A clean-slate accounting of the dollar (part 1)
There seems to be a bit of a commotion in the financial markets. As one mugwump put it: “this would be extremely interesting from an analytic perspective, if it wasn’t happening to us.”
And the yellow dog has barked. Woof! The Au/USD ratio or “gold price” rose about 10% on Wednesday. This may not sound like a lot, but it’s basically unprecedented. A Moldenstein moment? Possibly, but I wouldn’t write off the global financial system just quite yet.
Looking at the chart, what seems to have happened is that a little before 10 AM, someone chose to either acquire a fat slice of the yellow dog, or relieve themselves of a negative slice. I suspect the latter. “Actually, Brooke, we used to have our very own place in the Hamptons. But then Daddy decided a massive financial crisis was the perfect time to short precious metals.”
If the details of this matter interest you, see my comments Monday at Macro Man’s. The subject was the decoupling of gold and oil, which moved in sharply opposite directions Monday. Hopefully causing someone to lose a large amount of money. I would like to think that the events of the last year—or at least the last day—convince the wizards of Connecticut, or at least the wizards’ little black boxes, that the marginal market for gold is dominated by its role as a natural currency, and that it does not form reliable correlations with industrial raw materials (“commodities”) or official currencies (the hidden denominator of “the dollar,” as in “the dollar rose today”—meaning, basically, USD/EUR).
(Not that I expect raw material/USD ratios to crater. The basic driver of rising material/USD ratios is the US trade deficit, which ships large numbers of dollars to large numbers of people with a large marginal propensity to demand raw materials and products thereof. E.g., China. Since the basic cause of the trade deficit is underpriced currencies in emerging markets, and since this underpricing is not an accident but a policy, there is no reason to expect this trend to reverse. Of course speculators can overshoot it, but they can also undershoot it.)
You can also read this thread at Brad Setser’s, in which I explain my modified Austrian theory of banking, and the rather extreme crisis medicine it implies, to skeptical but friendly bankers. Here at UR we’re never afraid to go head to head with the heavy hitters.
But all of this is written for those with some pre-existing knowledge of the system. The great thing about being a skeptical generalist is that you trust your own ability to think about X, instead of trusting those with actual knowledge and experience in X. I am certainly no opponent of knowledge or experience, but it should be plain to everyone at this point that our financial system is, in at least one mysterious and incomprehensible way, broken.
So those who know it, know the broken thing. Worse, they believe that the dollar system is exactly what it purports itself to be—money is money, bonds are bonds, banks are banks, etc. And they have built a large superstructure of models on these understandings. Said models do not appear to be in perfect working order.
When we analyze the dollar with no assumptions about economics, finance, or accounting, we may make our own mistakes, but they will be our own mistakes. Think of it as rather like analyzing a Soviet nickel company in 1991. Does the company even exist? Does it even produce nickel? Does it own any mineral rights of any sort? If we buy it, do we really own it? Etc.
So think of our clean-slate dollar analysis as a sort of forensic accounting, on a very special Soviet nickel company: USG, also known as “America.”
First, let’s start by looking at a dollar. You probably have one in your wallet. This exercise may seem pointless. And to some extent it is. But it reminds us of the first rule of accounting: always trust your own eyes.
On the back of my dollar, there are some trippy engravings with Latin text. I don’t know Latin, but the context seems purely decorative and contractually void. Hopefully this is actually so. There is also the phrase “In God We Trust,” which sounds vaguely like a financial obligation. But we know not what liabilities are being ascribed to God, whether He has actually agreed to any such contract, or what will happen if He defaults on it. So we’ll call this one void as well.
On the front, more decoration, but also what seems like actual content. We learn that our dollar is a “Federal Reserve Note,” issued by “The United States of America.” The latter is our Soviet nickel mine—USG. The “Federal Reserve” is apparently some unit or subsidiary thereof. Perhaps it has some relationship to the “Federal Reserve Bank of San Francisco.” Also appearing is the “Department of the Treasury,” apparently some other unit, employing a “Treasurer of the United States” and a “Secretary of the Treasury.” Oddly, these are not the same person. Perhaps the latter is the former’s admin. Their signatures are reproduced—context, unclear. Are the Treasurer and her secretary personally liable for any obligation? If USG tanks, must they step in and make me whole? We’ll take a wild stab and say no.
