20 Translation of Central Banking to Developing Countries in the Post-World War II Period: The Case of the Bank of Israel

20.1 Introduction

During the post-World War II period (1946–1975), more central banks were established than in any other three consecutive decades before or after. Most of these were established in developing countries after they had become sovereign states.¹ On the nominal level, it was a case of global convergence. However, examining the policy instruments that central banks employed in different countries, one finds a pattern of divergence or regional convergence: central banks in industrial countries converged around the Keynesian model of central banking while central banks in the periphery converged around a different model.

The analysis of this phenomenon hinges upon several broader and fundamental questions. It raises questions concerning the encounter between economic knowledge produced in industrial countries on the one hand, and the unique institutional structures that prevailed in developing countries on the other. Secondly, it also raises questions concerning the respective roles of national interests and pressures in the international system, as well as processes of learning and experimentation in cases of policy transfer between countries. Finally, it raises questions concerning the relationship between processes of state formation and globalization. State formation is understood here as the transformation of local institutions and authorities as well as the construction of bureaucratic structures that enhance the capacity to govern the economy. Globalization is understood as the diffusion of global standards, rules and policy instruments.

In this chapter, I attempt to explain the fact that a large number of developing countries adopted similar policy instruments within a relatively short period of time. I argue that it can best be explained as a process of policy translation. Moreover, I argue that the capacity of states to translate depends on local factors as well as the legitimacy conferred by the international community on divergent policy instruments. The analysis is based on detailed research of the case of Israel, part of which is presented here, as well as on a comparative outlook.

20.2 Knowledge, Translation and the International Policy Discourse

Up to the present, most studies have described the process by which a large number of countries established central banks within the relatively short period of time following the Second World War as a case of temporal clustered policy reform Elkins and Simmons 2005 accompanied by convergence. Two causal mechanisms have been suggested to explain this process. First, the symbolic mechanism assumes that newly established countries emulated industrial countries for the mere purpose of appearing to be like them Helleiner 2003; Marcussen 2005. Central banks, which were symbols of independence, replaced the currency boards, colonial institutions that symbolized foreign control and oppression Uche 1997. Another explanation suggests that developing countries established central banks because they sought monetary flexibility, a policy instrument that industrial countries possessed in the international monetary regime after World War II Helleiner 2003.

Both causal explanations suggest that developing countries established central banks according to the northern, that is, the Keynesian model. Differences between northern and southern central banks, if at all apparent, are to be explained therefore in terms of ‘decoupling’ Meyer et.al. 1997 between functional and effective central banks originating in the North on the one hand, and poorly emulated and distorted central banks adopted in the South on the other. Decoupling and corrupted emulation are explained as ‘biases’ due to political constraints and “limitations in the learning process” which led to “inefficiencies” Elkins and Simmons 2005, 45.

However, the divergence of central banking from the northern model in developing countries was not random as is implied by the notion of coupling. Rather, as we will see below, there was a pattern of regional convergence: northern countries converged around Keynesian central banking practices while southern countries converged around a unique type of central banking which I call developmental central banking. The existence of such a pattern of divergence implies that it was not a product of mere ‘decoupling,’ but rather a product of structural factors and coherent logic.

In order to identify these structural factors, it is necessary to abandon the ‘point of view of the center’ and to take the ‘point of view of the periphery.’ The ‘center,’ in this context is the place at which original practices and policies were produced and legitimized on the basis of standardized, codified and scientific knowledge. Taking ‘the point of view of the center’ is an epistemological position, which assumes that the policies justified by the institutionalized and standardized knowledge produced in the center are necessarily ‘best practices,’ irrespective of prevailing local conditions in the environment in which they are applied. Moreover, this approach assumes that the criteria used to evaluate these policies are universal and value free.

Taking the point of view of the periphery implies that observers historicize and endogenize the processes by which bodies of knowledge are produced, institutionalized, standardized, codified, transferred and translated. Such an approach underlines the epistemic dimension of the gap between the North and the South. According to our approach, peripheral countries, unlike core countries, do not possess the adequate resources to produce standardized, codified and institutionalized knowledge. Therefore, they lack the resources to justify and legitimize policies that deviate from internationally standardized policy models. ‘Taking the point of view of the periphery’ implies that observers refrain from evaluating the effectiveness of policies in the periphery on the basis of theoretical knowledge and criteria produced in the center. Instead they evaluate policies on the basis of outcomes and experimentation, while taking into consideration local goals, constraints, values and the uncertainty regarding the effectiveness of certain policies during the policy-making process.  

This approach enables us to discern between cases of decoupling and cases of translation Djelic and Sahlin-Andersson 2008. Translation is an intentional, calculated and institutional process of localization of knowledge and practices Acharya 2004. In such cases, the imported policy instrument is not corrupted in the process of transfer, but rather is modified on the basis of local knowledge, experience, goals, values and constraints.

