The ICT sector’s significance in the Baltic states - An input-output analysis
S. A. Leonsaari
S3137783
s.a.leonsaari@rug.nl
University of Groningen, Faculty of Economics and Business
ABSTRACT
This thesis examines the economic significance of the information and communication technology sectors in the Baltic states, Estonia, Latvia and Lithuania, in the last decades in an input-output framework. The countries have increased their gross domestic product considerably since gaining independence in 1991 and all have fast-growing ICT sectors. However, we calculated net value added multipliers for the sector which showed that only the Estonian ICT sector can be considered a key sector for the economy. Similarly, calculating gross value added multipliers yielded much larger results for Estonia than for Latvia and Lithuania.
1
1. INTRODUCTION
The past few decades have seen the so-called information and communication technology (ICT) revolution completely change the way people, businesses and governments work and has had a great impact on economic development (Jorgenson and Vu, 2016). The ICT sector, which covers “computer hardware and software, telecommunications, consumer electronics, and Internet-based contents, applications and services”, is widely accepted as a key component for countries’ economies (Fransman, 2010: xi). It is an important sector in itself because it makes a notable contribution to GDP, trade and employment but also because it supplies the vital information and communications infrastructure that economies and societies rely on in order to operate (ibid.). The ICT producing sector is a dynamic, fast growing sector which in the EU more than tripled its real term value added from 1995 to 2015, whereas the total economy’s value added was only multiplied by a factor of 1.4 (Mas et al., 2018). The Digital Agenda for Europe in 2010, with the purpose of getting the most out of the economic and social capability of ICT, was named as one of the pillars of the Europe 2020 Strategy for growth in the European Union, which in itself is a testament to the ICT sector’s importance (ibid.). ICT’s effect on economies has been widely studied with significant evidence for its ability to generate growth (Jalava and Pohjola, 2007; Seo, 2009; Venturini, 2009; Vu, 2011). The effects on economic growth can be traced to it being used as an input through ICT capital services, the output from ICT production as well as the productivity advances it brings about (Jalava and Pohjola, 2007). Because of the pivotal role ICT plays in the modern economy, fostering it is an important issue for countries’ development.
2
were once kept isolated from the western Europe joined the EU in 2004, have since all adopted the euro, have greatly grown their per capita gross domestic product and still have economic structures that resemble one another (Brizga et al., 2014). However, Estonia can be singled out for its exceptional progress in the field of ICT which is largely credited to the policy steps taken by the Estonian government post-independence to transform the country into an information society (Kasekamp, 2010).
Therefore, the purpose of this thesis is to explore whether the ICT sector has had a significant role in the Baltic states’ economies between 2000 and 2014 with the remarkable development the countries have experienced in the past few decades, whether that role has changed over time and whether the countries’ are similar in that manner. The literature on ICT says ICT has been important for countries’ growth but it has not been specifically quantified for the Baltic states, which are small countries with relatively small ICT sectors, which is what this thesis does, thus filling a gap.
3
recognize the importance of specific sectors for the economy. Therefore, input-output analyses that allow one to take into account linkages between sectors are a useful framework when studying specific sectors’ significance, as is the case in this thesis, and is therefore the chosen method. Input-output data also allow us to make comparisons at an industry level between countries. (Miller and Blair, 2009)
Similar studies using input-output tables to study IT industries in various ways have previously been done for Korea (Cho et al., 2013; Hong et al., 2016), China (Xing et al., 2011; Li et al., 2018), Denmark (Strohmaier and Rainer, 2016), India (Roy et al., 2002), Singapore (Toh and Tangavelu, 2013), and jointly for different Asian countries (Irawan, 2014) as well as European countries (Rohman, 2013) as a comparison. However, no previous studies have researched the ICT sector in the Baltic states which is how this thesis adds to the existing literature.
4
than the other sectors are on it. If so, that would mean that the sector’s growth depends on the impulses received from other sectors instead of being able to produce growth impulses itself. Since we are interested in the importance of the ICT sector for the growth of the Baltic states’ economies, this is precisely what we want to know. For a comparison, the standard gross value added multipliers are also calculated for the ICT sector. Rohman studied the outputs of ICT sectors, but since output as a measure covers final as well as intermediate products and therefore double counts, it is not suitable for a sector’s economic performance evaluation (Oosterhaven and Stelder, 2002). Instead, we used value added as our variable of choice. The data comes from the World Input-Output Database (Timmer et al., 2015).
5
2. BACKGROUND AND LITERATURE REVIEW
This chapter first introduces the topic of information and communications technology by briefly outlining how the ICT revolution came to be, and then, in order to illustrate the importance of the ICT sector, summarizes the current literature on ICT’s economic impacts and describes the development of the Baltic states within the last decades.
