Incorporating carbon taxes into a Value strategy at a stock level is equivalent to imposing carbon footprint constraints on the overall portfolio. Our new research explores this mathematical relationship and provides more clarity on the implicit assumptions behind decarbonizing investment portfolios.
Given the growing impetus to tackle climate change, it is seems inevitable that carbon emissions will be penalized more heavily in the future. While various emissions mitigating measures already exist, including fuel excise duties and emissions trading systems, the scope and effectiveness of these methods is currently limited. Thus, one can expect an increasing focus on carbon taxation as a mechanism to curb emissions.
From a corporate perspective, the cost implications of carbon taxes will likely become more punitive going forward. And in terms of investments, traditional valuation metrics – such as price-to-earnings (P/E) ratios – do not yet reflect the costs of carbon emissions. In anticipation of likely future developments, we examined the impact of incorporating a hypothetical carbon tax in valuation ratios.1
Factoring in carbon pricing results in downward adjustments in valuations to account for unrecognized liabilities
Our research is related to a recent stream of literature that modifies traditional valuation metrics by incorporating intangibles, such as knowledge capital, organizational capital and brand value.2 However, whereas these studies adjust valuations upwards by identifying unrecognized assets, factoring in carbon pricing results in downward adjustments in valuations to account for unrecognized liabilities.
Our research demonstrates that applying a carbon footprint reduction constraint to a value portfolio is mathematically the same as following a carbon-tax-adjusted value strategy. This means that a certain percentage of carbon footprint reduction corresponds to a specific level of carbon tax, and vice versa. This insight provides more clarity on the implicit assumptions behind decarbonizing investment portfolios, while it also gives us an economic interpretation to portfolio footprint reduction targets.
In our investigation, we analyzed a sample of developed market stocks for the period December 1985 to August 2021. We first compared a carbon-tax-adjusted value strategy and a value portfolio optimized with a carbon budget constraint, to illustrate the equivalence between the two approaches.
For the carbon-tax-adjusted value approach, we sorted stocks into five quintile portfolios based on their carbon-tax-adjusted earnings before interest, taxes, depreciation and amortization/enterprise value (EBITDA/EV) ratios. For solving the optimization problem with a carbon budget constraint, we used the SciPy linear programming functionality, where portfolio weights were restricted between 0 and 5/N, with N being the total number of available stocks.
Figure 1 displays the weighted average EBITDA/EV and CO2/EV levels of top quintile carbon-tax-adjusted value portfolios at different carbon tax levels, along with portfolios optimized with various carbon budgets (as a percentage of market carbon footprint), as at end August 2021. The dots, which represent optimal portfolios given a specific carbon tax and carbon constraint, lie on exactly the same curve. This illustrates that the maximum exposure to EBITDA/EV for a certain carbon footprint can be achieved by using either a tax or a portfolio constraint. This curve can be seen as the value-carbon efficient frontier.
We also examined the long-term characteristics of carbon-tax-adjusted value strategies. In one of our analyses, we evaluated how carbon taxes reduce the carbon footprint of a value portfolio. Figure 2 shows the percentage reduction in the carbon footprint of the top quintile portfolio, measured against the base case top quintile portfolio without a carbon tax, as well as the equally-weighted universe of all stocks.
We observed that the largest carbon tax effect occurred in the USD 10-100 range. For the top quintile portfolio, a carbon tax of USD 10 led to an 18% lower carbon footprint compared to the base case; a USD 50 tax resulted in a 39% lower carbon footprint; and a USD 100 tax amounted to a 49% lower carbon footprint. We also noted that carbon tax levels below USD 10 did not have a significant impact, while those above USD 100 had a progressively smaller effect on the portfolios.
In our research paper, we also scrutinized the impact of carbon taxes on the portfolios’ value exposure. We observed that the EBITDA/EV exposure of the top quintile portfolio was virtually unaffected by carbon taxes up to USD 50, and slowly began to decay after that. Even at a carbon tax level of USD 1,000, a large part of the base case EBITDA/EV exposure still remained.
Furthermore, we analyzed the impact of carbon taxes on returns. To do this, we sorted stocks into five quintile portfolios on their carbon-tax-adjusted EBITDA/EV ratios at the end of every month, and then computed the return of the five quintiles over the subsequent month. Figure 3 illustrates that the outperformance of the top quintile portfolio was effectively immune to carbon tax levels up to USD 100. At higher tax levels, the performance began to deteriorate, but it took a carbon tax of over USD 20,000 to fully wipe out the performance of the top quintile portfolio.
In conclusion, our research empirically found that carbon taxes up to USD 100, corresponding with a portfolio carbon footprint reduction of about 50%, had little effect on the characteristics and the performance of the long side of an EBITDA/EV value strategy. However, to increase the carbon footprint reduction to 70%, the assumed carbon tax would need to rise from USD 100 to about USD 5,000. Such a level does not seem realistic, and it would also have an adverse effect on the performance of the value strategy.
1 See: Blitz, D., and Hoogteijling, T., November 2021, “Carbon-tax-adjusted value”, working paper.
2 See: Park, H., October 2019, “An intangible-adjusted book-to-market ratio still predicts stock returns”, Critical Finance Review, forthcoming; Lev, B., and Srivastava, A., March 2020, “Explaining the recent failure of value investing”, working paper; and Arnott, R. D., Harvey, C. R., Kalesnik, V., and Linnainmaa, J. T., January 2021, “Reports of value’s death may be greatly exaggerated”, Financial Analysts Journal.