In this chapter, I argue that the establishment of the Bank of Israel (BoI) represents such a case of translation and that the main impetus to establish the institution was not symbolic or based on the need for monetary flexibility, but rather the solution of the local problem of allocation of credit. The case of the BoI shows that local policy makers resisted the symbolic pressure to establish a central bank during the first years following the establishment of the state in 1948. During this period, monetary flexibility was achieved by the establishment of the issue department. The issue department was a small department within the largest commercial bank in Israel, Bank Leumi. It lacked any capacity to resist the government’s demands for loans. As long as it existed, the only limit to the government’s monetary flexibility was its own responsibility for and acknowledgment of the long-term cost of inflation. During this period the government deferred the idea of establishing a central bank by claiming that the time “is not ripe”² and that “objectively this is not the time.”³ In the early years of the state, the national priority dictated the maintenance of monetary flexibility, even at the expense of the advantages associated with joining the International Monetary Fund (IMF) and improving access to foreign aid.

What tipped the balance in favor of a central bank, I argue, was the government’s failure to solve a local problem in a closed policy domain: the problem associated with the allocation of credit. From 1950 onwards, the government made several attempts to employ selective credit controls. Up until 1953, most of these attempts had failed Bar-Yosef 1953; Bar-Yosef 1955; Bar-Yosef 1961. The supervisor of the banking system wrote, “the success of qualitative credit policies depends, first and foremost, on cooperation between credit institutions and the authorities” Bar-Yosef 1953, 188. However, the government did not have the necessary legal and administrative resources needed to realign the interests of the commercial banks with the national priority. In 1953, policy makers within the government identified the opportunity to use the central bank in order to control the banking system and the allocation of credit.

20.3 The International Discourse and Regional Convergence

The question arises as to what the conditions were that facilitated Israeli policy makers to effectively translate the industrial or Keynesian model of central banking and to use it in order to allocate credit. Effective translation of policy practices at the state level requires more than a devoted individual. Its success is dependent on structural and institutional factors. In this chapter I discuss two of these factors. Domestically, the recipient state has to have a minimal capacity of calculation and sufficient agency. Internationally, the likelihood of effective translation depends on the structure of the international policy discourse.

For a developing country to translate the Keynesian model of central banking and use it to solve locally defined problems, it had to have a minimal capacity for calculation and sufficient agency to act according to the results of the calculation. Such an approach assumes, as Bennett and Howlett put it, that state actors “can learn from their experiences and that they can modify their present actions on the basis of their interpretation of how previous actions have fared in the past” Bennett and Howlett 1992, 276. Or, as Hall formulates it, it implies that “elements within the state, acting, presumably, in pursuit of the national interest, decide what to do without serious opposition from external actors” Hall 1993, 26. Calculation capacities and agency are necessary conditions to creatively adopt policy instruments while altering them.

The literature of social learning has studied these aspects extensively. However, it has not given sufficient attention to the discursive international conditions that affect the capacity of countries to translate policies and knowledge. The likelihood that an effective translation takes place hinges, among other things, on the relative legitimacy conferred by the international community on the divergent policy instrument. If the international community harshly delegitimizes any deviation from standard models, the adoption of a translated policy instrument would entail high costs. For example, foreign investors would divert their investments to other countries, and banks would increase the interest rates for loans. Therefore states would not be inclined to translate. On the contrary, if the international community acknowledges the principle that peripheral economies require deviant policy instruments, the cost of deviation would be reduced and states would be more likely to translate.

The structure of the international policy discourse is the variable that captures the extent to which the international community legitimizes (or delegitimizes) deviant policy models. A homogenous discourse that legitimizes a very narrow range of policy models and strongly delegitimizes deviant models creates strong incentives to emulate, and punishes those countries that adopt deviant models. In cases where the international policy discourse is heterogeneous, deviating practices are not delegitimized and policy makers in developing countries are more likely to translate standard models.

In the post-World War II period, as I show below, the international policy discourse of central banking was heterogeneous in the sense that reputable experts acknowledged the utility of developmental practices of central banking in developing countries.

20.4 The International Policy Discourse of Central Banking

When policy makers in Israel, as in other new countries, faced the question of establishing a central bank in the post-World War II period, there were three alternative models of central banking to consider, each of which possessed a differing level of legitimacy in the international policy discourse: the traditional British model, the Keynesian model which was promoted by the newly established IMF and the International Bank for Reconstruction and Development (IBRD) and, as I argue, the deviant developmental model.

