Output, Investment and Capacity: An Empirical Investigation using Firm-Level Business-Survey Data in the United Kingdom

Mahony, Michael J (2022) Output, Investment and Capacity: An Empirical Investigation using Firm-Level Business-Survey Data in the United Kingdom. PhD thesis, University of Nottingham.

PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (11MB) | Preview


This thesis uses firm-level survey data to examine the decision-making of firms in order to gain greater insight into macrodynamics.

Chapter 2 examines the questions posed, the sample frame (i.e. details on the number and participation rates of respondents) and the characteristics of the firm participants of the Confederation of British Industry's (CBI) suite of business surveys. This dataset of firm-level survey responses is then matched to two external company accounts datasets (the Bureau van Dijk FAME dataset and the Office of National Statistics (ONS) Inter-Departmental Business Register (IDBR), including various ONS business surveys). Matching to external data sources often requires decisions to be made on how the matching should be conducted. Matching the CBI data to the IDBR yields a set of multiple matches when propensity-score matching is unable to select a definite match. Rather than dropping these firms from the sample, this chapter develops a decision rule to select a unique match from this set of multiple matches. Match results are around 50% when matching the CBI dataset with the Bureau van Dijk FAME dataset and around 90% when matched with the IDBR. However, match rates with the various ONS business surveys are lower than the corresponding match rates with the Bureau van Dijk FAME dataset (and in some cases far lower). Match rates are also reported for variation by geography, size and time-period. The matched dataset is then used in an illustrative exercise to examine the directional accuracy of firm output and employment forecasts. The results indicate the output and employment forecasts of firms in the manufacturing and mining and distributive trades sectors have value. However, this is not the case in either the service or financial services sector.

Chapter 3 introduces the new and novel meta-modelling quantification approach, which is used to produce quantitative industry-level measures of expected output growth, output disagreement and output uncertainty in the UK (using firm-level survey responses in the CBI dataset). This new quantification strategy provides more reliable estimates of expected output growth and output uncertainty compared to existing techniques such as the simple balance statistic (or the Anderson-Pesaran regression approach). These new quantified series are employed alongside actual output growth data in an analysis of the source of innovations and propagation mechanisms underlying output dynamics. These interactions are complex and out-of-line with those suggested by simple models embodying rational expectations. In addition, using a Beveridge-Nelson trends decomposition, this chapter shows there is a role for output uncertainty and output disagreement shocks in influencing business cycle dynamics - with these having relatively substantial effects of up to 4% in different sectors during the Great Financial Crisis (GFC), the sovereign debt crisis and the Brexit negotiations.

Chapter 4 extends the classic Abel (1981) paper to introduce capacity utilisation into a dynamic model with adjustment costs describing investment and hiring decisions of the firm. It provides an analytical solution for the theoretical model and then uses survey data from the CBI Industrial Trends Survey to test the model empirically. The results show that firms adjust their capital stock around a long-run equilibrium determined by sales over time. However, the speed of this adjustment depends on whether the model accounts for a capacity error correction term. Specifically, models which do not include a capacity error correction term overestimate the error correcting behaviour of firms, and imply a quicker adjustment speed of capital to its long-run equilibrium value. In other words, excluding capacity dynamics from an accelerator model of investment underestimates the time it takes capital to return to its long-run equilibrium value - providing an explanation for sluggish investment following recessionary periods.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Lee, Kevin
Mizen, Paul
Keywords: macrodynamics, macroeconomics, business surveys, CBI
Subjects: H Social sciences > HB Economic theory
Faculties/Schools: UK Campuses > Faculty of Social Sciences, Law and Education > School of Economics
Item ID: 68952
Depositing User: Mahony, Michael
Date Deposited: 01 Nov 2023 15:11
Last Modified: 01 Nov 2023 15:11
URI: https://eprints.nottingham.ac.uk/id/eprint/68952

Actions (Archive Staff Only)

Edit View Edit View