“I’ll be the greatest jobs president that God ever created.” – Donald J. Trump

The arrival of Donald Trump into the White House initiated a new wave of industrial policy in the West. Prior to Mr. Trump it was the 2008/09 recession that ushered in Mr. Obama’s economic nationalism: “Our first priority is making America a magnet for new jobs and manufacturing”, and before that it was the Great Depression which paved the way for FDR’s ‘New Deal’ reforms. In this way, industrial policy lies at the heart of a uniquely Anglo-American paradox (Wade, 2014): in a fundamentally neo-liberal environment, industrial policy has found its way – somewhat covertly – onto the manifestos of market fundamentalists such as Reagan and Thatcher. (Juhász et al., 2023). In times of hardship American policy has quietly abandoned its own so-called ‘Washington Consensus’ and embarked on the taboo business of government intervention, often in the form of an industrial policy. That is a testament to its allure of these policies but does not prove its success. Have industrial policies had their intended effect(s)? Are they more effective at different stages of development? And which types of industrial policy have been most successful historically? 

Despite the extensive literature and bipartisan support, industrial policy as a blanket term is still loosely defined; few mainstream policy fields elicit more controversy from economists (Pack & Saggi, 2007). Stiglitz et al. (2013) broadly label industrial policy as “government policies directed at affecting the economic structure of the economy”. Warwick (2013) outlines potential objectives of an industrial policy: 

“…any type of intervention or government policy that attempts to improve the business environment or to alter the structure of economic activity towards sectors, technologies or tasks that are expected to offer better prospects for economic growth or societal welfare than would occur in the absence of any such intervention …” 

Warwick proposes “economic growth” and/or “societal welfare” as two commonly applied justifications for industrial policy. Soete (2007) identifies that industrial policy “typically contains an element of… self-reliance in bringing about growth and development”. In the developed world, geopolitical tensions and supply-chain shocks have heralded the return of industrial policy in a dash for economic “self-reliance”. On the other hand, developing nations typically harness industrial policy to facilitate rapid “growth and development”, otherwise known as industrialisation. Possible industrial policy objectives might be categorised into the following four instances: 

a) economic independence in vital industries (often through ‘national champions’)

b) supporting manufacturing and job creation domestically (typically in lagging regions) 

c) protecting domestic ‘infant-industries’ until they are internationally competitive

d) catalysing industrialisation and development through structural transformation

It is this last category, commonly seen in the early stages of development, which has produced the most disparate results. After the second world war, late-developing countries set out to replicate the rapid industrialisation pioneered by Britain in the 18^th and 19^th centuries. The British industrial model, although commonly attributed to technological innovation, could not have taken place without the protectionist measures laid out 500 years earlier to cosset its wool industry (Kalish & Wolf, 2023). By 1820, Britain charged a 50% tariff on imported goods – the highest in Europe (Shafaeddin, 1998). Such measures did not cease until after the revolution. Classical liberals like Adam Smith (1776) and his neo-liberal ideological descendants argue that Britain was able to industrialise not because of industrial policy but despite it. But it is near-impossible to see how Britain would become the “workshop of the world” (Deane, 1965) without protection against textiles superior in price and quantity (Parthasarathi, 2011) from India. Two Calico Acts and countless tariff hikes later, the British cotton industry was able to innovate (Hargreaves’ spinning jenny/Crompton’s mule) and finally surpass India in global exports by 1820. Industrial policy allowed Britain to protect its infant cotton-textile industry. The protective barriers around the industry afforded it time for ‘learning-by-doing’, thus ultimately increasing domestic output with improved productivity, quality of output and cost of production. Of course, such egregious import tariffs in the latter half of the 20^th century and onwards would be incompatible with WTO trade rules, let alone Adam Smith, and would likely trigger destructive trade wars. It is important to remember that Britain was only able to impose such tariffs on Indian imports without consequence because India was under British subjugation. 

Both Stiglitz et al. and Warwick emphasise the impact of industrial policy on ‘economic structure’. Rodrik (2008) suggests that economic development is dependent on structural change in the economy, e.g. the shift from agriculture to industry to services (Aizenman et al., 2012). Industrial policy, being a facilitator of structural change, is oftentimes a precursor to development (note that well-implemented industrial policy is a necessary condition, but not a sufficientone.) According to Shafaeddin (1998), “no country has developed its industrial base” without some form of industrial policy, typically infant-industry protection. Furthermore, O’Brien (1998) claims “for real and sustained development there is no substitute for industrialisation.” However, this conclusion relies on the assumption of the ‘dual-sector’ model (Lewis, 1954) as fact. Lewis saw the ‘subsistence’ sector as fundamentally stagnant and inefficient. At an early stage of development, there is an excess of labour, a variable resource, in relation to the land available, a fixed resource. The result is a diminishing marginal product of labour and consequently lower wages. The disparity between wages in the industrial, ‘capitalistic’ side of the economy and the subsistence sector drives labour towards industry, driving up overall productivity and therefore profits which are then reinvested to increase rates of capital accumulation, a key determinant of economic growth. Crucially, the rate of this growth is dependent on the rate of transfer of the factors of production, primarily labour, to more productive industries. Industrial policy can play a role in catalysing this development. By introducing selective supply-side industrial policy in the form of wage subsidies or R&D tax credits for industrial firms, governments can increase incentives for new firms to enter the industry and reinvest their profits, creating development. 

Similarly, the theoretical argument for adopting industrial policies in developed economies aims to rectify inefficiencies in the free market. Markets fail. This is largely uncontroversial with both proponents and opponents of industrial policy. Take, for example, the case of positive externalities within certain industries. Due to the public-good nature of information (Stavins, 2011), it is inevitable that firms who invest into R&D operations (research and development) do not always receive a full return on their investment. Adjacent firms will, despite functional property rights, benefit from the innovation without incurring any costs in R&D. As a result of these appropriability concerns, the private sector will tend to innovate less than is considered optimal when left to its own devices. A ‘vertical’/’selective’ industrial policy in the form of R&D subsidies and tax credits is increasingly popular in “virtually all developed countries” (Takalo et al., 2013) as a policy response to this market failure.

In the graph above, the MPB (marginal private benefit) curve lies below the MSB (marginal social benefit) curve because an additional social benefit (distance from MSB to MPB) spills over to a 3^rd party outside the market transaction. Instead of producing at the equilibrium between MSC and MSB, the firm underproduces at MSC/MPB. The aforementioned industrial policies aim to internalise this externality by the shifting the market to the socially optimal equilibrium and eliminating the dead weight welfare loss at d.

Despite extensive theoretical and historical evidence affirming the crucial role industrial policy plays in achieving the aforementioned four objectives, there remains stiff resistance to its implementation, particularly in the cases of infant-industry protection and ‘vertical’ policies (targeted at a specific firm or industry). Industrial policies work best when they are ‘horizontal’, (i.e. refrain from ‘picking winners’ within an industry) and are deployed in the context of fostering industrialisation and development, rather than obstructing free trade.