Change Font: A A A A Contact Us What's New FAQs Subscribe ADB.org home
HomePublicationsCatalogFiscal Policy Issues for India after the Global Financial Crisis (2008-2010)Trends and Patterns in Fiscal Variables in India

Trends and Patterns in Fiscal Variables in India


A look at the trends and patterns over the last three decades (1980–2010), which span both the pre- and post-reform period, helps us understand the relationship between fiscal expansion and growth in the Indian economy. The first surge in India's economic growth rate came in the early 1980s, when it increased to above 5% from the average “Hindu” growth rate1 of 3.5% in earlier decades. Unfortunately, this spurt was achieved by unsustainable fiscal expansion financed by domestic credit and external borrowing. Growth accelerated to 5.8% during the 1980s, but in the second half of the decade, fiscal and current account deficits widened significantly, causing serious macroeconomic imbalances and culminating in the balance of payment (BOP) crisis of 1991. These triggered the series of economic reforms that have been introduced, starting in 1991, to bring about macroeconomic stabilization and implement structural measures2 to push up growth.

In the following section, we analyze fiscal trends in greater detail. The analysis is based on an annual time series corresponding to the fiscal year (1 April to 31 March). The data is drawn mostly from the RBI's Handbook of Statistics on Indian Economy and Annual Reports and National Accounts Statistics published by the Central Statistical Organization (CSO).

3.1 Deficit Indicators

The 1980s saw a sharp rise in the combined fiscal deficit of the central and state governments to 8% on average (see Table 3.1 [ PDF 19.2KB | 1 page ]). Along with high external borrowings, a sustained increase in the combined revenue expenditure to stimulate demand, particularly in the services sector, caused the fiscal deficit to rise during the 1980s. As a result, the combined public debt became 56% of GDP on average, with interest payments at 14.6% of revenue expenditure (3% of GDP on average) accounting for a large portion of government revenue expenditure and creating a debt trap in the 1980s. During the first half of the 1980s, these revenue expenditures averaged 18.5% of GDP. In the second half, they rose to an average of 22.4% with the bulk of the expansion coming under the heads of defense, interest payments, higher salaries (Fourth Pay Commission) and subsidies.

Studies by Srinivasan and Tendulkar (2003), Joshi and Little (1994), and others attribute the spurt in economic growth during the decade to demand-side factors. The flip side, however, was the spilling over of this into external balances. By 1990, the current account and fiscal deficits had risen to 3.5% and 9.4% of GDP respectively, leading to the BOP crisis of 1991 (Ahluwalia 2000, Arvind Panagariya 2004a and, 2004b; Balakrishnan and Suresh 2004; Nirvikar Singh and Srinivasan 2004). Containing this deficit was one of the key structural adjustments undertaken by the Indian government at the time. Economic reforms helped reduce the fiscal deficit, and the combined fiscal deficit fell to 6.3% of GDP in 1996–1997.

A sharp increase in government salaries and pensions in the next year halted the process of fiscal improvement until 2003–2004 when the government introduced the Fiscal Responsibility and Budget Management Act3 (FRBM) to control the fiscal deficit. The Act required the Government of India to bring down its revenue deficit by 0.5% of GDP each year until it touched zero, and to reduce its fiscal deficit by 0.3% each year to a level of 3.0% of GDP. The targets were to be achieved by 2008–2009. Further, it set an annual limit of 9.0% on the union government's total liabilities and capped union government guarantees for public sector units and state government loans at 0.5% of GDP. The targets laid out by the government's FRBM Act and state-level fiscal responsibility legislations were achieved in 2007–2008, a year ahead of schedule (except for the central government's revenue deficit target). The combined fiscal deficit came down to 4.2% of GDP in 2007–2008 (well below the prescribed 6.0%) and the primary deficit (fiscal deficit net of interest payments) turned into a surplus of 1.3% of GDP in 2007–2008.

However, there is a lot of disagreement among policy-makers about targeting a zero revenue deficit in India, for the reason that it sounds unrealistic to target a zero revenue deficit and a 3% fiscal deficit because this implicitly assumes that revenue expenditure does not contribute to growth. For a developing country, it may be argued that it is desirable to target a small revenue surplus to finance capital formation rather than target a zero revenue deficit. This means the government would be saving and contributing to capital formation (Chelliah 2000). Public finance experts like Dr. Chelliah have also questioned the wisdom of setting a 3% fiscal deficit target. He says, “… 3% is too low for a developing country as the government still has to spend large amounts of money on infrastructure investment, including social infrastructure such as hospitals and schools.”

