classical developmentalism
'Agriculture workers on a strawberry farm', Argentina. (Credit: Nahuel Berger/World Bank/Flickr, licensed under CC BY-NC-ND 2.0, original cropped) (via:

Between the 1940s and the 1960s, classical developmentalism – or development economics – made a major contribution to economics, as it defined economic development as structural change, or industrialisation, and advocated moderate intervention from the state. Since the 1970s, classical developmentalism has been in crisis, while from the 1980s, neoclassical economics has returned to the mainstream. At the end of the 1980s, Latin American countries engaged in neoliberal reforms; in the 2000s, some tried to return to classical developmentalism, but neither school of thought was able to provide policies that would cause growth – which had effectively stopped in 1980 – to resume. In the early 2000s, as this failure became clear, a new theory, new developmentalism, began to be established, with a new developmental macroeconomics and a new political economy. The paper compares new developmentalism – which builds on post-Keynesian macroeconomics and classical developmentalism – with classical developmentalism, and in the last section, presents a summary of policies that should be maintained and policies that should be changed in accordance with new developmentalism.[1]

In the framework of the 1930s Great Depression, the crisis of economic liberalism and the rise of the Keynesian revolution, economic development became a discipline in its own right in 1940s and 1950s with the contributions of economists like Rosenstein-Rodan, Raúl Prebisch, Arthur Lewis, and Celso Furtado. In the North, it received the name  ‘development economics’; and in Latin America, ‘structuralism’; but I prefer to call it ‘classical developmentalism’, following a suggestion from Ricardo Bielschowsky.[2] Classical developmentalism is a good name because the expression ‘developmentalism’ also applies to a real historical phenomenon: a form of economic and political organisation of capitalism that differs from economic liberalism. This has characterised industrial revolutions in every country that has experienced one.[3] Since the mid-2000s, in the framework of the return of neoclassical economics to the mainstream which had occurred twenty-five years before, and of the neoliberal reforms that have been causing low growth, financial instability, and increasing inequality, a ‘new developmentalism’ has been emerging from the contributions of Latin American economists like Roberto Frenkel, José Luis Oreiro, Martin Rapetti, Nelson Marconi, and myself.

Download this expert comment in PDF or continue reading online

The rise of a new developmental macroeconomics for middle-income countries: From classical to new developmentalism

Download expert comment

Classical developmentalism and new developmentalism are theories that seek to understand the real phenomenon of economic growth. Historically, this was a fast process leading to catch-up through a developmental form of organising capitalism. Classical and new developmentalism are theories that adopt a historical approach so it is natural that their historical reference point is the form of capitalism assumed during periods of faster growth, when the difference in income per person in relation to rich countries has been reduced.

Main contributions of classical developmentalism

Classical developmentalism has made major contributions to economic theory. First, in the form of Keynesian macroeconomics, classical developmentalism was based on a critique of neoclassical economics.[4] The main contribution of classical developmentalism to economics was to affirm that economic growth equates to industrialisation or ‘structural change’ or, as I prefer to say today, it equates to productive sophistication. This is explained with a series of arguments:

First, industrialisation has been the historical condition for all countries to have developed.

Second, in the process of growth or of increasing productivity, the transference of labour from agriculture to manufacturing plays a key role.

Third, as Raúl Prebisch (1949) and Hans Singer (1950) noted, increasing productivity in rich country manufacturing industries is not fully transmitted into a fall in prices which also benefit countries not producing manufactured goods, as neoclassical economics assumes, but causes a direct increase in wages in rich countries. Prebisch and Singer argued that while workers in these countries are organised and able to retain their productivity gains, workers in the primary sectors of developing countries are not, and this results in a tendency towards deteriorating terms of trade in developing countries.

Fourth, developing countries exporting primary goods confront a ‘foreign constraint’ that originates from two perverse income-elasticities: while the elasticity-income of imports of developing countries that export primary goods is higher than one, the import-elasticity of imports of primary goods by rich countries is smaller than one. Economists have derived simple facts from this, so clearly defined by Raúl Prebisch:  Firstly, the ‘two-gap model’, highlighting savings gap and the dollar gap, with the conclusion that developing countries should grow with foreign indebtedness using ‘foreign savings’: Secondly, the understanding of the foreign constraint as a ‘structural source of foreign vulnerability’ leading developing countries almost inevitably to cyclical crises. However, the only legitimate conclusion that we can derive from this constraint is that developing countries must industrialise to overcome this growth disadvantage. Developing countries certainly face a shortages of dollars, but the way out should not be to become indebted with foreign money; they certainly undergo cyclical financial crises, but, as we will discuss, the cause is not the foreign constraint but the mistaken belief that capital is not made at home.

Classical developmentalism didn’t calculate using macroeconomics. Its ‘structural theory of inflation’ had limited scope because countries are able to overcome bottlenecks in the productive system and production becomes responsive to prices. In practice, classical developmentalism adopted post-Keynesian macroeconomics: essentially the idea that fiscal policy must be countercyclical.

Classical developmentalism defends moderate but strategic state intervention in the economy, not only because there are non-competitive sectors in the national economies of even rich countries, but also because savings are insufficient and markets in pre-industrial countries are poorly developed, poorly regulated, and not sufficiently ensured by the state.

