Liquidity As Information

Tim Iacono very aptly titles a post about the much-discussed “liquidity” in world markets, Hard to Define and Measure. I have long thought the notion of liquidity was ill-defined and under-theorized. Never fear, because, as usual, I have the answer!

In loose talk, liquidity usually has something to do with the quantity or availability of money. From this perspective, liquidity means a high monetary base, low interest rates, and/or easy access to credit for prospective borrowers. The academic literature usually operationalizes liquidity in terms of the bid-ask spread and price impact. In a liquid market, the bid-ask spread is narrow, and price-impact small. (Price impact refers to the amount prices change disadvantageously when one attempts to buy or sell a commodity.)

My proposal is that liquidity should be defined very simply as certainty of valuation of an asset with reference to some currency or commodity. An asset whose value in dollar terms is 100% certain is perfectly liquid in dollars. An asset whose value is completely random or unknown would be perfectly illiquid in dollars.

This definition maps very nicely to the academic stand-ins for liquidity. One needs only assume the usual no arbitrage condition to see this. Suppose there were a market (in dollars) for $10 bills. The dollar value of a $10 is trivially certain. What would the bid-ask spread be in this market? If a market maker could consistently sell ten dollar bills for $10.001 or buy ten dollar bills for $9.999, the market-maker could make infinite, risk free profit by doing so in volume. The bid-ask spread on $10 bills must quickly converge to zero to prevent a tear in the fabric of the financial universe. Similarly, suppose I have a zillion $10 bills to sell. Will the price move against me? In a world without informational frictions or transaction costs, no. If some market shyster, seeing that I’m desperate to sell, offers only $9.999 a piece, some other entrepreneur, eying a perfect arbitrage, will quickly offer $9.9995, until the price converges to $10 nearly instantaneously. You can see all of this in action in the real world. If you ask to “sell” a ten-spot (that is to make change) most store owners will buy it for you for precisely 10 one dollar bills. If you have a hundred thousand tens, a bank will purchase that truck-load for one million dollars. (This sort of purchase is called a “deposit”.)

The relationship between an informational definition of liquidity and the popular notion of “lots of money sloshing around” is more subtle, but very much worth teasing out. In addition to requiring the no arbitrage condition, we’ll make two additional assumptions. We’ll presume that as the quantity of a currency increases, so too do transaction volumes in that currency. (This is equivalent to the conventional monetarist assumption that money velocity is resistant to change.) We’ll also presume that market transaction prices vary continuously, and that the rate at which prices change over short periods of time is bounded and not sensitive to changes in the quantity of money. Under these assumptions, an increase in the availability of money also leads to an increase in informational liquidity. Why? Because given a current price, a prospective buyer or seller of an asset is fairly certain as to a near-future realizable price, since transactions are frequent and the rate at which prices change is bounded. A current price represents a fairly certain near future value in the currency at issue. From an informational perspective, it’s not the extra money that represents the liquidity, but the frequent, near-continuous transactions provoked by the ready availability of the currency. I like to think of the sort of liquidity caused by extra money as “sample rate liquidity”, in that it decreases the uncertainty of valuation by increasing the sample rate of the fluctuating values.

I think that an information definition of liquidity can be made precise, and that many fruitful avenues for research that could be derived from it. If one assumes that markets are efficient, and that market prices reflect but do not alter the value of underlying assets, one can consider transactions to be samples of a noisy signal. Each trade price represents a sample, and the size of the trade is a measure of sample accuracy. From signal theory we know that for any signal whose maximum frequency in the Fourier transform is bounded, there is a sample rate that is sufficient to reconstruct the signal perfectly, such that further sampling would be pointless. If one views financial markets as decision-making institutions, devices whereby economies tease out information about the true value of potential enterprises and investors then devote scarce resources to the most useful, then a bound on the liquidity required to fully value an asset over time represents a bound on useful liquidity. If one also presumes the existence of “noise traders”, entities who engage in transactions for reasons detached from a valuation of the asset being traded, and presumes that noise trading is sensitive to money availability, a bound on informationally useful liquidity should become a normative bound to central banks or other currency issuers, as increases in the availability of a currency beyond this bound increases noise without contributing to asset valuation, increasing the likelihood that an economy will devote scarce resources to erroneously valued projects. Similarly, insufficient “sampling rate liquidity” could lead to “aliasing”, where the underlying signal and its sampled reconstruction may bear little resemblence to one another. Between aliasing and noise-trading, there should be an informationally optimal level of “sample rate liquidity”, and potentially an informationally optimal level of money and credit for a given stock of tradable assets and a maximum frequency of “real” value fluctuations.

There is much more to go from here. Suppose, counter to our assumption above, the rate at which prices fluctuate is in fact sensitive to the quantity of money and credit availabilty. Then conventional measures of liquidity, like the bid-ask spread, might either expand or decline in response to increased money, depending on a race between the increased slope of the price time-series and the increase in the frequency of transactions. In either case, this is a bad situation, as increased market activity, rather than more precisely valuing resources is simply decreasing the precision which with resources can be valued. I think a real world analysis would show that the effect of money and credit are non-uniform, that there are times and circumstances where additional money is likely to improve the informational resolution of markets, and times when it is likely to magnify noise, and that with a bit of effort, theoretical and empirical, these regions could be usefully characterized. I don’t think Taylor-rule-style monetary regimes even begin to capture this dynamic. Readers of this blog will be unsurprised to know that I think we are presently in a region wherein “lots-of-money-sloshing-around” is creating the appearance of liquidity (narrower spreads, less price impact) without the sine qua non of genuine liquidity: additional information or certainty about the real-economic value of the assets being exchanged and priced.


37 Responses to “Liquidity As Information”

  1. We2l3 writes:

    I can’t escape the feeling that you have left out one element of liquidity that is important in market design. Maybe I have misunderstood, but a big part of liquidity is the assurance that you will actually be able to move an asset quickly. “Noise traders” at Goldman probably aren’t going to be encouraged by what you are calling “genuine” liquidity. If I understand correctly, variance actually gives them a larger opening to make a buck. Lots of money sloshing around doesn’t seem to quite explain this part of the liquidity puzzle either, or at least it explains it in a very circular way. Obviously *some* things you won’t buy just because you can… ripe mangos, for example. Ripe mangos have a certain value, but it declines so quickly that money isn’t just going to slosh itself into the market for ripe mangos. Why isn’t the same thing true of the stocks that noise traders take? “Genuine liquidity” is out, and the noise traders *are* the sloshing money, so I think we need a third aspect of liquidity to explain why sloshing money is willing (or unwilling) to slosh its way into securities markets.

  2. We2l3 — Good call on the missing piece. The element of liquidity you refer to is a counterpart to price impact — that is in a liquid market, one can buy or sell immediately without moving prices. In an illiquid market, you can transact in a manner that does not move the market, but only by transacting over a prolonged period of time (accepting the risk that the market moves against you for reasons other than your market impact), or you can transact immediately, but will have to pay in the form of prices moving against you when you transact. I really should have made clearer the immediacy requirement when talking about price impact.

    I do think you are right that, in this real world, lots of people make money from “noise trading”, that is volatility without informational content regarding the assets underlying the prices. I think that is an error, a design flaw, a form of rent-seeking that should be abolished. Some would argue that the volatility attracts foolish gamblers, and that liquid markets require the subsidy of fools, but I am not than cynical. I think we can come up with better schemes for subsidizing liquidity in the markets for which a liquidity subsidy is really necessary.

    You also pose a wonderful question when you ask why some sorts of markets attract noise traders, and others don’t. Here’s a quick take: there are two sides to noise-trading, we’ll call them privileged parties and fools. The privileged parties expect to profit from the transactions of the fools. For a market to be amenable to this sort of thing, the valuation of the securities being priced must be inherently complex, so that a wide range of prices is plausible for the same security under the any set of real-economic circumstances. The securities must be long-lasting, as the primary strategy privileged players use is persuading fools to enter into positions they cannot hold indefinitely, and riding the volatility until weaker parties are shaken out. Money sloshes into these markets because the privileged parties are precisely in the business of lending and promoting, and they do both well. The ripe mango market wouldn’t work — the commodity is too perishable for players to ride long waves of volatility, and valuation of the commodity is probably too straightforward. Stock markets, as currently (ill-)defined are ideal. Given the same accounting and economic data, it is possible for smart analysts to come up with a very wide range of estimates for a share of common stock. The reasonable range for the fundamental value of an index or ETF is even wider (despite the lower volatility of indices than single issues).

    It would be possible to define securities by which firms could obtain at-risk equity funding, but whose valuation would be very straightforward and coupled in a fine-grained way to firm or unit-performance over a predetermined period of time. Such securities would be less amenable to rent-seeking by noise traders. It would not be rocket science to define them. Why aren’t such securities defined and traded? Ask the SEC. Regulatory capture is the most basic strategy employed by rent-seekers to ensure that the gravy train keep on coming. Our “sophisticated, modern” financial markets are, in my opinion, a crystalline palace for very well connected and protected rent seekers.

  3. The biggest areas for speculation (surely) are not long lived assets like equities but shorter lived assets like futures contracts and options. And I’ve always presumed that the main trade (and gain from it) is not between fools and privileged parties, but between those who are betting to hedge (in which case they’re after insurance and are prepared to pay for it) and those who speculate to pick up the money the hedgers leave lying around on the pavement.

    (Though I agree there is also the transfer of money from fools to the privileged – particularly in equities – by way of insider trading (of various degrees of legality and ethics)

  4. Nicholas, no argument on the speculative attraction of futures and options. And no argument on the traditional notion of hedgers and speculators, where hedgers pay for protection and speculators assume risk. That is how derivatives markets are supposed to work, and why they are not zero-sum games.

    I was responding particularly to the rationale for noise-traders: traders who are uninterested in the economics of underlying assets, but hope to profit from mere volatility. When speculators take on risk for hedgers at a reasonable profit, that is financial markets working well. When “technical traders” or deep pockets take positions in assets that are difficult to price, hoping to profit from behaviorally predicted fluctuations or the mere ability to outlast contrary moves, that is “noise-trading”. My contention is that noise-trading is a profitable racket for privileged players (and, like any casino, sometimes profitable for regular customers too), and that this is a problem. Noise-traders profit from distortions and misalignments of markets, and have an incentive to encourage mispricing, simply to get the volatility they need, or in hopes of profiting from a correction. This is fundamentally different from normal speculation against hedgers in futures markets. Normal speculators drive prices towards the zero-profit correct valuation of an underlying asset. They dampen volatility. Naive noise traders create volatility simply by being wrong, by mispricing assets and throwing money behind their errors. Sophisticated noise-traders look at markets as a game-theoretical battlefield, in which accurate pricing of underlyings is collateral damage.

    I do want to say, that in terms of security definition, futures and options are much more coherent than equity. Futures pricing in particular is very straightforward. Given an accurate understanding of the pricing, cash flow, storability, and storage costs of an underlying, a future can be normatively priced. Not so for equities, or equity indexes. Futures on equities or equity indices can be no more coherent than their underlyings. The problem lies in the poor definition of equities.

    (You might object that commodities are often much more volatile than equities or equity indices. And that’s often true. But the volatility of industrial and agricultural commodity pricing is real-economic volatility, not usually manufactured or exacerbated by financial markets, and financial markets serve as effective vehicles to hedge real volatility that would exist so long as, say, weather exists. My contention, though I grant reasonable people can disagree, is that for many financial assets, much volatility is an artificial outgrowth of poor security definition and market-gamesmanship, rather than being related to inherent volatility in the value of the underlying cash flows or enterprises.)

  5. Thanks Randy,

    I’m not really disagreeing with you. I am sympathetic to the argument, but that’s why I think it needs to be put very scrupulously. I agree that too many traders can destabilise and threaten the efficiency of a capital market. But I’m sceptical of the language you use. You speak of deliberate ‘manipulation’ of the markets and of ‘noise traders’ somehow introducing a volatility to exploit it. Doesn’t that call for awfully big traders or collusion between them? My impression is that there are a whole lot of noise trading programs out there all betting against each other and against the hedgers. This can be destabilising and so can reduce efficiency, but the picture of the privileged and the fools I think could be quite misleading.

  6. Oops – thanks Steve.

  7. Yeah, “the privileged and the fools” is overstatement and exaggeration, and perhaps unwise, since it invites marginalization.

    The main article, before the comments, tries to make a much more technical argument, that easy money may cause the appearance of liquidity (reduced spreads, less price impact to large trades) without actually serving the one of the main purposes of liquidity — more accurate pricing of assets, and the ability to actually trade at fair prices. To the degree that easy money induces volatility not related to a fair valuation of underlying assets, easy money increases the likelihood of misallocations of real-economy capital. “Noise-traders” in the main article are not necessarily manipulators or fools, just anyone who transacts in an asset market without valuing the underlying asset, on the (self-fulfilling) expectation that price fluctuations are loosely if at all coupled to any “fundamental value” of an asset. This is a less loaded, although still very controversial, sort of argument to make.

    That said, I think people underestimate the degree to which even very large capital markets are manipulable for short periods of time. Games really are played. Capital markets have never been and are not now the efficient calculators of academic theory. Those theories shold be regarded as prescriptive rather than descriptive — we should strive to modify market institutions to make them efficient, not simply presume that they naturally are efficient. The usual stories, that any mispricing will be quickly arbitraged away, have been very well debunked by an entire literature on “limits to arbitrage”, and by any practical investor’s experience. Lots of short tech investors were 100% correct on fundamentals in the late 1990s. It didn’t prevent any of them from going bankrupt.

    Capital market manipulation is not a conspiracy theory. A single hedge fund manager can purposefully alter security and index prices for a short period of time, if she is willing to risk direct costs on the hope of indirect gains. (The famous screaming Cramer recently admitted to having done so as a fund manager, and argued that managers who refuse to do so when it is arguably legal aren’t really doing their jobs.) “Momentum trading” amounts to a kind of collective, loosely coordinated form of market manipulation, in which traders rely on one another to contribute to the mispricing of an asset, while they compete with one another in a game of chicken to profit as much as possible from the mounting distortion but escape the inevitable but unpredictable correction. And in this game, there are more and less privileged players. Ones ability to be wrong in timing and still eventually profit is directly related to ones level of capitalization and ability to take on leverage with minimal risk. Some entities that make markets, provide credit, and trade on their own account have access to a lot of information that other players do not have and may, despite firewalls and whatnot, improve their timing in these games by virtue of that information.

    One can easily overstate this sort of case. But most mainstream writing and financial theory very much understates the real-world liability of markets to mispricing and manipulation. Near absolute faith in markets that frequently generate erroneous signals invites consequential errors in the organization of the real economy. (That invitation has already been accepted, in my opinion.)

  8. Thanks Steve,

    We’re not far apart, but I much prefer the idea that a bunch of momentum traders can create big distortions than the idea that “A single hedge fund manager can purposefully alter security and index prices for a short period of time, if she is willing to risk direct costs on the hope of indirect gains.” That seems to be a strangely risky way to make money. Why not head in the direction of some other market where it’s happening already?

    More to the point, do you see any policy measures that would deal satisfactorily with the problems you see?

  9. We2l3 writes:

    To Nick — the reason I made my original point to Steve is that I have a relative who was indirectly involved in, well, let’s call it market design. I took Steve’s basic position; noise traders (or to use our preferred metonym, Goldman) were not actually adding a great deal to the market. My relative defended a position that he couldn’t really deviate from, because it was his employer’s position; even if Goldman does zero research on fundamentals, introduces unnecessary volatility to the market, and diminishes the gains to research into fundamentals, they still are essential to the market because they provide liquidity for the people who do trade on fundamentals.

    So in short, Nick, if Steve and I are correct, the fact that markets are intentionally designed to attract noise traders is a serious problem. If my relative and his employer are correct, noise traders are adding a great deal of value to the market via a mechanism that is difficult to theorize articulately, and markets are designed quite well.

  10. moldbug writes:

    Steve –

    Sorry to butt in late here! This is a response to the original post, it has nothing to do with the very interesting conversation about noise traders.

    First of all, I dislike the word “liquidity” because it is, as you say, ambiguous. Trying to take a word that many people have many fuzzy overlapping definitions for, and trying to turn it into a useful and well-defined term, is like trying to turn Chernobyl into a luxury golf resort – no doubt with sufficient effort it could be done, but why? “Inflation” is in the same category.

    Accepting, for the moment, your use of the term and your definition, though, I think it can be improved.

    I think you’re missing an important dimension of liquidity, which is maturity. There is a qualitative difference between claims that are mature now and ones that mature in, say, 30 years. The yield curve is not an artifact – it reflects real time preferences. If an asteroid was scheduled to hit the Earth in 20 years, and there was nothing we could do about it, you couldn’t give away a 30-year T-bill.

    So if we parameterize claims to some commodity – whether it is monopoly currency, some natural substance, or whatever – we see that the money supply is no longer a zero-dimensional number. We can speak of the supply of money that will be mature at time T.

    Of course, we should apply the correction you suggest, and define this number not by the face value of these assets, but by their discount from par, reflecting their probability of default.

    One very interesting characteristic of modern fiat finance, which most people just take for granted, is that not only are there far more current claims to money than money itself, there are way, way, more future claims. Yet the supply of actual money, M0, or mature money, MZM, does not expand as one might expect with the maturation of these claims. Instead they are rolled over, in a mysteriously consistent manner.

    Clearly, if we draw any line between two sets of agents, labeling each A or B, and net out all claims between A and B, if we find that in claims maturing before time T, A owes net more to B than it actually has to give, A is insolvent. For B to rectify this insolvency by extending more loans would be the height of imprudence. Similarly, if the current risk-discounted value of claims on A exceeds A’s actual assets, these claims are overvalued.

    If you take A to be the banking system as a whole, you see where this line of reasoning is going. Clearly, A behaves as if it has an infinite line of credit to draw on. Neglecting this factor, and conceiving the financial system as a closed loop, leads one to postulate a reality that is clearly not real. Just as nothing in biology makes sense without evolution (Dobzhansky), nothing in fiat-currency finance makes sense without the Fed.

    So this is what people mean intuitively, I think, when they talk about there being a lot of liquidity. They mean that a lot of claims to dollars of a certain maturity are being issued – that the net quantity of such claims, as measured by some division between A and B, is rising.

    The interesting question is what happens when this same A issues claims that are denominated not in dollars, but in substances of which the Fed has a limited quantity in its hat. One cannot say categorically: it is an error to buy a T-bill. But I believe this statement is more open to question when one considers, say, a Comex gold future.

  11. moldbug writes:

    Let me be slightly clearer about relating my definition to yours:

    You relate the folk definition of “liquidity” (an increasing money supply, largely in claims of nonzero maturity) to the “genuine” or classical “liquidness” of an asset or asset claim, that is, the clarity of its price. Then you suggest that an increase in “folk liquidity” may in fact simulate “genuine liquidity” – do I have this right?

    My little response above is attempting to pin down the “folk liquidity” side a little better. It says nothing about “genuine liquidity.”

    However, it’s worth noting that the activities of a lender of last resort introduce an enormous amount of gamable uncertainty into the market. Consider the change in dollar asset prices if the Fed “broke its plates.” Consider the change if it ran the printers full out. The power of this political signal is essentially infinite.

  12. Wow. First, thanks to all for the thoughtful conversation.

    Nicholas — In theory, market manipulation is just a dumb idea. If a security has a well-defined fundamental value, bidding its price up above the value or selling it below is just a way of giving money away to arbitrageurs.

    But in the real world, most securities are not sufficiently well-defined to have an unambiguous fundamental value. When the current price and price movements are inputs into the market’s valuation process, creating a price movement may change the market value of a security. Suppose that you believe there are lots of “chartists” looking for “breakouts” from “channels”. Suppose you manage a large fund, and you note that a stock of you own a great deal of is near such a “breakout”. One fund can easily bid up the intraday price of a thin to moderately traded stock. Suppose you also have friends in the business press. By augmenting your already large position by a few percent, while phoning around with a positive rumor about what sparked the “breakout”, you might invite a lot of speculative interest in your stock, and create a stable revaluation in the market above the current price point, from which you can begin to unload your considerable holdings. The strategy is not without risk, but no strategies are, and it’s not ridiculously risky or expensive either. This is just an example. There are many other sorts of price-manipulation strategies, and more sophisticated (but often riskier) tools allow for transient manipulations even of whole price indices. Real-world trading is very psychological and game-theoretic, and simply does not resemble the efficient-markets of academic theory wherein manipulation never pays.

    Re policy proposals — I think the only way out of this mess are alternative economic and market institutions that compete favorably with the current nightmare. At the policy level, I think the most important thing that could be done is to permit experimentation with respect to new market institutions, and to permit failure with respect to the old. That would mean deregulation, and getting rid of the Greenspan/Bernanke/BoJ/PBoC put. (At a very “macro” level — with moldbug, I think, regarding the problem, though not the solution — I think monopoly fiat currencies will have to go the way of the dodo bird.)

    I’m a finance-interested computer programmer. I think I could define a small-scale securities market through which local business could raise capital and investors could put money into businesses they actually know. The securities bought and sold would look neither like traditional stock nor like traditional debt. I could define the technical infrastructure in a matter of months. But I’d be shut down by the SEC/CFTC/IRS within a matter of minutes, because a whole network of regulation wouldn’t be compatible with my novel securities. In theory, a process exists for trying new things. In practice, the barriers to entry are large (though perhaps not insuperable). Most experiments of this sort (including any I might come up with) would probably fail. That is the nature of an experiment. But finance is a field in which experimentation only happens at the initiative and in the interest of large established players. There is a lot of creative destruction in the finance industry that has been prevented from happening.

    We2l3 — It looks like we mostly agree. It’d be great to put the value of noise-trading to the test, by building systems more resistant and letting the best market win.

    moldbug — To be continued… (I’ll try to write tonight.)

