Keywords

This chapter presents a survey of the literature that has aimed to explain what caused the boom.Footnote 1 Most of this literature agrees that there was no single explanation for the boom and that the boom was caused by some combination of a number of explanatory factors, but there are widely varying views on which factors were important. The aim in Sect. 3.1 of this chapter is to present virtually all of the economic explanations that have been advanced, but it will not be possible to refer to every individual item of literature that discusses these explanations.

This chapter does not aim to reach final conclusions on how much each explanatory factor contributed to causing the boom. However, Sect. 3.2 includes some assessment of the suggested explanations, concluding that a number of them are not convincing or not very important whereas the others will call for further consideration and assessment in later chapters in this book.

3.1 Explanations for the Boom

3.1.1 Fiscal Stabilisation

Fiscal stabilisation was often cited as one cause of the boom. This was against the background of rising public debt in the 1980s which culminated with the debt reaching a peak of 118% of GNP in 1987. The government cut current and capital expenditure in 1987, reducing the budget deficit and the debt/GNP ratio from that year onwards. Economic growth began to improve almost immediately after years of very poor performance.

Giavazzi and Pagano (1990) suggested that this could have been an example of expansionary fiscal contraction (EFC), meaning that the fiscal contraction stimulated economic expansion. In principle, an EFC could occur if the deflationary impact of fiscal contraction on demand is outweighed by a boost to private investment and consumption resulting from improved confidence and expectations concerning future taxation, etc. (See Considine and Duffy 2007, for more on the EFC concept).

Bradley and Whelan (1997) used a small open economy macro econometric model of Ireland to assess this issue and they concluded that it was unlikely that an EFC could have occurred in Ireland. Honohan (1999) showed that the actual sequence of events was in the wrong order for an EFC. The economic recovery was led by very strong growth of exports in 1987, followed by smaller increases in the growth of consumption in 1988 and 1989, with investment beginning to recover only in 1989. It became largely agreed among economists in Ireland that there was no EFC in the late 1980s. Rather, the economic recovery was led by export growth which was stimulated by various conditions (e.g., Bradley et al. 1997; Bradley and Whelan 1997; OECD 1999; Honohan 1999; FitzGerald 2000; Bergin et al. 2010).

Despite the scepticism about fiscal contraction being the immediate cause of economic recovery in the late 1980s, it was nevertheless widely held that fiscal stabilisation had substantial long-run benefits. For example, it was seen as the “main precondition for a sustained economic recovery” (Mac Sharry and White 2000), and it “injected a crucial element of long-term confidence about the direction of policy” (Honohan 1999). Similar views were expressed by the OECD (1999), Honohan and Walsh (2002), Gallagher et al. (2002), NESC (2003) and Leddin and Walsh (2003).

3.1.2 Tax Cuts, Smaller Government

In political debate and media commentary it was frequently claimed that tax cuts were a major cause of the boom, but this view is scarcer or more qualified in the academic literature on the subject. Leddin and Walsh (1997) noted that tax declined from 41% to 34% of GDP between 1986 and 1996 and they described this as a “growth promoting factor”. Powell (2003) argued that cuts in government spending in the late 1980s reduced the size of the government’s role in the economy and he linked this to the recovery in growth at that time. He also argued that later reductions in income tax rates, as well as some cuts in corporation tax (see below), helped to bring about higher growth rates in the 1990s. He pointed out that by 1999 tax amounted to just 31% of GDP, almost the lowest level in the EU. Haughton (2005) similarly argued that the relatively limited role of the state was a pro-growth factor, with tax at just 31% of GDP in the early 2000s compared to 42% for the EU.

Leddin and Walsh (2003, p.481) expressed a significantly qualified view on the role of tax cuts by referring to the question of cause and effect. They said, “rapid growth of the economy has facilitated tax reductions and it is difficult to disentangle cause and effect”, and there was a “virtuous circle with faster growth leading to lower tax and public debt burdens” which in turn reinforced the economy’s performance. Honohan and Walsh’s (2002) view was similar. Walsh’s (2000) emphasis seemed somewhat different when he wrote that “no dramatic changes in tax rates or in the structure of taxation occurred in the late 1980s that can be identified as the factor that triggered the boom. And it is obvious that the rapid decline in the tax: GDP ratio during the 1990s was primarily a reflection of the large inflow of FDI and the exceptional growth of GDP rather than vice versa”.

Most of the other literature on causes of the boom does not mention tax cuts although some authors explicitly rejected the idea that tax cuts were a cause of the boom. They argued that strong economic growth came first—with rising employment, falling unemployment, etc.—and this made it possible to cut taxes (Ó Gráda 2002; Sweeney 2004).

In the specific area of corporation tax, a very low rate of tax had already applied to manufacturing and selected internationally traded services well before the late 1980s. However, Conefrey and FitzGerald (2011) pointed out that a low corporation tax rate was gradually extended to the remaining services sectors over the late 1990s and early 2000s. They estimated that this had the effect of increasing the level of GNP in 2005 by 3.7% above what it would have been otherwise. This may be compared to GNP growth of 90% over the decade prior to 2005.

3.1.3 Delayed or Belated Convergence

In the 1990s it was occasionally argued that one factor that made some contribution to the boom was a suggested natural tendency for the income level of poorer countries to catch up with, or to converge on, the level of broadly comparable richer countries—if they had adequate preconditions and sound policies (Leddin and Walsh 1997; Sachs 1997; de la Fuente and Vives 1997).Footnote 2

Ó Gráda (2002) put more emphasis on this natural convergence argument. Referring to Ó Gráda and O’Rourke (2000), he pointed out that in 1950–1987 other relatively low-income members of the OECD were converging towards the OECD average level of GDP per head, but Ireland failed to do so. Then rapid Irish growth in the period 1987–1998 put Ireland back “on track”, so that over the whole period 1950–1998 Ireland’s record of convergence was comparable to general OECD experience. Ó Gráda (2002) argued that Ireland had many of the conditions necessary for faster growth in the period before 1987 but it was held back initially by protectionism, and subsequently by “wrong-headed fiscal policy” in the late 1970s and early 1980s. However, by 1987 Ireland was beginning to overcome the earlier fiscal policy mistakes. Together with an inflow of US FDI and a stronger international economy, this allowed the “Celtic Tiger interlude” to make up the ground that had been lost.Footnote 3

