Abstract
This Selected Issues paper on Sri Lanka underlies the dynamics of growth and external competitiveness. The slowdown in the contribution of sectors that are labor intensive, together with faster growth in sectors that are capital intensive and have higher productivity levels, resulted in total factor productivity (TFP) as the main contributor to growth. Sri Lanka’s strong growth performance has brought positive benefits to the economy and has benefited from a high quality labor force. The labor productivity is low by regional standards and the internal terms of trade are skewed toward the nontraded sector.
II. Is Sri Lanka’s External Competitiveness Of Concern?1
A. Introduction
1. Recent appreciation of the real effective exchange rate (REER) together with a series of exogenous shocks to the external sector, has sparked concerns regarding Sri Lanka’s external competitiveness.2 Between 2005-2006, the CPI-based REER appreciated by 14 percent following a trend depreciation of 23 percent over 1998-2004. High domestic inflation and reduced nominal exchange rate flexibility accounted, to a large extent, for the recent appreciation in the REER. Over the same period Sri Lanka experienced a series of external and domestic shocks including the December 2004 tsunami, the expiration of the Multi-Fiber Agreement in end-2005, increases in global oil prices, and a resumption of the internal ethnic conflict in 2006. The implications of these developments for Sri Lanka’s external competitiveness are assessed in this paper.
Sri Lanka: Real Effective Exchange Rate
(CPI based, 2000=100)
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
2. This paper examines a broad set of indicators for the Sri Lankan economy to assess the implications of these developments for Sri Lanka’s external competitiveness. Trends in export performance, relative prices and indicators of structural competitiveness are examined. In addition, the analysis estimates the equilibrium real effective exchange rate (EREER) using several methodologies in order to determine whether the current level of the REER is appropriate. We proceed as follows: Section II discusses the performance in the tradable sector since 1980, followed by an assessment of production cost competitiveness in Section III. Estimates of the EREER are presented in Section IV, and indicators of structural competitiveness detailed in Section V. Finally Section VI concludes. Tradable Sector Performance
B. Tradable Sector Performance
Current Account Balance and External Sector Trends
3. Since 2001, Sri Lanka’s non-oil current account balance has been in surplus, improving from a deficit of 1.3 percent of GDP between 1990-2001 to a surplus of 3.5 percent of GDP over 2002-2006 (Figure II.2). Over this period the total current account balance also improved by an average of 2 percent of GDP, in spite of rising oil prices in 2005 and 2006. Two factors underlie these trends—strong growth in remittance inflows which have increased from 4.6 percent of GDP in 1990 to 7.7 percent of GDP in 2006; and improvements in the non-oil trade balance from a deficit of 6.5 percent of GDP over the 1990-2001 to 4.2 percent of GDP over 2002-2006. The latter has been driven by buoyant export growth and reduced non-oil import demand, following a collapse in both components during 1998-2001, a period of intense civil conflict. Taken together, the overall trade balance has remained broadly comparable at a deficit of 10 percent of GDP between the two periods. In the first half of 2007, strong apparel export growth strengthened the trade balance to 10.5 percent of GDP. Nevertheless, the current account deficit for 2007 as a whole is projected to remain high at 5 percent of GDP on account of high oil prices and mega infrastructure imports.
External Sector Developments
(Percent of GDP)
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
Sources: Central Bank of Sri Lanka (CBSL) and IMF staff estimates1/ Reserve coverage in months of imports of goods and services.4. Underlying external vulnerabilities exist and reserve coverage has remained below 3 months of imports (GNS) since 1999. Increases in oil prices, high public debt and external debt service, together with low levels of foreign direct investment, have contributed to Sri Lanka’s underlying external vulnerability. Between 2005-2006, oil imports increased by about 1.5 percent of GDP ($400 million) to over $2 billion. Over the same period, the government’s reliance on fairly short-term dollar-denominated domestic commercial borrowing added to balance of payments pressures and exacerbated external risks. By end-2006 reserve coverage was at 2.4 months of imports (82 percent of short-term debt), and is expected to remain at this level in 2007.
