Crises in ModellingArticulations of the Romanian Labour Market in the Long 1990sAlina-Sandra Cucu

Introduction

Needless to say, one hardly expects the technical documents of international organizations to be written in a confessional voice. But this was exactly the writing style adopted by Cătălin Zamfir, a Romanian technocrat and well-known social scientist, in the preface of a working paper requested in 1992 by the International Labour Organization. Zamfir introduced himself as ‘the man who had the unenviable task of being Minister of Labour in the first post-revolution government’, a government ‘elected in May 1990 with a promise to prevent unemployment and protect workers’ living standards’.11Cătălin Zamfir, The Romanian Wage System in the Transition to a Market Economy, working paper 54 (Geneva: International Labour Organization, 1992). According to him, this government had the historical responsibility to steer ‘a course away from the highly centralised and bureaucratic labour system prevailing in the latter years of the Ceaușescu regime, towards one based on social consensus, collective bargaining and a social sharing of the costs of the transition towards a regulated market economy’.22Ibid. After complainingBeginning of page[p. 182] about the ‘lack of trust’ and ‘the psychological distress’ that marked labour relations in Romania in the early 1990s,33By the time the working paper was published, Zamfir had resigned his position, less than a year after he was appointed, citing the same reason of not being able to protect the well-being of the workers during the transition to a market economy. Zamfir concluded by admitting that ‘[g]iven the social tensions and political situation, [preventing unemployment and ensuring workers’ well-being] was almost certainly an impossible task to achieve’.44Zamfir, Romanian Wage System.

A few years later, Emilian Dobrescu, a Romanian economist whose career has spanned more than six decades and involved a number of political roles, was complaining about his own historical responsibility, which was to build econometric models of the transition. Talking specifically about models of the labour market, he lamented that, in the worst-case scenario, his task was impossible and in the best-case scenario rather useless, since models are meaningful only when ‘a given economy does not register major disequilibria and develops in a predictable manner’.55Emilian Dobrescu, ‘Double Conditioned Potential Output’, paper delivered at the 28th General Conference of the International Association for Research in Income and Wealth (Cork, Ireland, 22–28 August 2004) <https:/​/​doi.org/​10.2139/​ssrn.1434909>. According to Dobrescu, the Romanian economy in the 1990s was a far cry from this description: market mechanisms were not yet functional; the culture of informality dominated ordinary economic life; the state maintained a high allocative power; and a deep fracture between the real and nominal sectors of the economy persisted.66The real/nominal distinction refers here to two aspects of measuring the economy: real production vs monetary expressions, and inflation adjusted vs non-adjusted monetary gains. Adding to these structural problems, the incompatibility between the indicators used to measure employment and productivity in late socialism and the ones imposed by the World Bank and the International Monetary Fund immediately after 1989 set further limitations for any effort to construct long-term econometric series.

Dobrescu pointed out that economists at the time knew little to nothing about ‘the microeconomic foundations of the transition’; they simply lacked ‘the necessary assumptions for building up econometric functions’.77Dobrescu, ‘Double Conditioned Potential Output’. Macroeconomic models could hardly say anything in aBeginning of page[p. 183] context dominated by high levels of uncertainty resulting from ‘the behavioural instability of economic agents’.88Ibid. The Homo economicus of the transition was hardly knowable, and most likely different from both the worker on which the numbers of the socialist plan had been configured, and from the agent of neoclassical economics presupposed by the World Bank advisors. The agents that the newly imported labour market models referred to were thus placed in a conceptual space between ‘not-yet-gone’ and ‘not-yet-arrived’.