We do not understand the internal structure of USG. We don’t want to understand it. We will treat it as a single entity with a single balance sheet. Presumably if we decide to do the deal, we’ll bring in our own people and reorg everything. For now, if USG is lending to itself, or whatever, these transactions will cancel out. Simplifying everyone’s life. Life is short. Accounting is boring. Or at least, it should be boring. If it’s not, something is probably up.
There is also what looks like a serial number. Mine is “L68082696L.” This suggests that there is a list of serial numbers somewhere. Probably on a crumbling, makhorka-stained notebook filled out by hand in Cyrillic. But still, this is excellent. It suggests that we may be able to know how many dollars are outstanding. As we’ll see, this is too optimistic, but one can hope.
It also suggests that the Cyrillic notebook might record that I, Mencius Moldbug, am the owner of L68082696L. But this can’t be so—I know that no one knows that I have this note. This is useful information; it means the dollar is some sort of bearer security.
Finally, there is a fascinating phrase. It reads: “this note is legal tender for all debts, public and private.” When we investigate this, we find it refers to a rule in the operating handbook for USG’s internal security forces—31 USC 5103:
United States coins and currency (including Federal reserve notes and circulating notes of Federal reserve banks and national banks) are legal tender for all debts, public charges, taxes, and dues. Foreign gold or silver coins are not legal tender for debts.
Does this actually mean anything? Or is it purely decorative? As we’ll see, it is decorative. Or more properly, historical. It doesn’t mean anything now, but it once did.
Today, if I sign a contract promising to deliver a hundred dollars, I need to deliver a hundred dollars. If I sign a contract promising to deliver a barrel of oil, or a tenth of an ounce of gold, or five shares of Cisco, I need to deliver a barrel of oil, or a tenth of an ounce of gold, or five shares of Cisco. All of this would be true with or without “31 USC 5103.”
So we have learned the following from our examination: the “dollar” is some sort of bearer security, issued by USG, attractive and portable, conveying no rights whatsoever. Well, it was worth a try.
But we know something else, which is not written on the note but is implicit in its design. We know that every dollar is equivalent. Each dollar is a member of a homogeneous class of goods, each of which confers the same rights as every other. Moreover, bills of higher denominations obey the same rule. A five-dollar bill will always be worth five dollars. It will never trade up to $5.23, or down to $4.65.
This minimal amount of information enables us to classify the dollar as a financial instrument. The dollar is equity.
Compared to your average equity issue—say, a share of Cisco—the dollar is a highly inferior class of stock. It never has paid any dividends and probably never will. It conveys no voting rights. We have no idea how many shares are outstanding. And we’re not at all sure what you’d own, if you owned all of them—but it is definitely not all of USG.
But this is par for the course for your Soviet nickel company. The forensic accountant is trying to make some kind of structured sense out of half a century of Communist nonsense. Is this a problem? It is. Does this mean you shouldn’t buy the company, or its shares? Well, nickel is nickel. Some of the world’s richest men got that way by buying Soviet nickel companies.
Any financial instrument is one of three things: a deed of ownership of some good (a title), a liability to fulfill some obligation, possibly contingent (a debt or option), or none of the above (equity). The dollar is equity because (a) it is definitely not a title or a debt, and (b) every dollar is created equal; whatever benefits USG may choose to shower on the holder of L68082696L, it will provide to the holder of any other dollar. It’s not much, but it’s a start.
Perhaps you think of the dollar as “money” or “currency,” and you are very confused by all this. “Money” and “currency” are nice words, but they have no precise accounting definition. They just refer to a good or security commonly held for the purpose of savings and/or exchange. Historically, we can identify four classes of currency:
(1) Direct goods, such as coins of a standardized weight of precious metal. (2) Titles or warehouse receipts to (1). (3) Obligations to pay (1) or (2), or redeemable currency. (4) Mere equity, or fiat currency.
