The economic literature has only recently begun to focus on environmental innovations, namely those regarding the intensity of emissions and environmental performance/efficiency. There are two main reasons for this increased interest. The first is that environmental performance is now one of the main economic policy goals of European countries. The second, which is partly linked to the first, has to do with the growing impact of environmental regulation on private sector activity in many European countries.

So far, the main approach has been to consider environmental performance as tending to have a negative impact on private sector performance. In some industries – particularly those at the beginning of the chain of production (such as extractive industries) – environmental costs account for a significant share of the added value of firms, thus negatively impacting their competitiveness. The idea that improved environmental performance can be a potential source of competitive advantage for the private sector has only recently emerged. In fact, both the higher costs associated with polluting activities and the advantages coming from being the first mover (e.g. by selling these innovations to other firms) may encourage firms to acquire new technologies and production systems, which in turn entail lower production costs and thus higher productivity.

The value added of the paper is manifold: first, we provide evidence on SME productivity drivers concerning techno-organisational and environmental innovations, such as product/process innovations, EMS/ISO, R&D, TQM, training, and local idiosyncratic factors such as industrial relations and environmental policies; we thus present a rarely found rich array of drivers, bringing together standard innovation strategies and specific environmental strategies. Secondly, we base our investigation on recent and official data deriving from balance account sheets, from which we obtain productivity performances. The possibility of exploiting real performance data is also rare for SMEs. Third, in order to bypass the flaws of cross section datasets and to assess causality links, the empirical model is based on diachronic relationships between productivity and co-variates/drivers.

The paper is structured as follows. Section two presents both a survey of the literature on techno-organisational innovation, industrial relations and their effects on firms’ performance, and outlines the realm of environmental innovation drivers and economic performance, that constitutes one of the main focuses of this paper. Section three describes the data and research hypotheses. Section four discusses the empirical model and comments on the main outcomes. Section five concludes offering insights for future researches.

The conceptual background that justifies the adoption of our integrated view on innovation activities stems from contributions that can be put under the heading of knowledge-based perspective of the firm (Foss 2005). Focusing the attention on a literature (Teece 1996; Teece, Pisano 1998; Coriat, Dosi 2002) that identifies in the (dynamic) organizational capabilities both assets and activities that co-evolve with technological innovations, we can understand the importance of “new organizational practices” coupled with and complementary to technological innovation. Thus, “new” organizational practices (EC, 2002), which require/imply a more skilled workforce, the flattening of the hierarchical structure, delegation of responsibility, some degree of decisional decentralization and autonomy in managing the job tasks, coupled with technological innovations contribute to sustaining firms’ competitive advantages. (Caroli and Van Reenen, 2001; Black and Lynch, 2001; Lynch, 2007; Hall, Lotti and Mairesse, 2007)

The human capital of employees becomes a fundamental resource since “innovating organization benefits from a strong skill-base” (Leiponen, 2005, p. 304). Thus, the importance of training activities (Zwick, 2004; Conti, 2005), which help generating and accumulating skills and competences complementary to technological innovations, is clear. Put another way, the organizational design should include human resource management practices that imply higher levels of employees’ commitment and involvement. A less explored idiosyncratic element that follows, somewhat linked to organizational and human capital strategies, and that potentially influences the firm performance, is the industrial relations system. Trade unions may be an element that enhances or hinders the productivity of a workplace as well as its profitability, through an influence on innovation activities (Booth 1995; Menezes-Filho, Van Reenen 2003; Metcalf 2003; Menezes-Filho, Ulph, VanReenen 1998; Machin and Wadhwani, 1991)

Although the conceptual literature does not provide unambiguous insights (Menezes-Filho and Van Reenen 2003; Metcalf, 2003; Naylor, 2003), it is common to find evidence of a neutral or positive influence of unions on productivity and negative influence on profitability. Whether the positive effect overcomes the other, qualifying unions as surplus-enhancing, should be analysed case by case. In fact, if industrial relations are cooperative then we would expect more workforce commitment to the firm, a higher morale and a more stable environment with positive consequences on the firm’s efficiency. On the contrary, if the industrial relations are adversarial, unions act in a conflicting way, the management may ignore the union voice and there is no kind of alignment between unions and management goals, then we would expect negative effects.