  13. Moldbug — Okay. I find very little in your comments to disagree with, so that might leave this a bit of a dull response. The headline essay here is a rough draft of a rough draft. I’d like to spell out more mathematically what I mean by “genuine” liquidity. If I were to do so, as you say, it would not be a simple scalar, like “the money supply is 10-gazillion dollars”. As both you and Nicholas have pointed out, liquidity needs to have a time dimension. But I think we are emphasizing different things. The most “liquid money” is zero maturity money, but to me it is a definitional question, “liquid in terms of what currency or commodity?”, and I’d be willing to consider liquidity in terms of some “broad money”, to the degree that can be defined. But the notion of nonzero maturity money, as you suggest, implies credit risk. And fiat money implies dilution risk. One way of thinking about “pseudoliquidity”, a false appearance of liquidity (low spreads at high volumes that suggest a degree of certainty in valuation that is inaccurate), is that it represents a market failure whereby agents underestimate the endogenous volatility of a currency, where despite increasing expected future volatility due to overprinting or extension of credit, agents value assets as though the “plates” had been broken already. MZM has dilution risk, but not credit risk. Liquidity defined in terms of broad money, which has both risks, is more likely to be compromised by currency volatility. But, paradoxically, since there is always many times more broad money than base money, to the degree market participants consider the two both interchangable and stable, broad money-based liquidity will seem greater than cash money liquidity. If broad money is not stable, this apparent liquidity is an illusion.

    I like your B <-> A &lt-> Fed picture, and agree with your point that A would be insolvent by definition without the Fed behind them. I also agree that what people mean by “folk liquidity” is that certain kinds of claims or financial assets are being generated, though I don’t think that maturity is the only criteria that matters. Sure, a 6 month note is always more money-like than a 30yr T-Bond, but I think it’s the depth, spread, and speed of the market between an asset and cash, rather than simple maturity, that defines how money-like, or “folk-liquidity-producing” a new claim is. I agree that there’s a stark qualitative difference between true cash and future claims on money, and that this difference creates the danger that much apparent liquidity will simply disappear someday, but still, the folk consider their eurodollar deposits to be liquidity, and will do so until those deposits can no longer be redeemed.

  14. (hmm…. i think i’m kind of confusing zero-maturity money and cash in the comment above. i’m treating deposits as claims on future money, but i think it’s conventionally zero maturity. or, more with more jargon, i’m taking MZM to be M0 or base money, which is not the conventional definition.)

  15. moldbug writes:

    Our confusions here are a good example of why I dislike the word “liquidity.”

    Liquidity in the “genuine” sense of the word – “low bid-ask spreads at high volumes” – has more or less been abandoned to the “pseudoliquidity” sense of the supply of credit. This is very analogous to the way “inflation” has come to mean “aggregate quality-adjusted consumer price appreciation” instead of “monetary dilution.”

    I fundamentally do not think these words are rescuable. I think they have become tref and need to be buried in the dirt for at least a few years. It is some effort to invent new words, but I think it is easier than fixing “liquidity” and “inflation.”

    There definitely needs to be a word for the cost, defined in terms of Y, of exchanging X for Y and back to X again. Would you agree that this is what you are trying to get at with your definition of “liquidity”?

    If so, perhaps a metaphor can be drawn from physics, in terms of friction, thermodynamic efficiency, electrical resistance, etc, etc. The concept is more or less the same as Menger’s “saleability” or Mises’ “marketability,” but neither of these words sounds right to me now.

    Back to the money supply sense: yes, MZM is broad money, not base money. But in practice, which is after all what counts, the difference between broad and base money is not a matter of credit risk. It is an implementation detail of our bank regulatory system, it is not related to any fundamental concept at all. (Mutual exclusion is a fundamental concept – lock files are an implementation detail.)

    To anyone who is not a bank, broad money is the same as base money, because the Fed says so. Fiat bux. Anything the Fed is willing to exchange for one of their pretty little notes is a dollar.

    What Austrians who define money supply as MZM miss is that the power of fiat applies not only to money of zero maturity. If you redefined checking accounts, for example, as time deposits that were rolled over every day, every hour, or nanosecond, or whatever, their economics would change only by this epsilon. Imposing a qualitative distinction at zero is bad praxeology.

    In other words, it is the power of fiat itself that has a time dimension. The Fed can “print” money not only at t=0, but at t=whatever.

    And it does this in the same way for t=0 and t=whatever: by implicitly insuring nominally private liabilities. Checking deposits – broad money – are actually supposedly “insured,” in the actual actuarial sense of the word, by the FDIC. But of course a bank run is not an insurable risk, so the real insurer is the Fed. (I’m really not sure what the people at the FDIC do, though I’m sure if you tried to take away their staplers they would have some kind of a story.)

    By implicitly insuring claims of nontrivial maturity – the canonical example, of course, is T-bills, but agency bonds apply as well – the Fed’s fiat power is responsible for a significant fraction of the market price of these instruments. If you imagine the price of any of these assets if “the plates were broken” – of course, this cannot actually be measured without actually doing it – you see how much of it is fiat. This is dilution without printing – a sort of quantum tunneling, if you will. And none, or at least hardly any, of it appears in MZM.

    Not even when it reaches maturity. Because it never needs to reach maturity. Claims on Fed-insured entities can be, and typically are, rolled over and exchanged for new debt. This is another fiat-world practice that would not exist in a hard-money economy. A disinterested creditor in an efficiently enforced free market should never roll over a loan. If the loan cannot be paid, liquidation is his best option. If it can be paid, and he wants to lend the money again, he should claim it and lend it to someone else.

    This phenomenon – fiat dilution in the market for immature claims to fiat money – is of crucial importance when we look at monetary reconstruction. If you are willing to look in any serious way at reconstructing the monetary system, the language that has to be used is the language of liquidation, which is fundamentally an exchange of claims for equity. My assertion is that many claims which are not, legally, claims against the Fed, have fiat dilution introduced by Fed insurance as a substantial part of their market price. If you disregard this form of fiat when you reconstruct, you introduce considerable disparities in the treatment of creditors, which obviously is to be avoided. Rothbard, for example, makes this mistake in his plan for returning to gold.

    Another way to say what I’m saying is that if you have a plan for monetary reconstruction, the first step in that plan needs to be nationalization of all banks, futures exchanges, and agencies, because in practice the market treats their liabilities as Federal liabilities, which makes them de facto parts of the state. Nationalizing them, and including their monetary liabilities in the book of fiat notes, is simply a formalization of reality.

    The reason I say this is that you seem to hint above that you have some kind of idea of what a stable monetary system might look like, and you seem to hint that your plan is not based on the usual tawdry metals. Hm. As it happens I also have some ideas in this department. I’ll show you mine if you’ll show me yours…

  16. MB — Most useful knowledge does arise from reflecting upon our confusions. To be confused constantly is the greatest gift.

    I think you are right that much of what ought to be credit risk has been converted to dilution risk by virtue of an implicit Fed guarantee. And while credit risk and dilution risk are two sides of the same coin, dilution risk more effectively socializes costs that ought to be borne by those who assumed risk. There is some rationale for nations/societies/organizations/whatever sometimes socializing private risks to encourage risk-taking. But the precise degree and manner in which some private risks should sometimes receive automatic guarantees by a larger community is an unresolved problem, which means in practice it falls to judgement calls and “public-choice”-ey gamesmanship. In my judgement, our collective decision-making institutions have proved inadequate to the task of making wise judgments in the face of these pressures. Polities have simply ceded the question of whether a private enterprise deserves public insurance to the private entrepreneurs themselves, and private entrepreneurs have learned how to capture the value of the free insurance policy without actually doing anything worthwhile for the public that pays the claims.

    (The public/private distinction is inadequate here. The cow that bankers and credit innovators are milking is the same one that congressmen are milking when they buy popularity with the proceeds of debt that will have to be monetized.)

    Re the shy ten-year olds playing doctor, mine is not all that interesting. The way out, as I see it, is to unbundle the various roles played by money and encourage actors to make wise and diverse choices. It is really the uninvention of money as a “natural monopoly”. (I think his is a very Austrian idea, but I’m much too ignorant to claim such a thing.) You, for example, may prefer to be paid in claims on allocated mold, managed by an institution of your choice. I store my wealth in the form of claims of year-in-advance resort accommodation, guaranteed by a clearinghouse I have chosen (and which I typically rollover). Payment intermediaries define direct cross-rates between the two sorts of claims, and negotiate a deallocation of my claims on hotel rooms and an allocation to you of claims in mold to consummate an exchange (in which I purchase the right to see yours, I suppose). As computer guys, we know that what I’ve described is a difficult sort of architecture to sustain: I’m suggesting a complete graph (nodes are claim types, edges direct cross rates, order n-squared connections) rather than a star topology (order n connections). A monopoly currency is apparently as inevitable as, say, an internet backbone: My network does not connect to your network directly. We choose a common intermediary to avoid complexity, chaos, and cost.

    But, wait a minute, the “internet backbone” is not really a monolith. There is no one device whose failure would bring down the system, and if some links started to behave flakily, the abstraction that is “the backbone” would evolve to incrementally exclude them, while incrementally relying on more dependable links, without any sort of edict. The internet architecture represents a sort of compromise between what would be ideal if there were no possibility of failure or corruption (a fixed hierarchy or star topology) and an alternative that would minimize the hazard of any failure but at very great cost (every computer connected to every other).

    Traditional money, whether gold or paper, represents a star topology for managing exchange. If any money were “perfect”, a money-centric star would be the ideal organization. But all forms of money are flawed, cooptable by different sorts of rent-seekers when either achieves a near monopoly on exchange, or, much worse, when claims on either are widely used as stores of value. Complete moneylessness, actual barter of goods and services, is uncorruptable but impossibly unwieldy. The best we can hope for is a compromise, a world in which a diverse array of money-like claims are used as stores of value, media of exchange, and units of account. Such a world can’t be created by fiat. There is no invention that would solve money’s problems.

    But, if the diversity of money-like claims increases, if the friction and cost associated with interconverting between such claims grows sufficiently small, if norms, regulations, and institutions evolve in a manner compatible with the exchange of and accounting for heterogenous claims, if tools and interfaces are developed that make navigating a multicurrency world tractable and intuitive for everyday anybody, then an ever evolving constellation of money-like claims could provide both for convenient, universal exchange and robustness in the face of monetary gaming. These are big ifs. That song in your head might be Imagine by John Lennon. But I’m not the only one.

    You see, rich people already live in the world I’m describing. They use a very diverse array of claims to store their wealth, and are not vulnerable to the first-order effects of monetary instability of any sort. (Of course come the revolution, all claims are invisible to the executioner. Second-order effects matter.) People accustomed to overt hyperinflation live in a multicurrency world, where wealth is constantly rebalanced among a variety of real assets and foreign exchange.

    Frictions can be incrementally diminished, and tools and interfaces improve over time. The main problem to be solved is how to defend against “temporary reliability attacks”, wherein some currency becomes popular, behaves well for a prolonged period of time, such that the logic of a star topology asserts itself and one sort of claim becomes hegemonic money. I think this problem is manageable, but more on that will have to wait for another time.

    By the way, re liquidity, our main difference is whether we think the word is salvageable. You have captured the essence of what I’m getting at with “the cost, defined in terms of Y, of exchanging X for Y and back to X again.” I think liquidity is properly defined as a surface in a three dimensional space whose axes are cost-in-Y-per-unit-roundtrip (“spread”), required-quantity-of-exchange (“quantity”), and maximum-permissible-duration-of-exchange (“time”). Just as supply in economics should not be described by a number, but by a curve, liquidity is described by the shape of a surface. If we let spread define the Y axis of such a space, the paradoxical effects of overzealous “folk liquidity” include (i) an upward movement of the entire surface (a increase in the spread due to “adverse selection” risk, owing to monetary volaitility) combined with (ii) a diminishment in the slope [d spread / d quantity ], or a decrease in the marginal spread required to motivate market-makers and pay transaction costs. Just as most changes in the shape of a supply curve can’t be unambiguously described as “increases” or “decreases”, the effect of money-printing on liquidity is complicated, and by looking only at (ii) above, one can claim an apparent “increase in liquidity”. But the asymptotic (as t goes to infinity) spread due to adverse selection is more important than the diminishment of marginal transaction costs, so the claimed of an increase in liquidity is worse than misleading.

  17. moldbug writes:

    Re liquidity, yes. It is interesting to observe the economic path by which the surface of exchange you describe is affected by dilution of the money supply, and the linguistic path by which the two became confused. But I still wish you luck in disentangling them!

    I think our main distinction on dilution is that I don’t think of the dilution as risk at all. I think of it as something that has already happened, a fait accompli.

    These are two equally valid ways of looking at the same problem. I just think that mine is simpler.

    The question is: why is a “broad dollar” worth exactly as much (to non-banks) as a “base dollar”?

    The answer in your sense is that the probability of a bank run and the probability of a dilution event (ie, the dilution risk – but “risk” is a slightly curious word, as this event of course is positive for the holder of said broad dollar, though it may not be desired by others who are using their dollars to compete for the same goods and services) are exactly, or almost exactly, equal.

    The answer in my sense is that the fact that these probabilities balance is not a coincidence, but a matter of policy. A political decision, enforced by the power of the state.

    Which is exactly the same power that stands behind the dollar. The US government creates considerable demand for dollars by demanding said instruments from us every April 15. It is just as happy to accept a broad dollar as a base dollar.

    My views may occasionally sound libertarian, but I am not a libertarian at all. I think of law as a practical rather than an ethical instrument. Property rights, to me, are just a way to keep people from fighting over property. If a better way could be devised – some drug, for example, that we could take, that would all make us love each other, and wouldn’t wear off the way MDMA does – I’d be first in line for the injection.

    The problem with a system in which a large part of the value of assets is provided by “winks and nods” – informal promises, such the Greenspan put – is that, in the terminology of de Jouvenel, it represents a step away from Law and toward Power. Goldman Sachs can make $20 billion a year or whatever off FICC because it is in the business of translating its informal proximity to this informal information into trading advantage. This is plunder.

    In my view, the way to fix such a system is not to dismantle the Greenspan put, but to formalize it. Confiscating the proceeds of crime is an ineffective and counterproductive way to restore law. All the real estate on earth, for example, has been alienated by violence many times over. You have to look at who has what now and give them a formal title to it.

    To me this equals formal dilution: replacing the informal value created by the Greenspan put with formal and accountable currency. Every claim on an implicitly insured financial institution (bank, agency, futures exchange) has two components of value: the value of the instrument and the value of the insurance. Since these cannot be separated without actually precipitating a market collapse, the correct formalist way to rectify the situation is to cancel them all and replace them with new dollars at the present market price.

    If you want to take the money that is returned to you in this flag-day closeout and buy a derivative from an uninsured actor, of course you should be free to do so. But my guess is that this business would be much smaller. Ideally it would involve the matching of equal and opposite risks, as in the purchase of a future by an industrial commodity user from an industrial producer.

    Back to money. Your views are actually in some ways more Austrian than mine. (The best Austrian primer, in my opinion, is the original – Mises’ _Theory of Money and Credit_. Full text is online.)

    I believe a star topology, as you put it, is unavoidable in monetary systems. Let me explain why. (This is all in this paper I’m writing, in my, um, copious spare time. NB: if you ever decide to design a new programming language and a revised theory of money, try not to do them in parallel. The only thing that saves me is that I have neither a job nor a blog.)

    Menger and Mises, like you, emphasize transaction efficiency (your “liquidity”) in the appearance of a monetary standard. Their view is that the most efficiently exchangeable good (“saleable” in Menger’s terms, “marketable” in Mises’) tends to become more efficiently exchangeable, creating, as you say, a star topology.

    This view appears very reasonable in the historical context in which money arose, which was hardly marked by sophisticated markets. In fact, today’s markets, as you point out, are lousy. (We don’t have a computerized financial system – we have a paper financial system that’s been ported to computers. Most people who do computerized finance seem to learn this system in much the way one might learn, say, VMS, or the Win32 API – in a practical context, that is, in which its lousiness is utterly irrelevant to their own personal objectives, and therefore need not be understood as such.)

    One way to forget this is to imagine the problem not in the context of real markets today, but in the context of a virtual world (MMORPG) such as Second Life or World of Warcraft. As you probably know, many of these systems have thriving economies which seem to obey all the same principles as the real ones. And they should, because they are also based on the axioms of human action in an environment of scarce resources.

    Of course, these worlds have implementation details of their own. But it somehow feels more grounded to construct an abstract virtual world than to construct an abstract real world. The planet has seen enough of the latter.

    So our gedankenexperiment will be to construct a premonetary equilibrium – your mesh topology, in which no asset has a central position in exchange valuation – in such an abstract world. Then, we will try and break the equilibrium in various interesting ways, and see what happens.

    My belief is that the following system is a premonetary equilibrium: a virtual world in which there are a fixed number of assets of every class (magic swords, etc), in which an asset can be teleported from any player to any other player at any time, in which there is no storage cost for inventory, and in which a central Walrasian auctioneer, limit-order book, bot, etc, provides automated exchange services.

    As an owner of assets, you could simply declare the goods that you were willing to exchange any of your goods for, set them as object annotations so to speak, and the exchange would happen automatically at any point at which it was demanded. (You might want to mark up your sword for a few minutes if you happen to be in the middle of fighting some orcs or whatever.)

    This design eliminates, I believe, the issue of “saleability,” “marketability,” “liquidity,” etc. The central auctioneer can discover and execute multipoint exchanges of arbitrary complexity. No numeraire is demanded. Clearly, this is a silly and impractical design not only in the real world, but in any reasonable virtual world. But this is beside the point.

    So what would break this equilibrium and create a monetary standard?

    The reason I have tried to eliminate transaction efficiency from this gedankenexperiment is that I do not believe transaction efficiency is the main cause of monetary standardization. In the presence of asymmetrical transaction inefficiency and self-reinforcing efficiency, and the absence of all other causes, the logic of Menger and Mises is correct – the cause is sufficient. But I don’t think it is necessary, and I think other causes are much more powerful.

    Recall the problem we are trying to solve. The problem is the superficially anomalous appearance of the demand for money. Whether it is little pieces of paper or discs of shiny, a
    ttractively colored metal, money would appear to command far more in exchange than its actual practical usefulness to anyone. Why do we trade real goods and services for this apparently useless junk? What’s up with that? What is the source of demand for money?

    Now review the nature of a monetary transaction. You have some good A which you exchange, in exchange E1 at time t1, for some good X. Then in exchange E2 at time t2, you exchange X for some other good B.

    This is a monetary transaction because you have a direct use for B, but no direct use for X. Therefore you have no concern for the quantity or nature of X. Your goal is only to maximize the quantity (or quality, or whatever), of B that you receive for your A.

    The interesting question is: what is the duration (t2 – t1)? Of course it is a continuum. But if this duration is short, we can characterize the purpose of the exchange as “trade.” If it is long, we can call it “hoarding.” (Some would say “saving,” but this word has the same kind of problem as “liquidity.”)

    If we look at the aggregate societal demand for goods of exchange (X), it is fairly easy to see that, in any normal human society where the world is not likely to end tomorrow, most of this demand will come from hoarders.

    From the perspective of the prospective hoarder, the choice of X is less likely to be controlled by the overhead of the exchanges E1 and E2. It is more likely to be controlled by the cost of storing X over a hoarding (eg, multiyear) timeframe, combined with the expected change over that timeframe of the exchange rate X:B.

    Back to our virtual world. Let’s introduce some reality.

    Consider Sven, a fisherman. He sails the virtual sea, he fishes the virtual fish. These fish are goods – perhaps you feed them to your pet orcs or whatever. But they are fish, so they do not keep. Sven cannot store his virtual fish in a pile in his virtual backyard.

    Therefore, if Sven wishes to hoard for his virtual retirement, the trivial option of X = A is not available to him.

    Here is our source of anomalous demand. Sven must choose some X to hoard, which is not a good that he would ordinarily demand. He has to sell his fish and exchange them for something. How does he choose that something? What is his strategy?

    Remember, he wants to maximize the exchange rate A:X:B, where A is fish. And – more crucially – he is not operating in a vacuum. He is not the only fisherman, or the only producer of nonstorable goods. He is operating in a world of many Svens. This must factor into his price projections.

    This is getting long. But my view is that if you look at this problem, you see four things.

    One is that Sven’s goal is best served by choosing the same X as everyone else. Otherwise, he is bidding up the price of an asset that is already fairly priced. If Sven and a few other Svens choose some X1, but the rest of the world chooses X2, X1 will appear to be an overpriced asset and its increase in price will not be sustainable. The mass exodus from X1 will become a self-fulfilling prophecy. There can be only one. (Bimetallic standards, for instance, are inherently unstable.)

    Two is that if society has not already made a clear choice of X, Sven should look for a storable good whose supply is rigid, or whose cost of production at least increases with the quantity already produced – because it is these goods whose buying power is most likely to be increased by an influx of hoarders, and thus will be favored in the Highlander-style competition.

    Three is that once X is chosen, it naturally becomes the basis for all monetary computation in this world. For example, the natural interest rate is defined in X.

    Four is that if some upstart Y intends to dethrone X, it can do so only by being a harder currency than X – by projecting a scenario in which the exchange rate X:Y will move sustainably over time in favor of Y. Once this consensus reaches a critical point, Y will become the new monetary standard and X will be valued solely on the basis of its utility. This transition is self-reinforcing and may be extremely turbulent.

    I’m making unsubstantiated assertions here, so I should stop. But I’d be curious to hear what your first reaction to this hypothesis might be – especially since it directly contradicts your assertion that the problem of preventing a star topology is manageable.

  18. MB – Ah, to sail the virtual seas… It is always a pleasure to read your parables.

    There is nothing that you say I disagree with, but it seems to me that you are not only describing the inevitability of a hard currency / star topology, but also its inevitable destruction. This isn’t Austrain economics, but German. On the day our virtual people are ejected, blinking and bewildered, from an eden of endless entactogenesis, a foreseeable process begins, whereby the industrious become hoarders, the early hoarders invent X and then hoard it, later industrious people have to pay ever greater rents to the idle descendants of early hoarders, who lay claim to most of current produce as well as the means of production by virtue of their near monopoly on X, until incentives to produce goods and services exchangeable for X are so badly distorted that the real economy fails, the hoarders learn that piles of X (magic sword though it may be) are insufficient to prevent the hordes that did not hoard from confiscating the X, perhaps along with their heads. Gold bugs (not mold bugs, mind you!) are fond of pointing out that fiat money systems are never stable. But they forget that hard money systems are not stable either. In real economies, distortions matter, in terms of production of genuine wealth foregone, and inequalities matter, in terms of the resentments and revolutions they give birth to. The worldwide transition from hard to paper money was not a technical error made by a few Keynesians. It was the product of a time where inequalities were large and property secure to those whose grandparents had hoarded it, and where communism, fascism, and eventually gold-confiscatory New Deals were sprouting up everywhere. Today’s gold bugs, nattering (as I do as well) about the Fed and moral hazard, are suffering from a serious case of the grass-is-always-greener.