The traditional model of central banking was shaped during the gold standard in the heyday of colonialism and a surging international trade Gallarotti 1995; Eichengreen 1996. During that period, gold served as an international means of payment and central banks played a key role in stabilizing it by maintaining the convertibility between national currencies and gold within the member countries of the gold standard club Gallarotti 1995. The principle of convertibility of currencies to gold and vice-versa was the anchor that guaranteed the stability of the value of national currencies at home and the stability of the exchange rates in the international arena Capie et.al. 1994, 10. Despite the image of the gold standard as an automatic mechanism, central banks played a guiding role in sustaining it by managing the demand for money and the international flow of capital through the use of interest rate policy and through the buying and selling of short-term bills (operations in the free market) Sen 1952, 2.

During the interwar period, the Bank of London with the support of the League of Nations, the Bank of International Settlements and the Federal Reserves made attempts to restore the gold standard by globalizing it. It exerted pressure on the governments of developing countries to establish central banks that functioned according to the same principle as the British model in countries whose economies were radically different Drake 1989; Marcussen 2005. In hindsight and from a domestic point of view, these central banks failed Sen 1952.

In the colonial countries, imperial governments established currency boards, which extended the logic of gold standard to the periphery. Currency boards funtioned by converting domestic currencies to British currency, while keeping a 100% reserve ratio of British currency. Other imperial powers, like the US and France, enabled more discretion Schwartz 1993. The strict British arrangement prioritized the stability and confidence of the domestic currency over monetary flexibility Williamson 1995, 5–11.

After World War II the art of central banking went through a paradigmatic shift. The shift was a particular aspect of the emergence of the Keynesian paradigm Hall 1989. Keynes rejected the view that business cycles could not be tamed. He suggested that appropriate fiscal and monetary policies could significantly increase the stability of capitalist markets Barber 1990; Collins 1990. The new paradigm redefined the objective of central banks. Previously, the role of central banks had been defined very narrowly and technically as maintaining the stability of the value of money in relation to gold and to other currencies. After World War II the common conception was that central banks should pursue, in addition to stability, the objectives of growth and employment as well. The balance of power between central banks and treasuries shifted in favor of the latter. Monetary policy had to follow the lead of governments’ fiscal policies Cukierman et.al. 1992; Eichengreen 1996, 94, 188.⁴

The instruments of central banks were also redefined. The incentive to use new instruments came from below, from practitioners of central banking, rather than from above, from economists. Due to the emergence of the welfare state, the growing size of public sectors and high taxation levels, the traditional instruments by which central banks used to influence the demand for money—interest rate policies and operation in the free market—lost their effectiveness Sayers 1949; Sayers 1950; Sayers 1956; Brockie 1954; Miller 1956. Therefore, central banks turned to managing the supply of money. For this purpose a more stringent control of commercial banks was needed. In particular, the reserve ratio of commercial banks became a key policy instrument through which central banks controlled the supply of credit.⁵ The new monetary instrument blurred the distinction between macroeconomic policies, aimed at managing macroeconomic variables on the one hand, and prudential policies, aimed at maintaining the stability of the banking system on the other Capie et.al. 1994, 25.

In developing countries the situation was different. Indeed, the shift from traditional to Keynesian central banking favored less developed countries as it acknowledged the legitimacy of governments to steer the economy in order to achieve locally defined goals such as employment and growth. However, the Keynesian paradigm assumed that the market was still the best mechanism for the allocation of resources through the price mechanism. In many developing countries markets existed in an underdeveloped form, and their performance was not consistent with the national priorities as conceived by local policy makers Hunt 2002, 454. Specifically, governments in many developing countries sought to industrialize the economy in order to close the gap with the industrial world, or at least to minimize it. This goal required deeper intervention in the market and in the price mechanism. Keynesian economics as a rule, rejected this type of intervention.

In the financial domain, governments in developing countries employed various types of developmental instruments such as mobilizing savings, providing credit to developmental institutions and nurturing domestic public sectors. Central banks fulfilled a significant role in executing these policies Hirschman and Rosa 1949; Brimmer 1971; Blackman 1979. In particular, developing countries employed preferential credit policies. The objective of such policies was to restrict the overall volume of credit and at the same time to channel cheap credit (with low interest rates) to nationally strategic industries and specific industrial projects. These policies required central banks to manage differential and multiple effective interest rates in the economy: low interest rates for industrial purposes and high interest rates for all other purposes Wade 1990; Haggard et.al. 1993. The effectiveness of such policies was dependent to a large extent on the bureaucratic capacities of the state. The fact that developing countries employed preferential policies has already been documented in the literature. This chapter makes the argument that the establishment of central banks in developing countries increased the administrative capacity of governments to implement these policies. This argument explains the incentive of local policy makers to establish central banks in the post-World War II era.

What distinguishes developmental from Keynesian central banking, therefore, is the consistent, continual, pervasive and premeditated use of selective credit instruments. Indeed, central banks in industrial countries also regulated the commercial banks, and some of their instruments—such as foreign exchange control—had selective side-effects on specific branches. However, in principle, most central banks in the industrial countries aimed at managing the overall supply of credit and were reluctant to execute multiple interest rate policies.