Eco-Innovation (EI henceforth) is becoming the conceptual reference point for many regional and international public policies and management strategies. One of the most recent definitions of EI defines it as the production, application or use of a product, service, production process or management system new to the firm adopting or developing it, and which implies a reduction in environmental impact and resource use (including energy) throughout its life-cycle. This definition includes innovations whose environmental effects are not intentional. A relevant distinction can be made between end-of-pipe technologies and clean technologies integrated in the production process. In addition to these two categories, there are eco-products, eco-processes, eco-management systems, and broader environmental innovation systems.

Studies on environmental innovation and its relationship with socio-economic performance originate within three analytical fields within the framework of environmental economics and management/innovation studies. The first research field regards the efficacy and dynamic efficiency of environmental policy tools as objectives to be evaluated along with static efficiency, in both economic and non-economic terms. (Requate and Unold, 2003) The analysis of efficacy, linked to a specific goal of ex-post evaluation and analysis of the effects on innovative paths induced by policies, is complemented by the analysis of dynamic efficiency. (Milliman and Prince, 1989; Hahn and Stavins, 1992) The second research field regards the “Porter hypotheses” on the links between environmental regulation and innovative strategies on one hand, and on the other hand between the performance/competitiveness of firms and CSR strategies. On the heels of the works of Porter many authors tackled the issues of competitiveness effects on the manufacturing sectors most vulnerable to competition and regulations (Jaffe et al., 1995), and the motivations of CSR, whether explainable by a profit-only motive or by the production of mixed public goods. (Reinhardt et al., 2008) An understanding of the following aspects is crucial: whether the firm is merely compliant with existing regulations or goes beyond them; whether it would have pursued innovation even in the absence of regulations; how costs and benefits are distributed along the dynamic path (from short to long term) that follows innovation.

For a detailed critique of the Porter hypothesis (Porter and Van der Linde, 1995) we refer among others to Jaffe et al. (1995) who empirically and theoretically address the hypothesis that environmental regulation may benefit affected firms. This hypothesis takes different gradual forms from “strong” (private costs are in the end lower than private benefits, even excluding social benefits) to light assumptions (net costs remain positive on average with some sectors turning out winners and other losers in the “environmental” oriented economy). The stream of diverse analyses stemming from this original debate is to some extent a reaction to the neoclassic frame, and points, from different perspectives to a possible complementarity/positive correlation between labour productivity and environmental efficiency. Complementarity may be opposed to the “substitution hypothesis” which derives from a usual neoclassic reasoning. In fact, if the firm is optimizing resource allocation in production before environmental regulations, any additional abatement cost or innovation cost deriving from policy enforcement leads, at least in the short run, to an equal reduction in productivity, since labour and capital inputs are re-allocated from “usual” production output to “environmental output” (pollution reduction). As stressed by Krozer and Nentjes (2006) this emphasis on substitution and negative relationship between the two efficiencies may stem from the role played in the neoclassic reasoning by both an assumption of optimal allocation of resources in the BAU (Business As Usual) and by the role of prices as innovation levers.

In the European setting, evidence on environmental innovation is recently provided by Frondel et al. (2004), who exploit OECD survey data for Germany at firm level (manufacturing industry), in order to investigate whether environmental auditing schemes (voluntary management-oriented organizational innovation) and pollution abatement innovation are correlated. The main conclusions are that the enhancement of corporate image is a potential force behind the adoption of EMS, while policy inputs do not seem to affect this organizational innovation.

Going on with works in the EU scenario, Horbach (2008) exploits a two year panel considering firms offering goods or services related to the reduction of environmental impacts, then subdividing between innovative and non innovative firms. A large vector of explanatory methods is tested, ranging from firms’ strategy to policy-related factors. Rennings et al. (2003) also provide evidence on Germany, deeply focusing on auditing schemes like EMAS and correlated environmental organisational innovations. They find that EMAS has a positive effect on all three forms of environmental innovation at firm level, with a key role played by the R&D department.

