104	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

I. INTRODUCTION

The literature widely recognizes that digital innovation in the form of information and communication technologies (ICT) has a significant influence on economic growth and productivity (see, for instance, Cette et al., 2005; Farkhanda, 2007; Venturini, 2009; Dimelis and Papaioannou, 2010; Heeks, 2010; Cortes and Navarro, 2011; Ahmed and Ridzuan, 2013; Rath, 2016; Niebel, 2018). As a consequence, rapid adoption of ICT in terms of access and use is observed worldwide (Chinn and Fairlie, 2007; Lam and Shiu, 2010; ITU, 2018). Adoption of ICT in advanced countries is much faster compared to underdeveloped and developing nations (Caselli and Coleman, 2001; Chinn and Fairlie, 2007). However, adoption of ICT, particularly the use of mobile phones and internet, has been rapidly increasing over time and now drives economic activity (see, for instance, Katz, 2009; Byrne et al., 2011; Avgerou et al., 2016; Rath, 2016).

While there is a considerable literature on the link between ICT development and economic growth in cross-country analysis and OECD countries, much less work has been devoted to understanding the link between ICT and economic growth with specific attention to single-country-level analysis for developing countries. We examine the impact of ICT development on economic growth in Indonesia. The research question is: Does greater access, use, and investment in the ICT sector foster economic growth in Indonesia? The motivation for undertaking this exercise is that, to best of our knowledge, there is no specific study that addresses this issue in the manner we do. Ahmed and Ridzuan (2013) is, to some extent, an exception in that it examines the impact of ICT on economic growth for ASEAN5+3 countries. By employing a panel cointegration and panel model, these authors find that labor, capital, and telecommunications investment positively affects economic growth. The limitation of this study is that it is a panel case study—a panel that includes Indonesia. There is nothing known exclusively with respect to Indonesia on this matter. That is, there is no study that takes a time-series approach to understanding the role of ICT in Indonesia’s economic growth. The other limitation of this study is that it uses telecommunications investment as a proxy for ICT, but the International Telecommunication Union (ITU) considers many other variables for ICT (ITU, 2009). Our main reason for focusing on Indonesia is as follows.

First, Indonesia is the largest country by population in Southeast Asia and it is an emerging country on the world stage. The country ranks 17th in GDP and 7th in GDP on the basis of purchasing power parity. Despite a gradual increase in economic growth and improvement in ICT infrastructure in recent years, Indonesia still lags in ICT as compared to neighboring Southeast Asian countries like Singapore and Malaysia. Indonesia is the fourth most populous country in the world. Figure 1 shows Indonesia’s ICT investment growing at an annual average of 20% over the 1980–2014 period, with GDP growing at 5.2% annually over the same time period. Thus, understanding the role played by ICT in increasing Indonesia’s economic growth is worth examining.

Second, many studies examine the role of ICT in propelling productivity or output per worker (Cronin et al., 1993; Dewan and Kraemer, 2000; Oulton, 2002; Hu and Quan, 2005; O’Mahony and Vecchi, 2005; Atzeni and Carboni, 2006; Kumar et al., 2016). Thus, examining the effect of ICT on per capita output for Indonesia

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	105

calls for investigation because the growth in ICT (as shown in Figure 1) both quantitatively (coverage, capacity, and accessibility) and qualitatively (adoption of advanced technologies, efficiency, and service quality) has been impressive.

Third, a plethora of studies examines the effect of ICT on productivity and economic growth, particularly via panel data sets (see, for instance, Roller and Waverman, 2001; Aker and Mbiti, 2010; Commander et al., 2011; Yousefi, 2011; Cardona et al., 2013; Jorgenson and Vu, 2016; Niebel, 2018). These studies find mixed evidence. This could be due to different methodologies, data sample periods, and measurement of different ICT indicators. However, studies on the ICT–growth nexus using time-series analysis for single developing countries are scarce (Kumar et al., 2016; Shahram and Woodside, 2017). One reason for this is lack of high quality data for these countries. On the other hand, the World Bank (2012) reveals that ICT has potential to increase productivity, boost economic growth, and thereby reduce poverty. Given this, the present paper fills an important gap in understanding how ICT development affects economic growth in Indonesia. Research on a single country provides in-depth analysis on the nexus between ICT and economic growth (see, for instance, Cronin et al., 1993; Khuong, 2013; Ishida, 2015; Kumar et al., 2016). Further, to best of our knowledge, the present paper is the first to examine the impact of ICT development on Indonesian growth.

Fourth, many studies investigate the causal relationship between ICT development and economic growth (see, for instance, Cronin et al., 1993; Datta and Agarwal, 2004; Shiu and Lam, 2008; Koutroumpis, 2009). Most of these studies employ bivariate analysis (i.e., only using ICT and economic growth). However, this bivariate approach may lead to bias due to omitted variables (Gross, 2012; Ishida, 2015). Therefore, the present study uses other important variables, such as total factor productivity (TFP), human capital, and ICT exports, that potentially affect economic growth.

