This paper is also available on SSRN as a PDF.
In the decades following the Great Depression, a new sort of monetary explanation of depressions came into vogue, and following the stagflation of the 1970s, gained dominance. It was a more straightforward explanation than those prior to the Depression that had called themselves monetary theories (e.g. Hayek 1933), in contradistinction to “real” theories. The older sort of monetary theory is usually called capital-theoretic nowadays, after its distinguishing feature of systematic capital misallocation brought about by monetary disturbances, a convention I will follow in this paper.
Since that time, capital theories have been eclipsed almost entirely. This is hardly to be lamented: though it would be rash to impugn the idea of a capital-theoretic approach in principle, the canonical explanations had a ptolemaic flavor due to their failure to distance themselves from their foundations in Böhm-Bawerk’s conceptual apparatus (cf. Samuelson 1966; Harwick 2016b). Nevertheless, along with Böhm-Bawerk’s bathwater has been thrown out the baby (babies?) of a number important questions that the more recent monetary theories are ill-equipped to address. In particular, systematic capital misallocation is a salient and much-ignored feature of business cycles, even if that misallocation 1) does not proceed along the lines indicated in the canonical capital-theoretic account, and 2) is not sufficient to explain other salient features of business cycles.
This paper is not an explanation to the effect of “why not both?” periodically offered as a conciliatory gesture from modern capital theorists. Instead, this paper offers an account centering in the dynamics of asset bubbles which subsumes the explananda of both the capital and monetary approaches. To return systematic capital misallocation to its historical place as a central explanandum of business cycle theory serves also to address several methodological shortcomings of the monetary approach as well.
It will be necessary to cover some well-worn ground in order to get to this destination on the other side. The first three sections are therefore primarily expositional; the core argument follows in the fourth and subsequent sections. Throughout, it will be helpful to keep in mind several stylized facts about business cycles, which constitute our explananda:1
The paper refers the first fact to monetary disequilibrium, and the third fact to capital misallocation. All three facts are brought together in an account I call “Mengerian” after Carl Menger’s (1892) groundbreaking account of the emergence of a commonly accepted medium of exchange as the self-reinforcing increase of a particular commodity’s “moneyness” – which I will call its liquidity in the context of financial markets. A similar process can explain the emergence of asset bubbles, a process of which the monetary and capital approaches to the business cycle are but two aspects.
Carl Menger, along with William Stanley Jevons and Léon Walras, is best known as one of the first proponents of marginalism in economics. His legacy, however, is primarily in the emergent explanation of social phenomena, of which his account of the emergence of money is the best known.
Like the more familiar account of institutional change in Demsetz (1967), Menger’s account begins by enumerating the inconveniences of an initial situation (barter exchange), explains how an institutional shift (the introduction of a medium of indirect exchange) would mitigate those costs, and concludes by demonstrating the actual predominance of this institution in our present world. Unlike Demsetz’ account, however, Menger’s does not take for granted the process by which social institutions converge to least-cost practices. Instead, his account proceeds from what we might anachronistically call micro-foundations to demonstrate explicitly that small-scale self-interested action can snowball into a macro-phenomenon which no one intended from the outset.
The primary inconvenience of barter exchange is the necessity of a “double coincidence of wants” – that is, in order to purchase something from my neighbor, I must have something he wants as well. If I want to purchase cabbage from him, and he is in need of a goat, but I raise cows, then we must find another party to the exchange who is willing to exchange cows or cabbage for goats. Needless to say, finding such a willing party for every transaction quickly becomes very inconvenient.
In order to avoid having to hire intermediaries for most transactions, agents in the economy can keep track themselves of the saleability of various goods. By holding an inventory of some relatively more saleable good, even if I have no use for it myself, I increase my chance of being able to overcome the double coincidence of wants and to acquire goods I do have a use for. The key to the story is that my holding such an inventory constitutes a demand for that good, which further increases its saleability. Because the desire of each individual to most effectively satisfy his own wants on the market leads him to hold an inventory of the most saleable good, one good will eventually be converged upon with universal saleability – i.e. money.