5. Sri Lanka’s external debt stock at about 53 percent of GDP in 2006, combined with a persistently high public domestic debt stock of 93 percent, is also of concern. Being largely concessional, external debt, with NPV rates of 35 percent of GDP, is below the debt-distress level of 40 percent of GDP applicable to Sri Lanka, leaving limited room for maneuver. Moreover, although external short-term debt is a relatively small share of total external debt, increased domestic dollar-denominated borrowing from commercial sources at short-medium maturities, has increased the underlying risks to reserves. At end-2006, the stock of such dollar-denominated domestic borrowing stood at $1.6 billion, about 63 percent of reserves.
Export Sector Performance
6. Between 1990–2006 Sri Lanka rapidly transformed from an agricultural to a manufactured goods exporter. Apparel and textiles, in particular, grew to dominate the export sector and comprised 45 percent of all goods exports in 2006. Nevertheless, despite a steady decline over the 1990s, agricultural exports (including tea, coconut products, and spices) remain an important component of Sri Lanka’s export base comprising about 20 percent of total exports in 2006. Shares of rubber exports and gem processing are also increasingly steadily.
Sri Lanka: Share in Total Exports, 2006
(In percent)
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
Note: The size of each balloon represents the percentage change in the category’s share in total exports.7. Export value growth boomed in the early 1990s, and since 2002 has gradually recovered from its sharp decline over 1998–2001. Export volume growth has followed a similar path, average real growth rates being about 5 percent (year-on-year) over 2002-2006 compared with 7 percent in the 1990s. The impact of the expiration in early 2005 of the Multi-Fiber Agreement is unclear although there has been a decline in real export growth rates from 7.7 percent in 2004 to 4 percent in 2006, driven mainly by the apparels sector. Data for the first half of 2007, however, indicate a rebound in both total and apparels exports. Nevertheless, the contribution to GDP of exports of goods and services at 2 percent of GDP is low, both from a historical and regional perspective.
Sri Lanka: Export Value Growth
(In percent)
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
8. Sri Lanka’s share of world exports which compares favorably with other South Asian economies has been on a steady decline since 2001 (Table II.1). Increased competition in the apparels sector and limited preferential access to the U.S. market has implied that Sri Lanka’s world market share (controlling for the effect of China and India) has not kept in line with GDP growth. This is particularly worrisome as world market shares of Sri Lanka’s regional competitors, for example that of Bangladesh, Vietnam, and Cambodia, have been increasing over this period.
Share of World Exports
(Excluding India and China)
Share of World Exports
(Excluding India and China)
| 1990-1997 | 1998-2001 | 2002-2006 | |
|---|---|---|---|
| Sri Lanka | 0.074 | 0.087 | 0.071 |
| Pakistan | 0.179 | 0.155 | 0.163 |
| Bangladesh | 0.060 | 0.087 | 0.095 |
| Cambodia | 0.006 | 0.019 | 0.024 |
| Vietnam | 0.105 | 0.222 | 0.316 |
Share of World Exports
(Excluding India and China)
| 1990-1997 | 1998-2001 | 2002-2006 | |
|---|---|---|---|
| Sri Lanka | 0.074 | 0.087 | 0.071 |
| Pakistan | 0.179 | 0.155 | 0.163 |
| Bangladesh | 0.060 | 0.087 | 0.095 |
| Cambodia | 0.006 | 0.019 | 0.024 |
| Vietnam | 0.105 | 0.222 | 0.316 |
9. These trends are reflected in the shift in exports since 2001 away from the United States and towards the European Union, both major destinations for Sri Lanka’s exports (Figure II.5). In particular, Sri Lanka’s share of the U.S. apparel and textiles market declined from 2.25 percent in 2001 to 2 percent in 2006. Over the same period, U.S. market shares of other garments exporters have been steadily rising—for example, Bangladesh saw an increase in her U.S. market share from 3.1 percent in 2001 to 3.5 percent in 2006. Market structure and limits to economies of scale in production and distribution may be important factors underlying this development. More encouraging is the growth of Sri Lanka’s E.U. apparel and textiles market share from 0.62 percent in 2001 to 0.73 percent in 2006 as Sri Lanka increased its utilization of the E.U.’s Generalized System of Preferences (GSP II).3 This is in spite of the competitive market and strong growth rates of market shares for other garments exporters to the European Union.