Honohan and Walsh’s (2002) main explanation for the boom was similar in some respects. They said that by 1973 Ireland had the preconditions needed for convergence and many foresaw a steady convergence towards UK and European levels “within a generation”. However, convergence was derailed for more than a decade by a series of fiscal policy errors in the 1970s. When these errors were eventually corrected, this allowed convergence to occur, facilitated by a pro-employment approach to wage bargaining. Honohan and Walsh also noted that, although GDP per head of population was far lower in Ireland than in the UK in the 1970s, non-agricultural GDP per person engaged was about the same in both countries. This meant that convergence, when it eventually occurred, was “a belated convergence not in productivity but in the share of the population at work outside low-income agriculture”.

Responding to Honohan and Walsh, Blanchard (2002) said that they went too far in saying that it was a “simple, run of the mill, catch-up story”. Ireland’s economic performance since 1987 was too impressive for that since it looked “quite miraculous”. Responding to Ó Gráda, Barry (2002) accepted that poor policy could inhibit convergence, but he argued that “there are few models that propose that inappropriate policies act merely as a dam behind which the thwarted convergence forces build up … so that when appropriate policies are eventually adopted the lost ground is made up for all the more rapidly”. As a further objection to the delayed convergence explanation for the boom, Barry asked why Ireland had not converged during the 1960s. He did not accept that this was due to delays in removing protection and improving education, because Ireland was ahead of Spain, Portugal and Greece in these respects yet those countries had quite strong convergence in the 1960s while Ireland did not. He also asked why the average income in Ireland had still been at the same level relative to the UK in 1960 as it was in 1913.

Haughton (2005) found that although there was a general convergence tendency among OECD countries over the period 1960–2002, there was not a significant convergence tendency during the second half of that period, 1980–2002. Therefore, Ireland’s rapid growth phase was exceptional and called for an explanation. NESC (2003, Chapter 1) also considered the delayed convergence perspective proposed by Ó Gráda (2002) and Honohan and Walsh (2002) but they decided that they were “not persuaded” by it.

3.1.4 Strong Demand Growth in Export Markets

It was often stated that rapid growth in overseas demand made a significant contribution to the boom. This point was made mainly in relation to the late 1980s and 1993–2000. It is commonly recognised that Ireland’s economic recovery in the late 1980s was partly attributable to strong growth in demand for Irish exports particularly in the UK (Bradley et al. 1997; Honohan 1999; Mac Sharry and White 2000). As regards 1993–2000, Kennedy (2000/2001) observed that overseas demand for imports grew far more rapidly than might have been expected from looking at GDP growth of the countries concerned. For example, in the EU, GDP grew by 2.5% per year while imports grew by 8.1% per year, perhaps because of the Single European Market. The volume of Irish exports increased by 16.5% per year in 1993–2000, which could be broken down into 7.8% per year being attributable to Ireland’s performance in gaining market share while 8.0% per year was attributable to the growth of the markets themselves (Kennedy 2000/2001; NESC 2003, Chapter 1).

3.1.5 Supply of Labour

Quite a number of studies observed that Ireland had the benefit of a plentiful supply of labour in the boom years. Some of them identified and quantified the principal sources of this growing labour supply—a relatively high birth rate until about 1980, a large number of unemployed people in the late 1980s, a large number of emigrants abroad many of whom were willing to return, and rising female participation in the labour force starting from an unusually low initial level (Bradley et al. 1997; FitzGerald 1998, 2000; OECD 1999). Some authors also noted that a plentiful labour supply in the past had commonly resulted in more emigration or unemployment, rather than rising employment (Bradley et al. 1997; FitzGerald 1998; NESC 2003, Chapter 1).

3.1.6 Education, Human Capital

Rising levels of education have been very widely mentioned as making a significant contribution to the boom (e.g., Sweeney 1998, 2008; Gallagher et al. 2002). There is plenty of evidence showing that there were increasing levels of educational attainment in the population and labour force before and during the boom years (e.g., Bradley et al. 1997; Durkan et al. 1999; OECD 1999; Duffy et al. 2001). Using the higher earnings levels of more highly educated people as an indicator of their higher productivity, Durkan et al. (1999) estimated that the rising education level of the labour force in 1986–1996 added 1% p.a. to the effective labour force, in the sense of making the labour force more productive to that extent (see also FitzGerald 2000).

Kennedy (2000/2001) observed that the same type of analysis showed that rising education levels were having an even greater impact on productivity in the early 1980s than during the boom years. He argued that rising education levels therefore could not account for the acceleration in economic growth that occurred in the Celtic Tiger boom (unless human capital affected growth in ways that are not captured by this type of analysis). He concluded that the outstanding new feature of the boom was the increasing utilisation of labour rather than the increase in its quality. Honohan and Walsh (2002) noted that education attainment levels had been rising for a long time before the boom and there was “no significant inflection point in the 1980s”. They said that, like some other long-standing factors, this was an important part of the policy environment, but it could not explain the boom.

It has sometimes been noted, with varying degrees of emphasis, that in order for increased education to have an impact on the economy there had to be a context of adequate demand for the resulting skilled labour (e.g., Breathnach 1998; NESC 2003, Chapter 1). In the past many skilled workers had emigrated, especially in the 1980s (Bradley et al. 1997; Breathnach 2004). As Ó Gráda and O’Rourke (2000) put it, “in an economy like Ireland, the key issue is not how many graduates can be created, but how many of them can be provided with jobs at home”.

Relating to this issue of demand for skilled labour, a number of authors suggested that there was an important interaction between rising levels of education and FDI, because new FDI created demand for skilled labour while at the same time the availability of skilled labour was a significant feature attracting new FDI into Ireland (FitzGerald 1998, 2000; Kennedy 2000/2001; Duffy et al. 2001; Barry 2005; Mac Sharry and White 2000).