    My contention is that a “state of nature” does indeed always pull towards a star topology, whether the central node is fiat or shiny, but that star topologies always collapse, usually very badly, as the productive masses are moved at some point to violently remedy the very real injustices perpetrated by either the hoarders or the printers.

    I have mixed feelings about your attitude that “everything has always been stolen, so let’s just ratify today’s theft and get on with it”. As a commenter in Setserland said, it’s hard to create expectations of future fairness by smoking one last cigarette of larceny and promising — promising — never to light up another. I agree with you that our approach should be forward-looking and not based on past grievance. But whether and how much of the produce either of hoarding or printing to ratify must take into account the distortions that inequalities impose on the real economy. If, post-flag day, goods and services are very disproportionately produced for enjoyment of yesterday’s forgiven malefactors, no problem will be solved, and yesterday’s malefactors will use their privilege to find some means of becoming tomorrow’s as well. I think that in the United States, for example, we are already well past tolerable levels of inequality (although we won’t notice until external producers stop subsidizing an apparent equality of consumption by giving us stuff for free). I think Americans underestimate the swiftness with which plenitude and stability can morph to civil strife and nonvirtual violence, once people can no longer afford all of the luxuries they had grown to think of as necessities.

    My scheme, conspiracy, proposal, conjecture is that humanity’s civilizing forces — social norms and institutions backed by the coercive power of the state — should be marshalled to maintain a desirable monetary arrangement that I concede would not be stable without those civilizing forces. I am more on the side of fiat money than hard money &mdash I fear private hoarders more than public printers, and in a world of fiat money, the damage done when control of the printing press falls too much in the hands of hedge-fund brats can always be undone by hyperinflation and taxation, which is gentler or at least better lubricated than outright confiscation. But I think we can do better than these cycles of monetary collapse, turbulence, and rebirth.

    The goals of a sane monetary system should include: 1) technocratic virtues &mdash minimal friction, easy exchange (as we have now with all our plastic and e-payments); 2) low to gently negative “real interest rates” to thoughtless general savers (these rates should always be much lower than the growth rate of the real economy); 3) significant positive “real” returns only to risk-taking investors, who make non-default wealth storage choices, and whose choices turn out to help support above-average real-economic production; 4) the ability (absent widespread catastrophe) to store wealth at near zero cost (but with near-zero real return) in terms of the goods and services one expects to consume in the future directly.

    Heterogenous money that includes claims against future nonstorable goods and services is, I think, the best way to achieve these virtues. I agree that such money is neither natural nor stable. Little interesting, good, or useful in the world is natural or stable. Human beings expend intellect and energy to design, produce, and maintain what is good. A monetary system should be no different.

    Lest this sound too idealistic and “great leap forward”-ish, one of the virtues of heterogenous money is that it can be implemented incrementally. Market-making robots, for real or virtual worlds, are tractable technological innovations. We are already very good at keeping track of and exchanging all kinds of virtual claims. Heterogenous money evolves spontaneously though transiently from homogenous fiat money during inflations. Institutions have already been constructed and are currently used to add artificial volatility to the exchange value of hoardable commodities in order to thwart the dynamic of self-catalyzing appreciation you describe that would lead to an inequitably distributed hard money. Reliable direct claims on the goods and services that define ones expected future consumption would provide ordinary savers with attractive alternatives to volatile scarce storables, especially if an expectation is established that claims whose exchange value grows egregiously inconsistent with present and future use value will experience artificial volatility, or in extreme cases more draconian interventions.

    (Of course this is dangerous, because states are corrupt, and it is hard to distinguish appreciation based on scarcity relative to expected future use-demand from appreciation based on self-reinforcing monetization. But there cannot be perfect algorithms for everything, societies will fail if their institutions are so corrupt as to be entirely unable to yield judgments that are at least good enough, and really bad states can always just kill us all and take our shit. All the alternatives on offer have weaknesses. I don’t claim this is a perfect solution, just better than either the monopoly-fiat-money or commodity-hard-money alternatives.)

    That’s enough for now. As you say, this is getting long. Though I am enjoying it.

  19. moldbug writes:


    It is a pleasure to see the perspective of 20th-century economics so elegantly and fairly stated.

    My feeling, however, is that this particular period of history was the most destructive since the aforementioned entactogenesis, and the fondness it still enjoys in our minds is a product of personal reminiscence and political propaganda, both of which can produce intense emotion but are not noted for secular constancy. And of course the impressive economic successes that occurred on this century’s watch can be attributed entirely to its technical prosperity, which did much to compensate for its monstrous crimes, but can scarcely be counted as an argument for or against it. 20th-century science and engineering was simply a continuation of the 19th century’s work, and there is very little factual basis for anyone to estimate how well this work would have proceeded had the aristocratic legal, political, and economic structures of the 19th century remained intact. Their destruction at the hands of an alliance of Gracchist politicians and moralizing intellectuals cannot be measured as either gain or loss by any objective method, though it is perhaps illustrative that the most successful states of the early 21st century, at least measured by standard indicators of government customer-service quality such as tax and crime rates, such as Dubai, Hong Kong, Switzerland, Andorra, Singapore, and so on, look a bit like “living fossils” of the 19th.

    In other words, I am not sure you have applied adequate consideration to the possibility that, contrary to the consensus perspective imparted I’m sure in your education, that the grass actually was greener.

    Because of my heterodox opinion of this period, I don’t consider the 20th century a good source of historical analogies, especially for anyone who means to praise some design feature that changed from the 19th to the 20th. It’s like saying that segmented addressing is great, because it worked so well on the 80286.

    Of course, there are certainly nontechnological aspects of 20th-century society that I think any reasonable person would conclude were improvements on the 19th. But to regard the general course of legal, economic, and political change over the last century as one of advance, even nonmonotonic advance, is a proposition with which it would be almost impossible to find any intellectual figure of 1907 to concur.

    Or suppose an alien mothership somehow could introduce 20th-century science and engineering into 1907. Pick up the contents of a university library and send them back in time. What do you think the result would be? Sure, there would be some dislocation, but I think it would be very difficult to argue for any result but an enormous economic boom.

    Now suppose 2007 society was reduced by some disaster to 1907 technology. Not only would we see a tremendous adjustment shock, there is simply no way that the world of 1907 could create and support the official-sector structures of 2007. Any political system in the world today, perhaps excluding the “living fossils” mentioned earlier, would be too heavyweight to function. Kind of like running Windows XP on a 286, in fact.

    I realize that this is not a point of view one hears a lot. Nor do I feel that this fact is entirely coincidental, although this is not by anyone’s direct or personal design. I think it is just a consequence of evolutionary selection in an extremely successful system.

    Three books I have read that have helped clarify my views on this subject are _Three New Deals_ by Wolfgang Schivelbusch, _As We Go Marching_ by John T. Flynn, and _On Power_ by Bertrand de Jouvenel – especially the last.

    So I think the large-picture perspective that it is difficult to see in the current political order of the world is that it is, as so many people with unfashionable prejudices have said, a system in secular decline.

    In other words, I am tempted to say, in a surely politically incorrect phrase, “who is this we?”

    I recently read a rather interesting book, Theodore H. White’s _Making of the President 1960_. This as it happens was written, in fact, in 1961, in other words before JFK’s assassination. It tells a story of the New Frontier that is the same in some ways as the one I grew up with, but not in others. And one thing that struck me in it, particularly, was White’s description of the Kennedy men as – I forget the exact words – basically, jaded and worn-out.

    Wow! I mean, this is the New Frontier we’re talking about, this according to NPR is the golden age of youthful brilliant energy. The Harvard faculty is seemingly descending en masse on the city by the river. Or at least this is what I was taught. But, when you read a contemporary description, it is actually cynical and disillusioned. What up wid dat?

    What’s up with it is that the real zenith of the American mandarin class – which is simply my definition of the kind of people who went to Brown with me – was actually around 1937. If that. The really essential documents are Bellamy’s _Looking Backward_ (1888) and House’s _Philip Dru_ (1912). But, since the window of personal memory only goes back so far, the aureate glow advances. Right now it is the civil rights movement and the resistance to Vietnam, a war which looks like it can be refought on an indefinite periodic basis with only minor variation.

    So here is my answer: at this point in the cycle, I do not believe that designing new solutions which are not stable is a useful exercise. Deck chairs, Titanic, etc. Since you know your approach is unstable I feel no need to elaborate on it.

    Another way to say this is that your answer sounds like an answer to the problem posed by John Rawls, of which I’m sure you are aware. This exercise is very useful in principle, but Rawls himself makes the mistake, which as one might expect has been generally copied by his disciples, of omitting that the most important and difficult aspect of the veil-of-ignorance problem is that of selecting, not from all possible organizations of society, but all possible organizations that can be formed and maintain themselves indefinitely without divine intervention.

    In other words, Rawls’ problem frequently serves as a surreptitious mechanism by which to reinject a God, and a strangely Christian God at that, into the problem of constituting legal systems. Which is no surprise, because the tradition which gave us Rawls, John Dewey, and similar thinkers, is a thoroughly Christian tradition. It traces back to the Unitarianism of John Adams and William Ellery Channing, forward from them to Emerson and the transcendentalist statism of Lincoln and the Bellamies, and of course earlier back to Jonathan Edwards, Oliver Cromwell, and Jean Calvin, and it is today epitomized by such influential and well-established institutions as Harvard, the United Church of Christ, Greenpeace, National Public Radio, etc, etc, etc.

    (Not that we didn’t already have a constitution. But strangely, I don’t recall it mentioning Mr. Rawls. Or the Federal Reserve, for that matter.)

    The fact that, on the basis of what any nontheistic person can only consider a minor doctrinal detail, this tradition no longer calls itself Christian, is most simply explained as a parasitic mutation strategy in order to evade the separation of church and state, and establish itself as the official creed of a monotonically expanding civil service and educational system. Historical precedents for such a strategy are nothing if not abundant.

    Hopefully all of this won’t someday leave me wondering what, exactly, I’m doing in an adult education facility in Alaska. But if you’re wondering why I start from different premises than you, well, this is why.

  20. moldbug writes:

    And still a third way to state this objection, which is probably as blunt as I can make it, is that in my opinion, democracy is a fundamentally criminal ideology. I am a formalist, I believe in the rule of law, and this is a system of affairs that the principle of popular sovereignty was born and bound to undermine and destroy, as did its logical predecessor the divine right of kings. (If this argument makes no sense to you, it is expanded at great length in Jouvenel.)

    It cannot be denied that the variety of inflationary currency you favor is a source of central revenue, and if that central revenue is given, as you presume it will be, to a good cause, the result is by definition good.

    The motivation and effect of dilution, is as you say, to capture the savings (in the sense of hoarded money) of the citizens who make the mistake of exchanging the products of their labor for the currency on which you are the monopoly producer. Va bene. They paid their money and they took their choice, and it is indeed possible that you, the cultured and benevolent official, have some better cause to which to direct their wealth than they, the consumer of big-screen TVs, tacky Florida condos and jacked-up pickup trucks.

    But what start as good causes do not always stay that way, and it is difficult to reroute their revenue streams. And I believe you will find dilution ethically (though not always politically) inferior to taxation in every possible respect. Götz Aly’s new _Hitler’s Beneficiaries_ is an excellent illustration of what these tactics look like when you really apply them thoroughly and without compunction, and the parallels to the dollar system are extremely troubling.

    Since in a monetary standard with negligible intrinsic value, unevenly redistributed dilution can be defined as a combination of neutral redenomination (replacing not only every dollar, but every contract written in dollars, with X dollars), selective confiscation, and psychological distortion (reprogramming everyone’s mind to think in terms of this year’s “real dollar”), you cannot ethically justify dilution unless you can justify this combination. Redenomination is no problem and sometimes even fun, and an adequate public relations effort can minimize the level of psychological distortion. But you still have to answer the question of why you prefer selective confiscation to be surreptitious, rather than explicit. This is not a position that even Rousseau could defend, or even in fact would care to.

  21. moldbug writes:

    And to be even still more pragmatic:

    You mention civil strife as a consequence of “unfairly” distributed resources.

    The case is exactly the converse. Civil strife is the consequence of the inability of a legal system in which property rights are defined as a function of “fairness,” rather than as formalized artifacts of history, to prevent violent disputes. This is ensured by the fact that there is no way to construct any standardized definition of “fair” property rights, on which all can agree and none can, cloaking himself in apparent and usually quite real sincerity, adopt the tools of physical or political violence to dispute. Or at least no such definition has been seen in the wild as far as I am aware, and it has not been for lack of attempts. This is especially compelling considering as exactly this concept of egalitarian distribution has been readily apparent in the religion which the West has followed for most of the last two millennia.

    Gracchism has always been, and always will be, presented as the cure for civil strife. Pay off the mob and it will quiet itself. If you apply this approach to the problem of impecunious and underprivileged youth at one’s adolescent educational institution, who at least in my experience can be quite insistent on the injustice of the structure of wealth in society, you will find yourself embarked on a course of consistent prandial asceticism, not to mention daily humiliation, that can be quite difficult to reverse. Or indeed so it proved for Rome, and it is hard to see how any different lesson should apply for us.

    In other words, civil strife is not the result of the rich denying their goods to the poor. It is the result of some of the rich offering some of their goods to some of the poor, in order to shape them into a political weapon. Gracchism, or Caesarism, or welfare economics, or whatever you want to call it, is a perfect iatrogenic therapy. It may not be necessary, but it is quite sufficient, to cause the disease it purports to cure. This keeps the doctor coming back, though of course he cannot and will never put it to himself in this way.

    Am I getting through here, or am I just belaboring a point which you have already considered and rejected? The effort does seem worthwhile to me, but perhaps I’d better stop here.

  22. MB — Your premises are not all that different, but I feel misconstrued. I didn’t mean to supply a normative justification for the politics and economics of the 20th century, only to describe it (accurately or not). I certainly made no reference to any boom. I do think there was a collapse of hard money in the early twentieth century, and that collapse was not incidental or exogenous, but arose because single-commodity hard money systems, though they do arise spontaneously, contain the seeds of their own destruction. To say that something sucks is not to suggest I like the replacement. Indeed, I believe the fiat money systems that followed are similarly unstable. As to the system I proposed, though I did concede that it would not be stable if certain supportive institutions failed to arise, I did claim that those supportive institutions were plausible, even incrementally accretable, in this real world, and that with supportive institutions in place, the proposal would be stable. You may of course disagree with the advisability, plausibility, or practicability of my proposal, but as stated, this was not a best-of-all-possible worlds exercise with no thought given from how we get to there from here.

    I’m agnostic to questions of whether a 19th century social orders were superior or inferior to 21st C institutions in some aesthetic/cultural way, given the technological differences in the possible. I am interested in questions of what forms of social organization will prove to be successful and maintainable as we roll towards the 22nd Century, and think that my own countrymen are so smug in presuming that ours is the best pattern for a society that we’re likely to doom our institutions out of sheer complacency.

    The 20th C was indisputably the bloodiest in human history in absolute terms, and I suspect in relative terms as well. There is little in 20th C economics I’m inclined to celebrate. Economic institutions and political institutions are of the same cloth and interwoven, and I blame failures of 20th C economics as much as any other cause for all that blood. These things we talk about, arcane videogame thought experiments, the political economy of overextended credit in the context of fiat money and competition for control of the printing press, these are urgent things from my perspective. The institutions of the 20th C have failed so completely that I fear we are doomed to repeat that century’s bloodiness in this the 21st. I do think the economics of the 19th C failed as completely as that of the 20th C, and am skeptical that a reversion to an older pattern will do anything other than restart a cycle.

    My contention is that a monetary “star topology”, whether hard or fiat, represents both a natural attractor (without countervailing institutions, a monopoly currency arises spontaneously) and an unstable state. I propose an alternative that is speculative, but is intended to be incrementally achievable via augmentation and reform of existing institutions.

    You’ve argued that monopoly hard money arises spontaneously, through a dynamic cannot easily be thwarted. I agree (but suggest it is worth trying the thwart regardless). But do you disagree with my rejoinder that hard money systems are themselves unstable?

    More interestingly, you promised if I showed you mine, you’d show me yours. I’ve showed you mine, and it seems you don’t like it much. What is your proposal for a monetary reconstruction (you hinted it wasn’t entirely mold-based)? Or must I wait for a paper to be published at

    BTW, I think our gut level predilections are more similar than dissimilar. I’m guessing you did not like my statement that, given a choice between hard and fiat, I’d pick fiat. But I would emphasize that I take that to be a choice along the lines of, “Which would you prefer, the guillotine or lethal injection?” I do have a preference, but I’d prefer to change the options. I suspect you also dislike, because I do as well, the requirement in my proposal for coercive meddling by states to ensure the stability of a system that would not be stable without meddling. You and I are in broad agreement, I think, about how destructive that sort of meddling is and has been under the current arrangement. I would prefer a solution that left no room for states to err in some well connected party’s interest. It’s just that, thus far, I’ve been unable to come up with one, and so instead I posit that a regime in which states meddle only to undermine burgeoning monopoly currencies will be less destructive than one in which states meddle to support the one they control, and undermine all the others.

    You wrote previously “My views may occasionally sound libertarian, but I am not a libertarian at all. I think of law as a practical rather than an ethical instrument. Property rights, to me, are just a way to keep people from fighting over property. If a better way could be devised – some drug, for example, that we could take, that would all make us love each other, and wouldn’t wear off the way MDMA does – I’d be first in line for the injection.” I’m mostly with you here. My instincts are libertarian, and “folk austrian”, in the sense that I think economic systems represent decision systems with respect to the organization of human production, and that most interventions bias the process and cause non-optimal choices. Both my gut and mind suggest that proposals for state intervention must overcome a very large presumption that any possible benefit would be undermined by error or corruption. But my stance is consequentialist and pragmatic, not based in law or principle. If Federal-Reserve-managed money were working at guiding the economic behavior of Americans to some approximation of optimal production and distribution, I’d be all for it, even though the Fed is nowhere in the Constitution. But it’s not working. Central bank managed money is leading to grave and growing errors in the organization of economic activity, in the US and throughout the world. I’m open to any alternative, sure, even a pharmaceutical solution would be fine, but there has to be an alternative. So far, I think that mere reversion to hard money would bring with it as much hazard as it would eliminate. But perhaps you can persuade me otherwise. Or perhaps you have something else in mind?

    I assure you, that should you find yourself someday in an Alaskan reeducation camp, I’ll be clamped into the electroshock therapy apparatus next to yours. Though we might argue over the merits and demerits of commodity money or the feasability of “mesh network” currencies, the new technocracy of economists would see to both of our post-orbital lobotomies, and for our own good.

    Anyway, I look forward both to your reconstitution of the monetary system, and your new programming language.

  23. MB — Oops! Just posted after seeing only the first of your series of comments! I guess I’ll read the rest now…

  24. Okay — First, a clarification. I am skeptical of states, consider dilution of a monopoly currency to be a form of theft, confiscation, or surreptitious taxation. It is not my proposal to praise monopoly fiat currencies, but to bury them. I did state that dilution is “gentler or at least better lubricated than outright confiscation.” I hope the implication of getting fucked, and not in a good way, was apparent here.

    We differ because I similarly think that rents extracted by early hoarders for the mere privilege of enabling others’ to exchange is also problematic.

    My preference for fiat over hard currency, if those are the only choices, is not normative, but pragmatic. It is my claim that both approaches to money are not stable, that they lead inevitably to turbulent periods of recalibration, and that the very surreptitiousness of confiscation under a fiat regime permits those recalibrations to be less bloody. I agree, in abstract moral terms, that confiscation is bad, and that it would be better to end confiscation than to hide it. But, as you suggest, we must agree about what constitutes confiscation in order to agree on a solution. I think you see (correctly) confiscation by the state (whether surreptitious or overt), but you fail to see confiscation by rentiers whose only claim to the produce of others is incumbency in the possession of a conventional means of exchange. You might, as a matter of law, decide one of these to be confiscation, and the other not to be. But that begs the question of how and why we define the laws. I would define both dilution of fiat currencies and the profits accruing to holders of commodity currency due to an increase in the demand for exchange as confiscation. I would define a law that eliminates both forms.

    Pre 20th C economics took dilution as illegitimate but thought rents accruing to gold-holders were perfectly fine. 20th C economics sees dilution as legitimate and even necessary, for many of the reasons you rip to shreds, but abhors rent-seeking by hoarders of the medium of exchange. I reject both, sure, due to my own notions of what is “fair”, but also based on an empirical claim that systems which permit either dilutions in favor of the state (and its friends), or which permit ever-increasing rents to holders of specie, are doomed to spectacular failure. I think that, ideally, we should define a legal regime that enables all exchange to occur that would occur if barter were frictionless, without rents being paid to third-parties, whether government diluters or private rentiers.

    I do not think you can prevent civil strife merely by relentless indoctrination to the notion that some legal regime is legitimate, and that alternatives are “Gracchism”. Nor do I think serious civil strife is inevitable under any non-equal regime, because any status quo is imperfect and arguable, and losers will always claim they have been shafted. It is my claim &mdash empirical, not philosophical — that notions of law and legitimacy do matter, and are effective, to a degree, but that there are situations and circumstances where an intuitive egregiousness in the distribution of fruits and labors will overwhelm any philosophic claim a social order has to legitimacy. The boundaries are fuzzy and probabilistic. I’ll concede that a culture of “Gracchism” might increase the likelihood of strife and disorder under a given set of circumstances, and that a culture of legitimate authority will widen the range of outcomes a society can permit without succumbing to rage and revolution. But I submit that in all circumstances, there are levels of intuitive maldistribution that will either foment revolution or else require great brutality to sustain. My claim is that both fiat regimes and hard money regimes lead to unsustainable maldistributions, because both create classes of capable rent-seekers who progressively lay claim to ever more of an economy’s production, without check or limit.

