FRANKFURT (Reuters) – Buying only green bonds as part of the central bank’s quantitative easing program would have only a small and temporary impact on pollution emissions, a European Central Bank study released on Monday showed.
The European Central Bank signaled it would consider shifting corporate bond purchases into sustainable debt to correct what it said was a market failure to price environmental risks. But the ECB’s own research shows that this “green QE” is only marginally effective.
Authors Alessandro Ferrari and Valerio Nispi Landi write: “The bottom line of our study is that temporary green QE can affect harmful emissions from the brown sector, But the impact on the pollution stock is minimal.”
“The main reason is that climate change and pollution are structural issues, while ad-hoc green quantitative easing … plays a role in the business cycle.”
The researchers first modeled a scenario in which central banks buy green bonds while simultaneously selling “brown” debt, leading to higher borrowing costs and lower output for more polluting firms.
They found that this would lead to lower emissions, but “has a negligible impact on overall pollution and on TFP (total factor productivity)”.
The study then simulated a “quantitative easing shock”, in which the central bank expands its balance sheet by buying either exclusively green bonds or both types of debt in proportion to their outstanding amounts, in line with the ECB’s own “market neutral” principle.
“We show that the differences between the two scenarios in terms of macroeconomic variables and pollution are quantitatively slight,” the researchers sadly said.
“Furthermore, even if QE were exclusively targeted at green bonds, its expansionary effect would boost brown production in the first phase, driving pollution slightly higher.”
The ECB has bought more than 300 billion euros worth of corporate bonds over the past four-and-a-half years, making it a driving force in the euro zone’s corporate debt market.
Reporting by Francisco Canepa, editing by Larry King