The account of bubbles that follows I call Mengerian not only because it explains a macro-phenomenon with reference to self-interested micro behavior. This it does, but more importantly, the key process is again the increase of an asset’s saleability, or liquidity. Two important differences must also be noted, however. First, Menger’s account is one of the emergence of social harmony, whereas the following account is one of the emergence of social disarray. And second, though a self-reinforcing quality is common to both stories, a bubble can obviously not persist forever. It is not a network good in the same sense as a basic medium of exchange. Thus, our story involves the further emergence of countervailing forces which must bring an eventual and abrupt end to the phenomenon at hand.
A money economy, though vastly increasing the ease with which people can satisfy their wants on the market, introduces its own unique problems, which can be seen best in the light of two accounting identities: Walras’ Law, and the Equation of Exchange.
Walras’ Law states that the excess demand for all goods in an economy must sum to zero. That is, unsold inventories of some good implies unsatisfied demand for another. It is related to Say’s Law, which has often been taken to deny the possibility of a general glut or recession. But, as Yeager (1956) notes, “while an excess supply of some things does necessarily mean an excess demand for others, those things may, unhappily, be money.” Because money constitutes one side of nearly every exchange in a modern economy, “an economywide excess demand for money shows up not as specific frustration in buying money, but as dispersed, generalized frustration in selling things and earning incomes” (Yeager 1968). This, of course, is characteristic of recession – our first stylized fact. Likewise, an excess supply of money does not show up as a specific frustration in selling money; it would cease to be money if one could be frustrated in selling it. Rather, it shows up as a general increase in spending – a “euphoria”, or boom.
Putting this in terms of the equation of exchange, MV=Py, will make clearer the macroeconomic response to an excess demand for or supply of money. Suppose, then, that the left-hand side falls (it does not matter whether the supply of money has dropped or the demand for money has increased), and we have an excess demand for money. So far as output has a “natural” level or trend to which it reverts in the long run, this change must result in the long run in a drop in P, the overall price level. But let us suppose that prices are “sticky” – that is, they do not respond immediately to changes in the fundamentals, but fall in the process of working through the excess demand for money. Because the equation is an accounting identity that must hold at all times, real output y must be depressed until P falls sufficiently to clear the market for money balances. This is Walras’ Law in action. By the same token, an excess supply of money implies an inflated y until P rises sufficiently to choke off the excess demand for goods.
In addition, let us suppose that various prices exhibit varying and idiosyncratic stickiness. The phenomenon is well-attested, though there are a number of explanations, all of which likely have some importance: the signals reach some sellers earlier than others (Mises 1966, p. 412ff); some are more alert than others to the signals (Phelps, et al. 1970); some interpret the signals as temporary and try to wait it out (Lucas 1972); there are costs to adjust prices (Mankiw 1985); firms are locked into long-term contracts and simply cannot adjust (Calvo 1983); etc. In any case, the upshot is that monetary pressure on P disarranges relative prices and thus impairs economic efficiency, in addition to temporary deviations from trend output.2
This story will form the basis of our account of the symptoms of business cycles. A general recession and a general boom must be referred to monetary disequilibrium – that is, to an excess demand for or supply of money, respectively. But the monetarist account leaves several important questions unanswered. In particular, the comparative statics mode of analysis means that the initial monetary disequilibrium must be regarded as an exogenous shock. This is, no doubt, often the case, as Friedman and Schwartz (1963) demonstrated of the Great Depression. Decisions of the central bank are usually treated as exogenous, and for good reason (Koppl 2002, p. 123ff). But a central puzzle remains: why do we still experience recessions, even now that central banks so scrupulously avoid falls in aggregate money stocks? In the absence of a fall in M, the accounting identity tells us we must look for a rise in the demand for money – a negative velocity shock. So far as this is the extent of our theory, we must regard it as exogenous. In order to further the explanation, we must weave in the third thread of our account.
