“If the market price of silver doubles to $20 an ounce, no rational consumer will use a $10 silver certificate to buy $10 worth of goods at retail, but will instead redeem or exchange it for an ounce of silver with a market price of $20.”

You get that result by making the certificates convertible into TWO things: silver and dollars. The same thing happened with silver coins after 1964.

If certificates were convertible only into silver, they would buy twice as many dollars. If they were convertible only into dollars, they would buy half as much silver. Even if money is convertible into only one thing, there is a risk of a bank run when the assets backing the money lose value. It is the maintenance of physical convertibility that is the problem–not backing.

fundamentalist and newson:

Regarding the Fed compensating its losses with gains, I don’t think that happens. This may be a bit long, but I think you’ll find it interesting. First of all, let’s qualify this by saying that when the Fed loses, for the most part it is still making a gain. But its gain is less than the gain it would have made if it had bought at the market clearing price.

To simplify this argument, let’s say that the Fed buys homogenous units of government bonds with a set duration. In reality it buys bills and bonds, 3 month all the way to 30 year, MBS, agency debt, etc and there are all sorts of interest rates for these securities. But let’s make it easy on ourselves.

Our homogenous government bonds yield 3%. They trade for $100. The Fed is keeping the federal funds rate at 2%. At 2%, they are paying approx $101 for bonds ie they are overpaying.

The Fed decides to raise the ff rate to 2.5%, government bonds still yielding 3%. How does it do this? It accepts less bonds that are submitted to it for purchase, letting their price drop to $100.50 or so from $101. More dealers return to the ff market, and with that market less slack than before, rates rise. The Fed is still overpaying though. Fundamentalist – note that in this example the Fed is tightening (from 2% to 2.5%) but is still overpaying for bonds. ($100.50 vs 100) In other words the Fed can raise interest rates, doing the opposite of what they’re doing now; and lose money on the bonds relative to the market.

Now the Fed decides to raise the ff rate to 3%. It offers to pay 100 for government bonds, ie. it offers the market rate. Primary dealers are indifferent to this price as they can sell in the open market at 100, or they can sell to the Fed. If the Fed offers 100.0001 it can still attract bids and influence the ff rate to fall to 2.99999%, which is close enough.

Now the Fed wants to raise the ff rate to 3.5%, above the bond rate of 3% and $100. It offers to pay $99 for bonds worth $100. This is the situation both of you brought up, the possibility that the Fed might make money on its bonds by paying less than the market, ie. it might make outsized profits relative to what market participants might make. But if the Fed offers $99, the primary dealers will just laugh at them. Why sell for $99 what can be sold for $100 in the market? The Fed will get NO offers, and the ff rate will stay at 3%. This means that the Fed cannot possibly make above market rates on its purchases of bonds. The market will not let them.

How does the Fed force the ff rate up to 3.5%, above the market rate for bonds of 3%? It must start to sell bonds, ie. conduct REVERSE repos. The Fed offers to sell its bonds at $99 and 3.5% while the market price is $100. The primary dealers will jump at this opportunity for free profit, withdrawing money from the ff market to purchase $99 bonds. The ff market is now much tighter and the ff lending rate will rise to 3.5%. But as before, the Fed loses money by transacting at prices below market. The only difference is that now it is selling rather than buying.

So my point is that no matter if the Fed is raising rates or dropping them, no matter if the ff rate is above the market rate or below, and no matter if it is conducting repos or reverse repos, each trade it makes is below the market, ie. it is losing relative to all other market participants.

Another interesting observation is this. To keep the ff rate BELOW the bond rate (or the natural rate, that determined by time preference) the Fed engages in open market purchases, ie repos. When the ff rate is ABOVE the bond rate, it must keep it locked there with open market sales, or reverse repos. As http://mises.org/daily/2676 says, “The Fed has only engaged in 16 reverse repos since late 2000, versus 1247 repurchases”. What this means is that not only does the Fed always overpay, but since late 2000 (as far as the data goes) is has consistently kept its market rate below the bond, free market, or natural rate. Interesting, eh?

