Proinnsias Breathnach: One thing that has always struck me about Irish economists is that, despite the importance of foreign direct investment and international trade to Ireland’s economy, they actually know very little about the activities of the transnational firms in question, or the structure of Ireland’s foreign trade and, above all, about the factors which attract foreign firms to Ireland and allow them to use Ireland as a base for serving external markets.
Rather than conducting real research in these areas, most economists who write on these topics appear to draw on undergraduate textbook models of how markets operate – models which in turn were originally devised to explain the kinds of competitive markets for commodity-type products (clothing, food, furniture) which were fairly typical of the British and US economies in the 19th century. Subsequent developments, such as industrial concentration, globalisation, rising living and educational standards, advertising and marketing and technological change, appear to have bypassed many of these people entirely.
Thus, a few weeks ago, we had Anthony Leddin of the University of Limerick (writing in The Irish Times) postulating trends in Ireland’s foreign trade which anyone with knowledge of this area would have realised right away were completely wrong. More recently (January 28), and again writing in The Irish Times, former Central Bank Chief Economist Michael Casey wrote: “At present the only bright spot in the economy is the output and exportation of pharmaceutical products”. This is an extraordinarily uninformed statement for a person of this status to make. Of the nine broad product categories into which the CSO divides Ireland’s merchandise exports, eight experienced growth in nominal export value in the first ten months of 2010 compared with the same period in 2009. Of the total growth in these eight categories, pharmaceutical products accounted for less than half (46.5%).
The growth in total merchandise exports, in turn, accounted for only one half of the overall growth in exports (including services) in the first three quarters of 2010. The growth in exports of computer services in this period exceeded that in pharmaceutical products by 24 per cent. Many economists have been unable to internalise the fact that services exports exist at all, never mind that they have been the main growth sector in Irish exports for many years and, in 2009, accounted for 46.5% of total exports. In the five years to 2009, exports of both computer services and business services grew much more strongly than exports of pharmaceuticals. In 2009, exports of both computer services and business services exceeded exports of pharmaceutical products in value terms.
In that year, these three sectors, along with organic chemicals, accounted for over half of Ireland’s total exports. If we throw in insurance & financial services, food & beverages and office & data processing machinery, the proportion rises to almost three quarters. If one were seeking the key to Ireland’s international competitiveness, one should be looking at why these seven very disparate sectors are able to use Ireland as a successful base for serving external markets.
But this would require some real research. Instead, our economists reach for simplistic and largely irrelevant statistics which are both readily available and tend to confirm deeply-entrenched prejudices. We got a good example of this in an article on Ireland’s competitiveness by The Irish Times’s chief economics journalist, Dan O’Brien, in the issue of February 4 last. While acknowledging that there are many ways of measuring competitiveness and that the National Competitiveness Council employs more than 100 competitiveness indicators in its annual reports,O’Brien then devotes most of the rest of his article to the old diehards – prices and labour costs.
Irish economists have an extraordinary tendency to rely on the EU’s harmonised index of consumer prices (HICP) as a measure of competitiveness, even though its relevance to Ireland’s export competitiveness is not immediately obvious – it is hard to see what bearing the price of a meal or a CD player has on the competitiveness of the organic chemicals or software sectors. Nevertheless, O’Brien regards the fact that Ireland’s HICP fell relative to the rest of the EU between late 2008 and early 2010 as evidence of Ireland “regaining” competitiveness.
O’Brien then suggests that trends in economy-wide unit labour costs (the ratio of wages to net output) are a better indication of Ireland’s improved competitiveness. However, the fact is that the vast majority of the Irish workforce are not engaged in export activity and, again, it is hard to see how the unit labour costs of a waitress or CD player salesperson have a bearing on the competitiveness of the main export sectors identified above. While Ireland’s economy-wide unit labour costs have tended to rise relative to the rest of the EU over the last ten years, the opposite has been the case in unit labour costs in manufacturing, the great bulk of whose output is exported. Yet, while the latter are to be found in the same page in the OECD website as the former, they are rarely, if ever, quoted by Irish economists.
There are no comparable data for export services, but the Forfás Economic Impact surveys indicate that payroll costs as a proportion of value added in foreign-owned export services (which account for 95% of the total) fell from 19% in 2000 to 9% in 2008 – a fall of over 50% in unit labour costs.
This is not to say that labour costs are important (never mind crucial) in the competitiveness of Ireland’s main export sectors. If general labour costs were a key determinant of competitiveness, then one should expect exports in all sectors to be influenced by labour cost trends. However, Ireland’s main export sectors have been very variable in their export performance, and there is no evidence that this variability has been influenced in any way by labour cost trends. Between 2000-2006 (the last year for which the relevant data are available), the chemicals & pharmaceuticals sector experienced volume output growth of 50%, despite a rise of 50% in the share of costs accounted for by pay (up from 12.6% to 18%). In the electronic components sector, there was a more modest rise in the labour share of costs from a lower base (up from 13.1% to 16.4%) yet production volume fell by 7%. In the office machines and computers sector, a fall in labour’s already very low share of total costs (down from 4.1% to 3.5%) produced volume growth of 22% - much more modest than that experienced by chemicals & pharmaceuticals.
The key point is that the idea of Ireland Inc. gaining or losing competitiveness is meaningless. Ireland’s exports are dominated by a small number of sectors whose characteristics are extremely varied and whose export performances are equally varied. Adding up these performances and then concluding from the total that Ireland is becoming more or less competitive is pointless. Over the last ten years Ireland has experienced strong export growth in a range of export services, and in pharmaceuticals and medical devices, modest growth in chemicals, and, overall, a sharp fall in exports of electronics hardware. A wide range of factors account for this export variability, of which labour costs play, at most, a minor role. In compiling its Global Competitiveness Index, the World Economic Forum employes no less than 113 quantitative criteria; in Ireland’s case labour cost factors account for less than two per cent of the total value of the competitiveness index.
The economists’ disconnect from the real world is nicely demonstrated from a passage towards the end of Dan O’Brien’s article. Referring to evidence that average productivity in Ireland has been raised by the collapse of the low-productivity construction sector, he suggested as an example that “a bricklayer produces less than the average assembly-line worker or office drone (sic)”. Productivity in the construction industry is indeed low when compared with manufacturing or business services, but O’Brien’s choice of bricklayers for his example was surely unfortunate. Assuming that earnings bear some relationship to productivity, in 2006, when industrial production workers were earning €601 per week, clerical and secretarial workers €540 and administrative civil servants €819, the average weekly earnings of all skilled construction workers came to €877 and there was one report of bricklayers at that stage earning over €3000 per week! The company making these payments was also reportedly paying its Turkish labourers €2.50 per hour. They might have been a better choice for O’Brien’s example.