The impact of foreign direct investment on sectoral adjustment in the Irish economy.

Ruane, Frances ; Gorg, Holger

1. Introduction

Foreign direct investment (FDI) has played a crucial role in the overall development of the Irish economy over the past three decades, as the Republic of Ireland, hereafter referred to as Ireland, has pursued an industrial strategy characterised by (i) promoting export-led-growth in Irish manufacturing through various financial supports and fiscal incentives, and (ii) encouraging foreign companies to establish manufacturing plants in Ireland, producing specifically for export markets. The significance of FDI for the Irish economy is now reflected in, inter alia, the significant gap between GNP and GDP; in 1994, GNP was roughly 88 per cent of GDP in Ireland. As regards the manufacturing sector, the high shares of output and employment in foreign-owned companies in Ireland also indicate the importance of foreign firms. As we discuss in some detail in Section 3, foreign companies produced roughly 69 per cent of total net output and accounted for 45 per cent of employment in Irish manufacturing industries in 1993.

From the early 1930s until the mid-1960s Ireland had a highly protected economy with some of the highest rates of effective protection of any economy in the world (McAleese, 1971). In particular, these tariffs protected Irish manufacturing industry against imports from the UK. Reductions in tariffs began in 1966 as a consequence of the Anglo-Irish Free Trade Area Agreement, and the process was eventually completed in 1978, following Ireland's membership in 1973 of the European Economic Community (EEC). In the absence of FDI, the reduction of high tariffs in Ireland would have required a massive devaluation against sterling, when policy at the time was fully committed to maintaining that link at parity. The scale of FDI coupled with some restructuring support for import-substituting Irish firms allowed the exchange rate to be maintained, and an inevitable shake out in the indigenous sector to take place without any massive decline in total employment in the manufacturing sector.

In addition, FDI has played a key role in the process of adjustment within the manufacturing sector and across geographic regions. The sectoral composition of FDI overall is less concentrated on traditional and food-sector activities than that of indigenous manufacturing, and thus the growth of the FDI component in manufacturing has meant that the share of these activities in total manufacturing output has declined.(1) The extent of restructuring is more apparent the lower the level of sectoral disaggregation. In terms of regional employment, FDI became a vehicle for dispersing manufacturing employment across the country, and away from the traditional manufacturing bases in Dublin and Cork. Since agricultural employment still accounted for a high share of total employment in Ireland in the 1960s, the regional dispersal of foreign-owned firms created new jobs in regions where existing employment opportunities were declining.(2)

This paper concentrates on FDI in the manufacturing sector, as the sector where FDI has hitherto had its greatest impact, in terms of output share and the associated high export ratios.(3) Where data allow, we include analysis of internationally-traded services, which have seen dramatic growth in more recent years. In the context of both the manufacturing and internationally-traded service sectors we assess the major impact of foreign-owned industries in Ireland by focusing primarily on the associated employment which has always been the major target of the policy makers. We also focus on a value-added type measure to try to capture the degree of linkages of foreign companies into the Irish economy.

The outline of the paper is as follows. In Section 2 we look at the characteristics of the policy which has been used to promote FDI and at some aspects of its implementation.(4) In Section 3 we examine the overall impact of FDI in Ireland in terms of employment, focusing on net employment changes as well as on employment gains and losses, and in terms of expenditures of foreign companies on Irish inputs. In Section 4 we consider some of the current concerns about industrial policy in Ireland and make some concluding comments.

2. Key Characteristics of Irish Policy towards FDI(5)

The decision to promote FDI actively in the 1950s represented a dramatic change in policy for the Irish economy, as FDI had previously been legislatively prohibited.(6) From this time onwards, attitudes in Ireland to FDI, irrespective of source, have been extraordinarily positive right across the socio-political spectrum. In this section we identify three major characteristics of Irish policy - its employment focus, its combination of automatic fiscal and discretionary financial policies, and its emphasis on policy certainty.

Employment Focus

In the mid-1950s FDI was seen as the key to providing enough jobs to counteract the falling number of agricultural jobs in rural areas and the inevitable decline in employment in indigenous manufacturing which would arise with free trade. At that time there were high unemployment and emigration rates, especially in and from rural areas. Consequently employment became and continues to be the key indicator of the success or failure of Ireland's industrial strategy. The regional distribution of companies and employment has been constantly monitored, and as we shall see below, the share of total manufacturing employment accounted for by foreign companies is now over 45 per cent. As regards the quality of employment, Ireland has marketed itself, particularly in the US, as a location for highly skilled workers for the last 15 years. This strategy is linked to the expansion of second and third level education. Some 40 per cent of school-leavers now go on to third level education and it is proposed to increase that ratio to 50 per cent over the next ten years, with an emphasis on technology-type education (Higher Education Authority, 1995).

In the past decade, other attributes of FDI, in particular profits (because of the growing significance of the corporate tax yield), technology (because of possible spill-over effects), and linkages (because of the potential positive effects for indigenous companies) have received increasing attention. Nonetheless, because of recurrent high levels of unemployment and emigration, employment remains and is likely to remain the key variable for measuring the success or failure of Irish industrial policy.