Two other papers deal with the correlation between EMS and technological innovation. Ziegler and Rennings (2004) analyse the hypothesis that EMS and other environmental organisational innovations correlate to process/product technological innovations, exploiting a sample of German manufacturing firms [1]. While R&D is positively related to both product and process environmental innovation, they find only weak evidence in favour of the correlation between EMS and innovation (for other evidence on the US see Florida et al., 2001; Khanna & Anton, 2002). Summing up, although evidence does not always robustly show a positive correlation, there is certainly not a trade off between technological and organisational environmental innovations; if not correlated, they may impact on the environmental and general performance of firms separately.

Frondel et al. (2004) instead focus attention on technological process innovation, testing whether end-of-pipe measures or integrated cleaner production processes are driven by different factors. They use an OECD survey-based dataset for 7 countries on manufacturing firms. Main results are that policy stringency is more relevant for end-of-pipe innovations, while “market forces” like R&D, environmental accounting systems and audits, and cost saving motivations are more relevant for cleaner technologies.

Gray and Shadbegian (1995) instead use as performance indicators total factor productivity and growth rates for plants in paper, oil and steel over 1979-1990, testing the impact of environmental regulations and pollution abatement expenditures. This is a typical example of a paper which is embedded in mainstream reasoning around environmental strategies and firm performances. They find that $1 of greater abatement costs is associated with $1.74 in lower productivity for paper mills, $1.35 for oil firms and $3.28 for steel mills. These figures are variations across plants in productivity levels. Instead, when analysing variations over time or growth rates, the relationship between abatement costs and productivity, as well as the impact of other regulatory measures, is statistically insignificant. The evidence on the “Porter hypothesis” is thus ambiguous: regulations do not increase long run firms’ performances, in this case productivity levels, but on the other hand a negative undermining effect is present only cross-sectionally. Greenstone (2001) estimates the effects of environmental regulations (Clean Air Act) on industrial activity, using data for 1.75 million US industry plants. The author stresses that “regardless of whether these policies pass or fail a cost-benefit test, this paper’s findings undermine the contention that environmental regulations are costless or even beneficial for the regulated”. (Greenstone, 2001, p. 1213)

Finally, we direct attention to recent EU-based studies which, though limited by lower data availability compared to US, focus the lens on the effects of more general environmental (innovation) strategies and performance. Some studies still focus on stock market performances of corporations, by using standard cross section/panel approaches. (Ziegler, Schroeder, Rennings, 2007). A rare set of studies that use real environmental performance to study the co-evolutionary effects between firms’ performances (turnover and profits) and firms’ emissions intensity are Earnhart and Lizal (2006, 2007, 2008), which exploit a 1996-98 panel of Czech firms. Finally, Cainelli et al. (2007, 2008) provide dynamic evidence for services and manufacturing firms, testing effects on various firms’ performance of green strategies and environmental performance captured by (NAMEA) emissions/value added ratios.

There is still room for research in fields that have not yet been fully explored, and there is a need for greater integration between analyses of the drivers and effects of environmental innovation. Overall, evidence obtained so far is neither robust nor univocal, and leaves room for new contributions that can identify empirically the direction and intensity of the relationship between environmental innovations and economic performance. Therefore, we feel that continued research in this field would achieve maximum added value by fully integrating the studies on the drivers and effects of environmental innovations (Mazzanti and Montini, 2010). Although valuable, and grounded on official datasets, we believe that evidence focusing on the corporate sector and stock market is limited since the great majority of firms, especially in Italy, are of medium and even small size, and are not present in stock markets. Then, even more relevant, we believe that innovation dynamics are closer to productivity trends which, in the end, form the leading engine of firms’ performances, including profitability.

The intersection of the first two surveys clearly left us with a dataset of 144 firms. Merging this set with the balance sheet data we were able to collect, leads us to a final sample of 89 firms with unique characteristics in the field of innovation and environmental studies. As far as the research hypotheses are concerned, the signs of the relationships among the sets of co-variates and the productivity variables are synthesized in table 1. Let briefly comment on the underlying hypotheses. First, we expect to find a positive effect of techno-organizational innovation and HPWP in general, including training, on labour productivity. Second, the highly relevant role played by management-union and management-worker relationships in highly unionized industrial areas such as the districts in Emilia Romagna cannot be neglected. The assumption is that good (cooperative) industrial relations and good (participative) workers’ involvement in some business strategies entailing techno-organisational innovations and HPWP (High Performance Workplace Practices) could cause higher performances in such “participative” oriented firms.

