Wednesday, April 02, 2008

Commodity derivatives: the new currencies

Unable to otherwise explain today's high commodity prices, Frank Veneroso and others fear we are in a commodity bubble. Exhibit A is the recent vast increase in the use of commodity derivatives[*]:

Here's how Veneroso observes the unprecedented state of our commodity markets:
Though the six-fold increase in such positions over a few brief years is dramatic , it is the magnitude of these positions that is most alarming...It is hard to know what to make of this data. But it is noteworthy that several years ago, at the then prevailing lower commodity prices, the entire above ground stock of all commodity inventories was only in the hundreds of billions of dollars. Even if only a fraction of the increase in global commodity derivative aggregates in recent years corresponds to a net long position of investors or speculators, implying speculative and investment positions of a “few trillions of dollars,” it would appear that this increased demand for commodity derivative positions has overwhelmed what have been relatively small markets...

No wonder, then, that this cycle’s bull market in commodity prices has gone higher in inflation-adjusted terms [i.e. in terms of the CPI and PPI -- a poor near-term measure of inflation, see below] and for longer than in all prior uninterrupted half-decade cycles in the past.
I agree with Veneroso that the recent commodities boom has a strong causal link to the recent rise in the amount and variety of commodity derivatives. But I couldn't disagree with him more about the role of inflation. Prices, and especially wages, react very slowly to changes in money supply, so that the CPI and PPI are trailing indicators. The leading indicators are increases in money supply in excess to increased demand for money and the associated (if disproportionate) increases in commodity prices.

M3, an estimate of the supply of dollars (including a reconstructed estimate for recent years). M3 is increasing at the highest rate since the 1970s, as are to a lesser extent the supplies of euros and other major currencies, resulting in a commodity boom not seen since the 1970s. Source:

The recent vast increase in commodity derivatives, and the resulting commodity boom, is a largely rational response to the vast increase in the supply, relative to demand, for the dollar and to a lesser extent the euro and most other major currencies during most of the last decade. This despite the fact that commodities prices have in the last year risen (at about 40%/year) much faster than the growth of money supply (about 15%/yr). To see why, observe that one of the most useful but least talked about uses of derivatives is their ability to simulate the economic behavior (primarily risk and return) of one kind of asset based on contracts for another kind of asset or performed in another legal environment. The classic example is Eurodollars. But let's take a general scenario:

Alice wants to promise Bob to transfer asset X. But the risks associated with Bob holding asset X on his balance sheet, or Alice holding liability X on her balance sheet, until actual transfer of the asset, are too high, or the transaction costs of actually transfering X from Alice to Bob are too high, or both.

(1) Solution for risks of holding asset X on books: derivatives. Alice has both assets and liabilities on her books. These have individual and net risks just like an investment portfolio. If Alice has the same amount of assets and liabilities all denominated in the same currency, her portfolio is balanced. But if, for example, she imports from Europe (thus creating liabilities in euros) and sells in the U.S. (creating assets in dollars), her portfolio becomes strongly subject to risk of dollar devaluation against the euro. If a new asset or liability creates such unbalanced risk, it can be hedged with derivatives.

(2) Solution for transaction cost of transferring asset X from Alice to Bob: use derivatives to create a synthetic asset. Construct from derivatives and asset Y a synthetic asset that economically behaves like asset X. These separate derivatives and asset Y that form the synthetic all can be transfered with low transaction costs, thus forming an asset economically equivalent to asset X but that, unlike X, can be transferred from Alice to Bob at low cost.

The reason these uses of derivatives are seldom talked about is that the main cause of problem (1) is the instability of government currencies that it is legally necessary (for a variety of reasons) for contracts to be denominated in. The cause of problem (2) is usually a legal barrier to a particular transaction -- a tax, exchange control, risk caused by regulation, or risk of confiscation in Bob's jurisdiction are four examples. People who route around the law tend to do so quietly, explaining their acts by euphemism, even if as here the new route is perfectly legal and is made necessary by deep flaws in the law or the way the law is executed.

There are often, in other words, huge discrepencies between the contract terms required or incentivized by law and the most economically efficient contract terms. Partly this is due to political stupidity, and partly to the inevitable profound imperfection of any body of rules.