Figure 3.1 [ PDF 68.5KB | 1 page ] provides a synoptic view of fiscal trends from 1990–1991, the year in which India faced its economic crisis. There was a steady improvement in central and state finances from 2001–2002, when the fiscal and revenue deficits of the combined central and state governments had peaked at 9.9% and 7.0% of GDP respectively.

3.2 Debt Sustainability

The trends in fiscal deficit were mirrored in the rising public debt levels. The combined debt of the central and state governments, which averaged 56% of GDP in the 1980s, rose to an average of slightly over 63% in the 1990s and climbed further to touch a peak of 81.4% in 2003–2004 (see Table 3.2 [ PDF 16.6KB | 1 page ] and Figure 3.2 [ PDF 91KB | 1 page ]). A notable feature was the drastic reduction in the share of the external liabilities to GDP from 6.7% (on the average) in 1980s to 1.7% in 2003–20044. After the introduction of the FRBM Act, public debt steadily declined until 2008–2009, when it stood at 74.7%. The concern now is that the high fiscal deficits of the past two years may suggest a long-term reversal of this trend. Budget estimates for 2009–2010 indicate an increase in the public debt to above 75% (about Rs 44,000 billion). It could be even higher if GDP growth slows down further.5 This also may reverse the downward movement in the public debt to GDP ratio in India, which has been achieved over the last few of years.

These trends also point to one of the main deficiencies in the FRBM Act, namely the failure to set a cap on public debt. There is little doubt that the FRBM Act put the country on a higher growth trajectory by reducing the fiscal and primary deficits, but a sound fiscal system also needs to have in place measures to control the debt to GDP ratio. We hope the next set of FRBM targets include policies towards reducing public debt.

The rise in public debt can be attributed to the sharp rise in the primary deficit (i.e. fiscal deficit minus interest payments). The basic rule is that the ratio of debt to GDP will keep rising if there is a primary deficit or if the interest rate on debt exceeds the growth rate of GDP. With the fall in the GDP growth rate because of the global financial crisis, concerns regarding the sustainability of such high levels of public debt have become stronger. Should economic growth slow down because of the crisis, debt servicing could pose a problem as interest rates decline only with a lag, which would result in a further deterioration in government finances. This may also point towards the need to adopt an early exit from the high fiscal deficit regime

There is little consensus on what the ideal debt to GDP ratio for an economy should be. Internationally, the Maastricht Treaty has set the tolerable debt level at around 60% of GDP for the European Union countries. The Twelfth Finance Commission had recommended an even lower target of 56% over a period for India. To this end, it had also recommended that the ratio be brought down to at least 75% by 2009–2010.

If one goes by the budget estimates for 2009–2010, the government is quite clearly not going to be able to meet the target. The primary deficit, which as Rangarajan and Srivastava (2005)7 pointed out was the core variable that led to an increase in public debt to GDP ratio in the period 1951–2003, turned positive in 2008–2009. The gross primary deficit rose from -1.3% in 2007–2008 to 3.5% in 2008–2009 and is expected to rise further to 4.6% in 2009–2010. Since capital expenditure has been declining over this period, it is apparent that the main factor accounting for the rise in the gross primary deficit is increased revenue expenditures for both the central and state governments.

3.3 Receipts and Disbursement of the Government

3.3.1 Central and State Governments' Expenditure

At the central level, average government expenditure8 stood at 17.6% of GDP in the 1980s (see Appendix 1 [ PDF 38KB | 1 page ]). The share fell by 1.6% immediately after the reforms, mainly because of the macroeconomic stabilization program that followed the 1991 BOP crisis. However, a sharp rise in salaries and pensions following the acceptance of the Fifth Pay Commission report9 in 1996–1997 pushed the expenditure level back to the 16–17% level the following year—a level at which it stayed until the FRBM Act in 2004–2005.