The third stylized fact requires a foray into credit markets. The Mengerian twist to the Monetarist story is to take credit markets as a blurring of the line between money and non-money. The usual emphasis in Menger’s story is on the snowball effect by which one good becomes far-and-away more liquid than all other goods. But, Anderson (1917) notes,
the development of money, while it adds to the saleability of the money-commodity, also adds to the saleability of other goods. . . . The fact that goods have money-prices, which can be compared with one another quite easily, in objective terms, makes barter, and barter equivalents, a highly convenient and very important feature of the most developed commercial system. [emphasis in original]
Kroszner (1990) sees this as a continuation of the process Menger described:
The process whereby money’s saleability brings about the increased marketability of other goods may be able to bring about the end of money itself as the medium of exchange. . . . Financial instruments besides money are accepted as the final means of payment, and a growing range of transactions take place without money. . . . More and more acceptable substitutes arise, and an ever growing number of transactions in the economy occur without money as we know it. Various assets – perhaps interest-paying bearer bonds and equity instruments – are used to settle accounts.
More precisely, various assets are employed as substitutes for various services of money. We can – with some amendment – use the textbook characteristics of money as a unit of account, a medium of exchange, and a store of value to generate a taxonomy of money substitutes.
The list, however, is somewhat misleading in that it consists of incommensurable properties. The medium of exchange and store of value are properties of concrete monies. When an exchange is made, it is facilitated with some medium, except in the case of direct barter. When value is stored, it is stored in some actually existing asset. Accordingly, we can speak of a demand for these services of monies. Other goods can also be closer or more distant substitutes for goods we call money, along either or both of these dimensions. The unit of account, on the other hand, is a property of money in the abstract. To say a good costs $5 does not commit one to settling the exchange with banknotes, a check, or deposit balances via a debit card. The unit of account is not a service for which we can conceivably speak of a demand.
The “privilege” of money, of course, lies in its unit of account function, not in its concrete services, for which any commodity or asset will do. Part of the self-reinforcing quality of Menger’s story is that people come to think in terms of some standardized unit of the medium of exchange. As Anderson notes above, this greatly facilitates the exchange of any commodity which can be exchanged against money. But it also allows the issue of provisionally scarce assets which, by fixing their price to the money commodity (i.e. pledging to buy and sell any quantity of the asset at a fixed price), can piggyback on its privilege and come to be used themselves as media of exchange. The vast majority of day-to-day consumer transactions, in fact, have for centuries now been settled with such an asset – bank liabilities, including deposits.
The term “money”, therefore, will be used in this paper to denote media of exchange whose price is fixed by a redeemability promise in terms of the unit of account, which includes the assets traditionally counted into M2. “Monies” is a set of assets which substitute for each other primarily in their capacity as media of exchange.3
In addition to these monies, there are additional “quasi-monies” whose price is not fixed in terms of the unit of account, and which therefore impairs their circulation as media of exchange. Though the most liquid do (as Kroszner argues) sometimes have a somewhat limited circulation as media of exchange within financial markets,4 they substitute for (and have an advantage over, insofar as they bear interest) proper monies primarily in their capacity as stores of value. In order to see their significance for monetary theory, suppose a fall in interest rates renders money a closer substitute for quasi-money. On the margin investors shift into monies as a store of value; the demand for money increases. The reverse is true for a rise in interest rates, hence the familiar negative relationship between interest rates and the demand for money.5
The liquidity of an asset, whether money or quasi-money, we will define as its saleability in the Mengerian sense.6 Specifically, liquidity is the ease with which one can sell an asset at its appraised market value; a joint function of its suitability as a medium of exchange and as a store of value.7 Monies and quasi-monies can be arranged together by their saleability – or more aptly, their liquidity – into an inverted pyramid, with the most liquid money on the bottom (“base money”), and various monies and near-monies built atop it. The quantity of each in terms of the unit of account is indicated by the width at that level. Each one, except for base money, is a liability to its issuer, and an asset to its holder. Mehrling (2012) calls this the “hierarchical” character of the credit structure. Much monetary economics takes account of only two levels of the pyramid, namely, base money and deposit balances, the ratio between which being called (misleadingly) the “money multiplier”. Bank deposits are “pyramided” atop base money, exchangeable at a rate one-for-one with it, though with a far larger supply, due to the practice of fractional-reserve banking.
The hierarchical character of the pyramid of credit is the rationale behind keeping track of multiple monetary aggregates M0, M1, and M2. However, there are two problems with using these aggregates: 1) With the exception of M0, these aggregates add together the value of imperfect substitutes. It is therefore not clear what the aggregate means for the economy. In addition, 2) given the central role of liquidity in our story, it will be more useful to imagine the pyramid extending much further up through the whole array of substituting quasi-monies: bonds, equities, and the securities pyramided atop them.
