«UK Economic Performance: How Far Do Intangibles Count? Rebecca Riley and Catherine Robinson March 2011 INNODRIVE Working Paper No 14. The research ...»
We begin with a standard Cobb Douglas function, where output is a function of labour (adjusted for quality/characteristics - QL), R&D expenditure (RD), capital (K), and intermediate inputs (M). In the case of our analysis, organisational capital workers are included, weighted by their wagebill share (equation 1; ORG).
Table 7 contains the results for the pooled random effects estimation of the production functions in order to construct the alternative, performance based measure of organisational capital. We estimate both with and without the organisational investment term and the GVA used for the dependent variable in both estimations is adjusted for the intangible assets. The first set of results we look at the productivity wage gap of organisational workers using the approach adopted in INNODRIVE, ignoring the suggest a relatively high marginal product for organisation workers, however, this approach ignores consistency in the treatment of organisational investment within the equation. In the second estimation which includes the organisational investment term to avoid double counting, ORG is removed from adjusted GVA. Relative marginal products still exceed wages but by less than the first column. The combined multiplier can be derived directly through this methodology and is 0.280, compared with the assumed 0.35 in the expenditure based model. Whilst this is a little lower, it broadly supports the initial estimate (an EU approximation rather than UK specific) and acts as a useful check on the plausibility of one of the assumptions made in the expenditure based approach. Combined with the estimate of relative marginal product, this yields a total multiplier of 0.33 as shown in Table 4. This is very similar the multiplier used in the expenditure based approach.
As well as a pooled regression, we estimate the production function within a panel framework, across time and broad industry groups. Table 8 contains the mean values of the estimated coefficients. These are generally consistent with the coefficients reported in Table 7 (first column) with the exception of ORG Asset, which is halved and tangible capital, which is almost doubled compared the pooled results. In table 9 results are presented separately by broad sector. We note that the marginal product of IT capital is largest with respect to business services and lowest in manufacturing. R&D capital has its lowest coefficient in other sectors (construction; utilities) and is highest in business services. With respect to organisation capital, the return appears to be lower in retail than in all other sectors. The relative productivity of organisational workers is highest in manufacturing and business services.
The performance based measure has a number of advantages, not least that the time profile appears more consistent with other studies. We compare the findings from both approaches – expenditure and performance based organisation capital in Figure 6. The performance based measure is more volatile than the expenditure based, peaking at 6 per cent of new value added in 2001, consistent with others findings. This compares with a share of only 4 per cent in the expenditure based measurement. Here we have measured only the organisation capital using this method, were we to estimate R&D this way, it is likely that R&D ratios 21 would be lower and therefore more consistent with existing studies. However, the performance based measure is perhaps also more all encompassing in the sense that it may also capture purchased intangibles as well as own account measures. In the pooled regression (table 7) we estimate the equation with and without the organisation investment term. When the investment term is excluded we see a much higher relative productivity, consistent with the findings in other innodrive countries. A caveat to our findings is the probability that our findings are likely to be affected by multicolinearity.
