In the summer term 2026, we are offering a Master-level seminar with a focus on the use of generative AI in investing and asset pricing.

 

Sustainability-linked bonds (SLBs) are bonds whose coupon increases if the bond issuer fails to meet predefined sustainability targets. Unlike green bonds, they do not restrict the use of proceeds, but directly link financing costs to sustainability performance. Our new paper, “When do sustainability-linked bonds lower default risk? A correlation threshold”, is the first to examine how this design affects corporate default risk. The key intuition is as follows: if the sustainability performance tends to be good if the company is doing not so well – e.g. if carbon emissions are low if the company’s revenues decline – then SLBs can lower default risk because the avoidance of an extra coupon will tend to provide relief when an issuer is under pressure. We analytically identify conditions under which financing with SLBs lowers default risk because of this mechanism.

Many variables have been proposed for forecasting stock market returns, e.g., the dividend yield or the price/earnings ratio. However, most of these variables have no lasting predictive power. Combination methods seemed to provide a remedy here. Forecast combination involves averaging the forecasts made with individual variables. Sebastian Denk and Gunter Löffler have now shown that these combination methods cannot be relied on either. They do not provide any added value when forecasting the US stock market. Curious? The article appears in the Review of Asset Pricing Studies and is also available here.