2.1 The ICT revolution
Just as the steam engine and electricity, ICT is a more recent example of a drastic innovation that changed the way of production as well as the composition of the economic system (Strohmaier and Rainer, 2016). Information and communications technology has radically transformed not only everyday life by altering people’s manner of communicating, working, and interrelating with each other but also the way businesses and governments work. These changes are tangible in all countries and this so called ICT revolution can be accredited to the swift improvements made in semiconductors. At the very source of the modern ICT is the invention of the transistor in 1947 which brought its inventors a Nobel Prize in Physics. A transistor is a semiconductor that operates as an electric switch converting information into code. The second ICT enabling milestone was the creation of the integrated circuit, which is another Noble Prize winning device formed of several data storing and handling transistors. These inventions were crucial, since the amount of transistors on a computer’s central processing unit is what estimates its processing ability. Therefore, the rapid development of semiconductor technology is what ultimately laid the groundwork for the growth of ICT. The evolution of Internet and mobile technology during the 1990s, in connection with the increased rate of globalization, is what consequently powered the striking expanse of ICT applications over sectors and countries. (Jorgenson and Vu, 2016)
6
have vastly raised their mobile phone and internet penetration rates in the last decades, the actual economic gains from ICT have been questioned in the past. In the late 80s, economist Robert Solow famously wrote that “you can see the computer age everywhere but in the productivity statistics” (Solow, 1987: 36), an observation that came to be known as Solow’s Paradox and inspired many a research study on the proposed lack of connection between the increasing amount of computers and productivity growth. The Solow Paradox has since been partly explained by lack of suitable data as well as improper use of statistical methods (Hitt and Brynjolfsson, 2002) but also by the long time it takes for organizations to rearrange their business processes along with the change from the existing technological regime to its follower in order to reap the productivity benefits (David, 1990). As can be appreciated, ICT’s impact on the economy has been widely, and not always unanimously, studied. Much of the literature on ICT’s ability to spur growth is focused on the level and quality of the usage of ICT (Evangelista et al., 2013). The following sections summarize the main issues in the literature.
2.2 The economic impact of ICT
7
This multi-factor (also known as total factor) productivity refers to “the residual or leftover part of productivity that can’t be directly inferred from capital and labor” which can be brought about by, for example, improved processes and new business practices (Brynjolfsson and Saunders, 2009: 46). ICT capital is also deemed a notable driver of long-run growth for modern societies, impelling GDP growth over time (Venturini, 2009). The same conclusion was reached by Seo et al. (2009), who observed a positive correlation with economic growth and investment in ICT in 29 different countries. Vu (2011) also found significant empirical evidence for ICT’s part in creating growth. An example of the remarkable economic impact of ICT is a study conducted by Jalava and Pohjola (2008) demonstrating that in the case of Finland, ICT’s contribution to the economic growth of the country from 1990 to 2004 was actually as much as three times bigger than that of electricity’s from 1920 to 1938. Because of this transformative nature and the wide-ranging influence of ICT, studying its economic impact has long been of special interest and a large part of ICT studies from the past two decades have concentrated on examining ICT’s effect on productivity.
ICT and productivity
8
This is a key finding in the literature of ICT’s effect on productivity, as the research is less unanimous on the reasons as to why countries within Europe generally had derived less advantage from the effects of ICT with regard to productivity growth in comparison with the United States (Draca et al., 2009). The impact of ICT on the growth of the US economy was a result of an increase in ICT investment that had a capital deepening impact, steadfast productivity influences from the industries that produce ICT equipment, as well as a slightly deferred multi-factor productivity boost in ICT-employing industries, especially in market services industries (Triplett and Bosworth, 2006; van Ark et al., 2008). Contrarily, van Ark et al. (2008: 25) accredit Europe’s productivity slump to a lagging multi-factor productivity growth, which the authors regard as a proxy for developments in technology and innovation, in the market services sector and “the slower emergence of the knowledge economy […] compared to the United States”. Thus, multi-factor productivity measures the indirect contributions to growth of the employment of ICT by amplifying the possibility of further innovations (ibid.).
Innovation is one of the paramount sources of productivity (Cardona et al., 2013) and the next section will explore the ways in which ICT links to innovation.
ICT and innovation
9
firms, enabling closer links between businesses and customers, reducing geographic limitations and increasing efficiency in communication”. This ability to enable innovation was also confirmed by Hitt and Brynjolfsson (2002). In addition, as mentioned previously, ICT as a general purpose technology also has a positive impact on growth through complementary innovations which means that “the productivity of R&D in a downstream sector increases as a consequence of innovation in the GPT technology”, thus amplifying the effects of innovation (Bresnahan and Trajtenberg, 1995: 84).
The ICT producing sector is distinctive for its high level of research and development and for productivity levels that exceed those of the rest of the economy (Mas et al., 2018). According to the OECD (2015), the persisting high expenditures of business enterprises on research and development within the ICT sector as well as the current growth in ICT associated patents indicate the essential character the ICT sector plays in innovation. The same conclusion was reached by Corrocher et al. (2007) who also mentioned the growth of patents as an indication of the ICT sector’s increasing importance for innovation.
To summarize the previous sections, there is significant evidence on the major impact ICT has on the growth of the economy through productivity and innovation advances, making ICT a key sector for growth. This, however, has not been explicitly quantified which is what this thesis does for the Baltic states. The following section outlines the development of the Baltic states’ economies by first giving a brief historical overview of the countries and then describing the state of the ICT sector with statistics, to give a frame of reference for the ICT sector analysis.