This chapter investigates what happens when the agent’s behaviour that normally underpins a family of models — in this case, models of the labour market — cannot be theorized. I argue that, in their everyday professional practice, the reference point of Romanian economists in the 1990s was not the self-interested, fully informed agent of neoclassical economics, an abstract figure populating ‘the world in the model’99Mary S. Morgan, The World in the Model: How Economists Work and Think (Cambridge: Cambridge University Press, 2012) <https:/​/​doi.org/​10.1017/​CBO9781139026185>. with his1010I use the masculine pronoun for the economic agent here for two reasons. First, because until recently and with the emergence of feminist economics, the language of economics has not been gender-neutral and the economic agent was presupposed to be male. Second, economics builds on the assumption of methodological individualism. In the Romanian 1990s, economists used masculine pronouns and the singular form. This is exactly how postsocialist economists talked about economic agents: using singular masculine pronouns (and nouns, which are also gendered in the Romanian language). ‘infinite ability to make rational decisions’ for maximizing utility and profit.1111James Chen, ‘What Is Homo Economicus? Definition, Meaning, and Origins’, Investopedia, 31 July 2021 <https://www.investopedia.com/terms/h/homoeconomicus.asp> [accessed 15 June 2024]. They also did not necessarily see the economic agent as the embodiment of humanity’s state of nature, a lost state of grace perverted by political intervention. The Romanian economists seemed to understand Homo economicus, the man in the models they operated with, as a real historical figure, one that emerges only in relation to a fully functional market. And fully functional markets need to be made.

I focus here on how, in the 1990s, markets — more specifically, labour markets — were made to fit the economic models that were supposed to describe them. I will show that, despite being rather useless forBeginning of page[p. 184] illustrating what was happening during the transition from a centrally planned economy to a market economy, labour market models were instrumental in bringing about the reality they were supposed to depict. In this sense, they worked performatively, as promissory utterances.

By constructing models of the labour market in the 1990s, the Romanian economists had two explicit aims: to assess the effectiveness of job allocation mechanisms and to predict the evolution of unemployment. Both issues were central to the debates concerning the privatization programmes in postsocialist Eastern and Central Europe. These programmes brought about massive deindustrialization and the rapid crumbling of the centrally planned system of manpower allocation, which had for decades matched people, skills, and jobs according to the ‘needs of the socialist economy’. During the transition, economists suddenly faced the challenge to model social phenomena that did not meet the most basic assumptions of the models themselves. Workers faced, in no ambiguous terms, a loss of future — one that produced not only existential fragility but also moral panic. Although bearers of different interests, aims, and world visions, workers and economists struggled to make sense of an omnipresent term — the market, a concept international advisors and investors seemed to have no problem with. In this chapter, I will sketch some preliminary thoughts as to how our understanding of the post-1989 economic transition in Eastern and Central Europe changes, if one takes seriously the fact that, at the time, the market was a rather puzzling idea for the social actors.

What follows goes against a large part of the scholarship on the region, which, for more than thirty years now, has exhibited the tendency to reify ‘the plan’ as the thing the region transitioned from and ‘the market’ as the thing it transitioned to. I follow up here on my long-term work on central planning, which has questioned the persistence of political and scholarly imaginaries that understand labour relations in state socialism through the absence of markets, and through a model of supply and demand for manpower that assumes that its reproduction was ever fully regulated. I have opposed this dominant perspective on two grounds. First, while discursively prominent, when analysed on the ground, neither perfect regulation nor the complete excision of market mechanisms were actual features of state socialism. Second, and most important, both regulatory and market mechanisms wereBeginning of page[p. 185] simply serving the broader aims of state socialist economies as accumulation regimes, in ways that uncomfortably resemble the everyday reality of capitalist societies.1212Alina-Sandra Cucu, Planning Labour: Time and the Foundations of Industrial Socialism in Romania (Oxford and New York: Berghahn Books, 2019). But the ‘market’ that the workers and the economists became acquainted with in the 1990s was of a different order, not only because it was supposed to reign supreme, but also because it had alarming consequences for people’s everyday lives.