Some studies examined international comparative indicators of educational expenditure, standards, participation rates, graduation rates, etc. and they found rather mixed results for Ireland with some strengths and some weaknesses. The education system looked good in certain respects, but not especially impressive overall compared to the EU or OECD (Ó Gráda 2002; Barry 2005; O’Malley et al. 2008). However, Barry (2005) and Crafts (2005) highlighted the apparent strength of the Irish system in meeting the specific requirements of foreign MNCs.

3.1.7 The Single European Market

In the late 1980s and early 1990s the EU implemented a programme of reforms aimed at establishing a Single European Market by 1992. This involved removing non-tariff barriers to cross-border business activity within the EU—barriers such as different national technical standards, nationalistic government procurement and different national regulatory regimes. The Single European Market is widely considered to have boosted Ireland’s economic growth (e.g., Sweeney 1998, 2008; OECD 1999; Bradley 2000; FitzGerald 2000; O’Donnell 2000; Ó Gráda and O’Rourke 2000).

Barry et al. (1999a) drew together the findings from a number of studies on this. They found that Ireland benefited to some extent from the general stimulus to the EU economy, but Ireland also benefited more than most other EU countries because of its own industrial structure and because of its ability to attract FDI. As regards industrial structure, Ireland had a relatively strong competitive position in the particular sectors that had previously been most constrained by the non-tariff barriers. Consequently, Ireland was well placed to gain more than most countries from the removal of the barriers. As regards FDI, flows of US FDI into the EU expanded quite dramatically in the late 1980s, and there was also an increase in intra-EU FDI flows. Barry et al. (1999a) concluded that much of the expansion of US FDI was due to the Single Market programme making the EU market more attractive. Ireland gained more than most countries from the increase in FDI, partly because it already had an ability to attract a disproportionate share of FDI and partly because the Single European Market made it more feasible for many companies to produce in a small peripheral country for the core EU markets (Mac Sharry and White 2000; Barry 2005).

O’Donnell (2000) and NESC (2003) argued that the Single European Market also had some other specific effects such as energising certain service sectors which had been relatively protected, stimulating free movement of capital, reforming competition policy, limiting state aids to weak sectors and causing radical change in public utilities.

3.1.8 EU Structural Funds

From 1989 onwards, Ireland received a substantial increase in its allocation from the newly enlarged EU Structural Funds, and it is widely agreed that this helped to increase Ireland’s growth (e.g., FitzGerald 1998; OECD 1999; O’Donnell 2000; Barry 2000). It has also been stated quite often that this positive effect was relatively small compared to the scale of growth during the Celtic Tiger boom (e.g., Bradley et al. 1997; Ó Gráda 2002; Burnham 2003; Honohan and Walsh 2002; FitzGerald and Honohan 2023).

Between 1989 and 1999, about IR£9.5 billion of structural funds (in 1994 prices) were transferred to the Irish exchequer. The bulk of this funding was spent on physical infrastructure, human resource development and aid to investment in the private sector. Barryet al. (1999a) reviewed some studies that assessed the impact of this EU funding. They concluded that it raised GNP by the late 1990s to a level about 4% above what it would otherwise have been, which was a contribution of about 0.5% per year to the GNP growth rate in the 1990s (when GNP was growing at about 8% per year). They acknowledged that this could be a somewhat conservative estimate.

Some authors made the point that the increased impact of the structural funds began at a very apposite time, because the drive to cut public expenditure had resulted in a severe reduction in public investment by 1989. The structural funds enabled a resumption of public investment to support future growth (Fitzgerald and Honohan 2023).

Some argued that the structural funds also had other more qualitative benefits for Ireland. It is said that the process of planning and administering the funds introduced more effective longer-term planning and consistent implementation of investment (Bradley et al. 1997; FitzGerald 1998, 2000 ; O’Donnell 2000; Honohan and Walsh 2002; Fitzgerald and Honohan 2023). In addition, Ó Riain and O’Connell (2000) considered that the structural funds facilitated the introduction of important new development programmes, particularly for the development of indigenous enterprises, without having to struggle for and win funding from established programmes.

A number of studies considered that the impact of the Single European Market was probably greater than the structural funds in boosting GNP over the long term (Bradley et al. 1997; FitzGerald 2000). In addition, Matthews (1994, Table 7) observed that Ireland’s receipts under the EU’s Common Agricultural Policy (CAP) were a good deal larger than its receipts of EU structural funds. He estimated that Ireland’s CAP receipts were worth about 10% of its GNP in the early 1990s (including the transfer arising from EU consumers paying higher prices for Ireland’s agrifood exports because of the CAP), while the structural funds were worth about 3% of GNP.

3.1.9 Social Partnership, Wage Moderation

From 1987 onwards, the social partners negotiated a series of multi-year national agreements covering various economic and social issues. The “social partners” initially included employers, trade unions, farmers and government and later broadened out to include other community and voluntary organisations. Many have argued that social partnership underpinned economic growth by delivering moderate and competitive national wage agreements together with industrial peace, as well as agreement on the public finances, tax reform, social welfare, health spending, public sector reform, social exclusion, exchange-rate policy, the Maastricht criteria, etc. (e.g. O’Donnell 1998, 2000; Mac Sharry and White 2000; OECD 1999; Ó Riain and O’Connell 2000; Ó Gráda and O’Rourke 2000, Leddin and Egan 2018/2019). It has been argued that social partnership made a significant contribution to the development of a coherent and consistent set of economic policies (NESC 2003). It has also been argued that social partnership led to a general acceptance of the importance of competitiveness as well as recognition of the many factors (not just labour costs) that combine to produce competitiveness (Sweeney 1998, 2008).

Disagreements about the importance of social partnership as a cause of the economic boom have generally focused on the issue of wage moderation or restraint. Some argued that partnership boosted growth by delivering wage moderation which enhanced competitiveness resulting in employment growth. This in turn generated additional tax revenues which were later partly used to reduce direct tax rates thereby facilitating further wage moderation (O’Donnell and O’Reardon 2000). Barry (2009) showed, at a theoretical level, how it is possible for national agreements that offer future tax reductions in exchange for current wage moderation to enhance wage competitiveness in a way that would not occur otherwise.