After the FRBM was passed, central government's total expenditure fell from approximately 16% to 14% of GDP over the next two years. However, this expenditure control was achieved by cutting down capital expenditure sharply while revenue expenditure showed only a marginal decline. Thus, the composition of government expenditure, which has always been a matter of concern, remains unchanged with revenue expenditure accounting for about 80% of total expenditures.10

Public capital expenditure as a percentage of GDP declined from an average of 6.2% in the 1980s to 3.6% in 2004–2005 and further to 1.8% in 2008–2009. By contrast, revenue expenditure, which was 11.4% of GDP during the 1980s, rose to 12.2% in 2004–2005 and to 15.1% in 2008–2009.

As in the mid-1990s, the reason for the sharp rise in revenue expenditure in 2008–2009 has been the implementation of the recommendations of the Sixth Pay Commission Report and measures such as the debt waiver on farm loans and subsidies. Interest payments, which account for over 30% of revenue expenditure, stood at about 4% of GDP until 2004–2005. However, these came down to 3.6% in 2005–2006 and continued at the same level until 2008–2009. This, however, was not really the result of a reduction in borrowings but rather an effect of softening of interest rates.

The other major item of revenue expenditure has been subsidies. Budget data do not indicate the actual expenditure on subsidies because several subsidies are hidden in the production of intermediate goods and services and the quantum of subsidy at the stage of final consumption of goods or services is not clearly known (Radhakrishna and Panda 2006)11. Explicit government budgetary subsidies like those on food, fertilizers, and petroleum products are only a small portion of the total subsidy.

Food subsidy as a percentage of GDP rose from 0.4% in 1990–1991 to 0.9% in 2003–2004. This has decreased since 2003–2004 and reached 0.6% in 2006–2007. However it started rising again in 2007–2008 (see Table 3.3 [ PDF 16KB | 1 page ])12, partly due to enhanced food security measures with a higher subsidy for the poor. A part of this rise in subsidy is due to the high minimum support price for food grain procurement and the inefficient operation of the Food Corporation of India. This indicates scope for reducing subsidy without hurting the poor (Radhakrishna and Panda 2006). The government has recently taken some measures to make the food subsidy more target-group oriented by revamping the public distribution system and introducing differential prices for the poor and non-poor groups. Food subsidy has increased further and reached 0.9% of GDP in 2009–2010. Fertilizer subsidies have gradually increased to 0.7% of GDP in 2007–2008 and further shot up to 1.4% of GDP in 2008–2009, whereas petroleum subsidies were constant at 0.1% of GDP until 2009–2010.

More importantly, the growing practice of issuing special bonds to oil and fertilizer companies to support low consumer prices means that at least part of the subsidy burden is off the budget. While these subsidies do not appear in the budget, they do result in additional costs and risk for the government.13 Oil subsides, which are included in off-budget bonds, not only affect the liquidity position but also change the fiscal position of the government itself. The off-budget expenditure incurred by the government has almost doubled to 1.80% of the GDP (Rs 970.19 billion) in 2008–2009 from 0.98% (Rs 403.61 billion) in 2006–2007.

Expenditures at the state level exhibit a trend similar to those at the central level. From an average of roughly 15.5% of GDP in the 1980s and 1990s, the total state-level expenditures rose to nearly 18.0% in 2004–2005 (see Appendix 2 [ PDF 40.8KB | 1 page ]). While expenditures fell steadily for the next three years to 15.5% in 2007–2008 on account of the Twelfth Finance Commission measures, they rose again to 17.3% in 2008–2009. Budget estimates indicate that the level for 2009–2010 will climb back to the 2004–2005 level.

An increase in revenue expenditure also accounted for the rise in states' expenditure. Between 2004 and 2005, there was some reduction in revenue expenditure but the trend reversed in 2008–2009 and it is expected to touch a high of 14% in 2009–2010. Capital expenditure has shown a more fluctuating trend. In the immediate post-reform period, there was a sharp drop in states' capital expenditures. This was an unhealthy development, because by reducing capital expenditure to achieve fiscal balance, they had effectively compromised on building the infrastructure capacity needed to promote growth. There was a moderate increase in states' capital expenditure in the three-year period from 2002–2004 but it slipped again thereafter. However, it has since increased from 3.5% in 2007–2008 to 3.9% in 2008–2009.

3.3.2 Central and State Governments' Receipts

The persistent fiscal expenditures reveal that total receipts of both the central and state governments have remained consistently below total expenditures. Tax receipts, which contribute the bulk of the central government revenues, fell sharply in the period following the introduction of the reforms in 1992. This was the result of the rationalization of the tax structure. Total tax revenue as a proportion of GDP declined from 10.3% in 1990–1991 to the lowest level of 8.2% in 1998–1999. It was only in 2005–2006 that tax revenue touched the level it was at in 1990–1991 (see Appendix 1). Tax receipts rose to 12.6% in 2007–2008 but again declined to 11.8% in 2008–2009.

