The International Monetary Fund in Sri Lanka: a critical dialogue

 

By Dr. MUTTUKRISHNA SARVANANTHAN*

In the year 2000 the LTTE overan the strategic Elephant Pass (a narrow strip of land area linking the densely populated Jaffna peninsula (Sri Lanka Army controlled at that time) and the sparsely populated Vanni mainland (mostly controlled by the LTTE at that time) of the Northern Province of Sri Lanka. In fact, the LTTE forced its way up to the Navatkuli bridge in order to regain its control of most of the peninsula, which it held between 1991 and 1995. Although the Sri Lanka Army was successful in halting the LTTE's forcible advance at the Navatkuli bridge and thereby saving the peninsula from falling into the hands of the LTTE yet again, the then Sri Lankan government went on a massive purchase of armaments, particularly from Eastern Europe (e.g. from Czech republic and Ukraine) beginning in later half of 2000, which strained the Sri Lankan economy so much so that the government budget for 2001 was postponed from being presented to the parliament in the usual month of November (2000) to February 2021. The foregoing delay was also due to the parliamentary elections held in December 2000 which the then incumbent government won narrowly. Massive armament purchases strained Sri Lanka's balance-of-payments in 2000, in spite of the fact that most of such purchases were on deferred payment running into the next few years. The economic crisis became critical after the successful attack by the LTTE on the then one and only international airport in July 2021. It was under the foregoing very trying circumstances, Sri Lanka sought IMF assistance in the latter half of 2000 (one of the 16 occations to date). Although this research note was published 20 years ago, it has contemporary relevance, precisely because even 13 years after the successful (from the viewpoint of the Sri Lankan government at least) end of the civil war, the annual defence budget (after public debt repayments) is the single largest ministerial allocation in the national budget to date. This is because, even in the absence of organised military threat to the Sri Lankan state, the armed forces are reported to maintain a combined strength of between 300,000 to 350,000 personnel, which is huge for a population of 22 million. Even the United Kingdom, with a population of around 65 million, is reported to have less than 200,000 armed forces personnel. 

Many Sri Lankans have had strong aversion or opposition to seeking assistance from the IMF ever since Sri Lanka became a member of the IMF. Even today, many Sri Lankans, especially leftists and nationalists, do harbour very serious reservations about seeking IMF assistance mainly on the perceived fear that any IMF assistance would be offered only on the condition of significant cuts on government spending on education, health, and social security provisions to the needy.

An important distinction should be made here. It is very true that any IMF assistance in the near future would be conditional upon cutting down budget deficits (ideally below 5% of the GDP) that was in double-digits in 2020 & 2021, and likley to be in 2022 as well. The IMF does not specify in which particular budget/s or sector/s  such cuts should be. It is absolutely the prerogrative of the incumbent government to make that strategic choice/s. Such cuts in goverment spending could and should come from down-sizing/right-sizing Sri Lanka's defence sector and public administration, cutting-down on wasteful public consumption expenditures (including massive corruption by politicians and senior-level public servants), divestiture of perenially loss-making state-owned enterprises (e.g. the national carrier Sri Lankan Airlines, Ceylon Petroleum Corporation, Ceylon Electricity Board, Litro Gas), etc. This is where this 👇 research note is of very high relevance even today for Sri Lanka which has formally requested for help from the IMF.

   - Editor in chief                                                           Jafffnafashion.com

On 20 April 2001, the International Monetary Fund (IMF) approved a US$253 million stand-by credit facility to Sri Lanka to stabilise the country's macroeconomic fundamentals. The government should be commended for requesting the IMF to make the stand-by arrangement a public document, facilitating public scrutiny. However, there are some critical discrepancies in the statistical data presented in the IMF Country Report No. 01/71 compared with annual reports of the Central Bank of Sri Lanka. These anomalies are highlighted with examples regarding defence expenditure, poverty reduction programmes, and public service recruitment. 

The International Monetary Fund (IMF) approved a US$253 million stand-by credit facility to Sri Lanka on 20 April 2001, to stabilise the country's macroeconomic fundamentals. An initial instalment of US$131 million has already been released, and the rest is expected to be released in four equal instalments (US$30.5 million each) depending on the performance of the economy. In addition to this, another US$250 million may be provided under the Poverty Reduction and Growth Facility, itself the successor to the Enhanced Structural Adjustment Facility.

