Sri Lanka, commonly referred to as 'the pearl of the Indian Ocean' is an exquisite little island, rich in natural resources, strategically placed in the world trade route and recognized as "one of the few functioning democracies in the Third World". (Grobrar, 1993). With a multiethnic population of approximately 20 million people Sri Lanka was at the heart of international media coverage over the last three years as a result of the intensifying internal armed conflict which some described as an 'unwinnable' war.
History
The Liberation Tigers of Tamil Eelaam (LTTE) began its journey to divide the country into two in the late 1970's with their "demands to form a separate state in the (mainly Tamil) northern and eastern provinces of the country" (Grobrar, 1993). In 1983 the LTTE initiated a full scale war against the democratically elected Sri Lankan Government (GoSL), which saw the GoSL losing its sovereignty for the first time after gaining independence. Since then, Sri Lanka has been under a state of emergency and over the years her population witnessed the brutality of terrorism as thousands of innocent men, women, children and a large number of eminent individuals lost their lives to suicide attacks by this barbaric outfit called the LTTE.
Prior to the commencement of the war, the local economy was one of the more successful South Asian economies and had a GDP growth rate of 8.2% between 1968 and 1978 which is the highest ever achieved by Sri Lanka. Tea and Rubber accounted for 70% of total exports while 50% of employment was in the Agriculture sector which was also the main source of income accounting for approximately 26% of GDP. The defence budget stood at less than half percent of GDP with an armed force of a mere 12000. According to Arunatillake (2001), defence expenditure rose steeply to 3.5% of GDP by 1985 and was at 6% of GDP in 1996 as the strength of the armed forces increased far beyond 200,000.
Between 1977 and 1982, Sri Lanka's investment to GDP ratio rose from 14% to 31% and multinationals like Sony and Motorola were interested in investing in Sri Lanka but later reconsidered their decisions as a result of the war. Real GDP growth averaged at 6.5% as inflation which read a high 26% in 1980 was restricted to 1.5% in 1985. According to Central Bank figures tourist arrivals were on the decline since 1982 and tourism had its first major blow on the 3rd of May 1986 when a bomb exploded on an Air Lanka flight on route to Male. The prevailing "security situation also contributed to a decrease in the momentum of foreign investment in the country" (Ross, 1986), in addition to having severe negative impacts on the Fishing sector and employment as a whole. Consequently data reveals that, between 1983-89 GDP growth declined to an average of 3¾ percent.
In 2000 Sri Lanka faced a foreign exchange crisis as a consequence of large imports of military equipment and escalating oil prices in the world market. 2001 saw the country's only international airport come under a terrorist attack and the by then end of that year, the Sri Lankan economy was in dire straits as it went into its first ever recession when a negative GDP growth rate of 1.5% was recorded. The newly elected GoSL was finding itself unable to continue with military operations and on the 22nd of February 2002, the then Prime Minister, Mr. Ranil Wickremasinghe signed a Cease Fire Agreement (CFA) with the LTTE. The optimism which surrounded the peace process lead to a positive impact on the Sri Lankan economy in terms of an increase in the country's GDP and is illustrated more clearly in the table below.
Table 1: GDP Growth with and without conflicts
Province
Pre Ceasefire
(1997-2001)
Post Ceasefire
(2002-2003)
Northern
3.4%
12.6%
Eastern
4.6%
10.1%
North Central
-0.2%
8.2%
Western
6.0%
6.2%
All Island
3.9%
5.9%
(Annual Report, Various Issues, Central Bank of Sri Lanka)
However, on the 26th of December 2004 a tsunami hit the coastal belt of Sri Lanka taking with it the lives of more than 30,000 people and leaving the economy in complete distress as The Asian Development Bank estimated the reconstruction costs at $1.5 billion US Dollars.
By the year 2005 the terrorists controlled one-fourth of Sri Lankan territory and two-thirds of its coastline. The picture below illustrates the ground situation during that period with the LTTE controlled areas being highlighted in grey.
(Figure 1: Ministry of Defence, Sri Lanka)
This meant that 29% of the land mass was in the conflict affected areas and 13% of the island's labour force was not fully contributing to the economy. The Presidential Elections which followed later that year saw Hon. Mahinda Rajapaksa being elected as the President of Sri Lanka. After repeatedly failed attempts to recall the LTTE to the negotiation table, peace talks collapsed on the 08th of June 2006. The ground situation worsened on the 20th of July 2006 as the Tamil Tigers (LTTE) decided to close the sluice gates of an eastern reservoir stopping the supply of water to 60,000 people and prompting the Sri Lankan Government to launch its first major offensive since the ceasefire.