The role of the environmentally-related activities on labor productivity is more ambiguous and we consequently may expect both positive and negative signs. On the one hand, if we focus attention on the role of EI factors in stimulating economic performance the hypothesis is that, taking a Porter/CSR perspective (Porter, van der Linde, 1995; Jaffe et al., 1995; Reinhardt et al., 2008), EI strategies may bring about positive effects on economic performance. On the other hand, the adoption of EI can merely be driven by non strategic reasoning (compliance with a policy).

A last factor we investigate is related to the impact of policies on firms’ performance. We test this link, which may be positive, not significant, or negative, depending on the same reasoning we carried out above. In fact, environmental regulations can spur positive private benefits for firms, in the medium long run, only if firms react by innovating and taking a full CSR pro-active action. Because the time lag we analyze is embedded in the short run, we do expect negative or negligible effects on productivity.

Table 1
Hypothesized signs of influencing variables on the productivity (VA/E)

As a general specification of the econometric model we estimate the following form:

(1) Prod (VA/E)_i,04 = β₀ + β_1i,98-01 [structural variables] + β_2i,98-01 [techno-organisational innovations/ HPWP] + β_3i,98-01 [industrial relations] + β_4i,01-03 [environmental innovations] + β_5i,01-03 [environmental policy/cost] + ε_i

where β is a vector of coefficients for each set of co-variates, the dependent variable refers to 2004, i identifies each firm, the sets of co-variates refer both to the periods 2001-2003 and 1998-2001. The variables in the above reduced form may be briefly described as follows, while for detailed information on their construction we direct the reader to table 2 [4].

Dependent variable. The productivity indicator is constructed on the bases of balance sheet information for the year 2004. The ratio between the firms’ value added and the number of employees (VA/E) is an expression of labour productivity.

Structural variables. The vector of structural variables captures as far as possible the firms’ specific heterogeneity in our “hybrid” cross-sectional environment, in order to reduce, to some extent, the likelihood of relevant variables being omitted.

Techno-organisational innovations/HPWP. Innovation variables encompass several spheres. In fact, the questionnaire administered allowed collection of information not only on product and process innovations, but also on organizational changes, including those practices that international literature defines as High Performance Work Practices (TQM, JIT, quality circles, job rotation and team working organized in a single index, plant flexibility, employees involvement), and training activities, such as the coverage in terms of employees involved in training practices. The expected sign associated with the variables belonging to this broad set is positive.

Industrial relations. The index of industrial relations takes into account both formal (e.g. presence of bilateral technical commissions or existence of firm level bargaining) and informal aspects of the firm level relations between management/union representatives and management/workers. It represents a synthetic index of the cooperation intensity between the social parts at firm level; thus, we can expect a positive relation with productivity, as suggested by theoretical insights.

Environmental innovations and policy/costs. The richness of environmental-related information represents a further value added of the data at our disposal, when compared to other international studies on the relationship between EI, environmental policies and firm economic performance. The set of variables here considered includes information on firms’ activities concerning both environmental innovations and aspects related to the “green” policy compliance.

Table 2
Descriptive statistics and variable contents (89 firms)

Quite interesting and counter-intuitive to some extent, group membership is related to productivity through significant negative links: the underlying reason may be that economic performance in such local systems has often resembled a bell shape with respect to size. Medium sized firms are those performing better; large firms are mainly involved in group strategies, then the evidence we find is consistent with other recent evidence on this manufacturing system (Antonioli et al., 2010).

Within the bundle of innovative and HPWP strategies, training and “organisational innovative actions” are those arising as more significant. More specifically, training coverage (COV), the additive total index of HPWP (INNO_ORG, embedding quality circles, TQM, team working, JIT and job rotation) adoptions and the use of monetary bonuses related to employees’ efforts (BONUS) emerge as main drivers.