Automatic Fiscal and Discretionary Financial Policies

Pro-active policy has taken two major forms: fiscal and financial. Initially the main element in the fiscal policy for foreign manufacturing companies was an automatic tax holiday on the profits from all new export sales.(7) When introduced in the mid-1950s, the tax holiday was given for ten years but this was subsequently extended to fifteen years with a further five years of partial relief; the holiday was set to terminate in 1990. Prior to 1990, however, Ireland was forced by the European Commission to alter the policy for new firms, as the tax holiday created a bias towards exports which was incompatible with the Treaty of Rome. Since 1982, all new firms have been entitled to an automatic preferential corporate tax rate of 10 per cent which has applied to total profits and not merely those arising from export sales.(8)

Comparing tax incentives with other incentives, various surveys have concluded that the tax incentives are the most important incentive encouraging manufacturing investors to locate in Ireland. For example, a recent Deloitte Touche Tohmatsu survey indicated that almost 60 per cent of foreign companies interviewed found the ten per cent rate to have been very influential in their location choice. This, however, does not imply that they have been quantitatively more important than other factors. Other factors mentioned in that survey were financial incentives, geographic location and infrastructure, skilled work force, stable currency, and low wage costs, though the survey does not give a ranking according to their importance.

Since part of Ireland's attractiveness for foreign companies lies in its being a base for exporting, especially for extra-EU companies, the switch to a trade-neutral incentive had little effect on firms' sales or marketing behaviour, with exporting remaining the driving force behind foreign investment in Irish manufacturing (see Table 1). The automatic fiscal incentives are backed by double-taxation agreements, intended to maximise the benefit of the tax incentives to foreign-owned companies.(9) The benefits are relatively less for UK and US firms as Ireland does not have tax-sparing agreements with them. However, in the case of the US firms, the preferential tax rate is still very important because of the possible benefits, given the scale and scope of their international investments, of tax deferment. Indeed, personnel engaged in industrial production would suggest that tax incentives are particularly important in promoting investment by US firms. McCutcheon (1995) examines how the complex methods used by US companies investing in Ireland can have an enormous impact on the benefits they derive from the preferential tax rate and hence on the effective cost of capital which they face.

Financial supports for foreign firms locating in Ireland in the 1950s were primarily in the form of cash grants towards the cost of the plant and machinery which would be used to manufacture products for export markets.(10) As with the tax holiday, the basis for grant eligibility had to be changed in 1982, so that the grants now apply to all manufacturing firms and not merely those which are exporting. Additional assistance in the form of employment and training grants, subsidised rents, technology-transfer supports, etc., have also formed an increasing component of the policy package since the late 1970s and in the case of some labour-intensive projects the employment and training grants can be the most important component of the total financial support given. By contrast with the fiscal incentives which are available to all firms automatically, these are discretionary grants, available up to certain maxima (determined by legislation) and implemented at the discretion of the Industrial Development Authority (IDA). Thus all FDI projects in Ireland involve bargaining between project executives and potential investors.(11) The grant maxima(12) ensure that the amount of assistance given is bounded and when the absolute amount of the capital grant support exceeds a defined number, presently [pounds]2.5million, the grant support requires Cabinet approval.(13) All grant payments are in the public domain so that transparency is assured about the final payment but not the process of arriving at it.(14)

While the discretionary option was rarely if ever exercised in the 1950s and 1960s, in the sense that the maximum grant was virtually always paid out, it has had a lasting effect on industrial policy in Ireland, in that for over forty years industrial policy has operated at project level and has increasingly become more pro-active and selective.(15) At project level this selection process has been market driven. It involves (i) identifying high-growth market niches, containing projects which are potentially mobile into Ireland; (ii) assembling information on companies which are operating successfully in those niches and which might be considering diversifying their production internationally;(16) (iii) initiating contacts with the companies identified through officials located in the relevant country; and (iv) getting the company to visit Ireland in the context of a specific project proposal. The precise pattern of FDI projects which come to Ireland is strongly influenced by this process, which might best be described as 'market-led intervention'.

With regard to sectoral selectivity, in the 1970s subsectors in the electronics and pharmaceutical sectors were identified as providing the most promising opportunities for foreign investment projects and the US was identified as the most likely market source for such projects. The creation of industrial clusters has been central to policy and since the late 1980s and early 1990s this strategy has begun to yield very significant benefits especially with US companies (see Section 3).(17) Promotion of outsourcing linkages with domestic firms in these sectors, as well as the attraction of head-quarter and R&D functions to Ireland are seen as important ways of achieving high-income jobs and of encouraging a deepening of the firms' commitment to Ireland. Finally, the world-wide growth in internationally-traded services has attracted considerable attention in recent years and has led to increasing resources being devoted to attracting projects, especially in software and financial services.

Policy Certainty

While the primary characteristics of the Irish FDI strategy are its employment focus and its promotional approach (combining automatic fiscal incentives with discretionary grants), a further characteristic is its emphasis on certainty. Recognising the negative effects of uncertainty on investment, Ireland has consistently striven to generate as much certainty as possible to the incoming and established investor through policy continuity.(18) With regards to fiscal incentives, certainty is achieved by providing the investing firms with a long and certain time horizon.(19) Promotional FDI fiscal policies are essentially independent of the annual budgetary process and of changes in government.(20) With regards to financial policies, company uncertainty is minimised by the centralised decision-making process operated by IDA Ireland and by the payment of the cash grant up-front, with repayment required if agreed employment objectives are not met.