In questioning the dominant framework of the analyses of state socialist regimes, I have relied on a performativity of economics perspective, which I also adopt here to discuss the relationship between econometric models and the making of a labour market in postsocialist Romania. In the simplest terms, performativity of economics approaches theories and models in the field as discourses with the power to shape economic reality, basically by telling other actors what markets do, by prescribing what they should do, and by predicting what they will do.1313See Michel Callon, ‘Introduction’, in Callon, The Laws of the Markets (London: Blackwell, 1998). The economic world ends up resembling the theory or the model not because of the quality of the model but because the economic discourse has previously influenced reality in specific ways. For instance, in order for a specific commodity market to emerge, concrete goods must appear first as commodities, that is, to be produced in order to be exchanged for a monetarized profit. This requires legal intervention, social acceptance, and a medium of exchange (like money).

The idea (along with its fierce critiques) that in capitalism economists ‘make markets’ rather than simply analyse them has given a fresh impetus to the social sciences and the history of economics in the last two decades.1414See Timothy Mitchell, Rule of Experts: Egypt, Techno-Politics, Modernity (Berkeley, CA: University of California Press, 2002) <https:/​/​doi.org/​10.1525/​9780520928251>; Marion Fourcade, Economists and Societies: Discipline and Profession in the United States, Britain, and France, 1890s to 1990s (Princeton, NJ: Princeton University Press, 2010) <https:/​/​doi.org/​10.1515/​9781400833139>; John L. Campbell and Ove K. Pederson, ‘Knowledge Regimes and Comparative Political Economy’, in Ideas and Politics in Social Science Research, ed. by Daniel Béland and Robert Henry Cox (Oxford: Oxford University Press, 2010) <https:/​/​doi.org/​10.1093/​acprof:oso/​9780199736430.003.0009>; and Franck Cochoy, Martin Giraudeau, and Liz McFall, ‘Performativity, Economics, and Politics: An Overview’, Journal of Cultural Economy, 3.2 (2010) <https://doi.org/10.1080/17530350.2010.494116>. However, the discourse on the performativity ofBeginning of page[p. 186] economics has never been seriously brought to bear on what happened on the other side of the Iron Curtain, either by the actors themselves, or by scholars of state socialism. The straightforward, linear, unidirectional functioning of economics as a prop for the political aims of the communist rulers continues to be taken for granted in most academic fields, and continues to shape our thinking about five decades of regional and global history. There is no mystery in the fact that the plan was explicitly performative — both in the broader sense of a discursive field intentionally producing societal effects, and in the more restricted sense of bringing together elements of specific economic models through various practices and relations and making them work in practice. However, the performativity of socialist economics was neither as straightforward nor as top-down as the literature on centrally planned economies usually assumes. As socialist planners were painfully aware, it was something that remained to be struggled for and achieved, always elusive, and often contested.1515Cucu, Planning Labour.

But it is the economic transition of the 1990s that offers one of the most striking examples of the performativity of markets. As I have already mentioned, the idea that in capitalism economists ‘make markets’ has been a fertile one. It can be summarized as follows: economists, rather than objectively depicting or measuring a particular aspect of reality, help shape that aspect of reality to resemble the model they propose. Economists help to make markets, rather than analyse them, because the assumptions of their models become hegemonic, and end up functioning as guides for action. The economic transition of the 1990s opens a strategic research site to explore how this happens in practice. However, what I will show is that it is not only economists who make markets. Union leaders, international organizations, and citizens, trying to make sense of their own historical encounter with unemployment, make markets, too. Various actors have to participate in materializing the most basic models of the labour market. This is something the performativity of economics thesis should not ignore, and is why I here adopt Georg Lukács’s historical materialist assumption that ‘all economic or “sociological” phenomena derive from the socialBeginning of page[p. 187] relations of men to one another’.1616Georg Lukács, ‘Technology and Social Relations’, New Left Review, 39 (1966), pp. 27–34 (p. 29), emphasis in the original. Formulaic images of the world, like models of the labour market, are thus always an intrinsic part of the world they describe. It is a world whose laws are subject to change, and whose knowledge regimes are rooted in praxis. The procedures and institutions that can sustain the reiterations, sedimentations, and variations that performativity requires1717Judith Butler, ‘Performative Agency’, Journal of Cultural Economy, 3.2 (2010), pp. 147–61 (p. 147) <https://doi.org/10.1080/17530350.2010.494117>. belong to the same historical experience that models attempt to objectify.

