Key Indicators for Improving the Resource Productivity in the Baltic States

The importance of resources

Natural resources are essential for the economy and human well-being and consumption of them should not exceed sustainable levels. They provide substantial raw materials and other commodities, and are an important source of production, income and jobs. However the use of materials has environmental, economic and social consequences (OECD, 2015).

There are lots of reasons why the resource problems should be placed in a dynamic perspective. One of the main objectives of natural resource economics is to better understand the role of natural resources in the economy in order to develop more sustainable methods of managing the resources in order to ensure their availability to future generations. Current use of non-renewable resources, such as oil or gas, determines future resource availability. Renewable natural resources regenerate in a dynamic ecological process, but it can be disturbed by commercial activities. Bretschger & Smulders (2007) found that macroeconomic dynamics become highly relevant for the resource scarcity and pollution. According to the researchers, capital accumulation and technological change are essential to offset the increasing scarcity of natural resources and to promote sustainable development. In particular, the development and adoption of new technologies allow improve the efficiency of use of resources. Social dynamics are also important. The behaviour of users of natural resources and polluters, as well as policymakers, changes over time because of learning behaviour, changing perceptions, the building-up of new information, and the reaction thereupon (Bretschger, & Smulders, 2007).

2.2 The conception of productivity

In general, productivity represents the relationship between the outputs of the production system and inputs that are necessary for the creation of these outflows (Aspridis et al., 2014). In more detail productivity is represented by the rate of input production factor for the production of a product or a service, and includes four factors – resources, workers, facilities, and technology and management (Joo, 2011). Therefore, productivity can be defined as an index of the product compared with the production factors, including labor productivity and resource productivity, or as the total factor productivity that reflects all production factors. The total factor productivity can reflect all the production factors, but it makes the calculation and analysis complicated, whereas the productivity based on a single production factor cannot reflect the effects of the other factors (Lee et al., 2014).

Labor productivity is the most widely used indicator of productivity and it is the result of a system related to human resources that produce the outcomes. Labor productivity reflects the overall effect of various factors (such as physical capital, technology, human capital, organization of work and other) to the result (Aspridis et al, 2014).

Mulder and Groot (2007) investigation on the development of cross-country differences in energy and labour productivity showed that it differs and cross-country variation of productivity levels is typically larger for energy than for labour. Energy prices and wages have positive affect on energy and labour productivity growth while the investment share, openness and specialization play only a modest role in explaining cross-country variation in energy and labour productivity growth.

Labour and energy productivity was also analysed by Achour (2016), Broadstock (2016), Calcagnini et al. (2016) and others. As well, the researches on total factor productivity (Farhadi et al., 2015; Long et al., 2015, Bah, & Fang, 2015) or consumption of biomass (Bringezu, 2012), water and forests can be found. The productivity is analysed at the national level, as well as various sectors, i.e. transport, manufacture, agriculture and so on.

However, the resource productivity is becoming the core element of the total factor productivity as the cost of production can be reduced in terms of resources via process innovation and design change. Resources change into the products that have added values through the general production process. The change in resources is directly applied to the added value of the product, while the changes in other factors are indirectly applied to the changes in the added values of the products. Therefore, the most efficient productivity evaluation method is to use the resource productivity index, which means the ratio of the resource to the resulting added value (Aspridis et al, 2014).

Efforts on improving the resource productivity

A growing population is getting richer and demands more and more products which require natural resources and create various environmental impacts (Bringezu, 2011). One key strategy is resource efficiency, i.e. to make more out of less, to generate more wealth and well-being with less input of natural resources. For industry, higher material and energy efficiency is a chance to reduce costs and enhance competitiveness. The search for eco-efficient technologies is thought to trigger innovation.

Resource productivity, measured as GDP output per resource input, is a widespread sustainability indicator combining economic and environmental information. High resource productivity is interpreted as the sign of a resource-efficient, and hence more sustainable, economy (Steinberger, & Krausmann, 2010).

The researches demonstrate that resource productivity is influenced by national income and its current use tends to support a simultaneous growth in economic productivity and resource consumption. A growing number of countries have already defined targets to increase resource productivity of their economy. This is requested by the Thematic Strategy for the Sustainable Use of Natural Resources (CEC 2005). OECD Council also adopted a recommendation that encourages its members to improve resource productivity by promoting environmentally effective and economically efficient uses of natural resources and materials at the macro, sectorial and micro levels as well as to strengthen capacity for analysing material flows and the associated environmental impacts. The nowadays challenge is to create the higher value with less natural resource input, and thereby do not compromise the needs of future generations.