The basic development strategy adopted by classical developmentalism came to be called the import substitution strategy. From the assumption that growth means industrialisation, it involved setting high import tariffs on manufactured goods, justified by the infant industry argument. Given the limits of domestic markets, and the big economies of scale, classical developmentalism acknowledged that the scope of the import substitution strategy is small unless the country is large. Nevertheless, development economists did not consider the alternative of exporting manufactured goods. Instead, they proposed regional integration to increase the size of domestic markets. As countries maintained high import tariffs on manufactured goods after one could reasonably consider the manufacturing industry ‘infant’, liberal-orthodoxy simply saw ‘protectionism’ in this practice. Yet, when we discuss new developmentalism, we will see that such tariffs – and therefore the import substitution strategy too – were not protectionist, but just an intuitive way of neutralising the Dutch disease on the domestic side, and thus a way of compensating for the competitive disadvantage represented by the disease, and assuring the national manufacturing industry of equal conditions in competition.

The crisis of classical developmentalism

Classical developmentalism, which was the mainstream economic theory in Latin America from the 1940s to the 1970s, came to a first crisis around 1970, when dependency theory became the dominant interpretation of economic development in Latin America and the United States, and to a second crisis ten years later, when neoclassical economics became mainstream again in the North.

Dependency theory was a Marxist interpretation of economic development on the periphery of capitalism, defined in the 1960s. Its distinctive thesis was the impossibility of a national bourgeoisie commanding a developmental class coalition, confronting imperialism, and achieving a national and capitalist revolution. Rich countries had counted on national bourgeoisies to realise their industrial and capitalist revolution, but for dependency theory the industrial bourgeoisies in developing countries would be intrinsically ‘dependent’ instead of ‘nationalist’. It was therefore a direct critique of classical developmentalism’s central political-economic proposal: the formation of a national-developmental class coalition.

Dependency theory was founded by André Gunder Frank with the paper ‘The development of underdevelopment’. He has been in Brazil at the time and wrote the paper just after the 1964 military coup, which had counted on the support of industrial business entrepreneurs. This coup, just as those that followed in Argentina (1967) and Uruguay (1978), was understood by Latin-American intellectuals of the left as a confirmation of dependency theory; of the ‘impossibility’ of national bourgeoisies in developing countries. Soon dependency theory was divided into two currents: one which remained Marxist, of Gunder-Frank and Ruy Mauro Marini; and the ‘associated dependency’ current, which was founded by Fernando Henrique Cardoso and Enzo Faletto’s 1969 book, Dependency and Development in Latin America, which defended the association of Latin American countries with the United States.

At a time when Marxism had become very influential, the associated dependency interpretation had a Marxist penchant because it worked with social classes, but it effectively defended not the resistance to American empire, as classical developmentalism did, but the association or subordination to it. This view, which soon became dominant in Latin America among Latin-American intellectuals – most of whom, including myself, not realising the dependent character of associated dependency – reflected the frustration of the Latin American left with military coups and a critique of classical developmentalism’s basic political thesis: that real or fast economic development on the periphery of capitalism depends on the formation of a developmental class coalition. The new interpretation saw multinational corporations investing in Latin American manufacturing industries as proof that the centre-periphery opposition was false. Not surprisingly, it was received with joy by the American academy, as Cardoso himself acknowledged (1977).

Dependency theory severely hit classical developmentalism, but the crucial blow it suffered was the major structural change that happened around 1980 in the central countries of the global economy – the change from post-war developmental and social-democratic capitalism to a neoliberal or rentier-financier capitalism and from Keynesian to neoclassical economics as mainstream economic theory. Several factors contributed to a change in the mainstream:

First, the permanent attraction that economists feel for the mathematical reasoning used by the hypothetico-deductive method adopted by neoclassical economics.

Second, the several theoretical innovations that strengthened neoclassical economics: Solow’s growth model in the 1950s; Milton Friedman with monetarist macroeconomics, and James Buchanan with the public choice school, both in the 1960s; Robert Lucas, with rational expectations macroeconomics, in the 1970s; and, Paul Romer with endogenous growth models in the 1980s.

Third, the new reality and the new ideas were a reaction against falling profit and growth rates in the 1980s in the United States and the United Kingdom, while the inflation rate rose (‘stagflation’) due to the increasing power of the unions in the 1960s and the first oil shock in1973. The rise of neoliberalism and neoclassical economics’ return to the mainstream were the response to this crisis.

Forth, they were also a reaction against the new competition originating from developing countries which first began to export manufactured goods to rich countries in the 1970s, benefiting from their low wages. What happened was change of ‘policy regime’ or, more broadly, the ‘form of economic and political organisation of capitalism’ from a developmental and social form to a liberal form.[5] This change was profound and soon encompassed all rich countries, irrespective of whether their governments were conservative or social-democratic. This made neoliberal ideology dominant in Western societies and neoclassical economics dominant in universities and financial markets. The United States asked the World Bank, which was the main think tank for classical developmentalism, to be the international agency that would implement neoliberal – or ‘Washington Consensus’ – reforms in developing countries. At the same time, powered by trade and financial liberalisation, by reduced transport and communication costs, and by the growth of multinational corporations and their foreign direct investment, globalisation became the material expression of neoliberalism, which profited from the policy difficulties that social-democratic and Keynesian policies were facing. Economics departments of major universities, using neoclassical economics, became an ideological instrument of neoliberalism at the service of a small elite of rentier capitalists and financiers, i.e., of a rentier-financier class coalition.[6] In this new world, classical developmentalism was reduced to “protectionist advocacy of the import substitution strategy”. Albert Hirschman, one of the pioneers of classical developmentalism, acknowledged its crisis in ‘The rise and decline of development economics’ (1981).

