Navigation – Plan du site

AccueilNuméros29The macroeconomic effects of covi...

The macroeconomic effects of covid-19:
the imperative need for a Keynesian solution

Les effets macroéconomiques de la covid-19 : l’impératif d’une solution keynésienne
Los efectos macroeconómicos de la covid-19: la necesidad imperiosa de una solución keynesiana
Mikael Randrup Byrialsen, Finn Olesen et Mogens Ove Madsen


Une très grave récession a frappé l’économie mondiale dans le sillage de la covid-19. L’ampleur exacte de ce choc est incertaine, mais beaucoup pensent que le recul du PIB sera d’une ampleur similaire, voire pire, à celle des premières années de la Grande Récession. Il est donc impératif de prendre des mesures politiques de nature keynésienne, comme celles qui ont déjà été mises en œuvre dans de nombreux pays. Compte tenu des circonstances atypiques de la crise, il convient de suivre un nouveau type de politique post-keynésienne orientée vers l’offre. L’objectif de cette politique devrait être de soutenir les entreprises en vue d’éviter la destruction de leurs capacités de production et de progresser dans la transition écologique. Cet article utilise le cas du Danemark pour présenter un modèle post-keynésien stock-flux cohérent (SFC) axé sur la demande et basé sur les restrictions du marché du travail. Il souligne certains aspects de l’impact économique négatif de la covid-19 sur l’économie danoise et des politiques adoptées par le Parlement danois.

Haut de page

Texte intégral

We would like to thank the journal’s rewiever and editors for comments on an earlier version of the present article. We would also like to express our gratitude to Robert Smith. The usual caveats apply.


  • 1 Moreover, this kind of empirical evidence might give way to other perspectives on macroeconomics th (...)

1The dramatic impact of the Great Recession on global economic performance, and the slow and irregular recovery of many countries over the past decade has led to the fear in a growing number of economists of a potential scenario of secular stagnation. Even with the inclusion of a number of mainstream economists. Amid growing insecurity and concern as to why post-recession levels of GDP had not recovered to the pre-recession trends, (Summers, 2014, p. 69) commented: “that the record of industrial countries over the last 15 years is profoundly discouraging as to the prospect of maintaining substantial growth with financial stability”1. Secular stagnation, various kinds of hysteresis and the monetary reality of a binding ‘zero lower bound scenario’ seem all to have found a place on the macroeconomic agenda for good few years to come.

2However, today’s scene of macroeconomics has changed dramatically. The global pandemic related to the covid-19 virus and the ensuing constriction of economic activity have presently, in 2020, brought the global economy to a stand-still with huge cuts in GDP-levels for all countries. The total magnitude of this shock to the world economy is of course uncertain, as is the length of the current recession. However, seen from a January 2021 perspective, one might hope for at least a slow recovery in GDP levels in the coming year as the covid-19 vaccine program is rolled out. Being hit hard by the covid-19 pandemic, all countries understood the imperative need for quick policy actions. A combination of strictly lockdown to break the chains of infection and stimulus to aggregate demand had to be introduced. The lockdown in most countries, however, can be interpreted as a source of exogenous supply shocks to the economy, which in turn is affecting aggregate demand through various channels. The direct effect of lockdown is obvious, however, the total damage associated with it may be much larger, if there are significant feedback effects in the economy.

3The aim of this paper is primarily to focus on a Danish case study. As a small and very open economy, the Danish economy is dependent on trade and a global crisis is therefore expected to hit the Danish economy very hard. Although not acting with a full membership of the EMU, the country has chosen to respect and abide by rules of the EMU with respect to monetary as well as fiscal policy, which restricts maneuverability in economic policy intervention. On the other hand, Denmark is characterized as a Nordic welfare state with a very active public sector and a strong tradition of countercyclical policy. This time, however, due to the nature of the shock, traditional demand policy needs to be mixed with initiatives focusing on the supply side. Furthermore, since the current pandemic is a zoonosis, investing in ecological transition may be a top priority, which requires policy actions in the short run to be consistent with a long-term sustainable economic policy strategy – respecting the urgent need of a necessary green transition. This analysis of how the Danish economy is impacted by the covid-19 pandemic, and how one could react policywise, is carried out using an empirical Post-Keynesian Stock-Flow-Consistent (PK-SFC) model for the Danish Economy. The model presented in this paper advances the standard PK-SFC by adding supply side restrictions to the labour market, which is very relevant for analysing the effect of a lockdown of an economy. This model allows us to investigate how exogenous shocks may propagate to the economy through various channels, and what the subsequent feedback effects are that could amplify the impact of these shocks. Furthermore, the model allows us to simulate various policy scenarios, which enable the discussion of the relevance of Keynesian policy.

4The two primary contributions of the paper are to add to the literature analyzing the macroeconomic effect of the covid-19 crisis, by offering a consistent analysis in a fully-fledged macroeconomic model. Second, we explore different policy options, and thus offer some useful information for shaping macroeconomic crisis recovery policies. This allows us to discuss the well-known trade-off between unemployment and the public budget.