(3) A third function of derivatives is simply as a store of value: a way to diversify or hedge inflation risk in an investment portfolio that otherwise contains either (a) debt denominated in a potentially inflating currency, (b) equity in companies with assets consisting mainly of debts denominated in a potentially inflating currency, or (c) both.

In performing functions (1) and (2), derivatives usually substitute for or augment currencies as a medium of exchange. For (3) they substitute for or augment the monetary function of a store of value. Most derivatives, in other words, perform largely monetary functions. They augment or substitute for a flawed method of payment or a flawed store of value.

Gold prices in dollars (and, not pictured here, the dollar prices of many other commonly traded commodities) roughly follow changes in the global monetary base of dollars. Source:

In a climate where one or more commonly used currencies are inflating (and by "inflation" I simply mean rise in money supply without concurrent rise in demand for that currency, not the CPI or any particular and usually trailing measure of inflation), derivatives, if they can be analyzed and traded at low cost, are very much in demand. Furthermore, if all of the currencies practically available for contractual payment terms are inflating, these derivatives will come to be increasingly based, not on currency-denominated debt, but on commodities, especially commodities for which supply and demand are relatively inelastic in the short and medium terms.

Why is it rational, rather than reflecting a bubble, for the recent rise in commodity prices to outpace the recent rise in money supply? The rise in commodity prices reflects the increase not just in supply of dollars per unit of commodity, but greatly increased demand for a commodity as money, i.e. for the monetary functions it serves either directly or indirectly via derivatives, on top of the relatively steady demand for consumption. As I have argued elsewhere, this increase is broad-based: all commodities that can be traded with low transaction costs on modern markets, and that exhibit short- and medium-term inelasticity of supply and consumption demand, are useful for and are being used as reserves for the new derivative currencies. Where computerized analysis lowers mental transaction costs, as with modern derivatives, it pays to diversify one's books, portfolios, and hedges among multiple currencies.

Of course, the traditional use of commodities futures and derivatives by commodity producers and consumers to hedge the assets and liabilities of the commodity sales and purchases inherent in their business continues. But these don't account for the vast majority of commodity derivatives use in recent years.

Before modern computers, networks, and software engineering, the cost of such derivatives was usually prohibitive. In the last period of monetary inflation and the resulting commodity boom, the 1970s, derivatives were available to few and only the very simplest hedges or synthetic assets could be understood, and even then very imperfectly. Today derivatives, and their use in very complex hedges and synthetic assets, are much better understood by most of their users, thanks to computerized game tree analysis, and are available to many. They can be designed, simulated, and traded electronically with very low transaction costs. Computerized derivatives are smart contracts that lower mental transaction costs. So much so, that they've created a vast new world of contractual relationships completely impossible with the brain alone.

Speaking perhaps a bit metaphorically, we are witnessing the rise of new and privately issued fractional reserve currencies. They need not and effectively cannot legally be called "money" by their "issuers", nor can they effectively be used directly in most contracts for payments. But they can be used indirectly to hedge payment terms or investments denominated in flawed, that is in inflating or otherwise unstable, government currencies in which normal contracts and instruments are generally denominated. The results are synthetic "currencies" that, in their economic behavior, may be almost indistinguishable from a tradtional commodity-backed and privately issued fractional reserve currency.

No wonder the Federal Reserve and other central banks are moving hard to usurp jurisdiction over derivatives (per, for example, the recently announced Paulsen plan here in the U.S. to consolidate a variety of financial regulation under the Federal Reserve). To effectively regulate derivatives, for better or worse, the Fed will have to understand them first. But they do not have this knowledge. Indeed no human brain holds this knowledge. The knowledge needed to understand the dizzying variety of derivatives exists mainly in the form of computer simulations.

Derivatives, the money of the digital era, perform monetary functions in competition with the Federal Reserve's dollars, an increasingly primitive holdover of the paper era. By regulating derivatives the Federal Reserve regulates its competition. This creates a profound conflict of interest. It's like putting Rust Belt executives in charge of Silicon Valley. Such regulation may cause, not the popping of a commodity bubble, but the destruction of a nascent economic order made necessary by the prediliction of the Federal Reserve and other central banks to inflate their fiat currencies. Whether ignorance or conflict of interest will play the greater role in the destruction of the new currencies, or whether one will cancel the other out leaving the derivatives markets in relative freedom, remains to be seen.

