VoxEU Column COVID-19 Economic history

The ‘Roaring Twenties’: Revisiting the evidence for Europe

Inspired by conspicuous historical parallels, some scholars and journalists have argued that GDP growth and productivity might boom in the aftermath of the Covid-19 pandemic. This column reviews the evidence for and against the ‘Roaring Twenties’ hypothesis, concluding that some countries might well experience a forceful economic expansion. But policymakers should avoid complacency and make the most of the Recovery and Resilience Facility funds, combining them with wide-reaching structural reforms to improve economic prospects for the decade to come.

The ‘Roaring Twenties’ was a decade (approximately 1921–29) of growing prosperity in the Western world, alimented by deferred spending, a boom in construction, and the rapid expansion of consumer goods, such as automobiles and electric home appliances. These factors materialised on the back of WWI devastation and, crucially, the N1H1 ‘Spanish Flu’ pandemic. As such, several pundits, journalists, commentators, and academics have been drawing parallels with that historical period, suggesting the post-Coronavirus recovery could be characterised by an economic boom, as illustrated in a recent cover story of The Economist. In this column, I review the evidence in favour and against a reiteration of the ‘Roaring Twenties’ in the 2020s, in order to draw policy lessons from the past.

There are indeed a host of parallels between current global conditions and those in the run up to the Roaring Twenties, albeit perhaps due to different underlying reasons. Analogies include the end of a pandemic (N1H1 then, Covid-19 now), the proliferation of new technologies (electricity then, AI/digitalisation now), a transport revolution (combustion engine then, electric vehicles now), political polarisation (rise of Communism and nationalism then, anti-establishment sentiments now), as well as emerging international rivalries, de-globalisation, and a soaring stock market.

Figure 1 shows real GDP per capita across selected Western countries. Two things should be noted. First, irrespective of a common trend across Western countries, it is evident that the Roaring Twenties materialised to different extents in various nations. For all, the challenge was to transition out of a centralised wartime economy, and back to a peacetime economic structure. The process was hardly smooth. In some cases, it led to moments of economic crisis, such as in Italy in 1921, when two large industrial trusts, Ansaldo and Ilva, collapsed after having outgrown during the war. As a result, the trend of a Roaring Twenties is particularly visible in France and the Netherlands, after WWI, but less so in Italy. 

In Germany, the decade was conspicuously characterised by hyperinflation, mainly due to money printing to finance the war, and then subsequent reparations. In the US, where the Roaring Twenties are most visible, GDP per capita went up from roughly $10,000 to almost $12,000 just before the 1929 stock market crash. Nonetheless, even there, the transition from a wartime to a peacetime economy was all but seamless. Lack of jobs for returning veterans led to strikes, social tension, and a rise in Communist sentiments – sentiments which were subsequently suppressed by the federal government.1 This is a period known as the ‘Red Scare’ (1919-1920). Eventually, the domestic rise of consumer credit fuelled consumerism, including for mass-produced cars like the Model T Ford. In parallel, the spread of electricity paved the way for a variety of home appliances, propelling the recovery.

Figure 1 Real GDP per capita for selected Western countries (2011 constant US dollars)


Source: Author based on Maddison historical data.

The second element to take away from Figure 1 is that the Roaring Twenties do not compare in magnitude to the post-WWII reconstruction boom in Europe.2 Perhaps what made them particularly ‘roaring’ – at least in terms of perception of progress at the time – was a comparison to a rather sluggish growth decade beforehand (something that holds true across countries). In 1921, US real income per capita was roughly at its 1906 level. This constitutes yet another parallel to the situation today, as Western countries in the aftermath of the Global Crisis have been characterised by so-called ‘secular stagnation’.

In what follows, I briefly list a set of arguments for and against the optimistic forecast of a high-growth decade for Europe following the end of the Covid-19 pandemic. In both cases, I will look first at the prospective for a cyclical short-term boost, and then the possibility that this feeds into a longer-term growth momentum.

The roaring enthusiasts

A variety of arguments have been used to push the view that we could experience a re-edition of the Roaring Twenties. These arguments often reach beyond standard macroeconomics, and include social psychology and management studies. 