It should be pointed out that the selective credit policy was not an invention of developing countries. The Federal Reserve System (FED), for example, used credit control in a negative way, that is, it restricted the flow of credit to specific branches. France used selective credit policies quite extensively Hirschman and Rosa 1949; Kriz 1951. It was left to developing countries “to carry this method of control to its conclusion.” Central banks in developing countries were granted the authority “to exercise a general control over the lending policies of the commercial banks, including the power to restrict the grant of loans for particular purposes” Sen 1952, 159.

20.5 The Worldwide Diffusion of Developmental Central Banking

When Israeli policy makers contemplated the idea of establishing a central bank, the British model of an independent and conservative central bank disintegrated and in addition to the orthodox Keynesian model, a new experimental model of developmental central banking began to emerge.

In Australia during World War II, selective practices were used. Initially, these were executed by a special governmental body, but later on, the Commonwealth Bank took charge of these policies. Similar measures were adopted by New Zealand, India and South Africa Sen 1952, 160–162. The Central Bank of Sri Lanka similarly employed these measures: its Annual Report for 1950 states, “there was every indication that bankers were co-operating with the Central Bank’s policy of restricting credit for non-essential purposes” Shuv 2003, 67. The State Bank of India was formed in 1955 with the explicit aim to allocate more resources to selected sectors of the economy to facilitate economic growth and development Capie et.al. 1994, 214. Similar measures were adopted by the Bank of Greece Capie et.al. 1994, 194.

The large number of developing countries that adopted similar policy instruments supports the claim that the deviation from the standard model was a case of translation rather than decoupling. Nevertheless, it suggests that the process of translation was not carried out on an individual basis—by each developing country in isolation—but rather that there were channels which coordinated the process: American experts who had heterogeneous ideas regarding central banking in developing countries and south-to-south transfer of knowledge.

20.5.1 American Experts

When the Israeli government took the decision to establish a central bank, it sought American advisors. Arthur Bloomfield was one of them. He was one of the experts to advise Israeli experts and politicians on how to formulate the central bank bill; he fulfilled a key role as a mediator between the international policy discourse and the Israeli experts. Bloomfield was a research economist at the Federal Reserve Bank of New York. In the late 1940s and early 1950s, he was sent to East Asian and Latin American countries to give advice on their central banks and financial systems Alacevich and Asso 2007. This type of job—money doctoring—situated Bloomfield in a strategic cross section of the processes of translation of policies. He played the role of mediator between the standardized principles of central banking and the unique conditions in developing countries.

As an American expert, Bloomfield was able to legitimize developmental central banking. In his missions Bloomfield insisted on taking into account the uniqueness of the economic condition of the countries in which he assisted the local authorities in establishing a central bank. He confronted other experts who, in his view, neglected the significance of such differences. The deviation of Bloomfield’s views from the orthodoxy was manifested in his exchange with the director of the research department at the IMF, Edward Bernstein. In the case of the Philippines, Bloomfield criticized Bernstein’s approach as being “too broad and general, and insufficiently oriented around Philippine problems.” He referred specifically to the lack of any reference by Bernstein to selective credit policies Bloomfield 1955.

Bloomfield publicly defended the notion that there was a theoretical justification for the use of alternative policy measures in developing countries, or at least that there was no theoretical foundation to the claim that such measures are ineffective. Central banks in developing countries, he wrote in an article, used measures that were “admittedly outside the traditional scope of central banking.” Moreover, the deviation of these instruments from traditional central banking did not render them ineffective. “Central banking in these countries should not necessarily be evaluated in terms of the standards and criteria applied in the more developed ones.” As the practices of central banking in developing countries had not emerged from a fully-fledged theory, Bloomfield characterized them as “experimental.” He expressed the hope that “out of this experimentation will develop a theory of central banking policy appropriate to the economically backward countries” Bloomfield 1957, 204.

Bloomfield was not unique in these views. An official report made by the Federal Reserve Bank of New York about central banking in developing countries recognized the utility of such measures, although it qualified their applicability by stating that their effectiveness is dependent on “the prestige and stature that the central bank enjoys within the financial community and the public at large” Fousek 1957, 78. Thus, although developmental central banking did not enjoy the international legitimacy of Keynesian central banking, it was not rejected out of hand by the international policy discourse of central banking. The existence of two legitimized models of central banking played an enabling role regarding the process of translation. Policy makers in developing countries were almost encouraged to treat the Keynesian model as a starting, rather than the end point, of the policy transfer process.  

Contrary to the underlying assumption of the flat view of globalization which assumes the prevalance of a homogeneous international policy discourse leading to convergence, it is assumed here that the international policy discourse, in certain policy domains, can be heterogeneous. It may consist of more than one legitimate policy model and therefore its diffusion is likely to lead to divergence and regional convergence rather than global convergence. This was the case in the policy domain of central banking in the post-World War II period.

