1 Introduction

T

2 Varieties of Deceleration Experience

Growth was adversely affected in all G20 countries. However, there were large variations among them in the timing and severity of deceleration; the timing, scale and composition of stimulus packages; and the response lag before they began to turn around. We refer to the “deceleration” experience rather than recession because as many as five out of 20 countries, or 25% of the G20, continued to register positive growth throughout the crisis period, while the others went into recession, in some cases a very deep recession.1

The beginning of growth deceleration in each G20 country, the point at which a stimulus package was first introduced (in some cases these were followed by a second or even a third package), and the point at which the country began its recovery after bottoming out have been marked in Figure 1 (p 86), which

Economic & Political Weekly

EPW

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1	2007Q2	2007Q3	2007Q4	2008Q1	2008Q2	2008Q3	2008Q4	2009Q1	200	9Q2	2009Q3	2009Q4	2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1	2007Q2	2007Q3	2007Q4	2008Q1	2008Q2	2008Q3	2008Q4	200	9Q1	2009Q2	2009Q3	2009Q4	2010Q1

2007Q1	2007Q2	2007Q3	2007Q4	2008Q1	2008Q2	2008Q3	2008Q4	2009Q1	200	9Q2	2009Q3	2009Q4	2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

4 0 -4 -8 -12

2007Q1 2007Q2 2007Q3 2007Q4 2008Q1 2008Q2 2008Q3 2008Q4 2009Q1 2009Q2 2009Q3 2009Q4 2010Q1

Beginning of Growth Deceleration ( ), Beginning of Stimulus (

juLY 16, 2011 vol xlvi no 29

tracks the annualised growth rate for each quarter from 2006Q1 to 2010Q1.2 Some countries like Indonesia, Saudi Arabia, Brazil, and Argentina did not experience any deceleration until the last quarter of 2008, i e, after the collapse of Lehman Brothers in September 2008. In other countries the GDP growth rate had actually started decelerating more than a year earlier, by 2007Q2 or Q3.3 However, among them a distinction has to be made between those countries such as the US and France, where deceleration was triggered by their high exposure to

Country Time of Deceleration Time of First Stimulus Time Lag Time of Recovery Time Lag Average Annual Minimum Growth Maximum Deceleration Size of Stimulus

(1) (2) (3) (4)= (3)-(2) (5) (6)= (5)-(3) Growth Rate1 (7) Rate2 (8) (9) =100[{(7)-(8)}/(7)] (in Billion) (10)

Indonesia 2008Q3 2008Q4 1 Quarter 2009Q4 4 Quarters 5.93 4.16 31.16 6.5 US$

India 2007Q2 2008Q1 3 Quarters 2009Q1 4 Quarters 9.60 5.76 40.57 4.1 US$

China Mainland 2007Q4 2008Q4 4 Quarters 2009Q2 2 Quarters 11.70 6.10 48.23 196 US$

Australia 2007Q2 2008Q4 6 Quarters 2009Q4 4 Quarters 3.31 0.87 77.62 52.4 US$

Saudi Arabia3 2008Q4 2008Q4 0 Quarter 2010Q1 5 Quarters 3.77 0.15 96.02 17.5 US$

Argentina 2008Q3 2008Q4 1 Quarter 2009Q3 3 Quarters 8.55 -0.77 109 36.5 US$

Brazil 2008Q3 2008Q4 1 Quarter 2009Q2 2 Quarters 4.85 -2.13 143.95 94.8 US$

South Africa 2008Q2 2009Q1 3 Quarters 2009Q3 2 Quarters 5.36 -2.47 147.77 3.7 US$

Korea, Republic of 2008Q1 2008Q4 3 Quarters 2009Q2 2 Quarters 4.79 -4.30 189.61 11 US$

Spain 2007Q2 2008Q2 4 Quarters 2009Q3 5 Quarters 3.83 -4.22 210.27 29 Euro

Canada 2007Q4 2009Q1 5 Quarters 2009Q3 2 Quarters 2.68 -3.18 218.52 43.5 US$

United States 2007Q3 2009Q1 6 Quarters 2009Q3 2 Quarters 2.63 -3.83 245.23 787 US$

Russian Federation 2008Q1 2008Q4 3 Quarters 2009Q3 3 Quarters 7.44 -10.94 247.11 20 US$

France 2007Q3 2008Q4 5 Quarters 2009Q2 2 Quarters 2.24 -3.93 275.42 33 US$

Turkey 2008Q1 2009Q1 4 Quarters 2009Q2 1 Quarter 6.75 -14.45 314.23 9.84 US$

United Kingdom 2008Q1 2008Q4 3 Quarters 2009Q3 3 Quarters 2.52 -5.90 334.2 30 US$