[Obligatory Wikipedia links: Eurodollars, elasticity]

[H/T: I learned of Venero's paper from Byrne Hobart]

[Here's a prior explanation of mine giving a broader overview of why commodity prices reflect demand for their monetary functions as well as for their consumption, but not explaining the role of derivatives].

[*] These numbers are a bit stale -- it would be great to see to what extent this growth has continued. If any reader has data on commodity derivative use through 2007, please let me know.

[UPDATE: here is more straightforward article that gets the basic facts right: production up, inventories up, consumption down, prices up -- inexplicable if consumption is the only major source of demand -- but gets the explanation wrong, replacing the rational and efficient use of commodity derivatives as money that I have described here with the typical speculations about "speculation" resulting in a "bubble"].

[Ad: I am now providing consulting services.]


Anonymous said...

On my blog I post: The revolution begins when you can send commodity based money over end to end encrypted instant messaging, and the money is good to pay for your web hosting and your prepaid cell phone minutes.

Anonymous said...

There are two big problems with extending commodity based money to consumers: (1) the mental transaction cost problem of trying to keep track of prices and transactions in multiple currencies, and (2) legal barriers: not just legal tender laws and harassment from regulators (a la e-gold), but assumptions in various commercial law codes (especially the UCC) that "money", meaning government-issued money, is by definition used in a "sale." And if it's not a "sale", you can get dumped into strange and often untested legal categories.

I've long thought about the possibility of consumer derivatives. Indeed this gets back to my first thoughts on smart contracts -- one of my big inspirations were derivatives and the synthetic assets one can construct from them:

"Another area that might be considered in smart contract terms is synthetic assets. These new [in 1994] securities are formed by combining securities (such as bonds) and derivatives (options and futures) in a wide variety of ways. Very complex term structures for payments (ie, what payments get made when, the rate of interest, etc.) can now be built into standardized contracts and traded with low transaction costs, due to computerized analysis of these complex term structures. Synthetic assets allow us to arbitrage the different term structures desired by different customers, and they allow us to construct contracts that mimic other contracts, minus certain liabilities. As an example of the latter, synthetic assets have been constructed that mimic the returns of stocks in German companies, without requiring payment of the tax foreigners must pay to the German government for capital gains in German stocks. It's important to note that these synthetics do _not_ confer voting rights as do the originals. It might be possible to add smart contract protocols to transfer voting rights to the synthetic. Of course, these protocols might have to be quite secure to withstand attacks from the third party jurisdiction, whose transaction cost (the tax) is being arbitraged away by the synthetic asset."

With consumer derivatives, consumers might conduct transactions like "big money" does as described above: a legal "sale" using government money in one contract, and one or more derivatives to hedge the risks created by using the government money on the side.

If the consumer is not to be overwhelmed by the added mental transaction costs of multiple currencies and derivatives, something like what I call a market translator, a.k.a. contract compiler, is needed to make these complex transactions "under the covers" in a way that the consumer need not worry about, i.e. in a way that doesn't add substantial mental transaction costs. The idea is that if we can reduce the mental transaction costs of using multiple currencies to reasonable levels, we can probably also reduce those costs for using derivatives. Indeed it's possible that arbitrarily small and complex transactions can take place "under the covers" as long as they reflect user preferences with reasonable accuracy and are communicated to the user by a trustworthy and reasonably accurate, even if of necessity highly simplified, metaphor.

To put it another way, the idea is to take the software "big money" uses and put it into a consumer's budget and shopping software, so that similar sophisticated hedges can be made under the covers when government money is used to pay for web hosting or to recharge cell phones, or when commodity currencies unfamiliar to the consumer (and different from the denomination used in the consumer's budget software, tax forms, etc.) are used to pay for these things.

Anonymous said...

Money inherently tends to be a monopoly, because everyone wants to use the same money. This monopoly naturally tends to be abused, and to fall under state control. This is a problem even with precious metal currencies: You need some authorities to stamp the gold and silver as being pure and of a certain weight, and pretty soon that authority, like the supreme court, comes to the postmodernist position that it is the stamp that makes the coin, rather than the true weight and purity.