The quantity of money relevant for our purposes, therefore, is not merely the addition of more asset classes into the quantity of money. A bond’s effect on the volume of spending, for example, is less than a banknote’s, although the aggregate value of bonds is much greater. Accordingly, the logic of the pyramid suggests the use of something like the Divisia aggregates, which weight various asset classes by liquidity (Barnett 1980). Indeed, Serletis & Chwee (1997) have found that movements in Divisia aggregates predict changes in output and interest rates according to the monetarist story better than simple sum aggregates like M1 or M2. This should not be surprising if the analytical core of this paper is correct: it is not the dollar value of some monolithic “quantity of money” which matters for the monetarist story, but the total volume of spending, which is determined by the demand for a large variety of imperfectly substituting assets.
Both within and without financial markets, risk factors negatively into liquidity. And as each issuer is another point of potential failure, risk increases with each level on the pyramid. A security, for example, will become worthless if any of the following happen: 1) a sufficient portion of the underlying assets fails to be repaid, 2) an institution holding a sufficient portion of the issuer’s deposits fails, or 3) the currency in which the security is denominated collapses. Each of these conditions corresponds to a level of the pyramid below the security. A bond, on the other hand, has fewer points of failure, and a bank deposit fewer still.
For those assets near the bottom of the pyramid whose liquidity is based in their usefulness as media of exchange, convenience will be an essential determinant of liquidity. This includes the traditional physical qualities of good media of exchange – divisibility, durability, etc. – but also, importantly, the time path of the asset’s purchasing power if values across time are to be reasonably comparable. Traditionally this is thought to demand a stable value, and is especially important for that asset in terms of which the unit of account is defined. Per Anderson’s argument above, financial markets could hardly exist at all unless denominated in a stable unit of account, hence their observed withering or crowding out in economies undergoing hyperinflation.
But at higher levels, trading is specialized and occurs on a sufficiently large scale that aspects of an asset’s usefulness as a store of value become more important determinants of liquidity. Fixity of value is an important quality for a medium of exchange, but not for a store of value which has the potential to earn interest. This difference cuts loose the asset’s nominal anchor, so to speak. Frankel and Rose (1995) explain the basic mechanism:
Expectations can be described as stabilizing when the effect of an appreciation today – relative to some long-run path or mean – is to induce market participants to forecast depreciation in the future. . . . Expectations can be described as destabilizing, on the other hand, when the effect of an appreciation is to induce market participants to forecast more appreciation in the future.
More aptly, we should call speculation stabilizing or destabilizing, with expectations determining which prevails for a certain asset. To speak of an asset’s “fundamentals”, for example, is to express nothing more than an expectation that the asset’s price will eventually revert to some mean.
Because a unit of account requires a certain stability of value over time, the asset in terms of which it is defined must be subject to stabilizing speculation, at least vis-a-vis the assets against which it is being used as a medium of exchange (Burns & Harwick 2015). Where this is not the case – for example in hyperinflations – the currency quickly loses its usefulness as a medium of exchange, and thus also as a unit of account, even if legal restrictions prevent any other media from being used. For such monies, the switch from stabilizing to destabilizing speculation, for example a bank run or a convertibility crisis, is a catastrophic event. But no such requirement binds financial quasi-monies. The switch from stabilizing to destabilizing speculation is not necessarily catastrophic, and may even signal the onset of a financial “euphoria”. A bubble is one such euphoric destabilization.
Some asset pricing models refer to any deviation from a calculation of fundamentals as a “bubble”. Our own definition will be narrower and closer to common usage: a prolonged instance of upward destabilizing speculation in a particular asset. Precisely what causes expectations to shift from stabilizing to destabilizing is difficult to pin down with much exactness. The “rational bubbles” that occupy the asset pricing literature, for example, where price increases of a sufficient magnitude can “launch” an asset into a self-fulfilling speculative bubble, are not much more enlightening than “animal spirits” or “sunspots” explanations. In both cases we have an external description of a pattern, but no internal insight by which the formation of expectations can be made intelligible at a sufficiently abstract level. In any case, we may satisfy ourselves with the pattern at this link in our chain of events, that a sufficiently abrupt unexpected rise in an asset’s price may induce the expectation of further price rise (Krugman 1991; Lachmann 1956, p. 30ff) and launch the asset into a bubble.