While Ireland is not unusual in Europe today in actively promoting FDI at project level, in the 1950s when it introduced this type of policy, it was unique and hence it provides an interesting case study of the effects of. operating this kind of strategy, which inevitably takes a long time to impact on an economy. The approach taken was certainly influenced by scale - indeed, Ireland's national approach is more akin to the regional approach adopted in the 1970s in the UK regions and more recently by the municipalities of many mainland EU countries. One of the major advantages of a national strategy lies in ensuring that different regions/municipalities within the same country are not duplicating efforts and competing against each other for the same projects.

3. Impact of FDI in Ireland

Overview

We now turn to look at some empirical measures of the significance and impact of FDI in the Irish manufacturing sector. Table 1 shows shares of net output, employment and exports by manufacturing sector for 1993, the most recent year for which CSO data are available; comparable data are not available for internationally-traded services. Over two thirds of total net output in manufacturing in 1993 was accounted for by foreign-owned companies with the share of total net output generated by foreign-owned companies varying quite considerably by sector.(21) This variability is to be expected, both because of differences in the degree of international mobility of investments across sectors and because of the selective way in which policy is implemented. Looking at sectors, we find that this share ranges from less than 20 per cent in Nonmetallic Minerals to over 90 per cent in Chemicals (including Pharmaceuticals) and in the high-tech sub-sectors of Metals & Engineering (M&E), namely, Electrical & [TABULAR DATA FOR TABLE 1 OMITTED] Electronic and Instruments. As noted above, these are the sectors targeted by Ireland's industrial policy.

To the extent that foreign companies understate, for transfer pricing reasons, the value of inputs used and overstate the value of the sales generated, output could be seen as this exaggerating the importance of FDI in real activity in Irish manufacturing (See Foley, 1991). The employment figures reported in the table can therefore be seen as giving a more conservative measure of the significance of foreign firms. They show the overall share of foreign firms in total manufacturing employment at 44 per cent, less than two thirds the share of net output. The foreign share of net output by sector exceeds the foreign share of employment in all but one sector showing a similar but not identical pattern.(22) These differences may be due to (i) differences in sub-sectoral activities, (ii) differences in factor intensities in the same sectoral activity, with foreign firms being less labour intensive than indigenous firms, or (iii) transfer pricing. Nevertheless, the employment figures confirm the importance of the high-tech sectors.

Table 1 also shows the export ratios for foreign-owned and indigenous manufacturing companies by broad industrial sector. Foreign firms have considerably higher export ratios than indigenous firms, with over 85 per cent of the gross output of foreign firms being exported - two and a half times the comparable figure for indigenous firms. Note that the export ratios in foreign firms are close to 90 per cent in most of the manufacturing sectors; only foreign firms in the Food, Drink & Tobacco, Non-metallic Minerals and Miscellaneous sectors appear to serve the local market to any great extent.

Sources of Foreign Companies

As discussed above, employment creation has been the overriding focus of Ireland's industrial strategy and we attempt to examine whether this has translated into employment growth in the Irish manufacturing and internationally traded service (mainly software and financial services) sectors over the past two decades. Overall, the figures in Table 2 suggest that this aspect of the strategy has been successful. While employment in Irish-owned firms decreased by approximately 14 per cent between 1975 and 1995, employment in foreign-owned firms increased by some 43 per cent over the same period.(23) Both effects sum up to a net increase in total employment in manufacturing and internationally traded services of roughly 5 per cent. As regards the broad sectoral distribution of total employment in Ireland, the share of employment in internationally traded services increased from 0.4 per cent to 8.6 per cent between 1975 and 1995.

Turning to the source of foreign investment, Table 2 shows that the UK and the US were the most important source countries in 1975 and 1995 respectively. However, their growth patterns over the period are dramatically different. Employment in UK-owned firms has declined by over 50 per cent since 1975, a decline which began as these firms lost tariff protection as the Anglo-Irish Free Trade Area Agreement came into operation in 1965. The free trade area removed the need for UK firms, which had established plants in Ireland prior to 1932, to maintain production in Ireland in order to serve the Irish market competitively, since this could now be achieved through exports from the UK. Furthermore, the de-coupling of the Irish punt from the sterling in 1979 and the subsequent punt-sterling exchange rate variability could be seen as having had an overall negative effect on UK plants located in Ireland.(24) Also, UK industry, for further reasons, sold a higher share of its output on the domestic market than other foreign firms, and consequently was adversely affected by the fiscal recession in the 1980s, which did not impact on other foreign companies to the same extent.