Sri Lanka: Direction of Trade
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
10. Going forward, Sri Lanka is expected to further increase its garments exports to the European Union as it improves backward interlinkages to the textiles sector and takes advantage of changes in the GSP II’s rules of origin. This preferential agreement has been expanded to include some of the ASEAN countries and is expected to be renewed to 2015 when it expires in end-2008. As further gains in the low-end apparels market share become increasingly difficult, Sri Lanka’s apparel industry is attempting to upgrade to higher-value items such as lingerie and knitwear. It has also launched an “ethical garments” marketing campaign in the United States to emphasize the high labor standards in place in the industry and the minimal use of child labor.
C. Production Cost and Relative Price Analysis
11. Labor productivity growth, including in the apparel industry, has been historically low in comparison to regional competitors. The apparel sector’s labor productivity growth averaged at 2 percent over 2002–2006, compared with sustained higher rates in India (about 3.5 percent total labor productivity growth) and Bangladesh (about 14 percent growth for the cotton and apparel sector).
Labor Productivity and Real Wage Growth
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
Source: Central Bank of Sri Lanka; and the International Labor Organization.12. Overall, Sri Lanka’s labor costs appear to be in the medium range of competitors. The available data on wages suggest that while real wage growth in manufacturing has not kept in line with productivity growth, there has been a marked slowdown since 2002.4 Moreover, according to the World Bank’s Doing Business Survey, in 2006 Sri Lanka’s nonwage labor costs (aside from firing costs which are unusually high) were one of the lowest in South Asia and compared favorably with those in the East Asian and Pacific countries (Table II.2).
Regional Nonwage Labor Costs
Regional Nonwage Labor Costs
| Employing Workers (2006) | Sri Lanka | Bangladesh | Cambodia | India | Pakistan | Vietnam | South Asia | East Asia and Pacific |
|---|---|---|---|---|---|---|---|---|
| Difficulty of Hiring Index | 0 | 11 | 56 | 33 | 78 | 0 | 42 | 24 |
| Rigidity of Hours Index | 20 | 40 | 60 | 20 | 20 | 40 | 25 | 25 |
| Difficulty of Firing Index | 60 | 40 | 30 | 70 | 30 | 70 | 38 | 20 |
| Rigidity of Employment Index | 27 | 30 | 49 | 41 | 43 | 37 | 35 | 23 |
| Nonwage labor cost (% of salary) | 15 | 0 | 0 | 16.8 | 12 | 17 | 7 | 9 |
| Firing costs (weeks of wages) | 178 | 51 | 39 | 56 | 90 | 87 | 72 | 42 |
Regional Nonwage Labor Costs
| Employing Workers (2006) | Sri Lanka | Bangladesh | Cambodia | India | Pakistan | Vietnam | South Asia | East Asia and Pacific |
|---|---|---|---|---|---|---|---|---|
| Difficulty of Hiring Index | 0 | 11 | 56 | 33 | 78 | 0 | 42 | 24 |
| Rigidity of Hours Index | 20 | 40 | 60 | 20 | 20 | 40 | 25 | 25 |
| Difficulty of Firing Index | 60 | 40 | 30 | 70 | 30 | 70 | 38 | 20 |
| Rigidity of Employment Index | 27 | 30 | 49 | 41 | 43 | 37 | 35 | 23 |
| Nonwage labor cost (% of salary) | 15 | 0 | 0 | 16.8 | 12 | 17 | 7 | 9 |
| Firing costs (weeks of wages) | 178 | 51 | 39 | 56 | 90 | 87 | 72 | 42 |
13. These developments, together with increased competition in the apparel sector, have reduced export profit margins although profit ratios remain high. Estimates of export profitability, using the ratio of export prices to manufacturing wages as a proxy for profitability, shows a trend decline of about 5 percent from 2003 to 2006. Analysis of profit ratios in the textile and apparel sector, calculated from estimates of the cost and value of total production, also reflects a slight decline from 2003 but points to profit margins of about 13 percent for the sector and 15 percent for industry as a whole. This suggests that while profit margins may be squeezed, there is still a fairly large margin for exporters to work within.