2.3 The economic development of the Baltic states
10
Union and turned into soviet socialist republics. The three countries were only able to regain their independence after the collapse of the USSR in 1991, after having been divided from the capitalist western Europe by the iron curtain for a half-century. Despite the independence, the Baltic states’ economies continued to be firmly linked to the USSR’s economy which was plunging into hyperinflation. The post-communist era started with a large economic downturn with the transition to a market economy, when regulators for prices and wages were eliminated, and expenditures and personal earnings started to be, for the most part, determined by individual enterprise and by supply and demand. It is estimated that up to 60% of the people in all three Baltic states were living below official poverty levels by mid-1990s, and all countries’ per capita GDP indices had experienced quick downturns. Per capita GDP did start to turn its descent around starting circa 1995, yet it did not result in an instantaneous improvement in most of the population’s living standards. (Plakans, 2011)
In 1995, the Baltic states applied for European Union membership. Property and infrastructures were being privatized as a basis for establishing a market economy, and all three countries were on the road to recovery. The second half of the 90s also witnessed the countries emerging as some of the most alluring foreign direct investment destinations in Europe. Estonia was the trailblazer and unilaterally introduced free trade as well as drastic taxation reforms - it was the first nation in Europe to institute a flat rate of income tax in 1994 and to eliminate taxes on reinvested corporate profits in 1999. Latvia and Lithuania were quick to conform by also instituting a flat tax system. (Kasekamp, 2010)
11
Figure 1. GDP per capita, PPP in constant international $. Source: The World Bank (2017)
The term could be argued as being justified, since the Baltic states region was growing faster than any other region within the EU in the first decade of the new millennium, and the GDP growth rates of 10 to 12 per cent attained in 2005-2006 settled the countries among the best economic performers in the world. The growth was largely driven by domestic consumption, specifically a construction boom pushed forward by inexpensive credit offered by the Nordic-owned banks that dominated the banking sector. The economic uptrend came to a halt in 2008, and the global financial crisis sent the Baltic economies into double-digit negative growth in 2009, with Latvia taking the hardest hit out of the three countries. (Kasekamp, 2010)
However, all three countries started to quickly recover and by 2011 they were displaying the highest GDP growth rates in the European Union, measured by Eurostat statistics, ranging from 5.5% in Latvia and 5.9% in Lithuania to a striking 8.3% in Estonia (Brizga et al., 2014).
12
2.4 ICT in the Baltic states
All three countries are small, with the population in Estonia being 1.32 million, Latvia 1.96 million and Lithuania 2.87 million (The World Bank, 2018). They are also geographically close and have a shared political past, but a field in which Estonia differs and has been, again, a trailblazer out of the three countries is the diffusion of information and communications technologies, with the Estonian government having taken major steps to promote ICT since the 90s. The eager interest in ICT was further boosted by the Estonian public sector which advocated the so called ‘paperless government’. Estonia was also the first nation in the world to vote online in a national election. (Kasekamp, 2010)
In 2015, the value added of the ICT sector was 1,317.5 million euros in Estonia, 1,492 million euros in Latvia and 1,802.2 million euros in Lithuania, measured in purchasing power standard. This meant a growth of 3.1% in Estonia, 12.6% in Latvia and 6.8% in Lithuania from the previous year. However, even though the Estonian sector was the smallest as measured in value added, its business expenditure on research and development at 79.2 million euros was more than triple that of Lithuania’s (24.0 million) and as much as seven times larger than that of Latvia’s ICT sector (11.0 million) for the same year. (Mas et al., 2018)
As of 2016, all three countries belong to the top quartile of the 44 highest ranking countries (out of a total of 175 economies) based on the level of ICT development as measured by The International Telecommunication Union, a United Nations agency specialized in ICT. The ICT Development Index, IDI, is a compound index comprised of 11 different indicators which are fused into one benchmark measure to observe and contrast advancements in ICT across countries and over time. (ITU, 2016)
2.5 Research questions
13
Research question 1: What was the significance of the ICT sector to the economies of Estonia, Latvia and Lithuania respectively in 2000-2014?
This question will be answered by calculating the net value added multipliers for each year from 2000 to 2014 to investigate how they have changed over time and how the countries compare amongst themselves. The results will also be compared to the ICT sector’s standard gross value added multipliers which yield the direct and indirect value added in the economy due to the final demand in the ICT sector.
Research question 2: What have been the key sectors for the economies from 2000 to 2014 and do these differ across the countries?
14
3. METHODOLOGY AND DATA
This chapter summarizes the concept of input-output analysis, explains the methodology used to answer the research questions as well as introduces the data source.
3.1. Introduction to input-output tables and analysis
The input-output analysis is an analytical framework created by Wassily Leontief in the late 1930s for which he was awarded the Nobel Prize in 1973. It is an interindustry analysis based on linkages between industries and provides a framework in which to analyze how industries rely on other industries for their intermediate products in order to produce commodities and services. The basic input-output model is usually composed from observed economic data for a specific region, such as a country or a state. (Miller and Blair, 2009)
The aforementioned linkages between industries increasingly traverse borders resulting in international division of production, and an input-output analysis allows one to study the full picture of these international interdependencies in the production structure (Timmer et al., 2014). Input-output analyses have been carried out for numerous topics ranging from emissions embodied in trade (Dietzenbacher et al., 2012; Zhao et al., 2016) to water use (Cazcarro et al., 2013; Liu et al., 2017) and labour productivity (Ochoa, 1986; Yang and Lahr, 2008), to mention a few.
An input-output table represents a full view of the flows of goods and services in an economy for a given year. The structure of a national input-output table is illustrated in Table 1. The production in an economy is done by companies that can be organized into n sectors or industries. In the rows are the industries that deliver goods and services, in the columns the ones that use them. There are k final demand categories, e.g. private consumption, government consumption and exports. Furthermore, there are m primary inputs, e.g. wages and salaries and imports. The n × n matrix Z presents the intermediate deliveries. A typical element in the Z matrix, zij, denotes the monetary
15
intermediate inputs in industry j’s production. Thus, industry j makes a payment to industry i. The subscripts i and j denote the row and column coordinates respectively. The n × k matrix F presents the deliveries to the different final demand categories, which are household consumption, government consumption, gross fixed capital formation (i.e. investments), changes in inventories and exports. Element fij presents the goods and services delivered by industry i to final demand
category j, such as government consumption. The m × n matrix W presents factor inputs, i.e. labor, capital, imports and indirect taxes less subsidies. Element wij, therefore, denotes the use of primary
input i for production in industry j. The n × 1 column vector x presents the gross output, i.e. production value, in each industry. Since input-output tables are acquired from double entry bookkeeping, each industry’s total sales (presented in the row of industry i, expressing its output distribution) equal the total of all inputs (presented by column i, illustrating the production process). Consequently, each industry’s row sum matches its column sum. (Dietzenbacher, 2012)
Sectors Final demands Total
Sectors Z F x
Primary inputs W (4) (t1)
Total x’ (t2)
Table 1. The basic structure of an input-output table. (Source: Dietzenbacher, 2012)
16
bold denote matrices and lower case letters in italics denote scalars, such as elements of vectors and matrices. (Dietzenbacher, 2012)
All of this information in an input-output table can also be expressed in the following form, with all equations below cited from Miller and Blair (2009).