As evidence that wage moderation occurred, Leddin and Egan (2018/2019) showed that wages’ share of the national economy fell very significantly, Kennedy (2000/2001) showed that profits were taking a rising share of net domestic product, Blanchard (2002) showed that real wages rose more slowly than the rate of technological progress, and NESC (2003) quoted Lane (1997/1998) to show that the rate of return on capital almost doubled between 1987 and 1996. At the same time, Kennedy and Blanchard mentioned that these trends in their respective indicators were already occurring years before the boom.

FitzGerald (1999) found that wage rates and labour costs in Irish industry had been rising a good deal faster than in the UK in the 1960s and 1970s, but then stabilised relative to the UK from around the late 1970s onwards, with labour costs having reached about 90–100% of the UK level. FitzGerald’s econometric work included a role for UK wage rates (among other things) in influencing wage formation in Ireland. His findings led him to suggest that the “impact of the partnership approach to wage formation has been less significant than many have assumed”, and “the partnership approach served more to validate the results which market forces had made inevitable”. However, FitzGerald (1999, 2000) acknowledged that partnership had other significant benefits relating to better industrial relations and economic policy-making.

Leddin and Walsh (1997, 2003) argued that high unemployment had imposed wage moderation on the Irish labour market from the early 1980s onwards, although they accepted that partnership contributed to reinforcing wage moderation after 1987. They also observed that some other countries such as the UK and USA had wage moderation and declining unemployment under a quite different system of decentralised wage bargaining. Honohan and Walsh (2002) expressed similar views. However, O’Donnell (1998) argued that Ireland’s experience with partnership was much more satisfactory than the UK’s approach, which had seen short bursts of fast economic growth followed by deep recessions imposed in order to reduce inflation.

McGuinness et al. (2010) analysed firm-level data from the 2003 National Employment Survey. They found that, despite the national wage agreement (NWA) negotiated under social partnership, there were still many firms that mainly used other types of wage bargaining such as individual-level or firm-level bargaining. Average labour costs were higher in these other firms than in NWA firms. They also found that foreign-owned MNCs that implemented the NWA enjoyed a particularly noticeable labour cost advantage. They concluded that, as the NWA was designed to protect employment in indigenous companies with lower productivity levels, MNCs implementing the NWA were able to set wages at levels well below what would be the case for them under firm-level or individual-level wage bargaining.

3.1.10 Foreign Direct Investment

It has been widely observed that rapid growth of FDI was a very prominent or central feature of the Celtic Tiger boom (e.g., Sachs 1997; Krugman 1997; OECD 1999; O’Hearn 2001; Duffy et al. 2001; Barry 2002, 2005). Ireland already had a record of attracting substantial amounts of export-oriented FDI in the decades before the boom, but US manufacturing investment in Ireland began to increase noticeably after 1987 and Ireland’s share of US manufacturing investment in the EU also began to rise (Barry 2005). Employment in foreign-owned manufacturing began to grow from 1987 onwards after declining for some years previously. By the late 1990s, foreign-owned MNCs accounted for over 45% of employment, about 65% of gross output and over 80% of exports in manufacturing industry (Barry et al. 1999b). Murphy (2000) estimated that, in the absence of the contribution coming from high-tech MNCs’ exports, Ireland’s GDP would have grown by only about 3.5% per year in 1990–96 instead of the actual rate of 7.6% per year. He also mentioned another estimate by McCarthy (1999) that inflows of FDI into Irish manufacturing boosted the growth rate of the economy by about 3% per year in 1993–1997.

Of the reasons that have been suggested to explain why FDI grew in Ireland, some are long-standing reasons that originally aimed to explain why Ireland had been relatively attractive for FDI long before the boom. Many commentators argued that most of these were still relevant during the boom. Such reasons included EU membership with access to large EU markets; low tax on profits; grant incentives; active and effective industrial development agencies; a suitable available labour force—English-speaking, reasonably well-educated, with labour costs below many other EU countries; historical and cultural links with the USA; and a less regulated business environment than many EU countries.

A number of further reasons were suggested to explain the acceleration of FDI in the boom years. These included the single European market; the development and rapid internationalisation of a number of important industries, originating particularly in the USA; declining tariff and transport costs; changing location requirements among MNCs in certain key sectors, with increasing emphasis on access to major markets combined with availability of technical skills; improvements in international communications and the benefits of the transformation of Ireland’s telecommunications system in the 1980s (Burnham 2003); more selective and focused state agency policies for attracting FDI; increasing agglomeration economies among growing clusters of related companies in Ireland, partly as a result of the state agency policies; and “demonstration” and “cascade” effects, meaning that the location decisions of prospective newcomer MNCs were influenced by the perceived successful experience of growing numbers of others already in Ireland.Footnote 4

Some authors also stressed that Ireland’s investment in education and training, resulting in rapid growth in the supply of skilled labour, became an increasingly important factor in attracting MNCs because their requirements for high-level skills were increasing over time (e.g., Breathnach 1998; FitzGerald 1998). This argument concerning education does not necessarily amount to a claim that the Irish education system was superior to those of other European countries in a general overall sense. For it has been argued, much more specifically, that the Irish system was unusually effective at producing sufficient numbers of graduates with the particular types of skills that were required by MNCs in the industries that were growing rapidly internationally (Mac Sharry and White 2000; Barry 2005; O’Malley et al. 2008). Shortages of such skills were not at all unusual elsewhere.

There have been a number of critical analyses of FDI in Ireland (e.g., O’Sullivan 2000; O’Hearn 2001; Kirby 2010), but they generally did not disagree with the view that FDI was a major factor in generating the Celtic Tiger economy. Rather they highlighted disadvantages and weaknesses in a form of economic growth that depended so heavily on such foreign MNCs, so that they were not impressed by the nature or likely sustainability of this growth process. But they generally accepted that FDI was a major element in the growth that occurred.