A heterogeneous structure of the policy discourse is the outcome of internal debates and disagreements within the international epistemic community. When members of an epistemic community do not reach a consensus regarding the best practices in a policy domain, more than one policy instrument would be legitimized for use by different countries. Such was the case in the international policy discourse of central banking in the post-colonial period.

20.5.2 South-to-South Policy Transfer

The structure of the international policy discourse was not only the product of epistemic communities located in the core countries: policy makers and experts from peripheral countries also had the capacity to affect it. After World War II, developing countries joined the international community and participated in international organizations such as the IMF, the World Bank (WB or IBRD) and the United Nations (UN). The international organizations served as key channels for the south-to-south transfer of knowledge and policies. In such a process, peripheral countries, instead of looking up to industrial countries, “have the choice of looking to other developing nations for programmes” Rose 1991, 14. The structural similarities amongst peripheral countries increased the likelihood of an effective transfer of policies, despite the fact that the policy instruments did not enjoy the same legitimacy as those used by industrial countries.

The IMF and the World Bank were two nodal points in an international network that connected industrial and developing countries and they faciliated the exchange of knowledge. In this network, knowledge flowed in all directions, essentially, from north to south Barnett and Finnemore 2004. But the annual meetings of these organizations, however, also served as spaces in which ideas spread among developing countries who shared their experience with unique instruments such as preferential credit policies. As the governor of the BoI announced at the IMF’s annual meeting in 1960, “qualitative control of credit is an imperative in order to achieve selectivity in investment and a proper order of priorities with a view to improve as rapidly as possible the balance of payment of the country” Horowitz 1960. The U.N. Economic Commission for Latin America (CEPAL) established in 1948, and the U.N. Conference on Trade and Development (UNCTAD), gave rise to a translational epistemic community comprising local experts and policy makers who exchanged knowledge and ideas about developmental practices Sikkink 1991, 55.

20.5.3 Local Problem-Solving Through Translation

The structure of the international policy discourse and the legitimacy conferred on developmental central banking practices by American experts and international organizations played an enabling role in the process of translation. However, the positive incentive to establish central banks in developing countries, I argue, was the product of a common-domestic problem: credit allocation.

Prior to the establishment of the BoI, the government used its regulatory powers as laid down by the Banking Act in order to force commercial banks to cooperate with the selective credit policy. In addition, the government exerted informal pressure on the directors of large banks through the Banking Committee, a body comprised of the directors of large banks and a government representative Bar-Yosef 1953. However, until the establishment of the central bank, the government enjoyed only partial success in enforcing its will on the commercial banks.⁶

In 1953 the Minister of Agriculture managed to convince several commercial banks to cooperate with the government to create cheap long-term credit for agriculture. Commercial banks committed to channel 20% of their credit to agriculture with government collateral.⁷ His experience negotiating with commercial banks on credit allocation convinced the Minister of Agriculture that a better solution to the problem of credit had to be found.

The Minister of Agriculture, the main figure involved in negotiations with the commercial banks regarding the allocation of credit, was also the one who urged the government to hasten the process of establishing a central bank. “We have to reach a decision whether the thing is important or not” he insisted,⁸ emphasizing that a central bank would be “an instrument of credit control.”⁹

The government reached a decision to hasten the process of establishing a central bank. It was in this context that establishing a central bank turned out to be a solution to the local problem of allocation of credit as well as a response to the external soft pressure to establish a central bank. A delegation was sent to negotiate with the IMF regarding conditions for Israel’s membership. The IMF promised that other than “persuasion” no restrictions or pressures would be exerted.¹⁰ At the end of the year the government nominated David Horowitz as governor of the Bank, and, one year later in December 1954, the BoI was officially inaugurated.

20.6 Mutual Interdependence Between the Government and the BoI

The BoI was a hybrid of a Keynesian central bank with some modifications which endowed it with the capacity to function as a powerful instrument for allocating credit. Local policy makers and experts were convinced that this kind of instrument would serve the national interests and goals more effectively than a standard—Keynesian—central bank.

So far I have attempted to explain the considerations which, given the national goals and local economic conditions, led local policy makers to translate the central bank rather than adopt it. However, the question arises as to why policy makers could not solve the local problem of credit allocation with original institutional innovation. In other words: why it was cheaper to translate rather than to innovate? To answer this question we need a more detailed analysis of those elements of the central bank that were adopted and those that were translated.