These are (excluding the common 4-to-1 transition) in order of historical evolution. Explaining the evolution is not of direct assistance in analyzing the dollar, but it helps us get our bearings—and it defines terms which will be useful later on.
Class 1 currency (standardized metal) evolves into class 2 currency (titles) because titles are more portable, secure and convenient. Digital gold is a modern class 2 currency.
Class 2 currency (titles) evolves into class 3 currency (redeemable notes), because the change is generally profitable for the currency issuer, and the marks are too dumb to know the difference. A title or warehouse receipt is a title because the issuer holds goods that match it; otherwise, he is a thief. A redeemable note is a mere debt, and does not default until redemption fails. And even then, the issuer is only bankrupt, not in jail.
Suppose I have 100 ounces of gold, and issue 100 titles against it, each title stating that the bearer owns one ounce of gold. This makes me a respectable issuer of class 2 currency.
Suppose I have 100 ounces of gold, and issue 200 redeemable notes against it, each note stating that I will issue the bearer one ounce of gold on demand. This makes me a scoundrel.
However, it also makes me wealthier than I was before, at least until more than 100 bearers show up to claim their gold at the same time. Will my notes trade at par? I.e.: will people accept them as equivalent to the class 2 titles? Well, every time someone starts to get suspicious and tries to redeem one, it works. So I don’t see why they shouldn’t. We have just reinvented the wildcat bank—a staple of early American finance.
There is a cure for wildcat banking: those who accept class 3 notes should ensure that the issuer is solvent. A financial institution is solvent if and only if the sum of all the payments it is obligated to make equals or exceeds the total price of all the assets it holds. So our Scoundrel Bank above is not solvent, because it is obligated to pay 200 ounces and it only has 100 ounces.
One way to see a redeemable note is to see it as a very short-term loan from the noteholder to the bank, which is automatically renewed or “rolled over” when the noteholder does not redeem. If the term of the loan is an hour, a minute or even a second, the effect is redeemability. And note that when making a loan, what you want to know is whether the loan will be paid back. And collectively, what all loanholders want to know is that they all can be paid back. First come, first served, is not solvency.
Scoundrel Bank can redeem for some of its noteholders. But not all of them. Therefore, all unfortunate holders of its notes can agree on a fair—or “equitable”—bankruptcy restructuring: every holder of a Scoundrel ounce note should receive half an ounce of gold. As for the scoundrel himself, trees abound, and perhaps the noteholders can pitch in for a rope.
But there is a tricky intermediate case—call it Questionable Bank—between Respectable Warehouse and Scoundrel Bank. Respectable Warehouse issued 100 one-ounce titles. It has 100 ounces. Verifying the quality of the titles is as easy as verifying these facts. Scoundrel Bank issued 200 one-ounce notes. It does not have 200 ounces. Verifying its insolvency is just as easy.
Questionable Bank also issued 200 ounce notes. It also has only 100 ounces. But it also has 100 old pianos. Each piano, it asserts, is worth an ounce of gold. “Easily. Easily. No sweat, man. These are fine pianos. Here, play this one. Hear that sound? That’s a sweet tone. Check out the action on the keys. Totally smooth. You could move this piano for an ounce fifty, no problem.”)
Suddenly our noteholder, or at least the accountant he hires, is required to be a piano appraiser. Should you trade a Respectable title to one ounce, for a Questionable note paying one ounce? It depends on the quality of the pianos. Obviously, every piano is different. You can’t just play the one up front in Questionable’s office. You need to go into the back room and tinkle away.
Moreover, even if each piano could be sold on the piano market for an ounce or more, it is hard to know that all the pianos could be sold at once. Since price is set by supply and demand, the appearance of a large splodge of pianos, even fine pianos, on the market all at once, is liable to depress the piano price. And “at once,” let’s not forget, is the term of the Questionable notes.
In real life, of course, the good in question is typically not pianos but loans. Usually long-term loans. Pricing a piano is difficult, but it is nothing on pricing a loan. And the result of dropping a glut of loans on the market all at once is even more astonishing and disastrous.