The 18th of May 2009 went down in history as the GoSL formally declared an end to the longest running armed conflict in Asia. A proud Sri Lankan President, H.E. Mahinda Rajapaksa addressed his Parliament as he said "this session of Parliament opens in a country where the writ of this august legislature spreads equally throughout the 65, 332 sq. km of territory of Sri Lanka."
"With an estimated 70,000 casualties over the years, it was a bitter and hard fought victory, one of the few instances in modern history in which a terrorist group had been defeated militarily." (US Senate, 2009, Pg1). Over the years, Sri Lanka's war related expenditure rose exponentially and by 2009 it stood at Rs. Million 187, 967 which accounted for a 3000% increase in comparison to 1985 figures. According to Ross (1986) this "generation of economic activity" comes at an opportunity cost as scarce foreign exchange is allocated for defence expenditure instead of the development sector. All defence equipment was imported from countries like Russia, India, Pakistan and the United States which meant that the increasing defence expenditure had no productive or multiplier effects on the local economy. While high defence expenditure has been the primary cause for budget deficits over the last three decades, future budgets will still have to bear the burden of reconstruction and rehabilitation following the end of the war.
Post-War Economic Outlook
Terrorism was clearly the most serious impediment for Sri Lanka's economic development over the last three decades. "It was a financial burden, a hindrance for investment and crippled several areas of production in the economy and the economy performed at less than its potential." (The Sunday Times, 2010). Following the complete eradication of terrorism, Sri Lanka - "the rising star of the Asian economy" (FOX News, 2010) is today striving to be Asia's next miracle while it is already showing great potential and is on track to becoming Asia's new Economic Hub with high foreign reserves of approximately US$ 7 billion and a budget offering a more friendly investment climate. The Central Bank of Sri Lank (CBSL) announced on the 11th of April 2011 that, the fiscal deficit was reduced to 7.9% of GDP in 2010 from 9.9% in 2009 while the economic growth rate reached 8% (the second highest growth rate recorded ever recorded in Sri Lanka) in comparison to 3.5% in 2009. Meanwhile the Sri Lanka Tourism Development Authority has set a target of 2.5 million tourist arrivals by 2016, it indicated an unprecedented increase of 49.9% in foreign exchange earnings from tourism for the year ended 2010 as "the three major sectors of the economy namely, Agriculture, Industry, and Services registered significant growth as 7.0 percent, 8.4 percent and 8.0 percent respectively, in 2010 over the previous year." (Press Note, 2010) while GDP Per-Capita increased to US$2,399.
The remainder of this paper begins in the second chapter which explains the aims, objectives and the significance of my study, followed by chapter three which introduces the data set and the theoretical model which I will use. The fourth chapter applies the model to Sri Lanka's data and presents the results of my analysis while the synthesis, recommendations and concluding remarks are made in the final chapter four.
Chapter Two - Literature Review and The Study
The "war ruin" school of thought argues that civil wars mainly impose negative costs on an economy and retards economic growth. Imai and Weinstein (2000) states that civil wars will reduce the rate of capital stock growth as a consequence of capital flight and a reduction of private investment. Grobar (1993) found that Sri Lanka paid for its defence spending by hefty reductions in investments and cutbacks in non military government spending. Furthermore, the government is compelled to reallocate expenditure focused on economic growth enhancing activities to that of warfare. The nation's output was directly affected as a result of the complete "destruction of capital assets and property, damages to infrastructure and loss of cultivable land" (Arunatillake, 2001, Pg.2), in addition to the drop in investments and lost human capital due to death, disability and emigration. However, Organski and Kugler (1980) assert that civil wars have positive impacts on the economy in the form of improvements in efficiency, enhancement of human capital and positive spill over effects from technological innovation. Knight et al. (1996, pp. 9-11) while suggesting that reallocating idle resources towards expanding the military could in theory result in a "short run simulative Keynesian impact on the growth rate by inducing a rise in capital utilization" went on to argue that high levels of military spending would lead to negative impacts on the long run economic growth rate in three significant ways. Firstly it would "lower the expected after tax return on productive fixed capital, while simultaneously reducing the flow of savings that is available to finance productive fixed capital formation in the domestic economy". Secondly, "research and development activities may concentrate on military progress at the expense of technological advances in economically productive areas" and finally via "policies implemented to support military programs" such as trade restrictions and nationalization "which are detrimental to efficient resource allocation and market growth".
