Empirical calibration of climate policy using corporate solvency: a case study of the UK’s carbon price support

ABSTRACT

Emission reductions improve the chances that dangerous anthropogenic climate change will be averted, but could also cause some firms financial distress. Corporate failures, especially if they are unnecessary, add to the social cost of abatement. Social value can be permanently destroyed by the dissolution of organizational capital, deadweight losses paid to liquidators, and unemployment. This article proposes using measures of corporate solvency as an objective tool for policy makers to calibrate the optimal stringency of climate change policies, so that they can deliver the least loss of corporate solvency for a given level of emission reductions. They could also be used to determine the generosity of any compensation to address losses to corporate solvency. We demonstrate this approach using a case study of the UK’s Carbon Price Support (a carbon tax).

Key policy insights

• Solvency metrics could be used to empirically calibrate the optimal stringency of climate policies.

• An idealized solvency trajectory for firms affected by climate change policy would cause corporate solvency to initially decline – approaching but not exceeding ‘distressed’ levels – and then gradually improve to a new ‘steady state’ once the low-carbon transition had been achieved.

• In terms of the UK’s Carbon Price Support, corporate solvency of energy-intensive industries was found to be stable subsequent to its introduction. Therefore, the available evidence does not support its later weakening.

KEYWORDS:

• Altman’s Z-score
• climate change policy
• corporate solvency
• policy calibration
• stranded assets

Acknowledgements

The authors would like to acknowledge the support of the Generation Foundation and the KR Foundation and also to thank the reviewers of previous drafts of this article.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. An example of this may be the 19 highly carbon-intensive coal plants currently at various stages of planning in the EU (Mathiesen, 2014).

2. In 2003 (well before the EU Emissions Trading Scheme’s launch in 2005) the UK government regulator Ofgem expected price levels to have risen to 33-49€/ton by 2010 (Ofgem, 2003), and the lowest price scenario for EU ETS permits forecasted for 2012 by McKinsey & Co was 20€/ton CO₂ (UBS, 2003).

3. Sensitivity to the business cycle would depend on the timeliness of financial reporting.

4. There are exemptions, mainly for small generators (<2WM) and plants that use coal slurry.

5. Energy price increases are most strongly felt in manufacturing, as energy expenditures are larger here than in any other industrial sector. However, within manufacturing there are a number of industries whose products are exceptionally energy-intensive. We examine the effect of the CPS on these sectors specifically.

6. Although one article (Luan & Lo, 2016) was found which estimated theoretical power plant LCOE increases due to the CPS.

7. More recent Altman’s Z-scores were identified such as Appiah and Abor (2009), but these failed to be both derived from EU firms and the private manufacturing sector.

8. Orbis classifies firms involved solely in the retail and wholesale trade of energy-intensive goods as in the same industry as those firms involved in manufacture. For our purposes this is an erroneous classification.

9. There are two differences between our methodology and Altman’s. These are that the Retained Earnings variable in X₂ and the Book Value of Equity variable in X₄ is proxied by the related variable ‘Shareholder Funds’ due to lack of data.

10. Listed firms were removed in order to ensure comparability. Researchers use different models and thresholds to categorise the solvency of listed and non-listed firms, and therefore it would not be clear how to integrate them together. In any event, only 6 listed firms remained after the selection criteria i–iii were applied.

11. This is done because we are interested in changes in solvency, not predicting which firms will go bankrupt.

12. Regression analyses yield the same results and these are available from the authors on request.

13. Discriminant Analysis was also run with a random sample stratified by country and industrial sector of 177 solvent firms on the 177 bankrupt firms. However, all coefficients on the five financial ratios from this exercise were unexpectedly negative, suggesting problematic sampling error due to this reduced sample size.

14. The EU ETS for instance used free permit allocations to affect this outcome, and countries imposing carbon taxes such as Finland, Sweden, and Norway all provided equivalent dispensations to their most vulnerable firms.

15. For instance, net of GDP changes, increases in the ‘Δ Unemployment Rate’ variable could in fact be proxying for lower wage costs to the firm.

Additional information

Funding

This research project would not have been possible without grants from the Generation Foundation and the KR Foundation.