BoE paper proposes fix for flawed rate expectation estimates

A model augmented with additional data can help solve some of the problems with existing methods of estimating interest rate expectations, economist Simon Lloyd writes in a new Bank of England staff working paper.

Lloyd says typical dynamic term structure models (DTSMs) used for such estimates suffer from an identification problem, caused by insufficient data. This leads estimates to be “spuriously stable”, and to vary as the sample length increases.

In the paper, Lloyd proposes augmenting standard DTSMs, which are based on bond yields, with information from the overnight indexed swap market. OIS contracts should reflect investors’ expectations of how short-term interest rates are likely to move.

Lloyd finds the OIS-augmented model outperforms three other designs – a standard, unaugmented model, a bias-corrected model, and a survey-augmented model. He says the model manages to obey the zero lower bound for the US in a less computationally-intensive fashion than other approaches.