«UK Economic Performance: How Far Do Intangibles Count? Rebecca Riley and Catherine Robinson March 2011 INNODRIVE Working Paper No 14. The research ...»
INNODRIVE WORKING PAPER SERIES
UK Economic Performance:
How Far Do Intangibles
Rebecca Riley and
INNODRIVE Working Paper No 14.
The research leading to these results has received funding from
the European Community's Seventh Framework Programme
under grant agreement n° 214576
UK Economic Performance: How Far Do Intangibles Count?
Rebecca Riley1 and Catherine Robinson2 1 National Institute of Economic and Social Research and LLAKES, London, UK 2 School of Business and Economics & WISERD, Swansea University, UK March 2011
JEL classification: M40, J30, O30, O40, M12, J24 KEYWORDS: Intangible capital, R&D, ICT, management, linked employer-employee data ________________________________
Acknowledgements: This paper is part of the INNODRIVE project financed by the EU 7th Framework Pro- gramme, No. 214576. Thanks to Richard Harris, Richard Upward and the MAUS team at ONS for making available useful data items, and to Geoff Mason, John Forth, Mary O‟Mahony, Andy Dickerson, Jonathan Haskel, and participants in INNODRIVE for comments and discussion of this work.
Disclaimer: This work contains statistical data from ONS which is Crown copyright and reproduced with the permission of the controller of HMSO and Queen's Printer for Scotland. The use of the ONS statistical data in this work does not imply the endorsement of the ONS in relation to the interpretation or analysis of the statistical data. This work uses research datasets which may not exactly reproduce National Statistics aggregates.
It has become increasingly clear in recent years that traditional productivity measurement, focusing on labour and capital inputs has been missing something. Specifically, research is turning to the significant expenditure on knowledge assets (such as R&D, software and marketing) which under the traditional national accounting framework is excluded from value added calculations, and as such is excluded from estimates of national income. Corrado, Hulten and Sichel (2009) (hereafter CHS) estimated that the exclusion of intangible investments from measured GDP from measures of capital stock in the US resulted in an underestimation of US capital stock of around $3 trillion (2003 data). Indeed, comparisons of UK and US estimates for the ratio of adjusted to unadjusted GVA figures reveal considerable similarities, particularly over the 1995-2003 period (Table 5, Giorgio-Marrano, et al, 2009) with a much smaller proportion of the growth for the US is attributable to tangible capital deepening, which experiences a bigger contribution from intangibles, compared to the UK.
Intangible assets by their very nature are difficult to identify. However, there is a general acceptance that they include a strong knowledge component which is difficult to measure in a market context. Hulten (1979) argued that such assets ought to be treated as investments rather than current expenditures. Moreover, whilst considerable research has focused on the aggregate importance of intangibles, it is evident that they are, to a large extent, firm specific inputs.
Much of the micro-analysis of intangibles in the UK and elsewhere has so far been relatively ad hoc; focusing on one or two aspects of intangible assets, such as R&D expenditure, innovation or management practices. Generally in these micro studies, investment in intangible assets is not capitalized. A second body of literature stems from the work of CHS, by Giorgio-Marrano et al (2009) which focuses on the macroeconomic impact of intangible assets.
Thus, given the firm specific nature of intangibles and the importance of capitalization of these assets, there exists a gap in current evidence for the UK.
3 The purpose in this paper is therefore threefold. Firstly, we provide new estimates of intangibles for UK firms. Our analysis is part of a wider European research effort and thus our approach is harmonised with others (e.g. Görzig et al, 2011; Piekkola, 2010). Secondly, we construct our intangible assets using both an expenditure based and a performance based measure. Thus, we can compare our findings from both approaches which offers a robustness check on the plausibility of our assumptions. Finally, our estimates of intangible capital are aggregated to be broadly nationally representative and incorporated into a growth accounting exercise to provide estimates of the effect this intangible capital mis-measurement has on productivity growth estimates for the UK. We compare our results with findings from other UK studies that have carried out similar exercises at an aggregate level, putting our firm level estimates into context. The period covered in our analysis is 1998-2006. This is dictated by data availability but covers the longest period possible.
The paper is organised as follows, we begin with a discussion of the definition of intangible assets and provide an overview of existing estimates of the magnitude and impact of intangibles that are currently available. In section 3, we present a detailed discussion of data sources we use to construct occupationally defined intangible assets. Section 4 provides a more information on the construction of intangibles for UK firms. In section 5 we illustrate the properties of firm level intangibles and include these within a growth accounting framework. In section 6 we include our measure of intangible assets within a growth accounting framework, providing estimates of how big an impact intangibles have on measures of growth for the UK and compare our findings to existing aggregate estimates. Section 7 contains our initial conclusions and directions for further research.
