a supply curve becomes meaningless with money. Money can be produced without inputs, without a production function, and indefinitely large amounts can be created or retired as easily as computer blips.
“I don’t understand. If checks are just as good as cash, and checking balances earn interest, why does the average American hold more than $1000 in cash at any given time.”
Maybe I don’t understand why you don’t understand, but it sounds like you are drawing some meaningless distinctions between paper dollars, checking account dollars, credit card dollars, etc. Consider the old Bank of Amsterdam, which operated on 100% reserves. Each depositor, for all practical purposes, had a stack of coins (guilders) in the vault. What did it matter if a shopper paid 5 guilder coins to a grocer, or wrote a check directing that 5 guilders be removed from the shopper’s stack and placed in the grocer’s stack? Or what if the bank had issued paper guilders for each coin deposited? In that case the shopper would hand 5 paper guilders to the grocer, but the effect would have been no different from moving coins around, either inside the vault or out. Even if the bank had issued credit or debit cards, a 5 guilder payment would still amount to nothing but a reshuffling of coins.
10. August 2009 at 06:19
Mike; You said:
“If the idea of fiat money were correct, then in principle, the central bank could very well have no assets, and nothing with which to buy back the paper dollars. In that case the paper dollars would lose some or all of their value. In that scenario, the value of paper money is too fragile to be believable.”
This is simply wrong. If there was a 20% chance that the value of money would fall to zero in 12 months, then the opportunity cost of holding fiat money would be 20%. Big deal, there are lots of countries that have averaged 20% inflation for decades and people keep using their non-interest bearing cash just the same because it is so convenient. So the demand for cash wouldn’t fall to zero even if unbacked and even if people thought it might later be replaced by near monies.
You still haven’t answered my question of why people hold so much non-interest-bearing fiat currency, if, as you say, interest-bearing near monies are just as good. I don’t see how your Bank of Amsterdam example relates to the question.
11. August 2009 at 14:13
Scott:
The backing theory view of why people hold paper money is no different than the conventional view: People hold paper money because it is the most convenient form for certain purchases. In my case, paper dollars are most convenient for about 10% of my purchases, while credit card dollars are most convenient for about 60%, and checking account dollars for 30% (not counting the check I use to pay off my credit card charges). The point of my Bank of Amsterdam example is that banks can easily issue money in whatever forms its customers prefer. There is nothing special about paper money.
That 20% chance would be closer to 100%. In the case of a true fiat dollar, where the central bank has no assets with which to buy back its paper dollars, the dollar would come under the same kind of short-selling attack that always happens when a bank is unable to buy back its paper money at par.
One scenario: A Cayman islands bank issues eurodollars, but instead of issuing them in the usual form of checking account dollars, the cayman bank issues them as blue paper eurodollars. As these eurodollars are issued, the value of US dollars falls, and the cayman bank, being in a short position in US dollars, profits from the fall of the US dollar. The cayman bank could even offer to share its profits with people who use the eurodollars in preference to US dollars.
(That is actually a similar situation to how private banks in the US issued paper dollars during the National Bank/Greenback era.)
Now, you might say that people would not use the cayman dollars, but if you believe in the existence of fiat money, you should believe that it can be issued by many countries, even ones that are small, weak, and close together. In this case, rivalry between various moneys would clearly be too strong for a true fiat money to survive.
12. August 2009 at 10:30
Mike; You said;
“Now, you might say that people would not use the cayman dollars, but if you believe in the existence of fiat money, you should believe that it can be issued by many countries, even ones that are small, weak, and close together. In this case, rivalry between various moneys would clearly be too strong for a true fiat money to survive.”
This still doesn’t address network effects. I think that people prefer to use one unit of account—dollars. We keep going around in circles. You talk about near monies. I point out that if they were really close substitutes then no one would use non-interest-bearing cash. You point to cash substitutes and I point to network effects. Since everything in stores is prices in US dollars, I’m not surprised that stores don’t want to want to mess with Caymen Island dollars.
Your last sentence doesn’t establish anything, it merely assumes away what we are debating. I don’t think an unbacked US dollar would be perceived as weak at all—I think people would keep using it as long as the government kept the supply of dollars roughly in line with demand, so that you didn’t have runaway hyperinflation.
14. August 2009 at 07:06
Scott:
I haven’t disputed network effects, or even the idea of money demand. What I have said is that if there is such a thing as fiat money, whose value is determined only by supply and demand, and not by backing, then the appearance of any rival money will reduce the demand for the fiat money and thereby reduce its value. I’ve mentioned two kinds of rival moneys: (1) foreign paper money, and (2) domestic derivative moneys, like checking accounts, credit cards, gift cards, scrip, travellers’ checks, etc.
In both cases, a crucial point is that the issuer of rival money profits from its issue of money. Foreign countries would profit by getting a piece of the free lunch that had previously belonged to the domestic country. Domestic issuers of derivative moneys profit because the issue of that derivative money puts them in a short position in the domestic currency, so they profit from the inflation that they themselves caused. The existence of fiat money thus implies an arbitrage opportunity. In finance, the presence of an arbitrage opportunity usually ends any debate.
Note that the inflation caused by rival moneys can occur only with a true, unbacked fiat money, and not with backed money. Thus, fiat moneys will self-destruct, and backed moneys will survive. If the idea of fiat money were correct, there should be some central bank somewhere that holds no assets against the money it issues, but there is no such central bank, and there never has been. On the Fed’s balance sheet, you will note an asset called “Collateral held against Federal Reserve Notes”. Enough said.
Your last sentence, about the government keeping the supply of dollars roughly in line with demand, assumes that in the case of a reduction in money demand, the government has the means to buy back some of its dollars. Buy them back with what? If the dollar were a true fiat money the government would have nothing with which to buy them back.
14. August 2009 at 11:35
mike, I agree with your first paragraph, but you lose me in the second when you talk about arbitrage. Arbitrage assumes away the issue we are debating, as it requires perfect substitutes. And I keep saying that interest-bearing near monies can’t be perfect substitutes if people still hold $800 billion in zero interest Federal Reserve Notes.