The tax reforms14 initiated since 1991 were part of the structural reform process after the 1991 economic crisis. The Tax Reforms Committee (TRC) concentrated on finding a suitable framework to reform both the direct and indirect tax structure. The committee recommended two major reforms on direct taxes—one was the simplification and rationalization of the direct tax structure (Chelliah 1992); the other was to introduce a service tax to widen the tax base (Chelliah 1994).

The Chelliah Committee (1992: 4) had, in its interim report, recommended that as a first step towards the rationalization of the personal income tax structure a three-rate slab structure should be introduced and later replaced by a two-rate structure. Further, the committee suggested reducing corporate income taxes. Both the recommendations were accepted and implemented in 1992. The maximum marginal rate of personal income tax was reduced to 40% from 56% in June 1991. Further, rates of corporate income tax, which were 51.8% for a publicly listed company and 57.5% for a closely held company, were unified and reduced to 46.0% in 1992. These rates were inclusive of a 15% surcharge.

The 1992 reforms radically altered the composition of tax revenue at the central level15. Direct taxes as a percentage of GDP rose from 2.0% in the 1980s to 6.5% in 2008–2009. However, this rise in the proportion of direct taxes was offset by a reduction in central indirect tax revenues as a percentage of GDP from 7.9% to 5.3% over the same period. The share of non-tax revenue16 in GDP at the central level fluctuated between 2.0–3.0% during 1980–2009 with the 3.0% recorded in 2001–2002 and lowest 1.8% observed in 2008–2009.

The government also introduced a service tax in 1994 in line with the recommendations of the Chelliah Committee17. Until then, the service sector had been totally left out of the tax net though the sector's contribution to GDP had risen to 36% by 1993–1994. Starting with three services, viz.,namely telephone, stock broking, and insurance services, the coverage has progressively widened over the years with about 80 services having been brought within the ambit of taxation tillto date. A few important services brought under the service tax net are banking and other financial services, management consultants, credit rating agencies, and market research agencies. Some important services that are still outside of the tax net are legal consultancy services, transport of goods by waterways, and cosmetic or plastic surgery. The rate imposed originally was a moderate 5% of turnover. This was, however, progressively increased to 12% and an additional education tax of 2% on service tax was imposed in 2006–2007. The 2008 crisis, however, forced a rollback in the service tax rate to 10% in February 2009. Collections from service tax have shown a steady rise from 1994–1995 (0.2% of GDP) to 2008–2009 (1.1% of GDP). However, in 2008–2009, they accounted for only 10.4% of the total tax receipts of the central government while the share of services in total GDP has gone up to 57%.

Major changes on the indirect tax side included a sharp reduction in import duties from extremely high levels to a range of 15–30% for manufacturers, reduction of multiple excise tax rates to three in the range of 10–20%, and extension of the then existing modified value added tax (MODVAT)18 credit to all inputs. In 2000–2001, the government converted the three excise duties into a single central value added tax (CENVAT), levied at the rate of 16%. Subsequently, state-level value added tax (VAT) replaced CENVAT in 2005–2006 though initially only 20 states accepted the proposal. The acceptability of the VAT has gone up to 28 states to date. Four slabs of VAT have been uniformly applied across all states that adopted it—0.0% on necessities and primary goods, 1.0% on bullion and precious stones, 4.0% on industrial inputs and capital goods and items of mass consumption, and 12.5% on all other items. Necessities and primary products were left out of the ambit of VAT.

The government now intends to move to a goods and services tax (GST) regime, which will replace state-level VAT and CENVAT. The tax will be imposed on final goods and services with a two-rate structure. The GST, which is being steered by an empowered committee of state finance ministers, was expected to be launched in April 2010. Its introduction will most likely be delayed until October 2010. This will mark a major step in unifying the tax regime across the country and do away with tax arbitrage that currently disturbs investment decisions.