2. INTANGIBLES: DEFINITION AND EVIDENCEIt is hard to be precise about the definition of intangible assets, which by their very nature are difficult to identify, trade and indeed see. The academic literature offers a variety of broad and more narrow definitions of what should and should not be included, however, following the seminal work of CHS(2006) the literature tends to agree broadly on three main sources of intangible capital (CHS, 2006; Giorgio Marrano, Haskel and Wallis, 2009; Haskel
et al, 2011; Jona-Lasinio, 2011). These are (CHS,2006):
1. Digitized information This is often measured as ICT capital, composed of software as well as databases.
2. Innovative Property This includes both scientific R&D and non-scientific R&D. By which we mean R&D into social sciences and humanities, mineral exploration, new motion picture films and other forms of entertainment, new architectural and engineering design and new product development in financial industries.
3. Economic Competences Such as brand equity, including advertising and marketing expenditures; market research.
This category also includes firm specific resources, including human capital (investments in training) and organizational structure (management).
Intangibles, some have argued are not easily verifiable, are not always visible, ma by non-rival in consumption (and thus display elements of public good) and (as with R&D) it is not always easy to fully appropriate the returns. Despite these problems, fundamentally, any input that reduces current consumption so that future consumption increases, qualifies as an investment and should be treated as such (CHS, 2009). Corrado, Hulten and Sichel are largely credited with developing the current „best practice‟ methodology on incorporating a wider definition of intangibles into the national accounts methodology. However, it is clear that with such asset, the approach to measurement is crucial.
In their macro analysis of the US, CHS make a number of assumptions in order to measure and capitalize intangibles, some of which we discuss in our methodology below. Their analysis highlights that by ignoring intangibles labour productivity growth is almost certainly mismeasured. Intangibles are estimated to account for around 26 per cent of the growth changes, on a par with the size of the tangible capital component. They also find that the unexplained multi factor productivity (MFP) component falls as the explanation of growth improves. Inclusion of intermediates as a new capital input does not alter the acceleration in labour productivity that the US experienced in the mid 1990s. Giorgio Marrano et al (2009) adopt a largely similar approach for the UK, constructing a measure of intangibles for the 5 UK over the period 1990-2004. They find that, similarly to the US, nominal business investment in intangible assets has grown over the period, increasing from around 6% in 1970 to around 15% in 2004. Intangible investments are estimated to be roughly equivalent in value to tangible investments. Broadly, their findings mirror those in the US (albeit, to a lesser extent) except in the fact that the slowdown in labour productivity growth in the mid 1990s is largely accounted for by the exclusion of intangibles from national accounts. Both papers, note that their measure of economic competences is far from perfect but nonetheless, their research contributes significantly to the debate.
Despite the fact that most of the economic studies so far have constructed aggregate, macroeconomic data, intangibles are embedded in the firm and in all these sources, perhaps the most constant source of intangible assets is knowledge. There have been a number of firm level estimates of intangibles however these stem from the accounting literature rather than economics. Lev et al (2009) highlight the contribution that accurate measurement of organisation capital can make to explain the discrepancy between market and book values of firms and to the generation of abnormal profits. They identify 4 types of firm level intangibles;
learning (R&D), customer related (brands), human resources (training) and organisation (business processes). Developments in the latter, they argue, are still in their infancy. Hulten et al (2009) have recently attempted to relate macro measures of intangibles to micro estimates for the US and Germany focussing on a small samples of multinationals. They find German firms to be more R&D intensive than their US counterparts, but in terms of overall intangibles were less intangible intensive because of high US organisation capital intensity particularly.
In this paper, we follow the lead of Piekkola (2010) in the measurement of firm specific intangible capital. Corresponding to the three main „assets‟ discussed above, we identify three groups of workers that are instrumental in determining intangible assets within a firm and thus these form the basis of our asset types in the construction of intangible capital. Organisation capital incorporates highly skilled management and marketing workers but in which we would also include social scientists. Research and development is our second category of workers which would incorporate all science based research, including architects. Our final 6 occupational category of worker is ICT personnel. Whilst previous growth studies have looked to incorporate a measure of ICT and R&D into their production function, the measure of economic competences, captured by organization workers is a relatively new addition to the literature.
In related microeconomic literature, Bloom et al (2010) review the existing evidence of the role that organizational factors play in accounting for productivity dispersion within sectors and between countries. Again, they perceive there to be a measurement error in the production function where organizational factors are not adequately accounted for. Their review concentrates on management quality and decentralization within the firm. Whilst citing the lack of usable data for such analysis, they do draw some conclusions from recent empirical analysis. They note that larger, more skilled and more globally engaged firms are better managed and more decentralized and it is implied that these factors are correlated with higher productivity. The more competitive the market in which firms operate also affects management quality and the extent of decentralization. The propensity to engage in IT use and in heterogeneous sectors are found also to be more decentralized (p.34, Bloom et al, 2010).
Thus, we see that it is not only within the growth accounts literature that the production function is perceived as mis-measuring inputs. We hope to offer further insight with our firm level measures of organizational capital derived from estimates of occupational structures within firms.