1. A physician reads the bank’s balance sheet
This note is a digression from transfer pricing. I spent a morning with Clément Juglar’s Des crises commerciales et de leur retour périodique en France, en Angleterre et aux États-Unis, in the ENS Éditions reprint (Lyon, 2014) of the second edition of 1889, edited and introduced by Ludovic Frobert. Like most economists, I came to Juglar through Schumpeter, who named the seven-to-ten-year cycle after him and placed him with Kitchin and Kondratieff in the triad of cycles that rhythm the process of creative destruction.
Juglar (1819–1905) was a physician. His Paris thesis of 1846 concerned the influence of heart disease on the lungs, and Frobert notes that the Paris anatomo-clinical school supplied his method: empiricism, the numerical method, specialization, and therapeutic skepticism. He abandoned medicine after 1848. His first economic series were demographic (marriages, births, deaths), and in searching for a cause of their fluctuations beyond famine, war and epidemic he found the discounts of the Banque de France since 1800, which he read as a clinician reads a pulse. A first paper appeared in 1856; a second, in the Journal des Économistes of 1857, predicted the crisis of that year six months before it arrived. The Académie des sciences morales et politiques put the causes of commercial crises to competition in 1860, and Juglar’s prize memoir became the first edition of 1862.
2. The volume
The 1889 edition is the text of record; a projected third edition was never completed. Frobert’s edition reproduces it with minimal correction and appends two short articles, “Crises commerciales” from Block’s Dictionnaire général de la politique (1863) and from Say and Chailley’s Nouveau dictionnaire de l’économie politique (1891). The 1891 article is the most compact exposition of the mature doctrine and the best place to start.
The book has three parts of unequal length. Part I, in nineteen numbered chapters, gives the definition, causes, mechanism and periodicity of crises. Part II is a crisis-by-crisis history read off the balance sheets of the Bank of England (from 1696), the Banque de France (from 1800), the banks of the United States, and Hamburg, Prussia and Austria. Part III traces the influence of crises on the economic state of societies through consumption, trade, clearing-house volumes, securities issues, railway receipts, public revenue, savings deposits, and vital statistics. A chapter map is given in the appendix.
3. Three periods
Juglar’s definition of a crisis is one phrase: the arrest of the rise in prices, the moment when no new buyers appear (1891 article, §12; the same phrase closes the mechanism section at §57, where it is called the principal, one could say the unique, cause of crises). Everything else in the book is the morphology of what precedes and follows that moment.
| Period | Duration | Markers |
|---|---|---|
| Prosperity | 7 to 10 years | Capital abundant and cheap (discount at 2%, 1%, ½%); easy credit; rising prices; metallic reserves declining from their maximum |
| Crisis (acute state) | 10 to 15 days | Drain of specie under adverse exchanges; discount rate raised to 7–10%; suspension of exchanges; failures of over-engaged houses |
| Liquidation | 2 to 4 years | Falling prices in all products and property; empty bank portfolios; idle working capital pushing down the interest rate; reserves rebuilding |
The sequence is invariant in order though variable in duration, and it is synchronous across the three great commercial nations, which Juglar takes as proof of the solidarity of markets. He refuses a fixed period. Chapter XV (§4) says Jevons went too far in attaching the cycle to a decade or to sunspots, and that Cobden’s rule of a crisis every ten years is of the same kind; §10 warns against confining the oscillations of business in formulas determined in advance, since fixing the return at 5 or 10 years would extract more from observation than observation gives.
The competing theories are dismissed on a single ground (1891 article, §§18–23): Peel, Tooke, Coquelin, Mill, Laveleye, Bonamy Price, Leroy-Beaulieu, Max Wirth and Yves Guyot each explain the liquidation and ignore the rise that preceded it. Wars, revolutions, harvests, tariffs and bank statutes are occasional causes, “the last drop of water that makes the basin overflow” (Introduction to the first edition, §9). The Crimean War did not interrupt the prosperity of 1854–1855; the revolution of 1848 followed the crisis of 1847 and could not have caused it; the wheat shortage of 1879 was paid for without incident and the crisis came in January 1882.
4. Credit as engine, specie as limit
The engine is credit, defined as the simple power to buy in exchange for a promise to pay (1891 article, §51). Everyone with credit can buy; purchasing power against promises raises demand and prices; rising prices enlarge collateral and therefore credit; the expectation of profit on purchases made on credit draws in new layers of buyers. Juglar rejects the quantity theory in both its forms: it is not the quantity of money but the rapidity of its circulation that matters (§43), and bills, notes and cheques do the work of money in every transaction above retail size (§44).
