23/2/2013: Irish Knowledge Economy: Sources of Funding

In previous two posts, I have covered the broader trends for R&D spending in Ireland over 2007-2012 and more specific trends in terms of R&D-related employment. In this, last, post I will illustrate some trends in relation to R&D spend and activity by nationality of the firm ownership.

First, two charts:

The charts above clearly show that for indigenous firms (Irish-owned):

• Use of own company funds has declined in overall importance between 2009 and 2011, although the category still plays more important role in 2011 than it did in 2007 in funding R&D activities. This longer-term trend is most likely a result of a combination of factors at work: 1) reduced availability of credit and equity investment as the result of the crisis, 2) reduced emphasis in R&D on tangible IP that can be used to raise equity funding.
• Meanwhile, public funding (despite the fiscal austerity) rose in overall importance in 2007-2011 period, although 2011 result is showing some moderation in overall reliance on public purse sources for R&D funding. This is also consistent with the points raised above.
• All other sources funding share accruing to the Irish-owned enterprises is volatile (per second chart above), but overall these sources of funding are becoming less important to the Irish-owned firms (first chart above). It is unclear whether supply (bust financial system in Ireland and collapsed investment) or demand (firms struggling with already massive debt overhang and facing the prospect of multi-annual Government deleveraging) drives this. My gut feeling - both.

The contrast between the Irish-owned and non-Irish-owned enterprises is difficult to interpret outside the simple realisation that the latter are predominantly MNCs and as such have no difficulty in sourcing internal funds for R&D activities. Public funding for these types of enterprises is ca60 percent less important than for Irish-owned enterprises.

Table below summarises the data:

I guess the main lesson here is that we need to more aggressively stimulate the use of 'other' sources of funding for the Irish-owned enterprises. I have been speaking about the need for enhancing Ireland's tax system to increase use of employee equity shares as a major tool for raising funding for indigenous firms, especially medium-sized ones (see presentation here).

23/2/2013: Irish Knowledge Economy and the Labour Market

In a recent post I looked at some troublesome trends in the overall R&D spending in Ireland. As promised, here are some more details, with the employment levels and R&D spend breakdown by nationality of enterprise ownership. This data, unfortunately, only goes as far as 2011.

Here's the chart showing the spending by enterprise type (Small Enterprises (SE) with <50 all="" and="" as="" base="" categories="" category="" employees="" enterprise.="" enterprises="" for="" here="" is="" of="" on="" other="" p="" select="" specific="" spending.="" taken="" the="" total="">

As the chart clearly shows, the bulk of R&D spend is allocated to Labour costs. I wrote about this earlier, so no need to repeat. But time trend is interesting in all costs:

• The importance of labour costs is falling in 2009-2012 for Small Enterprises (from 61.6% to 53.4%) and is rising for all other enterprises.
• The importance of Purchases (defined as expenditure on land & buildings, payments for IP licenses, instruments and equipment purchases and purchases of software) is rising for SEs (from 9.7% in 2009 to 24.4% in 2011) and falling for all other enterprises (from 17.5% to 6.8%). 
• In comparative terms, SEs are spending nearly four times more on purchases than other enterprises.
• The above is consistent with general theme around the world: SEs require more inward purchasing, while larger enterprises carry out more in-house activities. In turn, this means that SEs must generate more value-added to offset higher costs associated with purchasing.
• Remarkably, there is much less difference across enterprises types in terms of spending on own in-house software development. This suggests that in-house development is not associated with cost-shifting by enterprises (reallocation of normal business costs to R&D activity category to reduce tax exposures).

In terms of employment generated / supported by the R&D spending, the chart below shows distribution across the core categories of employees:

Despite the fanfares around 'Knowledge Economy' jobs, the chart clearly shows that the numbers of R&D employees with PhD qualification - the basic level in modern science to engage in advanced research - has declined in 2009-2011 period by 8.6%, although it is still ahead of 2007 levels for the Industrial & Selected Services sectors. It also dropped in 2007-2009 and 2009-2011 in the Manufacturing sector, with 2009-2011 decline of 10%.

