Navigation – Plan du site

Louis-Amable Petit and the 19th Century Literature on Crises and Cycles

Louis-Amable Petit et la littérature économique du XIXe siècle sur les cycles et les crises
Daniele Besomi
p. 208-237


Cet article étudie la théorie des crises développée par Louis-Amable Petit (1868) relativement aux autres théories semblables alors présentes dans la seconde moitié du XIXe siècle. Adversaire de la loi de Say, Petit propose une explication endogène des crises fondée sur les fluctuations de la part du revenu qui est consommée, par contraste avec une incitation continue à l’expansion de la production. Alors qu’à cette époque l’approche universelle du problème des crises y voit la liquidation des excès de la prospérité spéculative, sans être en mesure d’identifier un mécanisme de reprise économique, Petit clôt la boucle causale. En effet, chez Petit, le point de retournement de la conjoncture est une conséquence de la crise et de la baisse des revenus qui en découle. Par conséquent, Petit est l’un des premiers économistes à avoir proposé une théorie complète des cycles.

Haut de page

Texte intégral

1. Introduction

  • 1 Simonin has already supplied a biography of Petit and a bibliography of his writings, and I can saf (...)

1In a previous issue of Œconomia Jean-Pascal Simonin (2011) has skillfully presented the views on money, free trade and crises developed by Louis-Amable Petit during the 1850s and 1860s. Petit’s life and some aspects of his work have been thoroughly examined. Others, however, deserve further discussion, which this paper provides as a complement to Simonin’s.1 Although this article is self-contained, I highly recommend a preliminary reading of Simonin’s portrait of Petit’s life and works.

  • 2 Translations from French are mine. As Petit’s language is sometimes loose and not always consistent (...)
  • 3 As I am interested in this very fundamental level of interpretation, I refrain from comparing in de (...)

2This paper focuses on Petit’s theory of crises as expounded in La richesse par excellence (1868).2 Section 2 offers a brief outline of Petit’s argument. Section 3 discusses the conceptual fuzziness of some categories Petit refers to, in particular consumption, investment, and saving. This lack of precision makes Petit’s rejection of Say’s law somehow fragile on the analytical side. It is, however, argued in sections 4 and 5 that Petit’s theory of crises and their cyclicality has surprisingly modern features, and bears (apart from the conceptual naiveté) an astonishing resemblance with the nonlinear multiplier-accelerator mechanism proposed by Harrod in 1936. This similarity is pointed out not as a gratuitous search for forerunners, but in order to stress how, unfettered by Say’s law, one could devise original explanations of the cycle several decades ahead of later discoveries. Section 6 places Petit’s theory in the context of his contemporaries’ understanding of the relationships of crises and to cycles with the “normal” working of the economic system.3 Finally, section 7 briefly assesses Petit’s contribution against this background.

2. Petit’s Theory of Crises: a Brief Outline of the Argument

  • 4 Petit argues that exchanges would be formally equivalent to barter if the sole aim of producers wer (...)

3Petit’s theory of crises can be briefly summarized. It relies on the premise that precious metals are wealth in its quintessential form, la richesse par excellence. Petit defines wealth as ‘any product that is used to satisfy one’s needs’. Precious metals have a utility of their own, and as such have an intrinsic value; they are used as means of exchange, due to their durability, malleability, divisibility, etc. Being inalterable, they are the only form of wealth that can be preserved indefinitely, and as such they are the only means to set aside one’s wealth in view of future needs (1868, 6–7). Accordingly, precious metals play a double role, as means for transaction but also as a store of value. In recognizing this double role, Petit explicitly dissents from the économistes, i.e. the French liberal school (Bastiat, Rossi, Chevalier, Garnier, Clément, Baudrillart and others) and the authorities they rely on (Quesnay and Smith) (9); in particular, Petit rejects the idea Say’s law is based on, that exchanges are formally equivalent to barter and money is nothing but a means for transactions,4 which leads him to refute the law of markets, its premises and its implications on crises.

  • 5 Petit is adamant that the problem of overproduction crises is typical of capitalistic production: “ (...)
  • 6 The Sismondian theme of the increased production due to the application of more productive machiner (...)

4If the sole purpose of production were to satisfy the producer’s need (as, in part, it is), exchanges would be equivalent to barter: money would only serve as an intermediary for exchanges, and production would create its own market. Say’s law would hold, and overproduction would be impossible. In a capitalist economy, however, the purpose of production is to accumulate and preserve wealth, in the form of precious metals.5 In order to do so, one has to produce more than one consumes. Production is therefore pushed as far as possible, by means of more and more powerful machines. Up to a point, consumption keeps the pace with the increasing production. But as people become wealthier and primary needs are satisfied, they begin to save up to meet future needs. An excess of production over consumption, the sole cause of most commercial crises, inevitably follows (Petit also considers circumstances than can affect the balance on the consumption side, such as wars or famines; the latter was the case of the 1847 crisis).6 During the crisis production stops or slows down, workers are laid off, income decreases, and so does consumption. Nevertheless, some classes of consumers carry on consuming in reasonable amounts; after a few months the excess of production is therefore absorbed. It even causes scarcity, because production has slowed its pace, thereby inducing some producers to resume their activity. This stimulates the entire economy into another boost of production that can continue in so far as consumption keeps the pace.

5The general idea is plain enough. Nevertheless, its context and a number of details are worth discussing. Some have already been stressed by Simonin and can be referred to briefly, while others deserve additional reflections.

3. Consumption, Saving, Hoarding, and the Lack of a Theory of Investment

6Notwithstanding its apparent simplicity, Petit’s argument is based on a conceptual fuzziness that makes it difficult to grasp its exact meaning, in particular regarding the interconnected categories of consumption, économie and investment.

7Petit’s notion of “consumption” is ambiguous. In most passages he is clearly using “consumption” as consumption by final users, in particular when he writes about people having satisfied their basic needs. But elsewhere he seems to include also the consumption of capital goods incorporated in the production of consumption goods (if we exclude that iron is used for consumption purposes): “It is necessary that production be adjusted to the needs of consumption, to the market’s needs. Thus, producing iron there is no demand for, and building houses in a city where there are already too many and rents are declining, does not increase wealth and is not a good way to use money. For production to be useful, it is necessary to adjusts it to the needs of consumption and meet its requirements.” (Petit, 1868, 406) Elsewhere, however, he explicitly excludes such an interpretation, for instance when he describes industrial equipment as “a producer that produces as much as 100, 1,000 or 10,000 producers, but only belongs to one consumer. This automatic producer, that produces sometimes immense amounts of goods, and surely does so in a blind way, does not consume” (301).

  • 7 Precious metals “are, so to speak, an active form of wealth: they are the form of wealth leading to (...)
  • 8 This applies to individuals as well as to societies: “Precious metals are thus the real foundation (...)

8If we consider, following the latter statement, consumption as referring only to the purchase of consumption goods, Petit’s reference to the “equilibrium of production and consumption” (or lack thereof) clearly leaves investment out of the equation. This is one of the most troublesome aspects in his view of the way the economic system works. Petit is surely aware that the mass of more and more productive capital goods increases constantly, and indeed he places this process at the heart of his crises mechanism. He is also aware of the fact that the acquisition of new capital goods is financed by means of the producers’ savings. He describes in detail, in fact, how the accumulation process can only take place thanks to the existence of precious metals, that allow producers to preserve the wealth resulting from the difference between their production and consumption and acquire new and more productive machines capable of increasing production.7 In societies where there are no precious metals, the only means permitting to give a permanent form to the excess of production to consumption, “if nothing can be saved, it is also not possible to create the immense production means that can be produced or bought by societies made wealthy by gold” (1868, 298-299).8 While Petit is aware that investing savings continually in new machinery is necessary to push production further than consumption, he does not, however, formulate an explicit investment theory and does not seem to realize that investment, besides adding to the productive capacity of society, also absorbs part of its production.

  • 9 Again, Petit’s conceptual system is quite ambiguous. Économie, the term he used the most, is the la (...)