Bearing in mind this historical materialist correction to the performativity thesis, I will raise two questions about the 1990s economic transition in Romania. How did something called ‘the Romanian labour market’ come into being, precisely when socialist industry was disintegrating and little else seemed to be taking its place? And what challenges did economists face when trying to construct models that had to look as if they were describing actually existing social relations, and as if they were referring to a real thing?

In the following section of this chapter, I return to Emilian Dobrescu to understand how the material conditions in which the postsocialist labour market was articulated set limits to the performative capability of economic discourse. In the subsequent section, I re-examine the classical elements that had to be set in place in order for the economic models of the Romanian labour market to have a referent in reality: private ownership of the means of production; the theoretical and legal possibility of unemployment; and tripartite labour relations. I conclude by gesturing towards the sources of economic models’ generative dimension.

The economists’ troubles with the labour market

The Romanian 1990s were a historical moment at which the use of models was proliferating in policy circles. In Dobrescu’s words:

It seems strange to notice how the turbulent state of the economy, though creating serious limitations for the cognitive capacity of mathematical models, increases the need of the decisionBeginning of page[p. 188] makers and of a large part of public opinion for prospective works — programs, strategies, forecasts — which utilize such instruments. It is probably a reaction against the psychological stress of uncertainty.1818Emilian Dobrescu, ‘Unstable Processes and Macroeconomic Modelling’, Romanian Economic Review, 41.1 (1996a), pp. 17–34 (p. 27).

He called this a ‘paradox’ of the transition. The paradox was translated into methodological anxieties and doubts related to three issues: models’ capacity to accurately capture aspects of reality during the postsocialist transition; the appropriateness of ‘Western econometrics’ as a theoretical foundation for understanding the economic transformations of the 1990s; and the possibility of proposing a theoretical framework that would be rooted exactly in the social world and in the historical time it belonged to.

Educated as part of the late socialist technocratic elite, and specializing in growth models and the mechanisms of accumulation of planned economies, Dobrescu made his own transition in the 1990s, becoming a trainee at the Hoover Institute and the World Bank, tasked with modelling the most important macroeconomic phenomena in the Romanian economy. Among the available models were standard employment models that defined rather than described actually existing labour markets: the Beveridge curve, the Phillips curve, and the natural rate of unemployment. The Beveridge curve shows that, at any given moment, there are both jobs available and people unemployed, while the shape and the position of the curve provide important information about the functioning of the labour market.1919There are two key concepts associated with the Beveridge curve: labour market tightness and matching efficiency. Labour market tightness is the number of vacant posts per each unemployed person and matching efficiency reflects the market’s ability to match individuals to jobs. See Agostino Consolo and António Dias da Silva, ‘The Euro Area Labour Market through the Lens of the Beveridge Curve’, ECB Economic Bulletin, 4 (2019) <https://www.ecb.europa.eu/pub/economic-bulletin/articles/2019/html/ecb.ebart201904_01~9070de27a0.en.html> [accessed 12 June 2023]. The Phillips curve addresses the stable and inverse relationship between inflation and unemployment;2020Higher inflation is associated with lower unemployment and vice versa. and the natural unemployment rate is the lowest level of unemployment that can be sustained by an economy without creating inflation.2121The natural rate of unemployment condenses three types of unemployment: frictional unemployment, when people are between jobs; structural unemployment, when there is a mismatch between skills and job characteristics, or when labour-replacing technologies take over in a certain economic sector; and surplus unemployment, when state interventions like minimum wage legislation or price control make some categories of workers too expensive to be employed. The natural rate of unemployment is the unemployment rate that is consistent with price stability — defined as two per cent inflation in the US, for instance — and the economy growing at full capacity. In contrast to the actual unemployment rate, the natural rate of unemployment cannot be observed directly, but is instead estimated using macroeconomic models.Beginning of page[p. 189]