At the same time, 'Other Research Staff' numbers rose in 2009-2001 by 22.2%. 

This is probably consistent with the R&D activity shifting into ICT services sector, where share of PhD-led research is smaller and much of the research activity conducted is focused not on primary innovation, but adaptation, customisation, other incremental innovation. This is also consistent with much of the R&D activity in Ireland being secondary in nature - not patent-generating or new product development, but incremental improvement. The third potential factor driving these changes is possible expansion of collaborative work between academic institutions and producers.

A very interesting chart plots the sectoral R&D staff employment as a ratio to total R&D staff engaged:

It is very apparent that our flagship exporting sector, the 'Big Hope' for the 'Innovation Ireland' programmes - the agricultural sector is simply not engaged with much of R&D activity. The sector posts lowest share of R&D employment by far. Exactly the same holds for virtually every indigenous sector. The chart is extreme: MNCs-dominated sectors are clear leaders in R&D-related employment, domestically-oriented sectors are clear laggards. Remember renewable energy? We are supposedly filthy rich in inputs in the sector (wind, wave etc). We are also, supposedly, engaging in massive research in the area and have aggressive programmes to drive the alternatives energy sector into exporting electricity to the UK and selling know-how all around the world. However, even with the 'white elephant' project like e-cars from the ESB, sector employment of R&D personnel is simply inconsistent with the grotesque claims made about our alternative energy industry competitiveness or position.

Another worrisome fact is that for all the successes of the IFSC and international financial services in general in Ireland, the sector - a major source of innovation (good and bad) worldwide - is yet to put forward appreciable levels of R&D-related employment in Ireland. What is worse, while employment in the IFSC held up relatively well during the recent years, employment of R&D staff in financial services shrunk, relative to the levels of employment if research staff across the economy.

I will post on the breakdown of R&D activity by the company ownership (Irish v Non-Irish owned enterprises) later, so stay tuned.

20/2/2013: Irish R&D Spend 2011/2012 - Concerns >> Fanfares

CSO has recently published data on R&D spending in Ireland for 2011/2012. That's right: in the days of Big Data, Open Data, etc our 'Knowledge Economy' is operating in the environment where evidence is more than 13 months old. In fact, the reality is even more bleak: CSO data for 2012 covers only actual data on current spending, with capital spending covered by estimates. In brief, Ireland's pro-Knowledge Economy policy formation is backed by old and hardly impressive in scope data.

However, given we have nothing better to go, let's take a look at what the latest stats tell us about the Irish economy's R&D intensities. In what follows, I reference combined time series with both actual and estimated data points.

Overall, Total R&D Expenditure by all enterprises rose 5.49% y/y in 2012 to EUR1.96 billion. That's right, Irish economy is investing just 1.53% of its GNP on R&D activities. In 2009 that number stood at EUR1.87bn amounting to 1.41% of GNP. The miracle of the 'knowledge economy' or 'Innovation Ireland' is really quite feeble. In 2009-2012, therefore, the R&D spending rose 4.99% cumulatively.

However, the above growth is distributed unequally across a number of items of expenditure and types of enterprises:

1. 2009-2012 Labour Costs associated with R&D activities went up 15.18% (+8.45% in 2012 y/y alone), while total Current Costs rose 12.28% (+7.81% y/y in 2012).
2. 2009-2012 costs associated with Payments for Licenses on IP rose 356.84% (+0.51% y/y in 2012), while software purchases costs shrunk 47.3% (up 18.22% y/y in 2012). Meanwhile own Software Development costs incurred by all enterprises rose 148.01% in 2009-2012 period (up 0.03% y/y in 2012).
3. Total Capital Spending on R&D activities has declined in 2009-2012 period by 29.55% and was down 9.24% in 2012 in y/y terms.