Employment in US companies accounted for more than half of the employment in foreign-owned firms in Ireland in 1995, having more than trebled since 1975. Thus the policy of attracting FDI projects from the US may be considered to have been very successful over the last twenty years, a success which is also evident in Ireland's having increased its "market share" of US investment in manufacturing industries (measured in terms of capital expenditures by US affiliates) in the EU from 2.6 per cent in 1983 to 7.3 per cent in 1993 and its market share in financial services from 0.2 per cent to 5.6 per cent over the same period (Ruane and Gorg, 1996a). The growing significance of US firms in Ireland is not surprising given the fact that they use Ireland as a base to serve the EU market. The EU has become an increasingly attractive product market [TABULAR DATA FOR TABLE 2 OMITTED] for US companies, because of its rapid growth through enlargement and its increasing integration following the introduction of the Single European Market (see Aristotelous and Fountas, 1996).

While employment in German-owned firms increased between 1975 and 1995, the decline in employment numbers in firms from other European countries resulted in the employment share of continental European firms rising very modestly - from 11 to 11.4 per cent. Employment in companies originating in other non-European countries also increased considerably since 1975, though it still only accounts for a relatively small share of total employment in Irish manufacturing and services. In the remainder of Section 3, we focus on the employment performance of US and UK companies in Ireland, combining the other source countries in a category "others".

Sectoral Composition of Employment

Table 3 shows the sectoral composition of employment in indigenous and foreign-owned companies in Ireland. Following the policy orientation towards high-tech manufacturing industries, approximated here by the M&E and Chemicals sectors, and towards internationally-traded services, we would expect employment shares in these sectors to have grown particularly strongly. While the share of employment in internationally-traded services increased considerably since 1975, especially in foreign-owned firms, manufacturing is still the most important sector for FDI employment and, hence, we examine this sector in more detail. We calculate Gini coefficients to compare the relative patterns of employment across manufacturing sectors for different nationality groups in 1975 and 1995.(25) Note that the Gini coefficients for Irish-owned [TABULAR DATA FOR TABLE 3 OMITTED] and UK firms were much lower than for the other two nationality groups and that they increased between 1975 and 1995, indicating that the sectoral distribution of employment has become more concentrated over the period. In the case of UK firms, the sectoral concentration is towards the Food and Drink & Tobacco sectors, which account for almost 50 per cent of employment in 1995, rather than towards the high-tech sectors targeted by Ireland's industrial policy. In the case of Irish-owned companies the increased concentration has been towards the M&E sector. The Gini coefficient for US companies, already higher than for all other groups in 1975, increased over the 20 year period and the M&E and Chemicals sectors, which we use to proxy high-tech manufacturing sectors, together account for 80 per cent of total employment in US firms.

The data are consistent with the argument that a process of industrial restructuring towards high-tech manufacturing sectors has taken place in the Irish economy and that this process is led by US-owned firms and other foreign firms, excluding UK firms. This argument is also supported by an inspection of correlation coefficients between the sectoral distribution of employment shares for different countries in Table 4. The coefficients indicate that the sectoral distribution in Irish firms in 1995 has become much more similar to US (r = 0.58) and other foreign (r = 0.60) firms than to UK-owned firms (r = 0.37). Also, note that the correlation coefficients for 1995 between the UK and the US and 'other' foreign firms are very low (0.07 and 0.14 respectively), which indicates that the present sectoral distribution of employment is rather different.

Composition of Employment Change

Thus far, we have compared employment figures in 1975 and 1995, which provided us with a snap-shot at particular points of time. However, such an analysis cannot reflect the full extent of changes taking place within particular sectors over the period between the two benchmark years. Industrial policies can be expected to have some effects on job gains but only a weaker influence on the employment stock, since the latter reflects the impact of past history and policies. This is especially true in the case of UK firms. To take some account of these dynamics, we now turn to look at job gains and losses in manufacturing across sectors. A sector can show job gains and job losses in the same year due to (i) firms that expand (through firm entry or firm expansion) and (ii) firms that contract employment (through exit or contractions) over the same period. Thus, our measure of job gains includes all new jobs generated in firms in sector j over a given period (whether or not they still existed at the end of the period) while the measure of job losses includes all jobs lost in firms in the same sector over the same period. Net job change is the summation of job gains and job losses in a given sector. The data on job gains and losses per firm are compiled annually at firm level by the data collecting agency, and we aggregate these for the period 1975 to 1995.(26) In order to compare the employment performance for different nationalities we generate two indicators,

[A.sub.j] = [g.sub.j]/[n.sub.j],

where [n.sub.j] denotes total employment in sector j at the end of 1975, and

[B.sub.j] = ([g.sub.j] - [l.sub.j])/[g.sub.j].

Table 4. Correlations between manufacturing sectoral shares, 1975
and 1995

 Ireland UK US Other

Ireland 1 0.37 0.58 0.60
UK 0.39 1 0.07 0.14
US 0.27 0.30 1 0.99
Other 0.48 0.46 0.95 1

Note: Correlation coefficient for 1975 (bottom-left of diagonal)
compared to 1995 (top-right of diagonal)

Source: Own estimations based on data from the Forfas Employment
Survey.

The A measure expresses job gains relative to the employment base in 1975, which allows us to benchmark the extent of job creation across countries. The B measure expresses the net job change relative to job gains, indicating whether or not gross gains translated into net job creation. These are presented in Table 5.