Export Sector Profitability
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
14. Nevertheless, indicators of the internal real exchange rate suggest a steady loss of competitiveness of the traded goods sector.5 Figure II.8 suggests that the prices received by firms operating in the nontraded goods sector have increased relative to the traded sector, implying that the relative returns to investing in the traded goods sector have declined. This is expected given the declining trends in export profitability, the higher productivity growth in the nontraded/services sector, and the increasing trend of remittance inflows which are more likely to be spent on domestic goods and services. Estimates of the REER based on unit-labor costs6 in manufacturing corroborate this, showing a trend appreciation from 1998 (Figure II.9). The difference in the trend between the CPI-based REER and ULC-based REER may be due to the weight of administered prices in the CPI and also as ULC measures omit nonlabor costs (e.g. energy costs and the effective cost of funds) facing producers.
Internal Terms of Trade
(Relative price of nontraded to traded goods)
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
Real Effective Exchange Rates
(ILO Unit Labor Cost Based)
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
D. Estimates of the Equilibrium Exchange Rate
15. To fully address the question of whether Sri Lanka’s external competitiveness is of concern, an assessment of the exchange rate level is needed. This should ideally be based on an equilibrium level, such as is consistent with internal and external balance or a sustainable level of the current account. This section estimates the equilibrium real exchange rate (EREER) using several methodologies developed in the literature. In addition to the fact that estimates of the EREER tend to be sensitive to the methodology used, several caveats should be borne in mind: first, for the case of Sri Lanka, whose economy has been affected several times by decades of conflict and other exogenous shocks, whether an “equilibrium” such that internal and external balance are achieved has ever been reached at any point in the recent past may be debatable. Second, given this history, the use of the past to inform analysis of the post-peace agreement period since 2002 may lead to estimation errors. Finally, data availability together with quality and frequency issues hamper the precise estimation of the theorized relationships. These factors imply a fairly wide confidence band over the estimated levels of the equilibrium exchange rate and on the assessment of the actual exchange rate level.
Behavioral Equilibrium Exchange Rate Approach (BEER)
16. The BEER approach estimates the EREER by econometrically identifying structural determinants of the long-term real exchange rate using a reduced-form exchange rate model.7 Economic fundamentals which have been viewed in the literature as having permanent effects on the long-run real exchange rate include the net foreign asset (NFA) position, terms of trade, trade openness, government consumption, and relative productivity differentials. Figure II.10 shows the evolution of these variables over 1980–2006 and their bivariate relationship with the REER:
The Real Effective Exchange Rate and its Long-Run Determinants, 1980–2006
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
Source: Central Bank of Sri Lanka; and IMF staff estimatesi. The NFA position of an economy can affect the real exchange rate in two ways. A higher NFA position implies a positive wealth effect which raises domestic demand for and the price of nontradables. Secondly, higher income from foreign assets improves the external current account. In equilibrium this would need an offsetting lower trade balance from a more appreciated real exchange rate. Sri Lanka’s NFA position appears to track developments in the REER up to 2001 after which the series diverge substantially, with NFA rising by 3 percent of GDP over 2001-2004 while the REER depreciated by 7 percent. This is also reflected in the correlation coefficient between the REER and NFA, which is positive prior to 2001 and high and negative after.
ii. An improvement in the terms of trade would generally tend to appreciate the real exchange rate if the wealth effect dominates the substitution effect. Higher export prices result in higher wages in both traded and nontraded sectors resulting in higher prices for nontradable and a real appreciation. The wealth effect associated with improved terms of trade will also raise demand for, and the price of, nontradables in the economy. On the other hand, higher prices for nontradables may shift demand towards imported goods, depreciating the real exchange rate. Sri Lanka’s terms of trade in goods and services is correlated positively with the REER up to 2004, suggesting that wealth and price effects dominated the substitution effect. The deterioration of the terms of trade due to high oil prices in 2005–2006 and the expected REER depreciation (in the absence of pass-through of oil prices onto domestic goods) may have therefore somewhat offset the appreciation due to tsunami inflows and high domestic inflation.