Fifth, the International Telecommunications Union (ITU, 2009) provides a detailed methodology for constructing an ICT development index using 11 indicators. This ICT development index is very useful for comparing ICT development across countries as well as within a country over time. Numerous studies use either non-monetary indicators of ICT (teledensity, mobile phone subscribers, percentage of internet users, fixed broadband connections, and percentage of households using computers) or a monetary variable (ICT investment) to examine the nexus between ICT and growth (for instance, Caselli and Coleman, 2001; Baliamoune-Lutz, 2003; Chinn and Fairlie, 2007; Koutroumpis, 2009; Ishida, 2015). Studies are likely to face two problems using any of these variables separately. First, each of these ICT indicators will have mixed evidence on economic growth (i.e., some variables may affect growth, and some may not affect growth). Second, ambiguity may arise over the real contribution of ICT on growth in case of exclusion of some of these ICT indicators. In such circumstance, the ICT development index developed by the ITU can exactly assess its impact on economic growth. However, ICT development index data are not readily available for longer time-series in the case of Indonesia due to unavailability of data on these 11 indicators. Thus, the present study constructs an ICT development index using only three variables (telephone subscriptions, cellular phone subscriptions, and

106	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

fixed broadband subscriptions) employing principal component analysis.(Park et al., 2015; Rath, 2016).

Sixth, the present study differs from the literature by employing an Autoregressive Distributed Lag (ARDL) model with two structural breakpoints in the data, following Narayan and Popp (2010). To the best of our knowledge, no earlier study, including those that focus on single-country analysis, employs this technique. Examination of the long-run cointegrating relationship between ICT-related variables and economic growth in the presence of structural breaks would, ideally, reveal the true nature of this relationship. Gregory and Hansen (1996) nicely illustrate the importance of the cointegration model with the presence of structural breaks; for Indonesia, the importance of structural breaks in cointegration is demonstrated by Sharma and Syarifuddin (2019). The importance of modeling breaks in general is shown in Sharma et al., (2018).

Our approach provide the following insights. First, we find a long-run relationship between the ICT development index and economic growth. Second, the ICT index positively affects economic growth, whereas exports of ICT negatively affect growth in the long-run. Third, to check robustness, we use real GDP instead of per capita output as the measure of growth. We find that the ICT development index and ICT exports significantly affect growth positively and negatively, respectively, which corroborates our earlier findings. Apart from ICT development, this study uses TFP growth and human capital as other determinants of economic growth. Our results further reveal that apart from ICT development, both human capital and TFP growth boost economic growth in the long-run in the case of Indonesia, consistent with previous literature.

This paper proceeds as follows. Section II presents an overview of the performance of ICT-related variables and economic growth in Indonesia. In Section III, we explain our theoretical framework and empirical methodology pertaining to the nexus between economic growth and ICT. Section IV presents the data and results. Finally, Section V sets forth concluding remarks.

II. ICT IN INDONESIA

Figure 1 presents the annual growth rates of real GDP and ICT for Indonesia from 1980 to 2014. The graph shows that the annual growth rate of real GDP is lower than the annual growth of the ICT development index in most of the study period, except 1989, 2000, and 2014. The graph also shows that the gap between ICT development and real GDP growth rates was lower during the period 1981 to 1988, but thereafter widened drastically. This is due to tremendous growth of ICT development as compared to real GDP growth in Indonesia. The average annual ICT growth rate was quite impressive particularly over the 2001–2008 period, with annual growth of over 40%. Average annual ICT growth over the 1981–2014 period was observed at around 20%, whereas average real GDP growth was observed at 5.2% for the same period.

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	107

Figure 1. The Growth Rates of Real GDP and ICT in Indonesia

This figure exhibits the growth rates of real GDP and ICT in case of Indonesia from 1980 to 2014. The graph shows that the annual growth rate of real GDP is lower than the annual growth of ICT development index in most of the periods under the study except three years (1989, 2000 and 2014). The ICT development was growing slowly during 1980 to 1989, but thereafter, it has shown tremendous growth particularly from 2000 to 2008. The average annual ICT growth during 1981-2014 was observed around 20%, whereas, the average real GDP growth was seen at 5.2% during same periods.