Mexico 2008Q2 2009Q1 3 Quarters 2009Q3 2 Quarters 3.68 -9.97 371.03 20 US$

Germany 2008Q1 2008Q4 3 Quarters 2009Q2 2 Quarters 2.34 -6.70 386.66 29 US$

Japan 2007Q1 2008Q3 6 Quarters 2009Q2 3 Quarters 2.10 -8.57 508.48 150 US$

Italy 2007Q3 2009Q1 6 Quarters 2009Q2 1 Quarter 1.55 -6.52 519.56 2.56 US$

1- Simple average growth rate based on GDP data from 2004:Q1 to one quarter before the time of deceleration. 2 - Growth rate when the country experienced maximum fall in the GDP growth rate. 3 - Interpolation method has been used to derive the quarterly numbers for Saudi Arabia. Source: Authors' calculation based on GDP data from IMF, IFS, July 2010. Size and timing of stimulus for each country collected from respective country reports and various press releases

(see Appendix-A).

early events of the financial crisis, i e, the sharp rise in sub-prime housing mortgage loan defaults in the US, and those where deceleration was set off by other factors such as the spike in food and petroleum prices.

Collateral data on stock price indices, capital flows, exchange rate movements, and anecdotal information on their limited exposure to the US sub-prime mortgage loan market, suggest that countries like Australia and India belong to this latter category. The initial decline in growth, triggered by other factors, was later overtaken by the effects of the crisis. But the precise time from which the crisis started driving the decline remains a matter of judgment. In our view, this shift occurred in 2008Q4 in Australia and 2008Q3 in India.

The introduction of stimulus packages was closely coordinated. Most countries introduced their packages in the last quarter of 2008, in the immediate aftermath of the Lehman collapse. There are some outliers. Japan and Spain already had stimulus packages in place by the second quarter of 2008. India introduced its “official” stimulus package in 2008Q4. But its real and much larger stimulus, in the form of a large planned fiscal deficit, was already introduced in February 2008 in the budget for 2008-09, though it was not presented at the time as

Economic & Political Weekly

EPW

juLY 16, 2011 vol xlvi no 29

case of Australia and Saudi Arabia. The response lag was four quarters in India and Indonesia.

These differences in the time lag between stimulus and response are attributable to differences in the underlying economic environment that may have impacted on the efficiency of the transmission mechanism, differences in the size of the stimulus,4 and differences in the severity of the crisis in different G20 countries. The severity of the crisis in a country is frequently measured by the maximum output gap, the gap between potential output and actual output, when the country bottomed out. However, potential output is known to decline in a recession (Zhang and Zhang 2009) and there is some ambiguity about how best to measure potential output in a recession, hence also the output gap. In this paper, the severity of the crisis has been measured as the absolute difference between the average annual growth rate and the growth rate at the bottom of the growth trough in the country. The average annual growth rate has been estimated from 2004Q1 to the last quarter before the country experienced a sustained, usually monotonic, decline in growth rates. These estimates are also presented in Table 1. It turns out that the declin e in growth, compared to the pre-crisis average annual growth rate, has also varied a great deal. The smallest declines of 31% and 41% were recorded in Indonesia and India respectively, followed by 48% in China. At the other end of the range, the relative decline in growth has been as much as 520% in the case of Italy, and 508% in the case of Japan. More than half of the G20 countries experienced growth deceleration in excess of 200%, most of them being advanced countries. The two emerging G20 countries that are in this group include Mexico (371% decline) and Turkey (314% decline).5

There were large variations in the time taken for the stimulus to get transmitted to the level of real economic activity, partly because of these large differences in the severity of the crisis and partly because of differences in the size and composition of the stimulus and other conditions. It is arguable that there would be similar variations in time lags for the reverse transmission when the stimulus is withdrawn, since it would operate through the same mechanism in reverse. However, other factors that may have had an impact on the pace of the response may have changed meanwhile, including the growth performance of individual countries. Hence, each country’s readiness for stimulus withdrawal depends very much on its current growth, investment climate and other macroeconomic conditions that impact on the robustness or fragility of its recovery. Any “one size fits all” type proposition that all G20 countries must simultaneously initiate withdrawal of the stimulus, or that none of them is ready to initiate such withdrawal, seems quite arbitrary and irrational. Enforcing any such herd behaviour on the G20 runs the risk of eithe r another global deceleration triggered by a G20 wide

Table 2: G20 Countries: Selected Macroeconomic Indicators

negative shock of simultaneous stimulus withdrawal, or of overheating and high inflation in the countries ready for stimulus withdrawal if the withdrawal is postponed.