You said:
“Note that the inflation caused by rival moneys can occur only with a true, unbacked fiat money, and not with backed money. Thus, fiat moneys will self-destruct, and backed moneys will survive. If the idea of fiat money were correct, there should be some central bank somewhere that holds no assets against the money it issues, but there is no such central bank, and there never has been. On the Fed’s balance sheet, you will note an asset called “Collateral held against Federal Reserve Notes”. Enough said.”
I’m not so sure. The Fed doesn’t really “own” the bonds it holds. After all, the interest it earns must be given back to the Treasury. So the Fed is just part of the consolidated Federal government, de facto if not de jure. And Treasury bonds aren’t net wealth to the Federal government, as they are both an asset and a liability.
Regarding your last sentence, the fiscal view is right that if you have some sort of nominal target, the fiscal authorities must be willing to adjust taxes and spending so that the Fed can make that happen. I agree with that part of the fiscal view. My point is that the Fed usually decides first, and then Congress must scramble to do what it takes to meet its long run budget constraint.
17. August 2009 at 10:05
Scott:
Arbitrage still happens with imperfect substitutes. There are always people who are on the margin between using, say, dollars or yuan, or paper dollars vs checking account dollars. If stability in China makes the Yuan more attractive, or if debit card machines make checking account dollars more attractive, then on quantity theory principles, the demand for paper dollars will fall, and the value of the dollar will fall. Both the Chinese and the issuers of checking account dollars will profit from this inflation, so there is a strong force (arbitrage) pushing the value of the paper dollar toward zero. Note that this only happens if the money is unbacked. The value of backed money would be unaffected by this kind of rivalry.
If the fed were completely separate from the government, then the fed’s balance sheet would show dollars on the liability side, and gold and bonds on the assets side. The bonds, of course, are backed by the government. This means, for example, that if the government became financially stretched, and the value of the bonds fell, the dollars would lose value too.
If the fed and the government were one, then the government’s balance sheet would show bonds and dollars on the liability side, while assets would consist mainly of ‘taxes receivable’. It is these taxes that back the bonds and the dollars. In the case where the fed and the government were separate, we thought of the dollars as being backed by the bonds, and the bonds as backed by taxes. In the combined case, we see that it was taxes backing everything all along. The reshuffling of bookkeeping entries makes that clearer, but nothing important changes. In real life, of course, the distinction between the fed and the government is fuzzy.
About your last paragraph: Think of the combined case. When the fed issues more dollars, it usually buys government bonds with them. If the liability side of the government’s balance sheet originally showed $10 trillion of bonds, plus $2 trillion in currency, then the issue of another $1 trillion in currency would show up on the balance sheet as a $1 trillion increase in currency, offset by a $1 trillion DECREASE in bonds. Remember that those bonds, which used to be in the hands of the public, are now in the hands of the government. Once this happens, those bonds are no longer the liability of the government.
18. August 2009 at 11:54
Mike Sproul: “Agreed so far. I’d then add that if the interest rate is 5%, then a costlessly issued IOU promising 100 silver shillings in 1 year would be worth about 95 silver shillings today.”
Yes. Assuming someone wants such an IOU at that interest rate.
Mike Sproul: “If the costs of issuing that IOU amounted to 3 shillings per year, then the same IOU would sell for 98 shillings today and would therefore yield 2%, rather than 5%.”
Assuming a very deep and competatative market no profit for the operator. Again, this is fine assuming that 2% is the market interest rate.
If the market interest rate were 5% clearly the bill wouldn’t fetch such a high price.
Mike Sproul: “If the cost of issue were 5 shillings/year, then the IOU would sell for 100 silver shillings all year long, yielding 0% to the holder. If that IOU has some redeeming feature, like usefulness as money, people will hold it in spite of earning no interest.”
Yes.
Mike Sproul: “But the existence of that IOU will give people the false impression that money somehow follows different rules than ordinary IOU’s, when in fact it doesn’t.”
Clearly though money does follow different rules than ordinary IOUs. The things you have mentioned above don’t affect it’s special properties.
Money is widely accepted as a medium of exchange. Therefore it provides the holder with a store of value. The holder can dispose of that store of value as they wish.
You have hidden the most relevant feature of money in the phrase “If that IOU has some redeeming feature, like usefulness as money”.
A bill issuer is a very different organization to a banknote issuer. In places where there was free-banking though many businesses could issues bills and banknotes.
Would you agree with any of this?
18. August 2009 at 18:51
Current:
“this is fine assuming that 2% is the market interest rate.”
What I meant was that in a world where a costless bond would yield 5%, a bond with a 3% annual carrying cost would yield 2%.
“A bill issuer is a very different organization to a banknote issuer. In places where there was free-banking though many businesses could issues bills and banknotes.”
Since you wanted to talk about a free banking situation, we’d have to rule out fiat money. Nobody claims that privately-issued bank notes were fiat money. Rivalry between banks would assure that every banknote was worth its backing, and the backing theory would be applicable while the quantity theory would not.
19. August 2009 at 01:31
Mike Sproul: “What I meant was that in a world where a costless bond would yield 5%, a bond with a 3% annual carrying cost would yield 2%.”
I don’t really understand what you mean. When you are talking about cost I presume you are talking about the administration cost to the issuer.
Perhaps my rate of time-preference is 5%. It would cost me £1 to issue a 1 year bond. That means if I someone wanted to borrow £100 from me they would have to pay me >£106 in a years time to make it worth my while. Whether this transaction takes place or not depends on whether borrowers can get a better deal elsewhere.
Mike Sproul: “Since you wanted to talk about a free banking situation, we’d have to rule out fiat money. Nobody claims that privately-issued bank notes were fiat money. Rivalry between banks would assure that every banknote was worth its backing, and the backing theory would be applicable while the quantity theory would not.”