The column vector for total output x, the intermediate deliveries matrix Z and the total final demand column vector f (obtained from the row sums of the n × k matrix F, yielding the aggregate final demand of the individual final demand categories in matrix F) take the following forms:
x = [ 𝑥1 ⋮ 𝑥𝑛], Z = [ 𝑧11 ⋯ 𝑧1𝑛 ⋮ ⋱ ⋮ 𝑧𝑛1 ⋯ 𝑧𝑛𝑛] and f = [ 𝑓1 ⋮ 𝑓𝑛 ] (3.1)
The sales of the output in each one of the n industries can also be stated in equations using the elements as described above, as such:
x1 = z11 + … + z1j + … + z1n + f1
⋮
xi = zi1 + … + zij + … + zin + fi (3.2)
⋮
xn = zn1 + … + znj + … + znn + fn
Alternatively, the information on each industry’s sales can be summed up in matrix notation:
x = Zi + f (3.3)
17
This forms the basis of input-output analysis and gives structure to the table. What can be further derived from the table is a ratio called an input coefficient, typically denoted aij. It can be obtained
by simply dividing the intermediate delivery zij, i.e. the input, of industry i by the total output xj of
industry j, and thus illustrates the intermediate input from industry i needed for one unit of output in industry j. Calculating this for all n industries of an economy yields the n × n input matrix A that can be expressed as
A = Z𝐱̂−1 (3.4)
A circumflex over a vector indicates a diagonal matrix in which the elements of the vector are placed along the main diagonal with all other elements being zero. Therefore, 𝐱̂−1is the n × n
inverse diagonal matrix of output vector x. (Miller and Blair, 2009)
It follows from 𝐙 = 𝐀𝐱̂ and 𝐙𝐢 = 𝐀𝐱, that the total output of an economy in (3.3) can be represented in matrix notation:
x = Ax + f (3.5)
Using matrix algebra, the solution to that is
x = (I − 𝐀)−1f = Lf (3.6)
where I denotes an identity matrix consisting of ones on the main diagonal, zeros everywhere else, and the matrix (I − 𝐀)−1 = L is the Leontief inverse or the total requirements matrix. A typical
element in the L matrix, lij, denotes the extra output in industry i that is required to meet an
additional unit of final demand of industry j. These elements are customarily called multipliers. (Miller and Blair, 2009)
18
19
3.2. Multiplier analysis
Oosterhaven and Stelder (2002) found that net multipliers prevent the effect of double counting and can fix other conceptual difficulties associated with the customary application of gross multipliers when examining the economic importance of a sector for a region or a country. The practice is that people multiply the standard multiplier (like the value added multiplier) for an industry with an industry variable that indicates size (like value added). However, it is incorrect to multiply (at industry level) the value added multiplier with value added or the employment multiplier with employment. In the case of multiplying the output multiplier with industry output, it can even be shown that this leads to double counting. By construction, it only makes sense to multiply multipliers with industry final demands. Oosterhaven and Stelder’s response to this practice is: if we cannot change this (mal)practice, we perhaps can change the multiplier so that the result is fine and practitioners do not have to change their practice.
20
Examining the typical Leontief model, final demand for outputs f is exogenous and the model’s causality works in the following way. All variations in both final and total demand for sectoral outputs are matched, with no supply restrictions, by endogenous sectoral production x, i.e. the gross output in each sector. As a result of that, sectoral production dictates the endogenous intermediate demand for sectoral outputs Ax along with the endogenous demand for primary inputs, e.g. value added. To solve the total value added v, one would use the equation
v = vc’x = vc’ (I − 𝐀)−1f = vc’Lf (3.7)
where vc’ is a row with value added coefficients, which are obtained dividing the value added of an
industry by the total output and thus give the value added per unit of output, A is the technology matrix with intermediate input coefficients, and L the Leontief inverse. Evidently, the value added multipliers vc’L should exclusively be multiplied with f, which is the exogenously determined final
demand, instead of the endogenously determined gross output x. (Oosterhaven and Stelder, 2002)
Following some discourse on the topic (see for example De Mesnard, 2002), an economic interpretation for the net multiplier was given by Dietzenbacher (2005) and all following equations are cited from his study.
Following on the principles of input-output analysis elaborated in section 3.1., output multipliers are formulated as x’𝐱̂−1L(=i’L). Different multipliers, such as employment or emissions, can be determined likewise. Thus, the general description of multipliers is as follows
m’𝐱̂−1L (3.8)
21
order to indicate the size of the sector, these multipliers have to be adjusted for that method to be reasonable. As a result, Oosterhaven and Stelder formulated net multipliers in the following manner
m’𝐱̂−1L𝐟̂𝐦̂−1 (3.9)
Multiplying these net multipliers by 𝐦̂ produces an identical solution to the gross multipliers in (3.8) being multiplied by 𝐟̂. In this way, they adequately serve their purpose. Their economic interpretation can be explained as follows. Let us look at the matrix 𝐦̂ 𝐱̂−1L𝐟̂ and examine value
added, so that m = v. Element i, j of the matrix 𝐯̂𝐱̂−1L𝐟̂ yields the value added that is, both indirectly and directly, created in sector i as a result of the final demand fj in sector j. The matrix’s
ith row aggregate gives element i of the column vector 𝐯̂𝐱̂−1L𝐟̂i = 𝐯̂𝐱̂−1Lf = 𝐯̂𝐱̂−1x = 𝐯̂i = v. In
other words, vi signifies the value added in sector i required to meet final demands in their entirety.