3 Readiness for Stimulus Withdrawal

An assessment of the readiness of different G20 countries to initiate withdrawal of the stimulus is, therefore, critical for successful G20 coordination on this issue. Selected macroeconomic indicators of the G20 countries have been presented in Table 2. The performance of a country in terms of each of these indicators has also been graded. The exercise is analogous to that which rating agencies undertake to rate sovereign country risk, except that our exercise is much simpler, and undertaken for purposes of assessing a country’s readiness for withdrawal of the stimulus.

The countries have been grouped primarily according to their growth performance, since recovery of output and employment growth was the main objective of the stimulus packages. In general, based on the growth performance, all the G20 countries have moved out of recession with positive growth in 2010Q3. In the first group we have seven countries with high growth performance, all of them being emerging economies. Growth rates in these countries are well above 5% and in three of them over 8%. They are clearly ready for withdrawal of the stimulus. In some of these countries, e g, India, such withdrawal is even urgent because the inflation rate (CPI) is high. Withdrawal of the stimulus in India is also urgent because it has a large primary deficit (central and state governments combined) and a large

juLY 16, 2011 vol xlvi no 29

public debt in excess of 70% of GDP (IMF definition), though the sovereign external debt is quite modest. The health of the financial sector, reflected here in the ratio of non-performing loans, is quite robust in India. This allows room for monetary tightening, in addition to fiscal consolidation, without much risk of a collapse of the financial sector.

An interesting case in this group of countries is Turkey, which was one of the worst affected by the crisis (Table 1). It has recorded nearly 12% growth in 2010Q1, although this moderated to the pre-crisis level of about 5.5% in 2010Q3, implying a very sharp ‘V’ shaped recovery. This is partly due to its IMF assistance program. Inflation continues to be high in Turkey, making stimulus withdrawal both feasible and urgent. The primary deficit is quite modest in Turkey, as also its public debt, and sovereign external short-term debt is negligible. However, it has a high proportion of non-performing loans, highest among the seven countries in this group, which suggests that stimulus withdrawal initially could focus on fiscal measures, without too much pressure on the monetary-financial side. Argentina is another high growth emerging economy with high inflation, which makes stimulus withdrawal urgent. While it has a primary surplus, and a moderate public debt-GDP ratio, a fairly large part of public debt is shortterm external debt, which reinforces the case for further fiscal strengthening at an early date.

Next we have a group of 10 countries with moderate growth performance and moderate or low inflation except for Russia that has an inflation rate of over 8%, implying there is no immediate urgency to withdraw their stimulus packages. If such growth is sustained, then these countries could also move towards withdrawal of the stimulus. This may be particularly important in the case of Japan, which has a large primary deficit, over 8% of GDP, on top of a very high public debt-GDP ratio, much larger than for any other G20 country, a significant proportion of which is short term external debt. United States has the largest primary deficit of nearly 11% of GDP and a high public debt-GDP ratio, which is projected to rise further. Also, it has a very high ratio of short term sovereign external debt to reserves. This would have raised very serious concerns about solvency, but for the fact that the dollar happens to be the world’s reserve currency. Inflation is high in Russia at over 8%, but it has a very low public debt-GDP ratio, with a negligible external debt component and a weak financia l sector, as reflected in a non-performing loan proportion of above 8%.

Next, we have a group of three low growth countries, France, Italy and Spain. It would be premature to withdraw the stimulus in these countries, since the prospects of sustained recovery remain uncertain. Also, none of these countries is subject to strong inflationary pressures at present. On the other hand, these countries have a large public debt stock, and a significant portion of it is short-term external debt, raising concerns about solvency. The situation is potentially challenging for France, which also has a high public debt-GDP ratio, projected to grow further because of a large primary deficit and very large short-term sovereign external debt liabilities, one of the highest among all G20 countries. Similarly, in Spain where output growth is still negligible, the public debt-GDP ratio is huge and is projected to rise further

Economic & Political Weekly

EPW

because of a large primary deficit of around 8% of GDP. The ratio of short-term sovereign external debt to reserves is around 270%.