“But silver certificates (or a bank account that records deposits and withdrawals of silver certificates) are the only independent existing embodiment of dollars. They ARE dollars.”

Compare this to what you said earlier:

“If the market price of silver doubles to $20 an ounce, no rational consumer will use a $10 silver certificate to buy $10 worth of goods at retail, but will instead redeem or exchange it for an ounce of silver with a market price of $20.”

And of course there’s this:

“Let’s say we have a $10 silver certificate which can be redeemed for one ounce of silver or used to purchase $10 in goods at retail.”

This is quite a tangled web.

mike sproul says:
“We have two choices: (1) Issue paper money that is physically inconvertible. The bank’s loss of assets over time–resulting mostly from the cost of issuing paper money–will cause inflation, but no bank run. (2) Issue physically convertible paper money. The bank’s loss of assets will show up as a bank run.”

and here we have total agreement. i’m certainly in favour of #2. #1 has the problem of transforming localized and circumscribed risk (geography, time-frame etc) into long-term systemic risk. periodic bank-runs and collapses make the public very wary of banks, and less likely to entrust their entire capital thereto. the boom-bust cycle is likely to be much shorter under #2, malinvestments purged at more regular intervals. inflation neutralized over the course of the boom-bust cycle.

besides, “bank-run” scenario allows the possibility of 100% reserve banks to arise and thrive even during bank-run episodes.
not all banks collapsed during depressions, and it’s likely the survivors would enjoy a competitive advantage in terms of perceived deposit security.

i think the theft issue is a red herring, an entirely insurable contingency. a fraud/theft would have to first consume shareholders equityentirely before insolvency became a possibility. ratings agencies would fill information gaps with respect to these risks.

short of the law requring that banks be 100% reserved, i believe “bank-run” scenario to be much more benign (though politically fraught), than the institutionalized inflation scenario (our lot), with its masked and lagged downside.

to jp:
point taken. this perfectly illustrates what i said to mike sproul, the institutionalized inflation model has a complexity that the old bank-run model didn’t. the fraud on depositors is still there, just more opaque.

and yes, the number of repos vs. reverse-repos is less cryptic that the other points, and very revealing. thanks for the leg-work.

jp, Interesting analysis, and I think you’re right to some degree. In order to force money into the economy below the existing market rate, the Feds do have to lose money on its sale of notes/bonds. But the whole purpose of losing money is to pump huge quantities of dollars into the economy. How can you tell whether the inflation that follows is a result of loss of value in the bonds or the quantity of dollars pumped into the economy? In statistics, they call those “confounded effects.” They’re confounded because they’re correlated with each other as well as with the result–rising prices. How is it possible to tell which affect produces the result? Only by means of reason and observation of what actually happens.

No objective, automatic mechanism causes dollars to lose value. Valuation is always and at all times subjective. Dollars don’t lose value simply because the Feds buy bonds at a loss. Participants in the marketplace must decide that dollars are worth less. Do business people follow Fed sales and say to each other “The Fed is overpaying for bonds again. Let’s raise prices.” I don’t think so. What happens today is what has always happened since the days when the Spanish brought back boat loads of gold from the Americas. Merchants experience an increase in sales from the new dollars entering their stores and raise prices in order to keep limited inventory from disappearing too quickly and to profit from the increased demand.

But I have to admit that if the RBD means nothing more than that the Feds should always charge market rates for their notes/bonds, I would have to agree completely. The RBD can attribute the rise in prices to the man in the moon and I wouldn’t care, because the Feds couldn’t pump money into the economy without artificially lowering interest rates, which in your terms means losing money on the transaction. Austrians have always agreed on that. With respect to the Fed, the RBD would be a great improvement on how the bank operates.

But RBD means more than that. It defends fractional reserve banking, which is the second method by which the money supply expands and contracts. This expansion and contraction of the money supply causes booms and busts, destroys vast amounts of wealth, and persuades the people to accept greater state control of their lives and economy. It impoverishes the lower and middle classes at the expense of the wealthy and increases inequality of incomes and wealth.