9Finally, in Petit’s argument it is not clear what happens of the économies consisting in the unconsumed part of production. It seems that his producers are torn between behaving as capitalists, using their wealth to produce more wealth, and behaving as misers, refraining to spend and hoarding their money. Petit maintains that the purpose of production is the accumulation of gold, clearly in the treasure form: producers “consider only money. Surely they will devote part of their production to feed their industry and satisfy their personal needs. But the profits they obtain are in monetary form, or some substitute for it; they seek money, and they obtain money” (317), for “These metals are so seductive, offer such advantages, give rise to such solid fortunes, and enable one to set up assets eventually procuring fixed and solid incomes, that everyone does all they can to acquire them” (300). Yet the means to accumulate wealth consist in producing more and more, relying on increasingly powerful machinery in order to widen the gap between production and consumption: “One must produce more than one consumes and deliver to those who already possess [precious metals] the goods they desire and need. The more products one can deliver, the more precious metals one can obtain.” (300). Producers pay for their machinery “with the money they have made by selling their previous surplus” (300). The wealth acquired by selling their new products “gives [them] the means for expanding their means of production indefinitely” (300). For that reason there is a tension between the desire of hoarding and the necessity of investing all that is not consumed:9 “On the one hand, the desire of increasing their wealth induces producers to produce as much as possible; but on the other hand the desire of economizing induces a large number of consumers to limit their expenses.” (268) Crises result from this contradiction: “These opposite tendencies of production and consumption necessarily determine a glut.” (268)

10The overproduction problem only arises after a time of prosperity, when production has expanded enough after the previous crisis to increase the affluence of most people who, having fulfilled their basic needs, begin “to economize, to build up reserves for the future” (291). Who are these people, now setting their savings aside? “Everybody aims at saving, so that necessarily everyone tries to produce more than they consume.” (265) But the rentiers are excluded, as they do not produce and are already spending their accumulated wealth. The workers save very little, as their income scarcely exceeds their basic needs (266–67, 315). Those responsible for the largest part of the savings are

the producers who produce to economize, the ones producing more than they consume. They are the seeders of the crises. Most producers are in this condition, whatever the scale of their production; almost all of them contribute, each for his part, to determine these upheavals we are witnessing periodically, so to speak. However, the producers who contribute the most to the production excesses are no doubt those managing large factories and powerful equipment (Petit 1868, 347; see also 315–16).

11Petit specifies that even if “the consumers who are not owners of machines [this presumably includes merchants] spent their entire revenues and profits in products manufactured by machines, they could never buy them all” (316).

  • 10 “...periodically, at times more or less close, depending on the incidents affecting production or c (...)

12Here the gap in Petit’s account, by leaving out an investment theory and especially failing to consider the fact that investments not only add to the stock of capital goods but also to the demand for industrial products, apparently affects the cogency of his argument and prevents him to effectually reject Say’s law. A critic of Petit’s could argue that if savings were systematically and fully devoted to the acquisition of new machinery (producers “employ in trade and industry ever increasing amounts of money as wealth increases” (267)), there is no reason why the equilibrium between production and demand should necessarily be broken—think for instance of Tugan-Baranowsky’s interpretation of Marx’s reproduction schemes in the spirit of Say’s law, where he stretched the schemes to a paradoxical economy where only one worker attends to machines producing other machines (Tugan-Baranowsky 1905, 230, and 1913, 216–217). Ricardo’s dictum that “No man produces, but with a view to consume or sell, and he never sells, but with an intention to purchase some other commodity, which may be immediately useful to him, or which may contribute to future production” thus remains unchallenged, and his conclusion that “By producing, then, he necessarily becomes either the consumer of his own goods, or the purchaser and consumer of the goods of some other person” is not invalidated by Petit’s argument (Ricardo 1817, 1951 ed., 290). Petit and Ricardo rely on opposite assumptions as for the utilization of savings: Ricardo and all of the supporters of Say’s law take for granted that there is no sense in hoarding money, so that everything that is not consumed is invested and there cannot be any problem (except temporarily and partially, and due to miscalculations or external accidents) in selling the entire product of the country; Petit assumes instead that periodically10 there comes a point when savings are not invested but hoarded, inevitably glutting markets and generating commercial crises. Neither could prove their point: Petit supplies a sufficient but, due to the lack of an investment theory, by no means necessary, reason why capitalists may refrain from reinvesting their profits; while Ricardo does not admit that investment may turn out not to be profitable. For a composition and a criticism of both such views at once we will have to wait for Marx’s reproduction schemes (De Vecchi 1983, 234–5).

4. A Moving Equilibrium

  • 11 d’une manière continue (267).
  • 12 “When nations work more, they have more products to consume, and they

13Although Petit’s approach remains anchored to a primitive rejection of Say’s law, it is of interest for several reasons. The first one concerns his explicitly taking as a reference path a moving (though unstable) equilibrium of production and consumption. He repeatedly points out that as producers become wealthier more and more money is invested in machines; accordingly, production expands steadily.11 This process further increases the society’s income, and stimulates consumption.12 Up to a certain point, consumption keeps the same pace of production, until the desire to save (i.e., hoard) slows it down or even halts it. There is a “normal” moving level of consumption which Petit takes as a reference path. It is defined as follows:

One can think of an average level of consumption, a normal consumption so to speak, given the society’s wealth. This is the situation when nations are in peace, when bread is sold at a reasonable price, when trade and exchanges proceed at a regular pace, when prosperity advances and increases. This average level increases [usually at a slow pace] year after year, day after day as the affluence of the nation grows and as the sum of precious metals—the source of all durable wealth—grows and favors the development of general wealth, be it via foreign investments, be it via the creation of new fortunes inside the country, or via the expansion of profits and revenues. (Petit 1868, 275)

14A normal circulation of credit, namely, the amount of credit reflecting the purchasing power of consumers, is associated to this normal level of consumption (303, 350, 365). If credit exceeds its normal level, there is too much money in circulation, further stimulating the acquisition of new machinery, and thus an increase in production that will inevitably exceed the normal level of consumption (360, accepting this part of the overtrading theories). Note that the excess of credit is not necessarily created by the banking system but also by the mutual borrowing and lending among manufacturers and traders, often in the form of payment deferment (346 and 360). To normal consumption and the corresponding normal state of affairs is associated also a normal price level. During a crisis, when overproduction becomes apparent, “the supply of products exceeds demand, and prices fall below, and often much below, their normal level” (320).

15While the believers in Say’s law argued that production dictates the pace of advance by creating the purchasing power that will be used to buy the produced goods, Petit —like all adversaries of Say’s law— turns the relationship on its head and argues that demand (as represented by consumption at its normal rate of increase) determines the possible rate of growth of production. If production adapted to this normal level of consumption, it could proceed indefinitely; if instead it exceeds it, the system is glutted:

When affairs follow their ordinary course—namely, if there are no troubles, wars, political unrest—and if consumption follows its regular pace, there is overproduction if producers, rather than taking into accounts consumers needs, produce without assessing them and in an exaggerated way. (276)

There is a crisis because production, rather than ascertaining the needs of consumers, keeps manufacturing piles of products that cannot find the outlet they need and for which they are created. (335).

16Overproduction is thus defined with reference to the normal advance of consumption. It can arise in two ways. Either consumption is kept below its normal level by some abnormal circumstances such as a scarcity of food, which increases the price of primary goods and therefore absorbs an unusually high share of resources, depriving other businesses of outlets (276–7, 282, 287); or consumption proceeds at its normal pace while production exceeds it, due in particular to the tendency of production to expand continuously. Petit’s discussion of the first case is not particularly interesting, while the second has some intriguing implications.

17Petit does not define the normal pace of advance of consumption in a rigorous way. In the previously quoted passage (the most explicit one on the subject) he only specifies that it is somehow related to the society’s increasing wealth as permitted by the available amount of precious metals. However, it is clear that in Petit’s view the relationship between income and consumption is not linear. As income increases beyond the threshold at which basic needs are satisfied, producers start spending less than their income and hoard the difference. The proportion of consumed income thus decreases as income increases. Conversely, when income decreases in consequence of a crisis, consumption diminishes but at a slower pace, as holders of fixed incomes and people who can afford it keep consuming and are actually induced to consume more by the drop of prices below (sometimes very much below) their normal level (321). This is what Keynes later called the “fundamental psychological law” that consumption increases when income increases, but less than proportionately (Keynes 1936, 96–97).