The first question that the Romanian economists struggled with concerned the power of models to capture aspects of an extremely fluid socio-economic reality. When asked to produce deterministic, predictive macroeconomic models, Dobrescu complained that in such an unstable environment, econometric functions did not ‘satisfy the statistic tests for parametric constancy’.2222Namely, recursive least squares coefficients, one-step residuals, and recursive Chow tests. Emilian Dobrescu, ‘Unstable Processes and Macroeconomic Modelling’, in Macroeconomic Modelling, ed. by Emilian Dobrescu (Bucharest: Centre for Economic Information and Documentation, 1996b), pp. 3–4, 1–40, 3. This ‘intrinsic instability’ also meant that ‘the evolution of some indicators [wa]s marked by spectacular breaks, new connections appear[ed] between the real and nominal economies, [and] a new network of economic agents emerge[d], having atypical, even unpredictable, behaviours’.2323Ibid., p. 6. It also meant that, for a brief moment, the past was winning the battle, and ‘the pressure of highly inertial components of the economy prevailed upon that of highly dynamic components’.2424Ibid.

Economic agents, including households, were considered to be ‘the least stable in comparison with other determinations of the economy’.2525Ibid., p. 14. Fluctuations in agents’ behaviours were frequent, large, and erratic during the 1990s. Such fluctuations were actually higher in postsocialist Romania than any established theoretical frame would have allowed, and some economists envisioned the possibility that the sum of individual erratic behaviours would simply result in chaos. In addition, access to information was scarce and highly asymmetrical, making it difficult even for market radicals to speak about ‘rational expectations’. Hence, labour productivity, gross added value in industry, gross added value in construction and services, share of active population in the total population, and share of labour revenues in the totalBeginning of page[p. 190] gross added value (which is sometimes a good measure of exploitation) were among the indicators deemed problematic for building econometric models of the transition.

The second problem raised by the economists of transition was the appropriateness of ‘Western econometrics’ as a theoretical foundation for understanding the economic transformations of the 1990s. Although Dobrescu’s voice was more marginal when he critiqued the tendency of Western econometrics to become increasingly atheoretical and removed from the contexts in which its predictions were formulated, he was not alone among Romanian economists in expressing his doubts that a theory existed that would work for the transition from socialism to capitalism. Synthesizing these positions, Dobrescu stated:

A first cause of such a situation could be the absence of a suitable theory that would permit the correct definition of the really determining factors of a macroeconomic correlation, the direction and intensity of their influence. The intuition and the flair of the researcher alone are not enough because of a major risk: the most consistent functional relations can be overlooked (no matter how large the number of trials is). Regarding the mechanism of the transition economy, we do not have up to now, unfortunately, either assumptions, perfectly compatible with the theories generally accepted by the scientific community, or a coherent and secure set of hypotheses, corroborated with the present experience. This is the explanation for the ideatic eclecticism which characterizes most of the papers dedicated to the transition.2626Ibid., p. 21.

Crucially, Dobrescu led a disciplinary wave that proposed a theoretical framework rooted in the turbulent social world of the 1990s. The assumptions of this theoretical framework were very different from the neoclassical ones. The economy of the transition was defined by the following parameters: institutionally, it was a weakly structured system, with a mix of old and new rules,2727Factories, for instance, still had production targets based on the last five-year plan of the socialist period, but in many cases, demand had started to become ‘liberalized’, so many industrial units found it impossible to sell their commodities. a strong informal dimension, discretionary political authority, and no clearly defined property rightsBeginning of page[p. 191] (which, as we will see, led to the historical oddity of unions fighting for capital formation). It was a framework that gave no clear solutions for measuring and predicting phenomena like unemployment, job creation, or job matching, making all three models that the international organizations wanted to immediately implement in Romania — the Beveridge curve, the Phillips curve, and the natural unemployment rate — basically useless. The modelling problems were, in the economists’ words, ‘very complicated’.