The limit is metallic. During the rise, domestic settlements can be prolonged by renewals and rediscounts, so bank portfolios swell. Foreign settlements cannot: when manufactured goods are no longer accepted abroad at the prices at which they are held, the only merchandise accepted on every market is gold and silver, drawn from the banks’ reserves. The fall in reserves under adverse exchanges is the signal; the rise in the discount rate is the defence, not the cause (§§13–14, 30–32). Juglar had translated the British parliamentary inquiries on banking for the Banque de France (eight volumes, 1865) and takes the Bullion Report of 1810 as the authority on the exchanges.
Banks therefore suffer crises and do not produce them. The point is made against Peel’s Act of 1844, which Juglar describes as a brake that functions only when its action is null, suspended at every crisis it was designed to prevent (Introduction to the first edition, §8). The 1891 article closes with the image that governs the whole method: the balance sheets of the great banks are “a simple manometer indicating only the pressure” of a movement that dominates the entire social mechanism (§88).
Chapter XVIII reduces the observation to four balance-sheet items: note circulation, discounts of commercial paper, metallic reserve, and current accounts. Their annual oscillations since 1851 are read crisis by crisis. For the Banque de France the portfolio of commercial paper fell from 1,282 million francs to 335 million by 1876, while the reserve rose to 2,281 million the following year: liquidation complete, prosperity resumed, crisis in January 1882.
5. No equation
The volume contains no equation. Not one formula, functional relation or symbolic expression appears in the 1889 text or in the two appended articles. I verified this against the full text rather than by reading: the PDF’s embedded fonts carry no Unicode map, so I rebuilt the glyph-to-character mapping and searched the 1.4 million characters for équation, formule, algébrique, mathématique, coefficient, fonction, proportionnel, and for the equals sign. The hits are prose: Bastiat’s “formule” of what is seen and not seen, discount “proportionné” to commerce, and the like.
The apparatus is tabular and graphical. Annual balance-sheet series for the three central banks; price indices on a base of the 1845–1850 mean set at 100; the prices of the English and French 3 per cent rentes; wages; discount rates; exchange rates; and line charts, the “tracés graphiques” to which the text refers the reader. The only arithmetic is a base-100 index and the occasional ratio of loss to reserve.
The absence is doctrinal. Three passages fix it:
- Introduction to the second edition, §2: “Without bringing in any theory, any hypothesis, the observation of facts alone has sufficed to disengage the law of crises and their periodicity.”
- Chapter XV, §10: the oscillations of business must not be chained in formulas determined in advance; they must keep all their elasticity.
- Frobert’s introduction (§4, citing a later text): polemicizing against the use of means and coefficients, Juglar insisted that the figures speak for themselves and must not be touched, since they present themselves, with few exceptions, in ascending or descending annual series from maximum to minimum.
The “law” is therefore a verbal statement of sequence: prosperity, crisis, liquidation, always in that order, with amplitude growing with wealth. Juglar states the direction of each phase and its markers; he never writes a relation between two variables, a lag, a rate of change, or a period. His one quantitative generalization, that the amplitude of oscillation grows with the wealth of nations (1891 article, §§34, 70–72), is asserted from the comparison of the first and second halves of the century, not measured.
The counterfactual is instructive. Juglar possessed annual series for the four balance-sheet items over ninety years in three countries, which is more than Kitchin had in 1923 or Frisch in 1933. A second-order difference equation in the discount portfolio would have given him a period. He declined the instrument on principle, and the first formal treatment of his cycle waited for Tinbergen and Frisch half a century later.
6. Schumpeter’s Juglar and the historians’ Juglar
Schumpeter made Juglar’s reputation. In the History of Economic Analysis he credits Juglar with three innovations: the systematic use of time-series data, the discovery and complete morphological description by phases of the cycle as an endogenous phenomenon, and the attempt to tie the observed cycle to an original theoretical canvas. He called him the first to have a clear conception of how theory, statistics and history should cooperate in economics, and in Business Cycles (1939) attached his name to the seven-to-ten-year cycle.
Frobert’s introduction records the retreat from that judgment. Later analysts contrasted Schumpeter’s enthusiasm with the weak reception of Juglar’s work in his own time, observed that much of his theoretical and empirical material was then common knowledge, and set his indisputable originality beside a number of conceptual lacunae. Three studies carry the revision:
- Besomi, “‘Periodic crises’: Clément Juglar between theories of crises and theories of business cycles” (2010), places Juglar at the transition from a literature on crises as events to a literature on cycles as process, and shows how much of the transition he left incomplete.
- Bridel, “Juglar, Walras et Pareto: measurement without theory vs theory without measurement” (2009), gives the verdict in its title. Juglar measured and refused to theorize; Walras theorized and refused to measure.