Bubbles in particular assets, of course, most often coincide with general booms. This is to be expected: per the above monetarist account of boom and bust, an increase in total spending (i.e. a rise in MV in the equation of exchange) not only increases real income (y) until prices rise to compensate, but also raises some prices more quickly than others. The most violent price increases we will expect to occur in financial markets, for three reasons:
Stabilizing speculation will take over for many goods at this point. Asset prices increase before goods prices and then fall back somewhat as the new money permeates the wider economy. But – hence the difficulty in chronicling the switch at such an abstract level of analysis – investors often converge on one focal good with expectations of continuing price increases. As a matter of history, bubbles nearly always have an element of “it’s different this time”, an element which is almost by definition impossible to abstract away from its particular circumstances.
This being the case, an initial exogenous spending shock is sufficient both for the initial launch as well as the continuing rise of the bubble. The foundation of the bubble’s liquidity is its suitability as an investment vehicle. Rising demand propels its price higher, and as a store of value it becomes more saleable precisely because its price is rising, quite unlike monies further down the pyramid, whose liquidity is founded in part in a more or less stable value vis-a-vis other goods. The bubble finds ready acceptance so long as its price continues to rise at a rate sufficient to avoid disappointing its speculative investors.
The usefulness of the pyramid construct at this point is in highlighting the formal equivalence of a positive velocity shock at one level of the pyramid with a positive supply shock at higher levels of the pyramid. The total value of the supply of financial assets is generally demand-elastic. Where the price is pegged, as on bank money and other media of exchange, an increase in demand occasions an increase in price. Where the quantity is fixed, as in financial assets further up the pyramid, the price rises with demand. An expansion of the upper layers of the pyramid in response to a general substitution away from money and into quasi-monies, therefore, can be expressed equivalently as a rise in the total money stock, or a fall in the velocity of the narrower money stock.9 Importantly, both expressions are consistent with the Monetarist account.
A bubble therefore represents a tremendous expansion of the broad money supply, over and above the initial increase which launched it, reinforcing its own price rise with a further increase in the volume of spending. Mehrling (2012) argues that during the boom, the pyramid widens (more assets exist at each level), and the layers flatten (assets on different levels become closer substitutes). I would suggest a more apt metaphor for the latter phenomenon is not the flattening of the pyramid, but vertical additions. Closer substitutability between two layers of the pyramid reflects the fact that each level has become more perfectly liquid. This implies as well the addition of further layers at the top,10 as assets which were previously relatively illiquid become sufficiently liquid to substitute for proper monies. The bubble is itself one such asset.
The foundation of the bubble’s liquidity consists in its status as a focal store of value, driven by a rapidly increasing price. This means that, not only must investors eventually be disappointed, but also that the bursting must come as a sudden crash. Once the expectations of investors are disappointed, if its rise in price should slow down, the foundation of its liquidity vanishes at once: it loses its status as a focal store of value, and exits the pyramid entirely.
This in itself represents a significant collapse in the effective money stock. In addition, finding their portfolios on the whole suddenly less liquid, investors in the bubble asset scramble to move their asset holdings further down the pyramid. This depresses both the prices and the liquidity of the assets they sell, causing investors in diverse classes of assets to increase their demand for liquidity as well. The shift in demand down the pyramid leads, in addition to the removal of some of the top layers which are no longer sufficiently liquid to circulate as media of exchange, to the narrowing of the entire pyramid. The combined effect is sufficient to send the economy into recession and manifest the symptoms of the Monetarist business cycle.
This is the answer to our original puzzle: an economy can plunge into depression at the burst of a bubble, despite the best expansionary efforts of the central bank, because an immense quantity of money has either been destroyed or lost its moneyness – a shift which, reckoning further down the pyramid in terms of M1 or M2, will look like a severe velocity shock. As Hummel (2011) notes, the equivalence of a fall in velocity with the rise in the money stock also holds in reverse:
a fall in the broader money stock and a fall in the velocity of the monetary base are exactly the same thing . . . Thus, whether we label a particular decline in aggregate demand a monetary shock or a velocity shock can depend on how broadly or narrowly we define the money stock.