3.1.11 A Small Open Economy—Like a Region

A number of authors argued that one reason for Ireland’s exceptional boom was the fact that it is a small and very open economy, and like a regional economy in some respects. Nobody suggested that this in itself generates stronger growth, but the argument was that some factors that boosted growth were able to have a greater impact in the small/regional Irish economy than they could have in larger countries.

There are a few characteristics of a small/regional economy that are relevant for this view. First, external trade is relatively large and very important for the economy. Second, inward FDI can occur on a scale that is great enough to be far more influential than in larger countries. Third, the labour market is very open so that relatively large-scale migration occurs easily. Given this type of economy, it was argued that rapid export growth, largely driven by export-oriented FDI, was the major driver of the Irish boom. The very open labour market meant that the return of former emigrants and the growth of new immigration facilitated and prolonged the boom, by preventing labour shortages from emerging and by moderating wage increases (Krugman 1997; Barry 1999, 2002, 2005). It was argued that booms quite like this had occurred in regions within larger countries rather than in entire large countries. NESC’s (2003) interpretation of the boom was mostly consistent with this perspective, as was O’Leary’s (2011), although NESC added that alongside the leading role of FDI there was also a very significant improvement in the performance of Irish indigenous firms.

O’Leary (2015, Chapter 7) also suggested that Ireland’s small size was one reason why it was able to achieve EU agreement for its very favourable corporate tax regime, taking advantage of “the importance of being unimportant”.

Blanchard (2002) presented a somewhat different version of the same type of story, in which the main distinctive element was the emphasis that he put on wage moderation as being the principal cause of strong investment including FDI. Wage moderation in turn, he considered, was probably quite largely a result of the very open labour market and the consequent influence of UK wage levels in restraining Irish wage increases, while the same open labour market provided the extra labour supply needed to prolong the boom.

Using theoretical neo-classical modelling, Barry and Devereux (2006) found that an economy with very open capital and labour markets is affected more substantially than a less open economy by shocks such as an increase in the economy’s attractiveness to FDI, a reduction in labour market distortions, or an increase in total factor productivity.

Honohan and Walsh (2002), Leddin and Walsh (2003) and Crafts (2005) also mentioned the effects of the small/regional nature of the Irish economy, although with less emphasis than the other studies cited above.

3.1.12 Irish Indigenous Industry

Some studies stated, with varying degrees of emphasis, that a substantial improvement in the performance of Irish-owned or indigenous industry was a significant component of the boom (e.g., O’Malley et al. 1992; O’Malley 1998, 2004; Sweeney 1998; Barry 1999; Mac Sharry and White 2000; NESC 2003; Ó Riain 2004a, 2004b; OECD 2006).

Following years of very poor performance, Irish indigenous industry began to grow considerably faster than industry in the EU from 1987 onwards in terms of output and from 1988 in terms of employment. This trend continued throughout most or all of the boom. Exports of indigenous industry also grew relatively rapidly compared with the EU in the late 1980s, but then did no more than match the pace of EU export growth in the 1990s. It was suggested that the export comparison with the EU in the 1990s looked less impressive than the output comparison partly because EU exports were growing unusually fast in the 1990s (perhaps because of the Single European Market), and partly because Irish companies’ incentive to increase exports was diminished when the domestic market began to grow very rapidly in the 1990s (O’Malley 1998, 2004). Indigenous industry also showed other signs of increasing strength such as rising profitability and productivity, rapidly growing R&D, above-average levels of innovation by EU standards, increasing professionalisation, and particularly rapid growth (including export growth at above EU rates) in high-tech and medium-tech sectors rather than the more mature traditional sectors (Ó Riain 2004a, 2004b; O’Malley 1998, 2004; O’Malley et al. 2008).Footnote 5

A number of explanations were suggested for this performance of Irish indigenous industry, including the EU structural funds, rising education levels, social partnership, the secondary effects of FDI growth, etc. However, this literature generally put a particular emphasis on the role of the state’s industrial policy as a factor behind the improvement. It pointed out that industrial policy, from the mid-1980s onwards, had an increased focus on the development of Irish indigenous industry and that a series of new policy measures and approaches were implemented to further this aim. It also argued that the evidence indicated that such measures were successful (O’Malley et al. 1992; O’Malley 1998; Ó Riain and O’Connell 2000; NESC 2003; Ó Riain 2004a, 2004b).

Some authors doubted whether there had been a really significant improvement in indigenous industry (e.g., O’Sullivan 2000; O’Hearn 2001; Enterprise Strategy Group 2004). They cited concerns such as an unconvincing export performance, insufficient evidence of innovation capabilities and limited development of sales to foreign MNCs in Ireland, all leading to questions about the sustainability of this growth.

3.1.13 Other Explanations

A number of other explanations for the boom have occasionally been suggested. These include progress towards peace in Northern Ireland (Gray 1997), local development policies and active labour market policies (NESC 2003), new forms of work organisation (Sweeney 1998) and currency devaluations in 1986 and 1993 (Leddin and Walsh 1997, 2003; Honohan and Walsh 2002; Gallagher et al. 2002). Most of these suggestions were proposed by no more than two or three authors, sometimes with rather little supporting discussion. In most of these cases, it seems clear that the authors did not consider them to be among the more important explanations for the boom.

Some also argued that Ireland’s transition to participation in EMU helped economic growth (O’Donnell 2000; Leddin and Walsh 2003; NESC 2003). However, Honohan and Walsh (2002) argued that the stimulatory effects of the resulting lower interest rates came in the late 1990s which was relatively late and not very helpful timing.

Honohan (2006) considered whether financial sector development might have been a driver of Ireland’s twenty-year growth success but concluded that there was little evidence to support this idea.

Although privatisation and deregulation were a major focus for economic policy in much of Europe during the time of the boom, they were seldom mentioned as causes of the Irish boom. However, the specific case of airlines was mentioned sometimes. In the mid-1980s, deregulation introduced more competition in the Irish airline industry, and it has been argued that this provided a major stimulus for a subsequent tourism boom (Barry 1999, 2000; Murphy 2000; Burnham 2003). It has also been argued that stronger promotion and improved price competitiveness contributed to that growth of tourism (Honohan and Walsh 2002; Leddin and Walsh 2003).Footnote 6

It was also sometimes argued that the transformation of Irish telecommunications in the 1980s proved to be very important for attracting FDI in the service sector (Burnham 2003; Barry 2000). However, this transformation was carried out under the control of the state monopoly company Telecom Éireann, which was not privatised until much later.