60,0
50,0
40,0
30,0
20,0
10,0
0,0
-10,0
-20,0	GDP growth	ICT growth
-30,0
	1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Source: World Development Indicators (WDI) database

Figure 2 shows several key indicators related to exports of ICT for Indonesia. Since there is no availability of ICT export data prior to 1990, this figure presents three indicators of ICT exports from 1990 to 2016. The trend-line of high-technology exports (measured as a percentage of manufacturing exports) shows that the share of exports of high-technology products to total manufacturing products increased from 1.59% in 1990 to 16.54% in 2005. Thereafter, this share of high-technology exports declined, except in 2009, and reached 5.78% in 2016. Since high-technology export figures may not be directly attributable to ICT development of the country, we show exports of ICT goods and services, which are directly related to ICT development. ICT goods exports are measured as a percentage of total goods exports, and ICT services export figures are the share ICT service exports to total service exports in the balance of payment account. The trendlines indicate that ICT exports of services are higher than ICT goods exports, except for 2003 and 2004, where the share of ICT goods exports was slightly higher than ICT service exports. Figure 2 also indicates that ICT good exports show a downward trend over the period 2003 to 2016. ICT service export share shows an upward trend from 1997 to 2007 and thereafter declines till 2016. Share of ICT service exports was just 4% in 2016; it peaked in 2007 at 13.5%.

108	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

Figure 2. ICT Exports of Indonesia

This figures portrays the ICT related exports of Indonesia. The trend line of high-technology exports indicates an upward trend in share of its exports (measured as a % to manufacturing exports) from 1990 to till 2002 and thereafter it declined and reached around 7% in 2016. ICT service exports also shows a similar trend along with the trend line of high-technology exports.

20
					High-technology export							ICT Goods export							ICT Service export
15
10
5
0
1990	1991	1992	1993	1994	1995	1996	1997	1998	1999	2000	2001	2002	2003	2004	2005	2006	2007	2008	2009	2010	2011	2012	2013	2014	2015	2016

Source: WDI database.

Figure 3 presents four key indicators pertaining to overall economic performance of Indonesia. First, the growth of output per worker shows a zigzag pattern from 1981 to 2014. Output per worker growth rate declined from 1981 to 1985, then began to increase and peaked at 9.4% in 1995. This shows a fluctuating trend with increases and decreases until 2014. Growth of capital per worker shows a similar pattern, except that prior to 1997, growth of capital per worker was higher than output per worker. Post 1997, growth of output per worker was slightly higher than growth of capital per worker. TFP growth shows a declining trend in recent years as compared to 1981. TFP growth was 1.1% in 1981 and it almost consistently maintained the same rate until 1995, but TFP growth declined from 1.1% in 1995 to 0.84% in 2000. Although it began to increase after 2000, growth was relatively lower most years post-2000 as compared to the pre–Asian Financial Crisis period. Figure 3 also shows total investment made by Indonesia in the telecommunications sector, which we treat as the ICT investment. Investment in the ICT sector plays a critical role in production processes, and thereby in the economic growth of a country. ICT investment is measured as a percentage of GDP. Although the ratio is less than 1% of GDP, the ICT investment figures in absolute terms significantly increased from 1981 to 2017.

110	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

Table 1 presents basic data on key ICT development indicators for Indonesia. This table also clearly reflects the non-availability of key ICT indicators for a longer time horizon, which is the prime reason for the absence of country-specific research on the nexus between ICT and economic growth, particularly for developing countries, including Indonesia. Table 1 shows five indicators pertaining to ICT that are broadly used to examine the effect of ICT on economic growth or productivity. The second column of Table 1 shows data on fixed-telephone subscriptions per 100 inhabitants. The figures clearly show that access to telephone landline connections was very low in 1980 (0.25), increasing to around 17 in 2010. But this figure began declining from 2010 and reached 4.23 in 2017. The reason for this decline is high penetration of mobile phones. Subscription to mobile phones has increased phenomenally across the globe, and Indonesia is no exception in this regard (as shown in column 3). The other ICT use indicators, such as fixed broadband connections, internet users, and households using computers also increased, particularly from 2005 (last three columns of Table 1). However, these figures are very low as compared to developed countries. Overall, this preliminary analysis indicates a general increase in ICT development in Indonesia in recent years as compared to 1980 or even 2000.

III.FRAMEWORK AND METHODOLOGY

A. Framework

We employ the commonly used Cobb-Douglas production function, as stated in the

Solow (1956) framework and following Kumar et al. (2016, 2015) and Rao (2010).

The equation for output per worker (yt) can be written as:

(1)

where At is technological progress, kt refers to capital per worker, and α is share of capital. The Solow (1956) model assumes that the progression of technology is given by

(2)

where g and A0 are the growth of technological progress and initial TFP, respectively. πt refers to aggregate technology, At. The impact of ICT development on TFP can be assessed when ICT development arrives as a shift in production function. Thus, we use an augmented Solow growth model by including the ICT- related variables in the production function.

Let

(3)

Here θt is part of the technology component in equation (1), and At is redefined as follows:

(4) Thus, we can rewrite equation (1) as

(5)

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	111

Taking the natural logarithm of equation (5) yields the linear general production function equation for estimating the model as

(6)

where δ is the intercept term, β is the elasticity coefficient of ICT development, and e is the error term.