I agree that we have to rule out fiat money. However I think that both backing and quantity theory are relevant. The well thought-out froms of quantity theory still apply. We have a situation where quantity equilibrium is kept in place by the market.
19. August 2009 at 02:18
Mike,
Consider the equation:
Ms = k * p * S
S is the quantity of Swedes transacted on the market for swedes.
p is the prices of Swedes.
Ms is the amount of money transacted in the market for Swedes.
k is the ratio of the demand for swedes to the stock of swedes.
If we replace “Swedes” with “Money” the above is still true. So the quantity equation is generally true, its a tautology when applied to money or anything else. The question though is whether or not it’s particularly useful.
As Scott points out the quantity equation isn’t the quantity theory. Scott listed several quantity theories recently
http://blogsandwikis.bentley.edu/themoneyillusion/?p=2167
Scott’s first proposition #1 is generally not correct and not correct in the free banking case. In the free banking case exogeneous increases in M without other occurances are impossible, so #3 and #4 are irrelevant. However #2 remains reasonably accurate. Of course #2 is made true by the way banks behave, it is not “automatically” true.
19. August 2009 at 03:35
Mike, I’m looking for a new way to get at this, as we keep circling around to the same issues. Let’s assume the Fed is split off from the government. The Fed gives the Treasury all its assets, so the fiat currency is no longer backed. The Fed announces the following policy regime:
1. From now to the end of time the Fed will never reduce the monetary base. Instead the base will be increased every year by 3%. This will be done with open market operations. The Fed will not hold any assets, instead it will donate the bonds it aquires in OMOs to charity.
How do you see that working? I think that sort of fiat regime would not result in hyperinflation, despite the fact that the fiat money is not backed. The only qualification is that I wouldn’t want them to start from the current bloated base, but from a more normal level. Thus I am saying that had they adopted this policy in 2007, until the end of time, it would not have resulted in hyperinflation, at least in the near term.
20. August 2009 at 15:52
Scott:
The backing theory says the value of money is equal to the value of the assets backing it, so given the conditions you describe, the dollar would be worthless.
If the dollar held its value, then either the fed, the government, or your charities would get a free lunch. Given the usual theories about fiat money, any government, from the USA to some primitive tribe, is capable of issuing it. As they do so, and get that free lunch, rival moneys will vie for the free lunch, thus reducing its value to zero. As I said, backed moneys will survive,and unbacked moneys will not.
The quantity theory says that money is a ‘something from nothing’ deal. The backing theory says money is a ‘something from something’ deal. The backing theory says the dollar is backed but not physically convertible. The quantity theory says the dollar is unbacked but still manages to have value anyway. The backing theory is consistent with the fact that no central bank has ever operated without assets, the quantity theory is not.
In every case, the backing theory falls squarely in the realm of plausibility, and the quantity theory does not.
21. August 2009 at 09:47
Current:
1) If we rule out fiat money, and if we’re talking about a free banking regime, then the value of money can’t be anything but equal to the value of the bank assets backing it. Thus the backing theory applies and the quantity theory does not.
2) The equation of exchange is useless in any context. To extend your example, we could say M=# shares of GE stock, V=velocity of GE stock, P=# shares to buy a unit of y, and y= # units of goods bought with GE shares. The equation will be ‘true’ in the sense that the amount of GE shares spent=the amount of GE shares received, but no amount of tinkering with that equation will help you find the value of GE stock. That’s why no stock analyst uses the equation to analyze the value of stock, and it’s why no economist should use it to analyze money or anything else.
21. August 2009 at 11:03
Mike Sproul: “1) If we rule out fiat money, and if we’re talking about a free banking regime, then the value of money can’t be anything but equal to the value of the bank assets backing it. Thus the backing theory applies and the quantity theory does not.”
In a free banking regime nobody would accept for long the notes of a bank that isn’t properly backed. So, as you say backing theory applies.
Mike Sproul: “2) The equation of exchange is useless in any context. To extend your example, we could say M=# shares of GE stock, V=velocity of GE stock, P=# shares to buy a unit of y, and y= # units of goods bought with GE shares. The equation will be ‘true’ in the sense that the amount of GE shares spent=the amount of GE shares received, but no amount of tinkering with that equation will help you find the value of GE stock. That’s why no stock analyst uses the equation to analyze the value of stock, and it’s why no economist should use it to analyze money or anything else.”
The equation of exchange is a tautology used to illustrate points. It contain no real “content”.
My point was that in the situation I have described the quantity theory applies. As Scott said recently this is different from the tautological Quantity equation.
Scott put the theory in this way “2. The ratio of P*Y and M is relatively stable.” I would say, first take
M = k P T
Where T is total goods traded, P is prices and M is amounts of money transacted. k is the relationship between them. In classical terms it is the ratio between the stock for money and the demand for it.
Now, in the long run k is quite stable. It varies when there are recessions. There are other variation in the very long term due to changes in payment techniques. Wouldn’t you agree with this?
Now, to move the discussion on a little. Do you think that we now live in a world of effectively free banking?
22. August 2009 at 05:56
Current:
“Now, in the long run k is quite stable. It varies when there are recessions. There are other variation in the very long term due to changes in payment techniques. Wouldn’t you agree with this?”
There is also a relatively stable relationship between the number of shares of GE stock that exist, and the number of shares sold in any given period. It’s just a meaningless number.
“Now, to move the discussion on a little. Do you think that we now live in a world of effectively free banking?”
It’s a mixed bag. Features of our system that are anti-free banking include the fed’s monopoly of note issue, and various government restrictions on bank lending, reserves, etc. Features favoring freedom include the relative lack of regulation of credit card companies, gift certificates and scrip, eurodollars, and foreign currencies.
22. August 2009 at 09:05
I don’t understand why you think that it’s meaningless. What the number represents is the relationship between demand and stock.
This is always the problem with rival theories. A person who holds another theory cannot be persuaded that the concepts that rivals hold are useful. This is because they aren’t useful to the theory he holds.