On the other hand, in Latin America, growth had been stopped in the 1980s by a major financial crisis, the Foreign Debt Crisis – caused by the policy of growth through foreign indebtedness that was adopted in the 1970s – and its two major consequences: all countries falling into fiscal crisis and several experiencing high inflation. Given the economic stagnation and the ideological hegemony of neoliberalism, it was not difficult for liberal orthodoxy to attribute the crisis to the import substitution strategy, i.e., to state intervention and the protection of the manufacturing industry.

This contention was essentially false. It ignored the fact that reforms adopted – trade liberalisation and financial liberalisation – at the end of the 1980s by Latin American countries had the unpredicted consequence of dismantling of the pragmatic mechanisms which had, respectively, neutralised the Dutch disease in relation to the domestic markets and kept real interest rates low. Yet the idea seemed true because states in the region experienced a fiscal crisis, which the liberal orthodoxy simply explained with fiscal populism, but which had in fact also resulted from the Foreign Debt Crisis obliging the state to bail out the private and state-owned corporations, which were indebted in dollars.

Among the shortcomings of the neoliberal diagnosis of what was happening with developing countries, one was particularly relevant. The East Asian countries had continued to grow fast throughout the 1980s, and, they were definitely not examples of laissez faire policies, but examples of the developmental form of capitalism’s economic organisation. They followed the Japanese model of growth, which meant strong developmentalism. They adopted a nationalist approach and very active industrial policies. Nevertheless, the liberal orthodoxy tried to argue that the East Asian growth strategy was liberal, because it soon became export oriented. It is true that these economies became exporters of manufactured goods, but this does not mean that they had become liberal. They simply realised that the import substitution strategy was not a real alternative for them, not only because their domestic markets were small, but mostly because they didn’t have abundant natural resources from which to source commodities exports. Indeed, the four Asian tigers abandoned the import substitution strategy in the 1960s and rapidly became exporters of manufactured goods, while they gradually opened their economies. But at the height of their growth process, they kept firm control of macroeconomic prices and adopted capable industrial policies.

New developmentalism emerges

In the 1980s, developmental governments in Latin America failed to overcome the Foreign Debt Crisis, and in the 1990s the countries debilitated by the crisis bowed to the new ‘truth’ coming from the North. Countries engaged not only in required structural adjustment policies led by the IMF, but also in neoliberal reforms coordinated by the World Bank, whose validity was questionable.

Not surprisingly, the reforms were adopted but the countries failed to resume growth. Instead, there was deterioration: increased financial instability, low growth rates, and deepening inequality. On the other hand, in the 1980s and again in the 2000s, classical developmentalism was unable to achieve better results. This meant that the main schools of thought at the disposal of developing countries were proving powerless by offering inadequate guidance. This was true in relation to the two competing historical theories of economic development (classical developmentalism and neoclassical institutionalism), as well as in relation to the two competing macroeconomic theories: post-Keynesian macroeconomics and neoclassical macroeconomics.

It was in this context, one in which economic theories had no reasonable development proposals, whether developmental or liberal, centre-left or centre-right, that it became clear a new theory was needed: a new developmental macroeconomics.

Post-Keynesian economics may also be viewed as developmental macroeconomics, but its more celebrated model, Thirlwall’s law (1979), is nothing more than a formalisation of Prebisch’s two perverse elasticities. The model attracted post-Keynesians because it made development dependent on demand – essentially on exports. This assumption is only partially true but let us accept it. Nevertheless, the formalisation proved limited in explanatory value, and rich in wrong interpretations and mistaken policies. The formalisation allowed for an infinite number of econometric studies confirming the obvious: that the constraint really exists, i.e., that the growth of a country is limited by exports of commodities whose demand tends to grow at a slower pace than the increases in the country’s demand for imports. But the only legitimate conclusion that we can derive from this is that the country must industrialise in ordert to rid itself of this constraint, which will require an extra effort on the part of the country. Instead, Thirlwall and Hussain (1982, p. 1) sought to predict developing country growth rates from the income elasticity of the imports of each country, with poor results.

New developmentalism was a response to all these problems. It is a theoretical framework that explains both growth and growth failure in developing countries, particularly Latin American middle-income countries that suffer from the Dutch disease and from dependency in relation to the North. As a framework it was essentially macroeconomic, not because the supply side does not matter, but because, in terms of the supply side, developing countries – except when they have predator states, which are out of the scope of this work – are already involved in doing ‘their best’. They really are involved in developing education and healthcare, in building the best institutions, in investing in infrastructure, and in promoting science and technology. The correspondent outcomes only occur in the long-term, while the right macroeconomic policies produce almost immediate results.