5The rest of the paper is organised as follows. Sections 2 and 3 provide a short discussion on the possible return of Keynesian economics as an important paradigm in macroeconomics. Section 4 presents the Danish response to covid-19. The empirical SFC model for Denmark in presented in section 5, while the results from simulations are shown in section 6. Section 7 then presents a discussion of the results of the model scenarios, and the paper closes with some concluding remarks.

1. Back to basics?

  • 2 For a critical assessment of the modern macroeconomic mainstream, see (Marchionatti & Sella, 2017).

6In the aftermath of the Great Recession, many felt a need for a revision of the modern macroeconomic mainstream – the New Neoclassical Synthesis (NNS), and their DSGE-models. Some heterodox economists anticipated a change in the presiding paradigm. A transformation from the classical understanding imbedded in NNS to a revival of solid old-fashioned Keynesian understanding, built on the framework provided by Keynes in his General Theory. Based on empirical evidence and critical voices even from within – (Romer, 2016) as an example – even mainstream macroeconomists saw the need for development. And as we know today, mainstream macroeconomics has put some changes on their agenda (see amongst others, Galí, 2018 and Christiano, Eichenbaum & Trabant, 2018)2.

  • 3 That aggregate demand has an important role to play also in the longer run is known in general by a (...)

7Furthermore, non-mainstream economists have to acknowledge that today’s mainstream seems to have adopted a sort of Keynesian understanding of the importance of the role played by aggregate demand (although for most probably only in the short run)3. This seems to be the lesson learnt from the years of the Great Recession and the current covid-19 pandemic. As such, the Spring 2020 issue of the Review of Keynesian Economics addresses the status of Keynesian economics and, according to the various contributions, Keynesian economics is alive and well, at least in so far as policy response is concerned in times of crisis (see for instance Rowthorn, 2020 and Eichengreen, 2020). At the same time, non-mainstream economists have been critizied for not paying enough attention to supply side issues, but focusing solely on aggregated demand. In the current crisis, both mainstream and non-mainstream economists seem to agree on the need to combine short run supply side policy with traditional Keynesian expansionary policy to avoid a recession on the one hand and accelerating prices on the other hand.

8Therefore, as (Palley, Caldentey & Verningo, 2020, p. 21) seems right to conclude: “We are living in a time which many believe has a distinctly Keynesian character… its legacy remains in place in the sense that discretionary counter-cyclical fiscal policy is back… Likewise, activist monetary policy is back.”

9However, a Keynesian perspective on economic policy is broader than just arguing for expansionary policies as short run responses to economic difficulties, such as in the current crisis. It also has to do with the right strategy for the longer run. This concerns questions, amongst others, of inequality, environmental sustainiability, the design of a well-functioning welfare state and the aim to achieve a satisfactory level of output, close to that of full employment. In this respect, the mainstream is hardly as Keynesian in its content as many Keynesians would hope.

2. The need for collectively funded, comprehensive social protection systems

10Stiglitz has advocated for targeted assistance to help certain people and sectors. He calls on us to appreciate once again the important role of government, public policy and public values:

“It is clearly a case where targeted fiscal policy is what is needed. It’s been true for a long while that monetary policy has had only have limited efficacy… (but) this is a different kind of crisis than normal crises. It’s just not a problem of aggregate demand… because of the disease, people are shutting down their businesses… more demand is not going to save that particular problem”. (Tan, 2020)

11Proper functioning of a stronger public sector requires more creative thinking and institutional engineering than is currently offered. In many countries, governments have increased financial support for households and businesses.

12The world is – according to the World Bank – in the midst of the greatest economic recession since World War II, significantly deeper than the financial crisis and the Vice President Ceyla Pazarbasioglu said: “Our first order of business is to address the global health and economics emergency. Beyond that, the global community must unite to find ways to rebuild as robust a recovery as possible to prevent more people from falling into poverty and unemployment” (World Bank, 2020). According to the UN 2020 SDG report (Jomo & Chowdhury, 2020), the pandemic is expected to push an estimate of 71 million people into extreme poverty by the end of 2020, the first increase in global poverty since 1998 (UN, 2020).

13On the positive end – unlike during and after the latest financial crisis – there is a significant focus on a green recovery for economies. A less positive point is that the governments probably presuppose an overly rapid recovery of their economies (this also applies in the Danish case). This professional optimism is, of course, politically motivated. If we projected a different economic cycle, the message to voters would be completely different. These optimistic projections have also been used by the Danish government as the basis for its latest climate action plan. Had an alternative economic cycle been chosen, the compression predicted for 2025 would probably have been greater.

