Walras as an ordoliberal?

Abstract

Léon Walras and the ordoliberals share the opinion that State intervention in favour of a competitive order is a central element of economic policy. Hence, can Walras be regarded as a forerunner of ordoliberalism? This study performs a methodological and ontological analysis of Walras’ and Eucken’s thoughts and sheds light on another common ground: philosophical idealism. By taking different inclinations – Walras’ Teleological Realism vs Eucken’s Historicist Conceptualism – these authors reveal different relations with reality and methodological stances, which result in opposing philosophies of History. Paradoxically, by revealing tenuous epistemological bonds, we set a new distance between Walras and the ordoliberals.

Keywords:

• Competition
• Economic Policy
• Eucken
• Ordoliberalism
• Walras

JEL codes:

• B13
• B25
• B30
• B41

Acknowledgements

We would like to thank the members of the Centre Walras-Pareto for their supportive remarks on this article. The paper also benefited from the reading of Alain Alcouffe, Richard Aréna, Gilles Campagnolo, Patricia Commun, Pierre Dockès, Jérôme Lallement and Bertram Schefold. Finally, we are grateful to the two anonymous referees for their critiques and comments. Usual disclaimers apply.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 For a presentation of ordoliberalism see, among others (Commun 2016; Grossekettler 1989, 1994; Kolev 2015; Rieter and Schmolz 1993; Tribe 1995; Vanberg 1998, 2001).

2 This common interpretation – or ‘Schumpeterian’ interpretation – comes from the incorporation of Walras’ theory by the neoclassical school. Éléments d’économie politique pure, considered as the magna carta of economics, should only be taken into account. Others walrasian works are non-systematised and subject to various evaluation criteria that have been overlooked because they are not economics writings. The theory of general economic equilibrium (henceforth: TGE) is a model in the usual sense of the word, a heuristic representation expressed through a set of assumptions that are more or less faithful to a reality. It is a representation – a necessarily incomplete one – of the real. The ontological question is therefore not relevant. Economic policy is reduced to a minimum here: laissez-faire prevails. Given this interpretation, Walras and ordoliberalism do not communicate at all to any degree; this reading grid is therefore not relevant to this study. For a history of TGE’s autonomous path in modern economic science, see (Cot and Lallement 2006; Ingrao and Israel 1990).

3 Jöhr (1957). ‘Léon Walras als Vorläufer des Ordoliberalismus: Eine dogmengeschichtlich-vergleichende Untersuchung über das wirtschaftspolitische Werk von Léon Walras und seine Beziehungen zum Ordoliberalismus.’ Eberhard-Karls-Universität zu Tübingen, Baden-Württembergs: Deutschland. This thesis is unnoticed in the history of economic thought literature, but a copy can also be found in William Jaffé’s archives at the Scott Library, York University (Ontario, Canada). All references that are not originally in English has been translated by us.

4 Jöhr totally dismissed this interpretation, naturally. It results from the renewal of the studies on Walras that have accompanied the archiving work (Bridel and Baranzini 1996; De Caro 1980, 1985) and the publication of Œuvres complètes. For an overview – not a comprehensive one though –, see (Dockès 1996; Rebeyrol 1999; van Daal and Jolink 1993; Walker 1983).

5 As Mäki puts in general: ‘realism does not require realisticness’ (1998, 409–410). He defines ‘realisticness’ as the property of a theory – or of its hypotheses – to conform to the reality of the observed world.

6 To some extent, Eucken can be regarded as the epistemological benchmark for all the other ordoliberals, as they themselves explicitly referred to his concepts and analyses in this field; see for instance Wilhelm Röpke (1942, 143, 1944, 35, 38) and Alexander Rüstow (1950, 1980, 76); or Eucken’s students like Friedrich Lutz (1940, 1944, 1950) and Leonhard Miksch (1942, 1950).

7 Both Eucken (1940, 340) and Röpke (1963, 16) quote Cassel’s The theory of social economy (1918) as the finest and latest state of analyses on works on TGE, and whose importance is crucial to German economic theory. On the way German economists received Walras at the end of the 19^th century, see Alcouffe (2013).

8 Jöhr is nevertheless aware of some basic differences between Walras’ thinking and the New School of Lausanne’s interpretation (see Jöhr’s ‘Appendix’ dedicated to it). In light of its findings, the New School of Lausanne seems even closer, in its thematic as well as in its policy proposals, to German Ordoliberalism than Walras is.