It was in this context that a growing group of economists, mostly in Brazil and Argentina, started building a new development macroeconomics which, eventually, came to be called new developmentalism. The first attempt in this direction occurred in the 1980s: an initial step was to build a macroeconomics more adapted to developing countries with the theory of inertial inflation – a theory that is crucial to the understanding and control of high inflation, and today is part of the mainstream. This theory had Mario Henrique Simonsen (1970) and Felipe Pazos (1972) as pioneers and achieved its first complete formulation in Bresser-Pereira and Nakano (1983), who distinguished the accelerating, maintaining, and sanctioning factors of inflation, supply or demand shocks, and the distributive conflict responding for the accelerating factors, the formal and informal indexation of the economy for the maintaining factors, and the endogenous character of money, for the sanctioning factor.

After Argentina’s 2001 crisis, high inflation had been controlled in Latin America, and the problem now was resuming growth, which had stopped twenty years earlier. Considering the Fernando Henrique Cardoso administration (1995-2002) in Brazil, which formally adopted the growth-cum-foreign savings policy, but didn’t succeed in resuming growth – despite the fact the Foreign Debt Crisis and the high inertial inflation had just been resolved – I realised that this policy was essentially flawed; I realised that there was an inverse relationship between the current-account deficit of a country and the exchange rate. The higher the current-account deficit of a country, the more the national currency would appreciate. Thus, when President Cardoso decided to grow through foreign indebtedness, he implicitly decided that the currency would appreciate in the long-term, thus encouraging consumption and discouraging investment. I had written a short note on the subject in 1991 and a full paper the following year with Yoshiaki Nakano, ‘Economic growth with foreign savings?’.[7] In the same year, again with Nakano, I wrote a paper on the Brazilian economy where for the first time we made a severe critique of the high interest rates practiced by the Central Bank of Brazil, which opened room for the first serious public debate on the subject.[8]

In 2003, I used the expression ‘new developmentalism’ (Bresser-Pereira, 2003b) for the first time. I used this name not because some in Latin America were adopting developmental policies again after the obvious failure of the 1990s’ neoliberal reforms, which derived from the Washington Consensus. Indeed, several countries adopted economic policies based on a developmental approach, but the policies were a combination of classical developmentalism and economic populism in its two versions: fiscal populism and exchange rate populism. New developmentalism was so called in order to underline its theoretical difference in relation to classical developmentalism and its rejection of populist or vulgar developmentalism.

In 2006, I published the paper ‘New developmentalism and conventional orthodoxy’, which garnered some general interest, but principally the interest of political scientists who understood it as a generalisation of a form of policymaking, rather than as a theory: the policies that Lula in Brazil and the Kirchners in Argentina were practicing. I probably was not clear enough on that matter and created confusion. Instead of opposing new developmentalism to classical developmentalism, I compared it with ‘old’ developmentalism, expressing the latter negatively. This was not helpful: I put together the economists that I view as my masters – the economists with whom I had learned economic development like Celso Furtado and Arthur Lewis – with the populist practices that had plagued really existing developmentalism in the 1980s and were present again in the 2000s. The new thought gained ground in the debate and led to the approval of the ‘Ten Theses on New Developmentalism’ (2010), which was discussed and signed by a group of 81 academics, mainly economists.[9]

The theoretical innovations

In the following years, new developmentalism continued to be built step by step, and its distinction in relation to classical developmentalism became increasingly clear.[10] It is in obvious opposition to neoclassical economics and liberal orthodoxy. As to classical developmentalism, it is more of an addition than a substitution. New developmentalism’s main theoretical innovations evolved in the following sequence:

(a) From 2001-2006, the model rejecting growth through foreign savings as the additional capital inflows cause the national currency to appreciate, encourage consumption, discourage investment, and this results in a high rate of substitution of foreign for domestic savings;

(b) Between 2007 and 2008, the model of the Dutch disease, including definitions of the current and the industrial equilibriums, the disease’s neutralisation through an export-tax on the commodities that cause the disease, and the consequent current account surplus;

(c) In 2008, the model of the tendency towards cyclical and chronic exchange rate overvaluation, which shows that (c1) the exchange rate is not just volatile but that such volatility has sense, (c2) exchange rate overvaluation happens between financial crises, which cause sharp depreciation and are mainly caused by the policy of growth through foreign indebtedness, (c3) in between crises, the exchange rate remains overvalued for several years, (c4) and as a consequence, companies don’t invest in manufacturing, which explains why the exchange rate is a determinant of the expected profit rate and therefore of the investment rate too, thus becoming a key variable in the growth process of developing countries;

(d) In the early 2000s, the idea that in order to grow the country must ensure that the five macroeconomic prices (the interest rate, the exchange rate, the wage rate, the profit rate, and the inflation rate) are kept right, which the market definitely does not guarantee;

(e) The realisation that the right macroeconomic prices were essential for the catching up of the East Asian countries, and the endorsement of industrial policy provided that it is not understood as a substitute but as a complement to a competent macroeconomic policy;

(f) In 2013, the concept of value of the exchange rate, around which the exchange rate floats according the demand and supply of foreign money, which varies according to several facts, including variations in the terms of trade and in capital flows;

(g) In 2015, the model explaining the value of the foreign currency with the variations in the index of the unit labour cost of the country in relation to its main competitors; and the variations of the current equilibrium with, additionally, the variation in the terms of trade;

(h) In 2016, the completion of the model of exchange rate determination, where the structural component of such determination is the value of foreign money, and where, besides other variables, the demand and supply of foreign money vary according to three habitual policies often adopted by developing countries: the growth through foreign savings policy and the resulting capital inflows permanently in excess of outflows; the policy of using the exchange rate as an anchor against inflation; and the policy of high interest rates, which attracts capital inflows and is instrumental in relation to the two previous policies.