Equation (6) can be further expanded by adding the control variables ICT exports and structural break dummy variables, which can be presented as:

(7)

where ICTEXP is ICT exports, y is the coefficient of elasticity of ICT exports, TD is the structural break dummy based on two-stage structural unit root break test (Narayan and Popp, 2010) of the dependent variable (i.e., per capita output), σ is the coefficient associated with TD, b is the double break period, which is less than t, and t is the time period, t = 1980, 1981,….., 2014.

B.Methodology B1. Unit Root Tests

Let us now empirically analyze the stationary property of all time-series variables used in this study. First, Augmented Dickey-Fuller (1979) and Philips-Perron

(1988) unit root tests are employed to check stationarity. These conventional tests do not address the presence of structural breaks in the data. To account for structural breaks, we further employ the Narayan and Popp (2010) unit root test, which captures structural breaks.

B2. ARDL Model

After checking the unit root, we next use an ARDL bound testing approach (Pesaran et al., 2001) to inspect the long-run association between ICT and economic growth. We apply the ARDL model to equation (7). The main advantages of the ARDL model are: (i) it can use any data series irrespective of whether the variable is stationary I(0) or non-stationary I(1) or mixed order of condition I(0) and I(1);

(ii)it can generate the unknown parameters for both the long-run and short-run without dropping any information; and (iii) the model simply takes care of the endogeneity issue among the explanatory variables (Pesaran et al., 2001). The procedure for ARDL follows two steps. First, it examines the presence of a long- run relationship between the variables. Second, it obtains the long-run and short- run coefficients of explanatory variables using ARDL and error correction models (ECMs) together. Based on conjectural reinforcement and related work, we employ the following models:

Model I:	(8)
Model II:	(9)

112	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

where y represents output per worker, k refers to capital per worker, ICTINDX is defined as ICT development index, ICTEXP is ICT exports, GDP is real gross domestic product, TFP is the TFP, and HC is the human capital in equation (8) and equation (9). Thus, we can write the error correction representation of the ARDL bound testing for model I and model II as, respectively:

(10)

(11)

In equations 10 and 11, refers to the first difference operator of the corresponding variable. To inspect the cointegration relationship between ICT development and economic growth, the null hypothesis (no cointegration) can be elaborated as: H0:δ1 = δ2 = δ3 = δ4 = 0, and the alternative hypothesis (presence of cointegration) as H1: δ1≠ δ2 ≠ δ3≠ δ4 ≠ 0 by conducting the F test developed by Pesaran et al. (2001) and later refined by Narayan (2005). If the calculated F-statistics are less than the lower bound value I(0), then the null hypothesis of no cointegration cannot be rejected. On the contrary, if the calculated F-statistic is greater than upper bound value I(1), then we reject the null hypothesis. However, if the value of the F-statistic falls between the lower and upper bounds, then the relationship among the variables is inclusive.

Next, we examine the short-run relationship by employing an error correction mechanism. The short-run dynamic ECMs can be explained using equations (12-13).

(12)

(13)

In equations 12 and 13, σ is the speed of adjustment parameter and ECM is the residual series from the estimated model.

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	113

IV. DATA AND RESULTS

A. Data

This study uses annual data for Indonesia for the period 1980–2014.1 The data are collected mainly from three sources. Data on real GDP, GDP growth, and ICT exports are collected from the World Development Indictors. ICT exports (ICTEXP) are measured as exports of communications, computers, etc., as a percentage of service exports in the balance of payments account. Data on output per worker and capital per worker are estimated using Penn World Table 9.0, from Feenstra et al. (2015). Similarly, data on human capital and TFP growth are directly collected from the Penn World Table. Finally, all variables related to information communication technologies are collected from the ITU database. All variables related to ICT are presented in Section 2. Typically, the empirical literature on the relationship between ICT and economic growth broadly uses fixed-telephone subscriptions per 100 inhabitants, fixed broadband subscriptions per 100 inhabitants, fixed internet subscriptions per 100 inhabitants, mobile- cellular subscriptions per 100 inhabitants, and proportion of households with use of computer as measures of ICT development. We employ these variables in Section 2. Ideally, one could examine the impact of each of these ICT indicators on Indonesia’s overall economic growth. However, due to non-availability of data for longer periods and missing data for key indicators like fixed broadband use, use of internet, and computer use, the present study confines its analysis by constructing an ICT development index using principal component analysis.2

B. Results and Discussion

This section presents the results of the effect of ICT on economic growth. The empirical analysis includes examination of unit root test, cointegration, and ECM.

Table 2.

Results for Narayan and Popp (2010) Unit Root Test with Two Structural Breaks

This table presents the unit root test with two endogenous structural breaks proposed by Narayan and Popp (2010). The results indicate that the TFP and ICTINDX are stationary at the level form and the remaining variables are I(1), i.e. nonstationary at levels. M1 stands for a model which includes an intercept whereas the M2 model includes both an intercept and a time trend. TB1 refers to the first break while TB2 denotes the second break. Critical values for M1 are: −5.259 (1%); −4.514 (5%); −4.143 (10%). Critical values for M2 are: −5.949 (1%); −5.181 (5%); −4.789 (10%). k stands for optimum lag. T refers to observations. Finally, *** (**) represent statistical significance at 1% (5%) levels.