Table II.3.Correlations of Real Exchange Rate and its Fundamentals, 1980–2006
Source: Central Bank of Sri Lanka; and IMF staff estimatesTable II.3.Correlations of Real Exchange Rate and its Fundamentals, 1980–2006
Source: Central Bank of Sri Lanka; and IMF staff estimatesLog REER Openness Log Relative Productivity Log Terms of Trade Government Consumption % GDP NFA % GDP Log REER 1.00 Openness 0.17 1.00 Log Relative Productivity −0.34 −0.23 1.00 Log Terms of Trade 0.32 0.37 −0.39 1.00 Government Consumption % GDP −0.03 0.01 0.60 0.24 1.00 NFA % GDP 0.06 0.30 −0.10 0.83 0.35 1.00 Table II.3.Correlations of Real Exchange Rate and its Fundamentals, 1980–2006
Source: Central Bank of Sri Lanka; and IMF staff estimatesLog REER Openness Log Relative Productivity Log Terms of Trade Government Consumption % GDP NFA % GDP Log REER 1.00 Openness 0.17 1.00 Log Relative Productivity −0.34 −0.23 1.00 Log Terms of Trade 0.32 0.37 −0.39 1.00 Government Consumption % GDP −0.03 0.01 0.60 0.24 1.00 NFA % GDP 0.06 0.30 −0.10 0.83 0.35 1.00 iii. The effect of trade liberalization or openness also can operate via multiple channels. A more open trade regime generally is associated with a more depreciated real exchange rate when the removal of restrictions reduces the domestic price of tradables and discourages import substitution. On the other hand, if trade restrictions were primarily used to keep domestic prices of primary products low, then increasing trade openness may raise the overall price level and the real exchange rate. The bivariate relationship between trade openness and the REER in Sri Lanka is positive, fairly strong but volatile. As with the other variables there seems to be a divergence in 2004 where the relationship is negative. This may be due to endogeneity of the trade openness variable, measured by the share of exports and imports to total GDP, to the REER appreciation.
iv. Higher government consumption is associated with a real appreciation assuming a large share of government expenditure is on nontradables. For the case of Sri Lanka, this bivariate relationship is not strong and correlation is weakly negative. This could be because government spending is dominated by defense expenditure and interest payments, both of which would be associated with a worsening current account due to higher import spending and transfers.
v. Relative productivity differentials between traded to nontraded goods tend to appreciate the real exchange rate. As less developed countries converge to advanced economies, productivity improvements in the traded sector lead to increases in wages and prices in the nontraded sector and an real appreciation (the so-called Balassa-Samuelson Effect)8. Low productivity in the traded sector, compared to both the non-traded sector and other countries, has implied a trend decline in relative productivity since 1998 (Figure II.10).
17. A vector error correction framework (VECM) using the maximum likelihood estimator of Johansen is used to estimate the long-rum relationship between the REER and the structural determinants discussed above. Augmented Dickey Fuller unit root tests indicate that all the variables are nonstationary series in levels but stationary in first differences (I(1)). Johansen cointegration tests applied to several specifications including the endogenous variables indicate the presence of at least one cointegrating equation. This suggests there is a long-run equilibrium relationship among the variables and that the VECM framework could be reasonably applied. Nevertheless, the small sample size and the various episodes of intense conflict (controlled for using dummies), warrants some caution in the interpretation of the econometric results.