The government of Sri Lanka (GOSL) should be commended for requesting the IMF to make the stand-by arrangement a public document, hence facilitating public scrutiny. This is the first time the full text of an IMF—GOSL agreement has been made available to the general public, and is a very welcome step on the part of both parties towards transparency and good governance. However, it is disappointing and disturbing to note that there are some critical discrepancies in the statistical data, inter alia, presented in the IMF Country Report No. 01/71 (citing 'Sri Lankan Authorities') when compared with the Annual Reports of the Central Bank of Sri Lanka (CBSL).

Table 1. Defence expenditure as a percentage of GDP in Sri Lanka, 1996—2000

Fiscal year GDP at current prices (LKR billion) Defence expenditure (LKR billion) Defence expenditure as a percentage of GDP

Sources: Central Bank of Sri Lanka, Annual Report 1996. GDP, Statistical Appendix Table 1 ; Defence expenditure, Statistical Appendix Table 51. Central Bank of Sri Lanka, Annual Report 1997. GDP, Statistical Appendix Table 1; Defence expenditure; Statistical Appendix Table 51. Central Bank of Sri Lanka, Annual Report 1998. GDP, Statistical Appendix Table 1; Defence expenditure, Statistical Appendix Table 54. Central Bank of Sri Lanka, Annual Report 1999. GDP, Statistical Appendix Table 1; Defence expenditure, Statistical Appendix Table 55. Central Bank of Sri Lanka, Annual Report 2000. GDP, Statistical Appendix Table 1; Defence expenditure, Statistical Appendix Table 55.

Defence expenditure discrepancies

One striking instance of discrepancies in the statistical data pertains to the defence expenditure of Sri Lanka. The gross domestic product at current factor cost prices (nominal GDP) and the actual defence expenditure (recurrent plus capital) incurred in the past 5 years are presented in Table 1. Accordingly, the defence expenditure as a percentage of nominal GDP peaked at 6.8% in 2000. The figures for defence expenditure are for the Ministry of Defence, which includes the army, air force, navy, police, immigration and emigration, and registration of persons departments. Although normally the expenditure on police is for the maintenance of law and order, in the context of the civil war in Sri Lanka, a significant part of the police service is devoted to national security duties as well (e.g. the deployment of a police special task force in the eastern province). Therefore, the expenditure on police (law and order), immigration and emigration, and registration of persons departments should be considerably less than 1% of the GDP. According to the IMF, however, it has been 1% or a little more during the past 5 years. I (The IMF has provided a breakdown of the defence expenditure into 'wages and salaries' and 'goods and services', which is a welcome step because the CBSL does not publish this breakdown for public consumption).2

Other discrepancies abound. According to the CBSL, the nominal GDP (in Sri Lanka Rupees) of Sri Lanka in 1999 and 2000 was LKR995 billion and LKRI 125 billion, respectively (see Table 1). Yet, according to the IMF, the corresponding figures were LKRI 111 billion and LKR1263 billion.3 Similarly, whereas the CBSL shows defence expenditures in 1999 and 2000 as LKR54 billion and LKR77 billion, respectively (see Table 1), the corresponding figures

Table 2. Defence expenditure as a percentage of GDP reported by CBSL and IMF

Sources: Central Bank of Sri Lanka, Annual Report, various years.

International Monetary Fund, Sri Lanka Country Report No. 01/71 (Internet edition), May 2001, p 13.

(for 'security-related expenditure') for the IMF were LKR49 billion and LKR71 billion.4

Further discrepancies may be seen in Table 2. According to the CBSL, in the past 5 years defence expenditure as a percentage of nominal GDP ranged from 5.4% in 1999 (lowest) to 6.8% in 2000 (highest), whereas according to the IMF it ranged from 4.4% in 1999 (lowest) to 5.8% in 1996 (highest). Thus, defence expenditure as a percentage of nominal GDP, according to the IMF, was between 11% and 19% less than what was reported by the CBSL in the past 5 years. This discrepancy reached nearly 20% in the last 3 years!

The IMF data on military expenditure are considerable under-estimates. (The IMF and CBSL figures do not include payments to disabled soldiers and the pensions of retired soldiers, which is a further source of the under-estimate of defence expenditure in Sri Lanka.)5 The reason for these considerable underestimates is unknown, to the best of our knowledge. Whether it is an oversight, a typing error or an attempt to dampen the burden of defence expenditure in the public eye is anyone's guess.