[TABULAR DATA FOR TABLE 5 OMITTED]

There have been large numbers of total job gains and losses indicating considerable activity of job generation and destruction taking place which is not reflected in the net figures of employment change in Table 2 or implied in the share figures in Table 3. In the nationality comparison it is apparent that US firms have by far both the highest A- and B-ratios, while the UK has the lowest ratios in both cases. The B-ratio of -0.93 for UK firms indicates that the job losses were almost twice as high as job gains (-1 would indicate that job losses were double the job gains). Irish-owned firms have A- and B-ratios higher than UK-owned firms but considerably less than other foreign firms.

It is also evident from the table that job gains and losses were not uniform across manufacturing sectors. US-owned firms show very high A-ratios in the M&E and Chemicals sectors, suggesting that they have increased employment strongly in the sectors which proxy high-tech. However, the corresponding B-ratios (0.45 and 0.62) also show that job losses in these sectors were significant - a reminder that these sectors are not only innovative growth sectors, but, in individual project terms, relatively risky sectors. Irish-owned firms also have a relatively high A-ratio in the M&E sector, indicating considerable job creation in that sector. Also, the B-ratio is positive in this case, while it is negative for almost all other sectors. This supports the suggested process of sectoral restructuring towards the high-tech sectors for Irish-owned firms.

The figures for UK firms show net job reductions (negative B-ratios) in all manufacturing sectors; despite the quite considerable job gains in the M&E and Chemicals sectors, the job losses by far exceeded the gains even in these sectors. The A-ratio indicates that the relative job gains were among the highest in the M&E and Chemicals sectors, which also show some of the highest values for the B-ratio (though they are still very low compared with the other countries).

Overall, the empirical analysis suggests that the investment incentives offered in Ireland appear to have led to significant gross job gains in the targeted high-tech sectors, as proxied here by the M&E and Chemicals sectors, in Irish-owned firms and foreign firms other than from the UK. However, these gross gains have not necessarily translated into net gains of anything like the same magnitude. It should be noted that this analysis is likely to understate the extent to which policy has been influential because of the high level of sectoral aggregation in the data. A further sectoral dis-aggregation is likely to show that the job gains were in different sub-sectors to the job losses, for example, job gains occurred in the electronics and pharmaceuticals sub-sectors of M&E and Chemicals respectively, while the job losses occurred in the more traditional sub-sectors, such as mechanical engineering or basic industrial chemicals. Unfortunately, the particular data currently available to us did not allow further analysis of this issue.

Linkages Between Foreign and Indigenous Firms

We conclude this section by considering the issue of whether foreign-owned firms have created linkages with indigenous companies. We proxy the extent of linkages by analysing the values of Irish non-labour inputs purchased by foreign companies. As Kennedy (1991) points out, "[a] frequently made criticism of the strategy of attracting overseas industry to Ireland is that it did not succeed in encouraging linkages with the rest of the economy" (p. 82). Such a criticism was, inter alia, put forward by Stewart (1976) who, analysing data for the period 196470 from a random sample of 43 foreign and indigenous firms in the West of Ireland, concluded that "foreign firms in the survey sample have failed to develop forward or backward linkages to any significant extent" (p. 246). [TABULAR DATA FOR TABLE 6 OMITTED] McAleese and McDonald (1978), using data for 1974 from a census of firms in Ireland found that "new foreign-owned enterprises tend to have lower linkages than new domestic enterprises. [The] evidence also suggests that linkages tend to increase over time in new enterprises and particularly in new foreign-owned enterprises" (p. 336). O'Loughlin and O'Farrell (1980) get similar results in their analysis based on IDA data for 1976. Given the openness and smallness of the Irish economy, indigenous as well as foreign firms will be expected to use significant amounts of non-Irish inputs. As figures from the Forfas Irish Economy Expenditure data base indicate, the ratios of expenditure on Irish raw materials and Irish services respectively to total sales averaged approximately 12.6 and 10.8 per cent for foreign firms compared with 15.9 and 15.8 per cent for indigenous companies over the period 1989-1995.

Table 6 presents the data for the value of Irish raw material inputs purchased by foreign-owned firms in manufacturing industries as percentage of total raw material inputs purchased. Due to data availability, these figures exclude firms in the Food & Drink sectors, which seems reasonable since one would expect firms in these sectors to have higher linkages than firms in other sectors due to the availability of perishable inputs (McAleese and McDonald, 1978). The Tobacco sector is also excluded because of the sectoral aggregation in the data. Note that the extent of linkages increased between 1986 and 1994 by some 24 per cent, peaking in 1990. Also, note that linkages in the Electronics sector have increased even more considerably than linkages for the general manufacturing sector; however, even in this sector there appears to be some levelling off in the last two years. This levelling out or even decline in the scale of linkages, which is of some concern to policy makers, may be due to the rapid increase in new firms since 1990. As noted above by McAleese and McDonald, one would generally expect new firms to have fewer linkages with the economy, and a deepening of those linkages over time when the firm becomes accustomed to the host country. Thus, a firm level analysis might give better insights into this question; however, the present data available to us preclude such an analysis.