I have no problem with banks offering loans on collateral or without collateral, but the money loaned should come from the savings of businesses and consumers. In that way, savings regulates how much can be loaned and maitains a fairly constant money supply. But the RBD insists that restricting loans to savings in unecessary. Banks should be able to print paper and loan it out at will. And to some degree, that would be fine if the government didn’t force people to accept dollars as payment. In other words, if we returned to a state of free banking where each bank issued its own notes and people were free to accept, reject or discount them, then the RBD might work just fine.

But you know, we’ve been there and done that. for most of the 19th century we did that. The result was regular booms and busts, business cycles, every ten years on average with all of the evil results I described above. In the end, the people begged for protection from banks and the government gave us the super bank. So proponents of the RBD are doing nothing but asking us to repeat the 19th century because it was so much fun.

to fundamentalist:
i’m not sure that the nineteenth century is that bad, when you compare it to the twentienth. sure, the road was rocky, but the foundation for economic success was sound, and the inflation cycle’s duration was less (so less damaging).
last century we swapped the periodic, small runs for what will be the twenty-first century international monetary collapse.

i gather you, too, are a supporter of 100% reserve banks?

Michael A. Clem,

“Don, let’s not make this harder than it is. In your example, you say that $10 = 1 ounce of silver. If the value of silver increases, $10 will STILL equal 1 ounce of silver, but that $10 will now buy more goods. If the money is backed by silver, then the value of the money is equal to the value of silver. It’s impossible in your scenario for 1 ounce of silver to go up to $20 unless the banks specifically change that ratio. Silver IS the money, and the certificates simply represent a quantity of silver.”

No, money is the medium of exchange. It is the certificates that have a number of dollars embossed on them. The retail goods are priced in dollars, and only dollars are accepted for them. Silver COULD be money, but only if a dollar is a measure of weight for silver.

Money has value if and only if people and stores almost universally accept it in exchange for goods and services.

If you are about to hand a cashier a piece of paper embossed with ‘$10′ for bread and candy and another customer comes along and offers you silver with a market price of $9, this new backing of the paper money increases its value not one iota.

Regards, Don

Or, for example, look at the current price of gold. It’s over $1000 per ounce. If the dollar were still backed by gold, then the value of the dollar would be tied to the value of gold. The “value” of gold would always be equal to the dollar ratio that had been established ($32/ounce wasn’t it?), even if the value of gold increased in relation to other goods. Under such a gold standard, it would have been impossible for gold to reach $1000/ounce, no matter how much the value of gold increased, because the currency would be pegged to the gold. Gold can only increase to $1000/ounce because our currency is not backed by or pegged to gold, thus allowing us to value gold with a monetary price, instead of by how much goods gold can buy.

to mike sproul:
here’s a snippet that jp highlighted in another mises blog-post. alan greenspan in a speech to the university of leuven, belgium:

“That all of these claims on government are readily accepted reflects the fact that a government cannot become insolvent with respect to obligations in its own currency. A fiat money system, like the one we have today, can produce such claims without limit. To be sure, if a central bank produces too many, inflation will inexorably rise as will interest rates, and economic activity will inevitably be constrained by the misallocation of resources induced by inflation.”

perhaps it was just a lapsus, or perhaps the whole fiat-money/unconvertible money polemic is just semantics, as i’d thought?
in fact, it’s an amazingly frank admission from one whose style was deliberately gobbledegook.

to jp:
regarding the constant erosion of fed assets in its open market operations, there are some things fathom with what you’re saying. take a look at this, for example:

