Tuesday, 26 May 2020

Loss to Maharashtra’s Economy due to the COVID Lockdown




Mala Lalvani[i] and Ajit Karnik[ii]


As the global pandemic due to the COVID 19 inflicts enormous costs on countries of the world, efforts are being made to compute the loss to the economy so as to devise countervailing measures to mitigate the ill effects. Kapur and Subramanian[iii] (K&S henceforth), estimated that a two month lockdown would imply that about a month’s GDP would be lost due to the lockdown and that government expenditures to the extent of 5% of GDP would be required to make good this loss. CRISIL has estimated that India’s real GDP growth rate will collapse to 1.8%, which is quite similar to the forecasts of the IMF. ICRA expects the 2020-21Q1 GDP to fall by as much as 15% or more. While estimates are available for India as a whole, computations at the level of the states has been missing. This note makes an attempt to estimate the losses for Maharashtra.

We take the K&S approach as a starting point but add substantial detail to our computations. Specifically, where we differ from K&S is in the careful identification of sectors of the economy, which were exempt and not exempt during various phases of the lockdown. This allows us to take a more granular look at the losses as compared to the rather broad brush assumptions of K&S. Our computation is based on the Gross State Value Added (GSVA) which we employed to separate out the exempt and non-exempt activities. Once we had an estimate of the total contribution of exempted activities, we estimated the losses associated with activities that could not function (i.e. the non-exempt activities). It may be mentioned that we carried out these computation separately for the three phases of the lockdown.

Table 1 below provides a list of activities that were fully exempt and partially exempt during the various phases of the lockdown.

 
Table 1: Broad Categories of Fully Exempt and Partially Exempt Activities




While computing the GVSA of the fully exempt activities is straight forward, doing so for the partially exempt activities required some assumptions. We turn to these assumptions now:

(i) Manufacturing: As stated in Table 1, manufacturing related to food processing, medical instruments, and pharmaceuticals were exempt. The Annual Survey of Industries (ASI) data for 2017-18 was used to assess the share of such industries and it was observed that 42% of GSVA was attributable to these industries. The list of 3-digit industries in ASI which we considered as exempt included among others:  (i) support activities to agriculture and post-harvest crop activities; processing and preserving of meat, fish, fruit etc.; manufacture of vegetable and animal oils and fats, dairy products, starches, animal feeds, etc.; manufacture of textiles, refined petroleum products, basic chemicals, fertilizers, pharmaceuticals, medical and dental instruments; waste collection; waste treatment and disposal.

(ii) Hotel and Restaurants: We have seen that motels and Hotels which accommodated tourists who were stranded were exempt. Tourism is estimated to constitute 9.4% of GDP and, hence, this proportion of Hotels and Restaurants were seen to be exempt.

(iii) Public Administration: Police and Public Works together were found to constitute 72.6% of the budget of Home Department and General Admin Dept. This proportion of GSVA of Public Admin was considered as exempt.

Based on the above details, we obtained the aggregate amount of GSVA for ‘’fully exempted activities’’ and also GSVA of ‘’partially exempted activities’’.  Aggregating these gave us the GSVA associated with the ‘’totally exempted’’ industries.  We estimated this separately for Lockdown 1, Lockdown 2 and Lockdown 3. Using the proportions of exempt and non-exempt activities in the GVSA, we computed the losses as a percentage of GSDP. Table 2 gives these details.


Table 2: GSDP Losses during Various Phases of Lockdown




Having obtained the share of Non-exempt activities, the next step was to obtain the resources required, for which three scenarios were considered.  This would be a sum of two components: (a) share of the resources lost which could be compensated by way of support (we experimented with 25%, 33% and 50% levels of support) and (b) an additional 0.5% of GSDP which the state would need to spend on Public Health on account of the crisis situation.  K&S had assumed 2% of GDP being spent on Health. However, given that Maharashtra has been spending 0.5% of GSDP on Public Health and there are capacity constraints, we have been conservative and assumed that the State spending on health would be double the share of GSDP, hence we consider an additional 0.5% of GSDP when computing the resource requirement. The three scenarios we considered are as follows:

Scenario I: 25% of the losses of household income would have to be made good and an additional expenditure of 0.5% of GSDP was to be incurred on Public Health. In this case Rs. 83,533.52 crore would be required i.e. 2.9% of GSDP

Scenario II: 33% of the losses of household income would have to be made good and an additional expenditure of 0.5% of GSDP was to be incurred on Public Health. In this case Rs. 1,05,658.52 crore would be required i.e. 3.67% of GSDP

Scenario III: 50% of the losses of household income would have to be made good and an additional expenditure of 0.5% of GSDP was to be incurred on Public Health. In this case Rs. 1,52,674.13 crore would be required i.e. 5.3% of GSDP

Some of the options available to the State to make good the losses and raise resources are:

1. The Central government has permitted States to increase their borrowings up to 3.5% of GSDP without conditions and up to 5% with conditions. This will allow Maharashtra to borrow additional amounts, of course, bearing in mind the fact that future interest payments will rise and constrain future fiscal space. However, funds available through deficits are difficult to predict since revenue collections from taxes is likely to be adversely affected, which by itself will raise deficits.