6. INTANGIBLE CAPITAL AND PERFORMANCEGiven the results we observe in the construction of the performance based measure of organisation capital, we would expect to find a positive association between firm performance and all forms of intangible capital. In order to evaluate the contribution to sector level performance of our measure of intangible capital, we aggregate our firm level data to an aggregate sectoral level and perform a simple growth accounting exercise. This enables us to compare our findings with other UK studies (Giorgio-Marrano et al, 2009; Haskel et al,
2011) which apply a similar methodology to more aggregate data. In our growth accounting
approach we follow a translog production function, and following Diewert (1976), in applying tornqvist index as an approximation of the Divisia index (Jorgenson et al, 1987), decomposing growth into its various components:
Where jt is the share of labour in value added (labour costs divided by value added) averaged over period t and t-1. In studies relating to the impact of ICT on productivity, this has involved quality adjustment of capital, accounting for substitution between new technology and traditional capital. Details of the derivation of intangible capital shares are presented in Görzig et al (2011) The growth accounting results are presented in tables 10 to 12. In the first panel, standard growth accounting results are presented, where intangible capital is not included. This is our 22 base-line check. Over the period, we see average annual growth rates in labour productivity of around 3.6 per cent per annum, consistent with other studies. MFP is estimated to be growing at around 2.3 per cent over the period and tangible capital at 1.4 per cent per annum. Incorporating intangibles using the expenditure based measures sees a slight average decline in annual labour productivity, which is concentrated in the second half of the period (2003-2006). MFP also falls relatively to the baseline, and we see a very small decline in tangible capital too. Intangible assets each contributes around 0.1 per cent average annual labour productivity growth over the full period, for R&D and organisation capital, this is a slightly larger contribution in the earlier period. In any case, the magnitude of intangible capital is as much as one third of tangible capitals contribution to labour productivity growth. If we consider the performance based measure of organisation capital we see more variation, with organisation capital having a negative effect on labour productivity growth in the later 2003-2006 period. Overall, however, there is little difference between the contribution made in the expenditure and performance based approaches. What is noticeable is the differential impact it has on labour productivity growth across the two periods.
Our findings bear some similarities to existing measures of intangible capital, particularly in terms of time profiles. However, we note again that our data capture own account intangibles and not purchased intangibles. In addition, ours is a measure of firm specific intangible capital whereas other estimates for the UK have been constructed at the aggregate level.
There are reasons also to suppose that our estimates would benefit from further refinement.
The growth accounting exercise here makes no adjustment for labour quality; therefore there is a distinct possibility that human capital (individual rather than firm specific knowledge) is conflating our intangible capital measures. Even if these two terms are not highly correlated, at the very least, by omitting this control, the unexplained residual, the MFP component, is likely to be overestimated.
7. CONCLUSIONS In this paper, we present the method of construction used to create firm level intangible capital stock measures for the UK. We have followed the emerging macroeconomic literature 23 with regards to a number of assumptions, subject to UK data limitations and idiosyncrasies.
Specifically, our measure captures own account intangible assets and not those purchased directly from the market. This component is crucially important given the firm specific nature of intangibles, which leads us to believe that much of it is untraded and therefore difficult to measure. Having constructed the data, we explore the plausibility of our data and test alternative construction methods. Using the growth account methodology and by aggregating our firm level data to national estimates, we find that intangibles contribute 0.4 per cent to average annual labour productivity growth (tables 11 and 12). The time profiles differ depending on the method of organisation capital calculation; when the expenditure approach is used, the contribution is more even across the two periods, the performance based approach suggests that in the earlier period, labour productivity growth benefitted more from intangibles, with less of a contribution in the second half of the period under consideration.
Our data reveal a number of insights. Firstly, we observe a positive association between intangible assets and productivity at the firm level. Whilst expected, this helps us to explain sources of firm level heterogeneity more robustly than previously possible. We also note a tendency for firms to bundle intangible assets, particularly ICT and organisation capital.
This is consistent with the findings of Bloom et al (2007) which found complementarity between organisational assets and IT investment. Thus we indirectly observe evidence of complementarities between these intangible assets. Moreover, there appears to be complementarities between intangible and tangible capital, consistent with the idea of capital-skill complementarity (Goldin and Katz, 1998).
We find that whilst our growth accounting results broadly correspond to other UK figures, there are discrepancies which are driven in part by different data sources and subsequent sectoral coverage. In addition, we need to emphasise that our study focuses on own account intangible assets and not purchased intangibles. It is also worth noting that the UK results differ from those of other EU countries, in part the result of different industrial structure since the UK is more heavily involved in the service sector compared to the EU average..
The main purpose of this paper has been to chart the construction of these data and illustrate some of its key properties. Whilst aggregate sector data have been constructed elsewhere, this is the first time that firm level data have been constructed to explore the role that own account intangible assets play in the production process. In this way, the data have considerable potential to provide further insight, not only in terms of the direct associations between intangibles and productivity (Riley and Robinson, forthcoming) and how different inputs in the production process interact but also with respect to indirect associations with productivity through regional or industrial spillovers. These are areas of on-going research but depend heavily on the plausibility of the data constructed here.
