The few times that I have been to conferences in Mexico or in South America, I have been asked what advice I could offer on economic policy from a post- Keynesian point of view. My answer has always been the same: I bring the economic theories; you assess whether the theories and their assumptions apply to your country since you are the experts on its economic environment and constraints. I know fairly well the institutions and economic conditions of my own country —Canada—but, since I have little expertise in economic development, I only have a very vague idea of whether the problems, let alone the solutions, are similar in emerging countries. My intention is not to act like the technocrats from the International Monetary Fund who believe that whatever the problem or the country, the same standard recipe—their structural adjustments—can always be imposed.¹

This being said, there are policies that post-Keynesians are likely to recommend in most circumstances. I have long argued that one of the key presuppositions of heterodox economics is the belief that unfettered markets—markets left on their own—are unlikely to be associated with economic stability. While (most) post-Keynesians are ready to take the position that capitalism is likely to provide the dynamism Schumpeter highlighted, they are also keenly aware that capitalism must be tamed and that the State must play a key role, both at the macroeconomic level through its budgetary expenditure and by imposing proper regulations at the microeconomic dimension, but also through leadership at the mesoeconomic level. At the sectoral level, the key role of the State for technological development has recently been well documented by heterodox authors close to post-Keynesian economists such as Mazzucato (2013).

Post-Keynesians have also always been reluctant to join the fashionable excitement around capital flow liberalization. The IMF finally came to its senses by recognizing that free capital flows might be appropriate for developed economies but were most likely to be damaging for emerging economies with less sophisticated financial markets (Grabel, 2018). Besides the high dangers of financial crises and economic recessions generated by these uncontrolled capital flows, there has also been a realization by the IMF that the liberalization of foreign capital flows has led to a rise in income inequality (Ostry, Berg & Tsangarides, 2014), as it increases the negotiating power of firms to the detriment of labour. Indeed, when Jonathan Ostry, in 2016, made the above claims during a lecture he was invited to at the University of Ottawa, I pointed out that free trade was likely to have the same effect as free capital movements because it would also enhance the bargaining power of firms and weaken that of trade unions since firms could now import goods from anywhere in the world and threaten to move production abroad. Ostry reluctantly agreed that this was indeed a logical step to take. There are, thus, two reasons to be suspicious of free trade: authors such as Chang (2008) have long argued that free trade might be good for powerful countries whereas emerging countries may never get their opportunity to develop their own industries; moreover, we also now know that free trade is likely to increase income inequality. Thus, even mainstream authors, at least those at the IMF, now push aside the claim that we have to arbitrage between equality and efficiency, as was famously claimed by Okun (1975) in a little book that all of us were asked to read when I was finishing my undergraduate studies. Income inequality is now perceived as being likely to be harmful to growth.

This brings me to a topic which has generated a lot of interest among heterodox economists in general and post-Keynesians in particular and even among researchers in comparative political economy: wage-led versus profit-led growth. The topic has been brought beyond the realm of academia by the International Labour Office, which discussed it extensively in its 2012 annual report. I played a small role in all of this as I was asked to make a plenary presentation on wage-led growth at the 2009 ILO conference in front of the tripartite delegates. This was followed by a workshop on wage-led and profit-led growth that was held at the 2012 conference, which give rise to the book that I edited with Stockhammer (Lavoie & Stockhammer, 2013). One thing that we found hard to convey was the distinction between a wage-led regime and wage-led policies. A regime is how the economy reacts to an autonomous change in the wage or the profit share. It depends on propensities to consume, investment functions, the structure of international trade, and the way firms react to changes in unit labour costs. An economy has a wage-led demand regime if an increase in the wage share leads to an increase in economic activity; it is a profit-led demand regime if this leads to a decrease in economic activity. By contrast, policies are wage led (profit led) if they intend to raise or favour the wage share (profit share).

Heterodox econometricians have been particularly active around the question of whether or not various countries were empirically in a wage-led or profit-led regime. As one would expect, authors belonging to different research traditions have yielded opposite results, and not always because they were using different econometric methods. Anyway, my understanding of all this has been that most developed countries are wage led while countries exporting primary goods, even developed countries such as Canada or Australia, tend to be profit led. Surprisingly, at least for some observers, countries with a large export sector are not necessarily profit led. Almost all of the ten studies undertaken for South Korea have shown this to be a wage-led country (Lee, 2019). Also, a study has found that China used to have a profit-led demand before the 2008 financial crisis but has entered a wage-led demand regime since then as a result of changes in consumption and investment parameter values and because the proportion of trade has diminished (Jetin & Reyes, 2019).

However, I have become aware by attending various sessions at conferences that there is substantial scepticism regarding the feasibility of wage-led policies in emerging countries. These doubts arise both from a practical point of view, in the sense that governments are unlikely to endorse them, and from an economic point of view, in the sense that several heterodox economists are convinced that emerging economies have profit-led demand regimes. One of the arguments being advanced in support of profit-led demand regimes is that emerging economies would be supply-constrained. As I mentioned in my book (Lavoie, 2014, pp. 278), Kalecki himself believed that situations of excess capacity first and foremost concerned capitalist economies. Here is what he thought back in 1966, as summarized by Feiwel (1972) from an article written in Polish by Kalecki:

Whereas the crucial problem of a developed capitalist economy is the adequacy of effective demand, as such, an economy possesses a stock of productive capital which more or less matches the existing labor force and is therefore capable of generating a rather high per capita income, provided that its resources are fully employed. The cardinal problem of the underdeveloped (or less developed) economies is the deficiency of productive capacity rather than the question of its variable utilization (underutilization) (pp. 18-19).