I think that variations in the demand for money versus the stock of it are useful to economics in situations where there isn’t free banking.
Fair enough. There are a lot of regulations of the things that you mention though, at least in Britain and Ireland. I don’t understand how gift certificates come into the picture.
Anyway, you have agreed earlier that a person holds a reserve of money to pay for unforseen expenditures. That is, I take 100 euro from the bank to use for things like shopping, cinema and restaurant visits, etc where I can’t predict exactly when and where I will need the products. For situations where I can’t know beforehand exactly what I’ll want to buy. (If I did know I could buy it beforehand).
Now, would you agree that a person who lives in Ireland, for example, must hold this reserve of cash in Euros?
24. August 2009 at 00:54
Mike, You said,
“Given the usual theories about fiat money, any government, from the USA to some primitive tribe, is capable of issuing it.”
If they issued dollars they’d be imprisoned for counterfeiting. If they issued competing “tollars” they’d find no market. Stores wouldn’t accpt them in transactions.
You said,
“The quantity theory says that money is a ‘something from nothing’ deal.”
Suppose we are using gold coins as money. Isn’t that wasteful? That gold could make useful necklaces. And pieces of paper would provide transactions services just as well as money. We can replace gold coins with paper money. They sell off the gold coins that were bought back by the central bank. The government then has money to be used for other purposes. We do get something for nothing. A $20 gold coin can be replaced by a $20 bill that is produced at a cost of 4 cents. And then we use the gold for something more useful. Fiat money increases our welfare–it is like a technological improvement over commodity money—less real resources required.
Of course if we then further increase the quantity of fiat money, the QT tells us there is no free lunch, as it imposes an inflation tax on cash holders.
24. August 2009 at 02:08
Scott,
I think we both disagree with Mike. However, I think there are some problems with your arguments against him.
This is actually a valid criticism of some fiat money theories. Fiat money ultimately rests upon state force. If some agency issues “tollars” they may well find a market for them if the government is depreciating the dollar quickly. However the state can use it’s powers of force to stop this from happening. For example, in the UK many transactions can only legally be made in pounds. If the depreciation is very large though then even this becomes irrelevant.
Then reversion to barter occurs.
Surely the question is much more complicated than that?
Note though that what you say indicates that all money provides a form of “free lunch”. All types of money provide a means of indirect exchange which is vastly preferable to direct exchange. Sound money provides a means of indirect exchange over reasonable periods of time.
24. August 2009 at 09:25
Current:
The equation of exchange is ‘true’ of GE stock, but nobody claims that it is useful for finding the correct value of GE stock. The value of GE stock is determined by the value of GE’s assets. I’m saying that the equation is also ‘true’ of money, but it is not useful for finding the correct value of money, since the value of money is determined by the value of the issuing bank’s assets.
I mentioned gift certificates because they are money. Where I live, we have Disney dollars and Toys-r-us dollars. They are printed to resemble dollars but they have a picture of Mickey mouse or a giraffe instead of George Washington. They aren’t as widely accepted as green dollars, but if you owe me $20, I’ll accept giraffe dollars, because I know I’ll be able to use them when buying toys.
I haven’t disputed the fact that people hold stocks of cash. What I have said is that people hold cash because it has value, as opposed to the QT proposition that cash has value because people hold it.
Scott:
I wasn’t talking about counterfeiting. I was talking about money issued inside some government’s jurisdiction, but which crosses borders and is used elsewhere. The country whose currency invades another gets a free lunch, according to the quantity theory view of fiat money. (but not according to the backing theory)
“Suppose we are using gold coins as money. Isn’t that wasteful? ”
That’s missing the point. For example, suppose some paper currency is backed by land, with each shilling of paper laying claim to 1 square foot of land. The land can still be used, but people can also use the shillings to trade with. No resources are wasted. The quantity theory view of fiat money says the shillings would hold their value even if the land backing were lost, while the backing theory says that the shillings would lose value. So the QT implies that those shillings are a something from nothing deal, while the backing theory doesn’t.
24. August 2009 at 10:49
Certainly.
I understand what your conclusion is. I dispute it though I think that the situation of GE stock is different to that of money.
I don’t see how something as specific as a Toys-r-us token could be considered money. I have no need for toys so I couldn’t use a Toys-r-us token. I certainly wouldn’t accept it as money.
Surely these token are closer to being a form of symbolic barter than money. A Toys-r-us token grants the holder a certain amount of any goods from Toys-r-us, but not anywhere else. Anyone who doesn’t want toys must engage in an extra step of indirect exchange with someone who does want a Toys-r-us token, or give the token away.
I also think that people hold stocks of money because they have value. I hold a stock of money because I can use it for whatever purchases I want to make in the near future.
Would you deny that people hold money for this reason?
Would you deny that the citizens of one particular state can only hold that state’s money except in very rare circumstances.
25. August 2009 at 18:52
Current:
“A Toys-r-us token grants the holder a certain amount of any goods from Toys-r-us, but not anywhere else. ”
By the same token, a one shilling bank note issued by the bank of Edinburgh gives the holder the right to claim a shilling coin from that bank, but not from anywhere else. There is always a spectrum of moneys in circulation, some widely accepted and some of limited use. But the fact that some kind of money is only used in a limited area does not mean that it is not money.
Of course, people generally hold their own country’s money. But there are always people at the margin, who, from one day to the next, might switch from dollars to euros to shillings and back. That’s where competition between moneys is most active.
26. August 2009 at 02:01
I think there is still a difference though. If the Bank of Edinburgh is a sound bank then anyone who possesses a banknote can rely on it to provide them with a shilling coin. That shilling coin will have an exchange value that they are familiar with. The note is for a coin which is what Mises calls “money in the narrower sense”.
However, the note for Toys’r’us isn’t. It provides the owner with a claim against toys. It is not a simple matter to correlate the prices of toys at Toys’r’us with other goods.