9 On this topic, one can consult the Chapter ‘Monopoles’ of the Études d’économie politique appliquée (Walras 1992, 181–218), but also ‘L’économie appliquée et la défense des salaires’ (Walras 1992, 245–261).

10 The ‘authorities’ can either ‘intervene to exercise it [the economic monopoly] themselves’, i.e. a State monopoly, or ‘get it exercised by a concessionaire’, i.e. a private firm under public supervision (1992, 191).

11 Walras opted for a monetary quadriga besides State regulation of money creation (on Walras’ monetary theory, see Baranzini (2005)). Ordoliberals preferred gold as the single currency but their absolute automatism must be qualified: for example, Eucken’s favoured the adoption of the Graham Plan (the value of the currency is linked to several goods, and not just to gold). The issue of money and credit deserve a thorough investigation, which Jöhr has overlooked, and it exceed the scope of this article.

12 In his Esquisse d’une doctrine économique et sociale (An outline of an economic and social doctrine), Walras observed, ‘one second social issue would yet need to be resolved, that of the production of wealth by men in a society, preventing some entrepreneurs from making a profit otherwise than in normal conditions of free competition’ (Walras 1992, 425).

13 In part 2, we will have an opportunity to critique that type of interpretation.

14 Jöhr insists that Stackelberg specified this representation analytically: 1° independence of prices (as planning data) and 2° the freedom of trade principle. Jöhr subsequently presented various more or less sophisticated models, with different formalisations, but moving away, we believe, from the original ordoliberal project. Indeed, Stackelberg is based on Walras (and Pareto) as a theoretical reference the General Economic Equilibrium in a free competition regime (Stackelberg 1934, 3–6) without having a discussion of it.

15 On the question sociale in Léon Walras, see Lallement (2010, 2012). As a matter of fact, ordoliberals also intend to solve what they called the ‘new’ social question (see Fèvre 2017b).

16 Jérôme Lallement (2014) specifies matters in the field of ontology. Indeed, these are the ‘natural’ and co-constitutive origins of Man and Society – ‘People and society are two distinct facts, and they are inseparable’ (ibid. 19) – which rightfully gives them access to income.

17 Indeed, Walras asserts land redistribution does not make sense. The point is to determine the conditions for a fair allocation/distribution of social wealth, obviating, or even making unfair any redistribution, which by definition comes later.

18 Jöhr here gives a rather ‘liberal’ interpretation of Walras.

19 Against Jöhr’s reading, recent literature clearly evidenced ordoliberal’s concerns with equality of opportunity, and thus with a priori justice (Vanberg 2014; Wörsdörfer 2013b).

20 Indeed, Walras believes the State is the only entity entitled to land ownership. Income tax is illegitimate because it robs the individual of the remuneration of labour or capital. This solution makes it de facto obsolete by providing the State with another means of financing. Counter-intuitively, these are liberal principles – freedom and meritocracy – that cause Walras to opt for a State land ownership: a solution usually likened to a socialist position.

21 On relationships between State and market as ordoliberals saw them, see, amongst others (Bonefeld 2012; Streit and Wohlgemuth 2000; Vanberg 1998).

22 Walras may be considered as a part of ‘liberal socialism’ that begin in France at his time (Baranzini 2001; for a different interpretation regarding that issue, see Baranzini and Swaton 2013; Herland 2013; Potier 2012). Bourdeau (2005) showed how Walras’ political philosophy can be linked to post-revolutionary republicanism.

23 About this: Walras’ standpoint changed over time; he moved from the triptych to the four sciences (Baranzini 2005, ch. 4.2). Potier (1994), Dockès (1996), Jolink (1996) and Lallement (2000) provide a study of Walras’ pure, applied and moral sciences.

24 See Pierre Dockès (1999, 16).

25 See Robert Nadeau (1999, 154).

26 Walras refers here to Etienne Vacherot (1809–1897), a French philosopher who succeeded Victor Cousin (1792-1867) at the Sorbonne. Vacherot introduced, in particular, German idealism to French-speakers. Besides, he was considered as a neo-Hegelian, in his time.