“In going back to 1946 the market has lead the Fed at every major turn in interest rates. Yes, the market leads and the Fed follows. The latest example of this occurred during the 2000 to 2004 timeframe. In November 2000 the 3-month T-Bill was at 6.22% and the Discount Rate was sitting at 7.50%. By January 2001 the T-Bill rate had fallen to 5.70%, which widened the spread between the Discount Rate and the T-Bill rate from 1.28% to 1.80%. It was at that time that the Fed began cutting the Discount Rate and they continued cutting the Discount Rate as they followed the rates lower as was being set by the market. The 3-month T-Bill rate finally hit bottom in June 2003 at .82%.”

this is the link for a more extensive treatment –
http://financialsense.com/Market/wood/2007/0907.html

why does it follow that the fed loses on its on-market activity? i mean it’s trailing the market, not leading. also, from a cursory glance at the fed’s site, they impose a “haircut” on those using the repo facility.

“How can you tell whether the inflation that follows is a result of loss of value in the bonds or the quantity of dollars pumped into the economy?”

This question has been examined by Thomas Cunningham, Bruce Smith, Bomberger and Makinen, and a few others, all of whom are cited in my paper “There’s No Such Thing as Fiat Money”, at http://www.csun.edu/~hceco008/realbills.htm. Cunningham, for example, concludes that his results provide “clear support for the real bills doctrine”.

“But RBD means more than that. It defends fractional reserve banking, which is the second method by which the money supply expands and contracts. This expansion and contraction of the money supply causes booms and busts,”

If I pay for my groceries with my own paper IOU, and if the grocer uses that IOU to buy supplies, and if that IOU circulates a while before I finally pay it, then that IOU is money, backed by fractional reserves. Exactly what about that process should be illegal? One of the biggest complaints of the nineteenth century was that the currency was “inelastic”–that it failed to grow and shrink with the needs of business. That’s why the Fed’s first duty, as stated in its charter, is to provide an elastic currency. The quantity theory held sway in the nineteenth century, just as now, and quantity theorists always favored “tight money”. When they prevailed, the money supply was restricted and recessions followed. If fractional reserve banking is allowed to work, the money supply can grow and shrink to ACCOMMODATE booms and bust, without causing them.

“mandatory 100% reserve banking is defensible even from a libertarian perspective along rothbardian lines”

I repeat my question about my circulating IOU.

“A fiat money system, like the one we have today,”

No argument there. Alan Greenspan, like virtually every other reputable economist on earth, is a quantity theorist who believes that there is such a thing as fiat money. A few hundred years ago, those same people would have sworn the sun orbits the earth.

Newson:

Your 2000-2001 example is a good point. Let me summarize to make sure we’re on the same page. If the Fed was keeping the federal funds rate (not the discount rate, that’s a different thing) at 6.5% in 2000 (not 7.5% as per the article), significantly above the t-bill rate of 6.22% (indeed, it fell as low as 5.3%), then it would have been buying those t-bills at a discount to the market, after all buying at a higher yield implies a lower price. That means that in 2000-2001 the Fed should have been making profits in excess of what it would have made had it bought at market prices. Furthermore, the Fed was not keeping the ff rate above the t-bill rate via open market sales since the Fed has hardly engaged in any of these since 2000. Therefore, the Fed had to have been keeping the ff rate above the t-bill rate using open market purchases and presumably making money. End summary.

In my earlier example I simplified the analysis into a comparison of the ff rate to the rate on homogenous government bonds. The reality is that the Fed buys all sorts of securities with differing maturities and from different issuers. Unpack these and the picture broadens.

In your example t-bills were trading at yields below the ff rate. Given this criteria, when the Fed announces an open market operation, no primary dealer would willingly offer to sell t-bills as it would result in a loss to them. But there might be other bonds like the 5, 10, 20 or 30 year bonds that are still trading above the Fed funds rate, even though the 3 month isn’t (long term yields are often slower to react than near term ones). The primary dealers can thus keep their 3 month bills and sell these longer maturity assets instead to the Fed, continuing to earn outsized profits at the Fed’s behest.

In other words, dealers will substitute some assets for others depending on what relative interest rates are. The Fed rate may indeed be trailing market rates down, but there will often be some bond classes that are still above the ff rate.