5. The Instability of the Moving Equilibrium

18Here is where things become really interesting, because all in all Petit’s argument shows an astonishing resemblance to Harrod’s discussion of the trade cycle in 1936. Harrod started from the concept of a moving equilibrium, resulting from the interaction of the multiplier and the acceleration principle: the multiplier explains how income increases, depending on the volume of investment and the propensity to save, while the acceleration principle (which he called “Relation”) explains investment in terms of the needs of capital goods to produce a given quantity of consumption goods. Given the propensity to save and the technology in use, there is a “justified” rate of growth of income, consumption and investment such that the increase of consumption due to the rise of income generated by the ongoing investment proceeds exactly at the pace expected by entrepreneurs when they formulated their investment plans. If investment and consumption followed this path, there would be no reason to abandon it. But neither the technology nor the propensity to save are given once and for all. On the one hand, new investments give entrepreneurs the chance to introduce technologies increasing the proportion of capital per unit of output, so that additional investment increases output less than in proportion. On the other hand, the propensity to save depends on income distribution—which tends to move towards of profits in the ascending phases of the cycle—and on Keynes’s psychological law. Both factors suggest that the proportion of income consumed (and thus the multiplier) decreases in the ascending phases of the cycle and increase in the descending ones. Harrod’s moving equilibrium is thus unstable: even if the system found itself on the “justified” growth path, the very increase of income would cause a decrease of the propensity to consume. Consumption would thus slow down (or decrease altogether), thereby disappointing the entrepreneurs’ expectations on which investments were based, thus inducing them to slow down investment with consequences on income and on consumption, in a cumulative downward movement. This is first slowed down and eventually halted and reversed by the decrease of income itself, as it brings forth an increase in the propensity to consume and thus in the multiplier, and due to the decrease in the interest rate, which stimulates investment and the adoption of new techniques (Harrod 1936, Ch. 2; for a discussion see Besomi 1999, 72-77).

  • 13 Note that if savings were not hoarded but invested, they would create further income, part of which (...)
  • 14 “There is a certain class of consumers who continue to consume like before the crisis; other classe (...)

19Petit also starts with the notion of a moving equilibrium, ill-defined as it is for the lack of explicit discussion of the determinants of investment. As a society becomes richer, at first consumption advances along its “normal” line. At some point, however, having satisfied their basic needs, people begin increasing the proportion of income they save. As the corresponding portion of income is hoarded rather than expended, production exceeds demand, and the economy becomes glutted.13 Production comes to a halt, and income decreases. At first consumption decreases as well, but at a slower pace: the savings hoarded in the last stage of prosperity are spent, and rentiers keep consuming.14 Production excesses can thus be liquidated, entrepreneurs turn optimistic again, and a new expansionary phase can begin.

  • 15 Petit uses the word reprise (also elsewhere in the text, e.g. 4, 289-90, 345, 367, 481-2, 538-9). T (...)

Ordinarily, when there is manifest recovery15 in business, trade is profitable because there are many unsatisfied needs. The producers who have weathered the crisis, a smaller number than before due to the number of those who failed, increase their means of production. Soon new producers, excited by the prospects of gain, also join the fray.... At first, consumption easily absorbs the production on offer, because of the abstinence it was compelled to. It even increases gradually as wealth and the means of exchange grow. Producers see their horizon widen, and keep expanding production accordingly. Production increases due to both the development of machines and the arrival of new producers. But there comes a time when every consumer, thanks to the abundant production, has purchased all his necessaries except for food (which must be bought almost daily). He finds he has enough furniture, clothes and household utensils for some time and for his stand. Thus he aims at saving and building some reserves for the future. At this point, consumption declines, or remains stationary, or progresses slower than production. When economizing begins, at first the movement is not fully apparent, and production continues unabated. At this time, then, production and consumption start going, if not in opposite directions, at least on paths that cease to run in parallel for a while. What is more dangerous at this point, is that production can supply a mass of products that is more substantial every day, and indeed it does so. In this way, there is inevitably an excess, an overflow, a glut. Consumption builds an insurmountable obstacle to production. It forces products to ebb up the channels of production, so that the market, as it is flooded, suffers frightful losses. (Petit 1868, 290–1; a similar description is given on 267; some passages are quoted in the text)

  • 16 This is not necessarily to be interpreted as an investment function, a concept extraneous to Petit’ (...)

20On the supply side, again Petit is again not too clear. Production—he is only interested in the production of consumption goods, while the construction of capital goods is left out of picture—seems to depend on the number of producers and on the output of consumption goods per unit of machinery. “Number of producers” of course does not help, as what really matters is the amount of capital (or units of machinery) they activate. If we interpret Petit along this line (admittedly stretching his words beyond what they explicitly express, but arguably without contradicting them), we have that the output of consumption goods = capital volume multiplied by a variable coefficient indicating the units of consumption goods produced per unit of capital and depending on the technology in use. In equilibrium, where production increases at the same pace as normal consumption, with consumption dictating the pace, given the state of technology (and, one should add, the rate of utilization of capital) the rate of growth of capital (i.e., investment) must be proportionated to the rate of growth of consumption. This is nothing but a mechanical formulation of the acceleration principle, determining the required relationship between investment and rate of growth of consumption if a moving equilibrium is to be preserved.16 This is subject to three provisos. The first one is that consumption at some point begins to slow down (see below), so that investment should adapt. This explains how a discrepancy between production and consumption can arise from the intrinsic dynamics of the economy. The second proviso is that Petit is envisaging a lag in the entrepreneurs’ reaction to changes in demand, which is responsible for the persistence of the original deviation from equilibrium and for the widening of the gap, at least for a time; the crisis is a violent reaction to an excess that has become unsustainable. The third one is that capitalists constantly try to increase the output of consumption goods per unit of machinery. This alters both the ratio of output per unit of capital and the technical relations of production between branches producing capital and those making consumption goods, a discussion Petit’s text is completely lacking.

  • 17 The same transmission and amplification mechanism of troubles is also described by Bagehot, a few y (...)

21On the side of the demand of consumption goods, Petit’s discussion is clearer. Interestingly, we find not only an early expression of the fundamental psychological law, but also a clear understanding of the relationship between production and income in terms of a (non-numerical) multiplier process. While at Petit’s time it was already quite common to observe that troubles (or, conversely, progress) in part of the economy tend to expand to other branches of production, this was normally explained in terms of the intersectoral exchanges among producers: if a business goes bankrupt, all its suppliers of machinery and raw materials inherit part of its difficulties, in terms of loss of a customer and of default on debt, and in turn spread them to others. Petit is naturally aware of this mechanism (e.g., 290), but adds another one: salaries and profits from each business are spent in other branches of the economy, igniting production there and generating further incomes that in turn are going to be spent.17 On the downward side, one of the first consequences of overproduction is that production slackens, thereby creating unemployment. Then, “The first effect of this labor slow-down of is to reduce the purchasing power of the worker, since his wage decreases, and wages is the measure of consumption.” (309) This decrease of consumption is transferred from one branch of trade and manufacture to another, and eventually most industries are affected by it (311-12). Conversely,

Just as the distress is communicated from business to business when the crisis hits, in the same way consumption acts from trade to trade by promoting a favorable movement in business. The demand of consumption goods is addressed to the producer who, in turn, transmits it to the worker. The latter gets more work to do and, as a result, obtains a wage he spends immediately and in full to buy the things of which he has a pressing need after the sufferings he endured [during and after the crisis]. The shopkeeper sees customers in his store again, makes profits, resumes his purchases, and is able to go back to his consumption habits. Consumption spreads. Production follows this movement, and soon all industries pick up their course.

When consumption grows in the lower ranks of society, the increase is communicated progressively to all degrees of the production scale up to the highest peak; similarly, when consumption stops at these same levels, the consequences of this interruption pass on from one producer to another and are felt in the entire commercial and industrial body. (Petit 1868, 290)

22This “multiplier” mechanism is, needless to say, incomplete: Petit does not ask himself whether there are any leakages (he even excludes that workers can save in the early recovery stages) or how big the cumulated effect of successive expenditure rounds are, and he is even less interested in disputing the traditional causal relationship between saving and investment. While it is insufficient to express a precise relationship between the hiring of the first workers and the creation of income, this mechanism is nonetheless sufficient to explain the cumulative nature of growth. And when combined with the relationship between capital and the output of consumption goods, it enables one to conceive of a self-sustained pace of growth that may be, and actually is, disturbed by changes in the technology and in the proportion of income spent.

6. The Rejection of Say’s Law: Petit’s Position in the Development of Crises and Cycles Theories

23A second set of reasons why Petit’s approach to crises is of historical interest in spite of its fuzziness lies in its position in the development of crises and cycles theories of his time.

  • 18 I would, however, differ in emphasis from Simonin. He reports a passage from Petit’s book where he (...)

24Simonin has illustrated well how Petit explicitly rejected Say’s law18—the very subtitle of his book, Réfutation de la doctrine des économistes, refers exactly to this (as well as to free trade), les économistes being the French liberals writing in the Journal des économistes, staunch believers in free trade and in Say’s law and its corollaries. This approach was rather exceptional in France and in Britain, but more common and not undignified in Germany. In France, disputing the law of markets was a taboo and was guaranteed to create obstacles to publication, as the publication outlets were controlled by the liberals (indeed, Petit’s book was published chez l’auteur). In Britain, the endorsement of Say’s law by Ricardo and (not unambiguously) by Stuart Mill discouraged research along this line, and the opposition to it was confined to a fringe of heretics. In Germany, on the contrary, the law of markets was subject to academic debate and open criticism since the 1820s. This originated a different understanding of “crises”, different families of explanations, and even different definitions of the term.