In other words, there is some evidence of potential cost shifting via R&D credits onto workforce, away from physical investment, as well as evidence of re-orientation of our exports away from manufacturing toward services.

In terms of enterprises types (Small enterprises, SEs at <50 at="" employees="" enterprises="" large="" medium="" mles="" to="">50 employees):

1. SEs saw rapid growth in 2009-2012 in Licenses for IP costs (+3,997% and only up 0.87% y/y in 2012), followed by Software purchases (+112% on 2009 and up 29% y/y in 2012) and Software development by the company (+87% on 2009, but down 5.1% y/y in 2012).
2. SEs overall current spending rose 44.75% on 2009 in 2012 and 9.9% y/y, while their total capital spending rose 328.4% on 2009 in 2012 and was down 8% y/y.
3. Total R&D spending by the SEs rose 73.0% on 2009 and was up 4.9% y/y in 2012.
4. In contrast, for MLEs, the largest growth was recorded in Software development by the company (+174% on 2009 and up 1.65% y/y in 2012). There were significant declines recorded in all other categories, with a 86.5% drop on 2009 in payments made for licenses to use IP (also down 11.2% y/y in 2012), 64.9% decline on 2009 and 4.3% decline y/y in 2012 for Instruments & Equipment spending, and a 57% drop (on 2009) in Software purchases, although here there was a rise of 15.3% y/y in 2012. Total Capital spending on R&D by firms with more than 50 employees declined 65.8% on 2009 in 2012 and there was a drop of 10.8% y/y.

The above is consistent with the view that in 2011/2012 there was re-orientation in expenditure to either reduce labour costs and / or support services-focused sectors, away from traditional R&D spend on equipment, software and IP.

Table below summarises relative allocations to specific lines of expenditure by the types of companies:

For SEs I highlighted in color the areas of strength in the new data (green) and weaknesses (red). As can be clearly seen, Irish smaller enterprises are not at the races when it comes to overall investment and spending relating to R&D activities, with 26.5% of the total nationwide expenditure captured by SEa, although the good news is that this number has risen compared to 2007-2008 period. In particular, weak dynamics are present on the labour costs side. At the same time, Irish small enterprises tend to purchase more IP from outside (97.2% of total expenditure nationwide on IP purchases is by SEs) and  tend to develop less software in-house.

The above results show just how much more needs to be done at the SEs levels to drive forward knowledge intensification of the economy. At the same time, overall headline figure of 1.53% of GNP being spent on R&D related investment and expenditures is also a major, system-wide problem. It is even more egregious when one considers the fact that Ireland is the base for European operations of many major multinationals.

I will be blogging more on the analysis of the 2011/2012 figures in coming days, so stay tuned.

05/11/2011: Universities & Research: Europe v ROW

OECD recently released an interesting database on research and universities impact for 24 countries. Here are some insights.

First from the top, the US retained its absolutely dominant position in terms of high impact universities. The EU comes in as the second. The relationship between two in terms of specific categories of high impact instituions ('hot-spots') is plotted below:

Aggregating Medicine, Human Sciences and Sciences and plotting them against Social Sciences clearly shows that world-wide (within sample) and even excluding the US, there is a very strong positive correlation between the quality of Science-focused high impact academic centres and Social Sciences centres.

In fact, correlation between Sciences and Social Sciences hot spots numbers is 0.97 for full sample and 0.91 for sub-sample excluding the US. However, excluding UK and US, the correlation drops to 0.59. In my opinion, this strongly suggests that our policies, aiming at focusing in terms of building capacity in 'hard sciences' alone - the EU-wide and certainly Ireland-own agendas for research and development frameworks - is a misguided approach that ignores the important inter-links between two fields.

EU overall results in the charts above are significantly driven by the UK academic performance. Excluding UK from the EU numbers dramatically alters EU standing relative to the US:

Thus, overall, ex-UK, EU falls to the third place in global rankings in terms of hotspots, were it to be ranked as a singular country.

Here are some more detailed plots of sub-indices by more granular division of research areas: