| The goal of this paper is to demonstrate how one can compute, or explain, long-term stock return expectations (across the business cycle) using information from the credit derivatives market. We also show how stock portfolios formed based on these expectations outperform simple benchmark stock portfolios. We build our approach on a model suggested by Martin and Wagner (2016) that links stock return expectations and risk-neutral idiosyncratic stock return variances (SVIX indexes). While Martin and Wagner (2016) uses option-implied variances to compute the SVIX indexes we instead use credit default swap (CDS) implied variances backed out using the methodology described in Byström (2015, 2016); i.e. using implied variances that make empirically observed CDS spreads equal to model-predicted spreads. In addition to reflecting stock market expectations among the market participants in the credit market rather than those in the equity market, our approach has the advantage of allowing for a much longer-term focus than the equity market. If one uses ordinary call- and put-options, like Martin and Wagner (2016), the available option maturities limit the horizon of the expectation or forecast to a maximum of twelve, or perhaps twenty-four, months. Martin and Wagner (2016), indeed, look at horizons between one and twelve months. If one instead uses credit default swaps to back out the implied stock return variances then the available horizons are much longer. In most markets there are credit default swaps with maturities between one year and ten years and this allows us to back out stock market expectations at the same time horizons. Such long-term expectations and forecasts are obviously relevant for the strategic asset allocation of asset managers with long investment horizons such as pension funds and insurance companies. However, hedge funds and family offices also need to form long-term expectations on individual stocks. The literature on long-term expectations of individual stock returns is very limited though and the volume does not reflect the practical relevance and importance that real-life investors attribute to reliable long-term stock market forecasts. |