First, looking at the short term, some pin their hopes to the fact that the recovery from non-financial crises is generally faster than from financial crises such as the 2008/09 recession3 (Reinhart and Rogoff 2015). Indeed, early evidence from countries where the pandemic was brought under control, such as Australia,4 shows an extremely rapid recovery in growth and jobs. This is because non-financial crises generally produce less debt overhang, except in the private sector. While this could potentially curtail investment and weigh on long-term growth, the low interest rate environment makes debt burden more tolerable, not only for the public sector, but also for private companies. Finally, for Europe, Next Generation EU is expected to support investment and might end up being perfectly timed to kick in when the cyclical recovery will start to gather steam as the vaccine campaign reaches a critical mass.5 

Second, and related to the point above, roaring optimists note that the European response has been swift and decisive. One year into the crisis, the economy has proved resilient and adaptable to operating in a pandemic environment, while manufacturing exports are supported by a strong nascent recovery in Asia, where Covid-19 is more under control. Although bankruptcies might increase once extraordinary support measures are lifted, these should be contained by a cyclical upswing once restrictions are removed (the so-called ‘mini-corona boom’). In this respect, the summer of 2020 – when economic activity effectively returned to pre-pandemic levels and a sharp GDP expansion ensued – lends scope for optimism. Evidently, the cyclical boom could be supported by excess savings and pent-up demand, on the back of the strong income support measures rolled out by EU governments. By some estimates, these excess savings could constitute a stimulus worth 3-6% of GDP across Europe (Kennedy 2021).

Further underpinning the prospects for a short-term boost, social scientists have pointed out to psychological factors dominating in the aftermath of a shock like a pandemic. In his most recent book, Apollo’s Arrow, Yale physician-turned-sociologist Nikolas Christakis argues that the aftermath of pandemics is typically characterised by a positive psychological uplifting, spurring social interactions, creativity, experimentation, risk-taking, and flourishing arts. Indeed, the 1920s was a time of liberation, in which women won the right to vote in several countries, and starting to wear more clothes of their choosing, smoke cigarettes, and drink in public – liberties that previously could not be enjoyed. In America, at the same time, Black poets, authors, and musicians found wider audiences, as part of what is commonly known as the ‘Harlem Renaissance’. Christakis argues we might observe similar patterns following Covid-19, as people rush to re-connect.6

On the back of all this, optimists see reasons to expect that, at the very least, Europe will rapidly return to its pre-crisis potential output, following a short-term cyclical boost once vaccines allow for a full re-opening of the economy.

From a more structural long-term standpoint, the conversation on the Roaring Twenties is linked to the long-standing debate between techno-enthusiasts – such as Erik Brynjolfsson at MIT – and techno-sceptics such as Robert Gordon at Northwestern University.7 Brynjolfsson has long argued that the digital revolution has been long brewing, and it is just a matter of time before we see its GDP and productivity boosting effects. This view is gaining policy attention, and was recently embraced also by Bank of England Chief Economist Andy Haldane.8

The Covid-19 pandemic, which severely reduced inter-personal contacts, forcing firms to digitalise, could have served as an accelerator of innovation and technological diffusion.9 Indeed, the World Economic Forum this year found that more than 80% of employers intend to accelerate plans to digitise their processes and provide more opportunities for remote work, while 50% plan to accelerate automation of production tasks. A global survey of executives by McKinsey & Co. revealed that many corporations were a ‘shocking’ seven years ahead of where they planned to be in terms of digitalisation. Finally, firms that made digital investments at the height of the crisis are likely to want to reap the benefits also once the pandemic is over. This could imply protracted hefty savings in terms of office space and business travel for companies, as well as cuts in commuting time for workers, leading to a structural boost in productivity statistics. Recently, even techno-pessimist Robert Gordon had to concur that Covid-induced digitalisation will most probably lead to a surge in productivity during the early 2020s.10

The roaring pessimists

Unsurprisingly, many representatives of the ‘dismal science’ have a variety of reasons to be more pessimistic about the future. 

First, notwithstanding the strong crisis response at national and EU level, some damage in the economy will probably be unavoidable. Virus containment measures are likely to have affected individuals and workers through a variety of channels. In particular, there is ample evidence to suggest that young people who enter the job market during a recession face lower wages, which linger on for over a decade (Oreopoulos et al, 2012).11 

A similar effect could materialise for workers who probably experienced some skill losses while idle during lockdowns. According to research by the World Bank, countries struck by pandemic outbreaks (before Covid-19) experienced a marked decline in labour productivity of 8% after three years relative to unaffected countries (World Bank, 2020). Along similar lines, IMF research based on 133 countries between 2001 and 2018, shows how pandemics tend to reduce output and increase inequality, stoking social unrest, further lowering output and worsening inequality (Sedik and Xu 2020). 

In line with this pattern, the Roaring Twenties in the US were characterised by staggering inequality. This is best exemplified by F Scott Fitzgerald’s 1926 novel The Great Gatsby, slowly setting the scene for the Great Depression. In addition, it might very well be that the pandemic shock will lead to a structural increase in savings rates, as people become more cautious after having experienced unexpected hardship (Eichengreen 2021). If this were the case, ‘excess savings’ would be here to stay,  diminishing the power of the pent-up demand optimistic argument, and further compounding the savings glut behind Larry Summer’s ‘secular stagnation’ hypothesis for advanced economies (see also Jordá et al 2020). 