    I don’t deny that regimes that might otherwise survive a while longer are sometimes done in by “a perfect iatrogenic therapy… the rich offering some of their goods to some of the poor, in order to shape them into a political weapon.” But I don’t believe the converse, that even patently illegitimate wealth (by ill-defined, intuitive standards) would survive if only people would refrain from foolish demagoguery.

    I do not favor inflationary currency, and I do not support the use of dilution as a source of revenues for the state, or bank owners and other financiers. I’m under no illusion that government revenue, whether extracted via taxation or dilution, is money to a good cause. I am not any kind of Rousseauian. I see nothing definitionally or metaphysically righteous about state action, and see it as destructive, corrupt, and misguided more often than not.

    I think that wealth should, as much as possible, be tethered to production, and that our laws and institutions should be designed with that outcome as the goal. Like “fairness”, what I’ve stated is a bromide that cannot be pinned down and defined. In the real world, we often have to work for goals we cannot quite define. We have to make judgments.

    You agree with me, I think, about the badness of dilution, at least on moral grounds. (I don’t know if you believe the quasimarxist “contains the seeds of its own destruction” claim.) You don’t agree with me, it seems, about the badness of people collecting a continuing share of other people’s produce by virtue of holding a monopoly hard money, whose tendency to appreciate in value is certain and known to all, whose idiosyncratic allocation at any moment distorts rather than contributes to the efficiency of production, and which could simply cease to exist without affecting the productive capacity of the economy, so long as any other hard-to-produce-and-otherwise-useless commodity, by any random procedure, is nominated to take over the throne.

    I do see badness in this second case, and think the best resolution is to create uncertainty and heterogeneity with respect to what, precisely, qualifies as money, to protect private parties that wish to enjoy a private exchange from victimization by either the owners or printers of what is fundamentally an unnecessary intermediary.

    My wife hates me now. You’ve inspired me to fervor, I am punching keys loudly, and it is enough to annoy her to wakefulness at 4 a.m. Any continuation of this unmediated exchange will have to wait until the morrow. G’night!

  25. moldbug writes:


    While it is always a pleasure to assault on a broad front and meet with determined resistance along its full length, rather than the typical complaint that any system of thought which is both unconventional and general violates some intellectual Geneva Convention which requires us all to be either pragmatists, specialists, or preferably both, please be assured that any damage to either your keyboard or your marriage is the farthest thing from my mind. I speak as one who has actually worn holes into more than one keycap, and as for the tolerance of my girlfriend, sometimes it simply passeth understanding.

    I am afraid I have probably mischaracterized some of your opinions by identifying them with the closest conventional equivalent, a tactic that drives me into a white-hot rage of keyboard-destroying fury when it’s used on me, and I apologize for this as well. Please be assured that any such offense is unintended. You have clarified a bit and I will try to fill in the gaps.

    To start with, let’s look at your “quasimarxist” claim that something about the legal, social, or political nature of 19th-century Manchester liberalism was inherently unstable. An instability which you compare to the point, on which we agree, that the moneyless economy you propose is unstable.

    Well, clearly something about the system of the 19th century was unstable, because it did, in fact, collapse. Moreover, its collapse was predicted by numerous 19th-century writers whose worst fears were realized to an almost eerie precision, including Tocqueville, Burckhardt, Spencer, Acton, etc. So historical determinism aside, there is clearly some pattern in these events.

    This is not in any sense a scientific question. It cannot be settled by the presentation of evidence. It is a matter of perspective. You see one pattern, I see another.

    My interpretation of these events generally follows, as I’ve said, Bertrand de Jouvenel, though you can get much the same story from Erik von Kuehnelt-Leddihn, Albert Jay Nock, (later) James Burnham, etc. But Jouvenel attacks the problem in an extremely systematic and effective way without a trace of romanticism or Hegelian mystery, and I like his presentation the best.

    The tragedy of the 20th century, per Jouvenel, is a fragment of a trend that is continuous and almost, if not entirely, monotonic: the decline of law and the rise of Power. (Jouvenel generally capitalizes the word “Power,” which is a little Continental for my taste, but can be clarifying, so I will follow his usage.)

    All the centralist and demotist tendencies of the 20th can be seen, if you adjust your spectacles along the law-Power axis, not only in the 19th century, but also in the preceding four or five.

    To Jouvenel, for instance, the French Revolution was not the overthrow of the absolutist state created by Louis XIV, but its natural culmination. The First Empire was not a perverse and ironic reversal of the ideals of 1789, but their logical consequence.

    Huh? How does this make any sense at all? What is the difference between law and Power? Surely law, after all, is not self-enforcing? Surely it depends on Power? Aren’t they just the same thing?

    This is probably not the way Jouvenel would put it, but you can think of law as a special case of Power. Law is the subset of Power which is simple and well-defined. When law becomes ambiguous, it becomes indistinguishable from Power. And beyond a certain threshold, complexity is ambiguity.

    Simplicity is the cure for violence. For example, if we postulate an imaginary oracle that could predict the outcome of any battle, we could eliminate war. The predicted loser would have no incentive but to concede to the demands of the predicted winner. But since there is no such oracle, the only way to avert violence is to define an algorithm simple enough for both sides and an independent third party to execute with minimal chance of disagreement, and endow that third party with whatever Power it needs to enforce that result – and, ideally, no more.

    The word “violence” demands clarification. Violence does not always involve bloodshed. The point of violence is for side A to use “any means necessary” to convince side B that resisting some outcome is counterproductive. Thus we consider robbery a violent crime, even if no one is hurt.

    Similarly, a battle may be resolved by agreeing to fight not with real guns but with paintball guns, and respecting the outcome. Or by setting rules that reduce but do not eliminate the physical damage of battle, and have no significant effect on the result. Symbolic and restricted violence of this sort is common in primitive humans, and indeed is the general rule in the animal kingdom. In general the goal of symbolic violence is to figure out who would win in an unrestricted fight to the death. As long as the results of these processes are identical, there is no incentive for either party to deviate from the symbolic convention. And even if they are not quite identical, the parallel may be good enough.

    Another symbolic way to resolve a battle, for instance, might be to simply count the number of soldiers who show up on each side. We see here the roots of democracy, which once factions emerge – as, the Federalists notwithstanding, they always do – is a sort of regularized, symbolic civil war. It is no surprise that when you try to impose democracy on a country in which some faction believes that a real civil war will produce better results, the symbolic restraints tend to evaporate.

    Therefore, when we define Power as ambiguity and ambiguity as complexity, we can consider the decline from the elaborate precision of medieval law, with its principle of rex sub lege, to 20th-century democracy and Holmesian “legal realism,” as a self-reinforcing complexity collapse – a process surely not unfamiliar to any programmer.

    The interesting question, the “quasimarxist” claim you raise, is: what was the cause of this collapse? And did it have anything to do with hard money, laissez-faire, etc, etc?

    We can certainly identify the currents of demotist faith behind the disaster. I have done a little of this above. I would argue that demotism (ie, leftism) is an essential part of Christianity, maybe even its most important element. To me leftists seem if anything much more Christian than the traditionalists who call themselves by that name today.

    But this doesn’t really answer the question, because it doesn’t tell us why this ideology became so popular at this time. Certainly, as the history of Catholicism shows, Christian fervor is also quite compatible with extremely aristocratic and hieratic institutions.

    I have spent far too much time around leftists to doubt their sincerity. However, human psychology is very good at rationalizing its interests. It cannot be denied, for example, that the giant flocks of young aristocrats who have decided that NGO work is the capstone of their pyramid of needs tend to be deeply sincere about the causes they have chosen. But it also cannot be denied that these institutions of public service are the most obvious path to status and Power in today’s world, much as the Jesuits or Dominicans might have been in another century, and the pattern of a society in which young aristocrats put most of their non-hedonic energies not into producing goods and services, but into scheming for Power, is quite familiar from history.

    In other words, Gracchism attracted ambitious aristos because this strategy seemed likely to succeed, as indeed it did. If it had been clear that the assault on law would not succeed, it might not have happened at all. Certainly the Roman Republic had seen many a year before the Gracchi.

    This is a classic description of a feedback loop. Power, in my view, behaves much the way that bad code behaves in a software system. The “Big Ball of Mud” patte
    rn is always out there. There are many ways to the top of this mountain. And the only way to prevent it is to be absolutely vigilant, even paranoid, in rooting out complexity – Power – wherever it is found.

    For an example, let’s look at leftism a little more closely. As I have said, once you replace “justice” with “social justice,” Power’s nose is in the tent. There is no precise definition of “fairness” or “equality,” and that is what kills you.

    My contention is that once private charity becomes official policy, the battle is lost. There is no point of principle on which the decay of law into Power can be resisted, because you have already accepted the principle that whatever the People can extract, they will come to feel they deserve. (Spencer in _The Man Versus The State_ is particularly good on this.)

    Can this process be reversed? Let’s imagine restoring law to a welfare state.

    My position, as I’ve said, is Jouvenel’s: that the natural course of power is growth. When we aim to not only check but actually reverse this growth, we are trying to solve a very difficult problem. And beggars cannot be choosers – no pun intended.

    Therefore, when we try to formalize ambiguous power dynamics and return to a state of law, it is a very bad idea to try at the same time to redistribute scarce goods in a way that strikes us as superior for some abstract reason. Rather, we should take the approach that is most likely to succeed. Which is clearly an approach that confirms the ownership of everyone to whatever goods he or she has, in the period of violence, managed to lay his hands on.

    So the way to reformalize a welfare state is not to end welfare, and not to end the taxation that funds it. It is to consider that the welfare recipient has, by the use of power, acquired a claim on the state. If we try to cancel that claim, said recipient, and (more importantly) his or her political patrons, become our enemies, and we make the establishment of law more difficult. Which is not a problem we need to have.

    The state, freed of its mystical trappings, be they theistic, demotistic, or whatever, is simply a large corporation. So the welfare recipient gets a share in that corporation, or possibly a bond issued by it.

    What assets does the corporation own in order to make this claim valuable? Well, national forests, gold reserves, and so on, of course. But its most valuable assets are its claims on its citizens, who are simply corporate serfs. (You are a serf if someone else has a permanent lien on your income. For example, in the 19th century before the liberation, an effective enforcement apparatus allowed Russian landlords to let their serfs move to the cities, while still receiving a predefined fraction of the proceeds of their labor.)

    From this we see my own personal preferred path for the dissolution of the post-1945 system of government, which is not to destroy the existing states, but to demystify them, and migrate their ownership and management toward well-understood models of agent-principal separation. They can then be expected to dismantle and decentralize themselves simply for efficiency.

    I think the most efficient size of a government is a single city, and I think a global system of privately owned cities, competing on the basis of the usual customer service criteria, would be extremely stable and provide very good service.

    Especially if your citizens are mobile and can skedaddle, aggressive violence is unlikely to be profitable for any firm as a whole. My view is that aggression is common in state actors because states historically have not enjoyed a clean separation between agent and principal. Instead, they have been run for the benefit of their employees, invariably resulting on competition between employees for the unformalized distribution of the profits (something you can see all the time on “The Sopranos”), and aggressive war certainly brings glory and power to those employees who specialize in it. For the state as profit-making company, in theory there may be some paths by which aggression could increase the dividend rate, but in practice this is likely to be a very dubious business plan. (This is also why Europe in the age of monarchy, in which the state was to some extent a family-owned corporation, tended to be less violent – Hans-Hermann Hoppe is very good on this point.)

    Defensive and deterrent violence, however, is a necessity for corporate survival. And preventing either the corporation’s employees or its customers from seizing its assets and redirecting their profits away from the legitimate owners is a sine qua non, and hence a core competence, of the profitable state. The sort of salami tactics that we saw in the democratic degringolade are unlikely to be tolerated – Power will not gain its deadly foothold.

    I hope this answers your question as to why the 19th-century system of government was unstable and collapsed, and how a restoration of law can be made stable and effective.

    (I can’t tell whether or not you are expressing agreement with the conventional view that hard-money finance collapsed – ie, in 1929 – as a result of its own intrinsic instabilities. If so, I recommend Rothbard’s _America’s Great Depression_. Short summary: it didn’t jump, it was pushed. 1920s economics were an intermediate point between prewar economics and what we have today, and the failure to restore the Victorian financial system was exactly that, a failure. Indeed, 19th century American and British finance was itself constantly battling with fractional-reserve dilution, and this period should by no means be regarded as a paradise of hard money. If you want a hard target to practice on, the Bank of Amsterdam is the most recent clearing system with zero dilution.)

    Now when we turn to your monetary proposal, we see, I think, an entirely different form of instability.

    As I’ve explained, even completely eliminating transaction costs does not by any means eliminate the pressure for a star topology, which is primarily the result of the fact that humans want to save and not all goods are saveable.

    Intuitively, especially if you know something about investment, you would expect this “monetary energy” to distribute itself evenly across all assets, increasing the price of each by the same percentage. But when you look at the game theory of it, this is simply not how it works. Instead, the only equilibrium strategy is for all savers to pile into a single scarce good.

    Preventing this involves an enormous game of Whack-A-Mole. This game is being played as we speak, and it involves financial tricks, legal discrimination, and the management of public opinion. My view is that the first and the third are not very sustainable, and probably reliance on the second will have to increase. But in any case, here is Power aplenty and in all directions. Lower the bar and throw your citizens in the slammer if they keep gold, and watch them stock up on, say, palladium. (Or, more to the point, real estate.)

    Even today financial authorities are constantly trying to suppress asset bubbles. And this is with an existing fiat monetary system, that even though it happens to be diluting at 10 or 15% a year, still basically kind of works. It strikes me as very implausible that money – that is, a single overvalued asset – can be prevented from emerging in a moneyless society. Even with all the Power you are ready to apply to the task.

    And what is the end of this? The reason I assumed you favor dilution is that you seem so clearly opposed to saving. This is how the 20th century state funded itself – by confiscating savings, typically through dilution. If your goal is not to do so, what is the source of your opposition?

    “Distorts rather than contributes to the efficiency of production.” This is perhaps the center of your argument, if I may be so bold, and it is certainly
    familiar from the logic of many who have found money to be the root of evil.

    If there is anything I’ve learned from Austrian economics, it is that here, perhaps, is one place that the hackneyed analogy of Einsteinian relativity is actually useful. I believe that the central fallacy of 20th-century economics, really most due to Fisher, is that since monetary effects on pricing are unpredictable, since the aggregate demand for savings affects all prices, there must be some other scale that is more reliable (eg, Fisher’s price indexes).

    The history of 20th-century fiat economics is the history of the search for this fixed reference frame, this luminiferous ether. It simply does not exist. The only way to compute profitability is in terms of money itself. Any system that attempts to compensate for changes in the demand for savings, which is the demand for money, has no objective basis for calculation, and simply introduces additional sources of distortion into the signal. The composition of price indexes, for example, is arbitrary and meaningless information. It is little wonder that there was never any significant demand for price-indexed contracts.

    Money can always be priced precisely in terms of itself, and it does not appreciate. Defining a holder of money as a receiver of rents is a distortion of the concept of rent. Nor are declining prices, a normal phenomenon in the presence of increasing productivity, any impediment to moneylending. Money will be lent if it can be used to generate a positive risk-adjusted return. If it can’t, society clearly has no use for capital, which is surely not impossible in theory – maybe we really all can become Zen monks.

    So if you conclude that money is necessary, and that it must be restricted, you are left with two choices. You can use a commodity that is restricted by natural forces, or you can use one that is restricted by political forces – that is, Power.

    I favor the former because of my belief in the tendency of Power to expand itself. Because it can do so, and so easily, by shirking its duty to prevent creation of new money, it cannot be trusted with this duty.

    It is remarkable that there is a book which almost every Western intellectual has read – Tolkien’s _Lord of the Rings_ – whose core theme is precisely this. What is the Ring, but the Power of monetary dilution? Of course, I’m sure Tolkien had no such specific analogy in mind, but that is the greatness of great fiction. Or at least it is supposed to be. Yet the lesson of Boromir strikes me as if anything less understood than ever. (Maybe it’s just because the movies sucked so much.)

    Anyway. You ask: if not precious metals, then what?

    Well, in practice, I do think it will be precious metals. But, as I said, I do have some other ideas in this direction. However, I would like to keep them off port 80 for at least the present time. If you’re still interested, we can take this to port 25…

  26. MB — Fervor does not mean offense, and I certainly don’t hold you responsible for any insomnia I’ve inflicted on my dear wife! I just needed a way to end abruptly.

    ‘Tis true that I didn’t like the identifying of my views with a conventional set of views that I actively oppose, though I do understand how you got the wrong impression. But offense was not taken, only ruthlessness (and repetitiveness) in rebuttal.

    Moving on then.

    As always, your excursions are winding and the scenery is dense, lovely, and a bit strange. I must start by pointing out a danger, that we may frequently talk past one another because our definitions are so different. I enjoy your description of the relationship between law and Power, but I wouldn’t adopt your definitions. In danger always of mischaracterizing myself, it seems that you (after Jouvenal?) view law as a good thing, Power (as you’ve defined it, implying the absence of effective law) as bad, and complexity as an almost certain sign of the devolution of the former into the latter. Of course you know your audience, when you write of the big-ball-of-mud-problem. But there is a rejoinder, that a system ought be described in the simplest manner possible, but no simpler than is possible, given the complexity of an application domain. A system which for elegance of definition fails to exploit important potential opportunities in the application space is as surely deficient as a big-ball-of-mud that exploits every possible opportunity, but only for the microsecond before something in the world changes. We are left, in programming as in life, with directly opposing aphorisms, and we must come to terms with the possibility or the impossibility of balance.

    In practice, we are not very far apart. I smiled last night, when I read your description of the city-state-whose-citizens-are-customers as a promising approach to organizing human affairs. That is an idea I have also considered, favorably. Ditto for the notion that formal, contractual norms of customer / shareholder / member of a formal organization ought replace emotive / tribal / Rousseau-ish notions of “citizen” whose duties to the collective that in practice is the state are ill-defined and potentially inexhaustible, the discharge of which duties is often an evil masquerading as a good. We are even, in practice, very near one another on the question of money. My own wealth, such as it is, is largely held in a mix of claims (allocated and unallocated) against the “tawdry relic”. I do not think that gold is good money for the 21st C, but I do think that the latest experiment with monopoly fiat money is almost over, and that between the bloody struggle and the inevitable socialist paradise, there will be a moment for the dictatorship of the glittery stuff. Plus, at the moment, I consider it noble to root for gold, because (with reference again to Marxist metaphors that only serve to discredit the causes I support), today’s fiat-based prosperity is obscuring an awful sort of decay, which dynamic must be stopped as soon as possible, so “the worse, the better”, and gold has always been Kryptonite to the money printers.

    Putting aside the large definitional and philosophical issues, I think our primary difference lies here: “Defining a holder of money as a receiver of rents is a distortion of the concept of rent. Nor are declining prices, a normal phenomenon in the presence of increasing productivity, any impediment to moneylending.”

    Here is how I think about it: Mediating exchange is a service, and a very important service at that. Inventing institutions to mediate exchange is as essential to any society as coming up with forms of shelter that resist the wind. Like for shelter, there is a large space of possible solutions to a pre-existing problem. Human societies have used with success, at least for a while, commodity money of innumerable sorts, money defined as full or fractional claims against those commodities, government fiat money, plus myriad banking, lending, and transfer networks built around all of those monies, and to mention every variety of brokerage.

    Like any other necessary service, human welfare is best served when exchange service is provided as cheaply as possible in terms of the resources humanity at large finds itself endowed with, natural resources, but especially the opportunity costs of human time and effort. Just like the invention and reduction to practice of easier-to-construct-yet-more-resistant-and-comfortable forms of shelter improve human welfare, so too does human welfare improve when the service of mediating exchange can be performed more cheaply.

    The enemy of improvement is monopoly, public or private, whether naturally evolved or manufactured through state coercion. Suppose on a certain windy isle, it is discovered that the wood of a certain tree will resist windstorms so much better than any other known material that it would be foolish to construct homes of anything else. Suppose further that this tree can be made to grow on the property of a only a very few of the islanders. The discovery is a windfall for the well-placed property-owners, and bully for them. They should receive their windfall, for confiscating it would kill any incentive to foresee and store (hoard) what is not recognized to be valuable but in fact really is.

    I suspect we part company philosophically here. You, I am guessing, would say that the windfall is deserved as a matter of a law that can be unambiguously defined and which therefore demands adherence, rather than as a matter of consequences and arguments about time inconsistency.

    But to my world view, there is a problem here. While the tree-land-owners deserve a large windfall, should they form a cartel and resolve to sell trees in perpetuity with perfect price discrimination at the maximum price each of their fellows could be induced “voluntarily” to pay, and to not sell to any whom they might of developing an alternative material (wind-resistant laboratories being indispensable to the project of developing a replacement), I would support state action to coercively break the cartel. I see the exercise of (private, legitimately acquired) market power on the part of the landholders to be a form of Power, with that capital P that means “bad”. Similarly, if the state were to nationalize the productive land, denying the property owners their windfall entirely and declaring the trees “for the common use of the people” (defined inevitably the governing faction and its friends), that would be bad Power too. The welfare-enhancing potential of competition and innovation are simply too important, in my view, to be prevented by monopoly, even in the very rare cases where monopolies are privately and legitimately acquired, and long-term sustainable. (I should emphasize that my example is tortured because these cases are rare. Usually collaboration of the state and/or market-manipulative gamesmanship is involved in the acquisition or maintenance of a monopoly.)