The microeconomic innovations are more limited. New developmentalism borrowed the labour value theory and the tendency towards parity of profit rates from classical political economy. And it borrowed the definition of growth as industrialisation and the defence of industrial policy from classical developmentalism, although, in the case of the latter, with less emphasis. New developmentalism never starts reasoning from the general equilibrium model or pure competition, as it assumes competitive or relatively free markets, but it distinguishes a competitive and a non-competitive sector within modern capitalist economies and it defends economic planning and strict regulation for the non-competitive sector, which is formed by infrastructure, basic inputs companies, and large, ‘too-big-to-fail’ banks.

New developmentalism is also comprised of a particular view of political economy, understood as the relations between the market, the state, and politics. The components of this political economy were also developed gradually, out of much debate and reasoning. Some of them were already part of classical developmentalism, but are important in the new developmental framework:

(a) The identification of the beginning of economic development with the formation of the nation-state and the industrial revolution, with both these two major historical changes forming the capitalist revolution in each country;

(b) The distinction of economic from political populism, and the identification of economic populism not only as fiscal – the state irresponsibly spending more than it gets – but also, if not principally, exchange rate populism: the nation-state expending more than it gets and incurring current-account deficits;

(c) The affirmation of the possibility of national developmental class coalitions, accounting however for the ambiguous and contradictory character of Latin American business entrepreneurs;

(d) The definition of the developmental state as a state that intervenes moderately in the economy and practices industrial policy, and, although cooperating with other countries, adopts economic nationalism.

To this previous knowledge on political economy we must add other new developmental components, which may also be depicted in the sequence in which they have been elaborated:

(a) From 2006 to 2009, the definition of globalisation as a competition, not only among companies, but among nation-states: This induces an imperial practice on the part of the richer and more powerful countries, which explains why developing countries must resort to economic nationalism in order to grow;

(b) From 2010 to 2014, the more precise definition of the developmental state, which is not just characterised by economic nationalism and moderate state intervention in the economy, but also, if not principally, by an active macroeconomic policy which keeps the five macroeconomic prices right, particularly through exchange rate policy;

(c) In 2014, the classification of the developmental state into four models, according to whether they are central or peripheral countries and according to their degree of autonomy: the original central model, of England and France; the latecomer central model of Germany and the United States; the independent peripheral model, of East Asia; and the national-dependent peripheral model, of Brazil and South Africa;

(d) In 2015-2016, the definition of developmentalism as the alternative form of capitalism’s economic and political organisation to economic liberalism, and the definition of the phases of capitalism in its original countries as: mercantilism or first developmentalism; economic liberalism; the Golden Years of Capitalism or second developmentalism; and neoliberalism;

(e) In the same period, the definition of developmentalism as the default form of capitalism, to the extent that not only the original central countries, but all other countries have been developmental as they have industrialised;

(f) In 2017, the definition of contemporary capitalism as rentier-financier capitalism, and of the phases of capitalism according to dominant class coalitions: classical or business entrepreneurs’ capitalism; techno-bureaucratic capitalism, where techno-bureaucrats replace entrepreneurs in the management of corporations; and rentier-financier capitalism, when the heirs and speculators replace entrepreneurs in the ownership of corporations, while the financiers manage their wealth and play the role of organic intellectuals.

Considering these new contributions, we can compare classical and new developmentalism:

  • Classical developmentalism’s main objects are pre-industrial countries but for new developmentalism they are middle-income countries, which have already been through their industrial and capitalist revolution;
  • Classical developmentalism didn’t calculate with macroeconomics and reproduced post-Keynesian macroeconomics, while new developmentalism calculates with its own macroeconomics, although this is firmly based on post-Keynesian macroeconomics;[11]
  • Classical developmentalism was based on the thesis of the infant industry and defended an import substitution strategy, while new developmentalism assumes that middle-income countries are able to and should export manufactured goods;[12]
  • Classical developmentalism defended protectionism, while new developmentalism essentially demands the levelling of the playing field for the manufacturing industry – something that the market does not guarantee;
  • Classical developmentalism defended an overvalued currency and high import taxes, while new developmentalism defends relatively open markets and a right or competitive exchange rate which is only achievable with a low interest rate, and in countries exporting commodities, with a variable export tax on such commodities to neutralise the Dutch disease;
  • Classical developmentalism defended the growth through foreign indebtedness policy, while new developmentalism rejects it and defends balanced or, when the country faces the Dutch disease, surplus current accounts;[13]
  • Classical developmentalism defended the import substitution strategy, while new developmentalism defends growth based on the export of manufactured goods, and, thereby, competitive integration into international markets;
  • Classical developmentalism was sceptical about exchange rate policy, preferring high tariffs, while new developmentalism has a theory on the determination of the exchange rate and gives a major role to exchange rate policy in ensuring national companies equal conditions of competition.[14]

Ensuing new policies

I conclude this comparison between classical and new developmentalism with a short discussion of the new policies that derive from the new reality and from new developmentalism. I will use Argentina and Brazil as references, which are both middle-income countries and exporters of commodities, i.e., they both suffer from the Dutch disease.