Bringezu et al. (2012) proposed a comprehensive approach to account for the global land use of countries for their domestic consumption, and assess this level with regard to globally acceptable levels of resource use, based on the concept of safe operating space. They showed that the European Union uses one-third more cropland than globally available on a per capita basis and that with constant consumption levels it would exceed its fair share of acceptable resource use in 2030.

Hence governments need a reference point for assessing the sustainability implications of their policies on global resource consumption. Each country must be sure that the thresholds for sustainable resource use, emissions and waste disposal are not exceeded. In reality, however, only few countries have the capacities to monitor and control their domestic resource extraction, wastes and emissions (Bringezu et al., 2012). Nevertheless EU countries are making their efforts to improve resource productivity. Domestic material consumption indicator is routinely reported by Eurostat and analysed by various researchers.

Many countries in the world joined the initiative and have included resource productivity issues in their sustainable development strategies, established programmes on sustainable production and consumption, stewardship programmes for materials and natural resources, and have introduced integrated waste and materials management policies. Business sectors address these issues by establishing stewardship programmes for materials and products, investing in R&D and using advanced technologies in order to increase materials and energy efficiency, enhancing environmental management, promoting eco-design and coherent materials supply and use systems (OECD, 2008 and 2011). Moro and Stucchi (2015) state that changes in the distribution of resources across firms can have different effects on aggregate productivity depending on the elasticity of substitution among goods.

R&D is the main source of firm absorptive capacity and therefore, along with to introduce innovation, this factor is crucial to remain close to the technological frontier (Griffith et al., 2004). The contribution of foreign R&D to productivity growth often exceeds that of domestic innovation, especially when one considers R&D performed by third countries which is (indirectly) embodied in the products of our (direct) trading partners (Lumenga-Neso et al., 2005).

Giljum et al. (2015) found that the differences in material footprints per capita are huge, ranging from up to 100 tonnes in the rich, oil-exporting countries to values as low as 1.5 to 2.0 tonnes in some developing countries. Some researches indicate that the levels of material use per capita and the composition of materials used can be quite different even among highly industrial economies of similar per capita income levels. Weisz et al. (2006) have analyzed DMC for the EU-15 countries in order to identify possible reasons for cross-country variations in the levels of material use. They found that variability of DMC is in a similar order of magnitude as the variability of GDP per capita or total primary energy supply per capita. Linear correlation analysis reveals that national income and final energy consumption relate to material use but cannot fully account for the observed differences in material consumption (Weisz et al., 2006; Kalmykova et al., 2015). Steinberger and Krausmann (2010) showed that different types of materials exhibit fundamentally different behaviors, depending on their international income elasticities of consumption.

Wang et al. (2012) examined driving factors of changes in recent resource use in China and found that the affluence factor contributed most to the increase of direct material input. Schandl & West (2012) showed that Australia, China, and Japan have diverging patterns of resource use, and that these patterns can be linked to interdependencies between them and the very different roles each nation plays within a globalized system of natural resource exploitation. However, these studies only estimate domestic resources extraction and direct resource trade, without linking resources extraction with final products consumption within a globalized system of resources use (Wu et al., 2016).

Although resource consumption varies with income, resource productivity can be not dependent on income. This is because resource productivity is a ratio of income and consumption. So if consumption is proportional to income, there will be no variation in productivity. That‘s why the resource productivity is not a good indicator of resource efficiency according to Steinberger & Krausmann (2010). Other authors have also expressed skepticism toward resource productivity as a robust or informative indicator. Ang (2006) pointed out that both the resource and economic measures are rather arbitrarily weighted composites, with contributions from many sectors at varying prices for GDP.

Nevertheless the efficient use of resources and improvement of resource productivity are one of the main objectives nowadays. The improvement of resource productivity is a long lasting process and needs a good understanding of the material basis of the economy, greater coherence of policies relating to resource use and materials management, strengthening the partnerships with the private sector, research, and civil society and so on (OECD, 2011).

Kalmykova et al. (2015) after the analyses of the material flows at the national (Sweden) and urban scale declared that the resource consumption trends indicate that the implemented policies have failed to bring significant reductions in resource and energy throughput. Nations must urgently reduce the consumption of all resources. The EU’s long-term growth strategy, “EUROPE 2020”, chose efficient resources as one of its major implementation themes to emphasize efficient resource management for economic development (EC, 2010), and proposed the direction of the long-term policy for the increase in production efficiency and the use of wastes as resources (EC, 2011). Tarasyev et al. (2015) analysed a dynamic mechanism for optimization of resource productivity within the economic growth model framework. The optimal control problem is posed to optimize investment in capital, as a basic factor of production, and investment in technology for raising resource productivity. According to the authors, there is urgent need to explore new approaches in economics that take natural resource into an analytical framework. However Bringezu et al. (2012) and others think that globally acceptable levels of resource use can hardly be calculated and derived directly from any modeling.

