Efficient aggregation of panel qualitative survey data

Qualitative business survey data are used widely to provide indicators of economic activity ahead of the publication of official data. Traditional indicators exploit only aggregate survey information, namely the proportions of respondents who report 'up' and 'down'. This paper examines disaggregate or firm-level survey responses. It considers how the responses of the individual firms should be quantified and combined if the aim is to produce an early indication of official output data. Having linked firms' categorical responses to official data using ordered discrete-choice models, the paper proposes a statistically efficient means of combining the disparate estimates of aggregate output growth which can be constructed from the responses of individual firms. An application to firm-level survey data from the Confederation of British Industry shows that the proposed indicator can provide early estimates of output growth more accurately than traditional indicators.