One may wonder whether Kalecki’s opinion was overly influenced by his observation of socialist economies, in his case Poland. Kornaï (1971), observing Hungary, thought that capitalist economies had sales competition, with excess supply and sellers competing with each other to capture customers. Conversely, socialist economies were under purchase competition, with demand always being constrained by a lack of supply and with buyers (both firms and consumers) competing to capture commodities. Can past underdeveloped or current emerging economies be characterized along the lines of deficient productive capacity and deficient supply, as, for instance, is argued by Razmi (2016)? Perhaps they can. As I said earlier, I am no emerging economies expert, but what I find intriguing is that two of the first four authors in the Kaleckian tradition to build wage-led models came from India (Amitava Dutt) and Brazil (Edward Amadeo): both emerging economies. Taylor (1983, pp. 13-14), in a book on ‘applicable models for the third world’, describes his wage-led model based on excess or spare capacity by referring to ‘countries at middle and low levels of gross domestic product per head’.² In addition, Ro Nastepad and Servaas Storm, who have carried out very interesting theoretical and empirical work on demand and productivity regimes, became attracted to the Kaleckian model of growth as a consequence of their earlier studies on the Indian economy. In a recent interview, Dutt (2020) reasserts that wage increases in the informal sector are likely to have little negative effects on investment in the formal sector or on exports, and that small open economies, in contrast to mainstream assumptions, cannot export all that they can produce, as emphasized by balance-of-payment growth models à la Thirlwall.

Kaleckian models and empirical studies have perhaps put too much emphasis on wage and profit shares. Through the studies by Thomas Piketty and his numerous collaborators, inequality based on personal income has been in the limelight, and there has been a recognition that rising divergences within the wage share could be the cause of as large, if not bigger, income inequality compared to functional distribution between wages and profits. In fact, this is why I have always paid attention to the distinction made by Rowthorn (1981) between managerial salaries and the wages of direct workers, or what some call supervisory and non-supervisory workers, as have others more recently (Palley, 2017). Indeed, this distinction is not innocuous: there is evidence that the presence of managers, who are less directly related to production, may bias empirical studies towards findings of profit-led demand regimes (Cauvel, 2019; Lavoie, 2017). In addition, income inequality is also related to social benefits—social security, unemployment insurance, and pensions for the elderly—as well as to the income that can be obtained from the informal economy. The positive effects on aggregate consumption of a more equal income share also arise from improvements in these social benefits. If data for the Canadian economy can be generalized for other countries (Costantini & Seccareccia, 2020), the evidence is quite clear that individuals or households in the poorer quintiles have a propensity to consume which is very much higher than that of agents found in the richer quintiles, so improvements in the degree of equality in personal income distribution will for sure increase consumption demand.

Over the last few years, it has been recognized that the demand effects of an increase in the wage share are likely to be modest. From the very beginning, it was pointed out that to neutralize the potential negative effects on net exports, the positive effects of such wage increases would best be achieved if they could be pursued in a concerted way by countries belonging to the same economic zone: not an easy task (Capaldo & Izurieta, 2013). Coordination was also key in the response to the Global Financial Crisis in 2009. Such wage increases would also best be associated with a concomitant expansionary fiscal policy (Obst, Onaran & Nikolaidi, 2017). Perhaps the newly-found celebrity of the advocates of Modern Monetary Theory (MMT)—a branch of post-Keynesian economics that minimizes the financial dangers of public debt—will induce more governments to go ahead with such policies, although, once again, one may fear that MMT scriptures may not fully apply to emerging economies (Bonizzi, Kaltenbrunner & Michell, 2019).

Finally, one cannot be silent on the practical difficulties in moving towards less neo-liberal and more progressive economic policies, as well as having them accepted by the ruling class. I realized this when I went to a conference in Argentina in November 2013. In the car driving us from the airport to Buenos Aires, I was discussing the conjecture of the economy with an interpreter of the central bank of Argentina. To my surprise she told me how annoyed she was with the progressive social policies pursued by the Argentine government. She complained that social benefits had been improved so much that now her housemaid refused to spend four hours a day on public transport to come and work in her house at the current wage. She would need to pay her more. This led me to think that if middleclass citizens do not endorse these ‘income-led’ policies, the government was likely to be beaten in the forthcoming elections, which is exactly what happened. I am told that similar motives can explain the results of the presidential election in Brazil in 2018. Other South American countries—Bolivia for instance—also seem to be currently subjected to this sort of revenge of the ruling or upper-middle class.

Besides these political difficulties, even when wage-led or income-led policies are actually being implemented, as has been the case in Brazil or South Korea, they become easy targets for mainstream economists whenever the economy slows down. As it is very difficult to quickly disentangle the causes of such a slowdown between wage-led policies, monetary policies, banking behaviour, the exchange rate, a capital strike, degrowth by foreign customers or domestic fiscal restraints, it is easy to blame progressive economic policies, as recounted by Serrano and Summa (2015) for Brazil or by Lee (2019) for Korea.