25In spite of the multiple meanings of Say’s law (see, e.g., Baumol 1977), there was general agreement among both its supporters and adversaries that it stated the logical impossibility of overproduction.

  • 19 Petit summarizes the position of the économistes as never wanting to admit the existence of gluts a (...)
  • 20 The locution is Say’s, in the passage where he stated that crises can only be partial and temporary (...)
  • 21 Petit was aware that the acceptance of Say’s law forced economists to explain crises in terms of mo (...)

26Petit, for instance, quotes (397) Laveleye stating that “Crises are never brought forth by an excesses of production. We cannot admit a general excess of production, because, in this case, all products would be exchanged for others, as before, with the difference that everyone would have more of them.” (Laveleye 1865, 437; Petit, however, attributes the quote to Rossi)19 If overproduction can only be partial, self-righting and thus short-lived, only a limited number of explanations of crises can follow. Either the law is taken seriously, and crises are attributed to miscalculations, frictions or delays in the adjustment mechanism, exogenous accidents, or “political diseases”;20 or else Say’s law is circumvented, without being explicitly denied, by blaming the monetary side of the economy rather than the real side. If, instead, one admits the possibility of general overproduction, a different range of explanations can be envisaged, based on any mechanism that can get production and demand of goods out of gear.21

  • 22 Also in Italy where, however, the literature on the subject was scarce.
  • 23 The most elegant theory of crises belonging to this family, however, is due to a Frenchman, Charles (...)
  • 24 The most interesting examples in this tradition took the form of cobweb-like mechanisms. Around 181 (...)
  • 25 For references to Petit’s discussion of this approach see footnote 21.

27Accordingly, most explanations of crises elaborated in France and Britain in the first three-quarters of the nineteenth century were of the first kind.22 In France, where the grip of Say’s law was stronger, the favorite explanations relied on external causes or on state interference, especially in the first half of the century. Crises were seen as disconnected events, with nothing in common except for the distress they caused. In Britain, where the law of markets was not treated with the same deference, the most common approach relied on the Central bank’s mismanagement that could exacerbate the natural fluctuations in prices (this position was held, from different sides, by both Currency and Banking schools theorists) or, by means of its monopoly position, could create fluctuations and crises (free banking school23), or on the amplification of accidental fluctuations in trade due to institutional constraints,24 or on the credit-overtrading-speculation mechanism. The writers belonging to the overtrading tradition—by far the most common approach throughout the central part of the nineteenth century—maintained that during prosperous times traders ask for more credit than their own capital would warrant, as the larger their business the more profit they could obtain; the banks, on the other hand, are happy to lend them the money they ask for, as their profits depend on the amount of lent money; this increases the volume of trade and prices, which in turn feed speculation based on more credit, until the system collapses when, for any reason, trust is shaken. The latter approach, which gave rise to the most detailed explanations (for instance in the hands of William Huskisson, who seems to have been the first proponent of this theory in 1810, and with the endorsement of J. S. Mill: 1848, Book III, Ch. 12, § 3), eventually convinced the best French writers in the second half of the century, when the regularity of the recurrence of crises and their morphological similarity had made more and more difficult to support the view that crises were the result of accidental and disconnected events.25

  • 26 To honour the truth, one should point out that most of the literature on crises during the first ha (...)
  • 27 While it seems true that Ricardo’s view conquered England as the Holy Inquisition conquered Spain, (...)

28In France and Britain dissenters were a small minority. On the West side of the Channel, after the settling of the gluts debate and the defeat of the positions of Malthus, Chalmers and Lauderdale, over-productionist and under-consumptionist views are found only in the obscure26 writings of a few socialists, in particular Owenites (see e.g. King 1981), and return only at the end of the century with the writings of Hobson (and A. D. Wells in America) and a few minor figures. On the continental side of the Channel, Petit and Auguste Ott were the two only relevant exceptions to orthodoxy after Sismondi. In Germany the picture was rather different. Although the literature on crises was not as abundant as in France and Britain, towards the middle of the century writers critical of Say’s law were more numerous than those mentioned above, and some had a definite prestige among German academic economists: among the latter Rau (in his early writings), Roscher and to some extent Schäffle; others include Michaelis, Dühring, Rodbertus, Kirchmann, and Marx of course, whose views thus had roots in a consolidated tradition of dissent. This substratum of literature by the end of the century nurtured the writings of the authors participating in the transition to business cycle theories in the modern sense of the term: Herkner, Lexis, Tugan-Baranowsky and all the authors engaged in the debate on crises and the Marxian reproduction schemes, Schumpeter, Spiethoff, Bouniatian and many others.27

  • 28 For a discussion of the definition of crises in nineteenth century dictionaries, see Besomi 2011b, (...)

29This difference in approach also gave rise to different definitions of crises. The writers in Say’s tradition either defined the phenomenon in terms of some of its features (crises as “bad times”, the stoppage of payments, or when prices cease to rise), in terms of an unspecified “disturbance”, or as the consequence of such a perturbation. The stress was on crises as an anomaly, and the prevalent metaphor associated to them was an unhealthy state of affairs, explicitly implying that normally the system tends towards a healthy condition unless it is disturbed, and after being disturbed it recovers quickly. What is “healthy”, however, was not defined precisely. The notion rested on a circular understanding of “health” in terms of the absence of disease—just like medical dictionaries around the middle of the nineteenth century defined “disease” in terms of absence of health without carrying entries defining “health”. Writers who were prepared to think of crises in terms of an excess of production over demand (whether arising on the demand or on the supply side) also defined crises in terms of (or as the consequence of) a perturbation, specifying however that they meant a disturbance in the equilibrium between production and demand—implying that such a disturbance rather than self-righting feeds back the mechanism that had generated it.28

  • 29 Explicit definitions of “crises” were rather rare at the time; most of them were confined to dictio (...)

30In this context Petit’s anomaly is rather remarkable in several respects. Firstly for the very fact—extremely rare in France—of explicitly rejecting Say’s law. Secondly because it provides a (implicit29) definition of crises as the result of “the shock suffered by production, pushed to its extreme limits but knocking against the obstacle of shrinking, stationary or at any rate slower consumption” (Petit, 1868, 267-268). The equilibrium between production and consumption is clearly the reference notion:

The phenomenon of exchange follows three distinct stages. At first, the needs of consumers are larger than those production can meet; later, there is equilibrium; and then there comes a time when production, for the reasons we have indicated, oversteps consumption. It is at this point that crises begin. It is therefore certain that commercial crises are, as a general rule, determined by a glut of products, by production excesses in comparison with the needs of consumption. (273)

An excessive mass of products compared to the currency that circulates and to the normal needs of consumption; this is the cause of crises. (350)

A crisis breaks out because consumption is no longer able to absorb the products on offer, so that production can no longer obtain the money it needs in order to sustain itself. (365)

31The third and most remarkable aspect of Petit’s explanation is that it is not limited to the rise and liquidation of the crisis, but it extends to the recovery. He thus offers one of the very first explanations of the entire cycle, and one that differs from the few other attempts made at the time in this directions.

  • 30 One of the current metaphors was “speculative fever”: see Besomi 2011a, 83–89.
  • 31 Again, ex-post prosperity was recognized to be only “apparent”, being related to the speculation fe (...)

32Most of the theories of crises available during the first three quarters of the nineteenth century only aimed at explaining crises. This is apparent in the explanations relying on external events or political diseases. But also the approaches focused on explaining the recurrence of crises were build on the idea that prosperity is the normal state of the economy, from which the system periodically deviates but where it returns to once the cause of the original discrepancy is eliminated. The overtrading explanation, in particular, is of this kind. The narrative begins from a prosperous state, with the economy advancing at a more or less regular pace. At some point —and this is part of the human nature— traders want to become richer than they are, and start asking for additional credit in order to increase their trading volume. Speculative projects are more attractive, banks are happy to lend money, prices increase fostering even more speculation,30 money is diverted into unlikely ventures, all goes well apparently31 until confidence is shaken and collapses at once. Shortly after the crisis, liquidation of “bad debts” begins, and when they are all absorbed, confidence is restored among the surviving businesses. From this point on, new prosperity can set in. Recovery, however, is not caused by the previous liquidation, it is only made possible by it. After the crisis has eliminated the abnormal excesses of prosperity, at some point prosperity restarts not because of the system’s dynamics, but following some external stimulus or just because it is natural that trade progresses. Accordingly, most theories of crises did not even have a name for recovery: the cycle was divided into an odd number of phases, normally three or sometimes five, centered on the crisis. A statement, probably the first, that it would be necessary to provide a full causal chain is due to Juglar in the second edition of his famous treatise (1889), recognizing that the previous version (1862) was lacking it (see Besomi 2010b).