Moreover, defence expenditure as disclosed by the CBSL is only part of the total defence expenditure incurred by the GOSL. In addition to this apparent defence expenditure, there are considerable amounts of camouflaged defence expenditure that, by their very nature, are shrouded in mystery. These camouflaged defence expenditures are mostly expenditures in kind incurred in the course of civil war. For example, aid deflection seems to be a major contribution to these camouflaged expenditures. According to anecdotal evidence, heavy vehicles, machinery and equipment of donor-funded economic infrastructure projects such as highways, ports (air and sea), power and energy, and telecommunications development are occasionally diverted for the use of security forces during major military operations. In this light, the combination of power and energy with the defence ministry may not be a coincidence. Although these aid deflections may be temporary, it would undoubtedly delay infrastructure projects, which would cost the country dearly in terms of higher construction costs, greater interest payments, etc.

One point of potential evidence for aid deflection is the very low utilisation rate of foreign aid commitments in recent years. According to the External Resources Department of the Ministry of Finance and Planning, between 1997 and 1999 the utilisation rate of the Asian Development Bank (ADB) loans was only 21%, that of the International Development Association only 18%, and that of the Japan Bank for International Co-operation a mere 10%.6 It is important to note that the ADB is the single largest multilateral donor, and Japan is the single largest bilateral donor to Sri Lanka in recent times. Besides, the ADB and Japan are the greatest contributors to economic infrastructure development projects in Sri Lanka. Hence there may be a likelihood of negative correlation between aid deflection and the rate of aid utilisation of these multilateral and bilateral donors. This issue is of critical importance for foreign aid policy in Sri Lanka and requires more in-depth study.

Furthermore, some of the expenditures incurred for the defence sector may be classified under a different ministry. For example, health services provided for the armed forces may be accounted for partly in the defence ministry budget and partly in the health ministry budget. Supposing injured armed forces personnel are treated in a military hospital, the expenditures incurred may be accounted for in the defence ministry budget, whereas if they are treated in a civilian hospital, the expenditures may be accounted for in the health ministry budget. Likewise, we may be able to identify a number of camouflaged expenditures incurred in the war efforts that are accounted for in the non-defence sector. This would result in quite a serious under-estimation of military expenditure.

Yet another example of camouflaged military expenditure is that part of the expenditures incurred for the upkeep of the former Tamil rebel groups (Eelam People's Democratic Party, Eelam People's Revolutionary Liberation Front, Tamil Eelam Liberation Organization, etc.) by the government may be accounted for in the non-defence budget. This, again, would result in the under-estimation of total military expenditure. For instance, these paramilitary groups are provided with cash and perks in kind as reward for their allegiance to the government and state security forces. It is rumoured that loans at concessionary terms through the state banks are provided for the business enterprises of these paramilitary groups. Besides, some leading cadres of these groups are provided employment at different ministries. Strictly speaking, the cost of such concessionary loans, employment, and so on, should be borne by the defence ministry and not by other line ministries.

All the foregoing camouflaged military expenditures, inter alia, add to the formidable military budget of the GOSL. Therefore, the actual defence budget— both apparent and camouflaged—would be much greater than what the CBSL would want us to believe. To our understanding, the CBSL figures on defence expenditure are themselves considerable under-estimations. Hence, the IMF defence expenditure data are gross under-estimations of the actual total defence expenditure of Sri Lanka.

Defence expenditure in a comparative perspective

The IMF has also provided some comparative data on defence expenditure as a proportion of the GDP of selected countries, and points out that 'military expenditures in Sri Lanka, while high, is similar to that in many other countries'.7 We find it hard to stomach the IMF observation on Sri Lanka's defence expenditure as a proportion to the GDP compared with some other selected countries in the third world. First, the IMF has compared data pertaining to 1999 only, the year in which Sri Lanka's military expenditure as a proportion of GDP was the lowest during the period 1996—2000. Second, out of the seven countries with which the comparison is made, Egypt, Korea, and Nigeria have spent less (2.7%, 2.9% and 2.8%, respectively) while Cambodia and Ethiopia have spent marginally more (3.8% and 3.5%, respectively) than Sri Lanka (3.4%) on the armed forces8 as a proportion of the GDP in 1999. Only Pakistan and Turkey have spent considerably higher proportions (4.9% each) than Sri Lanka. Also, these seven countries with which the comparison is made are not similar to Sri Lanka in many respects. For example, Pakistan and Turkey have had a long history of military regimes where the army still plays a major role in governance; therefore, historically, their defence budgets have been high. Besides, Pakistan and Turkey have had major territorial disputes with neighbouring countries (namely India and Cyprus, respectively) for a very long time. Hence, the choice of countries for comparison with Sri Lanka by the IMF is very inappropriate. It would have been more appropriate to compare Sri Lanka with some other internal conflict-ridden countries.