The management of Toys’r’us may become poor, for example, and there may be few good toys to buy using the tokens. It is hard for the holder of a token to know if this will happen. The management have no commitment to offer a certain quality of goods for the token. The only thing that ensures that they do is that most transactions don’t take place through the tokens.
Still, I agree with you that there is a “spectrum” of things in circulation. In any particular place some are closer to money and some further away.
Yes, I’m one of those people. I’m English, but I live in the Republic of Ireland, which is in the Euro-zone. I own a property in England which I rent out. I have a job in Ireland and I live there. So, I have checking bank accounts in both countries. I have a UK account in pounds and an Irish account in Euros.
But, that doesn’t mean that to me there is much “competition” in money though. The problem money solves is the lack of knowledge over what immediate short-term expenses will be. To cover this uncertainty I must hold money in checking accounts in both countries. So, I must hold money in proportion to the extent of this problem in both countries. Since I know fairly well my ingoing and outgoings in pounds my UK bank account is normally quite empty.
Whenever I know I won’t need an amount of money for a reasonable time I invest it wherever I think is best in savings.
The only way competition comes in is that sometimes I can buy a product in euros or in pounds. There are probably agents for whom this competition is much more important because they can buy most of the things they want in either currency.
27. August 2009 at 03:00
Current, I agree with the first part of your comment. But my point is that you can have a lot of inflation before people start looking for alternatives like barter or foreign monies, even where those alternatives are perfectly legal.
I suppose money is a free lunch, in the same way that any technological improvement is a free lunch.
Mike, It is wasteful if gold coins circulate, and also wasteful if thay back government-issued paper money. The gold has better alternative uses than backing money. And I’d hate to see the government become a big landowner.
I still don’t think foreign money is a significant threat to an unbacked fiat currency, unless the government prints so much fiat money that you have hyperinflation. Otherwise stores prefer to deal with a single unit of account. Some Canadian stores do accept US money, but not many American stores accept the CAN$.
27. August 2009 at 03:32
I agree. The quantity theory provides for the “space” you describe. Where there is no inflation, or reasonable inflation, people don’t resort to foreign currencies the costs are higher than the benefits. This is a very normal case and that is why quantity theory is very important for monetary discussions.
28. August 2009 at 09:37
Scott:
“It is wasteful if gold coins circulate”
But the case in question is where some existing money is backed, but no gold circulates, or is held in vaults, so there is no waste. Historically, many backed moneys have fit this case, and a land-backed currency would be just one example.
Anyway, if the backing is suddenly removed, and the currency held its value, then that money would really be a something from nothing deal.
Also, if the US $ invades Canada, then according to the quantity theory, the US gets a free lunch while Canada loses one. That should leave you wondering why countries aren’t fighting like crazy over that free lunch.
28. August 2009 at 10:14
Though I don’t agree with Mike he is right in part. Historical gold standards have not being 100% standards. The amount of gold held for every note has been quite low, even less than 5%. For the rest of the liabilities the banks have held loans (bills) as assets against them.
Of course the situation with gold coins is different.
I would say that despite the “waste” of gold coins and gold standards they serve a very practical purpose. That is that they make inflation much harder for governments. Not impossible, but more difficult.
30. August 2009 at 08:33
Current:
Competing moneys will reduce the demand for so-called fiat money, and reduce its value. Closely competing moneys will have a strong effect and poor substitutes will have a weak effect, but the presence of any effect leads to absurd conclusions, like saying that legitimate issuers of foreign moneys or credit card dollars gift certificates are equivalent to counterfeiters. The backing theory doesn’t lead to those absurd conclusions.
30. August 2009 at 16:02
Mike, I don’t understand the land example. Are you proposing that it would be efficient to have the government become a big landowner?
You said;
“Also, if the US $ invades Canada, then according to the quantity theory, the US gets a free lunch while Canada loses one. That should leave you wondering why countries aren’t fighting like crazy over that free lunch.”
But doesn’t the US get a free lunch when other countries adopt our currencies (like Ecuador?) Isn’t it an interest free loan from Ecuador to the US?
They don’t fight like crazy because the gains from seignorage are small unless you have really high inflation. But in that case the harm from the high inflation exceeds the benefits of the seignorage. And in any case no foreign country would want to adopt another country’s currency if it was highly inflationary.
Current, You said;
“Though I don’t agree with Mike he is right in part. Historical gold standards have not being 100% standards. The amount of gold held for every note has been quite low, even less than 5%. For the rest of the liabilities the banks have held loans (bills) as assets against them.”
Not true, they were typically more like 40%. In the 1920s the US held gold reserves of far more than 50% of the base. I don’t have the exact figures with me, but I think it might have even been 75%.
Otherwise I agree with you.
Mike, You said;
“Competing moneys will reduce the demand for so-called fiat money, and reduce its value. Closely competing moneys will have a strong effect and poor substitutes will have a weak effect, but the presence of any effect leads to absurd conclusions, like saying that legitimate issuers of foreign moneys or credit card dollars gift certificates are equivalent to counterfeiters. The backing theory doesn’t lead to those absurd conclusions.”
Foreign money is not counterfeiting because it isn’t denominated in US dollars. Gift certificates are not generally accepted outside of the store in question. You can’t take them to any bank and redeem them for US dollars. So they aren’t close substitures for cash, nor do people regard them as such.
30. August 2009 at 18:59
If you take the 2007 base, then we’re still backed by 25% gold:
The H.41 reports the Gold stock of the Federal reserve system as 11,041B. Seems piddling right? Well its based at historic cost working out to an average of $42/oz. So the holdings are worth closer to 250B.
If we stretch our imaginations a bit, the US holds another 100B claim against the IMF gold. Another 100B of backing in the SPR, and another 100B of back in foreign asset holdings. Which brings us to your 50% backing.
See things aren’t so different… 🙂
31. August 2009 at 00:20
That’s not what I’ve read elsewhere.
Are you considering only notes here or notes and deposits?