  • 32 Overstone’s use of the term “cycle” should not mislead into thinking that he was considering someth (...)

33Meanwhile the observed features of crises, in particular their recurrence and the similarity in their unfolding, led some authors to suspect that they are subject to some law. The fact that crises are a periodical phenomenon was suspected very early in history. Some writers already made precise statements to that effect at the end of the eighteenth century, and the idea was more and more accepted as one crisis followed another during the first half of the nineteenth century, becoming commonplace afterwards. The term “periodic” was generally understood to mean “recurring”, admittedly at irregular times; only with Jevons it was sometimes meant in the strict sense of recurring at regular intervals (Besomi 2010a). From the late 1830s, it was also noted that crises arise, burst out and die away following the same pattern. This led to discuss the phases they pass through. Overstone’s well-known portrayal of the ‘cycle of trade’32 as revolving from quiescence through improvement, growing confidence, prosperity, excitement, overtrading, convulsion, pressure, stagnation and distress, to end eventually again in quiescence, was purely descriptive; so was a graphical depiction of a similar sequence by Longfield (1840) (see Besomi 2011e). There followed characterizations in three phases (Briaune 1840 and Juglar 1862) or five (Corbet 1841, 105; Cargill 1845, 27; White 1882, 525).

34Accordingly, the metaphors of crises as natural disasters, or as diseases (or as the consequence of feverish states), which were very frequently used up to the 1850s to indicate that something abnormal was (or has been) affecting the healthy working of the economic system (Besomi 2011a), started to be replaced by the metaphor of the wave or the ocean tides, suggesting something more regular and systematic, and subject to mechanical causation. More explicitly, Giffen submitted that “that there is ordinarily a cycle in all prices... may now be considered as an established doctrine in Political Economy” (1877, 103), and Cognetti de Martiis (1878) added that the fact that times of prosperity and depression, of plenty and of scarcity happen by turns, “has been established by several and protracted observations, and can therefore be accepted as a true law, which can be formulated as follows: to any time of general depression and stagnation there systematically follows a time of general progressive activity; this, after reaching at some juncture its acme and critical point, slows down and eventually peters out, gradually or suddenly, turning into a more or less durable stagnation” (438). The author, however, specifies that this applies to specific branches of the economy, and is not simultaneous throughout (ibid.). Yet crises (incongruously named “economic perturbations”) take place within these cycles (439). From Cognetti de Martiis’s account it is not really clear how and why this happens, but at any rate he submits that crises are subject to three laws: they amplify and spread throughout the system and across nations; they are periodic (again in the sense of “recurrent”, without implications as to the strict regularity that was being advocated by Jevons in the same year); and they purify the market by eliminating fictitious stocks, speculation, unsound businesses, etc. (446-9). He concluded that crises follow their laws, and accordingly they are suited for scientific inquiry (450). A few years later Juglar also defended the possibility of deriving a “law of crises” before the Société d’Économie Politique (Société d’Économie Politique, 1893).

  • 33 Although this explanation may rely to some extent on the mechanism developed by the writers quoted (...)

35These reflections took place after some earlier attempted explana- tion of the cyclical movement of the economy that enacted ante litteram Juglar’s dictum about the necessity to close the causal chain, with which the originality of Petit may be compared. The first example in this tradition is a cobweb-like explanation worked out by John Wade, who supplied two literary models in terms of prices and produced quantities in 1826 and 1833: rising prices beyond a certain threshold determine a fall in demand and a rise in supply, at first causing the increase in prices to slow down and eventually to reverse: “Thus, a rise of prices, like the ascent of a body on an inclined plane, is always accompanied with a counteracting power, which tends to draw them down to the point, or below it, from which they started” (1826, 32-33). Conversely, when prices descend, the opposite movement begins. In both cases, they are accompanied by fluctuations in confidence and in bank lending. When these smooth fluctuations are aggravated by speculation fed by easy credit, the system may undergo crises and panics (Wade 1833, 148; for further discussion Besomi 2008).33

36A second family of theories explains the turning points in terms of the fluctuations of loanable capital and of interest. When trade is depressed, savings cumulate in the banking system without finding remunerative investment outlets. Interest rates accordingly fall very low, and this eventually stimulates a recovery in investment. After some time all available savings are absorbed, as investment proceeds faster than saving; interest rates rise, but trade prosperity lures manufacturers and traders into further expanding their activity, each of them also providing a market to others. But as this happens, trade becomes increasingly fragile, and when any relevant accident interrupts its flow anywhere, the entire system breaks down (Bagehot, 1874). Another (possibly the first) version of this explanation considers, besides fluctuations in loanable capital, also price fluctuations and bullion movements. Prosperity brings higher prices, which upset the balance of trade, inducing the outflow of gold; this induces the banks to “turn the screw” on credit, which in turn creates pressure and shakes confidence. This causes a reversal. The opposite movement occurs during depression (Anonymous 1858, 286–287).

  • 34 This is in contrast with Overstone’s well known description referred to in the text (Loyd 1837, 31) (...)

37The most interesting example of this family of explanations is due to John Mills. His paper on credit cycles (1868) is arguably the first full business cycle theory, because it presents reflections on the epistemology of causation, describes an idealized cycle, also graphically represented (with Jevons’s aid), conceives the cycle as the normal state of the economy,34 supplied a description in four phases—as opposed to the odd number of phases centered around the crisis adopted by most contemporary writers—, and thoroughly discusses the relationship between cycles and equilibrium. The latter point is of interest to us. Credit is based on trust, and this trust has to rely on the equilibrium between credit, as demand for capital, and its supply. The equilibrating force is the price of capital, the increase of which curbs the tendency of credit to grow continuously while its decrease stimulates demand. Yet this force is outpowered by the psychological attitude of traders: during prosperity, their confidence gradually increases and degenerates into an excess, leading to overspeculation, while the panic suddenly destroys these beliefs and causes an interruption of credit.

  • 35 This factor explaining recovery probably relied on the branch of literature on plethoric capital em (...)

38The group of loanable funds theories is largely an elaboration of the overtrading-speculation theme, with interest fluctuations being called upon to remedy the omission in explaining recovery,35 while the credit-speculation-overtrading mechanism explains the cumulative advance, and the credit collapse explains panic and subsequent depression. This approach proved influential: Bagehot’s contribution, in particular as expounded in a more elaborate form in Lombard Street (1873; see O’Brien 2001) was often quoted (e.g., in the same decade by Duncan 1877, who reproduced in full Bagehot’s 1874 article in his appendix, Cognetti De Martiis 1878 and Giffen 1877—the latter, however, focused more on prices than on interest), Juglar incorporated the explanation of the trough in the second edition of his Crises périodiques (1889), it was at the hearth of Tugan-Baranowsky’s theory of the cycle (the first in the modern era), and we can find some of its elements in the monetary theories of the cycle in the first decades of the twentieth century.

39The factual premises of the loanable funds theory can be found even in Petit’s book. Loanable capital remains idle during and immediately after crises: on the lenders side, money “waits for the cessation of the storm before looking for opportunities. Like a ship in the harbour, it does not tackle the dangers of open seas while thunders rumble and the tempest roars” (346–347). Borrowers are also cautious, and “savings cannot be placed” (359). An abundance of idle capital causes a fall in the interest rate (359). The opposite happens during prosperity: the central bank is driven by the fluctuations in its deposits to offer capital at a low price when trade slows down and does not need it, while the Bank does not have much to offer and imposes costly conditions (370), eventually bringing about a financial crisis, when industry and merchants need money (386). However, Petit does not see this interest fluctuation as an alternation of incentive to production when it slacks off and as a restraint when it is exuberant: he only records the fact as a consequence of the cycle resulting from the fluctuations of production compared to normal consumption.

40The few explanations starting from an explicit rejection of Say’s law (rather than bypassing it by relying on monetary or credit factors) took a completely different approach and led to rather original explanations. Petit’s, as we have seen in Section 5, belongs to this class. Another imaginative theory is due to Auguste Ott. He too an explicit critic of Say’s law, Ott argued that production decisions depend on the entrepreneurs’ expectations, which are based on the state of trade at the time they made them. But the consequences of these decisions alter the state of trade; for instance, if goods are scarce on the market a number of entrepreneurs will be induced to increase production, with the result that goods will no longer be scarce and may actually turn out to be overabundant. Thus, acting on the basis of some expectations disappoints the very expectations that started the movement, and triggers its reversal (Ott 1854, col. 1389). At first Ott used this mechanism only to explain crises, but he later completed the causal circle by also considering the specularly reverse movement of recovery (1892, vol. 1, 139; for a discussion see Besomi and Colacchio 2010).

