By continuing on this site you have agreed to cookies being placed and accessed by this website. More information and adjusting cookie settings.

Robeco uses cookies to analyze your visit to this site, to share information via social media and to personalize the site and advertisements in line with your own preferences. By clicking on agree or by continuing on this site, you agree to the above. More information and adjusting cookie settings.

AGREE

Robeco uses cookies to analyze your visit to this site, to share information via social media and to personalize the site and advertisements in line with your own preferences. By clicking on agree or by continuing on this site, you agree to the above. More information and adjusting cookie settings.

AGREE

By continuing on this site you have agreed to cookies being placed and accessed by this website. More information and adjusting cookie settings.

Predicting bond returns with macro data

The Lux-o-rente fund of Robeco has grown to a size of more than 3.8 billion EUR. The fund offers the diversified return of global government bonds enhanced with market timing based on Robeco’s duration model. The model is able to protect the investor in the fund from negative returns when interest rates rise and to benefit from declining interest rates by having an additional duration exposure.

We continuously monitor the model and ideas to further enhance its predictive power for government bond returns. Over the years academics also have increased their efforts on this topic. One specific strand of the literature focuses on using macro-economic data to predict bond returns. Ilmanen argues that the stock market can predict future bond returns and is more effective than using statistical macro-economic data like the GDP growth. Duyvesteyn and Martens confirm the predictive power of the stock market since publication by Ilmanen.

Ludvigson and Ng use the general information in a large database of macro-economic indicators to predict government bond returns. More recently Cieslak and Povala use the long term inflation in a new framework to predict government bond returns with an amazing R2 of 53%! Thornton states in a recent working paper that a simple two factor model has the strongest out-of-sample predictions for bond returns.

Our research question is whether market or macro-economic data can better predict bond markets and why.  Related question is what the relation between these different sources of data is and to what extend these bond markets predictors are related to the two factor model of Thornton? Ultimate goal is to further enhance the predictive power of Robeco‘s duration model.

References
Cieslak, Povala, 2016, Expected Returns in Treasury Bonds, Review of Financial Studies, 28, 2859-2901.

Cochrane, J., and M. Piazzesi, 2005, Bond risk premia, American Economic Review, 95, 138-160.

Duyvesteyn, J. and M. Martens, 2014, Emerging government bond market timing, Journal of Fixed Income, 23, 36-49.

Ilmanen, A., 1995, Time-varying expected returns in international bond markets, Journal of Finance, 50, 481-506.

Ludvigson, S. and S. Ng, 2009, Macro factors in bond risk premia, Review of Financial Studies, 22, 5027-5067.

Thornton, D., 2016, Understanding the Predictability of Excess Returns, working paper.

Share this page:

Are you interested?

Let us know your motivation and send it together with your CV and list of grades to SQ@robeco.nl.