4. Concluding Comments

Ireland's recent success in winning large-scale FDI projects from the US, contributing to its "Celtic Tiger" acclamation, has generated concern in some quarters that the country is becoming too dependent on FDI. The fact that indigenous companies appear to be prospering at present with increasing employment and profits rates has reduced concerns about 'crowding out', so that the attitude to FDI is not hostile in any way but simply marks a concern with dependency per se - the "what would happen if FDI globally declined" scenario. Clearly it Ireland were to suffer a massive outflow of FDI it would be a disaster, given its importance in the economy. Of perhaps more significant concern is the dependency on the US market, as the portfolio of investment sources has concentrated dramatically over the years. That concern is somewhat counteracted by recognition of the real attractiveness of Ireland as a base for US investment in Europe and the scale and dynamism of the US economy at present. Also of concern is the increasing geographical concentration of foreign investment projects in Ireland which have led to new grant maxima being set for towns outside the major centres for high-tech industries. Effectively, these grants are available for investments outside Dublin, Cork, Limerick and Galway.

As noted above, Ireland was well ahead of the field in introducing incentives to encourage FDI in the 1950s and 1960s. Now governments throughout the world, and of particular relevance to Ireland, EU governments are promoting FDI projects by means of a range of incentives, typically offered by agencies at sub-national levels. Thus the relative impact of Ireland's incentives has been eroded over the past decade. Additional competition is to be expected from Eastern Europe, especially as EU membership becomes a reality for these countries. The implications of these developments have been recognised in Ireland (Forfas, 1996), reinforcing the search for projects which build on comparative national advantage as well as lower basic wage costs and fiscal and financial incentives.

It also seems inevitable that the increasing use of incentives to attract FDI projects across EU countries will soon receive more attention from the European Commission - at present only minimal attention is paid to it in connection with Article 92 of the Treaty of Rome. Furthermore, clarification of the future of the 10 per cent corporate tax rate, due to terminate in 2010, is now essential if the policy of ensuring investor certainty is to be maintained.

Irish policy towards FDI has evolved since the 1950s as a strategy driven primarily by the use of fiscal incentives to enhance the profitability of locating in Ireland, with grants as required to achieve a particular bargaining advantage in competing against alternative international locations. While employment in Irish-owned firms decreased by around 14 per cent between 1975 and 1995, employment in foreign-owned firms rose by approximately 43 per cent during the same period. Thus the original strategy of moving to free trade while retaining employment in the manufacturing sector could be said to have succeeded, with total employment in manufacturing having suffered a modest decline (3.2 per cent) between 1975 and 1995, which when combined with the recent growth in employment in internationally traded services resulted in an overall increase in these promoted sectors of 5.3 per cent. In terms of sectoral composition, there have been significant structural changes in foreign investment with an increasing share of employment in internationally-traded services and, within the manufacturing sector, a shift out of traditional sectors which were historically dominated by UK companies, towards high-tech sectors, currently dominated by US-owned companies.

Department of Economics, Trinity College, Dublin 2. Telephone x353 1 608 1325, Fax x353 1 677 2503, email fruane@tcd.ie and georgh@tcd.ie. We are grateful to Jim Bourke and Kieran McGowan for comments on an earlier version of the paper and to Ray Barrell, Nigel Pain and their colleagues at NIESR for comments on the penultimate version. We have also benefited from helpful discussions with John FitzGerald, ESRI and Donal de Buitleir, AIB Bank on issues raised in the paper. Frances Ruane gratefully acknowledges support from the Royal Irish Academy Social Science Research Council.

NOTES

(1) The major exception here is investment from the UK whose pattern of sectoral distribution is much more akin to that of indigenous industry. (See Section 3 below.)

(2) As Killen and Ruane (1993) discuss, Irish industrial policy in effect avoided the problems generated by industrial development in many LDCs where the growth of industrial employment concentrated in the larger urban areas led to massive rural-urban migration, a phenomenon modelled formally by Harris and Todaro (1970).

(3) The significance of FDI for output and exports in manufacturing have been well documented in the literature. See, for example, Foley and McAleese (1991) and O'Hagan (1995).

(4) Indeed policy can have quite different effects depending on how precisely it is implemented. While different governments offer various investment incentives, as listed by Yuill et al. (1995), it is widely recognised that the decision to invest in one country rather than another is influenced by a complex set of factors, many of which materialise during the actual implementation of policy.

(5) There are various extant reviews of policy with regards to FDI, the most recent substantial one being Foley and McAleese (1991). Other recent commentaries include O'Sullivan (1995) and Ruane (1991).

(6) Prohibition operated through the Control of Manufactures Acts introduced in the 1930s to ensure that only indigenous 'infant industry would benefit from the protectionist strategy and in particular that UK firms would not benefit by establishing new manufacturing plants in the protected Irish market into which they had previously been exporting. UK-owned firms which had already established in Ireland prior to 1932 were not forced to leave, however, so that there was a considerable amount of UK FDI in the 1950s.

(7) Since the 1970s, other fiscal policies have facilitated foreign companies in obtaining cost-effective tax-based financing on the Irish market. In some cases these policies have had a very significant impact on attracting foreign companies to locate in Ireland.