If you are interested in the details of this issue, see the discussion at bsetser’s. But in this case Questionable has an easy solution for its problem. Like most currency issuers, it is sovereign. It is not just a bank. It is also the government. One, nobody can go into its back room and tinkle its pianos. And two, if people don’t want to accept its one-ounce notes as equal to one ounce, it can make them.
This provides an interesting explanation for the mysterious “legal tender” phrase on the front of our dollar. It is easy for a sovereign currency issuer to transition its units from a class 1 or class 2 currency, to a class 3. The class 3 currency is still defined as the same weight of metal. But it is no longer backed by enough metal to redeem the entire currency issue (otherwise, it might as well be class 2, like the old Bank of Amsterdam). Besides metal, the issuer holds pianos, loans, Honus Wagner baseball cards, etc., all of whose total market price exceeds the sum of the notes. Or at least allegedly exceeds them.
It is highly desirable for the class 3 currency to trade “at par,” e.g., be exchanged one-to-one for the metal or the title. And for a sovereign issuer, this is no problem. If the subjects arrogantly persist in scorning the King’s honor by discounting his perfectly good notes, the King’s sword will discount their necks. Even the flagrantly insolvent Scoundrel Bank will do just fine as a sovereign issuer. If Jews and speculators try to stop the King’s loyal subjects from paying their debts with the King’s good one-ounce notes, which they maliciously insist are worth only half an ounce, their heads are forfeit.
And it is easy to see that the King’s notes are worth one ounce, because anyone can take them to the Castle and exchange them for an ounce of pure gold. No tricks or shenanigans. Of course, speculators and Jews may assert that King Scoundrel has used his sword to perform what is essentially an act of alchemy, an art of interest to all kings past, present and future. But swords speak louder than words.
However, when King Scoundrel pushes his alchemy too far, the seams begin to show. Foreigners, for example, have a nasty tendency to redeem the notes and skip the country with the gold. This won’t do at all. And it can be repressed in various ways, but it leaks. The domestic price of Scoundrel notes will always be one ounce, but the foreign price might be a bit lower. There is a perennial incentive to redeem notes and smuggle gold. Bad.
This problem can be solved in a very simple way. Cut the Gordian knot. Suspend redemption. Even better, terminate it permanently. Just default on the loan. You’re the King, after all. Who’s going to mess with you? Certainly not the Jews and speculators.
The result is a class 4 currency—equity. Conversion of debt to equity is the normal procedure in bankruptcy. Scoundrel Bank still has plenty of assets, after all, just not enough to keep its pesky subjects from redeeming their notes and selling the gold to speculators and Jews. The class 4 currency is not, by any means, worthless. Holders of notes are treated equitably. In theory they may even collectively own all of Scoundrel Bank, although the King is generous enough to manage it for them.
Is it possible that our present dollar evolved through this process? Today we are practicing accounting, not history. We do not know how the dollar came to be what it is. All we know is that it’s a class 4 currency. This conclusion is produced solely by our own eyes, above.
To regularize this equity structure, to retrieve it from the realm of Soviet nickel and bring it back into the realm of sane accounting, we need to do two things. One: we need to decide how many shares there are. Two: we need to decide what you own, if you own all the shares. For example, in the case of Cisco, the answer to the first question is “5.9 billion,” the answer to the second is “Cisco.” These answers may be slightly blurry around the edges, but only slightly.
Let’s start by declining to answer the second question. Clearly, if you own all the dollars, you do not own USG, including the highly lucrative (and nickel-rich) continent of North America. While it might be great fun to elect a CEO of USG with one vote per dollar, it is not essential to a proper accounting of the instrument. We would also like to avoid restructuring USG as a whole, at least in this blog post.
For the moment, let’s say our goal is to construct and spin off a new entity, DollarCo, whose shareholders are current dollarholders on a 1:1 basis—each dollar will become one share in DollarCo. As for DollarCo’s assets, this is a political question. But just to get in the right ballpark, let’s start by throwing in USG’s gold reserves: 264 million ounces.
The first question is all we have left. How many dollars are there in the world? There is a nice official answer to this question. The figure is known as M0, the monetary base, and its current value is about 825 billion. Thus, DollarCo will have 825 billion shares.