Quite counter-intuitively, product innovation adoptions do not impact on productivity, and process innovation is associated with a negative and moderately significant coefficient. This is nevertheless consistent with findings in recent investigations on the same local production system (Antonioli et al., 2010): technological innovations rank only third as performance drivers, with a role definitely lower compared to that of training and organisational innovations. Consistently with expectations, training and organisational innovations impact on labour productivity. All in all, organisational and human -oriented strategies seem to matter more as far as economic performance is concerned. The reason behind this lack of effects from technological innovations may depend on the different time lags technological and organisational innovations take to have impacts on productivity. In addition, the negative sign of INNO_PROC could also depend on the diverted investments necessary for adopting such innovations, that in the short run crowd out other productive value-creation strategies more dependent on current technological assets. Finally, we analyse, as in most works, the adoption in terms of binary variables. The intensity and “radicalness” of innovation are other aspects worth investigating in future works.

Analysing the single HPWP elements included in INNO_ORG, what we observe is that its significance is driven by team working, just in time, and quality circles, all highly significant, but not by TQM and job rotation (we do not show such ancillary regression with separated effects, but they are available upon request). The reason why TQM is not significant may depend on the quite general content of this HPWP, which may sometimes present forms of formal rather than effective adoption. When we consider the index capturing the intensity in employees involvement (INNO_PART) it emerges positively linked to labour productivity (table 3, column 2), as found in other studies. (Addison et al, 2000, Pencavel, 2003) Partially shifting decisional powers to the employees and according them higher levels of autonomy and responsibility through organizational re-designs, seem to be helpful in determining higher labour productivity. Contextually to the result of INNO_PART we have to stress that the index synthesising “good” industrial relations (IND_REL) is not significant, thus not reported for brevity. This result does not come totally unexpected, because other papers on Reggio Emilia found effects of industrial relations on firm performances, through impacts of industrial relations on innovation intensity at the level of the firm. (Antonioli, 2009) [6] The good quality of industrial relations is a factor that may more indirectly (through a stimulus to innovations) than directly increase firms’ performance. With respect to the first specification we note that in column 2 both F test and R² increase in value after including more co-variates, while the constant loses explanatory power, and training (COV) impacts more, while other evidence is untouched [7]. INNO_ORG is omitted in this case for its high correlation with INNO_PART.

Third, let us present what the role of environmental strategies is when included (table 3 columns 3,4). From a statistical viewpoint, we note that they further improve the fit of the regression, showing a potential case of omission of relevant variables if they were excluded.

Auditing schemes (CERTIF, containing both EMS and ISO schemes, adopted by around 30% of firms) adoption by firms is positively associated with productivity. The significance appears nevertheless to be driven by ISO14000 [8], when we disentangle the total variable by using specific dummies. This is somewhat counter-intuitive given that EMS is a stronger strategy in innovation terms. Nevertheless, this evidence may be consistent with the fact that EMS is rarer in comparison to ISO in small and medium firms. In addition, EMS showed a strict link to output process-product environmental innovations (Mazzanti and Zoboli, 2009b), which here in fact do not present connections to productivity, confirming the evidence we just found for non-environmental process and product innovations. Though some lack of significance for those technological innovations of either kind may depend on the short lag between adoption and our productivity performance, all in all the evidence we present supports stronger impacts for (eco-) organisational innovations. Alternatively, we may assume that organisational innovations tend to influence labour productivity through a closer lag with respect to technological innovations effects. ISO14000 obscures the role of INNO_ORG, though inside the general index we found that different elements play a role with different strengths.

As far as environmental R&D (expenses per employee) is concerned (table 3 column 5), a factor which has been seldom investigated in empirical studies (and a major result of this analysis) is that its inclusion causes another sharp increase in the fit in terms both of R² and F test. The RSN variable is highly significant at 5% if corrected for heteroskedasticity, a plausible result given its distribution, and does not affect the significance of other co-variates. If we consider jointly auditing and R&D, the latter does remain significant; then, environmental strategies appear to play a key role even towards firms’ performances, with an already noted relatively higher strength of eco-organisational innovations. All in all, both EI levers (technological and organisational) play a role in determining firms’ productivity. As for standard innovation, instead we do not find a significant effect for the adoption of EI (process and product). This may depend on the lag such adoptions need to exert effects, or on the need to investigate “intensity” of innovations (e.g. expenses, radicalness) rather than mere yes/no adoptions as done in most studies including ours.