    Monies, commodity, fiat, or otherwise, are inputs into the provision of exchange services. In a competitive environment, we would expect the cost of those services to fall over time (where cost is defined in terms of basic, nondepreciating commodities like human time at a given skill levels, or gold). But that doesn’t happen when money is either finite and privately owned or infinite and publicly owned. In the former, “hard money” case, as you’ve described, the money-commodity, once earned would never be exchanged by risk-averse rational actors who lack special information (regarding the prospects of a potential investment). Why not? Because it is expected to increase in value faster than the growth rate of the overall economy. Economic growth implies at least proportional growth in the need for exchange services, which implies growth in the demand for money, which given a nearly fixed supply, implies higher prices for money.

    You note that as productivity increases, almost everything producible sho
    get cheaper in terms of anything whose quantity is nearly fixed and whose intrinsic desirability does not much change. That’s true, natural, correct. But suppose there are two distinct metals, both of which have nearly identical uses, but one of which is used as money. In a growing economy with technological progress, the value of the both metals in terms of services and manufactured goods will rise. But the price of the money-metal (whether to purchase or to borrow) will rise much more than the price of its nonmonetary putative twin. The economy demands exchange, and money-metal holders will extract whatever the market will bear for the use of their property. And this is before the momentum trading dynamic you describe by “savers” (more below) creates increasing speculative demand for the money beyond the ever growing exchange-related demand! Conceptually the difference in price appreciation between our two metals represent undeserved rent to money-holders. It is as though they have won a lottery by picking the “good” metal. Which metal becomes money may not be random. Some smart people might be more skilled at guessing which metal would be blessed, and some skillful people might succeed at biasing the collective choice “their way”. Rent-seeking is called rent-seeking not because rent-seekers are without skill. Rent-seekers can be very clever: witness structured finance! What defines rent-seeking is not stupidity or laziness, but skimming the fruits of the real economy without contributing to their production.

    Rent-seeking is consequential. Genuinely productive economic activity that would otherwise occur will fail to occur if producers will find it unprofitable after rents are extracted from them.

    I could tell a similar tale with respect to fiat money, except that fiat money isn’t permitted appreciate. Instead, the rents extracted for increasing exchange-service provision is confiscated by whomever can best claim the output of the printing press. That is rent-seeking just the same, and it has just the same real-economic consequences. I won’t belabor this story, because it is one upon which I think we are agreed.

    I contend that it is not necessary that rents be paid to the holders or producers of any particular commodity, that in fact people should ask to be paid in claims on what they expect to require (which in the real world will mostly be services). Under such a regime, anyone can print money by producing or creating the capacity to produce anything lots of people would want to “save”. But if claims on any one product or service became too popular as money, “printers” — that is producers — would spring up to drive the price of the currency down in terms of claims on other goods and services. Speculative accumulators of that claim would thus be thwarted and punished precisely as real-economic wealth would increase. Non speculative accumulators of claims on what they expect to use are not harmed by increased capacity.

    There is as we’ve discussed the danger of a speculative bubble drawing too much interest in claims on commodities whose production cannot (or for some propriety claims, strategically will not) be increased despite appreciation due to monetary use. That’s where “whack-a-mole” is an apt metaphor. I would support state action to prevent any one sort of claim becoming the hegemonic medium of exchange. I see this, as in the island parable above, as a kind of antitrust enforcement (which I think is a legitimate function of the state, despite the inherent ambiguity of when such enforcement is warranted and when it serves as a weapon of the corrupt). Unlike you, I think that the state could win an indefinite battle against a single money arising. I suppose it’s an empirical question.

    You write that I am opposed to savers or to saving. I am not at all. Deferring present wealth to future consumption is a human essential. I just don’t think people need to be paid very much to have this service provided for them. If real interest rates were negative, people would still save, as most humans will have vacations, retirements, children with future tuition expenses, etc that they will want to prepare for. I think it’s a remarkable that people taking essentially no risk and doing essentially nothing to direct the investment of their funds expect and appear to have frequently earned positive real returns. Mere saving can be thought of as storing or hedging. Ordinarily one has to pay for storage and goods deteriorate over time. One also usually has to pay for insurance.

    When you save, someone has to invest what you have produced but not consumed in such a manner that, down the line, the investment will have covered any storage or depreciation plus any interest you’ve managed to demand, or else that someone takes a loss. If someone, thinking they have wonderful investment opportunities, freely assumes that risk, that’s great. If an FDIC bank accepts your savings, and they invest it poorly, then I as a taxpayer may have to cover the loss plus interest on whatever it is you produced but did not consume. You should have to pay me to assume your risk of loss. If you sell what you produced for gold and merely hold the gold, and if the real value of gold increases regardless of how well or poorly what you produced is eventually invested, then you are forcing future producers to assume the risk that their products will be worth less, because the economies meagre produce will have to be shared with “savers” who failed to ensure that their produce was invested wisely. A saver whose funds are not meaningfully at risk and who has not found a private party that willingly assumes (and can cover) the investment risk has created risk for someone else, and that someone else should be compensated.

    Investors are quite different from mere savers. Investors direct what they have produced to some use that they believe will magnify future production, and hope to not only store their wealth, but increase it. However, they willing assume the risk of error. Investors both increase the likelihood that future wealth will be larger than present wealth, and disproportionately bear the burden if that happy future fails to materialize. I love investors. Since human beings have thus far generally found it possible to increase wealth, most people who produce more than they consume should probably be investors.

    But investing implies risk. Putting funds into an insured account (which insurance is paid for by others), or holding gold and relying on the fact that a monopoly of exchange-service provision in a growing economy implies increasing rents to gold are both ways of externalizing the risk that your production cannot be stored or invested well to a widely dispersed public. They amount to the same thing in different denominations. In both cases, by subsidizing saving relative to the risky act of investing, economies decrease the number of true investors and (especially where bankruptcy is legal) increase the proportion of today’s production that is simply consumed with no provision for the future. The apparent lack of risk in holding an insured account or gold translates to an increase in the systemic risk that an economy will fail to grow in terms of real production. In the insured fiat case, this translates to inflation, which harms everyone. When savers hold gold money, unless the economy has contracted in real terms (very rare given population growth and technological progress), the savers do undeservedly well, as having harmed the economy by forcing others to invest and assume risk for them, the value of the under-mattress gold still appreciates, while those without assets bear the brunt of the downturn.

    I’m going to stop now, abruptly, and without editing. I’ll again use the hour (7 am after a long night) as an excuse, and apologize for he disorganized and unedited logorrhea that above. I wished to respond to a few more things in your note, the idea of a single fixed frame of reference (which I don’t adhere to or expect to find),
    and to talk more about how I can support state action to destabilize nascent money-commodities, while considering the same state too corruptible to manage fiat money. It’s my turn to hope I haven’t given offense, and to apologize for all the history and ideas that I haven’t responded to. Your arguments are wily and entertaining, and your erudition is impressive. This has been a very enjoyable (if timeconsuming!) exchange.

    I am very curious about your not-mold ideas. Let’s do try port 25.

  27. moldbug writes:


    I hope you’ll excuse the delay in responding. I am actually trying to write some code here, occasionally, and my girlfriend’s ability to suppress the fact that she thinks I’ve spent too long on this project cannot be tested too much longer.

    And of course your responses do actually call for some thought – they often seem clearer and more coherent than mine. Hopefully this is not because you are right and I am wrong, but just because you’re a better writer.

    Definitions, as you say, are often critical. This is why I like to make up my own, and avoid associations that others have. When I borrowed Jouvenal’s terms “law” and “Power,” I broke this rule. This was perhaps useful at the time, but now you are discussing these words based on the associations that you, and the other 99.999% of humanity who I assume haven’t read Jouvenal, ascribe to them. Obviously it is impossible to consider this foul play. So I would rather withdraw the words and start all over.

    I consider myself a logical pacifist and a formalist. Let me define these words as precisely as I can, so you know where I’m coming from. To reach my reference point:

    Start with John Rawls’ veil of ignorance. No abstract judgment of a society should be based on assumptions that can be expected to favor any one segment of a society over another. This does not mean that everyone will do as well as everyone else, but that any quality metric we define must be judged over its effect on the average citizen of the society with simple equal weighting. (This is not a consequence of logic – just an ethical assumption that I happen to share.)

    Do not confuse this algorithm with any kind of reification of the group, a la Filmer, Rousseau, Hegel, etc, into an individual. Linguistic tropes such as the “democratic we,” the word “society” without any article, etc, are extremely dangerous and confusing. The temptation to apply thoughts and motives to these constructs, which can exist neither physically nor ethically, is great, and the resemblance to theistic anthropomorphism is troubling.

    Remove the progressive fallacy implicit in Rawls’ actual solution to this problem, which passively postulates a “Rawlsian god” who can instantiate any social design and use his superpowers to make it actually work.

    Instead, judge designs by how well they maintain their intentions over time without any kind of divine enforcement, and how easy they are to migrate from present conditions to their intended steady state.

    To judge a steady state you need an ethical metric. This can be anything you want. If your ethical metric says that the best society is that which sacrifices the most prisoners to Huitzilopochtli, so be it. Since no system of ethics can contradict reason, reason works whatever your ethics may be.

    For example, let’s look at one design pattern for government: anticolonialism. If your ethical metric asserts that states should be judged morally by whether they are administered by genetic relatives of their citizens, decolonialization was a smashing success. If it asserts that states should be judged morally by their adherence to traditional European precepts of natural law, it was an appalling disaster.

    Note that the intentions of the people involved in the actions that resulted in decolonialization have nothing to do with this. With this approach, it is generally easiest just to assume that all political actors are sincere and have benign motives, or to be precise motives that they themselves consider benign.

    The simpler an ethical metric is, the easier it is to apply. And this is important, because social systems can often be very complex. A good evaluator is, like any good standard, as simple as possible – you want everyone in a society to perform the same computation and produce the same result. Evaluating a political design in terms of both quality and feasibility is an O(n*m) problem in terms of complexity. Ideally in practice it should feel like O(n).

    My ethical metric is that I prefer to evaluate societies by the level of predictability in their allocation of scarce resources. This of course includes the assignment of onerous tasks, which can be thought of as negative resources.

    Since, as I’ve explained, predictability is the converse of conflict, another way to state this goal is that it consists of minimizing violence. Basically, I feel that violence is the cause of such a large percentage of human ahedonia that all other criteria are effectively negligible. However, I am a logical pacifist, rather than a romantic pacifist, because I am concerned not with intentions but with results.

    Obviously, “violence” is a continuous scale which cannot always be simplified into a single dimension. A few vague discrete categories can, nonetheless, be discerned, and hopefully elaborating them will help define how I use the term.

    I’ll use the analogy of grades. Since these are blurry, we can speak of A-, D+, etc, etc.

    A “class A” society distributes and assigns resources on the basis of need and burden – as some might say, fairness. If you are the person to whom some burden is lightest, you will carry the burden. If you need something more than anyone else, you will probably be the one who gets it.

    A “class B” society distributes and assigns resources and burdens on the basis of formal rights. A right is “formal” to the extent that it can be computed by an algorithm that, when executed by any two reasonable people, will produce the same result. A class B society maintains Pareto-optimality, allowing any of its members to transfer any of their rights, and does not create artificial scarcity in resources which could otherwise be shared by all. Also, a class B society does not attempt to manage the psychology of its citizens.

    A “class C” society is like a class B society, but it does create artificial scarcities, limit transferability of rights, and enforce other non-Pareto-optimal constraints. It may also have some rights that are poorly formalized or enforced, resulting in unpredictability and hence conflict. And it may impose some level of psychological training on its citizens.

    A “class D” society is one in which conflict is widespread and its outcome is often unpredictable. However, this conflict does not result in widespread unlimited violence. For most disputes, a single monopoly of power still enforces a single outcome which no party has any incentive to challenge. For example, illegal executions do not violate class D, as long as they are ordered by a single coherent authority, and are the result of some considered process which prospective victims can understand and evade.

    A “class E” society is one which includes widespread unlimited violence (war), but parties to that violence agree on contexts in which their actions are constrained – for example, if they do not permit their soldiers to burn villages and bayonet children.

    A “class F” society is one in which members can and do use any effective tactic in any context to achieve their goals.

    Obviously, the most desirable level is class A. I would rather live in an A+ society. I think most people would.

    Any non-dysfunctional family, for example, is a class A society. Many extended families are as well. Possibly a kibbutz, even, is an A or A- system. We can call class A “fraternalism.”

    I personally do not believe that class A scales, at least not if your citizens are humans as defined by any common contemporary gene pool. You certainly could breed a line of bipedal pongids for which a large class A society would be viable, but I’m not sure “human” would be the right term for this animal.

    A class B society corresponds to the usual ideals of classical liberalism. I like the word “formalism.” Since I believe that class B is the highest grade achievable – any attempt, I think,
    to produce even B+ will only give you B-, at best – I consider myself a formalist.

    The US and Europe today are probably a solid C. Some of the best Communist countries might have gotten to C-. As long as you were an Aryan, the Third Reich before the war was a C- (if you were a Jew it was a D or D-). Hong Kong, Dubai, etc, might be C+ or even B-. We can call class C “nationalism,” “demotism,” or “statism.”

    A class D society is Stalin, Saddam Hussein, etc. A class E society is WW1, WW2 on the Western front, etc. Class F gets you into genocide, terrorism, bombing of cities from the air, etc, etc. Minor distinctions here are increasingly unimportant, as you just don’t want to go there.

    So to backtrack to Jouvenel: while his thinking is quite complex and I’m sure I’m trivializing it, I would define “law” as class B, and Power as C through F.

    Following Jouvenel, the pattern we notice in history is that the farther a society is below B, the more it is more likely to decline than rise on this spectrum. This tends to support the hypothesis of complexity collapse, which behaves exactly in the same way.

    (It is an ugly and disturbing fact that all societies in history that reached or at least approached class B evolved as survivals of some continuous tradition with preliterate roots. We do not really know the source of the Greek, Roman, Hebraic, Germanic, Chinese, etc, legal systems. Their designers, if such there were, left few if any notes, and their histories show a general and monotonic decline in simplicity.)

    In the context of programming, I like to use the analogy of semiconductor fabrication. If you set up a fab in a stable, there’s no reason to make the stableboy wear a bunny suit. Complexity breeds itself, like dirt, and the way to eliminate it is to focus on the biggest sources first and readjust your priorities as your results improve. In other words, get rid of the horse before you try to get rid of the horseshit. Then work on the hay, the oats, and so on down to submicron particles.

    So you invoke Einstein: “But there is a rejoinder, that a system ought be described in the simplest manner possible, but no simpler than is possible, given the complexity domain.”

    Perhaps. Although it’s not clear that Einstein’s views on political or economic subjects were exceptionally perspicacious.

    But my point here is that simplicity is – for me – not simply one quality of a good solution, subject to tradeoffs against other goals. For me, simplicity is the only nonnegligible measure of quality, because I equate complexity with violence, and I believe that all social problems are negligible compared to violence.

    (Obviously, this depends on a large number of ethical assumptions which I hope I have made explicit. If your assumptions differ, your mileage will vary. I am not trying to set myself up as the Pope here. I am just trying to explain where I’m coming from.)

    For me, therefore, one showstopper in a political design is any source of unbounded complexity. I believe in taking an extremely conservative approach – not in the political sense, but in the engineering sense – to any such object.

    One analogy for this task is bridge design, not in its present CAD-enhanced form, but in the traditional rule-of-thumb practice, entirely unenlightened by mathematics, that created so many of Europe’s beautiful old bridges. In fact even the earliest suspension bridges were built by rule of thumb. The way you work when you do not have math to guide you is that your margins of safety become very wide. I don’t really think of myself as a Burkean, but this attitude is surely an aspect of Burke, and it is present in much of the best conservative thought.

    Okay. Enough about this. Hopefully you now understand the style of design I’m trying to apply.

    Let me explain where, from this perspective, I see your approach going wrong. I will start with the problems I see that I think are largest and hardest to fix.

    My first problem is that you seem to be designing a monolithic solution, not a distributed one.

    One simple design pattern for class B systems is hierarchical symmetry. That is, if we can design a class B system that works for a limited number of actors, but does not scale to billions, or has problems with culturally heterogeneous populations, we can define each actor in the high-level system as a class B system itself, and customize or use a less scalable design within it. This keeps us from having to define one system of law that binds all 6 billion or whatever people on earth.

    Therefore, one very desirable property for a design that claims to implement class B is that it can coexist peacefully and openly with other class B systems which it does not control. We can call this property “compatibility.”

    For example, if a purportedly class B design has to resort to aggressive violence, exit restrictions, etc, etc, to prevent its citizens from jumping ship and moving to these other external class B systems, the design clearly violates class B in the context of compatibility. Even if it achieves class B as a monolithic system, it is less useful than compatible B designs.

    If your approach is designed to achieve compatibility, this is not obvious to me. You seem to be designing for a single world state. An interesting problem, certainly, but not I think as useful as the same problem plus the compatibility constraint.

    My second problem is, as you expect, with our differing definitions of rent-seeking. I say:

    Defining a holder of money as a receiver of rents is a distortion of the concept of rent. Nor are declining prices, a normal phenomenon in the presence of increasing productivity, any impediment to moneylending.

    You respond with:

    Rent-seeking is called rent-seeking not because rent-seekers are without skill. Rent-seekers can be very clever: witness structured finance! What defines rent-seeking is not stupidity or laziness, but skimming the fruits of the real economy without contributing to their production.

    Rent-seeking is consequential. Genuinely productive economic activity that would otherwise occur will fail to occur if producers will find it unprofitable after rents are extracted from them.

    This is precisely the root of our difference. I am working with a class B definition and you are working with a class A definition. My belief is that empirical observation indicates that the probable, if not certain, result of imposing class A rules on a large society will be a class C system.

    My “rent” – call it rent_B – is defined in a way that forces it to violate the rules of the class B system – to attract, as it were, the attentions of anything like a working immune system.

    Generally what I mean by rent_B, and what I think most public choice economists mean when they talk about “rent-seeking,” is the creation of artificial scarcities in order to create and extract scarcity rents. For example, a modern union or a medieval guild creates and appropriates an artificially scarce item of capital, a union card or mastership respectively, that is a superfluous ingredient in the productive process.

    The concept of rent_B has nothing at all to do with any change in the current market exchange rate of some good or right I own. It does not say that such changes are good or bad. It is quite independent of any such thing.

    Your “rent” – rent_A – is very different. “Skimming the fruits of the real economy without contributing to their production.” This is unfairness in a nutshell. It is eating, you might say, without taking out the trash or doing the dishes.

    Anyone in any kind of domestic relationship is very experienced at solving these problems in a way that all involved conclud
    e is fair. Thus it would certainly be an exaggeration to claim that the idea of a class A rule that works in a class A society is intrinsically paradoxical or otherwise ill-defined. There are billions of class A societies functioning in the world today with these kinds of rules. The problem is just that none of them is anywhere near as large as the political problem space.

    And what you will find, I believe, if you try to formalize this concept, is that it cannot be done. There is no way to promote it into class B or to identify an analogous rule there. In a class B design process, I believe, the idea of rent_A is at best a source of distraction and error.

    This is Mises’ calculation result in a nutshell. There is no precise way of calculating any concept of “value” intrinsic to any object. Since basically forever, economists had been looking for the absolute definition of value, just as physicists had been looking for the absolute frame of reference. What they both learned – or should have – is that no such thing exists.

    In economics this is the marginal revolution – more mentioned than understood. (I used to be a big fan of Tabarrok, Caplan, Kling, etc, but this was before I read Mises. And decided that what I was getting from the econ bloggers in this general group was an extremely contaminated and highly diluted low-grade counterfeit of the original Austrian marching powder, made to be smoked by giggling undergrads, as it were, and entirely unsuited to the serious praxeology junkie.)

    The only computable number that corresponds to the value of an object is the marginal rate it can be exchanged for in terms of some other commodity. Furthermore, this number has specific, and different, units in its numerator and denominator. It cannot be magically morphed into some kind of pi-like unitless constant. And it has a nasty habit of varying over time, and depending on any other variable in the world that may affect the individuals who buy and sell it.

    Of course, one can certainly come up with other numbers. You can invent any number you want and try to force people to obey it. But there is a fundamental and qualitative difference between administrative calculation and market calculation, and once your policies depend on the former, you are out of class B territory. Or to put it differently, if you can design a stable mechanism for price control, you have a political perpetual motion machine, and every stationary bandit in history would give his eyeteeth for a glimpse at your secret formula.

    I think a number of your points are affected by this difference in our reasoning, if often in ways that may not be obvious.

    For one example, anywhere you use the word “real,” you are using an invented number. Comparing the purchasing power of present money to that of past money is an entirely incalculable problem. Again, numbers derived in this way are administrative figures, and any calculations that use them are subject to GIGO. It is in fact only “nominal” numbers that can be used in calculation.

    For another example, in the case of your tree-land-owners, they have formed a cartel and they sell at a monopoly price. The bastards. But you have the Maxim-gun and they have not, so you can force them to sell at any price you want.

    So how should that price be calculated? Should you establish an equal and opposite artificial scarcity of tree-wood buyers, a monopsony with yourself (I mean “you” here in the official second person plural, a tense that violates, I concede, my rules, but which I do so prefer to the official first person) as sole buyer? And how is that any fairer?

    My answer in this case is that, in a free market, you will see that the tree-land-owners will earn an obscene windfall, but not an obscene profit. The yield on their land will converge with the natural rate of interest, which is the expected risk-adjusted yield on any asset. If they genuinely own a natural monopoly, this is how it is priced, and any attempt to second-guess this price runs into the computational fallacy.

    Of course, in practice, most cases like this are the result of artificial monopolies – such as patents. In the most commonly cited example of this case, such as a privately owned bridge, I think people just have the wrong scale. A city and its public services are a perfectly coherent natural monopoly, and you’ll notice that one case in which state-owned companies seem to provide perfectly decent, if not wonderful, service, is the case of urban-scale utilities, mass transportation, etc.

    (What is the difference between a “natural” and an “artificial” monopoly? I would just say a natural monopoly is a right to a real resource that is scarce, ie, cannot be shared without conflict, and an artificial one is anything else. It doesn’t seem to me that we disagree on this.)

    Let me briefly also explain my view of the loan and capital markets, one on which we do seem to disagree.