7. Concluding remarks

41Petit’s theory of crises is admittedly a strange creature. It starts on a mercantilist premise that leads him to emphasize the role of money as a store of value, and the concepts he uses are defined in a blurred and sometimes inconsistent way. His rejection of Say’s law, however, opened new horizons that were precluded to most of his contemporaries in France and Britain.

  • 36 One of the very rare exceptions, from the rank of free-bankers rather than that of overtrading theo (...)

42The panoramic view of trade cycle theories at the time of Petit’s contribution offered in Section 6—admittedly sketchy but, I would say, complete enough—enables us to assess his theory of crises and cycles within the developments taking place in the field. On the one hand, in Petit’s book the cycle is outlined almost incidentally, the main focus being on crises and their recurrence. Yet, while the overtrading explanation of crises started from the recognition that crises are a recurrent phenomenon but refrained from explaining the causes of the recurrence focusing instead on its conditions,36 Petit is concerned with the construction of a complete causal chain. Petit is surely among the first writers to do so, although John Mills’s contribution (published in the same year as Petit’s book) is surely more focused, less subject to ambiguous categories and more explicit. On the other hand, not being fettered by deference to Say’s law, Petit (as well as Ott) managed to work out a rather innovative explanation, based on real rather than monetary factors and anticipating approaches that were (independently) rediscovered (for sure in an analytically more rigorous setting) only several decades later, when respect for Say’s law was no longer one of the economists’ duties (Samuels 1972).

43One can thus hardly disagree with Jean-Pascal Simonin’s conclusion that Petit’s book did not deserve the oblivion it was condemned to by the dogmatism of the économistes and their firm control of publishing outlets. One is actually left wondering, in spite of the irreconcilable premises, whether an interaction between views such as Petit’s and Ott’s (his writings only circulated among Catholic associationists) with the empirical work of Juglar and the epistemic reflections on causation in connection with the recurrence of crises by Briaune, Coquelin, Juglar and others could have given rise to a French school of business cycle theorists paralleling the developments occurring in Germany.

I am grateful to Cécile Dangel, Jean-Pascal Simonin and two anonymous referees for helpful comments on the first draft; needless to say, I am responsible for any remaining errors and gaps in the argument.

Haut de page


Anonymous. 1818. Practice Opposed to Theory; or, An inquiry into the Nature of our Commercial Distress, with a View to the Development of its True Causes, and the Suggestion of a Suitable Remedy. By a practical man. London: Richardson.

Anonymous. 1858. Draft Report of the Select Committee on the Bank Acts, 1857-1858, as Moved by Mr. Cayley. In The Money Bag. Literature, Politics, Finance. London: Oakley, 265-294.

Bagehot, Walter [attributed to]. [1860] 1861. Memoir of the Right Hon. James Wilson. The Economist, Supplement, November 17. Cited as reprinted in pamphlet form. London: Effingham Wilson.

Bagehot, Walter [unsigned article]. 1873. Lombard Street. London: King.

Bagehot, Walter [unsigned article]. 1874. The Probable Money Market of 1874. The Economist, 3 January.

Baumol, William Jack. 1977. Say’s (at Least) Eight Laws, or What Say and James Mill May Really Have Meant. Economica, 44(174): 145-61.

Besomi, Daniele. 1995. From The Trade Cycle to the “Essay in Dynamic Theory”. The Harrod-Keynes Correspondence, 1937-1938. History of Political Economy, 27(2): 309-343.

Besomi, Daniele. 1999. The Making of Harrod’s Dynamics. London: Macmillan.

Besomi, Daniele. 2008. John Wade’s Early Endogenous Dynamic Model: “Commercial Cycle” and Theories of Crises. European Journal of the History of Economic Thought, 15(4): 611-639.

Besomi, Daniele. 2010a. The Periodicity of Crises. A Survey of the Literature before 1850. Journal of the History of Economic Thought, 32(1): 85-132.

Besomi, Daniele. 2010b. “Periodic Crises”: Clément Juglar between Theories of Crises and Theories of Business Cycles. Research in the History of Economic Thought and Methodology, 28(A): 169-283.

Besomi, Daniele. 2011a. Disease of the Body Politick. A Metaphor for Crises in the History of Nineteenth Century Economics. Journal of the History of Economic Thought, 33(1): 67-118.

Besomi, Daniele. 2011b. Naming Crises. A Note on Semantics and Chronology. In Crises and Cycles in Economic Dictionaries and Encyclopedias, edited by Daniele Besomi. London: Routledge, 54-132.

Besomi, Daniele. 2011c. The Analysis of Crises in Early French Dictionaries and Encyclopedias. In Crises and Cycles in Economic Dictionaries and Encyclopedias, edited by Daniele Besomi. London: Routledge, 177-196.

Besomi, Daniele. 2011d. Charles Coquelin: Banking Monopoly and Commercial Crises. In Crises and Cycles in Economic Dictionaries and Encyclopedias, edited by Daniele Besomi. London: Routledge, 209-224.

Besomi, Daniele. 2011e. Graphical Representations of Overstone’s Cycle of Trade. In C. Gehrke, N. Salvadori, I. Steedman and R. Sturn (eds), Classical Political Economy and Modern Theory. Essays in honour of Heinz D. Kurz. London: Routledge, 289-312.

Besomi, Daniele and Colacchio, Giorgio. 2010. Auguste Ott on Commercial Crises and Distributive Justice: An Early Input-Output Scheme. Review of Political Economy, 22(1): 75-96.

Briaune, Jean-Edmond. 1840. Des crises commerciales. De leurs causes et de leurs remèdes. Paris : Bouchard-Huzard.

Cargill, William. 1845. The Currency, Showing how a Fixed Gold Standard Places England in Permanent Disadvantage and Produces Periodical Domestic Convulsion. London: Ollivier.

Cayley, Edward Stillingfleet. 1826. Corn, Trade, Wages, and Rent: or Observations on the Leading Circumstances of the Present Financial Crisis. London: Rigway.

Cognetti de Martiis, Salvatore. 1878. Forme e leggi delle perturbazioni economiche, Giornale degli economisti, VI(6): 431-452.

Corbet, Thomas. 1841. An Inquiry into the Causes and Modes of the Wealth of Individuals: or, The Principles of Trade and Speculation Explained. London: Smith, Elder & Co.

De Vecchi, Nicolò, 1983. Crisi. In Giorgio Lunghini (ed.), Dizionario di economia politica. Torino: Bollati Boringhieri.

Duncan, William James. 1877. Notes on the Rate of Discount in London from January 1856 to August 1866: with Sketch since the Crises of 1847. With Continuation to the Close of 1876, 2nd ed. Edinburgh: Lorimer & Gillies.

Giffen, Robert. 1877. Stock Exchange Securities: An Essay on the General Causes of Fluctuations in their Price. London: George Bell and Sons.

Harrod, Roy Forbes. 1936. The Trade Cycle: An Essay. Oxford: Clarendon Press.

Harod, Roy Forbes. 1939. An Essay in Dynamic Theory. Economic Journal, 49(3): 14-33.

Huskisson, William. 1810. The Question Concerning the Depreciation of our Currency Stated and Examined. London: John Murray.

King, John E. 1981. Perish Commerce! Free Trade and Underconsumption in Early British Radical Economics. Australian Economic Papers, 20(1981): 235-257.

Juglar, Clément. 1862. Des crises commerciales et de leur retour périodique en France, en Angleterre et aux États-Unis. Paris: Guillaumin.

Juglar, Clément. [1889] 1967. Des Crises Commerciales et de leur retour périodique en France, en Angleterre et aux États-Unis, 2nd ed. Paris : Alcan. Reprinted, New York: Kelley.

Kates, Steve. 2003. Introduction. In Steve Kates, Two Hundred Years of Say’s Law. Cheltenham: Edward Elgar, 1-33.

Keynes, John Maynard. [1936] 1973. The General Theory of Employment, Interest and Money. In The collected writings of J. M. Keynes, Vol. 7. London: Macmillan.

Laveleye, Émile de. 1865. Les crises commerciales et monétaires. Revue des Deux Mondes, 207-233, 432-460.

Link, Robert Grant. 1959. English Theories of Economic Fluctuations 1815-1848. New York: Columbia University Press.

Longfield, Samuel Mountifort. [1840] 1971. Banking and Currency. Dublin University Magazine 15: 1–15, 218–233; 16: 371–389, 611–620. Reprinted in The Economic Writings of Mountifort Longfield, edited by R. D. Collison Black. New York: Kelley.