There are very few analysis available on the economic impact of the Sri Lankan war. The Richardson & Samarasinghe study (1991), Grobar & Gnanaselvam study (1993), Marga (1998) and the Arunatilake et al study (1999) are by far the most comprehensive pieces of work available and the focus of these studies have mainly been on the much broader topic of 'the economic cost of the war'. Detailed investigations have not been carried out into the impact the war had specifically on output which is a key macroeconomic variable. Bosworth and Collins (2003) and Miller (2004) have included Sri Lanka in cross country and regional assessments of sources of economic growth but these studies assume that the labour and capital shares are the same across countries, which is not an accurate assumption. I intend to overcome these issues and arrive at a more accurate end result. Finally, I intend to focus on the topic of total factor productivity (TFP), perhaps best explained as "a measure of our ignorance". (Griliches, 1994: Pg1) The measurement of which will require that I overcome a few empirical issues in terms of the measurement of inputs, outputs, weights for indices and a relevant concept of capital. TFP, capital intensity, increases in labour productivity and the labour force can be identified as drivers of GDP growth. The simplest explanation for TFP would be in the following form:
Which implies that,
Higher levels of TFP could be achieved through investment in human resources, technical progress and economic restructuring and if combined with higher capital intensity it would result in higher labour productivity which is a key driver behind economic growth.
Through this paper I will aim to achieve two objectives. "Wars result in a direct waste of output on the war effort, and an indirect loss of output through the destruction of human and physical capital." (Broadberry, Pg.26) Firstly I will concentrate specifically on analysing the impact of the war on output. For this purpose I will analyse the trends in output over the years from 1985 - 2009 using the growth accounting framework which is built on the foundations of the Cobb Douglas Production Function and the Solow growth model. I will begin by working out the TFP growth rates for the economy based on the current levels of output, labour and capital which will allow the comparison of the main sources of Sri Lanka's economic growth over the last 25 years. Murdoch and Sandler (2002) identified increases in military expenditure as a significant determinant of economic growth while Thompson (1974) went on to state that increases in military spending would enhance the enforcement of property rights and improve national security. Yet, the value GDP increases as a result of the increasing government expenditure on defence alone, whilst the military personnel who are included in the labour force do not add value to the economy in terms of production. Therefore I will go a step further and take out the effects of the defence sector on GDP while keeping the labour force constant which will enable me to see the impact the armed forces could have on output if they were engaged in more productive activities while paving the way for a better comparison on the impact of the war on output. Finally, I will endeavour to separate the effects of terrorism and weather on production as I seek to determine whether the War is a determinant of Total Factor Productivity.
Chapter Three - The Data and the Theoretical Model
The Data
The main source of data for this study was specially prepared by the Central Bank of Sri Lanka on a request made via the Governor of the Central Bank, Mr. Ajith Nivard Cabraal. It is fair to conclude that this study is conducted using the most accurate and up-to-date data available in Sri Lanka. Most of the data has been compiled using various issues of the Central Bank's Annual Reports which are available at www.cbsl.gov.lk. The secondary source for data was the Department of Census and Statistics in Sri Lanka which can be accessed via www.statistics.gov.lk .
Regardless of the source, I have been able to identify certain problems with the Data. Firstly, it is argued that defence expenditure data of most developing countries are unreliable (Sen & West, 1992), and in the case of Sri Lanka the defence spending excludes pensions and disability benefits of soldiers which suggests that the figure is underestimated. Secondly, data collection in the conflict zone (North & East Provinces) was adversely affected by the ongoing war and this has resulted in estimations and omissions in certain years. Finally the defence budget of the LTTE is not available for obvious reasons and it included the incomes of the people living in the North & Eastern Provinces and foreign contributions from a section of the diasporas sympathising with the LTTE, which could have otherwise been spent on output enhancing activities.
The Theoretical Model
"This paper takes Robert Solow seriously." (Mankiw, Romer and Weil, 1992, Pg.1) Robert Solow (1957) considered a neo-classical production function:
where, , and denote aggregate output, total factor productivity (TFP), capital stock and the size of the labour force. Solow "suggested that GDP growth can be decomposed into three shares: the capital, the labour and the TFP shares. The latter is computed as the residual of GDP growth once capital and labour contributions are removed" (Seetanah, 2011, Pg.369)
Solow's framework adopts the Cobb-Douglas production with Hicks-neutral technology which states that
(2.1)
This form of the production function enables the decomposition of the sources of economic growth in Sri Lanka. The model alone is built on certain crucial assumptions. Firstly, capital and labour are assumed to have positive and diminishing marginal products, thus limiting to a value between 0 and 1. Secondly which denotes returns to scale are assumed to be constant. For the Cobb-Douglas production function with competitive factor markets, the marginal product of inputs should equal its factor price. As such and where w and r are the wage rate for labour and the cost of capital. The assumption of constant returns to scale implies that.