Indeed, the Koran addresses this issue, commanding Muslims to use and issue coins that are of true weight, which on its face would seem to command all Muslims to use precious metals coined by weight and purity - which commandment has been universally ignored, even by those authorities that are pretty keen on the commandments to murder various people.

This problem, that money tends to monopoly, and monopoly tends to be abused, would be somewhat diminished if we had software that eased the management of multiple monies.

Anonymous said...

This problem, that money tends to monopoly, and monopoly tends to be abused, would be somewhat diminished if we had software that eased the management of multiple monies.

I wholeheartedly agree with this, but reach even further: there is no known limit to how much the monopoly money problem can be diminished with software, via the reduction of mental transaction costs. Indeed, for certain narrow contractual problems (estimating and hedging risk), computerized derivatives have reduced mental transaction costs by factors often of millions, making possible a whole new world of contracts that were impossible before.

As for consumer transactions, there is a vast space of possible user interfaces, and of different metaphors for representing contractual terms, only a small fraction of which can reduce mental transactions at all, but some of which may dramatically reduce mental transaction costs and put the monopoly-money problem on the dustbin of history. We've got to keep trying to see how far we can reduce the mental transaction costs of multiple currencies.

George Weinberg said...

I'm not convinced that money is a natural monopoly. Even if everyone used the same denomination of money in a region (say, money is measured in ounces of silver) it's quite possible that people would accept coins from multiple mints just as they accept checks from multiple banks.

If a money denomination with reasonably constant purchasing power can be defined I'm not sure that most users would have to understand the definition. After all, what exactly is a dollar?

Anonymous said...

“If a money denomination with reasonably constant purchasing power can be defined I'm not sure that most users would have to understand the definition. After all, what exactly is a dollar?”

Originally a dollar was one ounce of silver, and multiple authorities issued circular chunks of silver each weighing an ounce, or some exact number of ounces. This definition was gradually reinterpreted away, much like the US constitution. Wherever one authority gained dominance, the definition of a dollar quietly changed.

Anonymous said...

george weinberg: I'm not convinced that money is a natural monopoly. Even if everyone used the same denomination of money in a region (say, money is measured in ounces of silver) it's quite possible that people would accept coins from multiple mints just as they accept checks from multiple banks.

One can simply observe the history of coins to see that governments have generally come to dominate coin issue. I estimate, and challenge anybody to prove me wrong, that over 99% of the coins found in archaeological sites, since the very beginning of coin history, were minted by an entity closely associated with dominant tax collectors. You have to explain how your modern minter can avoid this fate, without invoking improbably libertarian governments.

Coins used directly as money are now silly. Even before e-commerce coins were considered too bulky and insecure and were largely replaced for serious transactions by negotiable instruments, both bearer and non-bearer, more or less (often less, even in private banking) backed by gold or silver.

Thieves, pirates, highway robbers, and government confiscation were endemic before the rise of paper notes which can be hidden much more easily, and non-bearer negotiable instruments which can't readily be passed by a thief.

Coin issue relied on trust and reputation and almost always started as or soon became a monopoly associated with the dominant tax collector. Without continual assays of coins a level of indirection based on trust and reputation is introduced. Once one introduces yet another level of indirection, whether of paper or bits, one introduces even greater reliance on trust and reputation. These cannot be standardized short of a monopoly issuer, but vary greatly from issuer to issuer. A one gram silver note from one issuer may be worth much less than a one gram silver note from another. Risk can be arbitraged, but moral hazard cannot be eliminated, and it's a big source of transaction costs in fractional reserve banking. (And if you insist on 100% reserves, that's an even bigger source of transaction costs).

(There's a possible future exception to many of the above claims, namely bit gold may be secure as bits with minimal reliance on trust and reputation).

Another more subtle problem, but perhaps an even a bigger problem, is that any single commodity is far from an ideal store of value, because neither a monopoly issuer nor a modern market can credibly commit to maintain a single standard for a very long period of time. As James Donald well observes, "Wherever one authority gained dominance, the definition of a dollar quietly changed." Modern markets can't credibly commit, OTOH, because they are decentralized and it's not economically rational to make such a commitment when one can invest in a diversified portfolio of commodities and deliver legally reliable commodity futures, based on commodities stored at the warehouse of a third party, at the redemption window.