Tendencies of resource productivity in the region

The analysis of resource productivity in Europe shows that the Baltic States are those where this question should be solved strictly as the resource productivity in the region is 1.5-3 times lower than the average resource productivity of all EU (2.1) and is one of the lowest in the region (Fig. 1). According to the data of 2014, the resource productivity in Lithuania is the highest (1.4) among three Baltic States and almost twice higher than the resource productivity in Latvia (0.9) and Estonia (0.7).

See Full Size >

Moreover, the progress to increase it during the last decade was low. The average resource productivity of EU increased by 34.2% since 2000 (till 2014), while Latvia was the only country among three Baltic States that managed to exceed this growth (Fig. 2). The resource productivity in this country increased by 43.5% during this period and outstripped the resource productivity in Estonia. Lithuania improved this indicator by 20.5%, while Estonia is one of a few that fixed negative change (-26.2%) of the resource productivity during fourteen years (the resource productivity has also decreased in Malta and Romania).

See Full Size >

The analysis tends to a conclusion that the resources in the Baltic States are used inefficiently. It raises a great concern and should be one of the most important questions that should be solved without delay. That‘s why it is important to identify the main factors that have influence on the resource productivity and put effort in supporting them in order to improve the situation in the region.

Identification of the factors that have influence on the resource productivity

Such indicators as national income, technologies used in business as well as investing in R&D are mentioned as the most important for the resource productivity in various researches. Gross and net national income per capita rose consistently in all three countries since 2000 (except it fell slightly in 2009) and increased three times during thirteen years (Fig. 3). In turn, it has had to increase the use of the resources.

See Full Size >

Domestic material consumption grew more moderately (Fig. 4). DMC in Latvia increased the least, i.e. 17.7% during fourteen years. DMC in Lithuania rose 49.1% while it increased twice (121.7%) in Estonia.

See Full Size >

Reduction in use of resources can be reached by improving the processes and technologies. This intent and effort can be represented by the expenditures on research and development. Total R&D appropriations per inhabitant in Latvia are the lowest in Latvia and they are 2.5 times lower than in Estonia (Fig. 5). Growth of R&D appropriations and decline of resource productivity at the same time leads to a conclusion that R&D expenditures in Estonia are not directed to reduce the consumption of resources or efforts to do that are not effective.

See Full Size >

In order to make the more detailed analysis of various indicators and their impact on the resource productivity the correlation analysis and Granger causality analysis will be done. Moreover more indicators related to outlays for research and development and national income will be taken into account. The further research covers the analysis of the following indicators:

• government budget appropriations or outlays for research and development (GBAORD) for environment, transport, telecommunication and other infrastructures, energy, industrial production and technology (all in euro per inhabitant and as a percentage of GDP),

• total R&D appropriations (euro per inhabitant and as a percentage of GDP), gross domestic expenditure on R&D (GERD) in PPPs USD, publically financed GERD as a percentage of GDP,

• total intramural R&D expenditure (GERD) of all sectors, GERD of business enterprise sector, GERD of government, GERD of higher education sector (all in euro per inhabitant and as a percentage of GDP),

• business enterprise expenditure on R&D (BERD) as a percentage of GDP, BERD financed by government (direct) as a percentage of GDP, BERD of small and medium enterprises (SMEs) as a percentage of total BERD, BERD of large firms as a percentage of total BERD,

• net primary income transfers with the rest of the world, primary incomes payable to the rest of the world, primary incomes receivable from the rest of the world, gross national income at market prices, net national income at market prices, net national disposable income, gross national disposable income, final consumption expenditure, domestic demand, gross value added at basic prices (all in euro per inhabitant).

The analysis covers an annual data from 2000 till 2014, but some indicators have shorter time series.

Correlation analysis

The correlation analysis between resource productivity in each country of the Baltic States and indicators mentioned above was made in order to define the indicators that are mostly related with the resource productivity. The correlation coefficients ( r ) and the probabilities of t -statistics ( p ) are presented in Table 1 . The linear correlation is significant if probability is less than 0.05.

The results show the great differences among countries. Most indicators are negatively correlated with the resource productivity in Estonia and net primary income transfers with the rest of the world are the only that significantly positively correlates with the resource productivity. Meanwhile most indicators are positively correlated with the resource productivity in Latvia. It has the strongest correlation with government budget appropriations or outlays for R&D as a percentage of GDP, national income at market prices (gross and net) and national disposable income (gross and net). Contrary to these, there are lots of positively, as well as negatively correlated indicators with the resource productivity in Lithuania, but there are only three indicators that correlates significantly, i.e. gross domestic expenditure on R&D (GERD) in PPPs, total intramural R&D expenditure of all sectors as a percentage of GDP and GERD of business enterprise sector, euro per inhabitant.