According to the literature, the origin of central banking during the financial revolution in England was based on the institutional innovation of “credible commitment” Williamson 1975 or “commitment through delegation” Cukierman 1994. The underlying principle suggests that governments committed themselves to respect private property rights and to avoid confiscation (or writing off loans) by delegating power to an independent institution which functioned according to constitutional rules and professional standards Cukierman 1994. The commitment of the government increased the confidence of capital owners in the government and reduced the risk premium associated with loans to the sovereign. Therefore, the innovation of independent central banks was associated with the protection of nascent market economies from the unpredictability of sovereign power. The sovereign relinquished their authority to nationalize private property and in exchange gained the capacity to raise cheap credit. Sociologists and political scientists would describe this development—following Weber—as a particular case of the emergence of the modern apolitical bureaucracy Weber 1965.

The principle of credible commitment and the concept of central bank independence were products of the codification and standardization of central banking practices in industrial countries. It was formulated under the assumption that a differentiation between states and markets and the protection of the latter from the intervention of the former was a necessary condition for sustainable economic growth.

However, during late industrialization, the protection of markets from the state was not a sufficient condition for development Gerschenkron 1962; Hirschman 1981. A distinction has to be made between the late industrialization of Europe in the nineteenth century and the industrialization of peripheral countries in the mid-twentieth century. France and Germany employed different strategies to industrialize the economy. In France the state functioned as both actor and planner Kriz 1951, while in Germany the state provided the infrastructure for the emergence of an oligopolistic market-economy Henderson 1975. However, in both cases, developed legal and bureaucratic infrastructure existed in which economic development was embedded. Contrary to the European countries, in the developing peripheral countries, an infrastructure did not exist. Therefore, many of the new states that were established in the 1950s and the 1960s were in a deadlock: on the one hand, they did not possess markets and entrepreneurial forces to drive their economies forward; on the other hand, they did not have the institutional capacity to take charge of the economy and manage it with a top-down approach.

The strategies of rapid industrialization required unique state capacities as governments had to confront powerful societal actors as well as market forces Wade 1990; Weiss 1998. These strategies thus necessitated considerable administrative, bureaucratic and institutional resources Evans 1995; Woo-Cumings 1999. In particular, they required the capacity of the state to control and manage the financial sector in order to channel credit to preferential purposes Zysman 1984; Haggard et.al. 1993. Most of the new states lacked such capacities and resources, and the establishment of central banks provided an opportunity to acquire them.

The establishment of central banks, I argue, provided an opportunity for new states to fortify their administrative and legal infrastructures in the financial domain in order to better govern their economies. Central banks functioned as a stronghold of domestic power that provided legitimacy and administrative resources in order to pursue industrialization. As was the case in industrial countries, central banks enjoyed the status of professional and apolitical institutions which were relatively autonomous of both party politics and the socio-economic domestic dynamics. However, unlike the central banks in industrial countries, their purpose was not to protect markets from the state, but rather to serve as an instrument to confront societal actors and to govern the market.

Developing countries lacked the resources that were required for purely local institutional innovation. They lack the experience, the know-how and the skilled personnel necessary to solve local technical problems, as well as the legitimacy and the necessary legal infrastructure for its effective implementation. Imported models can solve both problems: they are accompanied by technical and administrative aid and training programs that enhance the administrative capacities of governments. Moreover, imported models are endowed with international legitimacy Graziadei 2006; Watson 1974. In short, imported models have the advantage of providing local authorities with the administrative and legal resources that contributed to the capacity of government to confront local actors.

The translation of standardized policy models and institutions therefore enables local policy makers to enjoy—to some extent—the best of both worlds: unlike simple adaptation, translation takes into account local knowledge and contingencies; unlike pure local innovation, they are administratively, politically and economically cheaper to implement.

This mechanism can explain the coupling between the institutional units of the central bank and the exercising of selective credit policies: it was more effective and cheaper for governments of developing countries to employ these policies through an apolitical, relatively autonomous institution, which enjoyed international legitimacy, than to carry out this policy themselves.

The case of the BoI demonstrates this point. Its establishment contributed to the capacity of the state in several ways. Prior to the establishment of the BoI, monetary and supervisory powers were scattered among a number of different governmental bodies. In addition, five bodies were involved in managing the issues that later on were managed by the bank. The issue of money had to be approved by the finance committee and by parliament. The technical aspects of issuing money were taken care of by a special department within Bank Leumi (the issuance department). The supervision and regulation of the banking system was dealt with by a special department within the Ministry of Finance. With the establishment of the BoI, all these issues and powers moved to the central bank.

From the point of view of the state’s capacity to govern, such a centralization of powers had several advantages. As Zysman points out, the centralization of power, “enables a single agency to exert influence across a range of issues without having to develop regulatory or administrative apparatus for each specific case.” Moreover, such “multipurpose policy tool is outside the direct control of the legislature” Zysman 1984, 77.