18. In line with theoretical priors, the VECM results indicate that improvements in the terms of trade and higher government consumption appreciate the exchange rate.9 Moreover, openness has tended to appreciate the exchange rate, perhaps as previously, trade barriers were kept prices artificially low. Somewhat counter intuitively, NFA as a percentage of GDP is negatively related to the REER, a finding which may be driven by the sensitivity of NFA to the conflict Sri Lanka has experienced and the quality of the data series itself.10 VECM estimation including relative productivity (which is available from 1990 onwards) was not possible due to limited data points. However, OLS estimates suggest there is a negative but insignificant association between the REER and relative productivity.
Actual REER and Estimated EREER
Citation: IMF Staff Country Reports 2007, 374; 10.5089/9781451823592.002.A002
19. The VECM estimates indicate a trend appreciation in the equilibrium REER of about 13 percent over 1990-2006. Hodrick-Prescott filters of the structural determinants of the REER were used to calculate the equilibrium REER for Sri Lanka for several models specifications. Trade openness, the terms of trade, NFA as a share of GDP and government consumption all exhibit upward trends in their long-run path which may have appreciated the EREER. These estimates suggest that the REER was largely overvalued during 1998-2001 and moderately undervalued over 2002–2004 during the peace accord.
Based on the estimated EREER from this approach, the end-2006 REER appears to be slightly overvalued in the range of 2–7percent.
The Macroeconomic Balance Approach
20. Another approach to estimating the equilibrium REER is the macroeconomic balance approach which focuses on the REER that simultaneously achieves internal and external balance. This involves (i) estimation of the underlying current account, which is the actual current account deficit adjusted for output gaps and lagged effects of past real exchange rate movements; (ii) estimation of the structural current account, which measures the medium-term domestic savings and investment relation; and (iii) calculation of the real exchange rate which would bring the underlying current account to the level of the structural current account. This methodology is useful in that it encompasses a medium-term view of the economy, taking into account the net external position of the country, particularly relevant for Sri Lanka given its high level of public debt. Nevertheless, as mentioned previously, the assumption of internal-external balance may be debatable given the number of shocks Sri Lanka has faced and given the prolonged civil conflict. Therefore, these caveats should be borne in mind when assessing the estimation results.
Current Account Estimates
Current Account Estimates
| Structural | Underlying | Actual | |
|---|---|---|---|
| 1980-85 | −0.6 | … | −11.7 |
| 1986-90 | −0.6 | −3.2 | −7.8 |
| 1992-97 | 1.0 | −3.8 | −5.2 |
| 1998-00 | 0.9 | −1.8 | −3.8 |
| 2002-04 | 0.8 | 0.2 | −1.7 |
| 2005-06 | 1.3 | −1.7 | −3.9 |
Current Account Estimates
| Structural | Underlying | Actual | |
|---|---|---|---|
| 1980-85 | −0.6 | … | −11.7 |
| 1986-90 | −0.6 | −3.2 | −7.8 |
| 1992-97 | 1.0 | −3.8 | −5.2 |
| 1998-00 | 0.9 | −1.8 | −3.8 |
| 2002-04 | 0.8 | 0.2 | −1.7 |
| 2005-06 | 1.3 | −1.7 | −3.9 |
21. We find that the five-year average structural current account surplus is about 1.3 percent for Sri Lanka, and the underlying current account to be about 1.7 percent deficit.11 This latter estimate takes into account the effect of the output gap, remittances, and trade responses to past REER changes. There is therefore a 3 percent adjustment needed in the current account to achieve internal and external balance, which could be achieved by a mixture of mobilizing remittance transfers and adjustments in the trade balance. If we assume this is completely borne by the adjustments in the trade balance, then based on the estimated elasticities, a REER depreciation of about 10-13 percent is needed. On the other hand, given Sri Lanka’s development path, a balanced structural current account may be more appropriate. This would require a 1.7 percent adjustment in the underlying current account and a REER depreciation of about 7-8 percent.
The macroeconomic balance approach suggests that the current REER may be moderately overvalued in the range of 7–13 percent.
