The behavior of green exchange‐traded funds

Abstract

Purpose The purpose of this paper is to investigate green exchange‐traded funds (ETFs) and propose a market‐wide proxy for green returns and a green volatility factor. Design/methodology/approach Identifying a unique sample of green funds, this paper investigates the time‐series behavior of returns for these investment vehicles and their associated conditional volatility dynamics via GARCH methodology. In this study, green ETFs are defined as index funds replicating market indices that invest in stocks exhibiting positive environmental, social, and governance characteristics. Findings Cumulative market‐wide green returns are found to be positive from inception year 2005 through 2008. Estimating a t‐GARCH (1,1) specification, the author finds strong evidence in favor of volatility persistence for the 15 green ETFs identified in this study. Additionally, the results suggest that a 1 percent volatility‐based value‐at‐risk forecast ranges from $24,150 through $26,000 on a daily basis. Finally, the empirical evidence provides support for weak‐form market efficiency when examining the green universe of stocks. Research limitations/implications Green ETFs are relatively recent financial instruments and exhibit important implications for volatility timing strategies and the cross‐section of green stock returns. Practical implications Knowledge of green price behavior is important in constructing optimal hedging and risk management strategies. Social implications Green ETFs provide a natural channel for sustainable investing. Originality/value Green ETFs exhibit the advantage of including many companies that have undertaken positive measures towards the global environment in their respective business models. This paper provides the first investigation of green ETFs.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 32 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.