In 2001, the ff rate was actually set above all maturities of government bonds, 3 month all the way to 30 year. But the Fed also buys agency-guaranteed MBS. This is a new power and was only allowed in 1999 (see http://mises.org/daily/2676 for the story). Back in 2000-2001, the average Fannie Mae issued MBS was yielding around 7.5%. With the ff rate at 6.5%, the only assets the primary dealers could profitably sell to the Fed would have been MBS. By overpaying for MBS, the Fed lured primary dealers away from borrowing in the ff market to buying from it, thereby keeping the ff market tight and at the 6.5% target.

What do the numbers show? There was a big jump in MBS as a % of total open market ops, from 0% in 1999 to 20% in 2000, confirming the possibility that these operations were relatively more popular. Even though MBS sales to the Fed grew, there were still transactions in government bonds too. These would have been losing propositions for the primary dealers, and gains for the Fed. I have no idea why the dealers would have taken these trades and that does weaken my case. There could be some form of moral suasion by the Fed on primary dealers to take losing trades, but I have no evidence for this. It is also difficult to properly analyze Fed open market ops as the maturities of bonds bought and sold are not included, nor does the data series go back beyond 2000.

As for haircuts, the Fed does claim to apply them to open market ops, but I don’t think they make the actual rates public. The way I see it the haircut the Fed applies cannot bring the purchase price of bonds lower than the market rate for those bonds. If the haircut was onerous, the primary dealer would simply sell those same bonds in the market where haircuts are not applied, then transfer the cash into their reserves to meet requirements.

In the end, if the Fed doesn’t dangle some sort of carrot in front of the primary dealers, then it will lose the ability to control the fed funds rate. While it might be able to earn abnormal profits for a few months, at some point it’ll lose its ability to do so. I can’t see its abnormal profits ever outweighing its losses. While the 2000-2001 period does show evidence of some Fed profits amongst MBS losses, from 2002-2006 the ff rate was WAY below almost every market rate, resulting in what must have been pretty big losses.

“…debtor’s desire to avoid the stigma/costs of bankruptcy.”
low caffeine alert.

Because the laws of supply and demand, which apply perfectly well to commodities, don’t do so well when applied to pieces of paper that are claims to those commodities…This means that the fed can maintain financial convertibility at a rate below what its assets can support, and it can do so by printing too much money–which makes it look a lot like what the quantity theory says.
Not making sense. Why wouldn’t supply and demand apply to claims on those commodities? Or rather, since supply and demand apply to the commodities, why wouldn’t the claims be treated the same as the commodities that they represent?
And how can they print too much money, as long as they’re being backed by assets? Only if they engage in fractional reserve banking or some other fraudulent means, as far as I can see, but never with 100% backing.

Mike S,

DL: No, I said ‘Assume to start that the market price of silver is $10 an ounce.’
No, a $10 silver certificate will always yield a pre-determined weight of silver, assuming the redemption promise is still in effect. This may be less or greater than $10, where dollars are the monetary unit of account. It will always (?) buy goods and services whose prices sum to $10.

MS: Okay, I’m really not understanding you, Don. The only way that the monetary price of silver can vary is if it is not the backing asset of the currency, i.e., you are NOT using silver certificates, and banks are not holding silver. If you are using silver certificates, the certificates are necessarily defined as a certain weight of silver, and thus vary in lockstep with the value of silver.
I suppose there could be two different currencies on the market, silver certificates and something else. Then silver could have a monetary value that varies in the non-silver certificate currency. But that’s a much more complicated scenario, and I didn’t get the impression that you were setting it up that way.

A silver certificate has two properties which are only sometimes coincidently in alignment.

First it is a claim which can be redeemed for a specified weight of the commodity silver.

Secondly, it is embossed with a given amount of dollars, which determines how valuable it is as money, a medium of exchange, universally accepted.

At any particular point in time, the certificate will be more valuable used as money, or the reverse. It can’t be both at once unless they happen to be equal in value.