The total value added required indirectly and directly to meet final demand fj in sector j is given by
element j, which is the jth column aggregate of 𝐯̂𝐱̂−1L𝐟̂, of the row vector i’𝐯̂𝐱̂−1L𝐟̂ = v’𝐱̂−1L𝐟̂.
Consequently, the jth net multiplier gives
(v’𝐱̂−1L𝐟̂𝐯̂−1)j =
𝑗th column sum of 𝐯̂𝐱̂−1𝐋𝐟̂
𝑗th row sum of 𝐯̂𝐱̂−1𝐋𝐟̂ (3.10)
which is the ratio of value added created in all sectors by final demand fj for commodity j and the
value added created in sector j due to total final demands of all sectors. When studying the economic importance of certain sectors, a net multiplier for sector j that exceeds one signifies that the final demand in j creates more value added in other sectors than the other sectors’ final demands put together create value added in sector j. Should that be the case, sector j can be considered to be of more importance to the other sectors than those are to sector j. (Dietzenbacher, 2005)
22
Since total value added v = v’i = v’𝐱̂−1x = v’𝐱̂−1Lf, the jth element of v’𝐱̂−1L is interpreted as the
additional value added in all sectors collectively, resulting from one additional final demand unit in sector j. Then, these gross multipliers are multiplied with final demands to reflect the size of the sectors. Element j of the resulting row vector yields the aggregate value added attributable to the final demand in sector j. (Dietzenbacher, 2005)
3.4 Data
The 2016 release of the World Input-Output Database (WIOD) by Timmer et al. (2015) is used to conduct this research. The database consists of annual input-output tables for 43 countries from 2000 to 2014, covers 56 sectors that follow the International Standard Industrial Classification Rev. 4 and is available for free at www.wiod.org. Three of the available 56 sectors correspond to the definition of the ICT sector and all three Baltic states are covered from 2000 to 2014 which makes the data set suitable for the research.
Sector classification
The ICT sector definition used in this study is the 2007 OECD classification that states that “the production (goods and services) of a candidate industry must primarily be intended to fulfil or enable the function of information processing and communication by electronic means, including transmission and display”, and which is based on ISIC Rev. 4 (OECD, 2011: 59). The activities that fall within this definition are listed in Table 2, together with their 4-digit ISIC codes.
ICT Manufacturing industries
2610 Manufacture of electronic components and boards 2620 Manufacture of computers and peripheral equipment 2630 Manufacture of communication equipment
23
2680 Manufacture of magnetic and optical media
ICT trade industries
4651 Wholesale of computers, computer peripheral equipment and software 4652 Wholesale of electronic and telecommunications equipment and parts
ICT services industries 5820 Software publishing
6110 Wired telecommunications activities 6120 Wireless telecommunications activities 6130 Satellite telecommunications activities 6190 Other telecommunications activities 6201 Computer programming activities
6202 Computer consultancy and computer facilities management 6209 Other information technology and computer service activities 6311 Data processing, hosting and related activities
6312 Web portals
9511 Repair of computers and peripheral equipment 9512 Repair of communication equipment
Table 2. Activities within the ICT sector (OECD, 2011: 159).
24
WIOD sector ‘Publishing activities’ was also omitted. The repair activities could not be matched to the available WIOD data.
ICT sector in WIOD
C26 Manufacture of computer, electronic and optical products J61 Telecommunications
J62-63 Computer programming, consultancy and related activities; information service activities
Table 3. ICT industry in WIOD corresponding to the OECD sector definition.
25
4. RESULTS
Based on national input-output tables for Estonia, Latvia and Lithuania from the World Input-Output Database, the net value added multipliers of the ICT sector were calculated for each year from 2000 to 2014, the results of which are in Table 4. Following the explanation in the previous chapter, a net multiplier larger than one means the economy-wide value added created by final demand in the ICT sector exceeds the amount of the ICT sector’s value added that is created by all other sectors’ final demand. Therefore, the ICT sector is more important for the other sectors than those are for the ICT sector, and by this degree it is established a key sector. When the net multiplier is less than one, the sector is said to be more dependent on the rest of the economy than what the rest of the economy is on the sector (Oosterhaven and Stelder, 2002).
Estonia Latvia Lithuania
26
2013 1.09 0.85 0.90
2014 1.10 0.86 0.90
Table 4. Net value added multipliers for the ICT sector (own calculations).
The Estonian ICT sector’s net value added multipliers went from 0.84 in 2000 to 1.10 in 2014, with a rather consistent upward projection. Latvia’s multipliers showed very little change from 0.80 in 2000 to just 0.86 in 2014, with a downward projection from 2007 to 2010. This could be related to the financial crisis which was especially hard on the Latvian economy, as mentioned in chapter 2. The Lithuanian ICT sector’s multipliers actually decreased from 2005 to 2010, but still grew overall from 0.76 in 2000 to 0.90 in 2014.
Estonia is the only one out of the three nations that has ICT net multipliers larger than 1 from 2010 onwards, deeming it a key sector for the economy. This means that the ICT sector in Estonia in 2010-2014 put forth a bigger value added impact on the economy than the whole economy did on it. The Latvian and Lithuanian ICT sectors, by this definition, can therefore not be categorized as key sectors within their respective economies for the timeframe studied. Estonia’s net value added multipliers also grew the most from 2000 to 2014, whereas both Latvia and Lithuania show much slower change compared.
