Cyclical fiscal policy, credit constraints, and industry growth
GAICIL Insights
Austerity in a recession is usually judged by the jobs it costs this year.1 This paper counts the growth it costs for years after. Cyclical fiscal policy, credit constraints, and industry growth, by Philippe Aghion, David Hémous and Enisse Kharroubi, examines fiscal policy's effect on long-run industry growth. Industries relying more on external finance or with lower asset tangibility grow faster, in value added and labor productivity, in countries implementing more countercyclical fiscal policies.
The evidence applies the difference-in-difference methodology of Rajan and Zingales to a panel of manufacturing industries across 15 OECD countries over 1980-2005. Credit constraints, measured by external financial dependence or by the negative of asset tangibility in US industries, are interacted with country-level fiscal cyclicality. Measuring industry characteristics from US counterparts helps address endogeneity concerns.
The result is robust. More countercyclical fiscal policy enhances output and productivity growth more in financially constrained industries. It survives inclusion of structural macroeconomic variables such as financial development, inflation, average fiscal deficits, openness to trade and the net current account position. Either fiscal cyclicality's growth impact is comparable to or greater than that of structural features, or structural features work partly through fiscal cyclicality. The timing of the budget may matter as much as the structure of the economy.
The mechanism is intuitive. Financially constrained industries depend on external finance, which tends to dry up in downturns. Countercyclical fiscal policy stabilizes demand and funding, letting these industries sustain investment through the cycle. Without stabilization, constrained firms cut growth-enhancing investment during downturns, with lasting productivity effects. The recession ends; the forgone investment stays forgone.
AI industries share several features of the vulnerable profile. AI investment is growth-enhancing and often needs external finance, especially for young firms with limited tangible assets. AI-intensive activities rely heavily on intangibles such as software, data and skills, which are difficult to use as collateral. Such industries benefit particularly from countercyclical fiscal policy stabilizing conditions in downturns. A bank cannot repossess a training dataset.
Low tangibility tightens the link. Firms with fewer tangible assets face tighter credit constraints, and their investment is more cycle-sensitive. Fiscal policy leaning against the cycle can protect AI investment from recession cuts, sustaining long-run growth. A procyclical budget may quietly prune an economy's most promising branches.
The authors propose opening the black box of fiscal policy to see which budget components matter most. For AI, the question is important: public investment in research, education and digital infrastructure may have different cyclical effects than transfers or consumption spending. Which line items to protect in a downturn is an innovation question.
Policy implications are notable. Fiscal frameworks enabling countercyclical policy, building buffers in good times and supporting the economy in bad times, can enhance long-run growth in financially constrained industries. Procyclical austerity in downturns may harm these industries disproportionately, slowing productivity growth. Fiscal rules that force cuts at the bottom of the cycle carry an innovation cost nobody budgets for.
Macroeconomic and structural policy connect. Fiscal cyclicality may matter as much for growth as structural features of the economy. Policymakers fostering innovation should treat macroeconomic stabilization alongside structural reform.
For economies betting on AI, fiscal stability supports the technology's growth contribution. Keeping AI investment going through downturns, through stable funding conditions and supportive policy, helps economies capture AI's long-run benefits.
Seen whole, the paper's message is plain: macroeconomic policy affects long-run growth through its interaction with credit constraints. Countercyclical fiscal policy boosts growth most in industries depending on external finance or lacking tangible assets. As AI investment, intensive in intangibles, becomes central to growth, fiscal policy design over the cycle will help determine how fully economies benefit.
For finance ministries, the sting is this: in the next downturn, the AI startups with nothing to pledge but code and talent may be the first casualties of a procyclical budget. Fiscal rules written for an economy of factories may be mispricing an economy of intangibles.
Introduction
What are the effects of cyclical fiscal policy on industry growth? We show that industries with a relatively heavier reliance on external finance or lower asset tangibility tend to grow faster (in terms of both value added and of labor productivity growth) in countries that implement fiscal policies that are more countercyclical. We reach this conclusion using Rajan and Zingales׳s (1998) difference-in-difference methodology on a panel data sample of manufacturing industries across 15 OECD countries over the period 1980–2005.
Takeaway
In this paper, we have analyzed the extent to which macroeconomic policy over the business cycle can affect industry growth, focusing on fiscal policy. Following the Rajan-Zingales (1998) methodology, we have interacted credit constraints (measured either by external financial dependence or by the negative of asset tangibility in US industries) and the cyclicality of fiscal policy at the country level, and assessed the impact of this interaction on value added or productivity growth at the industry level. Using this methodology which helps address potential endogeneity issues, we provided evidence that a more countercyclical fiscal policy enhances output and productivity growth more in more financially constrained industries, i.e. in industries whose US counterparts are more dependent on external finance or display lower asset tangibility. This result appears to survive a number of robustness tests, in particular the inclusion of structural macroeconomic variables such as financial development, inflation and average fiscal deficits, to which one could also add openness to trade or the net current account position. This, in turn, suggests either that the growth impact of the cyclical pattern of fiscal policy is of comparable (or even greater) importance to that of more structural features, or that the effect of these structural features operates at least partly through their own effects on the cyclicality of fiscal policy. 4 3 The corresponding regressions can be provided by the authors upon request.1
Our analysis suggests at least three avenues for future research. A first is to open the black box of fiscal policy and investigate which component of government budget drives the relationship between industry growth and the cyclicality of fiscal policy. Appendix A below makes a first step in this direction by distinguishing between government expenditure and government revenues. A second question is whether the above analysis of the effects of fiscal policy countercyclicality on industry growth can be transposed from the fiscal to the monetary sphere of the economy. A positive answer to this question would be all the more important as presumably monetary policy can be more easily modified over time than fiscal policy, although transmission lags may be larger for monetary than for fiscal policy. Finally comes the question of the determinants of countercyclical fiscal policy, and especially the institutional features or arrangements that foster or prevent countercyclicality. Answering this question will shed new light on the ongoing debate about the relationship between growth and institutions.
Notes
- Philippe Aghion, David Hémous & Enisse Kharroubi, Cyclical fiscal policy, credit constraints, and industry growth, 62 J. Monetary Econ. 41 (2014), https://doi.org/10.1016/j.jmoneco.2013.12.003. Takeaway source: Conclusions, reproduced from a free author or repository copy. ↩
Cite as: Philippe Aghion, David Hémous & Enisse Kharroubi, Cyclical fiscal policy, credit constraints, and industry growth, 62 J. Monetary Econ. 41 (2014), https://doi.org/10.1016/j.jmoneco.2013.12.003
Header image: Port Credit Aerial view 2021.jpg by Canmenwalker, CC BY 4.0, via Wikimedia Commons.
