Post n°159

Current and past recessions: a long-term perspective

Published on 04/27/2020

The current recession is expected to be shallower than the Great Depression of 1929, but deeper than the Great Recession of 2008. It could be shorter than these two financial crises because of the temporary and exogenous nature of the shock that caused it. It could contribute to the deployment of the digital economy, thereby boosting productivity and growth.

Chart 1a: The current recession compared to previous ones. Euro Area.
Chart 1a: The current recession compared to previous ones. Euro Area. Source: Note: (GDP growth in %, periods of war in dotted lines, IMF forecast in orange)

The Covid-19 pandemic that is currently affecting all countries is having direct adverse effects on economic activity and thus on GDP growth. These effects are amplified by the need to deploy health policies, over varying lengths of time and with varying degrees of intensity across countries. These policies aim to protect the population by slowing the spread of the pandemic, but weaken both the productive supply and the demand of economic agents. The latest IMF forecasts expect GDP to contract by an average of 7.5% in the euro area (Chart 1a), 5.9% in the United States (Chart 1b), 6.5% in the United Kingdom and 5.2% in Japan (Chart 2) over 2020 as a whole. The IMF stresses that these forecasts may be revised significantly depending on the pandemic’s evolution. These forecasts also project a strong rebound in 2021, which would not however offset the loss of GDP in the previous year (4.7%, 4.7%, 4% and 3% respectively).

Chart 1b: The current recession compared to previous ones. USA.
Chart 1b: The current recession compared to previous ones. USA. Source: Note: (GDP growth in %, periods of war in dotted lines, IMF forecast in orange)

A recession shallower than that of 1929, but deeper than that of 2008

A long-term perspective, here since 1870, shows that, aside from the two world wars of the 20th century, only the Great Depression of the 1930s was characterised by an annual contraction in GDP of this magnitude, if not at this stage even greater (Chart 1). For instance, during the Great Depression, the current euro area equivalent experienced three years of successive falls in GDP, of 3.9%, 6.2% and 4.2% between 1929 and 1932, without any significant rebound before the end of the Second World War. At the same time, the United States saw its GDP contract by 8.5%, 6.4% and 12.9% over these three years, before, nevertheless, experiencing several years of strong increases. The impact of the 1929 crisis on the economy was thus, at this stage, considerably greater and more prolonged than that of the current crisis. For example, compared to its growth trend between 1919 and 1929, the cumulative loss of GDP over the Depression years (1930-1932) by the euro area was 740 billion in 2015 euro equivalent, i.e. 64% of its 1929 GDP. For the 2020-2021 period, the GDP loss compared to the previous trend could amount in the euro area to EUR 1.6 trillion in 2015, i.e. 14% of its GDP in 2019 (based on IMF projections).

Chart 3a: Euro area GDP during the Great Depression
Chart 3a: Euro area GDP during the Great Depression. Source : Note: (in logs, index 1 at the beginning of the period, trend over the start of the period in dotted lines)
Chart 3b: Euro area GDP during the 2008 financial crisis
Chart 3b: Euro area GDP during the 2008 financial crisis. Source : Note: (in logs, index 1 at the beginning of the period, trend over the start of the period in dotted lines)

The financial crisis of 2008-2009 seemed so far to be the second significant episode of GDP contraction in modern history and was thus called the "Great Recession". However, the magnitude of the current contraction in GDP is expected to be much larger. As such, the Covid-19 recession is expected to become, in terms of its impact on growth, the deepest for a considerable length of time, after of course the Great Depression.

There could, nonetheless, be a significant difference. The Great Depression, which was also a financial crisis, was followed by a catch-up process that took several years, despite the highly favourable technological context (Gordon, 2000). In the United States, the 1929 level was reached again in 1936, for the euro area in 1937 (Chart 2); growth returned to the pre-crisis trend after a few years (before the break due to the Second World War for the euro area), but remained very far from its trend level. The financial crisis of the late 2000s was followed by a slower return to growth than in the previous period, with no immediate significant recovery of GDP losses. In this respect, many studies show that financial crises have very lasting effects on GDP (Reinhard and Rogoff, 2009). This is notably because the period leading up to the financial crises is characterised by an unsustainable growth trend, for example since it is based on a continuous growth in debt ratios that boost demand.

A faster rebound than after the 1929 and 2008 crises?

The path out of the Covid-19 crisis will largely depend on health policies aimed at protecting populations, which may influence the strength of both supply and demand. It will subsequently depend on the state of the production base at the end of the crisis and on effective workforce dynamics as well as productivity dynamics.

As regards the production base, all countries have implemented monetary and fiscal policies on a scale unprecedented in history in order to reduce the risks of capital destruction resulting from business failures (see, in particular, the IMF’s policy tracker). It is true that investment is expected to be temporarily depressed during the crisis, but economic recovery should be greatly facilitated by keeping the production base in operation.

As regards the workforce, the responses differ greatly across countries, and are contingent on their different institutional contexts. Some, such as the United States, leave market forces to play a major role, with large numbers of dismissals and a surge in unemployment registrations, exposing these workers to economic risks as well as to loss of human capital for firms, but facilitating the reallocation of labour when the recovery takes hold. Others, such as France, have introduced powerful, albeit costly, protective measures such as short-time work, which make it easier to keep workers in companies and thus reduce the risk of losing human capital.

Lastly, the Covid-19 crisis started at a time when, outside periods of war, productivity growth rates were historically low in all countries. It is quite conceivable that these rates will remain low in the aftermath of the crisis and that they will be even more affected by an uncertain international environment for several years to come. From this point of view, the measures taken by multilateral institutions to stabilise developing countries (debt moratorium, IMF loans, etc.) or cooperation between countries to mitigate financial tensions (in particular swaps between central banks in dollars) could help to reduce the uncertainty shock caused by this crisis.

The scenario based on the deployment of the digital economy

For all that, another more favourable scenario is conceivable: that of an acceleration in productivity resulting from a positive shock from the use of digital technology that has supported health policies and, in particular, lockdown periods. This would imply a real, widespread transition into the digital work era. This is reflected, for example, in the massive increase in teleworking, i.e. almost 40% of jobs in France according to DARES, and an estimated 34% in the United States according to Dingel and Neiman (2020). These forms of work save time (transport in particular) and constitute a major shift in firms' demand for commercial real estate. The associated productivity gains would primarily concern the service sector, which now accounts for the lion's share of the economy.

According to Bart van Ark (2016), the digital economy is still in its "installation phase" and is yet to enter its "deployment phase". The Covid-19 crisis may facilitate this rapid transition to this "deployment phase" of the digital economy, associated with significant productivity gains conducive to a return to stronger growth. This would help to pay off the debt burden that has been skyrocketing due to the support and protection policies implemented during the crisis.