To summarise, the broad stylised facts are as follows. Barring some outliers such as South Africa and Saudi Arabia, most of the emerging G20 countries are growing quite rapidly and ready for withdrawal of their stimulus packages. In some of these countries stimulus withdrawal is also urgent because they have high inflation. In most cases, the withdrawal needs to focus on the fiscal side. But in some high inflation countries like India, which have a robust financial sector with adequate bank capitalisation and low non-performing loan ratios, there is a strong case for monetary tightening as well. In most of the advanced G20 countries, on the other hand, the recovery of growth is still moderate with manageable inflation. Hence, withdrawal of the stimulus would be premature. Continuing weaknesses in the financial sector raise the risk of double dip recession if there is premature monetary tightening in these countries. On the other hand, many of them have high public debt ratios that are rising rapidly because of large primary deficits, which raises concerns about debt sustainability in the absence of early fiscal tightening. Moreover, in some cases the short-term external component of this debt is also very high compared to reserves, raising concerns about their solvency – some of which has already been reflected in international financial markets.

The ongoing debate about stimulus withdrawal has to be seen in the context of this background, particularly the challenges being faced by some of the advanced G20 countries. Much of the debate has focused on the dilemma being faced by several advanced G20 countries. There is a consensus to move cautiously on the monetary side because of at least three reasons. First, it is possible that significant asset losses have not yet been discovered. Second, central banks may find it difficult to quickly unload the assets acquired in the course of extraordinary measures to shore up the financial market at the height of the crisis. Third, sudden monetary tightening and a sharp increase in interest rates could adversely affect the yield curves of banks that may be borrowing short and lending long (Giavazzi 2010; Minegeshi and Cournede 2010; IMF 2010a).

Table 3: Deterioration in Fiscal Balances and Debt Position in G20 Countries 2007-10

(% GDP)

On the fiscal front, moderate to one of the highest among all G20 countries, serious questions could be raised about its debt sustainability and solvency slow recovery of some advanced G20 countries would suggest that the fiscal consolidation should also be delayed. However, many of these countries are also quickly running out, or have already ran out of the fiscal space they need in terms of tolerable levels of public debt to delay fiscal consolidation (Table 3, p 89).6 Several papers from the IMF suggest that the advanced countries should aim at a stable public debt: GDP ratio of 60%, and based on likely debt dynamics this will eventually require a massive 8% reduction in structural primary deficits on average (IMF 2010b; 2010d). In a related argument Corsetti and others explain, with the help of a model that includes financially excluded households, that in the absence of an announced fiscal consolidation plan a stimulus based on short-term tax relief measures will have a declining impact. They therefore make the case for announcing credible spending policies that endogenise future spending reversals (Corsetti, Meier and Muller 2009).

Actual fiscal consolidation has to be postponed since the recovery remains fragile in these countries, leading to a further increase in public debt: GDP levels. At the same time, continuing the stimulus may not be effective, for, the fear of tax increases to balance the budget in the future may lead to a decline in current private consumption. It is therefore important to endogenise future spending reversals. How is this to be accomplished? There seems to be a wide consensus that the credible pre-commitment to future spending reversals should focus on entitlement spending reforms, particularly reform of public spending on pension and health (Barrell, Hurst and Kirby 2009; Giavazzi 2010; IMF 2010d). The compelling argument made here is that just a twoyear increase in the retirement age and consequent postponement of pension liabilities could bring down the public debt-GDP ratio by as much as 40%.

In an alternative perspective, Giavazzi (2009) points out that US savings are likely to rise from zero to about 4% as a consequence of financial sector reforms, withdrawal of extraordinary monetary policy measure and the fiscal stimulus, reflecting a corresponding reduction in consumer spending. This raises the risk of another slump unless there is a matching increase in demand elsewhere. Given uncertainties about public investment, Giavazzi (2009) underlines the importance of increased private investment to adjust the production structure of the US economy to match the shift in the structure of global demand with the emergence of very large but low per capita income economies like China and India. He recommends public policies to support a spurt in private investment to restructure the US economy.

While interesting in itself, much of the foregoing discussion is of limited relevance for the issues that matter regarding stimulus withdrawal in India, and presumably also some of the other emerging G20 countries. First of all growth is robust in these countries, making immediate stimulus withdrawal feasible, and inflation is also high in countries like India making such withdrawal urgent. Second, at least in India, the financial sector is healthy, with well capitalised banks, and there was no extraordinary asset acquisition by the central bank as part of its stimulus measures. This suggests that monetary tightening should be the policy of choice for curbing inflationary pressures. Indeed, the Reserve Bank of India has been tightening monetary policy and gradually raising nominal interest rates consistently during the past few quarters.

Third, the public debt: GDP ratio is high in India and the fiscal authorities have already announced their commitment to a programme of fiscal consolidation, which can include measures to both raise revenue as well as compress expenditure, in order to meet the fiscal consolidation goals set by the Thirteenth Finance Commission. The issue that really matters in the Indian context is the feasibility of a program that can combine fiscal consolidation with high and inclusive growth.