	Variable	T	M1				M2
			Test Statistic	TB1	TB2	k	Test Statistic	TB1	TB2	k
	TFP	35	-16.67***	1987	1996	0	-2.207	1997	2004	3
	lnGDP	35	-1.43	1997	1999	0	-4.105	1988	1997	2
	lny	35	-2.441	1994	1997	0	-3.914	1994	1997	0
	lnk	35	0.5849	1996	1998	0	-0.335	1996	1998	0
	ICTINDX	35	-4.855**	1997	1999	3	-2.689	1999	2005	5
	lnICTINV	35	0.7131	2000	2006	0	0.9176	2000	2006	4
	lnICTEXP	35	-2.789	1987	2003	5	-1.95	1988	2003	0
	FTS	35	-2.92	2005	2007	2	-4.167	2003	2006	4
	MPS	35	-1.52	1994	1997	0	3.534	1994	1998	5

1The empirical results in this paper stem from data prior to 2014 because of non-availability of data for various indicators used in the ARDL model.

2The factor scores and corresponding eigenvalues for constructing the ICT index are not presented here but the results are available upon request.

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	115

After identifying the cointegrating relationship, we now estimate the long- run coefficients using equation (10) for model I. The results of long-run elasticities are presented in Table 4. Table 4 offers the following insights. First, a 1% increase in capital per worker (k) on average increases output per worker by 0.28%. Theoretically, one would always expect a positive relationship between the capital–labor ratio and output per worker. As per capita capital increases, there is greater use of machinery and equipment per worker and thus an increase in output per worker. As an emerging country, both Indonesia’s capital stock and number of workers is increasing. Our preliminary results in Table 3 also show a strong positive correlation between output per worker and capital per worker, which supplements the finding of long-run elasticity obtained in Table 4. Second, the long-run elasticity coefficient of the ICT development index (ICTINDX) is positive and statistically significant at the 1% level. This result indicates that a 1% increase in overall ICT development on average increases output per worker (as a measure of economic growth) by 0.08%. Although this coefficient is very small, it nevertheless reflects the importance of greater access to and use of ICT in Indonesia’s overall growth process. As a developing country where overall ICT development is quite low (see preliminary analysis in Section 2) compared to developed countries or even neighboring countries like Singapore and Malaysia, the positive effect on output per worker is notable. Third, ICT exports1 negatively affect output per worker, which is bit surprising. But the coefficient has only a weak effect (only at the 10% level of significance) and its magnitude is negligible. The structural dummy (TD) variable introduced in the model shows a negative sign with 1% level of significance. The coefficient 0.30 implies that output per worker declines by 0.3% in the break periods as compared to the remaining sample periods.

Table 5.

Error Correction Representation for Selected ARDL for Model I

This table presents the error correction representation for the selected ARDL for model, and *** (**) indicate statistical significance at the 1% (5%) levels. The diagnostic tests are reported in the second half of the table.

Regressor	Coefficient	t-statistic
Δlnk	1.48***	7.30
ΔlnICTINDX	0.06	1.47
ΔlnICTEXP	-0.02**	-2.39
ΔlnTD	-0.009	-0.45
ECMt-1	-0.76***	-3.59
Diagnostic tests	Coefficient	p-value
R2	0.93
Adjusted R2	0.85
χ2 Auto(1)	1.52	0.25
χ2 Norm(2)	1.35	0.50
χ2 Hetero(1)	0.20	0.65
CUSUM	Not stable
CUSUM -SQ	Stable

4We also run an ARDL model by replacing ICT imports with ICT exports in Equation (8). However, we obtain no robust results. The chosen model (4,4,3,3,4) based on the Schwartz criterion with 4 lags yields an F-stat of 12.56, which rejects the null hypothesis of no cointegration at the 1% level of significance. Although there exists a long-run relationship among these variables, none of the long- run slope coefficients are statistically significant (at the 10% level). The cointegration equation can be

written as			. The error correction
coefficient turns out to be 0.02 with probability value of 0.06.

116	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

After studying the long-run elasticities of model I, it is important to discuss the speed of adjustment through ECM and the short-run relationship of model I. The results are illustrated in Table 5. The coefficient of the error correction term (ECMt-1) is (-0.76) and is significant in the short-run, which confirms that long-run equilibrium exists among the variables in model I. The coefficient -0.76 implies that once shocked, convergence to equilibrium is very high with 76% rectification occurring in the first year. The short-run impact of other explanatory variables is the same as the long-run relationship except that the ICT development index (ICTINDX) does not significantly affect output per worker in the short-run, which is as expected. Greater use of and investment in ICT will always impact the overall growth and productivity of a country in the long-run. Based on the diagnostic tests presented in Table 5, we can rely on the results of the ARDL bound testing approach.