Some authors suggested different types of explanation for the boom, of a social or political nature, but these are mostly outside the scope of this book which focuses on economic explanations. If these other explanations made an important contribution, they probably had their effects on economic growth mainly by means of driving or influencing some of the explanatory factors that are included here.

3.2 Assessment of the Explanations

This section undertakes a preliminary assessment of the explanations discussed above. It argues that a number of the suggested explanations are not convincing or not very important and that it will not be necessary to consider them further, whereas the others will call for further consideration and assessment in later chapters.

3.2.1 Necessary But Normal Conditions

Among the explanations for the boom discussed in Sect. 3.1, a few refer to conditions that could be described as necessary but normal. Rather like a reliable electricity supply or a functioning legal system, for example, these conditions would be necessary for the attainment of a satisfactory or average growth performance, while they have also been quite normal or common across many countries and time periods. Because they have been so normal in many places and times, they cannot actually help much to explain why Ireland had such a highly exceptional boom, which involved departing substantially from the contemporaneous experience of most other countries and from Ireland’s own previous experience.

Fiscal stabilisation is in this “necessary but normal” category. If the expansionary fiscal contraction argument is rejected, as seems to be largely agreed, fiscal stabilisation refers to the establishment of a condition that was necessary but was also quite normal or common. It could reasonably be argued that fiscal stabilisation was an important precondition for a return to moderate or average growth in Ireland after the slump of the 1980s. However, like the example of a reliable electricity supply, it is difficult to see how such a normal condition could help to explain why Ireland’s economic growth was much faster than average for a very long time.

Ireland’s plentiful supply of labour in the boom years was another necessary condition, but this was a condition that was also quite normal in the sense that many other European countries had plenty of labour available. In the years when the boom was occurring in Ireland, the unemployment rate in the EU-15 was usually above 8% (as was seen in Fig. 1.4), and many of those EU countries could also have attracted additional immigrant labour if they had required it. Furthermore, a plentiful supply of labour had been normal in Ireland itself for generations before the boom, but the previous results had always been a good deal of emigration and/or unemployment rather than exceptional growth of employment and the economy. What was new and distinctive about the boom period in Ireland was not the availability of a plentiful supply of labour, but the fact that there was strong demand within the country for the available labour.

As regards education or human capital, it is necessary to make a distinction between (a) the general process of raising the educational standards of the population and labour force and (b), within the broader system, the particular activity of producing a supply of graduates with the specific skills and qualifications that were required by rapidly growing and internationally mobile industries. The more general process was necessary but, as discussed in the literature on this topic cited in Sect. 3.1, it was also quite normal in the sense that it had already been going on for years before the boom and there does not appear to have been an acceleration or intensification that could be seen as a cause of the boom. It was also normal in the sense that the standards being attained do not seem to have been very exceptional for the most part, compared with other developed countries.

However, the more specific activity of deliberately providing for the skill requirements of fast-growing and mobile target industries was more exceptional in an international context, and was consciously intended to take advantage of new opportunities as they emerged and expanded. Hence this was potentially a cause of exceptional growth outcomes, in conjunction with changing patterns of industrial growth and FDI.

3.2.2 Delayed or Belated Convergence

If correct, the argument concerning delayed or belated convergence would be important. It could, on its own, provide much of the explanation for the boom. It would also transform the significance of fiscal stabilisation—described above as simply a “necessary but normal” condition—since the delayed convergence argument holds that Ireland was naturally overdue for relatively rapid growth as soon as such normal conditions for growth prevailed.

However, there are a number of problems with the delayed convergence argument. For one thing, as was outlined in Sect. 3.1, Barry (2002) identified significant problems with it.

Perhaps more fundamentally, the basic premise of the delayed convergence argument is questionable, i.e., the premise that there is a general tendency for convergence of average income levels among broadly comparable economies. There are conflicting theories on this matter—Ó Gráda and O’Rourke (2000) outlined some examples—so that the validity of the basic premise cannot be taken for granted. As regards empirical evidence, a convergence tendency certainly has been observed among some countries in some periods, but there is also other evidence that shows no such tendency, which calls into question the idea that convergence is the general tendency.

For example, Haughton (2005) found that although there was a general convergence tendency among OECD countries over the period 1960–2002, there was actually no significant convergence tendency during the second half of that period, 1980–2002. This means that there was a marked convergence tendency over one two-decade period, 1960–1980, but such a tendency was absent over the following two decades.

Ó Gráda and O’Rourke (1996) generally accepted the view that convergence was the prevailing tendency, at least among developed countries, but findings that they presented look consistent with the idea that convergence may actually have been confined to certain periods. Their Fig. 13.2 showed that, among Western European countries, GDP per head did tend to converge over the long period 1950–1988. However, when this is broken down into sub-periods, convergence was weak in 1950–1960, strong in 1960–1973 and absent in 1973–1988.

It was already seen in Fig. 1.2 in Chapter 1 of this book that there was no sign of convergence between the USA and the EU-15 during the period 1980–2007. In addition, Pain (2000/2001) showed that the EU’s GDP per head had risen from less than half of the USA’s level in 1950 to two-thirds of the US level by 1973, but then little further convergence occurred over the next 25 years, 1973–1998, as the EU’s GDP per head stayed within a range of about 65–72% of the US level. Thus, the overall story here—concerning the two largest economic blocs in the world—is that there was quite strong convergence between them in the 23-year period 1950–1973 followed by no further convergence in the 34-year period 1973–2007.

Taking account of the various observations outlined above, it may be concluded that the evidence here is too inconsistent to support the theory that convergence of average income levels has been a generally prevailing tendency. It seems more likely, from this evidence, that convergence may have been quite a common tendency among developed market economies from about the 1950s until the early 1970s, but there was no obvious convergence tendency after that time. Consequently, the evidence here would not justify the assumption that Ireland must necessarily have benefited from a strong convergence tendency in 1973–1986 if only it had adopted sound fiscal policies. There is also little support here for the argument that the Irish boom after 1986 was caused by the eventual arrival of the general convergence tendency.