Second, the fact that a wave of bankruptcies has not materialised yet is simply due to the fact that state aid – together with debt moratoria and some degree of financial repression – have been rolled out forcefully, based on the presumption that the Covid-19 recession was fundamentally a liquidity crisis with transient impacts. Most, if not all, firms have been propped up. But once the pandemic dust settles, some sectors might have disappeared or declined permanently, with classic examples being structural transformations of retail (which moved to e-commerce) or business travel. Both sectors are unlikely to return to previous glories following the pandemic.12 Economic policies that were appropriate in the early phase of Covid-19 could impede this structural transformation, supporting irremediably stagnant sectors, generating ‘zombification’. Based on this risk and due to a lower resilience or capacity for swift transformation,  in its latest WEO update the IMF expects the EU and euro area to return to pre-crisis levels of GDP much later than other advanced economies (such as the US and Japan). 

Third, there has been a political consensus on using fiscal policy to support the economy so far. But the fiscal tool could be constrained going forward, depending on a variety of factors such as the future evolution of inflation and the ECB’s reaction to it, the retention of trust on financial markets, and the future evolution of EU fiscal rules. Fiscal policy could either become a drag on growth, or not be in the condition of being supportive should a new macroeconomic shock occur. 

Fourth, Peruzzi and Terzi (2018) have shown that growth accelerations – meaning a structural and sustained increase in growth rates – remain rather elusive in advanced economies, and are typically associated with wide-reaching structural reforms. This argument speaks to Daniel Gros’ view that the effect of the EU’s Recovery and Resilience Facility will hinge mostly, if not fully, on whether it triggers wide-reaching structural reforms: something that remains far from obvious (Gros 2020). 

Finally, taking a more global perspective, advanced economies have bought the vast majority of available vaccine supplies for 2021, in the hope of being able to save lives and re-open their economies swiftly.13 But the global nature of supply chains suggests that production and trade will remain disrupted so long as developing countries are affected by the pandemic. Çakmakli et al. (2021) have recently shown that up to half of the coronavirus-related economic damage in 2021 could still accrue to rich countries, even assuming fast and complete vaccination rates for their citizens. Beyond trade and production, the continued proliferation of the virus in other countries implies a persisting risk of vaccine-resistant variants developing and spreading globally (Brown et al. 2021). In other words, Covid-19 could have a protracted impact over years, which will not be resolved by a fast vaccination campaign in the next few months. This could hamper the prospects of a roaring decade.14


After reviewing both arguments in favour and against, a replication of the Roaring Twenties looks far from granted for Europe as a whole. But what cannot be excluded is that an economic boom might materialise for selected countries within the EU – perhaps on the back of a more successful rollout of the Recovery and Resilience Fund (RRF) or an export-led model anchored to the resumption of global growth. We saw how exits from moments of crisis can pave the way for a differentiated recovery, even within a general supportive growth environment (as happened in the early 1920s). Further, large mega-trends like the shift to a new growth model based on the digital or green economy, and the challenges related to ageing, could further exacerbate the problem of uneven growth within the Union.

Different recovery paths, or a localised re-edition of the Roaring twenties, would have repercussions in terms of widening divergences in the EU on a variety of policy indicators. On the fiscal front, some countries might be better placed to rapidly re-absorb the high public debt levels incurred as a result of the Covid-19 crisis response. On the monetary policy front, diverging paths of economic success might once again create a ‘one size fits none’ problem for the ECB (Enderlein 2005). On the internal migration front, an uneven recovery could spark continued large involuntary migration of young people, as experienced following the euro area crisis, with vulnerable countries losing precious human capital. Evidently, all this would have rippling repercussions on political cohesion and consensual decision-making on the way forward in terms of EU policy integration. Within this context, national governments should avoid complacency inspired by parallels with the original Roaring Twenties. The ambition and quality of the public policy response to the crisis, including the degree to which Recovery and Resilience Fund is well spent, and coupled with wide-reaching structural reforms, is ultimately what will define economic prospects in the ‘twenties’ of the 21st century. 

Author’s note: The views expressed here are those of the author and do not necessarily represent those of the institution to which he is affiliated. 


Brown, C, M de Bolle and M Obstfeld (2021), “The pandemic is not under control anywhere unless it is controlled everywhere”, PIIE Real Time Economic Issues Watch, 2 February.

Çakmakli, C, S Demir, S Kalemli-Ozcan, S Yesiltas and M A Yilirim (2021), “The Economic Case for Global Vaccinations: An Epidemiological Model with International Production Networks”, NBER Working Paper 28394.

Eichengreen, B (2021), “An Infrastructure Inoculation for America’s Recovery”, Project Syndicate, 13 January.