Table 6.

Estimated Long-run Elasticities Using ARDL Approach for Model II (2,4,4,0,3)

Based on Schwarz Bayesian Criterion

This table depicts the long-run elasticities results of based real GDP as economic growth using an ARDL (2,4,4,0,3) model. Finally,

*** indicates statistical significance at 1% level.

Variable	Coefficient	t-statistic
lnTFP	1.53***	5.21
lnHC	2.18***	4.31
lnICTINDX	0.10***	2.72
lnICTEXP	-0.04	-1.4
C	25.47***	-64.49

After discussing the empirical results of model I, let us now discuss the results obtained from model II. We use model II5 as an alternative specification in which real GDP is treated as the dependent variable and TFP growth, human capital, ICT development, and ICT exports are included as the regressor. The bound testing results presented in Table 3 reject the null of no cointegration at the 1% level of significance based on both Narayan (2005) and Pesaran (2001) critical values. Thus, our finding indicates a long-run relationship among the variables. Next, the long-run elasticity coefficients are presented in Table 6. The effect of ICT development appears similar to the results presented in Table 4. The coefficient of ICT development (ICTINDX) indicates that a 1% increase in ICT development on average increases the economic growth of Indonesia by 0.10%. Similarly, note that the ICT export coefficient is negative but not statistically significant, which is similar to the results obtained in Table 4. In addition to ICT development and ICT exports, TFP growth and human capital positively influence Indonesia’s economic

5We also use an alternative form of model II, based on a reviewer suggestion. We run the model by taking labor and capital inputs along with other explanatory variables already presented in Equation (9). However, the results based on the ARDL bounds testing approach produces no convincing results. We select the ARDL (4,4,4,4,4) model based on Schwarz Criterion with 4 lags. The bound test result (F-stat. = 6.87) shows evidence of a long-run relationship between economic growth and other variables at the 1% level of significance. The cointegration equation can be written

as: The

error correction coefficient term is -3.01 with a probability value of 0.18. The model is also not stable based on the CUSUM and CUSUM SQ plots.

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	117

growth. In the literature, it is well established that TFP growth and human capital play a critical role in boosting national economic growth. Thus, the results in Table 6 are not surprising.

Table 7.

Error Correction Representation for Selected ARDL for Model II

This table presents the error correction representation for the selected ARDL for model and *** (*) indicate statistical significance at 1% (10%) levels. The diagnostic tests are reported in the second half of the table.

Regressor	Coefficient	t-statistic
ΔlnTFP	0.79***	12.75
ΔlnHC	0.43***	2.80
ΔlnICTINDX	0.02*	1.89
ΔlnICTEXP	-0.007	-1.27
ECMt-1	-0.19***	-3.28
Diagnostic tests	Coefficient	p-value
R2	0.87
Adjusted R2	0.85
χ2 Auto(1)	1.92	0.13
χ2 Norm(2)	2.76	0.25
χ2 Hetero(1)	1.76	0.25
CUSUM	Stable
CUSUM -SQ	Stable

After discussing the long-run elasticities, we present the results of ECM based on equation (13). The results are presented in Table 7. The coefficient of ECMt-1 is negative and statistically significant at the 1% level. The coefficient -0.19 confirms a long-run relationship between economic growth and ICT development, along with other control variables. The error correction term implies that the speed of convergence to equilibrium is relatively slow, with 19% adjustment occurring in the first year. TFP growth, human capital, and ICT development positively affect economic growth in the short-run, but the effect of ICT on economic growth is very weak in the short-run. The results of cointegration and ECM are quite reliable based on various diagnostic tests.

V. CONCLUSIONS

This paper investigates the impact of ICT on economic growth in Indonesia. Although many studies examine the impact of ICT development on economic growth or TFP, our focus exclusively on Indonesia is our main contribution to the literature. A plausible reason that the nexus between ICT and economic growth is less explored for individual emerging countries is the non-availability of data for all ICT-related variables. Our paper fills this research gap. To address this issue, we use the augmented neoclassical growth framework and the ARDL model to examine the long-run and short-run relationship between ICT development and economic growth by considering annual data from 1980 to 2014. The major findings can be summarized as follows. First, preliminary analysis indicates that the key indicators of ICT development have shown an increasing trend in recent years as compared to 1980 or even 2000. Second, empirical results based on the ARDL model

118	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

show evidence of a long-run cointegrating relationship between ICT development and economic growth. Third, we use two alternative models, one based on a production function approach and the other on determinants of economic growth by including ICT development as a regressor. The coefficient of long-run elasticity of economic growth with respect to change in ICT development is positive and statistically significant. This implies that ICT development positively affects the economic growth of Indonesia; however, the magnitude of this effect is very low. Other indicators like capital per worker, TFP and human capital also positively affect economic growth in Indonesia. Fourth, based on the export-led growth hypothesis, we use ICT exports as a regressor and assess its effect on economic growth. The results, surprisingly, show a negative effect on economic growth, although the impact is weak. From a policy perspective, our study shows that indicators of ICT are imperative for long-run growth in Indonesia. Thus, Indonesia should focus on not only increasing the use of ICT, but also target increased ICT- related investment. National governments always play a critical role in fostering adoption of ICT-related indicators; thus, an effective strategy to reap ICT’s benefits would promote improved economic growth.