2. Maharashtra's share in the State Disaster Response Fund will offer some resources.

3. Maharashtra can expect to get additional funding from the increased allocation of Rs. 40,000 crore to MGNREGA announced by the Central government. Earlier the allocation was Rs. 61,000 crores.

4.  Outlays on newly initiated capital projects, as announced in Maharashtra’s budget for 2020-21, can be postponed. For example, some of the new projects like the Konkan Marine Highway and the new metro line for Pune-Pimpri-Chinchwad could be shelved for the moment and resources re-directed to meet the crisis. 

******


[i] Mala Lalvani, Professor, Mumbai School of Economics and Public Policy, University of Mumbai mala.lalvani@gmail.com
[ii] Ajit Karnik, Professor, Middlesex University, Dubai ajit.karnik@gmail.com

Tuesday, 5 May 2020

SIX WEEKS OF LOCKDOWN IN INDIA: HAS COVID-19 BEEN TAMED?


INTRODUCTION

It’s been almost six weeks since the beginning of the lockdown in India, which began on 24 March 2019. As governments in India, both Central and States, begin the process of re-opening the country, it is time to evaluate whether the primary objective of the lockdown – slowing down of the number of confirmed COVID cases and COVID deaths – has been achieved. This is, of course, the so-called flattening of the curve that has been much spoken of during the last few months. In order to get a sense of how well or badly India has done, I will compare India with three countries, which have also experienced lockdowns of their own, more or less at the same time that India had its lockdown. These three countries are Italy, Spain and the United Kingdom (UK). The start dates of lockdowns in these countries are as follows:

Italy: 10 March 2020
Spain: 15 March 2020
UK: 23 March

In order to carry out a comparison, I label the date of the commencement of lockdown in each country as Day 0 and each day subsequently as Day 1, Day2 and so on. The time period considered for India is from 24 March till 5 May 2020. This yields 41 days of data for India and, for the other countries, I use six weeks of data from the beginning of their respective lockdowns. 

TOTAL CASES AND TOTAL DEATHS

The countries with which India is being compared have been ravaged to a far greater level than India has been. Figures 1 and 2 show these details. The problem with Figures 1 and 2 is that the difference in levels of cases and deaths in India and in the other three countries flattens the Indian curve relative to the other curves. This is especially so in Figure 2.



Despite the Indian curves being flattened excessively, we can draw some comparisons. On Day 0, Indian cases were just 5.8% of those in Italy, 6.2% of those in Spain and 7.6% of those in the UK. As far as deaths are concerned, the corresponding numbers were 2.4%, 3.8% and 2.9% of those in Italy, Spain and the UK respectively. By Day 41, the figures for cases were 25.6%, 22.6% and 24.9% of those in Italy, Spain and the UK while the figures for deaths were 6.5%, 6.8% and 5.5% of those in Italy, Spain and the UK. Clearly, the ratio of India’s cases and deaths to those of the other three countries has been growing over the last weeks. However, neither these numbers nor Figures 1 and 2 give any indication of whether the pace of the spread of COVID 19 is slowing down in India as compared to the three countries.


It has been suggested that taking logarithmic transformations of the number of cases and deaths helps clarify the trends better than do actual values. Figures 3 and 4 report the same data as in Figures 1 and 2 but using log values.




Taking logs does lift the Indian lines above the horizontal axis and shows that there is still a large gap between the situation of COVID 19 cases and deaths in India as compared to the other countries. But we already knew that. What we wish to see is whether the pattern of the progress of the disease in India is similar to the patterns in Italy, Spain and the UK. 

In order to eliminate the effect of these huge differences in numbers between India and other countries, I transformed all the data to z-scores. A z-score is defined as the value of cases on a given day minus the average number of cases for the period divided by the standard deviation and likewise for deaths. “For a standardized variable, each case’s value of the standardized variable indicates its difference from the mean of the original variable in number of standard deviations (of the original variable). For example, a value of 0.5 indicates that the value for that case is half a standard deviation above the mean, while a value of -2 indicates that a case has a value two standard deviations lower than the mean”.[i] Having done this, I report Figures 5 and 6.