What should be stressed about the net value added multipliers is that a value less than unity does not mean that the sector is unimportant - it simply means that the sector depends more on the rest of the economy than the rest of the economy depends on it (Oosterhaven and Stelder, 2002).
27
Figure 3. ICT sector value added in basic prices, millions of US dollars. (Source: WIOD)
Next, the standard gross value added multipliers were calculated. They were multiplied with final demands to reflect sector size and then scaled to 1 in order to identify key sectors in the economies. This means that the sector with the highest total value added generated directly and indirectly in the economy (that can be attributed to the sector’s final demand) got denoted as 1, establishing it the key sector by this measure, and the rest of the sectors’ gross value added multiplier effects are shown relative to that key sector in order to yield a relative comparison to the key sector. The results of this for the ICT sectors are found in Table 5.
Estonia Latvia Lithuania
28 2005 0.34 0.30 0.41 2006 0.30 0.24 0.32 2007 0.28 0.20 0.24 2008 0.36 0.21 0.23 2009 0.50 0.33 0.35 2010 0.63 0.38 0.36 2011 0.68 0.35 0.34 2012 0.62 0.35 0.33 2013 0.65 0.36 0.35 2014 0.69 0.37 0.35
Table 5. Gross value added multipliers of the ICT sector multiplied by final demand, scaled to 1 (own calculations).
The ICT sector is not the key sector in any of the countries according to this measure, which is understandable considering its relatively small size. What can be seen from the table is that these results are consistent with the net value added multipliers in that Estonia shows much more growth in the figures than Latvia and Lithuania. From 0.43 in 2000 to 0.69 in 2014, Estonia’s ICT sector has significantly grown in relative significance, with notably large advances from 2008 to 2010.
The key sector in Estonia was throughout the years the construction sector, with the exception of 2010 being the ‘public administration and defence; compulsory social security’ sector. This means that all values in Table 5, with that one year’s exception, are in relation to the total value added attributable to the construction sector’s final demand. That is, the total value added in all sectors of the economy due to the final demand in the ICT sector was 0.69 times that of the value added due to the construction sector’s final demand in 2014.
29
activities sector in 2011-2014. Therefore, the values in Table 5 for Latvia are relative to the total value added attributable to these sectors’ respective final demands. Over the years, the total value added in all sectors of the economy due to the final demand in the ICT sector has not surpassed 0.45. This means that at best, the gross value added multiplier effect of the ICT sector was still less than half of that of the key sector of the economy. Consistent with the net value added multiplier results, this can be interpreted as the Latvian ICT sector not having a major gross value added multiplier impact on the economy.
In Lithuania, the key sector in 2000-2001 was the manufacture of food products, beverages and tobacco products sector, in 2002-2004 and again in 2009-2010 the ‘public administration and defence; compulsory social security’ sector, and in 2005, 2011-2012 and 2014 the ‘retail trade, except of motor vehicles and motorcycles’ sector. In 2006-2008, the key sector was construction and in 2013 the ‘wholesale trade, except of motor vehicles and motorcycles’ sector. Year 2001 stands out for the uncharacteristically high result of 0.66 with no obvious explanation for it, especially since the rest of the results are not nearly as jumpy. Apart from that single year, the results have not surpassed 0.41 over the years. As with Latvia, this means that even at its largest, the gross value added multiplier effect of the ICT sector was still less than half of that of the key sector of the economy.
30
time, while both Latvia and Lithuania have been significantly slower movers in this sense. But as mentioned in chapter 2, the value added of the ICT sectors in all countries is growing fast, with Latvia and Lithuania actually experiencing faster growth than Estonia. Therefore, they do have the potential to catch up with the Estonian ICT sector.
The fact that the Estonian government has been actively fomenting the sector through, amongst other initiatives, nationwide computer projects since the 1990s (Kasekamp, 2010) is likely to be a contributing factor on the sector’s significance now which shows up on these results. Estonia’s ICT sector’s share of the GDP is also the highest out of the three countries. If Latvia and Lithuania had adopted similar ambitious policies of becoming an information society after the collapse of the USSR, the consequences could possibly be similarly visible in the results now.
Limitations
The input-output framework relies on a few assumptions that must be mentioned. A foundational assumption is that the flows from sector j to sector i within a certain timeframe, such as a year, are fully determined by the total output of sector i throughout that year. In other words, input coefficients aij do not change and can be understood as describing a fixed relationship between an
industry’s output and its inputs. Therefore, economies of scale through an increased production volume are not possible to achieve and the Leontief system production functions under constant returns to scale. What is also assumed in the input-output analysis is that inputs are used by sectors in fixed proportions, making input substitution impossible. These assumptions make the application of the framework easier, yet can also be considered a limitation as they restrict the capability of the framework to involve some essential elements of an economic system, like elasticities. (Miller and Blair, 2009)
31
the widely accepted OECD definition of the sector. More disaggregated data would have allowed us to also include, for example, the software publishing sector that was left out so as to not exaggerate the size of the sector.
5. Conclusions
This thesis has explored the economic impact of ICT and especially the significance of the ICT sector to the development of the Baltic states from 2000 to 2014. The study concludes that according to our chosen measurement, the net value added multiplier as introduced by Oosterhaven and Stelder (2002), only the Estonian ICT sector can be considered a key sector for the country’s economy, and only from the year 2010 onwards. These results were compared to the gross value added multipliers which were multiplied with final demands and scaled to one to identify key sectors. The two key sectors in common for all three countries identified this way were construction and ‘public administration and defence; compulsory social security’. The gross value added multiplier effects for Estonia were much larger than for Latvia and Lithuania.