Figure 3.3: Direct and Indirect Taxes and Non-tax Revenues of the Central Government [ PDF 78KB | 1 page ]

At the state level, fiscal health depends both on revenues from state taxes as well as constitutional and other transfers from the central government. There is a three-tier transfer mechanism in India. First, the Indian Constitution provides for mandatory transfer of revenue from central taxes on the basis of the recommendation of a Finance Commission that the central government is required to set up every five years. Each Finance Commission uses different criteria to transfer funds. Second, there are budgetary transfers made through the Planning Commission to implement plan projects19. Third, there are optional transfers through various union ministries and agencies. Fund transfers from the central government form a large part of revenue of the state governments. These transfers are accounted for in a state's revenue receipts. There are several issues related to the transparency of central government transfers and accounting problems. The discussion about these problems is beyond the scope of this paper.

A look at the revenue receipts of states shows that there was a steady improvement in the tax ratio during the study period. The revenue from states' tax receipts (including their share in the central pool) as a ratio of GDP was virtually stagnant throughout the 1980s and 1990s at around 7.7% (see Appendix 2). There was some decline from 1994–1995 and the low point of 7.2% was reached in 1998–1999, the year in which the states had to revise their pay scales, which exacerbated their fiscal problems. The pay revision in 1997–1998 created more fiscal stress for state governments as revealed by the increase in their revenue deficit from 0.9% of GDP in 1990–1991 to 2.6% in 2000–2001. The extent of the stress on state budgets can be gauged from the fact that, since the mid-1990s, salaries and pensions account for 80–90% of revenue receipts in most states. However, the tax ratio has steadily improved from 7.8% in 2000–2001 and reached 9.6% in 2008–2009.

A major development at the state level is the adoption of VAT from 2005–2006. The VAT would help to remove the cascading tax burden. Tax revenue20 is expected to rise as compliance improves under VAT. The state VAT has evidently helped tax revenues to increase from 8.6% in 2005–2006 to 9.6% in 2008–2009.

3.3.3 Combined Receipts and Disbursement

Taking the budgetary position of the central government and states together, one finds that the combined expenditure as a percentage of GDP rose from 26.8% in the 1990s to 27.4% in 2007–2008 (see Table 3.4 [ PDF 21.1KB | 1 page ]). The subsequent two years show a sharp rise in expenditures, with the budget estimates for 2009–2010 showing expenditure at almost 32% of GDP. This has been a consequence of a sharp increase in public expenditure in the run up to the general elections of 2009–2010.

Total receipts have also shown a similar increase from around 26% to roughly 31% from the 1990's to 2008–2009 (see Table 3.4). Over 60% of receipts are accounted for by revenue receipts (both tax and non-tax). The rest has come from capital receipts of which the two major components have been debt capital receipts (mainly borrowings) and disinvestment.

The share of the central government's capital receipts21 in GDP was just above 6% until 2000–2001 and thereafter increased until 2003–2004. Since then, it declined reaching 3.6% in 2007–2008. As Table 3.4 indicates, debt capital receipts have been the major contributor to capital receipts. The contribution from disinvestment has been about 1–2% of capital receipts in the post-reform period. Disinvestment was the highest in 2003–2004, amounting to Rs 169.53 billion (0.6% of GDP). However, it did not pick up momentum till until 2007–200822 where the disinvestment receipts were Rs 457.50 billion (about 1% of GDP)23.

3.4 Public Sector Savings and Investment 2424

The deterioration in the fiscal position of the central and state governments has impacted public sector savings and investment. The share of nominal public sector savings in nominal output25 averaged just above 3.5% in the 1980s (see Table 3.5 [ PDF 16.4KB | 1 page ]). This had reduced to an average of 1.5% in the 1990s. Public sector savings deteriorated further in the period after reforms were initiated, turning negative (-1.8%) in 2000–2001. Though there was some improvement in 2002–2003, public sector savings turned positive again only in 2003–2004, a trend that was maintained until 2008–2009. They peaked in 2007–2008 reaching 4.5% of GDP. There was a sharp deterioration in 2008–2009 when public sector savings turned negative at -1.8%. Budget estimates for 2009–2010 indicate a further deterioration.

The 1980–2009 period also saw a rapid decline in public sector investment, especially in the infrastructure and agriculture sectors. The fall was particularly sharp after the 1991 reforms. Since state governments in India typically handle both agriculture and infrastructure, declining public sector investment reflects in part the deterioration in the fiscal position of state governments.