Firstly, what does not change? Essentially, industrialisation or productive sophistication remains the main strategy; moderate state intervention and economic nationalism – i.e., the rejection of dependence – remain the main requirements for successful catch-up; and strategic industrial policy continues to make sense.

What changes? How should a developmental state act?

Firstly, industrial policy is advisable but is not a substitute for a developmental macroeconomic policy.

Second, macroeconomic policy must care for the equilibrium of the fiscal account, following a post-Keynesian approach on that matter. New developmentalism is critical of the common knowledge that distinguishes developmental from liberal-orthodox macroeconomic policymaking according to the belief that progressive developmentalists would be free-spending while conservative liberals would reject free-spending. New developmentalism is strongly against the state engaging either in fiscal or exchange rate populism – the former meaning the state to spend irresponsibly more than it gets, the second, the nation-state (the country) spending irresponsibly more than it gets. Free-spending is fiscal populism and is not a responsible macroeconomic policy that progressive and conservative governments in developing countries often practice. Engaging in current account deficits with the argument that this is growth through foreign savings is exchange rate populism, which so-called austere liberals always practice whilst they defend current-account deficits. The additional capital inflows that such a policy requires cause the national currency to appreciate in the long-term, and encourage consumption, not investment.

What to say of the austerity that liberals practice when there is a macroeconomic maladjustment characterised by fiscal and current-account deficits. This implies an ‘internal adjustment’: the government maintains the exchange rate, while making a strong fiscal adjustment involving current expenditures and public investment, which will directly contribute to the recovery of fiscal equilibrium if the fall in the tax revenue caused by the ensuing recession is not bigger than the reduction of the fiscal expenditures; this will indirectly balance the current account because the recession will cause a fall in wages and will increase the country’s competitiveness without a depreciation. As there is no depreciation, the only people that will pay for the adjustment will be the salaried class; the rentiers will pay nothing.

Instead, given the maladjustment, new developmentalism also defends a fiscal adjustment but only of current expenditure, coupled with an increase in taxes, a reduced interest rate, the Dutch disease neutralised, some capital controls, and, so, a devalued national currency. In this case, the rentier capitalists also pay for the adjustment, and pay more than workers. They pay more because with the devaluation, the fiscal adjustment will be limited to current expenditures, the recession will be milder, and it will cause less unemployment; they also pay more because the revenues of rentiers – dividends, real-estate rents, and interest – lose as much acquisitive power as austere wages lose in fiscal terms; the pay more because the wealth of rentiers in the national currency loses value, while the workers don’t have wealth to be reduced; they pay more because the reduction of the interest rate is anathema for rentiers, while it is welcomed by workers; and they pay more, finally, because with the adjustment, employment is soon restored, benefiting workers.

Third, new developmentalism also states that macroeconomic policy must guarantee the equilibrium of the external account, thus firmly rejecting current-account deficits, and, if the country faces the Dutch disease, the current account should not be balanced but portray a surplus. The rejection of current-account deficits is the more counterintuitive new developmental policy, but it is very important to the country to invest and increase the investment and savings rate. Deficits should be rejected because they involve additional capital inflows that cause the national currency to appreciate and make the country’s capable manufacturing companies non-competitive. In the case of the existence of the Dutch disease, the country should achieve a surplus, because this disease is defined by the distance between the industrial equilibrium – which is the competitive equilibrium – and the current equilibrium exchange rate, which balances the current account. As the industrial equilibrium is more depreciated than the current equilibrium exchange rate, when the country is successful in neutralising the Dutch disease, it will necessarily present a current account surplus.

Forth, macroeconomic policy must maintain the five macroeconomic princes correctly – something that the market is unable to do. Government-created central banks and monetary committees tend to keep only two of these prices right: the inflation rate and the interest rate. An active exchange rate policy and an exchange rate committee are equally necessary.

Fifth, the interest rate is not just an instrument used by central banks to control inflation; it is also a price whose level, around which the bank practices its monetary policy, should be low.

Sixth, in middle-income countries the tendency of wages growing below productivity disappeared because the economy had already reached the Lewis’ point, when the unlimited supply of labour ceases to exist. The policy should be that indirect wages grow with productivity, and so do not threaten the profit rate, while indirect wages – mainly social expenditures – and progressive taxation take care of the economic inequality that characterises capitalism.