14The covid-19-crisis makes it clear that not enough progress has been made. If the pandemic continues, it can be hoped that it will encourage governments to expand access to health care, sickness benefits and unemployment protection, since such spending has a greater positive multiplier effect on the economy than other measures. Primary sector workers are most exposed to the pandemic, as they often lack social and human rights and enjoy little or no social protection. This is the case for both those who cannot work (the unemployed) and those who do, but do not have guaranteed work or hours. Unfortunately, this extends to low-paid workers, with an overrepresentation of migrants, women and young people in certain sectors in the economy, such as cleaning, hospitality and retail.

15Governments must use the momentum created by the covid-19-pandemic to make rapid progress toward collectively funded, comprehensive social protection systems.

3. The Danish case study

16Like most other economies, the Danish economy has been hard-hit by the covid-19-virus. Denmark has had two pandemics. A smaller one in the spring and a larger one over the winter 2020/2021. All schools and public institutions, except for hospitals and doctors’ practices, were in periods simultaneously shut down. In the private sector, employers were asked to encourage their employees to work from home, unless their positions were essential for society.

17In order to avoid a catastrophic decline in economic activity and escalating unemployment, the Danish Parliament decided to introduce a number of aid packages on an enormous scale. This is backed by a government loan of DKK 300 billion.

18The implementation of aid packages are crucial in the very short-run to reduce the impact of the shock, while the economy requires a sustainable solution in the medium to long term. The question of whether the crisis will put the economy on a V-shape, U-shape or L-shape path is not entirely clear. This depends on a number of factors such as the health of the financial system and the degree of uncertainty in the economy. The decision to lock the economy down has increased the level of uncertainty in the economy, which may lead to a fall in private consumption as well as private investment. If there is prolonged uncertainty, consumer and business confidence will decline, leading to a prolonged recession.

19The response to the crisis in Denmark therefore seems to follow a two-stage strategy, as suggested by former chairmen of the Economic Council (Scheer, 2020): first to use compensation of business and workers to keep the factors of production intact. The main reason at this stage is to keep the production capacity of the economy as close to its pre-crisis level as possible. This should secure that production is ready to adjust to the expected increase in the aggregate demand. Second, more traditional fiscal stimuli are used in medium term to help the economy to recover from the covid-19 crisis.

20Our aim in the section that follows is to illustrate the need for discretionary counter-cyclical fiscal stimuli, combined with financial compensation as a tool to both minimize the initial adverse effects in 2020 and to promote a recovery from a recession. To do this, we use a Post-Keynesian empirical stock-flow consistent (PK-SFC) model, estimated for the Danish economy as presented in Byrialsen & Raza (2020).

4. The Model

21Before we present the results from the simulations, we will briefly discuss the properties of the model. Since the purpose of this paper is to discuss the need for counter-cyclical fiscal policy and not to provide a formal presentation of the model itself (A full description of the model can be found in Byrialsen & Raza (Byrialsen & Raza, 2020) and the model is presented along with a description of the variables in Appendices 1-3), the presentation is kept at a very general level. The structure of the model follows in the tradition of Godley & Lavoie (Godley & Lavoie, 2007), and other models that employ the Post-Keynesian SFC approach, and the behavioural equations are therefore based on Keynesian theory. This model, however, extends the Post-Keynesian SFC-model framework by integrating a labour market with restrictions on the labour force (labour supply constraints). This is done by adding a Leontief production function to the model, which places a restriction on the actual production capacity of the economy. The feature becomes especially relevant in the case of an economic lockdown, where both the participation rate of the labour force as well as the productivity are expected to be affected.

  • 4 The output from the econometric analysis can be provided on demand.

22Standard Post-Keynesian theory provides the guidance for the behavioral equations, while data determines the actual specification of the equations. The structural parameters in the model are estimated using annual Danish data from 1995-20164.

  • 5 The decision to employ or fire people is therefore closely connected to changes in the aggregated d (...)
  • 6 The consumption function is modelled in a standard way as function of disposable income and net wea (...)

23The model consists of the five institutional sectors; households, non-financial corporations, financial corporations, public sector and the Rest of the World. All production and employment take place in the sector of non-financial corporations, and the employment is positively related to the total production in the economy5. The income of the households consists of wages, surplus of the production, social transfers and income on capital, while the expenditures consist of consumption and investment in housing6.

24The income in the public sector consists mainly of taxes payed by the rest of the economy, while the expenditures consist of public investment, public consumption and net social transfers. The main flows on the real side as well as the financial side for a given year can be seen in the Sankey (Alluvial flow diagram) below:

Figure 1. Real transactions

Figure 1. Real transactions

Source: authors

Figure 2. Financial transactions

Figure 2. Financial transactions

Source: authors

  • 7 The two standard matrices: Balance sheet matrix and Transaction matrix can be found in the appendix

25As seen in the figure 1, the incomes and expenditures on the real side of the economy are relatively detailed in the model, which permits the identification of the channels through which a given shock to the economy is transmitted through the whole economy. Following the accounting principle behind the SFC-models, all flows out of a sector results in inflows of other sectors. As illustrated in the figure 2, financial assets are assumed to consist of three asset classes: equites, interest-bearing assets and pension assets. Aside from showing the transaction of the individual assets among the different sectors, the diagram also illustrates which assets belong to which sector, that is, which sector holds the asset for each counterpart liability (in net terms)7. The strategy in this model is to model the flows in the economy leaving the stocks as accounting identities, where the stock in the current period is equal to the sum of the stock in last period, capital gains and transaction in the current period. This consistent integration of stocks and flows on both the real and financial side of the economy enables the possibility of analysing the effect of covid-19 virus on both real side transactions (e.g. GDP, Investment and Consumption) and financial stock (e.g. public debt) in one comprehensive model.