25 REFERENCES Abowd, J. M., Corbel, P. and Kramarz, F. (1999) „The entry and exit of workers and the growth of establishments‟, The Review of Economics and Statistics, 81, 170–87.
Basu, S., J. Fernald and N. Oulton (2004) „The case of the missing productivity growth or does information technology explain why productivity accelerated in the United States but not in the United Kingdom‟, in NBER Macroeconomics Annual 2003, Volume 18, ed. M. Gertler and Rogoff.
Barnes, M. and R. Martin (2002) „Business data linking; an introduction‟, Economic Trends, 581, 34–41 Bloom, N., R. Sadun and J. Van Reenen (2007) „Americans do I.T. Better: US Multinationals and the Productivity Miracle‟, CEP Discussion Paper, 788.
Bloom, N., R. Sadun and J. Van Reenen (2010) „Recent advances in the Empirics of Organizational Economics‟, CEP Discussion Paper 970.
Corrado, C., C. Hulten and D. Sichel (2006) „Intangible Capital and Economic Growth‟, NBER Working Paper 11948 Corrado, C., C. Hulten and D. Sichel (2009) „Intangible Capital and Economic Growth‟, Review of Income and Wealth, 55(3), 686-716.
Criscuolo C., J. Haskel and R. Martin, (2003) „Building the evidence base for productivity policy using business data linking‟, Economic Trends 600, 39-51
Criscuolo C. and J. Haskel, (2002) Innovation and productivity growth in the UK. Paper presented at ONS / DTI productivity workshop, 2 November 2002. Available online at:
http://www.statistics.gov.uk/cci/article.asp?id=246 Corrado, C., C. Hulten and D. Sichel (2006) „Intangible Capital and Economic Growth‟, NBER Working Paper 11948 Diewert, W. E. (1976) „Exact and Superlative Index Numbers‟, Journal of Econometrics, 4, 115
Giorgio Marrano, M., J. Haskel and G. Wallis (2009) „What Happened to the Knowledge Economy? ICT, Intangible Investment and Britain‟s Productivity Record Revisited‟, Review of Income and Wealth, 55(3), 686-716.
Goldin, C. and L. Katz (1998) „The Origins of Technology-Skill Complementarity‟, Quarterly Journal of Economics 113 (August): 693-732 Görzig, B., Piekkola, H. and R. Riley (2011) „Production of Intangible Investment and Growth: Methodology in INNODRIVE‟, INNODRIVE Working Paper No. 1.
Harris, R. (2002), „Foreign Ownership and Productivity in the United Kingdom - Some Issues When Using the ARD Establishment Level Data‟, Scottish Journal of Political Economy, 49 (3), 318-335 Harris R. and S. Drinkwater (2000) „UK Plant and Machinery Capital Stocks and Plant Closures‟, Oxford Bulletin of Economics and Statistics, 62, 239-261 Haskel, J., P. Goodridge, A. Pesole, G. Awano, M. Franklin and Z. Kastrinaki (2011) „Driving Economic Growth. Innovation, knowledge spending and productivity growth in the UK‟, NESTA Index Report, January, 2011.
Hellerstein, J. D. Neumark and K. R. Troske (1999) „Wages, Productivity and Worker Characteristics: Evidence from Plant-Level Production Functions and Wage Equations‟, Journal of Labor Economics, 17(3), 409-446.
Hulten, C. (1979) „On the „Importance‟ of Productivity Change‟, American Economic Review, 69, 126-136 Hulten, C., J, Hao and K. Jaeger (2010) „Macro versus Micro Comparisons of Intangible Capital: The Case of Germany and the US‟, COINVEST Working paper.
Jona-Lasinio, C., M. Iommi and S. Manzocchi (2011) „Intangible capital and productivity growth in European Countries‟, Innodrive Working Paper 10.