Concern remains that high fiscal deficits would crowd out private investment by keeping interest rates high in the short-term. In the long term, the lack of critical investments would prevent the crowding in effect from becoming operative. A growing fiscal deficit will, therefore, adversely impact both the long and short-term growth prospects of the economy.

3.5 Structural and Cyclical Behavior of Major Fiscal Variables

3.5.1 Relationship Between Gross Fiscal Deficit and Growth

The relationship between the fiscal deficit and output growth has been of enduring interest for the Indian economy. In Figure 3.4 [ PDF 93.5KB | 1 page ], the annual data of the combined gross fiscal deficit (GFD) of both the central and state governments is plotted against GDP at market prices from 1980–1981 to 2009–2010 (BE Budget Estimates). Until 2002–2003, there appeared to be considerable long-run coordinate movement between these two series. This indicates that the relationship is structural rather than cyclical. However, for a short period over 2006–2007 and 2007–2008, fiscal deficit decreased as the output increased. This negative relationship could be attributed to the implementation and realization of FRBM targets. There is a sudden jump in fiscal deficit in 2008–2009 and 2009–2010 (BE), though output has grown at a slower pace27 , making the association between GFD and GDP horizontal in 2008–2009 and 2009–2010. Nonetheless, there is an upward linear trend exhibited throughout the study period implying a positive relation between fiscal deficit and output growth.

Interestingly, we find different results altogether when gross fiscal deficit as a share of GDP is plotted. Figure 3.5 [ PDF 39.4KB | 1 page ] shows the gross fiscal deficit as a share of GDP. The relative growth of GFD to GDP exhibits cyclical behavior through the study period. The cycle does not seem to coincide with the electoral cycle but the peaks coincide exactly with the pay commission recommendations28 and the troughs coincide with fiscal reforms29.

As discussed earlier in the paper, the relationship between the size of fiscal deficit and GDP growth has been an intensely debated one. There are those who believe in its “crowding in” effect in a developing economy. Their view is contrasted by others who see a high fiscal deficit as pre-empting domestic savings and discouraging private investment, resulting in a “crowding out” phenomenon. We have tried to test the validity of these arguments, by trying to quantify the relationship between GDP growth and fiscal deficit taken as a percentage of GDP. We estimated the simple equation given below.

1. Gr GDP = 8.63 + 0.07 Gr GCF - 0.41 GFD/GDPM30

(3.8) (1.8) (-1.5)

R2 = 0.17 DW = 1.92

Equation 1 yields a negative correlation, though a weak one, between GDP growth and fiscal deficit as a percentage of GDP. This substantiates the argument made by several Indian economists (Govinda Rao 2009; Rangarajan 2009).

But the long run relationship between GDP and fiscal deficit, using the logarithm of both to avoid non-stationarity problem, is surprisingly a positive one as given by Equation 2.

2. Log GDP = 1.28 + 0.64 Log GCF + 0.19 Log GFDR + 0.39 AR (1)

(2.6) (15.9) (3.4) (2.0)

R2 = 0.99 DW = 2.1

Apparently in conditions of unemployed resources and rising demand, an expansion in public expenditure, even when it increases the fiscal deficit, results in the positive impact of “crowding in” swamping the negative effect.31

3.5.2 Relationship Between Public Debt and Growth

Annual data on the combined outstanding liabilities and GDP at current market prices from 1980–1981 to 2009–2010 (BE) is plotted in Figure 3.6 [ PDF 104.1KB | 1 page ]. The scatter graph below depicts trends that are similar to that in the case of the fiscal deficit throughout the study period, confirming the structural behavior of public debt over decades. It shows that there is a positive relation between GDP and public debt from 1980s. However, there seems to be a marginal downturn from 2007–2008 to 2009–2010, implying rising public debt has had a negative impact in recent years.

Download this Paper [ PDF 753.9KB| 42 pages ].




[previous chapter] [next chapter]


Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [0] comment(s) for this entry. Post a comment.

    The views expressed in this paper are the views of the authors and do not necessarily reflect the views or policies of the Asian Development Bank Institute (ADBI), the Asian Development Bank (ADB), its Board of Directors, or the governments they represent. ADBI does not guarantee the accuracy of the data included in this paper and accepts no responsibility for any consequences of their use. Terminology used may not necessarily be consistent with ADB official terms.

    Back to Top 
    © 2014 Asian Development Bank Institute.