    The reason I used the term “hoarding” is that I wanted to avoid the value judgments that are conventionally applied in the case of “savings” and “investment.” Like many 20C economists, you want to differentiate productive investment from unproductive squirreling away of shiny metal disks.

    The belief that there is any sense in which lending money is an act of some net benefit to society, or that hoarding it is somehow antisocial, or that the first is somehow better than the second, is, I believe, entirely without foundation.

    We can attack this belief with a reductio ad absurdum. Suppose, in the extreme case, your miserly hoarder saves all his money and never spends it at all. He buries it, and when he dies it is never found.

    Clearly, an antisocial gentleman! Except that what he has done is precisely the opposite of the work of his neighbor two doors down, who is a counterfeiter, and who has printed exactly the same amount of money. It is curious indeed that the antisocial work of both these fellows adds up to a precisely neutral effect on the rest of the world – which it does, because due to an odd coincidence, they have exactly the same demand patterns.

    In my opinion, the hoarder is not antisocial at all. He is “giving back” by not using his money, at the present moment, to compete for goods and services. This increases the purchasing power of everyone else’s.

    Of course, another way to resolve this contradiction is to decide that it is actually the counterfeiter who is the social actor. This is the path of the inflationist, and I believe you have already rejected it.

    But the line of reasoning that distinguishes lenders over hoarders is exactly this. It presents a “shortage of money” as a social evil. This is simply a cry for subsidy via dilution.

    The right supply of loans, at any risk-adjusted interest rate, is simply determined by the set of people who know they will not need that amount of money for that period of time. If you have a lot of money saved but you fear that you could need to spend it all tomorrow, you shouldn’t exchange it for future money.

    Granted, the loan may be liquid – in your original sense. You may be able to exchange it back for present money in a pinch. But this just means you have changed places with another lender, and if there is no other lender who wants to assume your original deal, the price of the bond should go down – and it will. Again, if a comet is scheduled to destroy the earth in 20 years, the right price of a 30-year bond is zero.

    Most important, loans are priced in terms of money, not in terms of “purchasing power.” That is, they must be stated in calculable numbers that both parties can agree on. Since there is no other way to calculate profitability, the natural interest rate, the equalized yield on all assets, is simply a function of marginal supply and marginal demand.

    For the natural interest rate to go to zero
    , resulting naturally in zero lending, there would have to be no form of capital that was – again, in money terms, the only way that any sort of profit can be calculated – profitable.

    Profitable investment and falling prices go together like bacon and a cheeseburger. They are absolutely normal in the presence of rising productivity. The prices that matter to a borrower, in terms of his ability to achieve a certain return on capital, are not aggregate prices in the usual sense of “deflation.” The prices that determine return are the prices of the products his capital produces. And you’ll note that Moore’s Law has not yet put Intel out of business, despite the fact that it has to lower its prices every quarter or so.

    Okay, I too think I have written enough here. I hope this helps you figure out where our specific differences are – it has certainly been informative to structure my views as a response to yours. And I do hope the same is true for you. Again, I apologize for my slowness, which I believe is congenital, and henceforth not my fault.

    I’ll drop you a line about the unmold thing shortly, though my fingers do, I’m afraid, need a short rest…

  28. moldbug writes:

    I bounced you an email, or tried – you appear to have some kind of whitelist thing going on. It seemed to work, but let me know if it didn’t…

  29. MB — So, I think we are both trying to wind this down, though I can assure you that it has indeed been, for me as well, “informative to structure my views as a response to yours.”

    Overall I’d say that, although this exchange has been a debate, I am more struck with our similarities than our differences. I find your categorizations of different levels or grades of societies interesting. It’s not precisely how I’ve thought of things, but I don’t object. I don’t see my proposals as going for an “A”, but as messing around in B-space hoping to patch a what I see as a B- and falling fast societies (the US and traditional West broadly) to bring them to a stable B or B+. We are both, unfortunately, in rule-of-thumb land, which as you are right to suggest, implies broad margins of error. We talk past one another often, as our rules of thumb are very different. A rule of “conservatism” is insufficient. It might be conservative to build an airplane out of bricks rather than thin sheets of aluminum. But the damned thing just won’t fly. Medieval bridges are, I’m afraid, are too easy a case, just build them thicker and stringer when in doubt. We are neither of us let off so easy. Most real-world problems involve trade-offs, where being too conservative in one dimension implies taking too many risks in another.

    Given that we are in rule-of-thumb land, it’s clear that our rules are calibrated very differently. We would both agree that complexity increases the likelihood of devolution from “Class B”. I think we would both agree that there are possible systems, which while simple and unambiguous, would predictably devolve in the context of any real human society. My intuition, and it is only that, is that given the current level of technology and absent certain forms of brutality and suppression that I consider verboten (a priori, as a matter of ethical taste), the solution space of stable class B societies may both require and enable a greater degree of complexity than would have been necessary or stable in the past. Further, I’ll admit (and here you have me a bit on the accusation of utopianism) that within the universe of possible stable, law-governed societies, I do have preferences (notions of fairness, justice, and productivity), and I am willing to take some chances (in terms of complexity and transitional hazards), to go from worse to better. You probably have ready at hand a catalog of good situations turned to blood porridge by precisely this improving impulse. I could go all cliché and liberal on you, and talk about MLK and the civil rights movement and we-shall-overcome, or some other hippie crap. In finance-y terms, expectations of risk and return, plus degree of risk aversion, are exogenous and cannot be normatively computed. I think I’m less risk-averse than you are, and my sense of the degree that one achievable class B solution might be better than some other (as I define it, but I think my definitions are fairly conventional) is larger than yours. I say MLK, you say Robespierre, things only get interesting when we start talking specifics.

    Generally what I mean by rent_B, and what I think most public choice economists mean when they talk about “rent-seeking,” is the creation of artificial scarcities in order to create and extract scarcity rents. For example, a modern union or a medieval guild creates and appropriates an artificially scarce item of capital, a union card or mastership respectively, that is a superfluous ingredient in the productive process.

    The concept of rent_B has nothing at all to do with any change in the current market exchange rate of some good or right I own. It does not say that such changes are good or bad. It is quite independent of any such thing.

    Here we see more how our disagreements are largely definitional. You see the price of a thing as inherently its price, neither right nor wrong, but you are willing to acknowledge the idea of artificial scarcities. Your use of the word artificial is like my use of the word real — which you are quite right to call me on. Both economists’ mystical but conventional price indices, and any notion that an occurrence is artificial, try to impose an arbitrary viewpoint on a reality that, in the end, just is. If we are willing to talk about artificial scarcities, I’ll say that the speculative dynamic that occurs inevitably under nonfractional commodity money is an artificial scarcity, just as much as the rules imposed by a guild, union, or cartel. (These would be cartels controlling human labor or other rivalrous commodities; we are very far from the entirely manufactured scarcity you describe w.r.t. patents.) Let’s not argue about that though. Let’s leave aside notions of artificial, conventional ideas of real growth, and let everything have its price within a certain historical, institutional, and technological context. Because there is no price, no meaning or value to anything, without such a context. We oughtn’t argue about whether a price is wrong or right, whether gold is artificially scarce or fiat artificially abundant (and unfairly distributed). We should just talk about potential institutional and technological contexts, avoiding the Rawlsian fallacy that you dislike and limiting our discussion to what is at least arguably achievable. By a principal of conservatism, we might start with the present state of affairs, and ask whether, with no effort on our part (or on the part of others like us), the current state of affairs is okay, will work out well enough. I think we both agree that no, there will be a discontinuity or at least a transition, perhaps people like us are not entirely impotent to prevent a terrible devolution, or, more optimistically, to move things towards a location that is both more stable and qualitatively better. Then, the same principal of conservatism might move us to propose changes as incremental and plausible as we can muster to move us towards this, um, better place.

    Then to the engineering without math. You have a proposal that is thus far undisclosed, and a mold-based fallback involving I think a “flag day”, where, by fiat, all fiat currency claims are converted to nonfractional claims on an allocated stock of shiny penicillin. I propose increasing the “multilingualism” of exchange-institutions by technological and normative means (promoting multicurrency payment systems, plus broadening the range of electable currencies from government-sponsored fiats to claims on other commodities and services people might choose to trade), and more controversially relying on states to artificially introduce volatility into the market for claims that threaten to become universal and exclusive scrip. I’ll concede that my proposal is not ideal. I’d prefer that second part to be unnecessary, and I am optimistic in a way I suspect you would not be, that with sufficient ease-of-use and variety in multiple money-like claims, the emergence of hegemonic money might be prevented merely by people preferring to save in terms of claims of their own very different patterns of future consumption. If this is not enough, I’m still willing to rely on the imperfect institution of the state for help, but I’d like to engineer around it. You are pessimistic about my proposal (with and without the help of the state). I am pessimistic about a money-as-allocated-gold system, both out of my latent utopianism (that’s not fair!), but also because I believe that it is likely to collapse badly. We have different technical intuitions, and no good science within whose equations we could find a mutually persuasive answer. Alas!

    I think we should mostly leave it at that. I’d taunt you a bit about your hoarder / printer tale, and ask you to think about the case where our miser is a baker, and instead of hoarding “money”, whatever that is, he bakes more cakes than he eats, and hoards those. If we are telling intuitive tales about productivity, and “what’s worse”, we can’t credit any productivity to the baker unless he has directed
    his resource to an ill-defined good use. If he sells his cakes for money to the counterfeiter, who does not eat them, but prints money to pay for them, still nothing good has happened. I see too much focus on the horders (low-risk savers) and the counterfeiters (governments and financiers), and not enough focus on ensuring that real good and services serve real purposes. I’d also taunt you about your notion of an interest rate. The notion of a “natural interest” is as fraudulent as the notion of “real growth”. There can be a market interest rate in terms of some commodity (paper dollars, gold, or fresh cakes), but no “natural interest rate” associated with “rising productivity”. It is easy to make the natural interest rate in terms of fiat go large and positive, and perhaps even encourage saving: have the issuer stand ready to borrow, and pay interest with freshly printed chit. It’s a complicated set of facts and institutions (e.g. positive technological change and sticky prices) that sometimes makes this kind of scam seem like a good deal.

    Under gold, I see the market borrowing costs in gold on gold as growing without bound until transactions are sufficiently reduced that people lie, cheat, or steal, or start exchanging something else. “Fractional reserve” schemes are how private entrepreneurs solve this problem (a form of lie, cheat, and steal that is the best way out of a bad situation). Fractional reserve, of course (with a hidden and variable fraction), is just private fiat. Modern fiat is private fiat with a built-in “flag-day”. The banks do the money printing they have always done, the state ensures that those who have accepted their funny money are better off than those who have not when the crisis inevitably comes, so that no one wants to lynch the bankers. It’s a great system, for the bankers and their quantum-mechanical-structured-credit-engineer successors.

    There’s lots more to talk about. There always is with your posts, which are dense and insightful. Again, I think we more agree than disagree — we both seem to dislike the conventional arrogance of economists, who “magically morph[] into some kind of pi-like unitless constant” the output of product of endless tentative conjectures and poorly measurable data, defining the “natural rate” of this or that, and declaring nature to have changed when they deem it convenient. We disagree primarily on whether the speculative gains that accrue to mere holders of a fixed-quantity currency are normatively a good thing, whether that dynamic ought to be a part of a next-attempt at organizing convenient exchange (presuming, as I think we both do, that a next attempt will be required), or whether we should try to avoid that (presuming, arrogantly and improbably, that we have anything to say about what will be).

    Non sequitur, I do want to congratulate you on this, among your many insights: “… the assignment of onerous tasks, which can be thought of as negative resources.” That’s a delightfully elegant idea, and one I’ve not encountered before.

    [p.s. I’ve received your mail, but haven’t read it, wanting to finish this first. Sorry about the whitelist crap.]

  30. moldbug writes:


    Indeed many of our differences are definitional and unresolvable. But not all, and some of your comments, which are as usual insightful, still point to issues that are interesting and on which I am not precisely sure where our differences lie. I hope if it doesn’t cut too deeply into your quality time I can still explore a few of these.

    Your identification of my differentiation between “natural” and “artificial” scarcities, and yours of “productive” activity or “fair” prices, is very interesting. And I think very telling. It is certainly not an analogy that had occurred to me, although now, as usual, it seems obvious.

    Here indeed is complexity. We have hunted the beast to its lair. Roars and the yowls of cubs can be heard within.

    Take the oceans, useful for many purposes (thalassa! thalassa!), but especially renowned for their production of tasty fish. Until multinational corporations systematically strip-mine their depths with ten-mile drift nets which catch anything that swims and is bigger than a nickel. The well-known tragedy of the commons, and happening right now, I believe, as we speak.

    How in an ideal world would this be corrected? Well, I say, putting on my Friedmanite hat, we clearly have a case of natural scarcity here. Voila, tragedy of the commons, exhibit A, the prosecution rests, a property right is demanded.

    But what is the right way to rectify this? Should one entity own the whole ocean? If so, who should hold its shares? If not, how should the seas be divvied up? By rectangle grids? Or should one company own the right to fish each species or population of fish? Or what?

    I have no idea. I am not a fishing expert. But my point is to confirm yours, made earlier – that complexity exists. That we are not talking about a divinely instituted system of law, but one that has to be made, and made well.

    Yet there is still, I believe, a difference.

    The difference is that the complexity of defining a system of fishing rights – as in any other property rights system – is finite. There will always be little ambiguities and clever lawyers to sort them out. Even with a completely modern, computerized property rights system, legal overhead cannot be completely eradicated.

    However, when we resort to “just price” theory, I think, we have done something entirely different.

    Any model of a just price – any non-market assessment of “objective” or “normative” value which is enforced by the state – can be Pareto-optimized by converting the price control into an artificial scarcity which is defined as a property right.

    For example, if you want the price of carrots to be higher than it is today, issue carrot-growing licenses without which any farmer groweth the yellow root as his peril. If you want the price to be lower, issue carrot-eating licenses and require all carrot buyers to display them at point of sale. If you can force people to use these licenses, you can set any price you want.

    This abstract exercise is very useful, because it enables you to identify the political beneficiary of a new scarcity.

    But it is generally observed in the breach. Administrative measures are far more common, generally involving rights which are nontransferable, a clear violation of Pareto optimality, and one whose purpose can only be to disguise the looting that is going on.

    My point, in response to yours, is that since every administrative system of price control – including the cartel-busting operations you favor – can be Pareto-optimized into a rights system of this sort, the amount of loot that can be extracted from the creation of artificial scarcities is unlimited.

    If there is any ethical, or even merely political, ground on which to defend against this cancer, which I think is as good a definition of the 20C disease as any, it is that the only acceptable level of it is zero. New scarcities must not be created, which means (by inference from Pareto optimization) that new systems of administration must not be imposed. Furthermore, adjusting a moving price signal equates not just to the creation of a limited set of scarcity rights – taxi medallions, if you will – but to its constant adjustment over time.

    Thus the distinction between natural and artificial scarcities. My definition of natural scarcities is simple: the set of them is bounded and finite. You can look at the problem once and solve it forever, modulo minor legal interpretation. This is not my idea – it is simply the basis of the English common law.

    The definition of natural scarcities inherited from dusty history was probably not quite perfect. Witness the oceans and the drift-netters. But the case for fixing this once and for all is better, I believe, than it has ever been.

    Artificial scarcities, in contrast, admit a kind of Occam’s razor. What makes scarcities artificial is that an infinite and arbitrary set of such scarcities can be created. Admit one and you admit them all. According, of course, to ethical taste, conceptions of “fairness” and “productivity,” etc. But since no standard formula for these conceptions can, as we’ve agreed, be defined, the only rule that does not admit the arbitrary and infinite growth of these restrictions is that none at all are permitted. And this indeed is the pattern we see.

    Next let me expand a little on the problem of the counterfeiter and the hoarder. From your response I am not sure I got my point across.

    The method I am using here is a method of abstract equivalence. If you assign some moral judgment to outcome A, and outcome B affects all individuals equally in terms of their own desires, reason itself does not allow you to claim that A is any better than B or vice versa. Similar, if B is better than or equal to A, it cannot be worse. Nor can two outcomes with negative aggregate utility combine to an aggregate utility which is positive or neutral. Et cetera, et cetera. (Obviously as an Austrian I am not a believer in aggregate utility, but my aim is to show that even allowing this approach, your analysis is not without inconsistencies.)

    Let’s consider the case of the hoarder again with this in mind.

    First, it is clear that losing the money is equivalent to destroying it.

    Second, I believe it is clear that perfect and equitable redenomination is, if we assign negligible intrinsic utility to the currency, and we discount psychological effects, neutral. Certainly the Turks did not complain when each of their million old lira was replaced with one new lira.

    Therefore, we can combine the following two acts to reproduce the effect of the hoarder’s actions.

    First, the hoarder, rather than destroying his money, distributes it evenly among all other holders of money in society. It is hard to fathom how this could be deemed an antisocial act. It strikes me as the essence of charity.

    Second, the monetary system is redenominated by the ratio of the hoarder’s stash to the entire money supply. This is, as I’ve said, neutral.

    The net result of these activities – either way you look at them – is that the hoarder has no money, and no one else’s balance has changed.

    Note that again this is the precise opposite of the activities of the counterfeiter, which can be similarly redefined as a skimming of accounts.

    And, finally, I see I have made another mistake by referring to the “natural rate of interest.” This again conjures up some image of a Zeus or Gaia who bestows this number on grateful humanity, presumably in a press release after the market closes.

    This is not what I intended, I apologize for it, and I can hardly resent your mocking of the concept. All I meant was the risk-free market rate at a given maturity, where by “market” I imply an absence of artificial scarcities, fiat rights, etc, etc.

    Of co
    urse every investment has risk. And no free-market loan will be at the risk-free rate. This tempts us to think that the “risk-free rate” is another of these unattainable, uncomputable numbers.

    But this is not so – because it is easy to compute the risk of any boolean promise. You can set up a separate prediction market for the risk event. This allows you to, even in the absence of magical “risk-free” T-bills, separate interest rate from risk premium.

    You see “the market borrowing costs in gold on gold as growing without bound” – presumably meaning, as I do, “allocated gold without fractional-reserve dilution.” Let’s examine this claim, which I believe is false.

    Assume for simplicity that all notes are zero-coupon. The risk-free price of 1 gram of future gold – say, at one year – as denominated in present gold, must exceed 1, or there is no incentive to lend.

    Let’s say we start with an administratively enforced absence of lending, and relax that enforcement. I hear 1 gram. Who is willing to offer 1.01? Who will exchange 1.01 grams of 2008 gold, risk-adjusted for 1 gram of 2007 gold?

    It must be someone who has some business plan – as measured not just by him, but by our aforementioned prediction market – that is profitable, in terms of gold, within this timeframe.

    You argue, it seems, that if the buying power of gold, according to some administratively defined index of goods and services, is rising, there may be no sellers at 1.01. Or 1.02, or 1.03, or maybe even 1.30. Because the would-be entrepreneur would have to expend his loan in X “real gold” and return it in Y “real gold,” where Y exceeds X.

    But the entrepreneur is not buying or selling a price index. He or she is buying or selling certain very specific goods. He or she (just because I’m an Austrian doesn’t mean I didn’t go to Brown!) is buying capital and selling products.

    And the price of said capital is determined by exactly this interest rate. If a factory does not have the power to change 1 gram of present gold into 1.X grams of future gold, where X is the current interest rate, its price is too high.

    Yields of all assets converge. If the market (“natural”) interest rate is 67%, that means that any capital which does not yield 67% is simply overpriced. By decreasing the price of any capital asset to the right level, you can indeed get it to yield 67%.

    The reason we do not have 67% interest rates is that there is a supply side to this equation as well. If, as in my usual example, the world was about to end, and no one was willing to exchange future money for present money, the natural interest rate could indeed go to 67%, or to anything.

    Loans, in a hard-money economy, are supplied by people who have money, but know they do not need to use it right now – who, in other words, are willing to exchange 1 gram now for 1.X grams in the future. Because of the human lifecycle of savings there tends to be a lot of this. And if there isn’t – ie, if the world is really about to end and we all need our cash to blow on wild cocaine parties, or spaceship tickets, or whatever – then so be it. If the interest rate is X, in other words, it means, by definition, as a tautology, that loans at a rate below X are not productive.

    Note again that the demand side, as well, has nothing to do with the “purchasing power” of money. Suppose the amount of Weetabix I can buy with a gram of gold in 2008 will be twice the amount I can get with the same gram in 2007, perhaps as a result of tremendous advances in cereal technology. And suppose I am an incorrigible Weetabix junkie, that Weetabix is all I eat.

    But – none of this is relevant to my decision whether or not to lend my gram of gold. If I know I will not be exchanging it for Weetabix in 2007, it makes no sense to keep it, and have 1 gram of 2008 gold instead of 1.X grams of 2008 gold. In other words, we are comparing oranges to oranges here – 1.X is always better than 1, whether the exchange rate between gold and Weetabix stays constant, increases, or decreases.

    I do not know the actual statistics, but I believe that if you go back to the last actual period of hard money, which was the Dutch golden age under the Bank of Amsterdam standard in the 17th and 18th centuries, interest rates were quite reasonable. Remember, these people were financing multiyear trading voyages to Borneo in wooden ships – it can hardly be described as a period of pathological risk aversion. And it was stable for about 150 years.

    Obviously I cannot fault you for your views, or I would have to become quite the misanthrope – I think most people share them. I think our differences are just the result of the fact that you have learned your financial ropes from the orthodox economists, and I have learned mine from the Austrians. Certainly the above discourse on capital markets is 100% conventional Austrianology, as you might find, for example, in _Man, Economy, and State_.

    One assumption that may lie behind your views, and which I find is shared by many – possibly even some Austrians – is the idea that “inflation” adds a “purchasing power premium” to interest rates. From this we derive the idea that “deflation” should produce the opposite, and thence we run into the “zero bound” or “liquidity trap” so beloved by the Bernankes of this world.