27 The Realist position is opposed to nominalism. The debate was formulated in these terms in the late Middle Ages on the occasion of the so called feud over universals, claiming back the tutelary figures of Plato and Aristotle. We are mobilising the debate here as it is instrumental in its general formulation, i.e. as ‘a position rooted in the metaphysics of common sense’, in Libera’s words (1996, 17). In his seminal work on the subject, he gives himself the ‘historiographical formula: are universals things, concepts or names? […] Do they exist in the world at large? An affirmative answer is realism, a negative answer is nominalism’ (1996, 14, 18).

28 ‘Rationally’ must not be understood in a modern acceptance in terms of behaviour, but in the sense of loyalty to the principles obtained through Walras’ method, hence as opposed to the ‘experimental.’

29 For Walras, ‘free competition among entrepreneurs is not the only means of bringing selling price into equality with cost of production (…); and Applied Economics has to inquire whether it is always the best means’ (1988, 11 [n°1]).

30 Eucken is steeped – and this is another common point with Walras (Dockès 1999, 31) – in the German idealist approach. His main references are Kant and neo-Kantian philosophy, especially through his father Rudolf Eucken (recipient of the Nobel Prize for Literature in 1908) and Husserl (he was friends with and colleague of in Freiburg). On Eucken’s overall epistemological project, see (Broyer 2007; Campagnolo 2003; Herrmann-Pillath 1994; Klump and Wörsdörfer 2011; Kuhnert 2008; Weisz 2001a).

31 On the links between Husserl’s epistemology and Eucken’s, see, in particular (Campagnolo 2003, 134–138; Herrmann-Pillath 1994, 49; Klump and Wörsdörfer 2011).

32 It is interesting to note that the Methodenstreit can be read partly as a reenactment of the quarrel over Universals, insofar as the opposition between Menger and Schmoller had occurred fundamentally at ontological level, either between the Austrian’s realism and the German’s nominalism (see, amongst others, Hédoin 2013, 105–110; Mäki 1997; Pribram 1986, 224). Such superposition is only imperfect, as highlighted by Gilles Campagnolo (2011, 475–478) because Menger does not fit in a ‘pure essentialism.’

33 Conceptualism posits Universals do exist [Ideas, forms] that are nonetheless creations by the human mind and nothing corresponds to it outside the mind’ (Nadeau 1999, 77). Eucken’s opinion apparently mixes epistemological conceptualism (first definition) and ontological conceptualism (Ideas come from the immanence of concrete objects), usually opposed to each other (Vidal-Rosset 2010). Concepts do have a foundation in reality since they correspond to the objective essence of things, but they result from an encounter that is mediatised through our own subjective thinking.

34 Bertram Schefold (1995) gives an excellent presentation of Eucken’s views.

35 On that particular point, but also regarding methodology, German ordoliberalism was compared with the theory of French regulation; see their collective work (Labrousse and Weisz 2001).

36 A distinction should be established here between the ‘positive’ study of economic systems, including the regulated economy and the normative economic policies supporting a liberal order to a some extent, for example at the core of the German economist’s later works (Eucken 1948a, 1948c, 1949, 1951, 1952).

37 Michel Foucault, in his course on biopolitics (2004), particularly stressed this point : the inextricable relationship between State and Market(s), including the ordoliberal thought (for a review of the treatment of ordoliberalism by the French philosopher, see Wörsdörfer 2013a). Walras’ role in this genealogy of neoliberalism is more ambiguous, since Walras does emphasise the State’s foresight regarding organisational matters without adhering to a policy for society (see Bee 2008). Of course, Ordoliberals make one more step than Walras towards political theory, and they do so through their insistence on the significance of law for the economy to function properly. Beyond a few points of acquaintances, Foucault also gives us the limits of the comparison: ‘Ordoliberal analysis will not pertain at all to the line of thinking of that economic theory of competition (…) defined by Walras (…); it will be part of a whole line of law theory, the theory of the rule of law governed by the State, which has been very important in the history of German legal thinking and German institutions’ (Foucault 2004, 173).

38 On this, see (Arena and Ragni 1994; Baranzini 2011; Dockès and Potier 2005; Potier 1999).

39 On the irrelevance of the positive/normative divide in Walras, see (Baranzini 1993; Lallement 1997).

40 The ‘State socialism’ Wagner claimed seems rather close to the Walrasian approach when he suggested ‘avoiding errors of economic individualism and socialism as much as possible, strike a balance and stick to it’ (Wagner 1892, 26). This is the subject of another article. On that topic, see (Broyer 2003; Peukert 2000; Schefold 2003). We wish to thank Bertram Schefold in particular for this suggestion.