4 India: A Strategy for Fiscal Consolidation with High Inclusive Growth

As a prelude to the discussion of India’s fiscal consolidation strategy, it is useful to review the specifics of India’s deceleration experience and the policy response to date. The impact of the global crisis on the Indian economy was much lower, and the revival of the economy has been much faster, than in most other G20 countries. The slowdown had already started in India in the second quarter of 2007, but for reasons unrelated to the financial crisis.7 The crisis led to a much sharper deceleration from the last quarter of 2008. However, it bottomed out within two quarters. Furthermore, the deceleration was relatively shallow compared to most G20 countries, with the growth rate declining by about 41% from the second quarter of 2007 to the bottom of the cycle in the last quarter of 2008.

A major reason for the relatively low impact of the crisis in India, and its quick revival, is the fact that India had actually started its expansionary fiscal policy, long before the global crisis peaked. The global financial meltdown unfolded following the collapse of Lehman Brothers on 16 September 2008, but India’s expansionary fiscal stance had already been introduced in the budget for 2008-09 presented by the finance minister in February 2008. The combined impact of expansion in the scope of the National Rural Employment Guarantee Act, waiver of farm loans, implementation of the Sixth Pay Commission recommendations for central government employees, and the additional provision for food and fertiliser subsidies was to increase the fiscal deficit of the central government budget from 2.6% in 2007-08 to 6.1% in 2008-09.

Adding in off budget liabilities, the central government deficit increased by over 5 percentage points in 2008-09 compared to the previous year. In addition, there was substantial spending by political parties on the eve of the 2009 general elections through formal channels, and possibly even more through non-formal channels, though this is difficult to quantify. Although these expenditures were not declared as part of the stimulus package, they greatly helped in easing the impact of the crisis and in the fast recovery. By comparison, the formally announced stimulus package that is usually cited for India, mainly consisting of cuts in the union excise duty and service tax rates, amounted to less than 0.5% of GDP.

As noted earlier, there are several factors that make early fiscal consolidation imperative for India. First, growth has been robust at 8% in 2009-10 and it is estimated at 8.6% for 2010-11. Thus the

juLY 16, 2011 vol xlvi no 29

gap between past average growth and the current actual growth rate is modest, even though the external demand for Indian exports has not revived. Second, there are strong inflationary pressures in India which require fiscal compression to support monetary tightening. Third, both the fiscal deficit and the public debt stock in India are quite high. According to the IMF Fiscal Monitor (IMF 2010d), the cyclically adjusted (CA) fiscal deficit for India in 2009 was 10.5% and it is estimated at 9.2% for 2010. This is more than double the average for emerging G20 economies (Table 3). India also has a high public debt-GDP ratio of 69.5%8 compared to an average of only 37.4% for emerging G20 countries. Most of this debt is internal, with little impact on India’s solvency on the external account. However, servicing this debt consumes a high proportion of revenue, thereby crowding out other spending that could promote inclusive growth.

Thus, there are compelling reasons to withdraw the fiscal stimulus along with the withdrawal of the monetary stimulus that is already underway. For the medium term, the government has accepted the fiscal consolidation recommendations of the Thirteenth Finance Commission (Table 4). The commission has recommended that the fiscal deficit should be reduced from 9.5% of GDP in 2009-10 to 5.4% in 2014-15. The commission has also recommended separate targets for the central and state governments. For the central government, it has recommended that the revenue deficits should be eliminated and capital

Table 4: Consolidated Fiscal Reform Path of the Centre and States (% GDP)

2009-10 2010-11 2011-12 2012-13 2013-14 2014-15

Fiscal deficit – states 2.8 2.6 2.5 2.5 2.4 2.4

Fiscal deficit – centre 6.8 5.7 4.8 4.2 3.0 3.0

Net central loans to states 0.1 0.0 0.0 0.0 0.0 0.0

Aggregate fiscal deficit 9.5 8.3 7.3 6.7 5.4 5.4

Debt stock – states 27.1 26.6 26.1 25.5 24.8 24.3

Debt stock – centre 54.2 53.9 52.5 50.5 47.5 44.8

Outstanding central loans to states 2.5 2.2 2.0 1.7 1.5 1.3

Source: Thirteenth Finance Commission:2010-15, Table 9.7, p 141.