The study added to the current vast literature on different country analyses for ICT in that it explored the quickly growing ICT sector in these three countries that have gone from impoverished former Soviet states to technologically advanced, modern economies, whose ICT sectors have not been studied in this capacity before.
The Baltic states are an interesting comparison base, since their “starting points” as market economies after the collapse of the USSR in early 1990s were very similar, and their economies have grown in a similar manner, yet they differ in the progress of ICT.
32
Bibliography
Baily, M. (2002) ‘Distinguished Lecture on Economics in Government: The New Economy: Post Mortem or Second Wind?’, Journal of Economic Perspectives, vol. 16 (2), pp. 3-22.
Bresnahan, T. and Trajtenberg, M. (1995) ‘General purpose technologies ‘engines of growth’?’,
Journal of Econometrics, vol. 65 (1), pp. 83–108.
Brizga, J., Feng, K. and Hubacek, K. (2014) ‘Drivers of greenhouse gas emissions in the Baltic States: A structural decomposition analysis’, Ecological Economics, vol. 98, pp. 22-28.
Brynjolfsson, E. and Saunders, A. (2009) Wired for Innovation: How Information Technology Is
Reshaping the Economy. Cambridge, MA: The MIT Press.
Cardona, M., Kretschmer, T. and Strobel, T. (2013) ‘ICT and productivity: conclusions from the empirical literature’, Information Economics and Policy, vol. 25 (3), pp. 109-125.
Cazcarro, I., Duarte, R. and Sánchez Chóliz, J. (2013) ‘Multiregional Input-Output Model for the Evaluation of Spanish Water Flows’, Environmental Science & Technology, vol. 47 (21), pp. 12275-12283.
Cho, B., Cho, S. and Lee, J. (2013) ‘An alternative approach of input-output tables to dynamic structure changes in Korean IT industries’, Technological and Economic Development of Economy, vol. 19 (2), pp. 257-271.
Colecchia, A. and Schreyer, P. (2002) ‘ICT investment and economic growth in the 1990s: Is the United States a unique case? A comparative study of nine OECD countries’, Review of Income and
Dynamics, vol. 5 (2), pp. 408-442.
33
Cortés, E. and Navarro, J. (2011) ‘Do ICT Influence Economic Growth and Human Development in European Union Countries?’, International Advances in Economic Research, vol. 17 (1), pp. 28-44.
David, P. (1990) ‘The Dynamo and the Historical Perspective on the Modern Paradox’, American
Economic Review, vol. 80 (2), pp. 355-361.
De Mesnard, L. (2002) ‘Note about the Concept of ‘Net Multipliers’’, Journal of Regional Science, vol. 42 (3), pp. 545-548.
Dietzenbacher, E. (2005) ‘More on Multipliers’, Journal of Regional Science, vol. 45 (2), pp.421-426.
Dietzenbacher, E. (2012) ‘A crash course in input-output economics’, Working Paper, University of Groningen.
Dietzenbacher, E., Pei, J. and Yang, C. (2012) ‘Trade, production fragmentation, and China’s carbon dioxide emissions’, Journal of Environmental Economics and Management, vol. 64, pp. 88-101.
Draca, M., Sadun, R. and van Reenen, J. (2009) ‘Productivity and ICTs: A review of the evidence’, in: Avgerou, C., Mansell, R., Quah, D. and Silverstone, R. (ed.) The Oxford Handbook of
Information and Communication Technologies. Oxford: Oxford University Press, pp. 100-147.
Eurostat (2016) Percentage of the ICT sector on GDP. [Online] Available at: http://ec.europa.eu/eurostat/web/products-datasets/-/tin00074 [Accessed June 25, 2018]
Evangelista, R., Guerrieri, P. and Meliciani, V. (2014) ‘The economic impact of digital technologies in Europe’, Economics of Innovation and New Technology, vol. 23 (8), p. 1-23.
34
Hitt, L. and Brynjolfsson, E. (2002) ‘Information Technology, Organizational Transformation, and Business Performance’, in: Mairesse, J., L’Horty, Y. and Greenan, N. (eds.) Productivity,
Inequality, and the Digital Economy: A Transatlantic Perspective. Cambridge, MA: The MIT
Press, pp. 55-91.
Hong, J., Byun, J. and Kim, P. (2016) ‘Structural changes and growth factors of the ICT industry in Korea: 1995–2009’, Telecommunications Policy, vol. 40 (5), pp. 502-513.
Irawan, T. (2014) ‘ICT and economic development: comparing ASEAN member states’,
International Economics and Economic Policy, vol. 11 (1), pp. 97-114.
ITU (2016) Measuring the Information Society Report 2016. [Online] Available at: http://www.itu.int/en/ITU-D/Statistics/Documents/publications/misr2016/MISR2016-w4.pdf
[Accessed March 14, 2018]
Jalava, J. and Pohjola, M. (2007) ‘ICT as a source of output and productivity growth in Finland’,
Telecommunications Policy, vol. 31 (8-9), pp. 463-472.
Jalava, J. and Pohjola, M. (2008) ’The roles of electricity and ICT in economic growth: Case Finland’, Explorations in Economic History, vol. 45 (3), pp. 270-287.
Jorgenson, D. and Stiroh, K. (2000) ‘Raising the Speed Limit: U.S. Economic Growth in the Information Age’, Brookings Papers on Economic Activity, vol. 1, pp. 125-212.