For both, cases and deaths, the Indian line shows a worrying trend. While the other countries clearly show a flattening of their respective curves, Indian cases and deaths continue to show an upward trend. It is disconcerting that despite having suffered massive pain and incurred huge economic costs, India has not yet managed to exert control over the disease. It is true that the number of cases and deaths in India are substantially lower than in other countries but it is important to remember that India’s health services will be stretched to breaking point at far lower numbers than in other countries.

GROWTH RATES AND DOUBLING RATES

As a final basis of comparison, I look at the rates of growth of cases and deaths (Figures 7 and 8) and their doubling rates (in number of days) (Figures 9 and 10). It may be noted that for Week 6, the Indian growth rate has been averaged over only 6 days since data for the 42nd day of lockdown is not yet available.



The average daily growth rates have been computed based on weekly totals. The rate of growth for Week 1 is computed over the total for Week 0 which was taken to be the week that ended on the day lockdown was announced. For India, the average daily rate of growth of cases has fallen from 16.7% in Week 1 to less than one-fourth that level in Week 6. However, the fall has been much steeper in the other countries. A similar pattern is visible in the average daily growth rate of deaths.

Even in terms of doubling rates of cases and deaths, India has shown good progress but the other countries have shown a much greater increase in the doubling rates. See Figures 9 and 10.



CONCLUSION

In conclusion, we can see that India has managed to control the spread of the disease as measured by the number of confirmed cases and the number of deaths. However, concerns with respect to testing remain. Given India’s huge population, the number of tests performed remains very low. The ICMR has announced that it has performed over a million tests.[ii] However, there needs to be clarity about exactly what is being measured when ICMR reports a million tests: is it the number of tests performed or the number of persons tested? This is important because a COVID positive person is tested numerous times as his or her treatment progresses before he or she is finally declared free of the disease. Naturally, the number of tests performed will be greater than the number of persons tested.[iii] Table 1 below gives the number of tests performed per million.

Table 1: COVID Testing


Country
Number of Tests per million Population
India
758
Italy
36244
Spain
28898
UK
13925

Clearly, the rate of testing in India is very low and it is quite possible that the results shown above may undergo a change for the worse in the weeks to come. 

-------------------------------------------------------------


Thursday, 26 September 2019

Supply-side solutions for Demand-side Problems


Introduction

The flurry of reforms unleashed in the last few weeks suggests that the Government has finally accepted that the Indian economy is in trouble. The changes have been so dramatic that the Economic Times called it the “3rd Budget in 8 months”.[i] Of course, there is a disconnect between what the government says and what it does: even as the Finance Minister Nirmala Sitharaman announced changes to the FDI policy in August, she claimed that the Indian economy was doing very well.[ii] Never mind. The ruling BJP has its compulsions: it needs to keep its flock together and, hence, needs to paint a rosy picture of how well the economy is doing.

For months, economists and other responsible analysts had been pointing to the alarming state of the economy. Despite all the misgivings about the GDP especially after the base change and after numerous revisions,[iii] I use quarterly growth rates of GDP in Figure 1 below to capture in summary form the problems facing the economy.




The growth rate in 2019-20Q1 at 4.89% is the worst that the economy has performed since 2012-13Q4 when it was 4.21%. The BJP might still find solace in the fact that its worst performance is still better than the worst performance during the UPA rule. However, just the fact that one has to search for the worst performance of the UPA to make the latest growth rate look good shows the magnitude of the problem.  In addition to the falling growth rate, analysts have drawn attention to other grave problems: (a) unemployment[iv] [v] (b) stagnant rural incomes[vi] (c) falling consumption[vii] (d) falling investments.[viii] The combination of stagnant rural incomes, stagnant consumption and investment levels coupled with low inflation suggests that the primary cause of the slowdown in India is inadequate demand.[ix] Combating such a slowdown in the economy requires demand-side policies, such as, increases in government expenditure. This might well have an adverse effect on the quantum of fiscal deficit but, if the economy revives and GDP increases at a faster rate than it currently is, the fiscal deficit to GDP ratio may not experience too large an increase.

Flurry of Reforms
In response to the slowing down of the economy the government has announced a series of measures:
  1. Merger of banks, which has reduced the number of public sector banks from 27 to 12.[x]
  2.  Foreign Direct Investment: On 28 August 2019, the Finance Minister announced changes in the FDI policy which would now permit single-brand retail to commence online retail trading prior to opening of physical stores provided physical stores begin operations within two years of commencement on online operations.[xi] This is a major change since single brand retail FDI has been a “political hot potato” which no government had embraced so far.[xii]
  3.  Loan Melas: The Finance Minister has urged public sector banks to hold “loan melas" (loans festival) in 400 districts across the country.[xiii]
  4. Tax Rates: The biggest announcement has been the reduction in corporate tax rate from a highest rate of almost 35% to about 25%.[xiv]

The important question is whether there is a correspondence between the problems facing the Indian economy and the policy measures that have been announced. As stated above, the problem facing the economy is one of inadequacy of demand. Do the policies that have been announced address the problem of demand deficiency? I have my doubts especially because the measures announced are in the nature of supply-side policies.