Loyd, Samuel Jones (Lord Overstone). [1837] 1972. Reflections Suggested by a Perusal of Mr. J. Horsley Palmer’s Pamphlet on the Causes and Consequences of the Pressure on the Money Market. Reprinted in Overstone, Tracts and Other Publications on Metallic Paper and Currency. Privately printed, London 1857. Reprinted, Clifton: Kelley.

Malthus, Thomas Robert. 1815. Observations on the Effects of the Corn Laws and of a Rise or Fall in the Price of Corn on the Agriculture and General Wealth of the Country, 3rd ed. London Murray.

Marx, Karl. [1859] 1971. Zur Kritik der politischen Ökonomie. English translation, A Contribution to the Critique of Political Economy. London: Lawrence & Wishart.

Marx, Karl. [1867] 1974. Das Kapital. Kritik der politischen Oekonomie, vol. 1. English translation, Capital. A Critique of Political Economy. Moscow: Progress Publishers. Reprinted, London: Lawrence and Wishart.

McCulloch, John Ramsey. 1822. Agricultural Distress: Causes- Remedies. Edinburgh Review, XXXVI(72): 452-482.

Mill, John Stuart. [1848] 1965. Principles of Political Economy, with some of their Applications to Social Philosophy, 2 vols. London: Parker. Critical edition as volumes 2 and 3 of The Collected Works of John Stuart Mill, ed. V. W. Bladon and J. M. Robson. Toronto: University of Toronto Press.

Mills, John. 1868. On Credit Cycles and the Origin of Commercial Panics. Transactions of the Manchester Statistical Society, Session 1867/68: 5-40.

O’Brien, Denis Patrick. 2001. Bagehot’s Lombard Street and Macroeconomic Stabilisation. Scottish Journal of Political Economy, 48(4): 425-441.

Ott, Auguste. “Crise”. In Dictionnaire des sciences politiques et sociales comprenant la politique, la diplomatie, le droit naturel, le droit des gens, les rapports de l’Église et de l’État, l’administration, les finances, la police, la force armée, l’économie politique et la statistique : avec le texte ou le résumé des traités les plus importants, des constitutions et lois fondamentales des peuples anciens et modernes, et l’analyse des principaux ouvrages sur la politique et les autres sciences sociales, edited by A. Ott. Paris: Migne.

Ott, Auguste. 1892. Traité d’économie sociale ou l’économie politique coordonnée au point de vue du progrès. Paris : Fischbacher.

Petit, Louis-Amable. 1857. De la fixité de la valeur de l’or et de l’avantage d’un double étalon monétaire. Réfutation de la doctrine contraire de M. Michel Chevalier, par X. Rouen : Lecointe frères.

Petit, Louis-Amable. 1868. Richesse par excellence ; balance commerciale ; crises commerciales ; libre-échange, réfutation de la doctrine des économistes et des arguments présentés en faveur du libre-échange par MM. Rouher et Forcade La Roquette, lors de la discussion sur les traités de commerce. Écretteville-lès-Baons : chez l’auteur.

Ricardo, David. [1817] 1951. Principles of Political Economy and Taxation. In The Works and Correspondence of David Ricardo, edited by Piero Sraffa, vol. 1. Cambridge: Cambridge University Press.

Samuels, Warren Joseph. 1972. The Teaching of Business Cycles in 1905-1906. History of Political Economy, 4(1): 140-162.

Say, Jean-Baptiste. 1814. Traité d’économie politique, ou, Simple exposition de la manière dont se forment, se distribuent et se consomment les richesses. Seconde édition entièrement refondue et augmentée d’un épitomé des principes fondamentaux de l’économie politique. Paris : Chez A.-A. Renouard.

Simonin, Jean-Pascal. 2011. Un économiste hétérodoxe, théoricien des crises : Louis-Amable Petit, 1819-1871. Œconomia, History- Methodology-Philosophy, 1(2): 259-298.

Société d’Économie Politique. 1893. Y a-t-il loi ou accident dans la périodicité des crises ? Séance du 5 Janvier 1893. Journal des Économistes, 51(5e Séries, Janvier-Mars) : 1-13.

Torrens, Robert. [1815] 2000. An essay on the External Corn Trade: Containing an Inquiry into the General Principles of that Important Branch of Traffic; An Examination of the Exceptions to which these Principles are Liable; And a Comparative Statement of the Effects which Restrictions on Importation and Free Intercourse, are Calculated to Produce upon Subsistence, Agriculture, Commerce, and Revenue. London: Printed for J. Hatchard. Reprinted with the additions made in the 2nd edition in Giancarlo De Vivo (ed.), The Collected Works of Robert Torrens. Vol. 2. Bristol: Thoemmes Press.

Tugan-Baranovsky, Michail Ivanovic. 1905. Theoretische Grundlagen des Marxismus. Leipzig: Duncker und Humblot.

Tugan-Baranovsky, Michail Ivanovic. 1913. Les crises industrielles en Angleterre. Paris : Giard & Brière.

White, Horace. 1882. Commercial Crises. In J. J. Lalor (ed.), Cyclopædia of Political Science, Political Economy, and the Political History of the United States by the Best American and European Writers. Chicago: Rand, McNally & Company, vol. 1, 523-530.

Wilson, James. 1839. Influences of the Corn Laws as Affecting all Classes of the Community and Particularly the Landed Interests. London: Longman, Orme, Brown, Green, and Longmans.

Wilson, James. 1840. Fluctuations of Currency, Commerce, and Manufactures; Referable to the Corn Laws. London: Longman, Orme, Brown, Green, and Longmans.

Haut de page


1 Simonin has already supplied a biography of Petit and a bibliography of his writings, and I can safely refer the reader to his paper. For the sake of completeness, it is however worth adding to Simonin’s list of reference a further text by Petit, published anonymously in 1857 as De la fixité de la valeur de l’or et de l’avantage d’un double étalon monétaire. Réfutation de la doctrine contraire de M. Michel Chevalier.

2 Translations from French are mine. As Petit’s language is sometimes loose and not always consistent, I have chosen to pay more attention to the general meaning than to the precise wording.

3 As I am interested in this very fundamental level of interpretation, I refrain from comparing in detail the specific mechanisms devised by Petit and his contemporaries.

4 Petit argues that exchanges would be formally equivalent to barter if the sole aim of producers were to obtain the means for their survival, in a primitive stage of society (264), or a goldless economy (299), or “up to the satisfaction of the daily needs” (107) (e.g. when, in the early stages of recovery, most people are devoting their entire income to consumption); in such conditions, Say’s law would apply. Here Petit’s argument echoes those of Stuart Mill, Roscher and Marx, who noted that the intermediation of money in the process of exchange of goods for other goods essentially alters the nature of the circulation of goods, as it breaks the act of exchange into two distinct acts—sale (Goods-Money) and purchase (Money-Goods)—thereby introducing the possibility of crisis (even if this possibility in a simple mercantile system does not become an actuality: Marx 1859, Ch. 2, and Marx 1867, Ch. 3).

5 Petit is adamant that the problem of overproduction crises is typical of capitalistic production: “if the property of machines were communal, consumption would expand without limits, and we would not hear of excesses” (308). The distribution of wealth also plays a part in preventing consumption to follow the same path as production (336).

6 The Sismondian theme of the increased production due to the application of more productive machinery is easily recognized. In Petit’s view, however, the problem is not that workers cannot buy back their product, but that producers do not want to purchase it for, at some point, they prefer to save.

7 Precious metals “are, so to speak, an active form of wealth: they are the form of wealth leading to the acquisition of true wealth;. . . they are the means used by producers to obtain the different values that make people really rich” (321).

8 This applies to individuals as well as to societies: “Precious metals are thus the real foundation of thrift; one can affirm that without them there is no formation of savings, realization of any wealth, nor establishment of any fortune providing a constant source of revenue” (Petit 1868, 313).

9 Again, Petit’s conceptual system is quite ambiguous. Économie, the term he used the most, is the layperson’s expression for thrift. He also used thésaurisation (hoarding) and épargne (savings), but not is a fully consistent way. While he seemed aware that there is a difference between saving and hoarding (“in antiquity and in the Middle Ages. . . hoarding. . . was the mode of choice to build savings”, 40), he also identified savings with hoarding (“money is the only constituent of savings, which can only therefore be built upon hoardings”, 34). By thésaurisation he seems to imply literally the piling up of gold in a safe.

10 “...periodically, at times more or less close, depending on the incidents affecting production or consumption” (Petit 1868, 307); “periodically, so to speak” (347), “with an almost mathematical regularity” (137).