Dobrescu advised his fellow economists that the only ‘reasonable objective for such a fluid operational context’ was to limit themselves to short-term prediction and to use weaker criteria for their models, though he considered even these weak criteria difficult to specify.2828Dobrescu, ‘Unstable Processes and Macroeconomic Modelling’ (1996b). Pursuing these objectives was, indeed, likely the most reasonable thing to do. It was, however, not a widely accepted turn in the field of transition economics.

With imploding theoretical foundations and microeconomic assumptions, the Romanian economists obviously struggled to construct models that had to look as if they were describing actually existing social relations, and as if they were referring to a real thing. Dobrescu’s line of thinking recognized the impossibility of theorizing a subject of the econometric models of transition when everything was in flux. The reason was that the economic subject made sense only in relation to a predictable reality: the free market. Owing his position to his doubly technocratic education, Dobrescu militated for a knowledge regime that was rooted in praxis, and seemed aware of the fact that economic models are or should be integral parts of the reality they promise to represent, and that the epistemological limits of their claims to representation have to be specified. While never following the implications of his thinking to their logical consequences, Dobrescu wrote with a clarity given by the material conditions he was caught in, conditions that allowed him to see that, far from being a natural order of things, the market was a historically contingent arrangement, which had to be institutionally created in order to exist at all.

The following section focuses on how something called the Romanian labour market materialized in the 1990s, at a moment whenBeginning of page[p. 192] socialist industry was collapsing, the future was meagre, and the state continued to have high stakes in the ownership of the factories. I trace the results of various actors’ efforts to set in place the elements needed for its existence: private ownership of the means of production; the theoretical and legal possibility of unemployment; and the emergence of a framework for the relationship between labour, capital, and the state (henceforth tripartite labour relations). Many of these efforts were intentional. They provided the necessary fuel for the reiterations, sedimentations, and variations that allowed the economic models I have addressed so far to work performatively. What follows will allow the reader to understand how bringing a certain type of economy into being works concretely, and how this materialization requires that the relevant social actors perform specific actions, which economic models assume. I will also show that in order for economic models to have a referent in reality, new models of social relations also have to emerge and become institutionalized.

Unemployment and the making of tripartite labour relations

While Emilian Dobrescu was complaining about the fact that the labour market had little influence on the postsocialist reality, international organizations, newly founded trade unions, and state institutions were negotiating the emergence of the tripartite institutional framework for labour relations. The tripartite model was never questioned or challenged by the relevant Romanian social actors, and was considered the only path towards a functional labour market. It emerged as a dominant belief at the time, a dogmatic understanding of the new political economy of transition, in which there was only one way in which labour relations could be organized. This model had to be taught from abroad, and had to break completely with anything that workers had learned through experience during the socialist period and after.

The idea that tripartism was the only conceivable way to organize labour relations in postsocialist Romania was not relegated to institutions like the International Labour Organization or the World Bank. It was also a core element of the nascent postsocialist trade unionism.Beginning of page[p. 193] The uncritical acceptance of the tripartite model led to a historically unique pressure for privatization and for the creation of a class of employers coming from the trade unions themselves. Blaming every possible ill in the factories on the absence of private investors and owners, the unions never considered any political alternative in which the state would continue being their employer for the long run (not that such an alternative was tenable for political parties either).

Experts from the main international organizations — the International Monetary Fund and the World Bank — considered that the pace of Romanian privatization was too slow, calculating at one point that there was a ten-year delay between Romania and other countries in the region where the transition to market economies was concerned. The slow pace of privatization meant that the Romanian state still held ownership of the largest industrial units in the country, being thus trapped in the contradictory position it had inherited from late socialism — as capital, employer, and manager. Since, during the 1990s, the Romanian state was failing in all its capacities, organized labour could fight against all its capacities, with all the new legal and political instruments at its disposal. The unions’ structure, repertoires of contention, and styles of action reflected the fact that, for almost a decade after the 1989 Revolution, they negotiated almost exclusively with the Romanian state. It is not a coincidence that the power of the union leaders started to erode after 1997, when the privatization process was, to a certain extent, decentralized, and many attributions of the State Property Fund were transferred to regional authorities. The Romanian trade unions were thus practically born with the belief that the unions-government-employers triad was the natural state of affairs, and that no real negotiations for wages, working conditions, or unemployment protection could take place until the three pillars were in place. Among these topics of negotiation, unemployment represented a new threat, one that broke the relationship between the possibility of stable employment and people’s life courses that had characterized state socialist economies.