One of the nicest things about modeling a dollar as a DollarCo share is that the process of equity dilution—creating new shares—is extremely well-understood. Whereas the process of creating new dollars is, in classic Soviet nickel-company style, shrouded in deep mystery. Everyone has a sense that there are a lot more dollars around than there used to be, but few are quite sure why.
It is immediately obvious that creating new DollarCo shares cannot add more gold to Fort Knox. But the process of dilution is still useful for various events in the corporate lifecycle. Dilution confiscates a pro rata percentage of each share from all shareholders, collects it all together, and assigns it to whatever party is granted the new shares—presumably in exchange for some consideration which benefits the old shareholders, making the confiscation worthwhile for all. For instance, if DollarCo is to acquire PesoCo, it can issue new shares in exchange for all the shares of PesoCo, according to whatever ratio the dealmakers work out.
What we don’t want to see happening, however, is covert dilution. The party secretary of our Soviet nickel company may well be running off shares on his laser printer, and handing them out to his vodka buddies. This is a no-no. And we are not sure what’s happening in this “Treasury” place—but the number is definitely increasing. For now, we would like to not only ascertain, but also lock, the number of dollars outstanding. PesoCo will have to wait for another day.
It’s important to recognize that DollarCo is a product of the accounting process. It does not exist yet. All we have is USG, which owns Fort Knox among many other things, and has issued some number of dollars. Before we can form DollarCo, converting dollars to DollarCo shares and transferring USG assets to make DollarCo worth something, we need to ascertain and lock the dollar count—first.
A dollar need not be a physical object. It can be an electronic entry, as many dollars are at present. However, we need to know exactly how many dollars are outstanding. A foolproof way to do this is to assign every dollar, electronic or paper, a consecutive serial number—or as we programmers put it, a sequence number. If there are n dollars, the first dollar is sequence number 0 (we can frame this one), and the last is (n—1). Rocket science this is not. And, most importantly, n is final—it cannot be increased.
But we already know n, right? Isn’t it 825 billion—M0?
Wrong. The problem is that this assignment, n = M0, simply does not make sense. It is not consistent with economic reality. Of course USG can enforce it, as it can enforce anything, but the result will be social and economic disaster. North America will become a burned-out Mad Max wasteland, patrolled by marauding gangs and packs of radioactive mutant wolves.
To think about this, let’s make sure we fully appreciate the ramifications of permanently setting n = M0 by imagining that each dollar is not just a piece of paper, but an alien artifact, roughly the size of a poker chip but made using alien technology which cannot be duplicated. These alien poker chips are a strange, glowing purple color that no Earth technology can imitate. There are 825 billion of them, and no aliens are around to extend the supply.
Everything else about our world, however, remains the same. Including the fact that USG has a national debt of roughly $10 trillion—not counting unfunded entitlements. Moreover, this debt is not even discounted. Quite the contrary: it is considered “risk-free.”
Question: how, exactly, in a world that contains only 825 billion dollars, can a debt of $10 trillion be risk-free?
Moreover, USG runs an annual trade deficit of $750 billion. Even if it started each January 1 with all 825 billion of these dollars in the country, which it most certainly didn’t, its subjects should be feeling pretty impoverished by Christmas. But no. They run the same trade deficit, year after year after year. Perhaps the dollars are being lent back to them—but why?
There can only be one answer: this $825 billion number is just plain wrong.
825 billion is the number of formal dollars outstanding. It is not the droid we are looking for, though. What we need, in order to set up DollarCo and properly account for the dollar, is the number of fully diluted dollars. Somewhere out there in Soviet nickel-company land, there are trillions and trillions of covert, informal, virtual, contingent, or otherwise mysterious dollars.
The goal of any clean-slate accounting must be to (a) describe these virtual dollars, (b) figure out who owns them, (c) understand the process that produces them, (d) shut this process off, and (e) map the virtual dollars to sensible financial instruments, ideally just regular dollars.
Come back next Thursday for part 2 of this exciting series, in which we’ll solve these problems, or at least pretend to. (If you want the solution to actually be implemented, I’m afraid you’ll have to call your Congressman.)