5. Results

  • 8 We have also performed another scenario in the economy: a one-year (single period) shock, to identi (...)

26The purpose of the model is to analyse the possible effect of the covid-19 virus on the economy and discuss the effect of a counter-cyclical fiscal policy and financial compensation; a baseline scenario and two scenarios are therefore simulated in the model. In the baseline the economy is hit by covid-19, which has led to a rapid fall in real economic activity and an increase in the rate of unemployment. Deviations from this baseline are therefore identified as the effect of public intervention. This is similar to the procedure followed in a recent report from The Economic Council of the Labour Movement (AE, 2020), which builds on a scenario analysis from the Annual Danish Aggregate Model (ADAM)8. The simulation results below are compared and contrasted with this report and one conducted by the central bank of Denmark (Denmark’s Nationalbank) in the discussion that follows.

27In the baseline scenario the economy is expected to be affected in the following way by both a supply shock and demand shock, which is a combination of the negative impacts related to the covid-19 virus pandemic presented in Section 3, with no intervention by the public sector:

  1. The inputs of production have been hit on the labour side, since both the productivity and magnitude of the labor force have declined. This is as a result of lockdowns on production and reduced productivity of workers from home.

  2. A fall in demand for exports, the propensity to consume and propensity to invest.

28The two response scenarios develop progressively, are presented in greater detail below, and can be summarised as follows:

  1. Financial compensation to avoid a huge drop in income of firms (in order to avoid a huge drop in the demand for labour).

  2. Identical to scenario i), but in this scenario, the government introduces a expansionary fiscal policy.

29These two scenarios allow for the analysis of both the effect of the shock and the mitigating effects of the policies pursued by the public sector.

5.1. Baseline

  • 9 In the baseline scenario the value of the marginal propensity to consume is 0.513 and the propensit (...)

30In the baseline scenario, the Danish Economy is hit by a number of exogenous shocks, as presented above, in order to capture the possible negative impacts related to covid-19. From a supply-perspective the productivity of the labour-input and the labour force are lowered with 3%, which imposes a labour constraint on the economy, leading to a fall in production capacity. Firms respond to the crisis by reducing their demand for labour. Moreover, we assume the demand for Danish exports to be exogenously reduced by 10 percent. Since this scenario is assumed to affect both the behaviour of households and investment decisions by firms, the autonomous term in the consumption function is lowered by 5%, while the marginal propensity to consume is lowered to 0.4. In the investment function for firms, the autonomous term (often interpreted as ‘animal spirits’ to use the terminology employed by Keynes) is assumed to fall to -0.4 as a result of the sudden negative shock to the economy9.

31Hysteresis effects and a feedback mechanism to the exogenous shocks are integrated into the model after 2020. This is done by assuming that the shocks to follow an AR(1) process, with a parameter of 0.6, which creates a stationary process with a diminishing magnitude of the shock. With regards to the behavioural parameters this can, from a theoretical perspective, be explained by the impact of increased uncertainty generated by the shock on the economy. Particularly, of how long the virus is expected to take to circulate, or the extent to which vaccines can be administered across trade partners, and what the effects on current and future income will be.

32The table below illustrates the effects of this shock for the levels of GDP, the rate of unemployment, the public balance and public net wealth, both as a ratio of GDP.

Table 1. Baseline: real GDP, rate of Unemployment, public balance og public net wealth

Table 1. Baseline: real GDP, rate of Unemployment, public balance og public net wealth

Source: authors

33The level of GDP in the baseline falls by 10.7% from 2019-2020, which is much higher than the drop in 2009 of 4.9%. The hysteresis effect aggravates the negative effects of the shock through extended contractions in the propensity to consume, autonomous consumption and investment components of demand. This leads to a magnification of the negative effects of the fall in disposable income and a substantial fall in both real consumption and investment by households. The extended and magnified contraction of economic activity also has a corresponding effect on unemployment, as the rate of unemployment increases with 4 percentage points compared to the level in 2019. The high level of automatic stabilizers in Denmark creates a large deficit on the public budget of almost 4% of GDP in 2020 as a result of lower economic activity. In the medium term, the public balance turns positive as a result of the automatic stabilizers in the Danish Economy. As a result of the deficits, the public net wealth-to-GDP ratio falls by almost 4 percentage points compared to the level in 2019. The net wealth-to-GDP ratio improves after 2020, but is doesn’t reach the same level as in 2019 before 2024. This scenario illustrates the possible impact of the pandemic.