In Table 3 we produce comparative data on military expenditure as a proportion of the GDP of selected South Asian countries, as well as internal war-torn countries in the third world from 1991 to 1999, produced by the Stockholm International Peace Research Institute (SIPRI). It is important to point out that SIPRI data for Sri Lanka's defence expenditure as a proportion of GDP is marginally higher than the IMF data but considerably lower than CBSL data, as presented in Tables 1 and 2, respectively.

According to Table 3, all the selected countries—except Pakistan and Myanmar—have consistently spent less on defence as a proportion of their respective GDP than Sri Lanka throughout the period under consideration (i.e. 1991—1999). Myanmar spent more from 1991 to 1994 but, since then, has consistently spent less than Sri Lanka. Pakistan has spent more during seven out of 9 years, the same in 1996 and less in 1995 than Sri Lanka. Thus, during 1995, Sri Lanka was the highest spender of GDP on defence among the selected South Asian and internal war-torn countries. An important fact to note is that both Myanmar and Pakistan have had a long history of military regimes and, even today, the army rules both countries. Even during intermittent times of democratic rule in Pakistan, the military plays a powerful role in governing the country and, hence, military expenditures in Pakistan have been historically higher than other democratic countries in the region. Due to this factor, Myanmar and Pakistan are not strictly comparable countries with Sri Lanka.

* Expenditures on paramilitary forces are excluded.

** Not available.

Source: Stockholm International Peace Research Institute, SIPRI Yearbook 2001, Chapter 4, Table 4X4.

On the contrary, notwithstanding intermittent military dictatorships in Bangladesh, Colombia, Philippines, Sierra Leone, Sudan, and Uganda, their respective defence expenditure as a proportion of GDP is significantly less than that of Sri Lanka. For instance, the defence expenditure as a proportion of GDP in Sri Lanka is more than double that of Bangladesh, significantly greater than that of Colombia (especially since 1994), almost three times that of the Philippines since 1995, more than double that of Sierra Leone since 1995, more than three times that of Sudan between 1995 and 1997, and more than double that of Uganda during most years under consideration in Table 3.

In addition to this comparative military expenditure data, we produce a set of military development data of the South Asian region to see how Sri Lanka fares. Table 4 reveals military development of the five largest South Asian countries: India, Pakistan, Bangladesh, Nepal, and Sri Lanka. Although the data are a bit dated, they give an indication of the level of military development in Sri Lanka in a comparative perspective over a long period of time. The defence expenditure of Sri Lanka in 1996 (US$700 million) was the third largest in absolute terms after India and Pakistan, despite it being the smallest country among the five in terms of physical and population size. Sri Lanka experienced the highest annual percentage increase of defence expenditure in South Asia during the period 1985—1996 at 11%; fully double the rate of the second-placed country, Nepal. Defence expenditure as a percentage of the GNP in 1995 was highest in Sri Lanka at 5.3% (just above 5.2% in Pakistan). Sri Lanka's defence expenditure per capita of US$37 in 1995 was the highest in the region, with Pakistan trailing far behind in second place at only US$21. The percentage increase of armed forces personnel during 1985—1996 was highest in Sri Lanka at 81%; nearly double that of Nepal, which experienced the second highest increase. At 926 in 1996 (in 1985, it was 100), the Military Holdings Index of Sri Lanka was the highest in the region, nearly five-fold that of Bangladesh, the second highest. These regional comparative data are present up to 1996 only. Since then, the development of defence sector has accelerated in Sri Lanka. Therefore, the gap between Sri Lanka and other South Asian countries would have widened in the past 5 years.

The data presented in Tables 3 and 4, respectively, contests the observation of the IMF regarding Sri Lanka's military expenditure vis-å-vis other comparable countries. This is sufficient evidence of the rapidly growing militarisation of the economy and society in Sri Lanka despite its being a democratic polity. It is ironic that, while the IMF is quite rightly alarmed by the current account deficit in the balance-of-payments reaching almost 7% of the GDP, it does not seem to be overly concerned about the defence expenditure reaching almost 7% of the GDP in 2000. This apathy perhaps demonstrates where the priority of the IMF lies.