- Dal Pont Legrand and Frobert, “Le ‘prophète des crises’: économie politique et religion chez Clément Juglar” (2010), traces the providential reading of the cycle in Juglar’s Catholicism: the complete movement, rise and fall, is a natural regulator of the industrial system, and its normality signals an order in the world.
The balanced assessment Frobert quotes is worth retaining. Juglar’s principal title to glory, which is not small, is to have affirmed and attempted to demonstrate empirically that prosperity, crisis and depression are necessary phases of a dynamic and endogenous mechanism of economic activity, concomitant with growth: crises are no longer historical accidents but a recurrent and natural component of the mechanisms governing economic development, with their origin in the credit system and the interaction of credit, interest rates and nominal prices.
On the evidence of the 1889 volume, that is exactly right. Juglar supplies morphology and measurement; the dynamics remain in prose.
Appendix A. Chapter map of the 1889 edition
Part I. Commercial crises: definition, origin, causes
| Chapter | Title |
|---|---|
| Introduction | Des crises commerciales |
| I | Definition. Origin of crises |
| II | Causes of crises |
| III | Influence of credit on prices |
| IV | Influence of banks on crises |
| V | Variation of prices. Periods of rising and falling prices |
| VI | Rise and fall of prices |
| VII | The fall of prices and the current crisis |
| VIII | The rise of prices and the end of the crisis |
| IX | Price variations of the English 3 per cent |
| X | Price variations of the French 3 per cent |
| XI | Variations of wages |
| XII | Interest rate, discount rate, price of gold |
| XIII | Role of the foreign exchanges |
| XIV | Solidarity of the great markets |
| XV | Periodicity of crises (Jevons, Cobden) |
| XVI | Preventive means. Remedies |
| XVII | History of the principal balance-sheet items of the Banks of France, England and the United States |
| XVIII | Mechanism of crises. Movements of the balance sheets |
| XIX | Comparative study of the principal commercial crises in France, England and America |
Part II. History of crises. Bank of England (crises of 1696 to 1882); Banque de France (1804 to 1882); Bank of the United States and the national banks (1814 to 1884); Hamburg 1857, Prussia 1857, Austria 1873.
Part III. Influence of crises on the economic state. Opinions of Engel, Keleti, Neumann-Spallart and Kőrösi on measuring the economic state of populations; consumption; discount rates; exchange rates; prices; clearing houses; securities issues; imports and exports by product (cereals, cotton, wool, silk, precious metals); French and English railway receipts; coal, iron, gas, tobacco, beverages; public revenue; savings deposits; marriages, births, deaths.
Appended. “Crises commerciales” (1863); “Crises commerciales” (1891): sketch of a crisis in three periods; definition and causes; mechanism; relation of crises to the wealth of countries; influence on the economic state and how to measure it; bibliography.
References
Juglar, Clément. Des crises commerciales et de leur retour périodique en France, en Angleterre et aux États-Unis. 2nd ed., Paris: Guillaumin, 1889. Reprinted with an introduction by Ludovic Frobert, Lyon: ENS Éditions, 2014 (Bibliothèque idéale des sciences sociales). Open access: https://doi.org/10.4000/books.enseditions.1382.
Juglar, Clément. “Crises commerciales.” In Maurice Block, ed., Dictionnaire général de la politique, Paris, 1863. Reprinted in the above.
Juglar, Clément. “Crises commerciales.” In Léon Say and Joseph Chailley, eds., Nouveau dictionnaire de l’économie politique, Paris: Guillaumin, 1891. Reprinted in the above.
Frobert, Ludovic. “Présentation.” In the 2014 edition above.
Schumpeter, Joseph A. Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist Process. New York: McGraw-Hill, 1939.
Schumpeter, Joseph A. History of Economic Analysis. New York: Oxford University Press, 1954.
Besomi, Daniele. “‘Periodic crises’: Clément Juglar between theories of crises and theories of business cycles.” Research in the History of Economic Thought and Methodology 28A (2010): 169–283.
Bridel, Pascal. “Juglar, Walras et Pareto: measurement without theory vs theory without measurement.” Revue européenne des sciences sociales 47, no. 143 (2009), special issue “Clément Juglar: les origines de la théorie des cycles,” P. Bridel and M. Dal Pont Legrand, eds.
Dal Pont Legrand, Muriel, and Ludovic Frobert. “Le ‘prophète des crises’: économie politique et religion chez Clément Juglar.” Astérion 7 (2010).
Allisson, François, L. Bredan, and Pascal Bridel. “Bibliographie des écrits de Clément Juglar.” Revue européenne des sciences sociales 47, no. 143 (2009).
Page and paragraph references in the text follow the numbering of the 2014 ENS Éditions edition.