5.2. Scenario 1

34In this scenario the government intervenes in order to reduce the damage of the covid-19, by avoiding closures of firms due to loss of income, and reducing job losses by compensating firms in return for retaining employees. In this scenario these two initiatives are captured by:

  1. An increase in transfers to firms to compensate for the loss of income from the lower production in 2020. The size of this compensation is calibrated to 60% of the expected loss of production.

  2. An increase in transfers to firms to compensate for wages paid to keep people employed instead of firing them in 2020. This compensation is calculated as 60% of the expected drop in employment.

35The table below illustrates the effects of the intervention, for the level of GDP, rate of Unemployment, public balance og public net wealth.

Table 2. Scenario 1: real GDP, rate of Unemployment, public balance og public net wealth

Table 2. Scenario 1: real GDP, rate of Unemployment, public balance og public net wealth

Source: authors

36The intervention from the Government reduces the effect of covid-19 on both the short and medium term. The level of GDP in this scenario falls by 9.96% from 2019-2020, which is slightly lower than the damage in the baseline. The effect of the intervention is more striking if we consider the rate of unemployment. In 2020 it rises 2%-points compared to 2019, a much smaller increase compared to the baseline. As the compensation intervention is financed from the public budget, both the public balance and financial net wealth are affected negatively. Because the compensation only takes place in 2020, this scenario converges towards the baseline in the medium term.

5.3. Scenario 2

37This scenario builds on the same principles as Scenario 1, though but with one important difference. Government takes on the responsibility to dampen the negative impact of the covid-19 pandemic through the use of fiscal policy both in 2020 and in the medium term. As presented in section 4, the government has stimulated the economy in several ways. In this scenario, the economy is additionally stimulated by an increase in public consumption:

  1. An increase in the public consumption of DKK 50 billion (approx. 2% of GDP) in 2020.

38To analyse the effect of active fiscal policy, this reaction by government is also assumed to follow an AR(1) process with a parameter of 0.6, since the Government wants to use fiscal policy as a central part in the recovery phase to avoid too deep a stagnation of the economy like the situation in Denmark after the latest crisis.

39The table below illustrates the effects of the intervention, for the level of GDP, the rate of unemployment, the public balance and public net wealth.

Table 3. Scenario 2: real GDP, rate of Unemployment, public balance og public net wealth

Table 3. Scenario 2: real GDP, rate of Unemployment, public balance og public net wealth

Source: authors

40The effects of the actions taken by government clearly counteract the negative effects of the covid-19 pandemic, and the fall in real GDP from 2019 to 2020 is now reduced to 6.9%. The rate of unemployment is also at a lower level both in 2020, but also in the medium term. The 5-year multiplier of this expansionary fiscal policy can be calculated to 1.41, which can be explained by the openness of the Danish economy and high tax rates.

41This policy, determined by government, clearly has negative effects concerning the public sector budget, as expected, and results in a large government deficit (8.37% in 2020) and a corresponding fall in the net wealth-to-GDP ratio to 17.3% of GDP in 2020.

42In the following figures, the results from simulations corresponding to each scenario are illustrated and discussed:

Figure 3. Effects on GDP

Source: authors

43In figure 3, the development of real GDP for the three simulations in the period 2019-2025 is illustrated. While the fiscal compensation schemes (scenario 1) are most targeted at reducing the decline in economic activity in 2020, the increase in public consumption (scenario 2) both reduces the damage of covid-19 (to a larger extent) and speeds up the recovery phase to avoid another stagnation. As discussed above, since the compensation in scenario 1 is only being paid out in 2020, the baseline and scenario 1 converge in the medium term. This convergence can also be seen looking at the rate of unemployment. This rate increases in all three simulations in 2020, but both public interventions reduce the increase in the rate of unemployment significantly. In the medium term, the policy introduced in scenario 2 keeps the rate of unemployment at a lower level as seen in figure 4.

Figure 4. Effect on the rate of unemployment

Source: authors

44Since both intervention policies from scenario 1 and scenario 2 are financed from the public budget, they clearly have negative effects on the public sector budget, as expected, and results in a large government deficit and a corresponding fall in the net wealth-to-GDP ratio for the public sector as illustrated in figure 4 and figure 5.

Figure 5. Public balance-to-GDP-ratio

Source: authors

45Due to the automatic stabilizers the public balance is affected negatively by covid-19 as illustrated by the baseline-scenario. Because the increase in public expenditure in this model is not financed by taxes, the annual public budget is affected negatively in the short term. In the model, this budget deficit is financed by an increase in interest-bearing liabilities, which affects the net wealth position of the public sector negatively. Compared to the baseline, the choice of an active fiscal policy affects the public net wealth-to-GDP ratio negatively. Since debt is a stock, it captures the impact of cumulative processes in the model, and therefore it comes as no surprise that the net wealth-to-GDP ratio falls as a result of this policy. Compared to the baseline the ratio increases by 7 percentage points by 2025. Although this remains small relative to GDP, and accelerated economic activity reverses the trend within a relatively short window.