1. September 2009 at 08:26
Scott:
“I don’t understand the land example. Are you proposing that it would be efficient to have the government become a big landowner?”
You brought up that it’s wasteful to use gold coins, and that the issue of so-called fiat money prevents this waste. You then used this to claim that fiat money is not a ‘something from nothing’ deal. My land example pointed to a case where backed money doesn’t involve any waste. The question hanging fire is: In the case where backed money involves no waste, is fiat money a something from nothing deal? I say yes, which is one reason to believe that fiat money doesn’t exist.
“But doesn’t the US get a free lunch when other countries adopt our currencies (like Ecuador?) Isn’t it an interest free loan from Ecuador to the US?”
Ask yourself the same question about gift certificates from sears or walmart. If a sears dollar circulates in mexico, did sears get an interest-free loan from mexico? If they did, we’ve just violated the zero profit theorem. Competing stores would start offering gift certificates that bear interest, until the cost of issue just burns up the so-called interest-free loan. What’s true for gift certificates is true of government-issued paper money.
“Foreign money is not counterfeiting because it isn’t denominated in US dollars. Gift certificates are not generally accepted outside of the store in question. You can’t take them to any bank and redeem them for US dollars. So they aren’t close substitures for cash, nor do people regard them as such.”
No. Foreign money is not counterfeiting because foreign money is recognized by that country as its liability, not the fed’s liability. Whether they denominate in dollars is irrelevant.
And whether you call gift certificates money or not, the important fact is that they reduce the demand for base money, thus (according to the quantity theory) causing inflation. This leads to all sorts of absurdities, like saying that a counterfeiter has the same effect on the value of the dollar as a legitimate banker does.
1. September 2009 at 16:26
@ssumner,Sproul:
This is an interest-free loan only if the money is detained as cash. Otherwise, dollar deposits in a US bank are interest-bearing (or service providing) and dollar deposits in the country
This is one interest of currency boards compared to full-dollarization(euroization): the currency board makes an interest-bearing deposit in the US while his notes and coins are interest-free.
@Sproul:
“If a sears dollar circulates in mexico, did sears get an interest-free loan from mexico? If they did, we’ve just violated the zero profit theorem”:
That’s why there is not that much sears dollar in Mexico…
A good question however is why people buy gift certificates… The only reason I see is that there is some repugnance to make cash gifts.
“This leads to all sorts of absurdities, like saying that a counterfeiter has the same effect on the value of the dollar as a legitimate banker does.”: this is not an absurdity, this is true! But the banker *pays* to have the right to do this, either by providing services either by giving some interest
2. September 2009 at 00:46
Jean,
See my comments on gift certificates in above. Despite what Mike says, gift certificates are significantly different to money.
2. September 2009 at 16:49
Jon, Those numbers sound right. And they would have even more gold if they hadn’t sold some off.
Current, The government’s gold stocks only backed the monetary base (notes plus bank reserves.) They were not used to back private bank liabilities.
Mike, I’m not saying you are wrong about land, I am just having trouble following your argument. The standard argument is that gold reserves are wasteful because there is an opportunity cost to holding gold. What about land? Is it use productively? If so, how can the government engage in OMOs without incurring large transactions costs?
You said;
“Ask yourself the same question about gift certificates from sears or walmart. If a sears dollar circulates in mexico, did sears get an interest-free loan from mexico? If they did, we’ve just violated the zero profit theorem. Competing stores would start offering gift certificates that bear interest, until the cost of issue just burns up the so-called interest-free loan. What’s true for gift certificates is true of government-issued paper money.”
Ironically, I was going to argue (in a future post) that competition in currency is wasteful, because price competition is impossible. Thus you end up with inefficient non-price competetion.
North Korea competes with the US by producing Federal Reserve Notes, and many other countries would do so if they thought they could get away with it.
I use the counterfeiting example as a way of teaching the QTM. Any permanent increase in the money supply, whether by a counterfeiter or by the Fed, has the same impact on the price level.
Jean, I agree that a currency board avoids the loss of seignorage that you incur with full dollarization. Of course the currency board is also more susceptable to a speculative attack on the currency.
4. September 2009 at 06:53
Scott:
I’m saying that the fed could, in principle, hold land instead of gold or bonds. The land could still be farmed, so there would be no waste. In reality, the transaction costs of conducting open-market operations with land would be high, so it’s better for the fed to use bonds. But of course some bonds are backed by land anyway. Even government bonds are indirectly backed by land, since if I don’t pay my taxes, the IRS will take my land. In any case, the whole point of this was that in principle, money that is backed and convertible need not involve any waste of resources like gold. You had claimed that the free lunch created by fiat money was paid for by this elimination of waste, but in this case there’s no waste to eliminate, so who pays for the free lunch?
When a central bank, or any bank, issues paper or credit money, they put their name on that money, recognize that money as their liability, acquire assets to back that money, and stand ready to use those assets to buy back their money. North Korea and other counterfeiters do none of those things. The difference between banks and counterfeiters could hardly be clearer, and it is a sign of the underlying incoherence of the quantity theory that it implies that counterfeiters and bankers have “the same impact on the price level”. (I have to give you credit for saying that explicitly. Most economists start to equivocate on this point, since it is so fishy.)
If the fed has issued a trillion dollars, the fed will have a trillion of gold and bonds, which it can use to buy back the entire trillion. If the fed issues another $100 billion, the fed will get another $100 billion of bonds. Now it will have 1.1 trillion in assets backing 1.1 trillion in dollars, and it is still capable of buying back all the money it has issued. The backing theory says there would be no inflation. (It also says that the extra 100 billion would only be issued if the public’s desire for cash had increased 10%, so that people are willing to surrender 100 billion in bonds to get that cash.) But if that $100 billion had been issued by North Korea, the fed’s assets would still be just 1 trillion, and the fed could no longer buy back all the dollars. The backing theory implies 10% inflation in this case. The quantity theory makes no distinction between the two cases, and says there would be 10% inflation either way. It’s a bad theory.