Another type of problem with the delayed convergence argument arises from the fact that the theory underlying the suggested tendency to convergence (as outlined by Ó Gráda and O’Rourke 2000) actually applies largely to factor incomes. This means that the theory says that remuneration per unit of labour or capital should tend to converge. Even if this is true, the problem is that it does not necessarily mean that incomes per head of population have to converge. For example, Pain (2000/2001) showed that the EU’s GDP per hour worked rose from two-thirds of the US level in 1973 to almost 100% of the US level by the late 1990s. This would have provided a basis for convergence in hourly wage rates. But despite this, there was no convergence trend in GDP per head of population in that period because of differences between the EU and USA with respect to the proportion of the population in employment and hours worked per employee.

This point is relevant for Ireland because, as mentioned above, non-agricultural GDP per person engaged was actually about the same in Ireland and the UK in 1973, although Ireland’s GDP per head of population was 27% below the UK level. Thus, Ireland’s output per employee (and hence potentially wages per employee) across the bulk of the economy had already converged on the UK level by that time, while Ireland’s relatively low GDP per head of population was mainly due to a low proportion of the Irish population being in employment. Given this situation, it is not clear that Ireland had anything to gain from the convergence of factor incomes that is envisaged by the theory underlying the suggested general tendency to convergence.

In Honohan and Walsh’s (2002) version of the delayed convergence argument, they explicitly recognised that Ireland’s productivity had largely converged with the UK by 1973, and they recognised that convergence for Ireland in terms of GDP per head of population would require a large increase in employment, and more specifically non-agricultural employment, as a percentage of the total population. They believed that in 1973 the conditions in Ireland were right for such a convergence to occur by around the end of the century, before the convergence was derailed by fiscal policy errors.

As Honohan and Walsh made no explicit reference either to international evidence or to any generalised theory in order to justify their belief that Ireland in 1973 was on the way to convergence, their view is not undermined by the problems discussed above with the international evidence and the theory relating to convergence. However, this also means that the only rationale for their view appears to be a judgement that trends in Ireland were moving in the right direction, with rising participation by a better-educated workforce in non-agricultural sectors. This looks like an inadequate basis for their view because there were important trends in Ireland in the years leading up to 1973 that did not provide support for their argument.

A key point here is that total employment in Ireland had scarcely grown at all in 1960–1973, with an increase of just 0.2% per year, as discussed in Chapter 2. In that context, employment as a percentage of the population in Ireland had declined from 37.1% in 1960 to 34.8% in 1973—far below the corresponding figure for the UK. This trend was the opposite of the essential trend that was required for Ireland to achieve convergence in terms of GDP per head of population. Unless this trend could be reversed, there could be no prospect of Ireland’s average living standards reaching the level of the UK and other advanced European economies, because such a convergence in those circumstances would require Ireland’s output per person employed to rise far above the levels found in those countries.

Honohan and Walsh (2002) focused particularly on non-agricultural employment. If we look at the trend in non-agricultural employment as a percentage of the population, there was a clear rising trend in this indicator during 1960–1973, but this trend was of limited significance because it was too weak. In a context where the agricultural labour force was in secular decline, there would have needed to be a sizeable compensating increase in non-agricultural employment as a percentage of the population just to maintain total employment at a constant percentage of the population. In fact, the growth of non-agricultural employment in 1960–1973 was not even sufficient to do that since we have seen that there was a decline in total employment as a percentage of the population in 1960–1973.

Honohan and Walsh (2002) presented a chart which showed the rising trend in non-agricultural employment as a percentage of the population. If we extrapolate the rising trend seen in 1960–1973 forward in time, it seems that non-agricultural employment was heading towards about 33 or 34% of the population by the end of the century. However, this rising trend was well below what was needed to bring about convergence by the end of the century. When GNP per head converged on the EU level at the end of the century, non-agricultural employment had actually reached about 42% of the population, much higher than where it had been heading during 1960–1973.

Finally, there is another problem with the delayed convergence argument. The delayed convergence argument holds that the rise in Ireland’s GNP per head from about 65% of the EU level in the early 1970s to almost 100% by 2000 was primarily a natural convergence process. This convergence process would have been occurring more evenly throughout that period but for the fact that it was prevented from happening in the 1970s and early 1980s because of fiscal policy mistakes, with the result that the eventual convergence took the form of an exceptional boom from the late 1980s onwards.

To put this suggestion in perspective, consider some quantitative estimates by FitzGerald (2000) that are relevant here. FitzGerald’s Fig. 3.10 indicated that the expansionary impetus imparted by fiscal policy in 1977–1979 amounted to about 2.5% of GNP and this was followed by a deflationary impact of around 6% of GNP in 1980–1987, giving a net deflationary impact of about 3.5% of GNP by 1987. Similarly, FitzGerald’s Fig. 3.11 indicated that if a neutral fiscal policy had been pursued from 1974 onwards, GNP could have been 3 or 4% higher by 1987 than it actually was.

This implies that, with more orthodox fiscal policies, Ireland’s GNP per head could have reached about 66% of the EU level by 1987 instead of the actual figure of 64%. This gives little support to the suggestion that there was a convergence tendency present all the time that was strong enough to raise Ireland from about 65% to almost 100% of the EU level within a few decades provided that Ireland avoided policy errors. It is more consistent with the idea that there was no convergence tendency of any significance applying to Ireland.

3.2.3 Tax Cuts

Those who argued that tax cuts were an important cause of the boom generally supported this view by stating that there was a substantial decline in tax as a percentage of GDP between about 1986 and the late 1990s. However, this argument is questionable because the decline in taxation actually occurred relatively late in this period, long after the boom began.