Figure 6. Effect on financial net wealth-to-GDP-ratio

Source: authors

46As seen from these simulations, due to the expansionary fiscal policy, both GDP and the rate of unemployment are actually improved when compared to the baseline in the medium term. As seen in the development of the public balance and public debt, this improvement comes at the cost of a deterioration in public finances. The suggested policies are a combination of compensation to firms during the pandemic in 2020 and an increase in the public consumption, especially in the period of 2020-2022, where the negative effects of the crisis are expected to be the largest.

6. Discussion

  • 10 This claimed is also confirmed by a simulation of the model used in this paper, but in order to avo (...)
  • 11 To keep the process stationary, the value of the parameter is assumed to be smaller than 1.

47The question of whether discretionary countercyclical fiscal policy is back in fashion is ever more relevant, as the global economy feels the impact of an economic shock, the magnitude of which is expected to be similar, if not greater than that of the Great Recession. A major difference between the Great Recession and now, as discussed above, is the acknowledgement of the need to introduce discretionary countercyclical policy, by both economists and politicians, in Denmark and in a number of major economies. The strategy in many countries appears directed towards the financial support of businesses and workers in the short run to compensate a loss in income. From a monetary policy perspective, this has been coupled with additional cuts in official interest rates – although with limited space to cut with rates already at or close to negative rates. Furthermore, the access to liquidity for the firms and households has been improved in order to avoid bankruptcy and forced sales. The principle goal of is to keep the production factors (both labour and the productive capital) ready to meet the expected surge in demand as the pandemic abates. This strategy is therefore not to avoid any fall in GDP as such, but to reduce the magnitude of it as far as possible. As seen by comparing the different scenarios of the simulation, the fall in GDP can be reduced significantly by increasing the social transfers given to firms to compensate for both the loss of income from the fall in production and the costs of labour retention. As claimed by the current and a former chairman of the Danish Economic Council, the size of the fall in 2020 is highly depended on whether the behaviour of private consumption and investment is affected or not (Wang, 2020)10. Despite the already mentioned uncertainty regarding the size of the shocks, the results from Scenarios 1 and 2 highlight the fact, that the Danish economy is expected to be affected in both the short and the medium term by the repercussions of the pandemic. The recovery phase presented in Table 1 and 2 is naturally depended on the choice of the value of the parameters in the AR(1) process. A smaller value shortens the time period for which the shock affects the system, while a large value has the opposite effect11. Despite this dependency on the size of this parameter, the overall results from Scenario 1 and 2 are robust.

48After the first stage, to keep factors of production intact (the aggregate supply), the second should be to secure a sharp and sustainable recovery of aggregate demand with the implementation of Keynesian countercyclical fiscal policy. The effect of this strategy is illustrated in the comparison of the results from Scenarios 1 and 2. The purpose of step one, to reduce the magnitude of the drop in real economic activity in 2020, can be observed in the substantial improvement of GDP from -11.75% in Scenario 1 to -7.57% in Scenario 2. The deviation between Scenarios 1 and 2 is dependent on the size of the fiscal stimulus, but the overall results are robust to different choice of stimuli. The second stage is also accomplished in the model, as illustrated by the relatively faster recovery of the economy in Scenario 2 relative to Scenario 1. In scenario 1, a negative gap persists in in medium term as argued by Blanchard and others (Blanchard, Cerutti & Summers, 2015). This result, one the one hand, clearly highlights the need for a countercyclical discretionary fiscal policy, especially during a crisis. On the other hand, it also raises the question about how to finance this policy and if this policy creates a risk to fiscal sustainability in the longer run. To answer this question, it is important to consider that the public budget is already under pressure during a recession as a result of automatic stabilizers. With relatively strong automatic stabilisers, as in Denmark, a powerful shock or crisis can drive substantial negative changes in the public budget and the corresponding level of public debt, as seen in Scenario 1. An evaluation of the cost of a discretionary policy should therefore take these aspects into account. As presented in Table 2, the effect of fiscal policy clearly affects the debt-to-GDP ratio of the public sector. Despite an increase in the debt-to-GDP ratio of around 13% points compared to the baseline scenario, this should not pose a risk to the financial position of the public sector in Denmark. The net debt of the public sector is almost gone, which provides Denmark with the opportunity to access both foreign and domestic debt at very favourable interest rates. The long-standing stability of the fixed-exchange rate regime, and sound public financial position also make Danish public debt an attractive substitute for Euro denominated debt instruments, which allows for greater proportions of debt to be domestically denominated. Furthermore, due to the persistent high level of excess domestic saving over domestic investment in the Danish Economy (positive current account since 1989), debt taken up in the public sector as a result of active fiscal policy is most likely to be held by the domestic private sector in Denmark. The current situation in the Danish economy creates good foundation for applying an expansive fiscal policy, as stated by two former chairmen of the economic council (Scheer, 2020). The Danish government appears thus to be in a position where it can choose, to paraphrase Keynes, to take care of unemployment and to let the budget take care of itself as the government revenues and expenditures benefit from the reversion of economic stabilisers.