Acknowledgement: The authors gratefully acknowledge the suggestions from the Managing Editor and anonymous referees on an earlier draft of this paper. The first author is thankful to Dr. Dinh Phan for providing the ICT related data. The usual disclaimer applies.

REFERENCES

Ahmed, E.M., and Ridzuan, R. (2013). The Impact of ICT on East Asian Economic Growth: Panel Estimation Approach. Journal of Knowledge Economy, 4, 540-555. Aker, J. C., and Mbiti, I. M. (2010). Mobile Phones and Economic Development in

Africa. Journal of Economic Perspectives, 24, 207–232.

Atzeni, G.E., and Carboni, O.A. (2006). ICT Productivity and Firm Propensity to Innovative Investment: Evidence from Italian Microdata. Information Economics

and Policy, 18, 139–156.

Avgerou, C., Hayes, N., and Rovere, R.L. (2016). Growth in ICT Uptake in Developing Countries: New Users, New Uses, New Challenges. Journal of Information Technology, 31, 329-333.

Baliamoune-Lutz, M. (2003). An Analysis of the Determinants and Effects of ICT Diffusion in Developing Countries. Information Technology for Development, 10, 151-169.

Byrne, E., Nicholson, B., and Salem, F. (2011). Information and Communication Technologies and the Millennium Development Goals. Information Technology for Development, 17, 1-3.

Cardona, M., Kretschmer, T., and Strobel, T. (2013). ICT and Productivity: Conclusions from the Empirical Literature. Information Economics and Policy, 25, 109–125.

Caselli, F., and Coleman, W.J. II. (2001). Cross-country Technology Diffusion: the Case of Computers. American Economic Review, 91, 328-335.

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	119

Cette, G., Jacques M., and Yusuf K. (2005). ICT Diffusion and Potential Output Growth. Economics Letters, 87, 231-234.

Chinn, Menzie D., and Fairlie, R.W. (2007). The Determinants of the Global Digital Divide: A Cross-Country Analysis of Computer and Internet Penetration. Oxford Economic Papers, 59, 16-44.

Commander, S., Harrison, R., and Menezes-Filho, N. (2011). ICT and Productivity in Developing Countries: New Firm-Level Evidence from Brazil and India. Review of Economics and Statistics, 93, 528–541.

Cortes, E.A., and Navarro, J.A. (2011). Do ICT influence economic growth and human development in European Union Countries. International Advance in Economic Research, 17, 28-44.

Cronin, F.J., Colleran, E.K., Herbet, P., and Lewitzky, S. (1993).Telecommunications and Growth: The Contribution of Telecommunications Infrastructure Investment to Aggregate and Sectoral Productivity. Telecommunications Policy, 17, 677–690.

Datta, A., and Agarwal, S. (2004). Telecommunication and Economic Growth: A Panel Data Approach. Applied Economics, 36, 1649–1654.

Dewan, S., and Kraemer, K. (2000). Information Technology and Productivity: Evidence from Country-Level Data. Management Science, 46, 548–562.

Dickey, D. A., and Fuller, W. A. (1979). Distribution of the Estimators for Autoregressive Time Series with a Unit Root. Journal of the American Statistical Association,, 74, 427-431.

Dimelis, S.P. and Papaioannou, S.K., (2010). FDI and ICT Effect on Productivity Growth: A Comparative Analysis of Developing and Developed Countries. The European Journal of Development Research, 22, 79-96.

Farkhanda, S. (2007). The ICT Environment, Financial Sector and Economic Growth: A Cross-Country Analysis. Journal of Economic Studies, 34, 352-370.

Feenstra, Robert C., Robert I., and Timmer, M.P. (2015). The Next Generation of the Penn World Table. American Economic Review, 105, 3150-82.

Gregory, A.W., and Hansen, B.E., (1996). Residual-Based Tests for Cointegration in Models with Regime Shifts. Journal of Econometrics, 70, 99-126.

Gross, C. (2012). Explaining the (non-) Causality Between Energy and Economic Growth in the U.S. – A Multivariate Sectoral Analysis. Energy Economics, 34, 489-

499.

Heeks, R. (2010). Do Information and Communication Technologies (ICTs) Contribute to Development? Journal of International Development, 22, 625-640.

Hu, Q., and Quan, J. (2005). Evaluating the Impact of IT Investments on Productivity: A Causal Analysis at Industry Level. International Journal of Information Management, 25, 39–53.

Ishida, H. (2015). The Effect of ICT Development on Economic Growth and Energy Consumption in Japan. Telematics and Informatics, 32, 79-88.