Table 5: Impact of Tax Information Network on Income Tax Revenue

Years Actual Revenue Revenue Excluding the Percentage of Income Tax from Income Tax Effect of TIN Revenue due to TIN

(1) (2) (3)=100*{(2)-(1)}/(1)

2003-04 1,04,949 94,106.91 10.3

2004-05 1,31,948 1,17,086.7 11.3

2005-06 1,64,906 1,38,701.5 15.9

2006-07 2,29,007 1,67,891.8 26.7

2007-08 2,95,555 1,99,990.2 32.3

2008-09 3,19,441 2,41,143.4 24.5

Source: Estimated from the Regression Equation.

expenditures should be increased by 2.4 percentage points during the period. In the case of the states, those with revenue surpluses in 2009-10 are required to maintain the surpluses and those with deficits are required to eliminate them, such that the consolidated fiscal deficit of the states in 2014-15 is limited to 2.4% of GDP.

The essential logic of the finance commission recommendations is to combine fiscal consolidation with high growth by reducing the fiscal deficit while preserving growth promoting capital expenditure. This is to be accomplished by a gradual elimination of the revenue deficit, i e, a compression of revenue

Economic & Political Weekly

EPW

expenditure.9 There is a long held view that in India public investment crowds in private investment. Supporting evidence on the strong multiplier effects of government capital expenditure compared to current or revenue expenditure is provided by Das (2007) and more recently Guimaraes (2010). The latter estimates the initial multiplier for current government expenditures at 1, which declines to 0.5 after four or five quarters. In contrast the estimated multiplier for capital expenditure is greater than 1 initially and remains so even after 16 quarters. Recent simulation exercises by Mundle, Bhanumurthy and Das (2010) also confirm the strong multiplier effect of government capital expenditure via the crowding in of private investment. Additionally, over the medium-term infrastructure investment can also ease capacity constraints on the supply side in a country like India, where poor infrastructure is often a binding constraint on growth.

The main fiscal challenge in India today is to combine the financ e commission’s emphasis on capital expenditure preserving fiscal compression, i e, compressing revenue expenditure, with the present government’s emphasis on inclusiveness promoting public expenditure on education and health. Public spending on education in 2009-10 relative to GDP was about 3.1% as against the National Common Minimum Programme (NCMP) target of 6% and on healthcare it was 1.2% as against the NCMP target of 3%. This implies that public spending on human development will have to actually increase by 3-4% of GDP in the medium term. Similarly, ensuring food security to all could increase the outlay from the prevailing 0.8% of GDP to about 1.5% of GDP in the medium term. Thus, there would be additional expenditure requirement of at least 5.4% of GDP for human development and social protection by 2014-15. This is in addition to the 2.4% of GDP increase in government capital expenditure recommended by the finance commission. In other words, the expected bill of additional expenditure to achieve high and inclusive growth would add up to about 7.8% of GDP over the medium term.

This expenditure programme has to be somehow made consistent with the fiscal consolidation targets. Fiscal adjustment in 2009-10 was relatively easy for the central government because of the automatic elimination of large one time expenditures of the previous year on payment of arrears following pay revision in government and commitments on loan waivers. There were also significant one time non-tax revenue receipts on account of the first telecom spectrum sale and divestment of public sector equity. This is again true for the current year 2010-11, the base year of the finance commission reference period, thanks to the very large one time receipts from telecom spectrum auction and a buoyant trend in revenue from direct taxes and customs duties. However, for the next four years the fiscal consolidation effort will have to be much more concerted if the finance commission targets are to be met along with the increase in spending necessary to meet the requirements of high and inclusive growth.

Thus, maintaining the growth momentum and ensuring inclusiveness in the growth process would require additional government spending of about 7.8% of GDP over the medium term. In addition, according to the fiscal consolidation plan of the finance commission, compared to 2009-10 the fiscal deficit will have to be reduced by 4 percentage points to GDP by 2014-15. Financing additional expenditures of this magnitude, while at the same time achieving the deficit reduction target set by the finance commission, will require bold measures both on the expenditure side as well as on the revenue side.

The most important reform the government will have to introduce on the expenditure side is to drastically reduce explicit and implicit subsidies. Mundle and Rao (1991) estimated that the unrecovered cost of providing social and economic services amounted to 15% of GDP in 1987-88. Other studies that have followed (Mundle 2007) suggest that this proportion, if anything, may have increased. There is certainly a strong case for revising user charges on non-public good or non-merit services, revising administered prices in line with the market for several public

monopolies, and targeting food and fertiliser subsidies along the lines indicated in the Economic Survey 2009-10. At the state level, a major problem continues to be the burgeoning losses of electricity utilities arising from free and unmetered supply of power to the farmers. This haemorrhage clearly needs to be plugged. However, though economically rational, it is not politically easy to raise substantial resources by levying user charges on non-merit non-public goods and services. Assuming, optimistically, that the government is able to compress unintended subsidie s and raise resources from user charges to the tune of about 3% of GDP, this will require an additional 8.8% of GDP to be mobilised on the revenue side over the medium term to achieve the fiscal consolidation target. The finance commission recommends that the central government should raise about 1% of GDP from disinvestment of public sector equity. If the plan of divesting up to 10% of public enterprises equity is implemented, it may possible to meet this goal. This will still require raising additional resources amounting to about 7.8% of GDP from direct and indirect taxes.