E. Structural Competitiveness
22. Indicators of Sri Lanka’s structural competitiveness compare favorably with those in the region. Access to telecommunications is one of the highest in South Asia, second only to Vietnam; and over 90 percent of the road network is paved reflecting, in part, Sri Lanka’s small geographical area. Energy sector issues confound the region as a whole, particularly the high costs of electricity and its volatile distribution; in this regard Sri Lanka is not an outlier and the government’s planned reforms in the energy sector will go some way to ameliorate these barriers to competitiveness. Sri Lanka’s institutional framework and the quality of public sector management, measured by the World Bank’s CPIA, is considered to be moderate but ranks highly compared to its neighbors in South Asia. Moreover, the perception of corruption in Sri Lanka, as measured by Transparency International, is fairly low compared with the region.
23. Sri Lanka’s business climate indicators and legal framework are also fairly competitive (Table II.5). The number of procedures and cost of starting a business are one of the lowest in the region, and the duration (in number of days) is moderate. Similarly, enforcing contracts is cheaper and involves fewer procedures in Sri Lanka compared with other South Asian countries. These advantages have recently come to light within the international business environment which ranks Sri Lanka favorably with respect to its global competitiveness and as a location for global services: The World Economic Forum in 2006 ranked Sri Lanka (79th) as more competitive than Pakistan (91st), Bangladesh (99th) and Nepal (110th). In addition, A. T. Kearney’s Global Services Location Index for 2006 includes Sri Lanka at rank 29 for being one of the top 45 off-shore locations for global services noting its widespread use of English, strong education system, and increasingly open and well-regulated business environments.
Business Climate and Legal Framework Indicators
Business Climate and Legal Framework Indicators
| Starting a Business | Enforcing Contracts | |||||
|---|---|---|---|---|---|---|
| Procedure (number) | Duration (days) | Cost (% GNI pc) | Procedures (number) | Duration (days) | Cost (% of debt) | |
| Sri Lanka | 8 | 50 | 9 | 20 | 837 | 21 |
| Pakistan | 11 | 24 | 21 | 55 | 880 | 23 |
| Bangladesh | 8 | 37 | 88 | 50 | 1,442 | 46 |
| Nepal | 7 | 31 | 79 | 28 | 590 | 24 |
| Cambodia | 10 | 86 | 236 | 31 | 401 | 121 |
| Vietnam | 11 | 50 | 45 | 37 | 295 | 31 |
| India | 11 | 35 | 74 | 56 | 1,420 | 36 |
| Bhutan | 10 | 62 | 17 | 34 | 275 | 20 |
Business Climate and Legal Framework Indicators
| Starting a Business | Enforcing Contracts | |||||
|---|---|---|---|---|---|---|
| Procedure (number) | Duration (days) | Cost (% GNI pc) | Procedures (number) | Duration (days) | Cost (% of debt) | |
| Sri Lanka | 8 | 50 | 9 | 20 | 837 | 21 |
| Pakistan | 11 | 24 | 21 | 55 | 880 | 23 |
| Bangladesh | 8 | 37 | 88 | 50 | 1,442 | 46 |
| Nepal | 7 | 31 | 79 | 28 | 590 | 24 |
| Cambodia | 10 | 86 | 236 | 31 | 401 | 121 |
| Vietnam | 11 | 50 | 45 | 37 | 295 | 31 |
| India | 11 | 35 | 74 | 56 | 1,420 | 36 |
| Bhutan | 10 | 62 | 17 | 34 | 275 | 20 |
F. Assessment and Policy Recommendation
24. The analysis presented in this paper indicates that Sri Lanka’s external competitiveness is adequate. Estimates of the equilibrium REER using various approaches indicates that the current level of the REER is broadly in line with economic fundamentals. The appreciation in the REER over 2005-2006 reflected a moderate deviation from the long run path, driven by high domestic inflation and intervention in the foreign exchange markets to limit nominal depreciation. These effects have begun to unwind in the first half of 2007 when the REER depreciated by about 1.5–2.5 percent. While export growth had declined slightly in real terms over 2005–2006, they have rebounded in the first half of 2007, led by growth in the apparel and textiles. The tradable goods sector also continues to enjoy relatively high profit margins, which may be an indication of Sri Lanka’s comparative advantage in certain commodity markets.