Seventh, as to the rate of profit, it is necessary to consider that we are no longer in a time of classical capitalism, when there was just capitalists and workers, but we are in a time of rentier-financier capitalism, when the capitalists are either entrepreneurs, whose investment decisions are fundamental to growth and therefore must have a satisfying rate of profit, or rentier capitalists, idle people whose remuneration should be as small as possible. Given that, and given the tendency towards cyclical and chronic exchange rate overvaluation, which presses down the rate of profit and makes competent manufacturing companies uncompetitive, it is not only direct wages that should not increase above productivity, but, what is more important, the exchange rate must be competitive in order to keep satisfying the rate of profit.

As to the growth strategy that follows on from new developmentalism, it is clear that it is not the import substitution strategy, which only applies at the very beginning of economic growth and involves a reduction of the openness coefficient. But this does not mean that it is an export-led strategy involving an increase in such a coefficient. The exchange rate must be competitive, floating around the industrial equilibrium. Assured of that, the relative efficiency of the diverse industries that comprise the national economy will determine the openness coefficient.

Now the supply side variables that determine productivity must be considered. Industrial policy may play a role in this matter by encouraging exports of a given industry for some time, but the essential say will come from the market; an industrial policy is not a substitute for a competent macroeconomic policy; it is also not supposed to resolve the inefficiency of companies. The state is supposed to guarantee the general conditions of accumulation by investing in infrastructure, healthcare, education, and technology, and to permanently improve its institutions and plan the non-competitive sectors of the economy, mainly in infrastructure and basic input industries.

As to distribution, this should be achieved by making the tax system progressive, keeping the level of the interest rate low so as not to excessively remunerate rentiers and financiers, by increasing taxes to finance the social state, and by increasing the minimum wage to the extent that there is room for these two last policies – meaning that they do not harm the competitiveness of the competent companies. Yet, this room must be understood in open and dynamic terms, given the fact that increased direct and indirect wages means more demand, more growth, and more room for distribution.


Luiz Carlos Bresser-Pereira

Emeritus professor of the Getulio Vargas Foundation



Bresser-Pereira, L. C. (2001a). The new left viewed from the south. In A. Giddens (Ed.), The Global Third Way Debate (pp. 258-371). Cambridge: Polity Press.

Bresser-Pereira, L. C. (2001b). A fragilidade que nasce da dependência da poupança externa [The fragility resulting from the dependency on foreign savings]. Valor 1000, September: pp. 34-38.

Bresser-Pereira, L. C. (2003). Desenvolvimento e Crise no Brasil [Development and Crisis in Brazil], 5th edition. São Paulo: Editora 34. First edition, Rio de Janeiro: Zahar Editores (1968).

Bresser-Pereira, L. C. (2006 [2007]). New developmentalism and conventional orthodoxy. Nordic Journal of Latin American and Caribbean Studies, 37(1), pp. 27-54.

Bresser-Pereira, L. C. (2016). Models of developmental state. Working Paper, EESP/FGV 426.

Bresser-Pereira, L. C. (2017). The two forms of capitalism: Developmentalism and economic liberalism. Brazilian Journal of Political Economy, 37(4), pp. 680-703.

Bresser-Pereira, L. C. (2018). From classical to new developmentalism: Why a new macroeconomics and a new political economy of development? Retrieved from

Bresser-Pereira, L. C. and Nakano, Y. (1983 [1987]). The theory of inertial or autonomous inflation. In L. C. Bresser-Pereira and Y. Nakano, The Theory of Inertial Inflation (pp. 65/82), Bolder: Lynne Rienner Publisher. Original in Portuguese (1983), Fatores aceleradores, mantenedores e sancionadores da inflação [Accelerating, maintaining and sanctioning factors of inflation].

Bresser-Pereira, L. C. and Rugitsky, F. (2018). Industrial policy and exchange rate skepticism? Cambridge Journal of Economics, 42(3), pp. 617-632.

Bresser-Pereira, L. C., Oreiro, J. L. and Marconi, N. (2014). Developmental Macroeconomics. London: Routledge.

Bresser-Pereira, L. C., Oreiro, J. L. e Marconi, N. (2016). Macroeconomia Desenvolvimentista. Rio de Janeiro: Elsevier. 

Bresser-Pereira, L. C. e Nakano, Y. (2002). Uma estratégia de desenvolvimento com estabilidade. Revista de Economia Política, 21(3), pp. 146-177.

Bresser-Pereira, L. C. and Nakano, Y. (2003). Economic growth with foreign savings? Brazilian Journal of Political Economy, 22(2), pp. 3-27. In English for the website edition; in Portuguese for the printed edition.

Cardoso, F. H. (1977[1980]). The consumption of the dependency theory in the United States. Latin America Research Review, 12(3), pp. 7-24.

Cardoso, F. H. and Faletto, E. (1969 [1979]). Dependency and Development in Latin America. Berkeley: University of California Press. First Spanish edition in 1969.

Evans, P. (1992). The state as problem and solution: Predation, embedded autonomy, and structural change. In S. Haggard and R. Kaufman (Eds.), The Politics of Economic Adjustment. Princeton: Princeton University Press.

Evans, P. (1995). Embedded Autonomy. Princeton, NJ: Princeton University Press.

Frank, A. G. (1966). The development of underdevelopment. Monthly Review, 18(4), pp. 17–31.

Hirschman, A. O. (1981). The rise and decline of development economics. In Essays in Trespassing (pp. 1-24). New York: Cambridge University Press.