Supply-side policies

Supply side policies are designed to increase the potential output of an economy.[xv] In general, such policies affect an economy’s long run economic growth. Supply side policies typically involve reduced government expenditure, tax cuts for labour and business and policies designed to increase competition. Of course, the approach to supply-side policy has to be nuanced; for example, which type of government expenditure is reduced matters.

Considering the various measures that have been introduced in the recent past one could identify the following:
  1. Merger of banks: supply side measure to increase competitiveness
  2. Foreign Direct Investment: supply side measure to increase competitiveness
  3. “Loan melas": assuming that loans are actually given out, this is likely a demand side measure
  4. Cut in corporate tax rates: standard supply side policy

Barring “loan melas” all the others are supply side policies which are expected to increase long run economic growth. No doubt this is important but does it address the problems that the economy is facing? The hope of the supply side policies – especially FDI and tax cuts – that have been introduced is that foreign and domestic investment will revive which will lead to more employment and which will result in higher production which in turn will lead to more investment, more employment and so on in a virtuous cycle. The question is whether investment will work as an autonomous driver in a demand deficient economy. If producers are not confident about selling their products – remember the problems being faced by Parle-G biscuits as well as other FMCG products[xvi] and Maruti Suzuki cars?[xvii] – would they take the risk of investing? The tax cuts that have been announced will have immediate beneficial effect only if the money is spent on additional investment. In the absence of that, it is merely a bonanza for companies and huge strain on the public finances.  

Fiscal Multipliers

A change in fiscal policy (change in government expenditures or change in taxes) has an impact on GDP. The size of this impact is measured by the multiplier.[xviii] It may be noted that a change in taxes or a change in expenditure has an impact on the fiscal deficit which is defined (in simple terms) as:

Fiscal Deficit = Government Expenditure – Government Revenues

Expenditures include both capital and current account expenditures while government revenues include tax revenues, non-tax revenues and other non-debt-creating revenues (e.g. disinvestment proceeds).

The recent reduction in corporate taxes is expected to reduce tax revenues by Rs. 1.45 lakh crore. Keeping government expenditures constant, fiscal deficit will rise by the same amount. However, if the government wishes to keep fiscal deficit constant, expenditures must be reduced by the same amount that taxes have gone down. It may be noted that a reduction in taxes and increase in expenditures are said to be expansionary i.e. they are designed to expand the economy. The question is, by how much? This is computed by reference to the multiplier. Denoting government expenditures as G and tax revenues as T, the relationship between GDP and G or GDP and T is given by:

  1. Change in GDP = (expenditure multiplier)*(change in G)
  2. Change in GDP = (tax multiplier)*(change in T)


It can be shown that the tax multiplier is smaller than the expenditure multiplier. Hence, for a unit change in expenditure, the change in GDP will be greater than for a unit change in taxes. Bose and Bhanumurthy of NIPFP have estimated that the multiplier associated with capital expenditure is as high as 2.45. [xix] If instead of cutting taxes to the extent of Rs. 1.45 lakh crore, the same amount had been spent of capital expenditure, GDP would have risen by Rs. 3.55 lakh crore. Bose and Bhanumurthy estimates for the multiplier associated with the corporate tax indicate that for the Rs. 1.45 lakh crore cut in taxes, GDP will increase by the same amount. Clearly, you get more bang for the bucks with a unit change (i.e. increase) in expenditures than for a unit change (i.e. decrease) in taxes.

Conclusions

In conclusion, I will summarise my evaluation of the recent policy changes:

  1. The problems facing the Indian economy have arisen due to a deficiency of demand and, hence, solving these required that demand-side policies be enacted
  2. Unfortunately, the government has enacted supply-side policies which will be beneficial in the long run by raising India’s potential output. However, it may not do anything to reverse the slowdown in the economy.
  3. Reducing taxes as an expansionary measure does not have as much of an impact as raising government expenditures due to the much smaller multiplier effect.
  4. If government expenditures are reduced to keep the fiscal deficit at its announced level, the economy will face a further problems. Note that the expenditure multiplier works in reverse as well and a reduction in expenditures will reduce GDP. In the current context, if expenditures are cut by Rs. 1.45 lakh crore, GDP will fall by Rs. 3.55 lakh crore. Coupling this with the increase in GDP due to the tax cut, the net effect on GDP will be a fall of Rs. 2.1025 lakh crore. This is bound to worsen the economic situation in India.


[iii] Do listen to this very informative podcast: http://www.seenunseen.in/episodes/2019/7/22/episode-130-demystifying-gdp