Since trust is limited, and thus the degree of fractional reserve limited, standardizing on a single commodity will create vast demands for that commodity, driving the price through the roof. The vast majority of its value will be based simply on using it as a standard, and any change of standards to another commodity will cause the price of the former standard to plummet. The mere long-shot risk of an imminent change of standards will generate great volatility in the standard's exchange value versus other commodities. Thus, quite ironically, using a single commodity as a standard makes that standard very risky as a store of value, unless one can credibly commit to maintaing that as a standard for decades to come. Nobody can do that, especially where one has monopoly or near-monopoly issuers or where mental transaction costs are low enough (as the are on international currency and commodity markets) to maintain portfolios diversified across currencies backed in many different commodities.

Diversification is a huge win, both for one's own portfolio and for overall economic stability, and we can be certain that smart investors with computers and electronic markets will take full advantage of it.

Gold and silver, and even better the most valuable gemstones, do have advantages where physical delivery is required, for example a redemption window where one can redeem gold notes for actual gold, But with a reliable legal system, one simply delivers legal rights in the form of paper or bits -- this is vastly cheaper than any physical delivery and is the same cost for any commodity. Without a reliable legal system, there is no guarantee that a redemption window based on either commodity futures or precious metals will be secure. (Again, bit gold might provide an exception to this. It makes possible a secure and cheap redemption window that doesn't rely on a legal system to function).

A diversified basket of commodities is far more secure, especially commodities that have a "floor" value that comes from consumption demand rather than just demand as money. Thus, gold and silver are for the most part the worst standards for modern markets to use, and we have observed and can expect them to exhibit both a high volatility and a slow long-term decline in value relative to other commodities as long as modern markets remain reliable. (The only exception is an improbable political disaster that makes electronic and paper instruments, including commodity futures, insecure but somehow allows precious metals to be used securely).

There is also an important practical consideration: gold and silver that are not issued by governments as official currency are not money according to the UCC and many other legal codes. e-gold and other commodity-backed systems that function as money are caught between a rock and a hard place -- considered "money" for the purposes of money laundering regulations, but not considered "money" under standard contract and other commercial law such as the UCC. They (to greatly oversimplify a wide variety of regulations, precedents etc. as usual when discussing law) get most of the burdens of the law while getting few of the benefits of the law. Changing this would take a vast lobbying effort and a change in the political wins that strikes me as quite improbable.

George Weinberg said...

I think historically things like coin debasement have been effective not so much because people have been fooled into thinking that the new coin really was the equal of the old, but because the issuer had the power to compel merchants and creditors to treat them as equal. Gresham's law requires that people understand that the "bad" and "good" monies are in some sense not really equally valuable but they are obligated to accept them as such. The fact that there is such a "law" indicates to me that this is historically a common situation.

As for realistic plans for getting around this, well, I'll try to get back to you on that ;-)

Anonymous said...

Bit gold has the problem that it will suffer rapid predictable inflation — which may still be better than unpredictable inflation, but I doubt people will wish to shift unless we get a fair bit of Mugabe style monetary management.

Before World War II Hong Kong had a truly free market banking system based on silver (even though England was on the gold standard) Every day the banks would cancel out their obligations to each other and demand physical delivery of physical silver from each other, because people kept an eye on a banks fractional reserve from day to day, so banks always urgently needed to top up their stash of physical silver. This had the effect that bank runs were frequent, but less catastrophic. Efforts to make bank runs impossible always seem to lead to less frequent but bigger bank runs.

There may be an impossibility proof here: Financial institutions will always sail close to the wind. If guaranteed by the fiat money issuer, must be regulated, if regulated, fractional reserve institutions will materialize in the loopholes outside the guarantee and outside the regulation, and there will be a run.

Bit gold has the advantage that in a fractional reserve banking system running on top of bit gold, we would reliably know the bank’s reserves from minute to minute, which would increase the frequency of runs while limiting the havoc.

Anonymous said...

james: Bit gold has the problem that it will suffer rapid predictable inflation

This problem is easily solved, because the bit strings (puzzle problem/solution pairs) are securely timestamped by their time of publication More recent (and thereby cheaper to produce and thus in greater supply) bits will be discounted by a market. To create fungible units one has to bundle strings of different value into pools of a standard value (i.e. collect strings into a pool so that the sum of the market values of the strings in the pool add up to the standard value).