ITU., (2009). Measuring the Information Society: ICT Development Index. http:// www.itu.int/ITU-D/ict/publications/idi/2009/index.html.

ITU., (2018). Measuring the Information Society Report -2018, Volume 1. Jorgenson, D. W., and Vu, K. M. (2016). The ICT Revolution, World Economic

Growth, and Policy Issues. Telecommunications Policy, 40, 383–397.

120	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

Katz, R.L. (2009). The Economic and Social Impact of Telecommunications Output. Intereconomics, 44, 41-48.

Khuong, M.Vu. (2013). Information and Communication Technology (ICT) and Singapore’s Economic Growth. Information Economics and Policy, 25, 284-300.

Koutroumpis, P. (2009). The Economic Impact of Broadband on Growth: A Simultaneous Approach. Telecommunications Policy, 33, 471-485.

Kumar, R.R., Kumar,R.D., and Patel, A. (2015). Accounting for Telecommunications Contribution to Economic Growth: A Study of Small Pacific Island State. Telecommunications Policy, 39, 284–295.

Kumar, R.R., Stauvermann, P.S., and Samitas, A. (2016). The Effects of ICT on Output per Worker: A Study of the Chinese Economy. Telecommunications Policy, 40, 102-115.

Lam, Pun-lee., and Shiu, A. (2010). Economic Growth, Telecommunications Development and Productivity Growth of The Telecommunications Sector: Evidence Around the World. Telecommunication Policy, 34, 185-199.

Narayan, P.K, and Popp, S. (2010). A New Unit Root Test with Two Structural Breaks in Level and Slope at Unknown Time. Journal of Applied Statistics, 37, 1425-1438.

Narayan, P.K. (2005). The Saving and Investment Nexus for China: Evidence from Cointegration Tests. Applied Economics, 37, 1979–1990.

Niebel, T. (2018). ICT and Economic Growth – Comparing Developing, Emerging and Developed Countries. World Development, 104, 197-211.

O’Mahony, M., and Vecchi, M. (2005). Quantifying the Impact of ICT Capital on Output Growth: A Heterogeneous Dynamic Panel Approach. Economica, 72, 615–633.

Oulton, N. (2002). ICT and Productivity Growth in the UK. Oxford Review of Economic Policy, 18, 363–379.

Park, S.R., Choi, D.Y., and Hong, P. (2015). Club Convergence and Factors of Digital Divide Across Countries. Technological Forecasting and Social Change, 96, 92–100.

Pesaran, M.H., Shin,Y., and Smith, R. (2001). Bounds Testing Approaches to the Analysis of Level Relationships. Journal of Applied Econometrics, 16, 289–326.

Phillips, P.C.B., and Perron, P. (1988). Testing for a Unit Root in Time Series Regression. Biometrika, 75, 335-346.

Rao, B.B. (2010). Estimates of the steady state growth rates for selected Asian countries with an extended Solow model. Economic Modelling, 27, 46–53.

Rath, B.N. (2016). Does the Digital Divide Across Countries Lead to Convergence? New International Evidence. Economic Modelling, 58, 75-82.

Roller, L.H. and Waverman, L. (2001). Telecommunications Infrastructure and Economic Development: A Simultaneous Approach. American Economic Review, 91,

909- 923.

Shahram, S, and Woodside, J.M. (2017). Emerging Markets: The Impact of ICT on the Economy and Society. Digital Policy, Regulation and Governance, 19, 383-396.

Sharma, S.S., Tobing, L., and Azwar, P. (2018). Understanding Indonesia’s Macroeconomic Data: What Do We Know and What are the Implications? Bulletin of Monetary Economics and Banking, 21, 217-250.

Sharma, S.S. and Syarifuddin, F. (2019). Determinants of Indonesia’s Income

Velocity of Money. Bulletin of Monetary Economics and Banking, 21, 323-342.

Do Information and Communication Technologies Foster Economic Growth in Indonesia?	121

Shiu, A., Lam, P.-L. (2008). Causal Relationship Between Telecommunications and Economic Growth in China and Its Regions. Regional Studies, 42, 705–718.

Solow, Robert. M. (1956). A Contribution To the Theory of Economic Growth. The Quarterly Journal of Economics, 70, 65-94.

Venturini, F. (2009). The Long-run Impact of ICT. Empirical Economics, 37, 497-515. World Bank. (2012). ICT for Greater Development Impact. Available at http:// documents.worldbank.org/curated/en/285841468337139224/ICT-forgreater-

development-impact-World-Bank-Group-Strategy-for-2012-2015.

Yousefi, A. (2011). The Impact of Information and Communication Technology on Economic Growth: Evidence from Developed and Developing Countries. Economics of Innovation and New Technology, 20, 581–596.

122	Bulletin of Monetary Economics and Banking, Volume 22 Number 1, 2019

This page is intentionally left blank