The aggregate production function shown in (2.1) enables us to decompose growth rates into contributions from productivity, capital and labour. "Only capital and labour are actually observable in the data and productivity serves as a catch-all for anything else that is left unexplained by the other two factors." (Seetanah, 2011, Pg369). "The growth of total factor productivity is estimated as a residual from the production function (Statscan 13-568: 50-51). By rearranging equation (2.1) we can arrive at the following relationship for gross total factor productivity.
(2.2)
The value of could be approximated under the assumption of perfect competition amongst firms in the economy. Using the production function in (2.1) we could look at the profit maximization problem of a firm that rents capital at rate and hires labour at wage rate :
The marginal product of each factor must equals its price (first order condition), therefore:
(2.3)
(2.4)
We could now calculate the capital and labour shares for the economy as:
(2.5)
(2.6)
We could now calculate gross TFP values for any given year using the equation (2.2).
In order to analyse the impact the war had on potential output I will need to decompose total output into separate contributions from productivity, capital and labour. This can be done by taking logs of the production function in (2.1).
(2.7)
The growth rate of output can now be expressed as:
(2.8)
As such the growth rate of TFP can be expressed as:
(2.9)
Finally I use the Solow Growth Model. The first three components of this model are of interest to this paper. The first of which is the production function which I have already laid out in detail above.
The second component links Investment and Savings :
(2.10)
The third component describes the savings pattern. It is assumed that people save a constant fraction of their income:
(2.11)
Where is the savings rate, a constant number (0< <1).
The fourth and final component is the assumption that the capital stock depreciates at a constant rate , such that:
(2.12)
Construction of Variables and Discussion
Measures of Output
Output is measured by GDP which is defined as the market value of all final goods and services produced in a country, in a given year. Historically different measures of output have been used in literature and these include output per capita, output per employee and output per capital. However I will use real GDP growth rates as an output growth indicator. The data is available from 1950 onwards in the Annual Report of the CBSL at current market prices. Although real GDP growth rates have been given, I will have to use the following equation in order to compute the values for real GDP.
Even though GDP is regarded as an important economic indicator it is not the most efficient as it ignores many factors such as health, and standard of living which are mandatory for the well being of a society.
Measures of Labour
Labour is generally measured in a number of ways depending on the availability of data. I have chosen labour hours per capita as a measure of labour input. Data on the labour force is available via the Department of Census and Statistics of Sri Lanka, ranging from 1985 - 2009.
Measures of Capital
Sri Lanka's capital variable is not readily available and is therefore estimated using the equation (2.12). A study conducted by Vikram & Dhareshwar (1993) estimates Sri Lanka's capital stock for the year 1985 to be Rs. 419,810 million which I will use as the base figure for my calculations. Data on investment was made available in current prices from 1985 onwards by the CBSL. The depreciation rate I will use is estimated by Duma (2007) at 25% which he has justified as follows.
"This is a reasonable depreciation rate for Sri Lanka, taking into consideration the level of development of the economy and the many disruptions in capital formation in recent decades due to an unstable political environment, and the civil conflict. Also, this takes into account the depreciation rates allowed for tax purposes in Sri Lanka, ranging from 6â…“ percent for buildings to as high as 33â…“ percent for plant and machinery". (Duma, 2007, pp.8)
Finally I will adjust the figures for inflation and the study will be using the real values instead of the nominal values.
Labour and Capital Share for the Economy
Bernanke and Gurkaynak (2001) worked out Sri Lanka's labour share to be at 0.78, which I will use in my study.
This implies that which is the capital share of the economy.
Military Expenditure
This data was provided by the CBSL from 1985 - 2009 in current prices. I have adjusted these for inflation and will be using the real values in my study. "Since the conflict escalated in 1983, there has been a phenomenal build up of military spending by the Sri Lankan government." (Grobar, 1993, pp.3) This build up is expected to have slowed down Sri Lanka's economic growth due to the decreasing expenditure on investment, education and other productive resources which in turn reduce the productive efficiency of the economy.