Jorgenson, D. and Vu, K. (2016) ‘The ICT revolution, world economic growth, and policy issues’,
Telecommunications Policy, vol. 40 (5), pp 383-397.
Kasekamp, A. (2010) A History of the Baltic States. Basingstoke: Palgrave Macmillan.
35
Liu, S., Wu, X., Han, M., Zhang, J., Chen, B., Wu, X., Wei, W. and Li, Z. (2017) ‘A three-scale input-output analysis of water use in a regional economy: Hebei province in China’, Journal of
Cleaner Production, vol. 156, pp. 962-974.
Mas, M., Fernández de Guevara, J., Robledo, J., Cardona, M., López-Cobo, M., Righi, R. and Samoili, S. (2018) The 2018 PREDICT Key Facts Report: An Analysis of ICT R&D in the EU and
Beyond. JRC Technical Reports.
Miller, R. and Blair, P. (2009) Input-output Analysis: Foundations and extensions. 2nd ed. Cambridge: Cambridge University Press.
Ochoa, E. (1986) ‘An Input-Output Study of Labor Productivity in the U.S. Economy, 1947-72’,
Journal of Post Keynesian Economics, vol. 9 (1), pp. 111-137.
OECD (2011) ‘OECD Guide to Measuring the Information Society 2011’ [Online] Available at: http://dx.doi.org/10.1787/9789264113541-en [Accessed March 22, 2017]
OECD (2015) ‘Digital Economy Outlook 2015’ [Online] Available at: http://dx.doi.org/10.1787/9789264232440-en [Accessed March 18, 2018]
Oliner, S. and Sichel, D. (2000) ‘The Resurgence of Growth in the Late 1990's: Is Information Technology the Story?’, Journal of Economic Perspectives, vol. 14 (4), pp. 3-22.
Oliner, S. and Sichel, D. (2002) ‘Information Technology and Productivity: Where Are We Now and Where Are We Going?’, Federal Reserve Bank of Atlanta Economic Review, vol. 87 (3), pp. 15-44.
36
Plakans, A. (2011) A Concise History of The Baltic States. Cambridge: Cambridge University Press.
Rohman, I. (2013) ‘The globalization and stagnation of the ICT sectors in European countries: An input-output analysis’, Telecommunications Policy, vol. 37 (4), pp. 387-399.
Roy, S., Das, T. and Chakraborty, D. (2002) ‘A Study on the Indian Information Sector: An Experiment with Input-Output Techniques’, Economic Systems Research, vol. 14 (2), pp. 107-129.
Seo, H., Lee, Y. and Oh. J. (2009) ‘Does ICT investment widen the growth gap?’,
Telecommunications Policy, vol. 33 (8), pp. 422-431.
Solow, R. (1987) ‘We’d better watch out’, New York Times Book Review, 12 July, p. 36.
Spiezia, V. (2011) ‘Are ICT Users More Innovative? An Analysis of ICT-Enabled Innovation in OECD Firms’, OECD Journal: Economic Studies, vol. 2011 (1), pp. 99-119.
Strohmaier, R. and Rainer, A. (2016) ‘Studying general purpose technologies in a multi-sector framework: The case of ICT in Denmark’, Structural Change and Economic Dynamics, vol. 36, pp. 34-49.
Temurshoev, U., & Oosterhaven, J. (2014) ‘Analytical and empirical comparison of policy-relevant key sector measures’, Spatial Economic Analysis, vol. 9 (3), pp. 284-308.
The World Bank (2017) World Bank national accounts data. [Online] Available at: https://data.worldbank.org/indicator/NY.GDP.PCAP.CD?locations=EE-LV-LT [Accessed March 14, 2018]
The World Bank (2018) World Bank population data. [Online] Available at: https://data.worldbank.org/indicator/SP.POP.TOTL?locations=EE-LV-LT [Accessed April 5, 2018]
37
Timmer, M., Dietzenbacher, E., Los, B., Stehrer, R., and de Vries, G. (2015) ‘An Illustrated User Guide to the World Input-Output Database: the Case of Global Automotive Production’, Review of
International Economics, vol. 23 (3), pp. 575-605.
Toh, M. and Tangavelu, S. (2013) ‘An input-output study of the Singapore information sector’,
Economic Systems Research, vol. 25 (2), pp. 233-244.
Triplett, J. and Bosworth, B. (2006) ‘'Baumol's Disease' Has Been Cured: IT and Multifactor Productivity in US Services Industries’, in: Jansen, D. (ed.) The New Economy and Beyond: Past,
Present and Future. Cheltenham: Edward Elgar Publishing, pp. 34-71.
van Ark, B., O'Mahony, M. and Timmer, M. (2008) ’The Productivity Gap between Europe and the United States: Trends and Causes’, The Journal of Economic Perspectives, Vol. 22 (1), pp. 25-44.
Venturini, F. (2009) ‘The Long-Run Impact of ICT’, Empirical Economics, vol. 37 (3), pp. 497-515.
Vu, K. (2011) ‘ICT as a source of economic growth in the information age: empirical evidence from the 1996–2005 period’, Telecommunications Policy, vol. 35 (4), pp. 357-372.
Xing, W., Ye, W. and Kui, L. (2011) ‘Measuring convergence of China's ICT industry: An input-output analysis’, Telecommunications Policy, 35 (4), pp. 301-313.
Xu, Y. and Dietzenbacher, E. (2014) ‘A Structural Decomposition Analysis of the Emissions Embodied in Trade’, Ecological Economics, vol. 101, pp. 10-20.
Yang, L. and Lahr, M. (2008) ‘Labor Productivity Differences in China 1987-1997: An Interregional Decomposition Analysis’, The Review of Regional Studies, vol. 38 (3), pp. 319-341.