While the issue of supply and demand for manpower was also fundamental for socialist planners, it was only when unemployment became a legal and theoretical possibility that something that could be recognized as a labour market could emerge. The Romanian authoritiesBeginning of page[p. 194] began to register unemployment only in 1991. The beginning of the registration process was followed in 1994 by the first employment survey conducted according to International Labour Organization standards. Registered unemployment rose continuously until 1999, and has been constantly underestimated because of the high level of temporary and seasonal migration to Western Europe, as well as because of the large number of people working in the informal economy. Under the guidance and supervision of the International Labour Organization, the World Bank, and the Organisation for Economic Co-operation and Development, the government created the institutional infrastructure that would deal with unemployment: labour offices,2929Of course, employment offices were not invented in the 1990s. Under state socialism, every enterprise and every cooperative centre had its own labour office. and several institutions charged with organizing training programmes for those affected by collective dismissals.3030Especially under the aegis of IROMA — the Romanian Management Institute. See ILO RL 52-3-01, Rapport de mission de Sergio Ricca en Roumanie (12–19 April 1991), p. 4.

The chaotic processes of the 1990s act as a magnifying glass for an understanding of the relationship between models of the labour market and the reality they try to represent. It is hard to say, however, how models like the Beveridge curve, which is supposed to describe the efficacy of job allocation mechanisms, could represent anything in postsocialist Romania, a context where the institutions tasked to collect data about unemployment were still to be created. It is hard to understand how the Phillips curve could show an inverse and stable relationship between inflation and unemployment, when inflation was rocketing but workers were still employed in state factories, and their incomes were indexed at a fixed rate. At the most fundamental level, these models are supposed to have a referent in reality, an already-existing, out-there, stable-enough labour market. These models come with their own stories; they are, as Mary Morgan would claim, narrative in nature. They certainly come with their own characters: job offerers and job seekers, which are supposed to abide by rationality norms in which prices are paramount. It is only under these conditions that talk of the neoclassical Homo economicus even begins to make sense at all.Beginning of page[p. 195]

Models and sources of performative work

The tendency to reify ‘the plan’ as the thing Eastern and Central European countries transitioned from, and ‘the market’ as the thing they transitioned to, has been a common feature of scholarship on the region. First, because the lack of temporal (as well as ideological) distance made it difficult to analyse the postsocialist 1990s as a moment that belonged to a centuries-long regional trajectory of class formation, on the one hand, and to far broader global transformations, on the other. And second, because social scientists have rarely if ever looked critically at the economics of transition, a field whose methodological puzzles at the time could teach us something about how problematic the fundamental terms taken for granted in our scholarly conversations actually were.

To the extent that they described the labour market as it was supposed to be, the models that circulated in policy circles in the 1990s could be considered as promises, and functioned as particular cases of what John L. Austin calls ‘performative utterances’.3131John L. Austin, ‘Performative Utterances’, in Austin, Philosophical Papers, ed. by J. O. Urmson and G. J. Warnock (Oxford: Clarendon Press, 1961), pp. 233–52 <https:/​/​doi.org/​10.1093/​019283021X.003.0010>; How to Do Things with Words [1962] (Oxford: Oxford University Press, 1976) <https:/​/​doi.org/​10.1093/​acprof:oso/​9780198245537.001.0001>. Eve Kosofsky Sedgwick further specifies the nature of promises as ‘explicit performative utterances’, which reveal the ‘directly productive aspect of language’ even in cases ‘when the utterances in question are closest to claiming a simply descriptive relation to some freestanding, ostensibly extradiscursive reality’.3232Eve Kosofsky Sedgwick, Touching Feeling (Durham, NC: Duke University Press, 2003), pp. 4–5; emphasis in the original <https:/​/​doi.org/​10.1215/​9780822384786>.As classical speech act theory has taught generations of scholars, in order to make things happen, utterances have to meet certain felicity conditions. I suggest that felicity conditions are predicated on broader historical conditions of possibility, which allow some economic ‘models of’ to become ‘models for’, or, to use Mary Morgan’s terms, to make the transition from ‘the world in the model’ to ‘the model in the world’.3333Morgan, The World in the Model, pp. 378–412.Beginning of page[p. 196]