The authority of the BoI was derived largely from its role as mediator between the government and the two major international financial institutions, the IMF and the IBRD. International organizations exerted strong influence over policy-making on a state level Barnett and Finnemore 2004. The influence of the global organizations is based on the flow of expertise, knowledge and data to other countries Stone 2004, 554. Moreover, they had leverage on countries due to the financial aid programs Lal 2001. The power, the relative autonomy and the legitimacy of the international organizations were conferred institutionally and symbolically on central banks as representatives of the international organization in the public arena. Moreover, central banks also represented the country in its contacts with the international organizations. The mediatory role of the central bank contributed to its informal authority and legitimacy in the local arena.

These features were pertinent to the case of Israel. According to the Central Bank Bill, one of the BoI’s roles was to act on behalf of the government as a member of the IMF and the IBRD.¹¹ The BoI therefore maintained continuous contact with the IMF and the IBRD, which in turn supplied it with knowledge and expertise in various forms. The BoI’s governor also used IMF publications, which emphasised the main goal of the BoI—price stability—as an important condition of sustainable economic development in order to augment his own persuasive powers Horowitz 1975. The prestige of the international financial institutions was therefore conferred on the BoI and it increased its influence in the local arena.

The BoI’s persuasive power was further ameliorated by its capacity to generate economic data, knowledge, analyses and forecasts relating to the Israeli economy. This was necessary to identify problems, to formulate goals, to assemble policies and programs and to implement them, and in addition, it was essential to mobilize support for programs and to legitimize them. The research department of the BoI was the only source of empirical economic knowledge that encompassed the whole economy. Its Annual Report provided an essential source of information and analyses in the Israeli economic discourse. The department maintained contact with prestigious foreign economists and institutions, a practice that kept it up-to-date and contributed to its professional prestige Gross 2007, 181–185. The BoI’s capacity to produce empirically-based and up-to-date theoretical knowledge was an extremely significant factor in its ability to present as non-political its analyses, recommendations and policies.

Due to its autonomous budget and unique terms of employment the BoI also provided an opportunity to improve the meritocratic recruitment of personnel. The budget was managed by the bank itself, as was the employees’ pay scale. The budgetary independence of a central bank is an index of its actual independence Cukierman et.al. 1992, 366, table 4. As it offered employment conditions that were unprecedented in the public sector, the BoI managed to attract high-quality staff, including the leading graduates from the Hebrew University’s Economic Department. This was one of Weber’s conditions for effective state power and autonomy (Evans and Rauch 1999, 751, Weber 2009, 241). As a result, the BoI nurtured an autonomous community of economic experts speaking one language, a phenomenon that enhanced its image of objectivity and professionalism.

With the establishment of the BoI, many powers that previously had been dispersed among several other authorities and institutions were transferred to the bank. The centralization of power increased its “coherence and corporate identity,” which in turn increased its capacity Evans 1989, 573. As Johnson points out in his classic study about the Japanese developmental state, its capacity relied to a large extent on the existence of an informal network of experts who graduated from elite universities in Japan Johnson 1982, 57–59. In the case of Israel, a young state, a network of elites did not exist and the BoI contributed to the consolidation of such a network.

The institutional unit of the central bank was thus imported and it provided further capacities for local policy makers. However, while in the industrial world the autonomy of central banks restricted the legitimacy of the government to manage the economy, in developing countries the apolitical status of the central bank and its professional status were used in order to control, restrain and mobilize local societal actors.

20.7 Conclusions

In this chapter I have discussed the epistemic and institutional aspects of the translation of central banking to peripheral countries in the post-World War II period. The analysis suggests a few directions for generalization and for further research.

The post-World War II period was a unique period in terms of cross-country transfer of economic knowledge and policies. During this period the international community became more heterogeneous than in any other period since the emergence of the international system in the seventeenth century. From the four nations that constituted the Concert of Europe in the nineteenth century and around sixty members who were involved in the League of Nations in the interwar period, the number of member nations in the UN reached 140 in the 1970s. The majority of the new countries did not have developed markets or industrial infrastructures and their socio-economic structures differed radically from those in the industrial world. While those countries possessed nominal sovereignty and were equal members in the international community, as far as their structure was concerned, they were essentially different. Therefore, it is justifiable to argue that the community of nations became highly heterogeneous from a structural point of view.

The heterogeneity of the international system posed an unprecedented challenge to macroeconomics, a body of knowledge whose primary object was national economies: on the one hand, as a standardized, codified and theoretical body of knowledge the macroeconomic discourse assumed—and it had to assume—a certain level of “homogeneity of nature,” that is, a “spatial and temporal permanence of general laws” Desrosières 1998, 282. It had to assume that despite differences between national economies, there are essential structural similarities that made comparisons justifiable. One fundamental similarity was the epistemic condition of macroeconomics, just as the epistemic condition of biology is that all individuals within a species are essentially similar; on the other hand, empirically, the institutional, social and economic differences between national economies were enormous.