At present, if I pull a $10 bill out of my wallet, I can buy a total of $10 of priced goods and services with it, including silver at its market price.

However, it is not a claim to any particular weight of silver. If it were, the specified weight of silver claimable might be very large, or very small. If the claimable weight of silver is sufficiently large, the highest value use of the $10 bill will be as a silver claim. If the claimable weight of silver is sufficiently small, then the highest value use of the $10 bill will be as a medium of exchange, money.

The exact same mechanism is true of actual silver coins as well. A few years ago Canadian 1 oz Silver Mapleleaf coins, marked with a face value of $5 CDN had less than $5 CDN worth of silver content. At that time, they were actual Canadian money, and $5 CDN served as a price floor no matter how low the market price of silver might go. As the market price of silver rose above $5 CDN (ignoring dealer markups) the coins tracked the price of silver due to their increased melt value, and no rational holder of a coin would spend it like it was just worth $5 CDN as money.

Regards, Don

to jp:
ok, tim wood has used the discount rate by way of example, but as he says, the same applies for ffr. for week ending november 24, 2000 the ffr was 6.5%, the 3mth bills were trading @ 6.18%. come week ending jan 26, 2001 ffr is 6% and bills trade @ %5.1. now fast forward to the lowest point for yields in this cycle – june 2003, ffr stands @ 1.25%, but the 3month bills are yielding between 20-40 basis points lower. so the fed has been out of the market over this whole cycle, as far as this part of the yield curve is concerned. now this i find confusing, as i had always believed that the fed’s interventions were mainly at the short end of the yield curve. and yet over this period, to deal with the fed would have been a losing proposition. any transactions with the fed would have strengthened the feds balance sheet at the time of the operation. or am i going crazy?

p.s. i don’t know where tim wood gets his figure of 5.7% for 3month bills in january 2001. all the figures i quoted are from the ny fed’s site.

Newson:

I don’t think the central bank should have a monopoly on the issue of paper money, and given free banking, I don’t think the central bank serves any purpose that couldn’t be handled by the private sector. Standardizing the look of paper money comes to mind as an advantage of central banks, but I doubt that private banks would have much trouble with that. Given the ubiquity of central banking, I can’t quite rule out the possibility that they serve some unknown useful purpose, so I don’t favor abolition, but if private banking were freed of constraints, I’d expect that people would soon have no use for a central bank.

Mike:

“Well, the fed could buy all the world’s farmland, but the land would probably still be there, still being farmed, so the fed could buy all the farmland without affecting its value. But this is beside the point of what happens to the value of the dollar when the fed issues more dollars for miscellaneous assets that it does not actually consume.”

Please just stick to my specific qustion: How can you think that an additional demand for something, ceteris paribus, will not result in an increase in prices. If the fed buys assets it exercises an upward pressure on those assets’ prices. It is simple economics 101. On another note: the federal reserve does ultimately consume the assets. It earns a profit via the interest payments on the assets. The profits are disbursed to its shareholders and used for consumption on the open market.

“I could buy 10,000 loaves. I could buy a car or a house. If the treasury issued them in small denominations, then I could buy single loaves. The point is that when I exchange my T-bill for paper dollars from the fed, my purchasing power is substantially unchanged, so there is no upward pressure on prices.”

Same here, my simple question again: It seems like you agreed that you cannot purchase a loaf of bread with a treasury bill? Yes or No? But for your own arguments sake I will address your car example: It is wrong. Plain and simple. Your car dealer is not going to accept treasury bills as means of payment. If you believe otherwise, feel free to prove it.

I have written a post as to what is and what is not to be included in the money supply http://nimamahdjour.blogspot.com/2008/03/money-supply-watch.html

I have not read all of your posts regarding free banking, but I do agree that a fractional reserve system could be maintaned with very little inflation and price changes. But this is not contingent upon the money supply rising in step with the banks assets. It is contingent upon the money supply rising in step with its demand. And it is completely out of the question in an unfree, compulsory legal tender money system.