The insightful paper by Andersen (2024) provides a thorough examination of some key shifts in Denmark’s fiscal policy framework during the last three or four decades. Central to his analysis is the discernible transformation towards a “stability-oriented” approach, characterised by a pronounced focus on medium-term plans and long-term fiscal sustainability rather than short-term fine-tuning. This strategic shift has seen discretionary fiscal interventions reserved for significant shocks, such as the Financial Crisis and the COVID-19 pandemic. Otherwise, Denmark has relied to an increasing extent on automatic stabilisers to manage moderate economic fluctuations, underscoring the increasing confidence in the efficacy of these mechanisms. In this brief comment, I first attempt to make some qualifying statements about some of the points made by Andersen (2024) before turning to a discussion of the future of fiscal stabilisation policies in Denmark.
Andersen (2024) posits that automatic stabilisers in Denmark have weakened over the last 30 years, citing reduced marginal and participation tax rates as evidence of this weakening. However, it is important to keep in mind that Danish tax reforms during this period have generally focused on reducing tax rates while broadening the tax base. It is not clear that such reforms necessarily reduce automatic stabilisers. One way to shed light on this issue is to consult the computations of tax and spending elasticities (with respect to changes in economic activity) regularly calculated and published by the OECD. A comparison of the two most recent versions (Girouard and André, 2005; Price et al., 2014) yields little or no evidence that automatic stabilisers in Denmark have been declining. Corporate taxes offer an illustrative example, as Denmark has reduced its corporate tax rate notably over the period in question. Yet, the OECD finds that the output gap elasticity of corporate tax revenues has increased from 1.65 to 3.15 due to a more cyclically responsive tax base. I believe more evidence is required before we can draw the conclusion that automatic stabilisers in Denmark have generally weakened. Herein lies an important message for policy makers: It is possible to introduce reforms aimed at reducing tax rates without weakening the automatic stabilisers as long as the reforms broaden the corresponding tax base.
My second remark is mostly a call for further work. Andersen (2024) rightly points out the difficulties associated with fiscal stabilisation policy in real time, not least due to data uncertainty. He provides illustrative evidence for 2023 while leaving out a more systematic assessment of this issue. The existing literature has found some evidence that fiscal policy in OECD countries tends to be countercyclical based on ex-ante (or real-time) data but procyclical (or perhaps acyclical) when using ex-post (revised) data (see, e.g., Cimadomo, 2012; or Bernoth et al., 2015). It would be interesting to analyse whether up-to-date Danish data would yield a similar result.
My final short remark concerns the fiscal response to the COVID-19 pandemic. While Andersen (2024) does not focus on the fiscal policy response to the pandemic, I find it worth pointing out that this response was, effectively, an illustration of the benefits of the “stability-oriented” approach to fiscal policy cited above. During the pandemic, Denmark’s Finance Minister Nicolai Wammen promised to do whatever it would take to guide the Danish economy safely through the crisis, an approach made possible only by the country’s fiscal stance before it.
Andersen (2024) raises a number of concerns associated with the use of government consumption and investment as well as tax policies for fiscal stabilisation purposes in the future. I share these reservations. Spurred by this, the rest of this comment offers some reflections on the remaining options for fiscal stimulus in a deep recession (other than automatic stabilisers). Inspired by recent academic research and by experiences from other countries, I will consider three somewhat unconventional types of fiscal policy that may be attractive from a Danish point of view.
First, it is not entirely unfair to say that Denmark was lucky in each of the last two deep recessions: In 2009, the Special Pension (SP) savings were available and could be readily released with a view to stimulate private spending. In 2020, a similar situation arose with respect to the “Holiday Savings”, which were released at that time. Both of these had been accumulated for reasons largely unrelated to their potential role as fiscal stimulus, and the government found itself able to stimulate private spending at essentially no cost in both cases. This begs the obvious question of what to do next time. It appears wise to think carefully about this question well before the situation arises. In theory, one option would be for the government to impose systematic “recession savings” during boom times and release them when appropriate. In practice, this is unlikely to be a good idea for a range of reasons, including the distinction between idiosyncratic and aggregate uncertainty and insurance, the fact that anticipation effects may erode the marginal propensity to spend the money, and other potential precautionary effects. A better case can probably be made for direct cash transfers from the government to households. Various forms of this policy have been implemented in the last three recessions in the US, with some success (see, e.g., Parker et al., 2013). In addition, an empirical study of the SP release has found high marginal propensities to spend in Denmark, too (Kreiner et al., 2019). Finally, while such transfers require a certain amount of fiscal space, this constraint currently does not appear to be a major issue for Denmark.
Another type of unconventional fiscal policy is to impose a temporary reduction in value-added taxes (VAT) to induce households to bring consumption forward in time. In theory, such a policy essentially works like an expansionary monetary policy (e.g., D’Acunto et al., 2018). A recent empirical study documents large stimulative effects on consumer durables from a VAT reduction in Germany in 2020, indicating that this policy was largely successful (Bachmann et al., 2023). The obvious caveat when applying these results to the Danish economy is that Denmark produces far fewer consumer durables than Germany. As a result, a reduction in the Danish VAT may end up stimulating the German or Swedish economy rather than domestic economic activity. However, given the recent German experience, it would be wise to estimate the amount of “leakage” associated with such a policy in a Danish context (again, ideally, well before the next recession hits).
Finally, in recent work (Druedahl et al., 2022), my co-authors and I use an open-economy New Keynesian model with heterogeneous households and two sectors to study the policy response to foreign demand shocks. One key takeaway is that a fiscal devaluation – in the form of an increase in VAT combined with a reduction in the payroll tax or a higher employment subsidy – may be a very successful stabilisation policy under a fixed exchange rate regime. We show that conventional fiscal policies may stimulate the domestic non-tradeable sector but at the expense of the tradeables sector due to the associated appreciation of the terms of trade. Instead, a fiscal devaluation successfully mimics a monetary expansion, as it depreciates the terms of trade at the same time as stimulating the non-tradeable sector. Our findings, therefore, suggest adding fiscal devaluations to the standard fiscal policy toolkit in small open economies with a fixed exchange rate (or in a currency union).
In conclusion, Andersen (2024) provides a profound reflection on Denmark’s fiscal policy over time. While certain dimensions merit further exploration, the paper serves as a useful starting point for further discussion of the country’s fiscal toolkit.