4. September 2009 at 07:07
Regardless of Mike’s other views he is quite right about land. Central banks have used land backed notes on several occasions in history. In Germany after the hyperinflation of the 20s for example, and also in France after a hyperinflation.
The problem is that the most common reason to redeem currency is to obtain another sort of money than can be used abroad. Land in France isn’t a very liquid commodity in, say, South Africa. So, as I understand it, an importer who wanted gold had to sell the land backing his money in exchange for gold.
Such a form of backing also makes local agricultural output relevant to monetary policy. Oddly it means that farm subsidies could be used in nefarious ways to change the value of money. (I don’t think this was an issue when it was done, but it would be now).
I don’t think it’s the best monetary system, but it’s not the worst either.
9. September 2009 at 02:30
Mike, You said,
“The land could still be farmed, so there would be no waste.”
That’s a pretty big assumption, as the biggest waste in all of human history occurred when the Chinese government owned land. But it’s a moot point as in the next line you correctly note that that OMOs would not be practical with land.
Regarding my free lunch argument, that applied to alternative monies that use real resources. If you imagine a government that can raise money at zero resource cost through lump sum taxes, and imagine Ricardian equivalence holds, then fiat money may not result in an efficiency gain, but I don’t see how that supports the backing theory, as under Ricardian equivalence the government bonds that back money would not be net wealth from the public’s perspective.
You said;
“When a central bank, or any bank, issues paper or credit money, they put their name on that money, recognize that money as their liability, acquire assets to back that money, and stand ready to use those assets to buy back their money.”
I don’t follow this, I thought modern central banks no longer made promises. Indeed I think that one cause of this recession is that the Fed doesn’t promise to maintain any particular price level, and lets prices wander around unpredictably.
You said;
“If the fed has issued a trillion dollars, the fed will have a trillion of gold and bonds, which it can use to buy back the entire trillion.”
If I am a counterfeiter and printed a billion dollars I’d buy a billion in T-bonds and live off the interest, so my money would be fully backed. You might say that not all counterfeiters would be so wise, some would blow it on consumption. But then again so do some central banks in banana republics.
Current, It may not be the worst system, but it’s pretty close. Imagine open market operations.
14. September 2009 at 09:20
Scttt:
If there is a case where fiat money does not create an efficiency gain, and yet the issuer of that fiat money gets the (asserted) free lunch, then the issue of fiat money steals wealth from the public, just like a counterfeiter. Doesn’t that contradict common sense? Or do you accept that?
Central banks used to promise to buy back their dollars for a certain weight of metal, at the customer’s option. Now they promise that they will probably, at their option, buy back their dollars with bonds. That’s still enough of a promise to give the dollar value. Corporate stock has value because of similarly vague promises.
“If I am a counterfeiter and printed a billion dollars I’d buy a billion in T-bonds and live off the interest, so my money would be fully backed.”
If you were a counterfeiter you would not have printed your name on those dollars. You would have fraudulently printed the fed’s name on them. You would not use your bonds to buy back your dollars, and they would not appear as a liability on your personal T-account, so they would be fully UNBACKED. If your dollars could be told apart from the fed’s dollars, they would be worthless. If they could not be told apart, then they would be a claim against the fed. The fed would have less backing per dollar, and there would be inflation.
16. September 2009 at 04:44
Mike, You said;
“You would have fraudulently printed the fed’s name on them. You would not use your bonds to buy back your dollars, and they would not appear as a liability on your personal T-account, so they would be fully UNBACKED.”
But I could have backed the bonds. Suppose I decided to counterfeit for 10 years, buy T-bonds, and live off the interest on the T-bonds. But I didn’t touch the principal. Then after ten years I bought back all the cash with the bonds I held. Isn’t that type of counterfeiting backed?
I am not sure about the free lunch issue. Is it like Samuelson’s old argument that a PAYG social security system was a free lunch if it lasted forever, because the public could borrow from infinitely far in the future? Do I have to assume that the fiat currency system would be expected to last forever?
Suppose I agree that at some future date the money stock is redeemed because of changes in technology. Say 23,456 years in the future. Does that mean that the QT cannot be approximately true right now, just because the money will eventually be backed? I suppose I still don’t understand the backing theory, as the Fed is actually part of a broader government. Even if the Fed lacked assets, there might be an implicit promise that in a pinch the broader government would raise enough revenue through taxes to prevent hyperinflation (by transferring those resources to the Fed.) But that makes the backing theory seem almost tautological, as you could never prove that money is not backed. But maybe I’m not understanding the key assumptions that make the theory tick, and differentiate it from the QT.
17. September 2009 at 17:45
Scott:
I think you meant to say you could have backed the dollars. If you were able to live off the interest, then issuers of rival moneys would start issuing interest-bearing moneys of their own, until the cost of issuing money made it a zero-profit proposition. Thus, if you had spent the interest, you would be unable to buy back the money at the proper price. If Walmart issued 100 gift dollars for a year, and invested the $100 paid for them at 5% for a year, then it would have $105 after a year. If it had cost $5 to issue those WM$ for a year, Walmart would have to spend $100 buying back the WM$, plus $5 on printing costs, leaving it with zero profit. But if the cost of issue was only $3, Walmart would have a $2 profit. In this case, rivalry from competing moneys would force Walmart to pay 2% interest on its dollars. Walmart would have to buy back its 100 WM$ for $102 at year-end, which with the $3 printing cost again leaves zero profit.
I’d rather not be side-tracked with the social security free lunch, except to say that I don’t believe that free lunch story either.
Before thinking of redemption in 23,456 years, try thinking of a 1-year horizon. A costlessly-issued dollar that promises 1 ounce of silver in 1 year would be worth .95 ounces today. It would bear 5% interest. Now add a simple complication: Suppose it cost .05 oz. to issue that dollar. Then that dollar would be worth 1 oz. today. It would bear no interest because the cost of issue exactly burns up the interest. Now extend the redemption date to 23,456 years, and the dollar will still be worth 1 oz. today. (I’m leaving out a lot of complications here, not least of which is the case where the bank suspends physical convertibility while maintaining financial convertibility.)