In this context, the two major tax reform initiatives, the enactment of direct taxes code and the introduction of goods and services tax (GST) are extremely important. On the former, with the government withdrawing many of the base broadening measures earlier envisaged, additional revenue mobilisation in the short and medium term may be difficult, though simplification of the laws is likely to reduce arrears from litigations. The most important tax reform for additional resource mobilisation is introduction of the GST. If a broad-based GST is introduced, supported by a good technology platform to track input tax credit and interstate transactions, it may be possible to generate significant additional revenues. It should be noted in this context that a major factor which contributed to significant fiscal consolidation during the period 2003-04 to 2007-08 was the spectacular growth of income tax revenues. On average, income tax revenue increased at over 30% per year. As a proportion of GDP, the revenue from personal and corporation income taxes increased from 3.8% in 2003-04 to 6.5% in 2007-08 (Table 5, p 91). This was partly attributable to general buoyancy of the economy, but a significant proportion was also due to the application of new technology through the introduction of the electronic tax information network (TIN).

The impact of this innovation on tax revenues has been estimated by regressing the revenue from income tax on GDP from non-agricultural incomes in a log-linear regression model over

Figure 2: Impact of Tax Information Network (TIN) on Income Tax Revenue

3,50,000 3,00,000 2,50,000 2,00,000 1,50,000 1,00,000 50,000 0 Without TIN Buoyancy excluding TIN: 1.22 Buoyancy of actual revenue: 1.50 Actual

90 92 94 96 98 00 02 04 0608

the period from 1990-91 to 2009-10 with dummies introduced to capture the impact of TIN. Dummy variables were specified for both the intercept and change in the slope from 2003-04. The regression results are summarised below: Ln (Y) = –6.2323 + 1.2214 Ln (NAGDP) – 3.9290 D1

(–14.27) (38.16) (–3.24)

+ 0.2791 D2*Ln(NAGDP)

(3.40) (Adj R-squared = 0.997 F(3, 16) = 2081.31 Y = Direct tax revenue; NAGDP = Non-agricultural GDP; D1 = Intercept Dummy for technology from 2003-04; D2 = Slope Dummy for technology.

The regression coefficient of the slope dummy is positive and significant. The analysis shows that buoyancy of income tax revenue with respect to non-agricultural GDP increased from 1.22 for the period prior to the introduction of TIN to 1.50 for the period after its introduction (Figure 2). The application of TIN technology led to an increase in the probability of detection, and this significantly increased tax compliance. Not surprisingly, the revenue from income tax as a ratio of GDP increased by 2.8 points in just four years. These results underscore the importance of technology in tax administration. The introduction of GST, if supported by a good technology platform, could raise tax compliance significantly, and perhaps make it possible to increase the tax: GDP ratio by around 3 percentage points.

Even with these reforms in expenditure and revenue there will still be a shortfall in the revenues required to meet the finance commission’s target of reducing the fiscal deficit to 5.4%, raise the allocations relative to GDP for capital expenditure and also for education and health as indicated in the NCMP. There is a trade-off between these three fiscal policy priorities. Perhaps, instead of loading the entire burden of adjustment on only one of these priorities it may be appropriate to allow some adjustment of all three. Settle for a slightly lower level of social sector spending than 9% of GDP, a slightly lower share of capital expenditure than the 2.4% of GDP additional spending proposed by the finance commission, and allow for a slightly higher fiscal deficit than that proposed by the finance commission.

5 Conclusions

There were large variations in the timing and intensity of growth deceleration and the recovery lag of G20 countries following the financial crisis of 2010. This variety of deceleration experience suggests that “a one size fits all” approach to withdrawal of the stimulus would be flawed. In fact, the negative shock of such a simultaneous withdrawal of the stimulus from all G20 countries could seriously raise the risk of double-dip recession. Instead, it is essential that the programme for stimulus withdrawal be tailored to the specific circumstances of each G20 country. As it turns out, barring some outliers, most emerging G20 economies are ready for withdrawal of the stimulus, and in some cases it is even urgent because of high inflation and other factors. In contrast, most advanced G20 countries are not yet ready for stimulus withdrawal. Recovery has been tentative in most of these countries, and some of them are still in recession. At the same time, the existing high levels of public debt, including external sovereign debt, and the debt dynamics of large primary deficits make it difficult to postpone fiscal consolidation without risking unsustainable levels of debt.