(8) This compares with the standard rate of corporate tax which has been reduced gradually from 50 per cent in the 1970s to 36 per cent in 1997. The yield from corporate taxation has increased dramatically since 1990, and corporate tax revenues now account for over 10 per cent of total net receipts from taxation, compared with under 5 per cent in 1989. The official estimate of the tax revenue foregone attributed to the preferential 10 per cent corporate tax rate on manufacturing, on the assumption that there would be no project supply effect, is just under 10 per cent of total tax revenue. However, this is clearly not a very plausible assumption!

(9) Double tax agreements are very important in this context, since the value of a tax incentive against corporate tax in the host country depends crucially on how it is viewed by the tax authorities in the home country.

(10) While companies were not totally prohibited from selling on the domestic market, they could not do so using grant-aided equipment and the return to domestic sales was significantly lower because of the high rate of corporation tax such profits attracted (circa 50 per cent).

(11) While foreign project investors argue for more grants on the basis of the attractiveness of alternative international investment locations, Irish project executives limit the grants they offer on the basis of a form of cost-benefit analysis, which considers factors such as employment potential and location of the projects within Ireland

(12) The grant maxima differ across areas, being higher in the geographically-peripheral regions and in locations which have suffered significant industrial decline.

(13) For firms with large employment/training grants, there is a maximum also subject to Cabinet review.

(14) When the grant policy was introduced, it operated as a straight capital grant. Gradually when the grant system came to be operated in a more discretionary manner, me rate of grant given (i.e., ratio of grant to capital expenditure) was influenced by the number of jobs expected to be created but it continued to be paid as a capital grant. Following further refinements (to overcome incentives for firms to exaggerate the number of expected jobs), grant limits were set in terms of both grant per unit capital and grant per unit labour, and whether the grant was paid in the form of a capital grant or an employment grant was agreed as part of the negotiation process. Where grants are still paid out in relation to capital purchases, firms are legally contracted to repay grants if they do not reach their stated employment target. Thus while the grant system in the 1950s, 1960s and into the 1970s had a capital bias, today it operates more or less in a factor-neutral way.

(15) While in many countries intervention is increasingly made at a project level, in the 1950s and 1960s this was extremely rare, except in, say, the nationalised industries in the UK. Such intervention as occurred was only likely in the case of extremely large projects, whereas in Ireland, the smallest of projects was analysed and evaluated for grant purposes.

(16) In a sense IDA personnel were collecting and responding to market information about particular firms which were likely to expand. Implicit in their approach to looking at potential foreign investment was the kind of framework developed by Dunning (1988), who suggests that foreign investment depends on (i) special firm characteristics which enable companies to produce profitably abroad, (ii) an incentive to internalise this advantage, and (iii) location characteristics in the host countries. IDA Ireland's approach involves (i) identifying industries (and firms) which had the ability to profitably locate in foreign locations, (ii) considering the means of how this foreign involvement could be achieved (in general through FDI), and (iii) analysing whether Ireland could offer locational advantages for these particular industries (and firms).

(17) For example, recent high profile investors in Ireland, such as Intel, Hewlett Packard, and IBM have been targeted by IDA for more than a decade, with regular presentations, etc. This points to the long lead time required to realise the benefits of such a strategy and the fact that the success in this area has been heavily influenced by the method of implementation.

(18) The continuity in policy has been possible primarily because there is widespread consensus in Ireland on the strategy of promoting FDI and on the use of the financial and fiscal incentives in that process. That is not to say that there are not disagreements about the support given or not given to particular projects, but by and large there is no major political debate on the fundamental issues and consequently policy has remained broadly unchanged over long periods.

(19) Thus, for example, firms locating in Ireland in the 1960s were given a fifteen-year tax holiday and those locating in the 1980s were assured that the corporate tax system which they would face would be unchanged until 2000; in 1990 this was extended to 2010. Furthermore, the ending of the tax holiday was 'grand-parented' so that individual firms who had been given tax-holiday status were able to hold onto it, despite the changes for new firms after 1982.

(20) Where changes have been made, such as the reduction in the rates of asset depreciation for tax purposes as rates of inflation declined in the late 1980s, these have been pre-announced and done in a very steady and gradual way, minimising uncertainty for firms.

(21) The share of output produced by foreign firms in Irish manufacturing industries was 58 per cent in 1983, the first year for which these data are available from the Central Statistics Office. Net output produced by foreign companies increased by some 95 per cent between 1983 and 1993, while net output in Irish companies grew by 42 per cent over the same period.

(22) The correlation coefficient between sectoral shares of net output and employment is 0.796.

(23) In that period employment in indigenous manufacturing declined by 18.9 per cent while employment in foreign-owned manufacturing firms grew by 26.9 per cent, leading to a decline in employment in total manufacturing of 3.2 per cent - employment in manufacturing in the EU declined by some 20 per cent in the same period. Thus the strong growth in employment in internationally-traded services significantly boosts the employment growth achieved by foreign companies and reduces the losses associated with Irish companies.

(24) No such sensitivity would be expected by US firms which price entirely in US$.

(25) We calculate the Gini coefficient using the formula:

G = 1 + (1/n) - (2/[n.sup.2][Mu])([y.sub.1] + 2[y.sub.2] + ... + n[y.sub.n]),

where y denotes the share of employment in the sector and [y.sub.1] [greater than or equal to] [y.sub.2] [greater than or equal to] ... [greater than or equal to] [y.sub.n], n denotes the number of observations and [Mu] denotes the mean of the variable.

