ESG-Proxy Shocks and Cryptocurrency Returns: Reduced-Form Evidence for Green-Tilted versus Conventional Crypto Baskets

Abstract

This study examines whether an aligned ESG proxy is associated differently with the returns of green-tilted and conventional cryptocurrency baskets. We combine an annual France-based ESG score from Refinitiv Datastream with daily cryptocurrency prices spanning 12 December 2021 to 27 September 2023. Daily log returns are aggregated into two equally weighted baskets - a green-tilted basket (GC: ADA, XTZ, THETA, ETH) and a conventional basket (CC: BTC, LTC, ETC) - and then averaged within month, yielding monthly average daily basket returns and 22 monthly observations. Because the ESG proxy is annual, it is mapped to daily frequency using a stepwise (forward-fill) transformation and then averaged to monthly frequency; the empirical exercise is therefore interpreted in reduced-form descriptive terms. We estimate a time-varying parameter VAR and compute generalized impulse responses (GIRFs) and generalized forecast-error variance decompositions (GFEVDs), complemented by connectedness and minimum-variance hedge metrics. A one-standard-deviation shock to ΔESG generates an immediate negative return response for both baskets (h = 0: -0.003385 for GC and -0.002674 for CC), followed by rapid reversion toward zero. GFEVD results indicate that ΔESG shocks account for 24.0% of GC's forecast-error variance at H = 10, compared with 13.794% for CC. Connectedness at H = 10 highlights strong within-crypto spillovers, with CC's variance largely explained by GC innovations (69.187%). Pillar decompositions imply smaller Social effects and comparatively larger short-run responses to Governance innovations. Hedge ratios are close to one and cross-hedging reduces variance by 78%, indicating limited diversification between the two baskets. Overall, the aligned ESG proxy is more tightly linked to GC than to CC, although the evidence remains descriptive given the annual proxy and short sample.