“I suppose I still don’t understand the backing theory, as the Fed is actually part of a broader government.”
You would understand the backing theory if you read “Backed Money, Fiat Money, and the Real Bills Doctrine”, followed by “There’s No Such Thing as Fiat Money”, both of which can be found by clicking my name above. And yes, the fed is part of a broader government. This just means the backing is larger than what shows on the Fed’s balance sheet.
17. September 2009 at 18:03
Mike You said;
“You would understand the backing theory if you read “Backed Money, Fiat Money, and the Real Bills Doctrine”, followed by “There’s No Such Thing as Fiat Money”, both of which can be found by clicking my name above. And yes, the fed is part of a broader government. This just means the backing is larger than what shows on the Fed’s balance sheet.”
But that is exactly my point. Suppose we try to falsify your theory. We find a fiat currency that has value despite the central bank having zero assets. You could just argue that there is an understanding, maybe just tacit, for the central government to step in and bail out the central bank if there is a drop in the demand for money and the central bank must redeem some money in order to prevent inflation. But in that case are the backing and anti-backing theories observationally equivalent?
On the counterfeiting, I thought we were debating something different, whether counterfeiting could be profitable if the currency was backed. If you are going to allow many counterfeiters and competition then I agree that is different, but I don’t recall that as being the question that triggered the discussion.
21. September 2009 at 11:34
Scott:
Here’s a simple example: The fed has 20 oz. of silver, plus US bonds worth 80 oz., as backing for 100 paper dollars that it has issued. Each dollar is worth 1 oz. The federal government, meanwhile, has assets (‘taxes receivable’) worth 800 oz., while its liabilities consist of US bonds worth 800 oz, so that the government’s net worth is zero. If we imagine that the government would bail out the fed, then we should combine the balance sheets of the fed and the government. The combined balance sheet of the resulting entity would show assets of 800 oz. of taxes receivable, plus 20 oz. of silver. On the liability side, the fed’s 80 oz. worth of US bonds are canceled, leaving liabilities of 100 paper dollars, plus 720 in bonds. The value of the dollar would then depend on both the assets and the liabilities of the combined institution. Empirical testing of the validity of the backing theory then becomes harder, but it is not ‘unfalsifiable’. There are some cases where the backing theory will be observationally equivalent to the quantity theory, and others where it won’t. The backing theory is not as slippery as the quantity theory in this regard, since quantity theorists can explain any anomaly by saying ‘Money demand must have changed.”, “Expectations must have changed.”, etc.
The thing that triggered the discussion of counterfeiting was your assertion that the issue of money by a bank has the same impact on prices as the issue of money by a counterfeiter. My point was that that is incorrect, even in the case where you imagine that you can live off the interest and ultimately buy back the dollars you issued.
26. September 2009 at 12:45
Mike, It’s hard for me when I come back to an unfamiliar area of scholarship after a few days away. But the bottom line is that I think backing matters when there is a serious question as to whether a government has the resources to pay its bills without resorting to monetizing the debt. If, on the other hand, the country has great ability to raise revenue and cut spending (like the US) then money-holders won’t worry about backing issues, and a forward-looking QT model will be best. Inflation will depend on the current money supply, and the future expected money supply relative to demand. In the example you gave the combined Treasury and Fed were almost broke, but I don’t see that as having relevance for the US right now.
Sorry if this doesn’t address your point, but this is a hard topic for me to handle. I have trouble seeing how the predictions of each theory differ, especially in cases relevant to the US today.
28. September 2009 at 09:00
Scott:
A government at the end of its fiscal rope might decide to issue new dollars and use them to pay its bills, or it might decide to issue new bonds to pay its bills. I expect your forward-looking quantity theory would predict that if new money were issued the money would depreciate, while if new bonds were issued then the bonds would depreciate. The backing theory says it would make no difference. Liabilities rise relative to assets either way, so depreciation of the money and the bonds happens either way. The backing theory goes a step further and says that if the new money or new bonds were spent on valuable assets, then assets would rise in step with liabilities and there would be no depreciation, even if the government were at the end of its rope and printed more money. Those are just a few of the differences in predictions made by the two theories. I also note that the backing theory explanation is more elegant in every case, since it uses simple accounting tools, rather than a lot of hand-waving about expectations.
30. September 2009 at 03:20
Mike, That’s a good summary of the differences. Yes the backing theory is more elegant. But it is also true that flexible price models are more elegant than sticky price models. Elegance doesn’t always equate with usefulness.
20. April 2012 at 05:33
[…] what I said three years ago: So let’s imagine an FOMC with the above 4 members [Bernanke, Svensson, Woodford, Krugman], plus […]
11. September 2012 at 06:28
[…] in 2009 I wrote the following: Someone either understands that the only reliable speedometer is NGDP (or inflation) or they […]
12. September 2012 at 23:15
[…] let their other responsibilities go by the wayside. Ignorance will damn us, not incompetence. As Scott Sumner has said of the 1930s, the last time we got things this wrong: The elite bankers and financiers of Wall Street were […]
23. December 2012 at 19:56
[…] are unqualified people serving in the most important economic policy position on the planet. I also argued that we should do whatever it takes to attract the best: I don’t care how much is costs, even if […]
24. December 2012 at 21:06
[…] They are unqualified people serving in the most important economic policy position on the planet. I also argued that we should do whatever it takes to attract the best: I don’t care how much is costs, even if […]
24. December 2012 at 21:06
[…] They are unqualified people serving in the most important economic policy position on the planet. I also argued that we should do whatever it takes to attract the best: I don’t care how much is costs, even if […]
25. December 2012 at 07:13
[…] They are unqualified people serving in the most important economic policy position on the planet. I also argued that we should do whatever it takes to attract the best: I don’t care how much is costs, even if […]