Pazos, F. (1972). Chronic Inflation in Latin America. New York: Praeger Publishers.

Prebisch, R. (1949 [1950]). The Economic Development of Latin America and its Principal Problems. New York: United Nations, Dept. of Economic Affairs. Original Spanish publication in 1949.

Przeworski, A. (2001). How many ways can be third? In Andrew Glyn (Ed.), Social Democracy in Neoliberal Times (pp. 312-333). Oxford: Oxford University Press.

Rosenstein-Rodan, P. (1943). Problems of industrialization in Eastern Europe and South-Eastern Europe. Economic Journal, 53, pp. 202-211.

Simonsen, M. H. (1970). Inflação: Gradualismo x Tratamento de Choque [Inflation: Gradualism x Shock Treatment]. Rio de Janeiro: ANPEC.

Singer, H. (1950). The distribution of gains between investing and borrowing countries. American Economic Review, 40, pp. 473-85.

Thirlwall, A. P. (1979). The balance of payments constraint as an explanation of international growth rates differences- Banca Nazionale del Lavoro Quarterly Review, 128, pp. 45-53.

Thirlwall, A. P. and Hussain, M. N. (1982). The balance of payments constraint, capital flows and growth rates differences between developing countries. Oxford Economic Papers, 34(3), pp. 498-510.


[1] This expert comment builds on arguments the author has developed in ‘From classical to new developmentalism: Why a new macroeconomics and a new political economy of development?’ (2018).

[2] In an international conference organised by the Centro Celso Furtado in Rio de Janeiro, in 2015, I asked Bielschowsky for a more specific name for the school of thought than ‘development economics’ and not as regionally specific as is ‘Latin American structuralism’, and he suggested ‘classical developmentalism’.

[3] The first industrial revolution – the British – is often assumed to have emerged from liberal ideas, but this is not the case. The British Industrial Revolution happened in the time of mercantilism, that was the first developmentalism. Great Britain only turned an open economy, thus following the precepts of economic liberalism, in 1834.

[4] Actually, Keynes does not distinguish classical from neoclassical economics in his General Theory, because both were supply-sided, assuming Jean-Baptiste Say’s law, which holds that supply creates its own demand.

[5] Policy regime is a concept developed by Adam Przeworski (2001), while Bresser-Pereira (2001a), writing on the ‘new left’, remarked that in the 1980s, “the political center moved to the right” and social-democratic countries adopted very similar policies to liberal ones, while in the post-war period the inverse had happened: conservative political parties participated in the building of the social and developmental state.

[6] Even so, Keynesianism remained relatively present in the mainstream, because it was divided into a radically liberal current formed by monetarist, neoclassical, and Austrian school economists, and the New Keynesian school, which is basically hypothetico-deductive and neoclassical, but reserves some room for demand and acknowledge greater market failures.

[7] See Bresser-Pereira (2001b) and Bresser-Pereira and Nakano (2002; 2003).

[8] See

[9] See

[10] The more complete exposition of new-developmental macroeconomics is in Bresser-Pereira, Marconi, and Oreiro (2014; 2016). I also cite the 2016 Portuguese edition of the book because new developmentalism is a work in progress and this edition is more complete than the English edition because it was published two years later. The political economy of new developmentalism is mainly in Bresser-Pereira (2016; 2017).

[11] Except in relation to the ‘structuralist theory of inflation’, which eventually proved to have limited scope.

[12] Classical developmentalism’s pessimism in relation to the exports of manufactured goods was a major mistake that Latin American developmental economists made. When, in 1967, Brazil abandoned such pessimism and created an export subsidy that neutralised the Dutch disease on the export side – high tariffs had already neutralised it on the domestic market side – Brazilian exports of manufactured goods soared. They went from 6% of GDP in 1965 to 62% of GDP in 1990.

[13] In Rosenstein-Rodan’s (1943) big push model, which founded classical developmentalism, the huge and simultaneous investments that would benefit from crossed externalities, become internationally competitive, and trigger economic growth were supposed to be financed by foreign money. Some development economists defended some conditions for the admittance of foreign investments, but none rejected foreign borrowing. Up to 1970, they viewed the shortage of foreign capital as a major obstacle to growth. When, after the 1973 first oil shock, the major international private banks resumed finance to Latin-American countries, which had been unavailable since the 1929 crash and the Great Depression, development economists in Brazil commemorated the “good news”.

[14] See Bresser-Pereira and Rugitsky (2018). In this paper we have citations of Prebisch clearly showing this scepticism.


Copyright © 2018 by Dialogue of Civilizations Research Institute

The right of Luiz Carlos Bresser-Pereira to be identified as the author of this publication is hereby asserted.

The views and opinions expressed in this publication are those of the original author(s) and do not necessarily represent or reflect the views and opinions of the Dialogue of Civilizations Research Institute, its co-founders, or its staff members.

All rights reserved. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the publisher, except in the case of brief quotations embodied in critical reviews and certain other noncommercial uses permitted by copyright law. For permission requests, please write to the publisher:

Dialogue of Civilizations Research Institute gGmbH

Französische Straße 23

10117 Berlin


+49 30 209677900