(26) To put it formally, we calculate

gains: [g.sub.j] = [summation over i] [g.sub.ji]

losses: [l.sub.j] = [summation over i] [l.sub.ji]

net: ([g.sub.j] - [l.sub.j]) = [summation over i] ([g.sub.ji] - [l.sub.ji]), i [Epsilon] {76,77...,95}

Note that the first gains and losses are reported in 1976, since 1975 is our base year. Note also that the calculation of job gains and losses is not readily comparable to the calculation of gross flows of jobs as introduced by Davis and Haltiwanger (1990) and used recently by, for example, Konings et al. (1996) and Konings (1995). In particular, the two indicators described below are different from the measures used in those papers.

REFERENCES

Aristotelous, K. and S. Fountas (1996), 'An empirical analysis of inward foreign direct investment flows in the EU with emphasis on the market enlargement hypothesis', Journal of Common Market Studies, vol. 34, pp. 571-583.

Davis, S.J. and J. Haltiwanger (1990), 'Gross job creation and destruction: microeconomic evidence and macroeconomic implications', NBER Macroeconomics Annual, vol. 5, pp. 123-168.

Dunning, J.H. (1988), Explaining International Production, London, Unwin Hyman.

Foley, A. (1991), 'Interpreting output data on overseas industry, in: Foley, Anthony and Dermot McAleese (eds.): Overseas Industry in Ireland, Dublin, Gill and Macmillan, pp. 29-53.

Foley, A. and D. McAleese (eds.) (1991), Overseas Industry in Ireland, Dublin, Gill and Macmillan.

Forfas (1996), Shaping our Future: A Strategy for Enterprise in Ireland in the 21st Century, Dublin, Forfas.

Forfas (1996a), Annual Survey of Irish Economy Expenditures: Results for 1994, Dublin, Forfas.

Harris, J.R. and M.P. Todaro (1970), 'Migration, unemployment and development: a two sector analysis', American Economic Review, vol. 60, pp. 126-142.

Higher Education Authority (1995), Report of the Steering Committee on the Future Development of Higher Education., Dublin, Higher Education Authority.

Kennedy, K. A. (1991), 'Linkages and overseas industry', in: Foley, Anthony and Dermot McAleese (eds.): Overseas Industry in Ireland, Dublin, Gill and Macmillan, pp. 82-105.

Killen, L. and F. Ruane (1993), 'The regional dimension to industrial policy and performance in the Republic of Ireland', paper presented to the Conference: Economic Change and Development: Prospects for Peripheral Regions, Queen's University Belfast, 2 to 4 September 1993.

Konings, J. (1995), 'Job creation and job destruction in the UK manufacturing sector', Oxford Bulletin of Economics and Statistics, vol. 57, no. 1, pp. 5-25.

Konings, J., H. Lehmann and M.E. Schaffer (1996), 'Job creation and job destruction in a transition economy: ownership, firm size, and gross job flows in Polish manufacturing, 1988-91', Labour Economics, vol. 3, pp. 299-317.

McAleese, D. (1971), Effective Tariffs and the Structure of Industrial Production in Ireland, Dublin, Economic and Social Research Institute.

McAleese, D. and D. McDonald (1978), 'Employment growth and the development of linkages in foreign-owned and domestic manufacturing enterprises', Oxford Bulletin of Economics and Statistics, vol. 40, no. 4, pp. 321-339.

McCutcheon, M. (1995), 'The tax incentives applying to US corporate investment in Ireland', Economic and Social Review, vol. 26, no. 2, pp. 149-171.

OECD (1995), Employment Outlook: July 1995, Paris, OECD.

O'Hagan, J.W. (ed.) (1995), The Economy of Ireland: Policy and Performance of a Small European Economy, Dublin, Gill and Macmillan.

O'Loughlin, B. and P.N. O'Farrell (1980), 'Foreign direct investment in Ireland: empirical evidence and theoretical implications', Economic and Social Review, vol. 11, no. 3, pp. 155-185.

O'Sullivan, M. (1995), 'Manufacturing and global competition', in: O'Hagan, John W. (ed.): The Economy of Ireland: Policy and Performance of a Small European Economy, Dublin, Gill and Macmillan, pp. 363-396.

Ruane, F. (1991), 'The traded sector: industry', in: O'Hagan, John W. (ed.): The Economy of Ireland: Policy and Performance, Sixth edition, Dublin: Irish Management Institute, pp. 345-377.

Ruane, F. and H. Gorg (1996), 'Aspects of foreign direct investment in Irish manufacturing since 1973: policy and performance', Journal of the Statistical and Social Inquiry Society of Ireland, forthcoming.

Ruane, F. and H. Gorg (1996a), International Patterns of Foreign Direct Investment - Putting Ireland in Perspective, Report, Trinity College, Dublin.

Stewart, J.C. (1976), 'Linkages and foreign direct investment', Regional Studies, vol. 10, no. 2, pp. 245-258.

Yuill, D., K. Allen, J. Bachtler, K. Clement and F. Wishlade (eds.) (1995), European Regional Incentives 1993-94, 13th edition, Glasgow: Bowker Saur.