This is certainly the case during a scramble for liquidity. To explain a depression with a velocity shock, especially one so severe as to mark the beginning of a financial panic, is just so much hand-waving. A satisfactory explanation of the event must look further up the pyramid for the assets that have lost their moneyness.
The central bank can, in its lender of last resort capacity, prevent institutions with issues further down the pyramid from failing, and therefore prevent an even more severe contraction as all its monies as well as those pyramided on top of them become valueless, as occurred during the Great Depression when a string of bank failures cut the effective U.S. money supply by a third. In other words, it can in principle prevent the narrowing of the pyramid by providing liquidity (whether it can in fact act fast enough to do so is an open question), but it cannot prevent the exit of the top layers.
None of this, of course, appears in the official monetary aggregates. Taking M0 as the relevant monetary aggregate would lead one to expect heavy inflation since 2008. Taking M2, one gets the impression that nothing at all has been amiss since the turn of the century. It is only when one includes a sufficiently broad array of substitutes for the services of money that the supposed velocity shock of the Great Recession reveals itself as – after all – a shock to the quantity of money.
A focus on bubbles and broad monetary aggregates can also subsume the capital-theoretic observation of systematic capital misallocation over the course of the business cycle. Industries whose investments are susceptible to bubbles are not necessarily “higher order” – if such a term can even have a meaning in an economy sufficiently advanced in the division of labor – though such industries as are usually considered higher order may have characteristics that make assets based on their investments more susceptible to launch, in particular a more violent price response (contemporaneous or lagged) to shifts in demand. Instead, investment flows into the bubble asset and funds investment in that area. A bubble in stocks funded overinvestment in mechanization during the Roaring 20s. A bubble in dot-com stocks funded overinvestment in internet startups in the late 90s. And most recently, a bubble in mortgage-backed securities funded overinvestment in residential housing during the 2000s.11
If this is the case, then it will be fairly straightforward in principle to separate out the effects of capital misallocation from those of depressed aggregate demand, and thus to delimit the applicability of the monetary and capital explanations. A general recession must be referred to an insufficient effective money stream. Downward pressure on prices originating from insufficient spending power must have the effect of depressing economic activity, and must do so more or less evenly across all industries, with due allowance for correlations of consumption patterns with the distribution of demand for money (Harwick 2016a). The magnitude of misallocation, on the other hand, can be gauged by the eventual crash in the bubble’s price and by the time necessary for the affected industry to return to normal operations – or, if a general depression also prevails, to a level of operations in line with the rest of the economy.
Business cycles are traditionally seen as the domain of macroeconomics, and bubbles as the domain of finance. While macroeconomics has incorporated much from finance in the last several decades, especially as the relevance of financial markets came to be appreciated more fully, bubbles have been viewed at best as a small aspect of business cycles – as symptomatic, but not central. This account, by contrast, places them at the center of the story, as leading the process of economic fluctuations. The same sort of theory that explains the emergence of a single money can explain as well the emergence of further monies and their dynamics over the course of the business cycle.
This analysis suggests many lines of further research, first of all in economic history, linking the development of liquid financial instruments with the emergence of a business cycle, and identifying the particular shocks that effect the switch in particular bubbles from stabilizing to destabilizing speculation. In addition, the criterion by which misallocation may be distinguished from excess demand for money must be econometrically operationalized in order to determine the relative significance of each, and to what extent they may be correlated. This division, in turn, bears on the risks of reflation. To the extent that an excess demand for money dominates, liquidity expansion will ease the monetary pressure on prices and speed recovery. To the extent that systematic capital misallocation dominates, expansion risks a destabilizing demand shock which may launch another bubble. Some indeed have worried this was the case when the Federal Reserve flooded the market with liquidity following the collapse of the dot-com bubble, only to face the collapse of a far more severe housing bubble less than a decade later. If both factors are significant, the central bank may face a tradeoff between depression and misallocation that can be improved – perhaps only over the course of several business cycles – by a credible commitment to the stability of the currency’s exchange against a well-accepted anchor (cf. Kydland and Prescott 1977).12