Correlation analysis still does not show the causal relationship between these indicators. Furthermore, some indicators can have the delayed effect on the resource productivity. That‘s why Granger causality test will also be made.

Granger causality test

The Granger causality test is a statistical hypothesis test for determining whether one time series is useful in forecasting another. In other words the Granger approach to the question of whether x causes y is to see how much of the current y can be explained by past values of y and then to see whether adding lagged values of x can improve the explanation. y is said to be Granger-caused by x if x helps in the prediction of y , or equivalently if the coefficients on the lagged x ’s are statistically significant. Two lags were taken into account in order to test the delayed effect of various indicators on the resource productivity.

The results showed that the resource productivity in Estonia is Granger caused by total R&D appropriations as a percentage of GDP, publically financed GERD as a percentage of GDP and intramural R&D expenditure of government (in euro per inhabitant and as a percentage of GDP). As the correlation between these indicators and the resource productivity is negative, it is obvious that R&D expenditure does not help to improve the resource productivity and this can be as a cause of inefficient use of these outlays (Table 2 ).

The resource productivity in Latvia is Granger caused by government budget appropriations or outlays for R&D for energy (in euro per inhabitant), total R&D appropriations (in euro per inhabitant), total intramural R&D expenditure of all sectors (in euro per inhabitant), GERD of government as a percentage of GDP, GERD of business enterprise sector, government and higher education sector (all in euro per inhabitant) and all the indicators related to national income, i.e. net primary income transfers with the rest of the world, primary incomes payable to the rest of the world, primary incomes receivable from the rest of the world, gross national income at market prices, net national income at market prices, net national disposable income, gross national disposable income, final consumption expenditure, domestic demand and gross value added. It tends to a conclusion that growth of income has positive effect on the resource productivity, as well as the expenditures on R&D in all the sectors are used efficiently in this country.

Meanwhile the resource productivity in Lithuania is Granger caused only by R&D of small and medium-sized firms as a percentage of total BERD and R&D of large firms as a percentage of total BERD. By analogy to the case of Estonia, it shows that R&D expenditure does not help to improve the resource productivity in Lithuania. Only large firms can be distinguished as leaders as their expenditures on R&D tend to increase the resource productivity.

Regression analysis

Hereinafter autoregressive distributed lag models (ADL) will be created in order to present the more detailed relationship between the resource productivity and the indicators related to national income as well as expenditures on R&D. In order to insure the logical interpretation of the results and avoid the situations when the results tends to the recommends to worsen the certain indicators that would have negative impact on the economics of the country, the requirement that the parameters of the model (except intercept) must be positive will be set. All the indicators that can be significant for the resource productivity according to the results of correlation analysis and Granger causality test will be taken into account when making the ADL model. Also two lags of dependent and independent variables will be analysed.

Unfortunately no one significant model for the resource productivity in Estonia was found. It confirms again that Estonia has great problems with the resource productivity and R&D expenditures are not directed to solve it.

In the case of Latvia several similar models can be created, but the best significant model (at the significance level of 0.05) is as follows:

(1) here yt is the resource productivity in Latvia at the time moment t and xt-1 is domestic demand at the time moment t-1. So, growing demand of products (as well as national income since they are also strongly correlated with the resource productivity) encourages the nation to take care about the resource productivity. The coefficient of determination of the model is 0.86 and it means that variation of xt-1 can explain 86% of the variation of yt. Lags of yt (i.e. yt-1 and yt-2) as well as other indicators don’t improve the model.

The resource productivity in Latvia is also strongly correlated with the R&D expenditure. The best significant model is written below:

(2) here y _t is the resource productivity in Latvia at the time moment t , x _{1 t-2} is total R&D appropriations in euro per inhabitant at the time moment t-2 and x _{2 t-1} is GERD of government in euro per inhabitant at the time moment t-1 . So, the resource productivity in the country grows respectively to R&D expenditures and this fact indicates effectiveness of such outlays. The adjusted coefficient of determination of the model is 0.81. Lags of y _t (i.e. y _t-1 and y _t-2 ) as well as other indicators don’t improve the model.

The best significant model for the resource productivity in Lithuania is presented below:

(3) here y _t is the resource productivity in Lithuania at the time moment t and x _t-2 is R&D of large firms as a percentage of total BERD at the time moment t-2 . It means that the benefit of R&D expenditures on the resource productivity best manifest after two years. The coefficient of determination of the model is 0.48 and it means that variation of x _t-2 can explain almost half of the variation of y _t . Lags of y _t (i.e. y _t-1 and y _t-2 ) don’t improve the model. So the results indicate that only large companies significantly contribute to improvement of resource productivity in Lithuania.