To deal with this dilemma, several strategies have been proposed, especially the introduction of reforms in entitlement spending such a pensions and health expenditure that would endogenise future expenditure reversals without immediately withdrawing the stimulus. This discussion of approaches to address the dilemma of the advanced G20 countries, however, is not very relevant to countries like India that are not only ready for withdrawal of the stimulus, but urgently need it to curb over heating and high inflation. Apart from gradual withdrawal of the monetary stimulus, India has also initiated a strategy of fiscal consolidation with high growth as proposed by the Thirteenth Finance Commission. Essentially, it entails phased reduction of the fiscal deficit, while preserving and even enhancing growth promoting capital expenditure.

The main fiscal challenge in India today is to combine this strategy with the NCMP of the present government that focuses on enhanced social spending to make growth more inclusive. Despite strong reform measures on both the expenditure and revenue side, it turns out that a realistic programme of fiscal consolidation with high and inclusive growth may require some downward adjustment in all these three fiscal priorities.

Economic & Political Weekly EPW juLY 16, 2011 vol xlvi no 29

constitutes a stimulus. Technically, on the fiscal side, any increase in public expenditure is a stimulus, even if it is fully financed by additional revenues, because of the balanced budget multiplier. In actual fact stimulus packages in most countries have typically involved large deficits. However, large budget deficits have not always been counted as a part of the stimulus package, as for instance in the case of India’s FY 2008-09 budget cited above. Similarly, on the monetary and financial side, any policy intervention that encourages greater private spending, whether it lowers the cost of funds or raises the value of real balances, is a part of the stimulus but these have not always been officially been recorded as such.

5 For an analysis of cross-country differences in the impact of the financial crisis see Berkmen, Gelos, Rennhack and Walsh (2009).

6 For an analysis and definition of fiscal space in terms of the projected level of public debt based on debt dynamics simulation, compared to the tolerable levels of public debt see Ostry, Ghosh, Kim, and Qureshi (2010). Also see Freedman Kumhof, Laxton, Muir, and Mursula (2009).

7 See the discussion in Section 2 above.

8 Estimated on the basis of the information contained in “Government Debt: Status and Road Ahead”, Ministry of Finance, Government of India, 2010.

9 For the model simulations underlying this logic see Mundle, Bhanumurthy and Das (2010).

References

Barrell, R, I Hurst and S Kirby (2009): “How to Pay for the Crisis or Macroeconomic Implications of Pension Reform”, National Institute of Economic and Social Research Discussion Paper No 333.

Berkmen, P, G Gelos, R Rennhack and J P Walsh (2009): “The Global Financial Crisis: Explaining Cross-Country Differences in the Output Impact”, IMF Working Paper, December.

Corsetti, Giancarlo, André Meier and Gernot J Muller (2009): “Fiscal Stimulus with Spending Reversals”, IMF Working Paper, May and CEPR Discussion Paper 7302.

Minegeshi, M and B Cournede (2010): “Monetary Policy Responses to the Crisis and Exit Strategies”, OECD Economics Department Working Papers No 753, ECO/WKP(2010)9, France.

Mundle, Sudipto (2007): “Government Subsidies” in Kaushik Basu (ed.), The Oxford Companion to Economics in India (New Delhi: Oxford University Press), pp 227-29.

Mundle, S, N R Bhanumurthy and Surajit Das (2010): “Fiscal Consolidation with High Growth: A Policy Simulation Model for India”, Working Paper No 2010-73, NIPFP, August.

Mundle, Sudipto and M Govinda Rao (1991): “Volume and Composition of Government Subsidies in India 1987-88”, Economic & Political Weekly, 26(19), May 4: 1157-72.

Mukherjee, Chandan (2009): “Recent Trends in Indian GDP and Its Components: An Exploratory Analysis”, Economic & Political Weekly, XLIV (41), 10 October, 13-15.

Ostry, J D, A R Ghosh, J I Kim and M S Qureshi (2010): “Fiscal Space”, IMF Staff Position Note, 1 September.

Rakshit, Mihir (2009): “India amidst the Global Crisis”, Economic & Political Weekly, XLIV, 28 March, 94-106.

Zhang, Z and W Zhang (2009): “The Road to Recovery: Fiscal Stimulus, Financial Sector Rehabilitation, and Potential Risks Ahead”, Working Paper 18/2009, Hong Kong Monetary Authority, Bank of International Settlement, 10 December.