اثرات سیاست کمک مالی در سرمایه گذاری فناوری در ایتالیا

This paper investigates the effect of subsidies on the adoption of information and communication technology in Italy. We argue that the important geographical differences between North and South Italy are something that public policy should be concerned of. A matching estimator is used to explore the effects of financial assistance, by comparing the granted with non-granted firms. Our results suggest that public assistance has positive effects on ICT adoption. We also find that firms in the South are more dependent on public grants, and that duality would be more pronounced if subsidies were not given. Furthermore, because firms prefer traditional rather than ICT investment, subsidies should be specifically ICT oriented.

The adoption and diffusion of ICT capital goods throughout the productive system has assumed a core position in the new economy. New technology and internet can have effects which can be compared to reductions in transport costs, as well as facilitating access to more developed markets for marginal regions. This might favour the reduction of disparities between different areas. From this perspective, it is of foremost importance to pinpoint any constraints that might discourage or reduce the rate of ICT adoption (Salvatore, 2003). Incentives represent critical factors in the development of information technology and its effects on a faster productivity growth (Feldstein, 2003). Moreover, although ICT is global, public policy may play an important role locally (Iammarino, Jona-Lasinio, & Mategazza, 2004). In Italy several studies have been carried out as the new national data have become available. Becchetti, Londono Bedoya, and Paganetto (2003) investigated what determines the decision to invest in ICT and the impact of the ICT component on labour productivity and efficiency. They found that ICT investment is affected by the industry, the geographical location and the characteristics of the firm. Bugamelli and Pagano (2004) found that the relatively low value of ICT capital in Italian firms is due to certain barriers to investment, such as the low level of human capital and the organization of the firms. Fabiani, Schivardi, and Trento (2005) investigated the role of firms’ specific variables in determining the optimal rate of ICT investment. They concluded that the size of the firm, the level of human capital and the presence of large firms in the local environment played the most important role. Atzeni and Carboni (2006) analysed the link between ICT productivity and the innovative level of investment and found that computers have more effect on output growth than conventional capital. In this work we test the effectiveness of investment subsidy policies on ICT adoption in Italy, one of the advanced countries lagging behind in the adoption of new technologies. Given its economic and financial duality, Italy is a litmus paper for testing the effectiveness of investment subsidy policies on ICT adoption across regions. In this context, the Southern regions are more backwards in almost all technology indicators than the Northern ones. The fact that Southern firms are lagging behind in adopting ICT contributes to the increase of the disparity, which is something that should influence public policy. Although there is voluminous literature on the role of public support in reducing territorial disparities (Faini & Schiantarelli, 1987; Gabe & Kraybill, 2002; Harris & Trainor, 2005) there is no agreement on the effectiveness of investment incentives. The extent to which investment incentives affect economic performance has been investigated for decades, and it still is an open question, since several support plans are now being implemented in many EU countries (Braunerhjelm, Faini, Norman, Ruane, & Seabright, 2000; Yuill, Bachtler, & Wishlade, 1999). Although many empirical investigations have been carried out on the impact of subsidies on R&D investment at firm level, no specific research has been conducted on the effect of public aid on ICT. This study attempts to advance our knowledge in this area. Forms of market failures in real and financial markets provide, in principle, scope and justification for public intervention: (i) insufficient private investment in public or semi-public goods, such as monitoring (Stiglitz, 1993) and technological knowledge (Grossman & Helpman, 1991); (ii) information asymmetries leading to financial constraints and credit rationing (Bond & Meghir, 1994; Devereux & Schiantarelli, 1989; Fazzari, Hubbard, & Petersen, 1988; Hoshi, Kashyap, & Scharfstein, 1991); (iii) learning-by-doing externalities in equipment investment (De Long & Summers, 1991). For all these reasons the private return may be too low (the costs too high) to justify private investment. In these cases giving a subsidy would partly compensate for the investment disincentives. Grants might affect the financial sources that firms have access to. If these increase, the incentive has a positive effect on investment, but if they decrease, subsidies turn into simple substitutution of financing, with little effect on overall investment. In this paper a matching estimation method for the average treatment effect is employed to measure the impact of subsidies on ICT adoption. This allows us to determine whether the supported firms would have invested the same amount of ICT if they had not received assistance. We find evidence that generic subsidies to investment support ICT. Our investigation shows that subsidies are more effective for firms in the South than those in the North although these latter are more efficient in the use of grants. The paper is structured as follows. Section 2 illustrates data characteristics and descriptive statistics. Section 3 contains the methodology. Section 4 investigates the determinants of ICT adoption. Section 5 outlines the results. Section 6 is the conclusions and policy implications of the paper.