Enderlein, H (2005), “One size fits none”, Central Banking 16(1).

Gostin, L O, D E Shalala, M A Hamburg, B R Bloom, J P Koplan, B K Rimer and W A Williams (2020), “A global health action agenda for the Biden administration”, The Lancet 397(10268): 5-8.

Gros, D (2020), “Europe and the Covid-19 crisis: the challenges ahead”, CEPS Policy Insights 20. 

Jorda, O, S R Singh and A M Taylor (2020), “Longer-run Economic Consequences of Pandemics”, NBER Working Paper 26934. 

Oreopoulos, P, T von Wachter and A Heisz (2012), “The Short- and Long-Term Career Effects of Graduating in a Recession”, American Economic Journal: Applied Economics 4(1): 1-29.

Peruzzi, M and A Terzi (2018), “Growth Acceleration Strategies”, Harvard CID Working Paper 91.

Phillips, N (2021), “The coronavirus is here to stay — here’s what that means”, Nature 590: 382-384. 

Reinhart, C and K Rogoff (2014), “Recovery from Financial Crises: Evidence from 100 Episodes”, American Economic Review 104(5): 50-55. 

Sedik, T and R Xu (2020), “A Vicious Cycle: How Pandemics Lead to Economic Despair and Social Unrest”, IMF Staff Working Paper 216. 

World Bank (2020), “What Happens to Productivity During Major Adverse Events?”, in A Dieppe (ed.) Global Productivity: Trends, Drivers, and Policies, Washington, DC.


1 Between 1919 and 1920, federal troops were employed to assist in putting down labour disputes and other public disturbances 29 times.

2 The first use of the term ‘roaring twenties’ dates back to 15 August 1923, and was not used to define the social and economic boom at large, but rather the development of the movie industry, specifically for women. Intriguingly, it was first used in the Lincoln Journal Star (Lincoln, Nebraska) to note that “During the war and after the war up to the big movie slump of 1920, Hollywood and Los Angeles were literally swamped with would-be movie queens. New companies were springing up like mushrooms […]. But the "roaring twenties" have gone, and the movies are now on a different and very much more efficient basis. Casting directors and welfare workers are unanimous in declaring that, with the greatest talent and the best of looks and the rarest of luck, no unknown girl nowadays can hope to make even a small reputation in Hollywood inside of two years”. This nostalgic use of the term suggests that the current bonanza we associate with the ‘Roaring 20s’ was far from clear in the first years of the 1920s, even to those living through them.

3 Evidently, this rests on the assumption that the economic fallout from Covid-19 will not generate a subsequent financial crisis: something that will require careful timing and sequencing of policy support withdrawal. 


5 This point of view stresses the structural, rather than cyclical, nature of the Recovery and Resilience Facility. Something that former governor of the Bank of England Mervyn King has recently highlighted as the fundamental characteristic of any recovery package. 

6 A similar point was recently made by Janan Ganesh (2021), “What really roared about the Twenties”, Financial Times, January 8. 

7 Bob Gordon’s view, expressed in The Rise and Fall of American Growth, is that productivity growth is over, because the latest innovation, including AI and computers, cannot match the great innovations of the past. These include those that characterised the 1920s, such as electrification and the internal combustion engine. More recently, when reflecting on the Roaring 20s hypothesis, Gordon remarked that “While it is likely that productivity growth will revive somewhat in the 2020s from the dismal record of the 2010s, there is no chance of sustained decade-long growth that matches the achievement of the 1920s.”


9 More broadly, the pandemic has probably pushed several firms, including SMEs, to enter the digital market, creating a more competitive environment. It is therefore possible that following the peak of the pandemic, and once the economy reopens, some Schumpeterian creative destruction effect could be at play.

10 https://www.anderson.ucla.edu/centers/ucla-anderson-forecast/forecast-direct/february-2021

11 Relatedly, albeit on a longer time horizon, distance learning might have affected the overall quality of human capital formation, and particularly so for students coming from disadvantaged backgrounds.

12 This structural transformation needs to be read within the context of wider changes in the economic system, not least the policy acceleration towards low-carbon technology as part of the European Green Deal. This further compounds the challenge related to innovating business practices and it is unclear yet what type of impact it will have on GDP and productivity growth.

13 High-income countries, representing 16 percent of the global population, have bought at least 70 percent of COVID-19 vaccine doses available in 2021 (Gostin et al. 2020)

14 In January, the academic journal Nature asked more than 100 immunologists, infectious-disease researchers and virologists working on SARS-CoV-2 whether it could be eradicated. Almost 90% of respondents think that the coronavirus will become endemic, meaning that it will continue to circulate in pockets of the global population for years to come (Phillips 2021).

19,737 Reads