11 d’une manière continue (267).

12 “When nations work more, they have more products to consume, and they

actually consume more. Ever-increasing affluence and well-being results from this consumption increase.” (106)

13 Note that if savings were not hoarded but invested, they would create further income, part of which would be consumed. Whether this would be sufficient to sustain growth depends not only on the proportion of income saved, but also on the capital-output ratio of the new investment. This issue is completely beyond Petit’s discussion, which is in terms of consumption rather than in terms of demand.

14 “There is a certain class of consumers who continue to consume like before the crisis; other classes, although moderating their consumption somewhat and in various degrees, nevertheless still consume in a reasonable proportion. This consumption flow eventually absorbs—after a few months, sometimes after a year, but never much more than that—the excess that became apparent when the crisis broke out. This process is quickened because production has slowed down.” (289).

15 Petit uses the word reprise (also elsewhere in the text, e.g. 4, 289-90, 345, 367, 481-2, 538-9). The very presence of a noun (from the verb reprendre, meaning to recover, pick up, return (to normal), resume) to indicate this phase of business activity is rather extraordinary for the time: see Section 6.

16 This is not necessarily to be interpreted as an investment function, a concept extraneous to Petit’s thought. Such a technical relationship does not represent the amount of capital entrepreneurs buy or want to buy, but only what is necessary in order to maintain equilibrium, given the dynamics of consumption. In this, the relationship resembles better Harrod’s ex-ante acceleration coefficient C of (1939) than the Relation of (1936): for a discussion see Besomi (1995, 318–24).

17 The same transmission and amplification mechanism of troubles is also described by Bagehot, a few years after Petit, in an article (Bagehot 1874) the importance of which will be described below. Among the antecedents there is a pamphlet by Cayley, where both the interconnectedness of trade (1826, 9) and income spending (14-15) are evoked to explain the spreading of good or bad trade.

18 I would, however, differ in emphasis from Simonin. He reports a passage from Petit’s book where he qualifies Say’s law as “admirable” but criticizes it as “incomplete” (Simonin 271 ; Petit, 106–107). Petit’s criticism is not as mild as it may seem from this passage. He argues at length that Say’s law does not apply where it matters most, in a monetary capitalist economy where there is accumulation of wealth and capital. Either the law guarantees that a general overproduction is impossible, or fails to do so. Petit clearly maintains that Say’s law fails to. (see Petit, 296-311 and passim).

19 Petit summarizes the position of the économistes as never wanting to admit the existence of gluts and declaring that they are impossible (288; see also 312 and 356).

20 The locution is Say’s, in the passage where he stated that crises can only be partial and temporary unless some “natural or political calamity, or the avidity or incompetence of the Government, force a scarcity [in some branches of production] which causes a glut [in some other branches]. As soon as this cause of political disease is discontinued, the means of production find their way towards the empty channels, and their product absorbs the excesses produced by the other [trades]. Equilibrium is re-established; in truth, it would rarely fail if the means of production were always granted their full freedom” (Say 1814, vol. 1, 149–50).

21 Petit was aware that the acceptance of Say’s law forced economists to explain crises in terms of monetary or credit factors: “The économistes maintain that the cause of crises cannot be attributed to general overproduction. Yet some cause must be assigned to them. According to the économistes, this cause is related to money. It is money, and only money, which, according to them, gives rise to all the commercial crises we witness at short intervals and with a distressing regularity.” (322) He thus discussed at length the theories blaming the banking system (292–294), troubles in monetary circulation (322–27), or credit (351–61), and offered his own explanation of the banking, monetary and credit phenomena as accompanying, rather than causing, the crises, whose “sole cause” (328, 349, 354) is overproduction.

22 Also in Italy where, however, the literature on the subject was scarce.

23 The most elegant theory of crises belonging to this family, however, is due to a Frenchman, Charles Coquelin (see Besomi 2011d).

24 The most interesting examples in this tradition took the form of cobweb-like mechanisms. Around 1815, a number of writers (including Malthus, 1815, McCulloch, 1822, and Torrens, 1815) argued that the Corn Laws amplified the normal fluctuations of the prices of cereals by hindering the adjustment between supply and demand. John Wilson’s agricultural cycle (1839; see Bagehot 1860, 15-16, and Link 1959, 104–107), and Corbet’s explanation extending to the entire economy the oscillations in trade arising from the movements of capital seeking the highest remuneration (1851, 33–35 and 104) also belong the same family.

25 For references to Petit’s discussion of this approach see footnote 21.

26 To honour the truth, one should point out that most of the literature on crises during the first half of the century was rather obscure, being highly topical and scattered in pamphlets and in the periodical press.

27 While it seems true that Ricardo’s view conquered England as the Holy Inquisition conquered Spain, Keynes’s conclusion (accepted not only by his friends, but also by his foes: see Kates 2003, 7-8 and passim) that the puzzle concerning effective demand ceased to be discussed and vanished from the literature (Keynes 1936, 32) largely (but with relevant exceptions) applies to Britain, France and Italy for the first three quarters of the nineteenth century, but not to the German speaking area.

28 For a discussion of the definition of crises in nineteenth century dictionaries, see Besomi 2011b, 80-84; more specifically in French dictionaries in the first part of the century: Besomi 2011c, 178-179 and 184-6; on the medical metaphor see Besomi 2011a.

29 Explicit definitions of “crises” were rather rare at the time; most of them were confined to dictionaries. See Besomi 2011c, 184–186.

30 One of the current metaphors was “speculative fever”: see Besomi 2011a, 83–89.

31 Again, ex-post prosperity was recognized to be only “apparent”, being related to the speculation fever: “the flush of fever [is mistaken] for the glow of health” (Anonymous 1818, 23–24; for a discussion see Besomi 2011a, 86–88).

32 Overstone’s use of the term “cycle” should not mislead into thinking that he was considering something like a theory of the cycle in 1837. He was clearly using the term as indicating “A round, course, or period through which anything runs in order to its completion; a single complete period or series of successive events”, as opposed to the meaning implicit in the twentieth century’s business cycle theories of “A recurrent round or course (of successive events, phenomena, etc.); a regular order or succession in which things recur; a round or series which returns upon itself” (Oxford English Dictionary, meanings 4 and 3a). Overstone’s emphasis was on the succession of events within a cycle (the description of which has a precise beginning and an end), while later the focus was set on the repetition of such cycles, that is, on the causal chaining of each phase with the succeeding one.

33 Although this explanation may rely to some extent on the mechanism developed by the writers quoted in footnote 25, the interpretation is completely different. The Corn Laws amplify existing fluctuations in the production of cereals, while Wade’s mechanism describes free oscillations: once in motion, his mechanism accounts for self-sustained fluctuations.

34 This is in contrast with Overstone’s well known description referred to in the text (Loyd 1837, 31). Overstone’s cycle has a well defined starting and ending point, and leaves unexplained how the next cycle resumes from the first. This is because progress was assumed to be the normal state of the economy, from which the cycle represents a deviation (for a discussion Besomi 2011e). The same applies to Juglar, who explicitly qualified prosperity as an equilibrium and a normal state of the economy (Besomi 2010b).

35 This factor explaining recovery probably relied on the branch of literature on plethoric capital emphasizing that the abundance of monetary capital unsatisfied by low interest rates induces its owners to invest it in speculative enterprises rather than lending it via the banking system: see e.g. Wilson 1840, 28–29 and 45–47. Wilson was certainly a source for Bagehot. This passage is rather explicit: “It has often been a subject of remark by practical men, without referring to the real causes, that there appeared periodically to exist a plethora in the currency or available capital of the country, which brought about these periods of excitement and wild speculation, opening up channels through which the redundancy flowed, not only to the cure of the disorder, but much to the danger of the patient. Such were the years 1825 and 1836” (Wilson 1840, 45).

36 One of the very rare exceptions, from the rank of free-bankers rather than that of overtrading theorists, is Charles Coquelin: see Besomi 2011d.

Haut de page

Pour citer cet article

Référence papier

Daniele Besomi, « Louis-Amable Petit and the 19th Century Literature on Crises and Cycles  », Œconomia, 2-2 | 2012, 208-237.

Référence électronique

Daniele Besomi, « Louis-Amable Petit and the 19th Century Literature on Crises and Cycles  », Œconomia [En ligne], 2-2 | 2012, mis en ligne le 01 juin 2015, consulté le 28 mai 2017. URL : ; DOI : 10.4000/oeconomia.1583

Haut de page


Daniele Besomi

Centre Walras-Pareto, Université de Lausanne, Email:

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus d’Œconomia sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Association Œconomia
  • Logo CNRS
  • Les cahiers de