This work evaluates the effect of incentive policies on firm investment spending and provides evidence of a positive effect of subsidies on ICT. Using a large sample of Italian micro data we find that firms would have adopted less ICT had they not received grants. This result is particularly evident in South of the country where subsidies show a greater impact. Firms in the South, though having a higher capital output ratio, adopt less ICT compared to Northern ones and this is particularly evident when they are not subsidised. The weaker penetration of ICT in the South is associated with less reorganisation of work practices, size and sector specialization. Traditional sectors which are dominant in this area, are notoriously less information intensive, have a capital stock less adaptable to ICT and have a low human capital level which also strongly discourages ICT investment. Given these characteristics of the Italian productive system and the growing importance of ICT for output growth and labour productivity the role of government support and policy design represents a foremost issue. It is of particular importance on the one hand to identify possible constraints that might discourage or slacken the rate of ICT adoption and, on the other, to stimulate ICT investment. Although ICT is global, public policy can play an important role at firm level in order to attain faster productivity and growth. The rationale for the subsidy is that the private return is too low to stimulate private investment. Given the few benefits that firms may receive from new technologies, they would not be keen to invest in ICT, especially considering the amount of complementary investments that ICT requires. In such cases granting a subsidy may help to compensate for these investment disincentives. In fact, improvements in technology influence production only when they can be put into practice through appropriate capital formation or through the replacement of old processes with the latest ones. Such implementation is costly and investment specific. A well-defined set of technological purpose-oriented activities needs to be established in order to make ICT work. To the extent that these costs are mainly fixed, it may be the case that full complementarities are financially precluded for many firms, which may not access ICT or do so only partially. This becomes even more likely if firms operate in a backward environment. In such circumstances public intervention for backward areas where firms suffer from structural weaknesses may be desirable to mitigate the negative effect on technological change and growth deriving from underinvestment in ICT. ICT produces effects that are similar to transport cost reduction, since it ameliorates market access for peripheral regions. This might potentially turn into new opportunities for backward areas. In this case policy becomes crucial for reducing the regional economic gap in countries like Italy, where territorial disparities are marked and persistent. We find that the digital divide between North and South in Italy would be more pronounced if subsidies were not given. Firms without grants in the South appear to be particularly backward in the level of ICT investment, even compared with those not granted in the North. As a corollary firms in the South are more dependent on public aid when making investment decisions. This may supply an explanation (albeit partial) for the fact that despite higher interest rates, heavier credit constraints and more pronounced systematic structural weaknesses, the amount of ICT per worker for subsidised firms in the two areas is not as marked as one would expect. Subsidised firms in the South reap particular benefits from public assistance, which may help to partly overcome their external and internal constraints. Incentive policy is likely to produce a reallocation in the composition of financial sources. The availability of low-cost or totally free funds helps firms by improving their borrowing capacity or inducing entrepreneurs to direct cash flows to finance new capital acquisitions. However, when financial constraints are severe firms may simply substitute credit and internal funds with the costless financing. As a consequence the final effect of subsidies on the level of funds that firms have access to is not straight. If other funds increase, the incentive has a positive effect which is greater than the subsidy itself. Conversely, if funds decrease subsidies turn into simple substitutution of financing, with little effect on overall investment. In this study we investigate this issue considering two possible effects deriving from subsidies: efficiency (how much additional investment occurs per euro) and the composition of this additional effect (how much of this additional part is due to internal or external funds). Our results show that the overall effect of assistance is attributable to a direct effect (roughly two thirds) of the subsidy and one third to an indirect effect. A great share of the latter is due to an increment of credit and internal financing. During the period 2001–2003 these effects are less pronounced and in some cases quite the opposite. In fact, grant-supported firms in the South reduced their internal financial effort and credit availability, thus neutralising the potential effect of the program. However, non-significance of parameters does not allow a reliable interpretation for that period. Though public assistance is more efficient in the North, where treated firms achieve relatively higher additional investment, the composition of this additional indirect effect is very different in the two areas. In the South nearly three quarters is attributable to an increase in self-financing, while in the North this share is only one fifth. Differences in the access to credit and the heavy reliance on internal funds are indirect evidence of more severe financial constraints in the South. Clearly, if firms use subsidies mainly to ‘avoid’ credit, their positive role is greatly undermined. Incentive policy is more efficient in the North also because grants increase firms’ access to credit compared to those in the South. This source of inefficiency can be particularly serious for technology adoption, where complements are crucial ingredients for the profitability of the investment. The diversion of funds from other complementary investments reduces profitability of new technology and, hence, a firm's solvency. Programs with a project-related incentive involving the partnership of firms, local authorities and local development agencies (GROWTH) seem to perform better when compared to other capital grants. This is probably due to the application procedure which requires a planning activity both at firm and territorial level under the supervision of development agencies, which favours investment attainability. The methodology employed in this study and the results obtained comparing backward with developed areas supplies some useful indications for the modelling of industrial and regional policies. Given firms’ far higher propensity to general investment we suggest that public intervention should be specifically ICT oriented. Moreover, because of complementarities surrounding ICT, policy should follow a framework-oriented design. The risk is that firms invest in ICT only because of the availability of incentives while neglecting all the required complementary assets, reducing in this way the potential effects deriving from new technologies. Furthermore, incentives should affect present value calculation, increasing firm solvency and avoiding distortion in the mix between internal and external funds. Finally, policy design should take into consideration possible sources of inefficiency arising from the imperfections of financial systems. In depressed areas, where financial system is more risky and access to credit is limited, grants might need additional measures (such as development agencies for instance) in order to help firms to plan their financial function. Without grants the debt constraint may put in danger the solvency of the firm, which is forced to skip the investment opportunity. Negative effects in terms of growth and productivity are likely to occur.