Figure 3.1 shows tax as a percentage of GNP and GDP each year from 1982 to 2007. It can be seen that there was no clear declining trend in either of these series before 1995. In 1994 tax as a percentage of GNP or GDP was actually higher than in 1986 when the boom was about to begin and, even if we deduct the exceptional revenue received from a tax amnesty in 1994, tax as a percentage of GNP or GDP in that year was at just the same level as in 1986. Thus, it was not until 1995—the ninth year of relatively fast economic growth—that the overall tax level began a declining trend.Footnote 7

Fig. 3.1
A line graph plots tax as a percentage of G N P and G D P. The line for tax as a percentage of G N P peaks at 41 in 1988 and declines to 35 in 2007. The line for tax as a percentage of G D P peaks to 36 in 1988 and declines to 30 in 2006. Approximated values.

(Note: Once-off receipts under tax amnesties increased the tax yield by 2.1% of GNP in 1988 and by 0.7% of GNP in 1994. Tax here is defined to include taxes, employers’ and employees’ social insurance contributions and health and training fund contributions. It does not include other exchequer non-tax revenue. Source Derived from Department of Finance, Budgetary and Economic Statistics, October 2012, Tables 4 and 12)

Tax as a percentage of GNP and GDP, 1982–2007

Full size image

Much of the discussion about tax cuts as a cause of the boom focused more specifically on taxes that have a bearing on labour costs. A useful way to examine trends in such taxation is to consider trends in the “tax wedge”, which measures the difference or gap between the cost to an employer of employing someone and the take-home pay received by the employee. The size of this wedge or gap is determined by the amount of social insurance (PRSI) paid by the employer plus the amount of income tax and social insurance paid by the employee. Figure 3.2 shows the trend in the tax wedge in 1982–2007, as measured by the average cost to the employer divided by the average take-home pay of the employee. Thus, a reading of 1.5 in Fig. 3.2 would mean that on average it costs the employer 1.5 times the amount that the employee receives.

Fig. 3.2
A line graph plots the tax wedge between 1982 and 2007. The line peaks at 1.6 in 1988, then declines to 1.38 around 2003, and then slightly increases. Approximated values.

(Note The tax wedge is defined here as (1 + RGTYSE)/(1 − RTYPTOT) where RGTYSE is the average rate of employer social insurance contributions and RTYPTOT is the average rate of personal taxation, including social insurance, paid by employees. Source Derived from Economic and Social Research Institute (ESRI) databank, based on data from the Central Statistics Office (CSO). Creative Commons Attribution BY 4.0)

Irish labour market tax wedge, 1982–2007

Full size image

Rather like the trend already seen with respect to overall taxation in Fig. 3.1, there was no declining trend in the tax wedge before 1995. The figures for 1993 and 1994 were somewhat higher than the figure for 1986 when the boom was about to begin, and it was not until 1995—nine years into the boom—that the tax wedge began a declining trend.

It is clear from Figs. 3.1 and 3.2 that tax cuts did not help to start the boom or to sustain it through its first decade or so. Therefore, the boom was generated by other factors. When the declining trend in tax began, it was an effect or a consequence of the boom since fast growth of incomes and employment had led to a declining unemployment rate, a declining dependency ratio and a declining national debt burden—all of which facilitated tax reductions.

In the literature discussed in Sect. 3.1, it was suggested by a couple of authors that a “virtuous circle” effect could have developed, such that tax cuts were initially a consequence of the boom but the tax cuts then helped to stimulate further growth, which in turn facilitated further tax cuts and so on. However, this suggestion is not convincing. The economy had already proved capable of prolonged rapid growth without tax reductions being part of the explanation, so evidently that could have continued to be the case. At the same time, no evidence was offered to support the suggestion that tax cuts could have helped to stimulate growth further.

3.2.4 Minor Explanations

The range of “other explanations” for the boom that were mentioned at the end of Sect. 3.1 were clearly of no more than minor importance even if there was some truth in them. Most of them were mentioned by few authors, and it seems clear that nobody considered most of them to be among the more important explanations for the boom. Here we briefly consider two of them that were mentioned more than most—currency devaluation in 1986 and 1993, and deregulation of airlines.

It was suggested that the devaluation of the Irish pound in August 1986 helped to stimulate the growth of Irish industry’s exports over the following years. However, any such effect must have been minimal. The main reason is because the devaluation was relative to the ECU (European Currency Unit) while there was no significant devaluation relative to other currencies that were important for Irish industrial exports. If we go by the trade-weighted exchange-rate index for the Irish pound published by the Central Bank of Ireland, the value of the Irish pound changed rather little each year in 1986–1989 declining by an average of just 1.1% per year. Since it then rose the following year, it ultimately increased slightly by 0.6% per year over the period 1986–1990.

The devaluation at the end of January 1993 had a more substantial effect at first, with an initial decline of about 10% in the Irish pound’s effective exchange rate, but most of this decline was reversed over the next few years so that the net decline over the period 1993–1996 was just 3%. It seems unlikely that this could have been of significant lasting benefit for the economy.

It was claimed that deregulation and greater competition in the airline industry led to the growth of air traffic and consequent benefits for the tourism industry. However, since tourism accounted for just 4% of GNP by 2000, its growth cannot have contributed very substantially to the overall boom. Perhaps more importantly, since the growth of air traffic facilitated outward tourism as much as inward tourism, the net contribution to Ireland’s GNP would have been significantly less than the contribution coming from the growth of inward tourism.

3.3 Conclusion

For the reasons discussed in Sect. 3.2, it may be concluded that a number of the explanations for the boom that were proposed in the literature are not convincing or not very important. Consequently, those proposed explanations will not be considered further in this book, whereas the others will call for further consideration and assessment in later chapters.

The suggested explanations that will not be considered further are fiscal stabilisation, a plentiful supply of labour, the general process of education (other than education specifically for fast-growing industries), delayed convergence, tax cuts and the “other explanations” that were mentioned at the end of Sect. 3.1.

The rest of the suggested explanations remain as potentially significant causes of the boom and they will be considered and assessed further in later chapters in this book. These remaining suggestions are strong demand growth in export markets, foreign direct investment, the Single European Market, education for fast-growing industries, the small/regional nature of the economy, Irish indigenous industry, EU structural funds and social partnership/wage moderation.