49Since the use of expansionary fiscal policy seems like a necessity to reduce both the initial damage and to speed up the recovery of the Danish economy, the final point to discuss is the aim of the expansionary policies. In this respect, the ecological transition is argued to be a top priority for several reasons. From a health perspective, the covid-19 crisis must be seen as a humanitarian treat and it is most likely that such treats should be faced in the future. The focus on public investment in sustainable projects and the ecological transition are ways to reduce the risk of these events in the future. From a Keynesian point of view, the multiplier of fiscal policy is assumed to be high during economic crisis (the multiplier on an increase in public consumption is 1.41 in this model). It is however reasonable to assume, that the multiplier of public investment in the ecological transition is even higher. This means, that the actual effect of the expansionary policy may be even higher, if investment goes towards projects supporting ecological transition. Exploring the effect of investing in “green projects” requires a more complex model, as the model used in this paper does not allow for different investment strategies.

Concluding remarks

  • 12 In an analysis made by the Danish Central bank published on the 1st of April, the Bank estimates th (...)

50Once again, the global economy faces great difficulties. Just a decade since the start of Great Recession, from which many economies have yet to recover, the covid-19 pandemic has triggered a second collapse in global production. The full extent to the decline is presently unknown. There is additional uncertainty on the expected duration of the both the health pandemic, and the economic consequences. However, as the present analysis shows, presenting Denmark as a case, it could take years for production and employment levels to recover to pre-crisis levels. The analyses presented above is, as is necessary, subject to uncertainty of a fundamental or an ontological kind. This is the case for Denmark as for other countries. The estimated shocks are scaled to represent a middle ground between optimistic and pessimistic possible outcomes, and both the negative effects of the pandemic and the strength of the response could be stretched in either direction12.

51Be that as it may. It seems certain that all countries have implemented some form of policy to dampen the negative economic impact by the covid-19 pandemic, particularly of an expansionary nature. Similarly, many economists – mainstream and non-mainstream alike – advocate that present responses should further be supplemented by traditional fiscal stimuli to boost aggregate demand, and to bring economies back to or close to their potential output levels. Furthermore, the covid-19 recession may regenerate a most needed focus on the crucial importance for countries to have collectively funded and comprehensive social protection systems, as argued above. At least in this respect, modern times seems to be inspired by Keynesianism.

Haut de page


Arbejderbevægelsens Erhvervsråd (AE) (2020), « Coronakrisen vil koste tusindvis af arbejdspladser », AE Analysis, March. URL: [accessed on 12/08/2020]

Blanchard O., Cerutti E. & L. Summers (2015), « Inflation and activity – two explorations and their monetary policy implications », National Bureau of Economic Research, no w21726.

Byrialsen M. R. & H. Raza (2020), « An empirical stock-flow consistent macroeconomic model for Denmark », Levy Economics Institute, Working Papers Series (2020).

Christiano L. J., Eichenbaum M. S. & M. Trabant (2018), « On DSGE models », Journal of Economic Perspectives, vol. 32, no 3, p. 113-140.

Dimand R. (2020), « The much-exaggerated death of Keynesian economics », Review of Keynesian Economics, vol. 8, no 1, p. 36-45.

Eichengreen B. (2020), « Keynesian economics: can it return if it never died? », Review of Keynesian Economics, vol. 8, no 1, p. 23-35.

Galí J. (2018), « The State of New Keynesian Economics: a partial assessment », Journal of Economic Perspectives, vol. 32, no 3, p. 87-112.

Jomo K. S. & A. Chowdhury (2020), « covid-19 pandemic recession and recovery », Society of International Development, p. 1-12.

Marchionatti R. & L. Stella (2017), « Is neo-Walrasian macroeconom(etr)ics a dead end? An assessment of recent criticisms of DSGE models », Journal of Post Keynesian Economics, vol. 40, no 4, p. 441-469.

Palley T. I., Caldentey E. P. & M. Vernengo (2020), « Do current times vindicate Keynes and is New Keynesian macroeconomics Keynesian? », Review of Keynesian Economics, vol. 8, no 1, p. 21-22.

Romer P. (2016), The Trouble with Macroeconomics. URL:

Rowthorn R. (2020), « The Godley-Tobin Lecture: Keynesian economics – back from the dead? », Review of Keynesian Economics, vol. 8, n1, p. 1-20.

Scheer S. (2020), « Topøkonomer på stribe : Slip tøjlerne og brug den aktive økonomiske politik », Netavisen Pio. 19.03.2020. [accessed on 16/05/2020]

Skott P. (2016), « Public debt, secular stagnation and functional finance », from the book Macroeconomics After the Financial Crisis – A Post-Keynesian perspective, M. O. Madsen & F. Olesen (eds), Routledge, p. 20-37.

