RDP 8810: The Relationship Between Actual Investment and Survey-Based Expectations 5. Conclusions

A simple model was developed to explain the relationship between actual investment and expected spending as measured by official surveys.

In general it was found that the relationship between survey expectations and actual investment outcomes can be well explained by surprises in key economic variables. The most important of these variables, particularly for equipment investment, is short-term interest rates but profits (for equipment) and demand factors (for construction investment) can also be important.

In the case of expectations collected closest to the reference period the relevant variables explain about three quarters of the variation in the realisation factors and in general correctly predict the direction of movements in investment. Errors do not appear to be serially correlated.

However, the usefulness of the model is related to the timing of the collection of the expectations data. The further the collection of the expectations are away from the reference period then the less surprises in key variables can explain the difference between the actual outcomes and expected values.

Further research in this area could usefully explore the relevance of a wider range of “news” variables. It could also refine the measurement of expectations for the explanatory variables.