25. Nevertheless, greater exchange rate flexibility is essential to safeguard reserves in light of Sri Lanka’s near-term external risks. Given the persistent nature of high oil prices, a rising debt service burden going forward, and low levels of reserves, growing pressures on the foreign exchange market could occur in the period ahead and should not be resisted. In this context, greater flexibility in the nominal exchange rate is advisable, while limiting intervention to smoothing excessive volatility. This would go some way to contain external vulnerabilities and safeguard reserves.
26. Ultimately, although competitiveness may not be of immediate concern, it warrants monitoring going forward. The export base remains narrow, vulnerable to global commodity and apparel price changes, and dependent on preferential access for market share. Moreover, Sri Lanka’s labor productivity is low by regional standards and the internal terms of trade are skewed towards the nontraded sector suggesting that investing in the tradable sector is becoming increasingly less attractive. Looking forward, ensuring peace and macroeconomic stability will be key to attracting the needed foreign direct investment to develop the export sector, necessary for economic development and transition to an emerging market economy.
References
Balassa, Bela, 1964, “The Purchasing Power Parity Doctrine: A Reappraisal,” Journal of Political Economy, 72, pp. 584– 96.
Chinn, Menzie and Iro Hito, 2005, “Current Account Balances, Financial Development, and Institutions: Assaying the World’s Savings Glut,” NBER Working Paper 11761 (Cambridge Massachusetts: National Bureau of Economic Research).
Clark,Peter, and Ronald MacDonald, 1999,“Exchange Rates and Economic Fundamentals: A Methodological Comparison of Beers and Feers,” in J. Stin and R. MacDonald (eds.) Equilibrium Exchange Rates (Kluwer: Boston), pp. 285– 322.
Edwards, Sebastian, 1989, “Real Exchange Rates, Devaluation and Adjustment,” (Cambridge: MIT Press).
Hinkle, Lawrence E. and Peter J. Montiel, 1999, “Exchange Rate Misalignment—Concepts and Measurement for Developing Countries’” World Bank Research Publication (Oxford: Oxford University Press).
Isard, Peter, Hamid Faruqee, Russell Kincaid, and Martin Fetherston, 2001, “Methodology for Current Account and Exchange Rate Assessments,” Occasional Paper No. 209 (Washington: International Monetary Fund).
MacDonald, Ronald, 1997, “What Determines Real Exchange Rates?: The Long and The Short of It”, International Financial Markets, Institutions and Money, 8, 117– 53.
MacDonald, Ronald, 2007,“Exchange Rate Economics: Theories and Evidence”, (London and New York: Routledge).
Samuelson, Paul A., 1964, “Theoretical Notes on Trade Problems,” Review of Economics and Statistics, 46, pp. 145– 54.
Prepared by Lupin Rahman.
External competitiveness is taken to describe a country’s capacity to export goods and services more cost-effectively than others. Edwards (1989) defines external competitiveness more narrowly as the capacity of firms in the tradable sector to export and remain profitable in the long-term.
The E.U.S. GSP (II) provides for preferential duty treatment (a reduced rate of import duty or, even, duty-free) of imported goods originatinh in beneficiary countries.
Trends in Sri Lanka’s nominal and real wages are limited by the coverage and quality of data which focuses on the workers in wages boards only and hence does not reflect the complete distribution of private sector wages. The data used in this paper are from the International Labor Organization based on an annual survey of industries.
The internal exchange rate (or terms of trade) measures the relative price of nontradable to tradable goods and is an indicator of the relative attractiveness of production in the nontraded goods sector. See Hinkle and Montiel (1999) for a detailed discussion.
ULC-based REERs were calculated using rough estimates for unit labor costs from total value added to wage bill in manufacturing.
Although the VECM results are somewhat sensitive to the specification used, the general tendencies are as described.
A similar result was found when using the stock of reserves as a share of GDP.