Summers L. (2014), « U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero Lower Bound », Business Economics, 49(2), p. 65-73.

Tan, H. (2020): « The coronavirus outbreak is a ‘different kind of crisis’ says Nobel laureate Joseph Stiglitz », CNBC. [accessed on 18/06/2020]

UN (2020), « Estimates of the impact of COVID-19 on global poverty », WIDER Working Paper 2020/43.

Wang L. F. & M. Bahn (2020), « Økonomer : Coronavirus rammer økonomien hårdt, men en ny finanskrise ser ud til at kunne undgås », Information, 13.03.2020. [accessed on 21/08/2020]

World Bank (2020), « Global Economic Prospects », June 2020. Washington, DC: World Bank.

Haut de page

Document annexe

Haut de page


1 Moreover, this kind of empirical evidence might give way to other perspectives on macroeconomics than that of the dominant mainstream understanding of the New Neoclassical Synthesis. The lesson learnt seems to be that “it is increasingly clear that the trend in growth can be adversely affected over the longer term by what happens in the business cycle” and that makes Lawrence Summers wonder “how satisfactory would the recovery have been in terms of growth and in terms of achievement of the economy’s potential with a different policy environment, in the absence of a housing bubble, and with the maintenance of strong credit standards” (Summers, 2014, p. 66-67).

2 For a critical assessment of the modern macroeconomic mainstream, see (Marchionatti & Sella, 2017).

3 That aggregate demand has an important role to play also in the longer run is known in general by any Post Keynesian; see e.g. Skott, 2016.

4 The output from the econometric analysis can be provided on demand.

5 The decision to employ or fire people is therefore closely connected to changes in the aggregated demand. The level of Investment to fixed capital ratio is assumed to be a function of the capacity utilization as suggested in most Post-Keynesian models.

6 The consumption function is modelled in a standard way as function of disposable income and net wealth. The investment function is assumed to be a function of disposable income, actual stock of housing, house prices relatively to cost of new houses and credit available.

7 The two standard matrices: Balance sheet matrix and Transaction matrix can be found in the appendix.

8 We have also performed another scenario in the economy: a one-year (single period) shock, to identify through which channels the combination of supply and demand shocks (covid-19) may propagate through the economic system. The effects of this shock is as expected: a strong fall in GDP in 2020 but a quick recovery already from 2021 due to the rebound-effect from the temporary shocks.

9 In the baseline scenario the value of the marginal propensity to consume is 0.513 and the propensity to invest is 0.02.

10 This claimed is also confirmed by a simulation of the model used in this paper, but in order to avoid too many scenarios, a scenario without changes in the consumption and investment behaviour is not presented in the paper.

11 To keep the process stationary, the value of the parameter is assumed to be smaller than 1.

12 In an analysis made by the Danish Central bank published on the 1st of April, the Bank estimates three runs ranging from a rather quick and smooth recovery to a more sluggish and slow recovery. As such, they give estimates between -3% and -10% in the Danish GDP for 2020.

Haut de page

Table des illustrations

Titre Figure 1. Real transactions
Crédits Source: authors
Fichier image/png, 147k
Titre Figure 2. Financial transactions
Crédits Source: authors
Fichier image/png, 71k
Titre Table 1. Baseline: real GDP, rate of Unemployment, public balance og public net wealth
Crédits Source: authors
Fichier image/png, 107k
Titre Table 2. Scenario 1: real GDP, rate of Unemployment, public balance og public net wealth
Crédits Source: authors
Fichier image/png, 110k
Titre Table 3. Scenario 2: real GDP, rate of Unemployment, public balance og public net wealth
Crédits Source: authors
Fichier image/png, 108k
Crédits Source: authors
Fichier image/png, 30k
Crédits Source: authors
Fichier image/png, 26k
Crédits Source: authors
Fichier image/png, 31k
Crédits Source: authors
Fichier image/png, 23k
Haut de page

Pour citer cet article

Référence électronique

Mikael Randrup Byrialsen, Finn Olesen et Mogens Ove Madsen, « The macroeconomic effects of covid-19:
the imperative need for a Keynesian solution
Revue de la régulation [En ligne], 29 | 2021, mis en ligne le 17 mai 2021, consulté le 18 juin 2024. URL : ; DOI :

Haut de page


Mikael Randrup Byrialsen

Aalborg University, MaMTEP, Department of Business and Management. Fibigerstraede 2, 9220 Aalborg East, Denmark;

Finn Olesen

Aalborg University, MaMTEP, Department of Business and Management. Fibigerstraede 2, 9220 Aalborg East, Denmark

Mogens Ove Madsen

Aalborg University, MaMTEP, Department of Business and Management. Fibigerstraede 2, 9220 Aalborg East, Denmark

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Search OpenEdition Search

You will be redirected to OpenEdition Search