    As you know, I am very unhappy with the word “inflation.” I prefer to speak of “dilution,” which I define as just the increase in the quantity of outstanding money or risk-adjusted claims to money. But this does not invalidate the above; one could just as easily imagine a “dilution premium.”

    This also, however, is the wrong perspective, I believe. Because when we look at the supply of and demand for money, we see very clearly where the dilution has an effect. And it has nothing to do with the suppliers, whose time preference has not changed at all, ceteris paribus, and who will still take as much 2008 money as they can get for their 2007 money.

    Rather, the rise in interest rates due to dilution is due to the increasing demand for high-rate loans, on the part of actors who can pay these high rates because they are tapped into the dilution vehicle. For example, if the dilution comes not from a government but from some amazing new gold-mining technology, the gold miners will be able to pay very high prices for capital.

    Finally, I want to take a quick step back and look again at your nonhierarchical mesh topology of financial assets, with no single hegemonic currency.

    My view is that, even on the class-A grounds you want to judge it by, this world will be very undesirable.

    The problem is that you may not get one single “terminal bubble” – you may have administrative powers sufficient to prevent this, although again I think this implies a global superstate, and I am not sure what cure will be worse than this disease – but you will, nonetheless, have a continuing series of smaller limited bubbles which eventually pop.

    If they can be called that. And I have certainly seen a bubble or two in my day, and I’m sure you have, and I am not sure they can be called harmless.

    The problem, once again, is that you have a large amount of monetary energy which you want to see distributed evenly across all savable assets. But saying so will not make it so, and you do not have the computational tools – here I agree with Bernanke et al – to distinguish a fair price from a bubble price. No formula will tell you when to pop.

    Of course, if you recognize an incipient terminal bubble that threatens to become a new monetary standard, it will be pretty obvious. But this is a very blunt instrument. There are many cases short of it, and by the time any bubble is recognized – terminal or not – and popped – you are guaranteed to be inflicting a nontrivial level of pain and suffering.

    Is there a purpose to this suffering? I don’t think so. Because I do think a monetary system with a fixed quantity of money is stable. In both the political and economic senses. And I think I have explained why pretty much as clearly as I can.

    Please let me know if any of this is unclear, because as usual, I am happy to clarify. Again it is a huge relief to debate someone who is willing to defend the precepts of the status quo – or at least many which I regard as generally so – without resorting to the tactic of assigning the burden of proof to the defense, so to speak. And to see it done with such eloquence only adds to the pleasure of the task.

  31. MB — It is endearing to be termed someone willing to defend the precepts of the status quo. You, my friend, and I as well, are both money cranks to the rest of the world, our differences notwithstanding. (We have also both been trying to gracefully wind down this exchange, but alas, we have no grace.)

    First, I must object and disown, because again, words have been put into my mouth that I have not said. I dislike price indices of any sort. I think they are dumb, and to the degree they are relied upon they represent single points of failure and magnets for corruption. Nor do I believe in any kind of administrative price controls or price setting. I have (though I would prefer I had not because I don’t think heterogenous money is inherently as unstable as you think) argued for the threat of government action to prevent monetary monopoly. But this does not imply any sort of price setting. On the contrary, my suggestion was about price unsetting, about introducing volatilty into markets to prevent a scenario where a monotonic increase becomes a self-fulfilling one-way bet. Again, I want to back of from this a bit, though, both because I don’t think it’s likely to be so necessary, and because I think better market institutions could make it possible for nascent medium-of-exchange monopolies to be arbitraged away. I do not think governments ever know the “right” price of anything. In fact, I do not think there ever is a single price for anything. That is my whole point. Homogenous money is economics’ original sin. Like with sex, humanity could not have gotten very far without it. But, if I may make yet another self-discrediting comparison, we are ready to move beyond tradition, it is the dawning of the age of the orgasmatron.

    Your parable of dilution is not enlightening to me only because it is so close to how I conceive of the same phenomena. I have never argued that there is much wrong with destroying money, per se. As you suggest, it is just a redenomination. (There is a small problem with it, in that random distribution of wealth is informational noise, and reduces the efficiency of the information system that defines, according to my view of things, an economy.) But squirreling money under the mattress is not destroying it; the money is absented only temporarily. My problem is precisely with the fact that it can come back some day with claims that make no sense.

    Suppose that Entrepreneur X invents an amazing thing and builds an productive business around it. In doing so, he amasses 50% of all money (under a fixed money regime), which he saves under the mattress. Everyone is wealthier thanks to X’s work.

    Some time later, entrepreneur Y invents an amazing thing and builds an productive business around it. In doing so, he amasses 50% of all money circulating, which he saves too under the mattress.

    Some time later, it is entrepreneur Z’s turn. He invents an amazing thing and builds an productive business around it. In doing so, he amasses 50% of all money circulating.

    Now, X, Y, and Z all go to retire. Each, in turn, has done some wonderful thing that has in some ill-defined manner multiplied the world’s wealth severalfold, and has captured only a fraction of the value he produced in the form of 50% of all money. But Z’s wealth is only 1/4 of X’s. This is not a time value of money issue. X, Y, and Z may have done their work nearly simultaneously, or with many years intervening. None of them invested their money, or even lent it. Why should X’s claim be worth 4 times as much as Z? It’s a mere incumbancy effect, an artifact of successive redenominations. X by saving may have temporarily merely given money to his fellows, but when he dilutes by redeeming, he will redenominate in a manner that impoverishes all by more than he enriched, because he will be capturing the value of Y and Z’s innovations (as well as other productive changes that have occurred since he made his fortune). The saver is not only the hoarder, who destroys his money (little problem there), but also the eventual diluter, who contains effectively the power of the press, but withholds it until he can capture a great deal of wealth he’s had no hand in creating.

    [Would X, Y, or Z lend their gold? Only under genuinely risk-free circumstances, i.e. never. Why? Because although ’til true that regardless of gold’s inevitable appreciation, 1.1 X is better than 1.0 X in tomorrow’s gold, the risk-adjusted return on any individual investment is (by standard CAPM / portfolio theory arguments) worse than the risk-adjusted returns of the diversified market portfolio. But merely holding gold is equivalent to investing the market portfolio! In other words, an investor should do precisely as well holding gold as engaging in a diversified investment strategy, so the average investor will simply hold. Sure, some investors will have special opportunities and information (and others will be duped into thinking their opportunities are special), so some lending will occur. But markets will be thin and rates volatile, because by definition most opportunities are not above average, and taking a loss on an appreciating asset is much harder than taking a loss on a stable or depreciation scrip. BTW, A deus ex machina of using prediction markets to compute the risk of wide varieties of events is problematic. It comes around to the original post this debate is attached to. The scarce resource in finance is informational liquidity. Attracting sufficient liquidity for prediction markets is a challenging problem even for events “naturally” of widespread interest (e.g. US presidential elections), when the currency gambled is depreciating, and there’s no natural proxy for the ideal “market portfolio”. In a gold economy, where liquidity is inherently diminished because squirreling is a fairly optimal investment strategy, liquidity to support manifold fine-grained prediction markets is likely to be hard to come by without some subsidy. Indeed, under current circumstances, much is most interesting in prediction market research is how to attract sufficient subsidy to compensate skilled investors for their time and deploy without distorting prices.]

    I also don’t claim that all exchange should be spread evenly over all tradable assets. “Evenness”, as you say, is an impossible computational and definitional challenge. I only suggest that there must be more than one medium of exchange, that there must be at least several always (though that several may vary over time), to prevent monopoly rents being paid to the holder of an arbitrarily anointed resource. (I don’t think the value, or stability, of heterogenous money will be fully realized until there are direct money-like claims on goods and services that correspond to peoples’ planned consumption. I think that when such claims do arise, we will find that the lack of a fixed point of reference, rather a conundrum, will turn out to be a great source of wealth. Diversity of desire means wealth-seeking need only be partially rivalrous; a single medium of exchange is a bottleneck that creates avoidable zero-sum games.)

    We will not persuade one another, I’m afraid. We are both fairly attached to our viewpoints. That’s not all bad. I’m pretty serious about this idea of heterogenous money. I want it to happen, and I think it will happen. I’m interested in helping to develop the sorts of claims, markets, and trading systems that would make it possible. I think fixed-quantity commodity money will crash under the weight of “empirical injustices” — defined not because I think they are unjust, but because in the real world, as the economy plays out, enough people will consider outcomes sufficiently unjust or constraints sufficiently onerous that they will overthrow or subvert the system. Perhaps I would change my mind if I were to learn more about the Dutch golden age. But I doubt it. That the age didn’t survive, and that technological changes have arguably accelerated the dynamic by which inherent flaws in a monetary system lead
    to an inexorable undoing, are enough to keep my priors set firmly on “skeptical”.

    But, please do feel free to prove me wrong. I’m not standing in your way. As I’ve said, I see the current fiat credit bubble as wreaking great harm and sowing the seeds for terrible disorder in the not too distant future. I said at one point that I prefer fiat to gold, if those are the only choices, but that was meant in as a damned-if-you-do-damned-if-you-don’t kind of choice under abstract circumstances. At this moment, I would love to see the credit bubble crash (because, “the worse, the better” — ugh! — sooner is better than later when terrible errors viz the real economy are being made on a very large scale) and a flight to gold-money would not bother me at all. (For one thing, it would make me wealthy.) I’d still try to prepare and promote alternative monies, and that’s all. If I’m wrong, my error would be harmless. If I’m right, after a while people will be looking for ways around an artificially scarce medium of exchange, and my schemes will find adherents. (I more and more think that the state-as-antihegemonic-guarantor part can be dispensed with, so let’s leave that part out.) I suppose that we’ll actually see. We have the fortune, or misfortune, of living in times where these discussions may turn out to actually be relevant.

    More on port 25. Tho’ probably not tonight!

  32. moldbug writes:

    I have a little bit of trouble capturing the time relationship between your X, Y, and Z in this case. Their shares of the money supply add up to 150%, which implies somehow that money passes from X to Z, but when, and why?

    I think the problem, still, is that you are trying to compare currencies at different times by some sense of objective value. The conventional way to do this is a price index. You protest that you do not believe in price indexes, and for all the same reasons I don’t believe in price indexes. But without that, what are you using?

    Suppose someone tells you that on Planet 6 of Alpha Centauri their currency is the foobar. They hand you a super-powerful telescope they have built, you squint into it, and sure enough – you can see them, just barely, exchanging little notes that say “One Foobar” on them. Presumably the same telescopes can be built there, and they can observe with amusement our monetary woes.

    The question arises: what is the exchange rate between the foobar and the dollar? How many dollars is a foobar worth? Or vice versa?

    Well, of course, the question has no meaning, because in this toy example, faster-than-light travel remains impossible. There is no way to exchange these goods for each other, even if we wanted to.

    And obviously, the same would be true if the Alphans called their currency the “dollar,” just like us. Names do not matter.

    But suppose you learned that Planet 6 is actually an old backup of the Earth system. To be exact, it is a snapshot taken right after our Civil War. So their “dollar” is in fact, in a sense, our “dollar.”

    The question, however, remains meaningless. And it is also meaningless if the snapshot was taken in 1965. Or in 2005. Et cetera.

    We cannot compare the exchange value of present dollars with the exchange value of past dollars, no matter how much easier it would make historians’ jobs, because just as there is no way to trade with Alpha Centauri, there is no way to trade with the past.

    When referring to past dollars, we can note that if you had a past dollar in 2005 and you did nothing with it, you would have a present dollar in 2007. But you can think of the past as just one algorithm for deciding who owns what. It just happens to be the simplest algorithm – and, again, this is what matters to me. Here again is a core point of disagreement.

    You seem to have a very strong desire to arrange the consequences of trade in a way that corresponds to a certain concept of fairness that you have. It is presumptuous of me to guess where this desire comes from – I have done too much of that already. However, I think it is still worth noting that this general ideological, or perhaps idealistic (for the last few months I have been going around, in a rather annoying way, and trying to find people who can explain to me the difference between an ideologue and an idealist – if you can do this, you may win a valuable prize!) tendency is deeply ingrained in the New England traditions that are the backbone of American, and since WWII global, intellectual thought.

    I myself was raised in these traditions. My father worked in the old spearhead of the American mandarin class, the Foreign Service, and I have done time in all the usual educational sancta sanctora, if never in a context that was really engaged in their strident ideological mission. But I was, I think, not left entirely without a certain appreciation of this ancient, rich in every sense of the word, and tremendously successful school of thought.

    It does, however, show a consistent pattern of thinking that is more moral than practical, that displays an affection for “progress” which is ahistorical often to the point of hubris, and has frequently produced results that are contrary to its intentions. All of this is on display in the two strange pseudo-novels that are the unjustly neglected bibles of the American century, Bellamy’s _Looking Backward_ and House’s _Philip Dru: Administrator_. (They are both on Gutenberg.) Neither is terribly well-written, but just about every ideal or idea that was ever tried by American or American-inspired political and economic systems in the 20th century can be traced or connected to one of these odd old books.

    And the family resemblance is not, in my opinion, flattering. The reason I favor returning to proven designs is that these experiments have, in my view, failed, and restoring the belle epoque, the World of Yesterday (Zweig’s little book is absolutely invaluable), is the logical alternative.

    Certainly if you had a future-observing telescope and you could lend it to all the readers who were so enchanted by Bellamy, or the statesmen who admired House (Wilson’s right-hand man and more or less the founding spirit of the CFR), they would recoil in horror. And if you offered them the choice between keeping the world as it was, with all its minor aches and pains, or plunging into what they certainly would have defined, though their intellectual heirs today do not of course so view it, as a century of barbarism, I am sure they would have found it easy. Why not respect this? What else can we possibly have to respect?

    Now certainly this old world failed. And restoring it in a stable way involves understanding why that happened, and making sure it doesn’t happen again.

    But my view, and perhaps here too we disagree, was that the failure of this world was due to politics, not economics. The Bank of Amsterdam did not fail because its noteholders refused to lend. It failed because the Princes of Orange raided it to fund their wars. You can argue that the distinction between politics and economics is a fuzzy one and not qualitative; it remains my belief that the history of law, especially pre-20C law, shows that the job can be done within a limited complexity budget. Here again, perhaps, we reach disagreement.

    A good book, although it has nothing to do with economics and is just a well-written cultural history, on the Dutch golden age is Schama’s _Embarrassment of Riches_. Don’t miss his Rembrandt biography, either, it’s a masterpiece.

    We are not there on the gold. If merely holding gold gets you 1 gram of gold in a year and lending gold gets you 1.X grams of expected gold – risk-adjusted, by your own definition – the only reasons not to lend are that you expect to need the gold this year, not next, or that X is 0. It is true that capital yields equalize, but gold in this case is cash, not capital. Cash has no yield by definition.

    You also seem to be mixing my meaning of risk, which is just the probability of default in one particular note, with the meaning in the sense of diversified as opposed to covariant risk, that is, assuming an investor with a nonlinear utility function who has an incentive to diversify. Fine. But this assumes the money is lent out at all, not just held.

    Your point about prediction markets, however, is well taken. They are not at all magical, and I have improperly imputed quasidivine powers to them. Not only may these markets be thin, they may simply be wrong.

    Lenders who cannot accurately assess risk may charge an overly high risk premium. Or, of course, an overly low premium. Given the parlous state of today’s financial system, it does seem unwise to rely on market assessments of risk, as today’s market assessments are clearly completely out of whack. But this is of course an artifact of the odd state of fiat finance today, and if you look at historical markets, the task of assessing loan risk, which in the dark medieval days before the yen carry trade (Cassandra is just blogging up a storm lately…) banks actually used to think of as their job, is quite doable.

    I think one of the reasons we differ on homogenous, “Highlander” money is that you are looking at the process of appreciation in which one currency becomes dominant, and I am looking at the steady st
    ate point at which the issue has already been settled.

    Transitions between currencies are, without a doubt, singularities, and they will include a serious amount of chaos. From the perspective of the old currency they always count as hyperinflation. And there is no doubt that during the transition, lending of the new currency will be the exception rather than the rule. (Look at gold lease rates these days!)

    The point of a flag-day approach is to make the transition process surgical rather than traumatic, and instantaneous rather than protracted. I am not Philip Dru and I do not run the world, so I do not expect this to be executed perfectly according to my precise ideological dicta, but I do think that at some point someone will have to pull the switch. And it strikes me as not insignificant that someone in the position of a Benn Steil is thinking about it.

    As for heterogeneous money, I still don’t think you have answered my main complaint, which is that either

    (a) monetary energy – that is, demand from producers of nonsavable goods who are looking for a store of wealth, but have no direct demand for the monetary commodity – will be distributed evenly across all savable goods; or

    (b) monetary energy will be distributed unevenly across savable goods, and will not be distributed statically, but will oscillate, causing price fluctuations which are inherently unpredictable and tend to disrupt productive activity.

    If your choice is (a), I do not see how you enforce it. If your choice is (b), I do not see how it fits your value system, which seems to ascribe considerable importance to the problem of associating reward with productivity.

    And I do not see how there are any choices other than (a) and (b), so I must still be missing something. (Surely if you could enforce some static distribution of monetary energy, you would choose to enforce (a).)

    Perhaps this too is best explained over port 25…

  33. Alas, we are fairly hopeless in our resolutions. Port 25 has become a rhetorical promised land to which we make only ritual genuflection. “Next year in Jerusalem!” (You, at least, did once try.) But we do (or did for the first 25 comments or so) have lurkers to our hyperverbosity, so for their sake, I think the topic of mold vs. heterogeneity stays here (while unmold and other matters can move elsewhere).

    Re X, Y, and Z, I think I explained myself poorly. Suppose there are 1000 grams of gold in the world when X does his thing. He accumulates 500 grams,and goes into cryogenic hibernation for 3t years, keeping his gold in the vault. Some time later, Y does an equally amazing thing, and manages to gather 250 grams of Au for his trouble. Y is perfectly happy with this for the moment, as money, like “the sign” is arbitrary and differential in its nature. Y is happy enough: 250 grams in his day has the same meaning that 500 g had in X’s (in relative terms; in absolute terms it probably buys a little more due to technological change and population growth). Y sets the dial on his cryostore to 2t. Now Z is tiresomely brilliant yet again, accumulating through sheer marvelousness 125 grams of yellow. Again, for his day, he’s as rich as X or Y ever were, which is only fair (uh oh, that word!), since by hypothesis our entrepreneurs have done essentially the same thing just at different times. Z yawns, and looks forward to a good t of sleep.

    3t time units after X turned in, all three wake up. While all three did approximately the same thing, X is 4 times as rich as Z. You might argue that this is a time-value-of-money thing. After all, X produced but did not consume long before Z did, and his relative fortune comes from unconsumed production earning interest (here in the form of relative capital appreciation, actually the depreciation or dilution of value of others’ holdings). After all, if those Indians had just put their beads for which they sold Manhattan in a bank, they’d be worth more than all those skyscrapers now, right?

    [That last is one of the core myths of finance. What currency could they have chosen, and which banks, such that a chestful of beads would actually have been transmuted by the magicks of compound interest to trillions of dollars of present value? It simply would not, and could not, have happened in this real, historical world.]

    Back to XYZ, no time-value-money explanation works here, quite simply because our outcome is entirely insensitive to the value of t. t can be a year or a millenium, but X ends up 4 times as rich a Z, just because he was around first. There is a mere incumbency effect, that, absent draconian action by states (confiscation of inheritances, property taxes, or whatever) will lead to plain inequities and constraints that will lead to the subversion of the system, or its maintenance by brutality. Obviously, X, Y, and Z here are like Gates, Buffet, and Jobs — too rich to pine about the injustice of their circumstances, just ’cause Gates is #1 in fortune. But the same dynamic holds at smaller scales. Suppose everyone starts with equal quantities of gold. Exchange occurs, because people need stuff. But over time, by luck or great deed, some accumulate more gold than any consumption needs might require, and store the rest, since mere storage is a good investment. Over time, new accumulators, regardless of their talent or diligence, find it harder and harder to build up caches of the stuff of comparable scale to that collected by early entrepreneurs. The first rounds of winners become noblemen and everyone else serfs, until the monopoly of gold is broken, which will not be long. Ambitious pretenders will write surreptitiously fractional notes to support living the lifestyle of their richer betters, and we are back in the world of fiat.

    We are talking past one another about lending. My claim is that any lender under a gold standard has no investment portfolio whose risk-adjusted return is better than holding gold. Suppose someone offers you 1.1 grams next year for 1 gram today. As you say, if that’s a certain bet, you’d be foolish not to take it. But it’s not a certain bet. Suppose that independently, the risk of nonpayment is known with certainty to be 5%. Then you should take that bet. Now suppose a competitive market for the use of funds. No one now will offer 1.1 grams for this opportunity. In a nonfractional gold world, the price will approach 1.052 g, as any investors with no use for gold today will bid the price down towards that level. But at that level, a risk-averse investor is better off just holding than investing, as the expected value of the investment is his initial 1 gram! In the real world, the risk associated with an investment would not be so certain, so there would be some difference-of-opinion liquidity of the kind that helps prediction markets to sort of work. If I think that the real risk of nonpayment is 2%, and the market thinks it is 5%, than I have a profitable opportunity. But these are nonequilibrium anomalies, and prediction markets provide good examples of how liquidity is hard to come by when the expected profit on playing relative is no greater than expected profit on holding money for the average, no-special-information investor.

    In today’s fiat world (and any good world), where money-holders expect stable to depreciating, rather than growing, investing is positive sum relative to the alternative of mere storage. If someone offers me $1.1 for $1, a competitive market does not imply that there’s a 10% risk of nonpayment. If dollars are being diluted by 5% of their value annually (in terms of a diversified portfolio of storable goods, for example) then a $1.1 investment implies a risk of default of about 5%. I might choose to buy and store a diversified portfolio of storable stuff, but actually that’s very hard in practice to do, and since I have no comparative advantage in storing, I’m likely to achieve something less than my expected 5% profit. Much better to lend money into a competitive marketplace, and thus taking advantage of those with a comparative advantage in storing (who will borrow to buy storables, paying back just less than the value of the storables after a year, given a spread for their minimal storage costs, and a very small profit spread). In a fiat world, savers have to invest in order not to lose. And competitive markets for money don’t happen magically. Investors have to evaluate and investment opportunities to ensure they are well and competitively priced, or else they also lose. Default opportunities (“index funds”, CDs) arise for savers uninteterested in informational work. They must, over time, earn something less than the dilution of the currency, otherwise no one would do the information work required to define the appropriate weightings of the index, but are expected to earn more than mere holding.