The challenge was aggravated by the fact that it had not only theoretical but also practical implications. The policies that were formulated on the basis of macroeconomic knowledge affected international politics and the lives of millions. It is beyond the scope of this contribution to discuss the ways in which macroeconomics dealt with this challenge. However, we can draw several generalizations on the basis of the policy domain of central banking.

During the interwar period, the Bank of England, with the support of the League of Nations, the Federal Reserve and the Bank for International Settlements, made a concerted effort to export the British model of central banking to Latin American countries Drake 1989. The mission of globalizing the gold standard and central banking practices was obviously driven by UK interests. The UK, which based its power on international trade, had a clear interest in the maintenance of a stable international means of payment. However, the drive to restore the gold standard was also the product of a stiff policy paradigm that prevailed amongst UK civil servants. The British bureaucracy was very closed, hierarchical and immune to change Weir 1989. Moreover, until the 1930s, there was no alternative policy paradigm to the one promoted by British policy makers. It took some time until the Keynesian alternative permeated the inner circle of the British state apparatus Weir and Skocpol 1985; Weir 1989.

Similarly, in the neoliberal period during the 1980s and 1990s, national interests were combined with epistemic factors. In the 1980s after the oil price crisis and the international debt crises, the IMF began to implement the Structural Adjustment Programs which conditioned financial support by domestic reforms Goldstein 2001; Drazen 2002; Khan and Sharma 2003; Akonor 2006. Among other requirements, developing countries had to reform their central banks’ bills and upgrade their independence vis-à-vis governments Polillo and Guillén 2005.

The ambition to coerce central banking reforms was supported by newly formulated theories of central bank impendence. Academic economists demonstrated, on the basis of the principle of rational expectation, that high central bank independence was essential to curb inflation Barro 1976; Barro and Gordon 1984. The theory was supported by evidence regarding the correlation between a high level of independence and low inflation and growth Berger et.al. 2001. This conception dominated the international policy discourse throughout the 1990s and the 2000s.

The Bretton Woods period was unique in the sense that the combination of geopolitical constellation, on the one hand, and the heterogeneous structure of the policy discourse on the other, played a permissible role in regard to acts of translation. One could not escape the question, though, of whether the structure of the international policy discourse was not only an epiphenomenon of the geopolitical constellation. Helleiner, for example, explains that:

The sympathy that U.S. officials exhibited towards the nationalist monetary goals of Southern governments often reflected their desire not to alienate key allies in the context of the Second World War and then the Cold War. It also helped them to gain influence in newly independent Southern Countries, particularity ex-British colonies. Helleiner 2003, 268

It is undeniable that the geopolitical conditions and the national interest of world leaders influenced, to a certain extent, the openness of experts to the idea that deviant models were effective in a heterogeneous world. However, the transmission mechanism between national interests and the international policy discourse is not that simple. If it were, we would expect that during the post-World War II period the dominant view among American experts would support deviant models of central banking in developing countries. In practice, this was not the case. Rather, the American discourse of central banking was characterized by heterogeneity, that is, by internal disagreements and debates among experts. The heterogeneity of the discourse cannot be explained on the basis of political influence. Therefore, we may conclude that the view of experts was not completely determined by the national interests of the country they served and that there was uncertainty among experts regarding the best practices that suited developing countries.

According to this line of reasoning, what was unique in the policy discourse during the Bretton Woods period was not only the fact that it consisted of Keynesian ideas, rather than, for example, laissez-faire or liberal ideas. No less important was the fact that the policy paradigm was flexible and heterogeneous enough to embrace a wide range of policy instruments. Keynesian ideas were interpreted in various and very different ways. Some, like the neo-Keynesians, concluded from Keynesian economics very conservative policies, while others interpreted Keynesian economics as conforming to the principle of social democracy Hunt 2002. Keynes, some argue, was even a source of inspiration to developmental economics Toye 2006. The heterogeneity of the discourse, I argue, affected the capacity of local policy makers to translate and localize economic knowledge and policies.

It has already been established by various authors that globalization processes do not lead to a demise of the state Weiss 1998; Sassen 1999; Polillo and Guillén 2005 nor to global convergence Radaelli 2005. However, there is still much to be studied regarding the transmission mechanisms that translate global pressures to domestic actions. Bodies of knowledge, and particularly bodies of policy-relevant bodies of knowledge play an essential and significant part in these mechanisms. The international policy discourse and its structure play a crucial role in such a transmission mechanism.

Acknowledgments

I would like to thank Stephen Bell and Simone Polillo for reading previous versions of this paper. I am also grateful to the members of the reading group Transfer and localizaton of knowledge at the Max Planck Institute for the History of Science, Malcolm Hyman, Anna Perlina, Albert Presas i Puig, Jürgen Renn, Margareta Tillberg and Milena Wazeck, for their comments.

References

Footnotes