The Romanian 1990s provides a striking case for testing the idea that ‘economics does not describe an existing external “economy”, but brings that economy into being: economics performs the economy, creating the phenomena it describes’.3434Michel Callon, quoted in Donald MacKenzie and Yuval Millo, ‘Constructing a Market, Performing Theory: The Historical Sociology of a Financial Derivatives Exchange’, American Journal of Sociology, 109.1 (2003), pp. 107–45 <https://doi.org/10.1086/374404>. But this chapter has also shown something else: that in order for this economic reality to materialize, other actors need to perform the required gestures. In this case, quite surprisingly, alongside international organizations and state institutions, trade unions were powerful actors working for the actualization of the conditions assumed by these economic models. These conditions were themselves caught in a model of a different order, this time, explicitly, a model of social relations. This means that the real source of the performative power of economics must be found somewhere else, in the material conditions within which economists’ ways of thinking, modelling, and writing emerge.

Whether and how models become ‘engines’ that lead decision-making in certain directions (as economic sociology envisions them);3535See Donald A. MacKenzie, An Engine, Not a Camera: How Financial Models Shape Markets (Cambridge, MA: MIT Press, 2006) <https:/​/​doi.org/​10.7551/​mitpress/​9780262134606.001.0001>; and Donald MacKenzie, Fabian Muniesa, and Lucia Siu, Do Economists Make Markets? On the Performativity of Economics (Princeton, NJ: Princeton University Press, 2008) <https:/​/​doi.org/​10.1515/​9780691214665>. whether and how models work as prescriptions, guides, or rhetorical dispositifs that legitimate particular practices, institutional arrangements, and subjectivities (as the new cultural economy affirms);3636See Deirdre McCloskey, The Rhetoric of Economics (Brighton: Wheatsheaf, 1986); Catherine Chaput, ‘Rhetorical Circulation in Late Capitalism: Neoliberalism and the Overdetermination of Affective Energy’, Philosophy and Rhetoric, 43.1 (2010), pp. 1–25 <https://doi.org/10.1353/par.0.0047>; Catherine Chaput and Joshua S. Hanan, ‘Economic Rhetoric as Taxis: Neoliberal Governmentality and the Dispositif of Freakonomics’, Journal of Cultural Economy, 8.1 (2015), pp. 42–61 <https://doi.org/10.1080/17530350.2014.942349>; Joshua S. Hanan, Indradeep Ghosh, and Kaleb W. Brooks, ‘Banking on the Present: The Ontological Rhetoric of Neo-Classical Economics and Its Relation to the 2008 Financial Crisis’, Quarterly Journal of Speech, 100.2 (2014), pp. 139–62 <https://doi.org/10.1080/00335630.2014.961529>. and whether and how they become ‘true’,3737John Searle, ‘How Performatives Work’, in Searle, Consciousness and Language (Cambridge: Cambridge University Press, 2002), pp. 156–79 <https:/​/​doi.org/​10.1017/​CBO9780511606366.011>. all dependsBeginning of page[p. 197] on processes that are inherently uneven, agonistic, and usurpative. In other words, whether and how a specific model of the market becomes successful depends upon the outcome of political struggles, including here, of course, the politics of knowledge production within the field of economics.Beginning of page[p. 199]