Regarding R&D relationships with productivity, it is a sign that environmental investments are not undermining firms’ performances even in the short run, and they also bring about significant effects. They might be targeted to increasing efficiency at a general level of reasoning, and they also may generate partially appropriable rents [9]. The significant role played by R&D as a technological driver in explaining performance brings back the potential of innovation for increasing firms’ performance. We note the fact that it is the level of eco-R&D expenditures per employee, and not merely the presence of some R&D in the firm generally (measured and tested by a dummy, not shown and not significant), which becomes significant. This may suggest that it is the intensity of innovation, and not just the simple adoption of innovations, that matters. This may explain to some extent why some other output innovation factors, here captured by dummies, end up with being not really significant. This intensity is usually difficult to elicit and capture, especially for intangible and non monetary factors. Further research should in any case focus more on the differences between both radical and incremental innovation and between intensive efforts and less intensive ones. Monetising expenditures may be one way among others to ground the analysis on more specific and concrete figures.

Finally, the “policy proxies” we include show mixed evidence. On the one hand, the number of years waste policies (WASPOL) are experienced in a given firm seems to lower performances. This is not emerging for emission-related policies, maybe more recent policies as well. The interpretation may be also that regarding emissions, firms tend usually to first introduce end-of-pipe technologies rather than fully restructuring processes. The latter option may have at least in the short run a negative effect on productivity, which is not otherwise present when end-of-pipe technologies are implemented. End-of -pipe strategies are less likely to be feasible for waste, given the nature of the environmental problem, which needs from the beginning a structural approach. It is also relevant to stress that, while waste policies were already implemented in the early and then late nineties (packaging waste EU directives), emission policies have witnessed a more recent development, and often SMEs are not covered by emission taxes or emission tradable permit schemes at EU and Italian levels.

We also note that the variable capturing current environmental expense/costs per employee (COSTN) is not significant in any regression. This evidence is against a mainstream hypothesis by which environmental policy costs decrease productivity, as discussed in the literature survey above. Added to other evidence, it seems to corroborate some sort of Porter reasoning, by which competitive strategies and investments in innovation assets, including environmental strategies and environmental innovation, may increase firms’ performance, or at least do not depress it.

Table 3
Labour productivity effects of (eco) innovation strategies

Results show that regarding standard innovations, training and organisational innovations are, as expected, the main significant drivers. The fact that control structural factors do not exert effects on performance (besides some negative performance linked to the chemical firms) is a signal that innovation strategies are more relevant than “exogenous” facts, at least in the medium run, such as size and sector of production. The role of technological innovation is instead less relevant here, signalling a possible weakness in terms of innovation intensity and/or a longer time span for such techno-innovation in determining real productivity effects. This relative ranking nevertheless confirms the results of previous studies on the same firms.

Environmental strategies, specifically auditing schemes, especially ISO14000, and expenses in environmental R&D, appear to improve the fit of the model, and both impact positively on productivity. This is the key result of the paper. Eco-R&D, a specific factor we test here and which is seldom accounted for in its specificity (the next CIS covering EI for the first time will not deal with eco-R&D), re-integrates into the picture the role of technological innovation. It is also worth noting that current expenses associated with environmental costs do not seem to exert any negative effect on performances, as possibly expected. Some criticalities emerge concerning waste policies, but not for emissions. Comprehensive innovative SME strategies thus seem to effectively impact on firms’ performance reducing, or even eliminating, trade-offs between standard business actions and strategies encompassing environmental objectives. In addition, local idiosyncratic factors such as participation and involvement of unions and employees in managerial decisions towards innovation actions correlate positively with productivity. Environmental innovations seem to be well embedded in the multi-innovations set of potential drivers firms may exploit to enhance their competitiveness in the medium-long run, and environmental strategies do not even have a negative impact in the short run we here focus on.

Future analyses could improve the understanding by studying medium-long run type of lags between the time innovations occur and consequent performance is experienced. The analyses on complementarities or trade-off between different eco- and not environmental innovations, and within categories, is also a fruitful direction for research. Finally, the construction of panel data sets derivable from consequential surveys on the same firms is another potential value added from a methodological ground, also highlighting the role of the dynamics.