    There’s a lot oversimplified in the above — the quantity and value (in terms of any other commodity) of “storables” is not fixed. If money is diluted by the same amount by which the quantity of storables expands, then investors lose only relative wealth, rather than absolute-wealth-in-terms-of-storables by holding. Most people are interested in both relative and absolute wealth, and in terms of some idiosyncratic variety of goods and services they hope to consume, not merely storables or a monolithic hypothetical portfolio. But the core case holds — Assuming people want to save (ie they have no use for their wealth today, but want it tomorrow) and will compete for the strategy that yields the greatest future wealth, the risk-averse average saver will invest in a fiat world, but hold under nonfractional gold.

    As for heterogeneous money, I still don’t think you have answered my main complaint, which is that either

    (a) monetary energy – that is, demand from producers of nonsavable goods who are looking for a store of wealth, but have no direct demand for the monetary commodity – will be distributed evenly across
    all savable goods; or

    (b) monetary energy will be distributed unevenly across savable goods, and will not be distributed statically, but will oscillate, causing price fluctuations which are inherently unpredictable and tend to disrupt productive activity.

    If your choice is (a), I do not see how you enforce it. If your choice is (b), I do not see how it fits your value system, which seems to ascribe considerable importance to the problem of associating reward with productivity.

    The difference between monopoly and competition is not an even division of monopoly rents. I don’t think there needs to be any rent paid (“monetary energy”) for use of claims on a commodity as medium of exchange, and therefore no price fluctuations based on the current popularity of a commodity for the purpose. When gold is the only medium of exchange, those who have it hold it, unless someone pays a substantial premium to use it to undergird their exchange transaction. If we assume that there is a natural, direct barter rate between corn and cows and also between wheat and cows and also between corn/chickens and wheat/chickens, if I have chickens and will want a cow, I can sell my chickens for wheat. If too many people do this, wheat holders might raise their prices, but even with infinitesimal moves in this direction, I’ll switch to corn. If both cornsellers and wheatsellers try to profit from exchange mediation, I’ll choose something else. By hypothesis, the number of commodities that potentially serve as a currency is infinite — there are low barriers of entry into the field of exchange mediation. While at certain times certain commodities may be popular, and the price of that commodity might be slightly affected by medium-of-exchange use, as soon as that elevation becomes significant at all, I’ll choose another commodity for exchange. For any given commodity, price-fluctuation due to popularity of medium of exchange approaches zero as the difficulty of choosing alternative commodities approaches zero. Thus, there need not be any sort of wide price swings, even though some commodities may be widely used for exchange over some period of time. Since any attempts to extract much rent for that use would simply force a switch to competitors, holder of the exchange commodity have little interest in trying raise prices.

    Anyway, I hope this makes some sense, though I’m under no illusion that I’ll have persuaded you. As always, it is a pleasure neither persuading nor being persuaded by you…

  34. moldbug writes:

    Okay, I understand the point of your XYZ example now.

    My answer, as always, is that you continue to use words like “inequity” in a distinctively 20C sense. You are clearly a believer in social justice over formal justice – a legal realist, so to speak, rather than a legal formalist.

    (I am happy to be able to report that legal formalism seems to be in again this year. Well golly gee, it’s about time. Soon I’ll be able to get the uber-narrow Mod Squad ties out of the closet.)

    If X’s actions, you assert, are no more meritorious than Y’s, than it is unfair that X should live better than Y. Sure. I mean, as usual, it all depends on your definition of fairness. Which I think you will find very hard to pin down.

    I think I have made my point, which is that the degringolade of law that you and I seem to concur in decrying is precisely the result of replacing the scalable, formalizable, but morally unfounded, principle of natural law, with the essentially Calvinist principle of moral deserts.

    Concurring with most 20C thinkers, you suggest that the degringolade (I’m sorry, this is just my new favorite word, we can use another if it strikes you as too pompous) was due to the morally justified revenge of the masses at the behavior of the stiff-necked aristocrats in refusing to relinquish their ill-gotten loot, not generated by any kind of productive activity, but simply inherited from ancestors in the distant past who were no more than murderous thugs.

    Concurring with many 19C thinkers, I believe that the cause was the abandonment of medieval formalist law, with its deep Roman roots, in favor of more authentically Christian criteria of democracy and social justice. Which because they could not be formally defined, simply resulted in complexity collapse and the victory of the powerful. It proved impossible to give an inch without giving a mile.

    And I think an understanding of this principle of law has to be a core feature of any new consensus. Obviously this is no small task, but neither is the one you propose. When prominent mainstream scholars of law say things like “my name is Lawrence Solum, and I am a legal formalist,” I cannot help but be encouraged.

    Chesterton once said that Christianity had not failed, it had never been tried. Well, in the 20C, it was tried – I think the basically Unitarian creed I was taught at Brown, for all its atheism, has a heck of a lot more connection to the teaching of Jesus than anything that comes out of Pat Robertson etc – and sure enough, it did fail. No one really believes in government by the whim of the masses anymore. After Communism and the Third Reich, how could you?

    So when you write:

    But over time, by luck or great deed, some accumulate more gold than any consumption needs might require, and store the rest, since mere storage is a good investment. Over time, new accumulators, regardless of their talent or diligence, find it harder and harder to build up caches of the stuff of comparable scale to that collected by early entrepreneurs. The first rounds of winners become noblemen and everyone else serfs, until the monopoly of gold is broken, which will not be long. Ambitious pretenders will write surreptitiously fractional notes to support living the lifestyle of their richer betters, and we are back in the world of fiat.

    I think you would be well-served by a historical look at how “the monopoly of gold” was broken. Fractional-reserve banking is an explicit violation of Roman law, and it cannot exist and has never existed without the protection of the state. It originated with commercial rather than aristocratic interests, who certainly cannot be faulted as a class for accepting the opportunity that the growing central power offered them.

    Besides, remember, either your aristocrats hoard their gold, they spend it, or they lend it.

    If they lend it, you approve.

    If they spend it – which is typically what happens in practice – the economy is certainly more oriented toward the production and consumption of luxury goods than a Calvinist would want, nor are the individuals who get to enjoy this luxury lifestyle the ones who your moral criteria, or mine, or anyone’s would select for this distinction. But through this process the gold leaves their vaults and does not return, and the result is the genteel but penurious aristocrat who became quite a stereotype in the 18 and 19C.

    If they hoard it, as I’ve demonstrated, they simply decrease the effective money supply for the duration of their squirreling behavior, thus increasing the purchasing power of everyone else’s money. Of course this squirreling may at some point end, but then we are back to one of the other two cases.

    Then you write:

    My claim is that any lender under a gold standard has no investment portfolio whose risk-adjusted return is better than holding gold. Suppose someone offers you 1.1 grams next year for 1 gram today. As you say, if that’s a certain bet, you’d be foolish not to take it. But it’s not a certain bet. Suppose that independently, the risk of nonpayment is known with certainty to be 5%. Then you should take that bet. Now suppose a competitive market for the use of funds. No one now will offer 1.1 grams for this opportunity. In a nonfractional gold world, the price will approach 1.052 g, as any investors with no use for gold today will bid the price down towards that level. But at that level, a risk-averse investor is better off just holding than investing, as the expected value of the investment is his initial 1 gram!

    What you are showing in this paragraph is that assuming an infinite supply of loans (that is, of moneyholders who accept a certain maturity of exchange) the risk-adjusted interest rate will be bid down to epsilon. This is true – but the supply is anything but infinite.

    Similarly, as the risk-adjusted interest rate approaches epsilon, we can expect quite a multitude of ways of using capital to meet this very relaxed standard of profitability. So as you increase the supply of loans, the demand will increase as well. The system is self-stabilizing.

    Then we reach the nub of your views of heterogeneous money, and mine of homogeneous. In terms of my earlier taxonomy, you seem to be defending the position that condition (a) is self-stabilizing, which I dispute. Here at last is a really meaty argument. Let’s mix it up.

    You write:

    By hypothesis, the number of commodities that potentially serve as a currency is infinite — there are low barriers of entry into the field of exchange mediation.

    Well, “infinite” is a strong word.

    First, “exchange mediation” is certainly a correct definition of the role of a monetary standard – but it can also be a little misleading. Because it leads one to think of monetary demand – the demand to hold a good not for its own use, but to exchange later for other goods – as dominated by, as the old inflationists used to say, the “needs of trade,” that is, short-term exchange.

    But as I think I’ve shown, the dominant source of monetary demand, in a quantitative sense, comes not from traders, whose quick ins and outs tend to balance on an aggregate scale, but from long-term savers. There is an enormous demand to hold storable goods for time periods that are significant on the scale of human life.

    Nor does the existence of a loan market change this. A loan market does not affect the net monetary demand for a commodity. The lender exchanges the money for a note – but we have simply moved the money from one holder to another. If anything a loan market in some commodity makes it more attractive as a monetary good, because for those who are willing to lend, which (due
    to the human lifecycle) is many, a loan market increases the relative yield of that commodity next to another good that does not have a loan market. (For example, this is one advantage of fiat over gold today.)

    So there are well-known criteria for satisfaction of monetary demand. First and foremost these are low transaction cost (because two transactions are needed for a round-trip exchange) and low storage cost (because the period of storage may be low). There are, as I’ll discuss, more criteria. But I think these two already get you out of the “infinite” category.

    Thus, since the set of potential monetary standards is finite, we can expect the price of at least one such good to be increased, ceteris paribus, by monetary demand.

    The question is: will it be one good, or many? Will the distribution of monetary demand across potential monetary standards be uniform, or will it exhibit spontaneous symmetry breaking? And will it be stable, or unstable?

    So you write:

    While at certain times certain commodities may be popular, and the price of that commodity might be slightly affected by medium-of-exchange use, as soon as that elevation becomes significant at all, I’ll choose another commodity for exchange. For any given commodity, price-fluctuation due to popularity of medium of exchange approaches zero as the difficulty of choosing alternative commodities approaches zero.

    This is the very heart and essence of our disagreement. We are down to the bare metal of game theory.

    I prefer to do my game theory in English. I’m sure you could express the same in math, but who cares. Ever since I was fourteen and I realized that I wasn’t smart enough to be a mathematician, I have resented the discipline and its practitioners. So there.

    The question is: I, the saver, the fisher of fish, wish to exchange my nonstorable goods for storable goods in such a way as to maximize the value of other goods I can exchange these goods for at some later date. What is my strategy for selecting a monetary commodity?

    What we are searching for is a Nash equilibrium. This is a strategy that, when all players in the game use this strategy, allows no alternative strategy which will be superior.

    Let’s review your strategy, which we’ll call strategy A, and mine, which we’ll call strategy B.

    Strategy A tries to avoid bubbles. In other words, it tries to exchange out of commodities whenever they experience appreciation due to demand which is monetary, not intrinsic. Due to your hotshot computerized central-limit-order-book or Walrasian auctioneer or whatever, we will assume that transaction cost is zip, zilch, zero, so a user of strategy A can make any set of exchanges he wants to stay out of the way of the hoarders.

    One significant problem with strategy A is that, from the information that the market price provides, there is no way to know whether the exchange rate vector (in the heterogeneous world, remember, price is not a single number but a vector; arbitrage makes the matrix of all such vectors degenerate, but it does not provide a reason for choosing one exchange rate and nominating it as “price”) of some commodity is affected by monetary demand, and if so how much. If this makes you think of the games that are played with base metals and non-LME warehouses, it should. We are in administrative whack-a-mole territory here.

    But set this aside and assume we can somehow know this information.

    I still contend that strategy A will lose to strategy B – which is to find the terminal bubble. Money, in my view, is the bubble that doesn’t have to pop. Players following strategy B try to find a Schelling point which is a commodity that is naturally scarce, and thus can appreciate in price without monetary demand being satisfied by equal and opposite supply. They they value all other goods in terms of their yield in this commodity.

    We know that strategy B is a Nash equilibrium, because it is the concept of “money” we all are used to. Absent dilution, which imparts a ceteris paribus upward momentum to all prices, there is no incentive to switch from the Schelling point commodity – clearly, for us, gold – to, say, iridium. If iridium is a fixed-quantity resource and its projected yield in gold exceeds the market interest rate, its price just has to increase. But, barring projected dilution in gold, this is a one-time adjustment – there is no reason to extend it indefinitely into the future. I think if you google “Hotelling curve” you will find a good explanation of this phenomenon.

    If everyone follows strategy A, assuming again that they have the imaginary information necessary to follow it uniformly, your heterogeneous market will survive and do fine.

    But this is not stable. All it takes is a small band of renegades who follow strategy B and start pushing the gold price up. The adherents of A will begin to note that their returns, on the non-Schelling potential monetary commodities across which strategy A has distributed monetary demand evenly, are lower than the return on strategy B. They will also note that if everyone switches to strategy B, they will have missed out on a remarkable alternative investment strategy.

    So, if they are rational actors, they will sell their wheat or copper or whatever, restoring these commodity markets to a state of purely intrinsic supply and demand, and flood into gold. At any point before this process terminates, it benefits the followers of strategy A to abandon their doomed vessel and leap into the yellow sea. And once it terminates, it makes no difference – they are holders of a commodity basket which will have lost any price appreciation due to monetary demand. The last follower of strategy A has no incentive at all to sell non-gold commodities and move to gold; the second to last follower has an incentive which is epsilon; and so on.

    Of course, Friedrich Hayek was on your side, not mine. (Oddly enough the market price on this item was much lower when I bought it last year. Hm, hm, hm…) But if we believed in argument from authority, would we be here?

    Please bear in mind: this argument I have given you is just mine. Not only is it not Hayekian (Hayek was really a second-rate Austrian, a popularizer at best), it is not even Austrian. If you follow the monetary theory of Menger or Mises, with your super-duper computerized exchange you have eliminated coincidence of wants as a cause of monetary standardization, which is the only cause that either of these fine Germanic gentlemen mentions. There is no authority at all behind my claims – just an IP address. You are being asked to accept the unorthodox monetary theories of someone who has been studying economics for a little over a year. So, if I wuz you, I wouldn’t be too quick to believe them! I believe that if you work through my logic, you will arrive at the same destination – and that is all I can say.

    I’m very curious to hear what results you derive from this exercise, if all this has not depleted your finger strength too much to perform it…

  35. MB — The last thing that interests me is the authority of a speaker. I am arrogant enough to believe I have things worth saying besides having no claim to authority, and arrogant enough to choose for myself who is worth arguing with and whose ideas are worth taking seriously without the distraction of knowing their “credentials”. My fingers have already proven I find you worth taking seriously.

    Here are some quick rejoinders:

    1) My notion of heterogeneous money rests on a widespread expansion of the sorts of things that can be monetized. In particular, as I’ve mentioned, claims on services, which are not storable at all, should be collectable as claims in arbitrary small fractional increments, and exchangeable without friction. Obviously, there are technical/definitional issues here. Service provision is not homogenous. Here I think the recent burst of financial innovation will help us, that some of the ideas behind futures-markets/clearinghouses and ETFs will be of assistance. (In practice, you can buy future claims to lots of nonuniform commodities, from Treasuries to pork bellies, and markets have been defined imperfectly but sufficiently well that the nonhomogeneity of the underlying has not impeded liquid markets from arising. There are lots of futures markets in non-storables as well, energy or weather being extreme examples, perishable agricultural commodities more mixed.)

    2) I think that many savers are more concerned with minimizing risk than maximizing returns. In the present scheme, the two goals are difficult to distinguish, because the returns on “low-risk” investment strategies (gov’t fixed income, even “inflation protected”) relative to volatile prices are such that accepting low “certain” returns implies accepting a large risk that I won’t be able to afford my expected consumption when I wish to redeem my savings. If I could purchase directly, one year of desirable accommodation redeemable any time; one year of food and food service, one year of transporation services at least equivalent to owning a small car, etc. etc. I (and many others) might find these attractive regardless of the volatility of their prices viz-a-viz other commodities. The speculative dynamic you describe (which you describe very well, which I don’t disagree with as a likely outcome given past technologies and current social habits) requires that many savers feel like “failing to keep up with the Joneses” — failing to achieve gains others are achieving — is equivalent to losing wealth, since in the end the economy will be what it is, and wealth will be allocated in proportion to money held relative to others. With money of this sort, the question is not how much I will have when I wish to redeem my savings. What matters is how much I have in relation to others. If I am a billionaire when I turn 80 and wish to retire, I may still be a pauper if others are trillionaires. But if I have collected claims on approximately what I will require in my dotage (or for my children’s education, or anything else), my returns in terms of any money or commodity relative to someone else’s will be of less interest to me than they otherwise might. If enough people choose this “low risk” approach to savings, and if people are willing to make exchanges against these sorts of claims, then the dynamic that leads to a hegemonic money disappears. If my unit of account is Hilton-years, their price in Iridium might be very volatile, might rise and fall dramatically, but so long as I am gathering Hilton-years, I have more reason to be content with my circumstances than I would have been with a growing dollar bank account, and am less likely to act as an absolute maximizer, as required of most participants for “Strategy B” to work.

    Here’s Strategy C: define what future goods and services you want, and make claims on those future goods and services your unit of account. Work and get paid at the “market rate” in whatever currency is on offer, enduring as you must the inflations and deflations and taxes and adjustments that will mangle the value of each paycheck. No matter. Immediately convert all current income, in whatever form its given, into claims against your chosen unit of account, which is precisely tailored to your expected consumption. (There are complications: You’d need to rebalance occasionally as expectations change, and with sufficient wealth, the expected consumption to be purchased incremental revenue will differ from “average” expected consumption, as many of your future consumption goals are already purchased. But these are refinements at the margin.) If enough people follow Strategy C, Strategy B won’t work. Even if we posit that most savers will wish to devote a fraction of saving to speculating, rather than hedging future consumption, speculative money will be drawn to creating capacity to make good on hedging savers’ claims, as this is where the most certain demand arises. Mere speculation on some metal, that is neither especially useful as a basis for exchange, nor especially widely used as a unit of account, will be putting wealth into a very volatile store of value.

    I’m going to go back to liquidity on the main page of this blog now. You’ve gotten me way ahead of myself (I say that with gratitude, not pique). These are things I expected to write about, but was too nervous, and the ideas were (still are) too half-baked to present (to all four of my regular readers, really, it’s terrifying). There are still a lot of holes, and a lot of work to do, but you’ve inspired me at least to overcome the greatest barrier to all intellectual work, nervous inertia, to get a rough draft of an outline down. I do thank you for that!

    When “liquidity surfaces” has surfaced, I’ll try to hit you back on port 80 re: unmold.

  36. err, port 25, that is…

  37. moldbug writes:


    I’ll take the hint – you are officially released from my ugly, rotting teeth ™.

    Your idea of trying to push savings into representative claims on future goods is certainly very interesting. It is certainly better than most of the “amonetary” economic designs I have heard (don’t even start me on LETS).

    As I see it, you are essentially trying to eliminate all long-term indirect exchange, and replace it with futures contracts. Certainly, if you can pull this off, monetary pressure will by my definition be considerably lightened.

    There are a few questions worth asking yourself about this model, though.

    One is how you write futures on goods that don’t technologically exist yet, and cannot even be defined in the present.

    Two is how successful this replacement of futures with savings needs to be to eliminate the monetary pressure that makes my strategy B successful. By my analysis, this pressure needs to be entirely eliminated or at least reduced to a very small epsilon, or followers of strategy B will get way more time in the Hilton than even buyers of Hilton hours, and strategy C will collapse just as strategy A did.

    Three is how these futures are underwritten and insured. The CFTC does not like you to think about this, but futures have default risk – and in most markets, the commercial shorts are practicing the equivalent of fractional-reserve banking, ie, have written far more futures than they can possibly redeem. Paging Dr. Hunt…

    Fundamentally, the choice of a savings instrument is about exchange rate projection over time. Suppose you intend to buy yourself a long vacation in the Hilton, H. You are a rat hunter, you produce the commodity R, rat skins. Your question is: if I exchange R for X now, hold X for 20 years, then exchange X for H, what choice of X maximizes my expected H?

    Your point (2) that since a scalar expected value does not capture the full informational scope of this decision, the word “maximizes” above is a false simplification, is certainly accurate. In the sense of mathematical economics, you have to take the full probability distribution and integrate its product with your utility curve. Sure. So there is some sense in which X = H, or X = some future claim to H, may start with an advantage.

    But, in practice, in real life, very few people plan their retirements in this way. I certainly have no idea what I intend to be doing in 30 years. And many of the goods I will demand will, hopefully, be luxury goods for which my utility curve is relatively linear.

    So the basic problem of saving, in which you compare X to Y simply by comparing their projected exchange rates forward, still probably has to assume a fairly high importance in your model. And in this analysis, it does not matter if X or Y is H itself, or even R (do rat skins keep?), or the usual G. The Schelling point strategy remains a threat – you need some way to make sure people buy these futures rather than flooding into storable commodities.

    And problem four is how you prevent standardization from occurring just as a matter of monetary compatibility, for the purpose of standardizing calculation. I certainly find it hard to imagine a shopping interface at Amazon in which I trade rat skins directly for the mimeographed, right-wing fringe publications to which I am so partial. At a certain level this is just a UI problem, and the net demand it creates is certainly minor compared to the net demand for savings. But you are defining a fairly unstable system here, as I think you realize, and a little bit of standardization pressure seems quite capable of making it go up in smoke a la strategy B.

    Anyway, this seems like enough – you are still working on this and it is not productive, for either of us, to bang on it too hard while it is in a tender and juvenile state.

    I remain quite curious as to your thoughts on unmold. But this has the opposite problem – while not tender, it is dormant. So there is no great urgency.

    Note that in Safari, if you print this exchange out as PDF, it comes to almost 100 pages. I do hope one or two of your other readers has followed along, but even if they haven’t, it’s been fun…