We are aware of a decision of the Executive Board of the IMF not to take military expenditures into account in evaluating the performance of their lending and other conditions related to IMF-supported structural adjustment programmes.9 The IMF is entitled to decide on their policy on defence expenditures, although we may disagree with that decision. Nonetheless, considerable under-estimates in the military expenditure data provided by the Sri Lankan authorities to the IMF, and the acceptance of such data by the IMF, is a very serious concern to citizens of Sri Lanka, let alone to the credibility of the IMF. The lack of transparency in the huge military budget of Sri Lanka is also of critical concern. We appreciate that military procurements cannot be entirely disclosed for security reasons. However, the lack of a proper and open tendering procedure for military procurements (as in other public sector procurements) is a crucial drawback in the management of public finances in Sri Lanka.

The GOSL has given an assurance to the IMF that defence expenditures for the current fiscal year (2001) would be strictly limited to LKR63 billion as allocated in the budget, and that large military procurements would require the approval of the finance secretary. Imports by the defence ministry are expected to be cut back from about US$400 million in 2000 to about US$140 million this year. 10 How successfully the government can keep up this commitment depends on the military situation on the ground. In the past several years, defence expenditures have significantly overshot budgetary outlays. ll Even the present budgetary allocation of LKR63 billion for defence is very high, accounting for 19% of the total budget for 2001. If the Liberation Tigers of Tamil Eelam decide to return to the offensive, then the GOSL will be forced to respond. Consequently, the commitments made by the government on restraining military expenditures may not hold. In such a scenario, the GOSL has vowed to increase taxes (possibly including the rate of Goods and Services Tax), and to cut expenditure on goods and services and domestically-financed capital spending commensurate to the potential increase in defence expenditure, 12 all measures that may negatively impact on businesses and the general public alike.

Poverty reduction

Another key statistical anomaly in the IMF country report is the data pertaining to the poor in Sri Lanka. The IMF claims that the number of poor in Sri Lanka is 20% of the total population.13 This is a gross under-estimation. Applying the official poverty line, the GOSL poverty reduction framework revealed that almost one-third of the population is poor, while fully 50% of the people receive cash payments under the Samurdhi poverty alleviation programme.

The IMF also claims that Sri Lankan authorities have given an undertaking that Samurdhi will be reformed and better targeted. 14 Yet, if the GOSL is serious about reforming the Samurdhi programme by cutting down wastage and better targeting, it logically follows that the total expenditure on Samurdhi should decline. On the contrary, according to the appropriations bill presented to the parliament, the budgetary allocation for Samurdhi in 2001 has increased to almost LKRII billion (almost 1% of the GDP). The budgetary allocation for Samurdhi as a proportion of total budget of the government already grew from 2% in 2000 to 3% in 2001 (a 50% rise). Hence, the government's undertaking to reform Samurdhi seems unreliable.

The Samurdhi poverty alleviation programme is a big political enterprise employing about 30,000 persons. The vast majority are supporters of constituent parties in the ruling coalition, and the employees of the programme are a reserve army of political workers. Earlier poverty alleviation programmes, such as the Janasaviya, were the same. In Sri Lanka, national poverty alleviation programmes have always been heavily politicised. Therefore, the most crucial decentralisation-cum-privatisation programme Sri Lanka should undertake with the utmost urgency is that of the national poverty alleviation programme. The Samurdhi programme cannot be reformed or restructured by the state; it can only be resurrected by decentralisation-cum-privatisation. In Sri Lanka, social safety net mechanisms are largely political safety net mechanisms. We are told that, in many instances, the poor have to bribe Samurdhi officers to get access to benefits under the programme. In this context, we beg to differ with the IMF mission's view that 'if the above programs were implemented efficiently, they would significantly improve targeting of the poor and the vulnerable and provide protection for living cost increases' . 15 To us, the targeting of the poor is so wide of the mark, not because of a lack of knowledge or inefficiency; on the contrary, it is a deliberate political act.