It's a bit indirect, but comptuers can easily handle these logistics. Leaving aside the gold metaphor for a minute, one can think of these bit strings as digital rare postage stamps. Each stamp might trade for a different price, but one can sort stamps into pools so that the prices of stamps in each pool add up to the same total price. Then divide each pool into tranches to create your standard currency denominations.

One misleading aspect of the rare stamp metaphor is that unlike stamps, but like gold, there is little or no ongoing change in subjective valuation between bit strings to worry about, but instead it's purely based on how scarce the supply of puzzles solved during a given time period was and is: thus pooling will work much better for bit gold than it does for actual rare postage stamps.

There will be a problem defining futures contracts for yet-to-be produced bit gold: how much it might cost to solve a given puzzle a year later, or even a month, will be a very uncertain matter. But the pools that define currencies will be based on spot prices for already produced bit gold, not on futures.

Anonymous said...

the bit strings (puzzle problem/solution pairs) are securely timestamped by their time of publication More recent (and thereby cheaper to produce and thus in greater supply) bits will be discounted by a market.

If the market discriminates between one item of bit gold and another, they will be like collectibles - of uncertain value, not readily exchangeable.

Early in human history, people used collectibles as money, with the value of a collectible deriving in part from its visual interest, use as adornment, in its value as mnemonic for history and conversation, and from its uniqueness and irreplaceability. However, they tended over time to adopt standardized interchangeable monies.

one can think of these bit strings as digital rare postage stamps. Each stamp might trade for a different price, but one can sort stamps into pools so that the prices of stamps in each pool add up to the same total price.

Except that one can never know or predict the value of a rare postage stamp.

The great things about gold is that there is one and only one stable isotope of gold, and that gold is relatively easy to purify. To be useful as money, a bit gold needs to be all alike, but any such bit gold will suffer the inflation problem.

Anonymous said...

james: If the market discriminates between one item of bit gold and another, they will be like collectibles - of uncertain value, not readily exchangeable.

I'm afraid this is a profoundly mistaken conclusion, but since it keeps coming up I suspect gold sellers will keep trying to trash bit gold with it long into the future. So even though I long ago refuted it, I'll keep trying to communicate the proof.

There is a perfectly objective, measurable, and inelastic supply curve, completely derivable from the relative scarcity of bits (puzzle solutions) on the week (or the day, or the hour, or the minute, if necessary) of their publication. Arbitrage to set the different prices of different weeks (or minutes) can be computerized on this basis. The demand curve, the demand for puzzle solutions for their monetary value, will be based on recognition of the superiority of bit gold as a form of money that is more secure and has a far less elastic (indeed, perfectly inelastic) supply curve than traditional commodities such as precious metals. This demand curve will be the same function of scarcity for all weeks (or minutes), so it won't affect the simple scheme of automated arbitrage between epochs with different supply curves.

Early in human history, people used collectibles as money, with the value of a collectible deriving in part from its [aesthetic value]

The "in part" part is crucial, as is the question of why humans evolved such aesthetic values. The value of gold today is almost entirely based on its monetary value rather than mere decorative value. There are plenty of metals that are as shiny as gold, but people don't demand them as a store of value or medium of exchange because they are common. There are plenty of rocks that look as good as diamonds, but "diamonds are a girl's best friend" because they are hard to obtain and thus hold their value. It is their secure scarcity, not their aesthetic features, that allows them to be more securely used as money and thus gives them a monetary value far above the often trivial value they have as an aesthetic object. Furthermore, the asthetic instincts to which you refer, for example the instinct to collect shiny things, evolved just because in the evolutionary environement they approximated an instinct to collect scarce things, and to distinguish hard-to-find from easy-to-find things, i.e. an instinct to collect objects of monetary value, as I describe here, in the "Evolution..." section early in the paper, and the "Attributes of Collectibles" section late in the paper.

Anonymous said...

Good Job!: )

Anonymous said...

The upward movement of commodity prices is mainly due to the diversion of food for bio fuel leading to chaos. The demand-supply scenario is such that any diversion for non-consumption of human leads to shortage of agri products for human feeding.

There are also reports that futures too has played a role in lifting up the commodity prices incl. crude but it cannot be confirmed.