2015 07 09 OECD: the future of productivity

Editor's Notes

• #3: Methodology: -- this analysis marshals a wide-range of statistics and varying methodological approaches, most importantly firm-level data, to analyse this problem -- It is important to recognise that currently available firm-level data sources are not ideal, particularly for analysis of the productivity dynamics of laggard firms. -- to compensate, the report relies on a mix of critical review of existing evidence, descriptive analysis and when possible, firm level econometric analysis to try to provide insights. -- Nevertheless, it is reassuring that the results from policy analysis using incomplete firm level data are often confirmed by analysis using official industry level data.
• #4: background and a reminder as to why we care so much about productivity growth
• #6: Productivity growth slowed in many OECD countries even before the crisis Concerns of a structural slowing in productivity growth.
• #7: Looking back over the last 4 cycles since 1973 (green), 1981 (red), 2000 (orange): the current cycle, dated to 2008 (grey) has been distinguished by lacklustre investment Notes: Gross investment, PPP weights, data are for OECD countries for which the breakdown of investment is available.
• #8: Knowledge-based capital (KBC) which has been more resilient during the downturn has declined since 2007 as well. Notes: Sourced from Corrado et al. (2012) … KBC includes R&D, firm specific skills, organizational know-how, databases, design, and various type of intellectual property.
• #10: Which is where the new OECD work that I will present fits in: Analysis led by Chiara Criscuolo (OECD STI) and Dan Andrews (OECD ECO) find evidence to support both views
• #11: While Gordon is right about the average which for manufacturing is rather lacklustre, at 1.7% since 2000; Brynolffson is right about the “most advanced firms” whose productivity is robust at 3.5%; It is simply that the average is dragged down by the “other firms” at 0.5% This bifurcation is even more pronounced in services where the average is a low 0.3% while the global frontier is 5.0% and the “other” category is actually negative. Note that this divergence between the global frontier and the rest has grown over time.
• #12: 11
• #13: 12
• #14: The paper examines 3 mechanisms for stimulating diffusion
• #15: Let’s assume that MFP growth at the frontier is 2% -- roughly the same as the US in the 2nd half of the 1990s This Figure presents estimates of how the benefits of this productivity growth at the global frontier diffuse to countries, depending on differences in the policy setting for an array of policies. For example, countries that trade very intensively with the frontier economy (e.g. Canada with the US) would realize 0.35 percentage points higher productivity growth per annum from more rapid diffusion, compared to a country with fewer such trade linkages (e.g. Austria). Similarly -- higher efficiency of skill allocation, R&D investment and managerial quality have similar effects and these gains are economically significant I will use this framing...trade, reallocation and KBC…to draw out some policy implications and directions for future work Notes: The chart shows how the sensitivity of MFP growth to changes in the frontier leader growth varies with different levels of structural variables. The diamond refers to the estimated frontier spillover effect associated with a 2% MFP growth at the frontier around the average level of the policy. The label “Minimum” (Maximum) indicates the country with the lowest (highest) value for the given structural indicator in a given reference year. For example, countries that trade intensively with the frontier (Canada) would realise 0.35 percentage points higher productivity growth per annum than a country with fewer linkages (Austria).
• #17: 16
• #18: But if you focus on traditional gross trade you are not getting the right picture. OECD TIVA For example: China appears to be the world beater in terms of gross exports of ICT & Electronics (dark blue) = traditional measure, but in terms of value-added (light blue) it falls behind the US.
• #19: Also GVCs and the connections that they propagate are an important vector for innovative collaboration: Those firms that are “innovative” report that collaboration with up-stream suppliers or down-stream clients are the most important source…
• #20: 19
• #21: Pushing the frontier also requires enabling experimentation with radical new technologies and business models. Since innovation is about trial and error, failure needs to be recognised as an opportunity to learn and rebound, rather than being seen as the end of the game. Thus, the policy environment should enable successful firms to grow, but also let weak firms exit the market, so that scarce resources can be released to underpin the growth of the successful ones.
• #22: The average productivity gap that some countries face can be due to two different factors: Country A is the Global Frontier to which all aspire; Country B and C have the same average Productivity, but the distribution across firms is very different – leading to different policy prescriptions Country B – has a problem of a large gap between the National Frontier and the Global Frontier reallocation is not the central problem but lack of innovation in general; Country C – which has the same average productivity as B, has a problem of weak market selection bad firms are surviving and locking in resources which should be re-allocated to more productive firms
• #23: 22
• #24: 23
• #25: This likely reflects the inability of firms to grow to scale and the fact that low potential firms can survive when market selection is weak, with significant cross-country differences emerging in the relative sizes of old and new businesses. Notice, while French firms enter at almost twice the size of US firms (the blue bar), after 10 years (the square), the size of the firms in the US are basically twice as big; this may be because many French regulations become binding at 50, and incentives are withdrawn Why does it matter? Because many attributes of being a productive firm come with the additional resources that are commensurate with “scale”
• #26: One example – the adoption of Enterprise Resource Planning software (ERPs) like SAP which are the glue that allows SMEs to interconnect with GVCs which have been shown to be a productivity enhancing characteristic. But the adoption of ERPs are not easy – they can be complex and require specific skills that many SMEs cannot afford. Not surprisingly SME uptake lags – red line – only about 25% of 10 to 49 employee sized firms have ERP these firms lack the resources While closer to 80% of large firms (250+); blue line But what is interesting is that the 50-249 (medium – green line) are at 60% – scale matters.
• #27: More generally, those firms that are innovative (patents increase) are more successful at attracting capital (resource re-allocation) when good “framework policies” are in place:
• #28: Both under and over skilling Underskill = shortage; Overskill = allocation problem, underutlised HC; High rates of skill mismatch often coincide with the presence of many small and old firms. For example, skill mismatch is estimated to account for up to one-fifth of Italy’s labour productivity gap with the United States. Yet, the research in this book suggests that around one-quarter of workers report a mismatch between their skills and those required to do their job. A better use of talent could translate into significant labour productivity gains in many OECD economies. Around one-quarter of workers report a mismatch between their skills and those required to do their job. A better use of talent could translate in up to 10% higher labour productivity in some economies. However, the most advanced firms often have productivity levels close to the global frontier, but their aggregate impact can be muted if they are small. High rates of skill mismatch emerge as a key constraint on the growth of innovative firms. On average across countries, roughly one-quarter of workers report a mismatch between their existing skills and those required for their job – i.e. they are either over or under-skilled – but this figure is closer to one-third in Italy, Spain and the Czech Republic (Figure 4). Over-skilling is generally more common than under-skilling, with the average likelihood of being over-skilled roughly two and a half times greater than that of being under-skilled. When firms draw from a scarce and fixed pool of skilled labour, trapping highly-skilled labour in relatively low productivity firms can make it more difficult for more productive firms to attract the workers necessary for their expansion. This could close about one-fifth of the gap in labour productivity between Italy and the United States (or Sweden).
• #32: 31
• #33: 32
• #34: 33
• #35: Firms at the global frontier are more likely to be part of a MNE group than other firms. This suggests that global co-ordination policy measures might be relevant