Green is Good? An Empirical Analysis of Incentives in the Green Bond Market
Abstract
Financial innovation has generated many new ideas which has contributed to solve challenges
the world, as a community, is facing. Anything from sharing wealth by making the financial
markets more accessible for everyone to structured products that has diversification benefits.
Today, the world is facing a climate crisis which requires innovative thinking and through
that, new products incentivizing to green investments coping with climate change. This thesis
aims at quantifying whether a new financial innovation has succeeded in incentivizing
companies to focus on green investing. This financial innovation is termed green bonds. More
specifically, this thesis will study whether there exist incentives for riskier firms to utilize the
green bond market to a wider extent than their counterparts. The riskiness of firms will be
characterized by their credit rating prior issuance of the green bond.
We perform two analyses. First, we use ordinary least square regression to examine whether
riskier firms have issued more green bonds in the period 2013-2021. Second, we perform a
logistic regression estimating the probability of issuing a new green bond the next year as a
function of the yield spread around issuance for a company’s green and conventional bond.
For this second regression, we use a matching methodology to find pairs of green and
conventional bonds.
Our findings show statistically significant more issues from medium and upper-medium grade
firms. They issue, on average, 44.92% and 44.77%, respectively, more green bonds than high
quality-rated firms. On the contrary, no definite conclusion can be made for firms with noninvestment
grade. These results show that there is an indication of more issuances from riskier
firms, especially when considering the small sample size for non-investment grade firms.
Thus, we can certainly predict a different conclusion with a bigger sample. At least, we
cannot exclude an economic relationship between credit ratings and the choice of green bond
issuance. Furthermore, we find no evidence on a relationship between yield spread and the
probability of issuing a new green bond the next 365 days following a green bond issuance.