The nature of poverty and the poor differ from village to village, town to town, and village to town; therefore, a top-down national poverty alleviation programme based on a single official poverty line cannot address different varieties of poverty and deprivation. An independent Poverty Alleviation Commission (PAC) with constitutional provisions should be set up and the government in power should allocate a sum of money annually to it. Professionals and experts on poverty should head this commission who, in turn, should be democratically selected by concerned citizens and not appointed by any government. Each village/town should identify the nature of poverty and the poor in their village/town. Any local government (village/urban council), semi-government, non-governmental, or private organisation should be entitled to bid for money from the proposed PAC for poverty alleviation programmes in their respective village/town. The commission should evaluate the proposals from different bidders on their merits, with absolutely no political interference. Perhaps such a fundamental change in the way in which the perennial issue of poverty is addressed would make a real impact on poverty in Sri Lanka. Having said that, we are doubtful whether there is sufficient political will—within the ruling or opposition political parties—for such a fundamental overhaul of the national poverty alleviation programmes in Sri Lanka.

The IMF is quite correctly concerned about the fallout of the present stabilisation measures that would adversely affect the poorest segments of the population. However, it seems to be satisfied with the assurances of the Sri Lankan authorities that the existing social safety net mechanisms such as the Samurdhi programme would take care of such a situation. Given the performance of Samurdhi in the past 6 years, it would be a serious gamble to rely on that for providing effective social protection for the marginalised groups in rural, urban, and war-torn areas.

Public service recruitment

The GOSL has pronounced several policies and given a number of undertakings to the IMF to get the stand-by credit facility approved by the latter's Executive Board. However, some of these are too ambitious for the present coalition government with not even a simple parliamentary majority, which depends on three minor political parties (Ceylon Workers Congress, Eelam People's Democratic Party, and National Unity Alliance) for its survival. Already, there are several instances of faltering on public policy pronouncements and undertakings given to the IMF, one of which is now outlined.

Public sector reforms are one of the cornerstones of the IMF—GOSL stand-by arrangement. Accordingly, the government has given a firm undertaking to the IMF that there will be a moratorium on recruitment to the public service during 2001. Nonetheless, recruitment to the armed forces continues unabated. In April and May, the air force and navy placed advertisements for recruits, which continues to date. Furthermore, the police special task force launched a recruitment drive during late-June and early-July 2001, 16 while the defence ministry launched a recruitment drive for the army on 1 July 2001 to recruit 10,000 additional soldiers. 17 This upsurge in recruitment to the armed forces flies in the face of government's commitment to a moratorium on public service hiring.

Conclusion

The country analyses by the IMF require structural adjustment due to deficiencies in key statistical data. The political will of the GOSL to undertake far-reaching reforms of the economy as enunciated in the IMF country report is suspect as per early indications indicated in the present paper. At the same time, it is also quite unrealistic for the IMF to expect a coalition government without a parliamentary majority to fulfil such a broad reform agenda. Nonetheless, the stand-by credit facility by the IMF has provided a temporary respite to an ailing economy.

Courtesy: Contemporary South Asia (2002), 11(1), 77-87 Carfax Tay%r&FrancisPublishing

*The views expressed in this paper are solely those of the author, and not of The International Centre for Ethnic Studies.

Correspondence: Muttukrishna Sarvananthan, The International Centre for Ethnic Studies, 2 Kynsey Terrace, Colombo 08, Sri Lanka. E-mail: sarvi@slt.lk

ISSN 09584935 print; 1469-364X online/02/010077-11 @ 2002 Taylor & Francis Ltd

DOI: 10.1080/0958493022000000378

Notes and references

I. International Monetary Fund, Sri Lanka Country Report No. 01/71 (internet edition), May 2001, p 13.

2. Ibid.

3. International Monetary Fund, op cit, Ref 2, p 33. 4. Ibid.

5. International Monetary Fund, op cit, Ref 2, p 13.

6. Foreign Aid Review: Sri Lanka (Government of Sri Lanka, Ministry of Finance and Planning External Resources Department, Colombo, 1999), p 3.

7. International Monetary Fund, op cit, Ref 2, p 13.

8. Military expenditure here includes only the army, air force and navy.

9. BUFF/91/186, cited in International Monetary Fund, op cit, Ref 2, p 13.

10. International Monetary Fund, op cit, Ref 2, p 19.

11. Annual Report (Central Bank of Sri Lanka, Colombo, various years).

12. International Monetary Fund, op cit, Ref 2, p 15.

13. International Monetary Fund, op cit, Ref 2, p 24.

14. International Monetary Fund, op cit, Ref 2, pp 17, 24.

15. International Monetary Fund, op cit, Ref 2, p 17.

16. Ceylon Daily News, 13 June 2001, p 3.

17. Virakesari, 13 June 2001, p 1.