An autonomous research institute under the Ministry of Finance

 

(Coauthored with Lekha Chakraborty) 
 
Often, fiscal stimulus is launched through the tax side than expenditure side, assuming that the buoyancy of the former will ensure minimum fiscal slippage, while shoving the economy out of a glut. The general idea is that a reduction in rates will increase the tax base and compliance.  The fiscal stimulus programme announced by finance minister Nirmala Sitharaman is also premised on a similar idea. An IMF working paper titled ‘How Buoyant is the Tax System? New Evidence from a Large Heterogeneous Panel’ by Paolo Dudine and Joao Tovar Jalles, published in 2017, finds that tax buoyancies are generally equal to unity or greater for developed as well as for less developed economies. 
 
In our economy, the tax-to-GDP ratio has hovered around 14-17% for the last few decades, which is the combined figure for the Union and states. Direct and indirect taxes contribute almost equally to the total tax revenue, although the share of direct taxes is slightly higher at 52% during 2017-18. The Union collects about 10% of GDP as tax revenue and the rest is by all the states together. The finance minister’s stimulus package is premised on the buoyancy of these taxes. Hence, it is imperative to look at the tax buoyancy factor both at the Union and the state level during the recent past.
 
Tax buoyancy measures the response of tax revenue to a change in national income and the tax policy. Economists generally define it as the ratio of percentage change in tax revenue to a percentage change in income. Buoyancy can be estimated for the long-term as well as for the short-term. Short-term buoyancy above unity signifies that the tax system acts as an automatic stabiliser.  In other words, the short-run buoyancy measures the instantaneous effect of a change in GDP on the tax revenue.
 
Long-run buoyancy is important in gauging the impact of long-run growth of the economy on fiscal sustainability. Long-run buoyancy above unity would mean that faster growth would lead to better fiscal balance through the revenue side. This would be an important guiding principle while considering countercyclical fiscal measures, meaning an increased fiscal deficit would trigger growth, which can, in turn, generate more tax revenue, leading to the easing of fiscal pressure.
 
The Auto Regressive Distributed Lag (ARDL) model allows us to estimate the long-run and short-run buoyancy, along with the speed of adjustment—which tells us how fast the buoyancy converges to the long-run equilibrium value.
 
BUOYANCY OF TOTAL OWN-TAX REVENUE BY STATES AND GROSS TAX REVENUE OF UNION
 

State

Long-Run Buoyancy

Short-Run Buoyancy

Speed of Adjustment

 

<1

1

>1

<1

1

>1

 

Andhra Pradesh

 

 

1.17***

 

1.12***

 

-0.88**

Bihar

 

 

1.28***

0.29

 

 

    -

Chhattisgarh

 

 

1.09***

 

0.94***

 

-0.92***

Goa

 

0.95***

 

0.08

 

 

-0.26*

Gujarat

 

1.09***

 

 

2.4***

 

-0.09

Haryana

 

0.85***

 

0.04

 

 

-0.34*

Himachal Pradesh

 

 

1.17***

 

1.3**

 

-0.98***

Jammu & Kashmir

 

 

1.28***

 

1.1*

 

-0.51*

Jharkand

 

 

1.24***

0.54***

 

 

-0.93***

Karnataka

 

1.04***

 

 

1.28***

 

          -0.5*

Kerala

 

 

1.06***

 

1.59***

 

    -

Madhya Pradesh

 

1.15***

 

 

0.5

 

-0.58*

Maharashtra

 

0.99***

 

 

0.97***

 

-0.52*

Odisha

 

1.08***

 

0.29

     

 

-0.43**

Punjab

 

1.03***

 

 

-0.63

 

-0.59*

Rajasthan

 

0.99***

 

 

0.52***

 

-0.33**

Tamil Nadu

 

0.94***

 

 

1.19***

 

-0.46**

Uttar Pradesh

 

 

1.17***

 

1.75***

 

-0.88***

West Bengal

 

1.12***

 

 

0.66

 

-0.52**

 

 

 

 

 

 

 

 

Arunachal Pradesh

 

 

1.39***

 

0.77*

 

-0.51**

Assam

 

1.07***

 

 

1.24**

 

-0.42***

Manipur

 

 

1.58***

 

2.04**

 

-0.68**

Meghalaya

 

 

1.36***

 

0.83**

 

-0.41*

Mizoram

 

 

1.42***

 

-

 

-0.19**

Nagaland

 

1.24***

 

 

0.81

 

-0.25

Sikkim

 

0.85***

 

0.17

 

 

-0.35*

Thripura

 

 

1.33***

 

0.24

 

-0.41*

 

 

 

 

 

 

 

 

Union

 

1.05***

 

 

2.05***

 

-0.47**
*** p<0.01, ** p<0.05 and * p<0.1
GDP and GSDP data are from RBI database
Tax revenue data are from NIPFP database of Finance Accounts. 
 
The estimates for the period 2001-17 show that the long-run and short-run buoyancy are 1.05 and 1.74, respectively, for total tax (the Union and states combined).  The slowdown will have a heavy impact on the Union tax revenue.  This will also have a deleterious effect on the fiscal health of states as the shareable kitty will shrink substantially. Now, with the 15th Finance Commission (FC) asked to consider the impact of the award of 14th FC on Union finances, any fall in the share of states would adversely affect state finances. Short-run buoyancy is found to be either equal to or less than unity for all states. Bihar, Goa, Haryana, Jharkhand, Odisha and Sikkim will be the ones that would be least affected in the short-run, with a buoyancy factor less than unity. For the long-term, all states have buoyancies either equal to unity or greater than unity. Goa, Gujarat, Haryana, Karnataka, Madhya Pradesh, Maharashtra, Odisha, Punjab, Rajasthan, Tamil Nadu, West Bengal, Assam, Nagaland and Sikkim have long-run buoyancy equal to one, making them less vulnerable in the long-run. Interestingly, most of the richer states fall in this category.
 
The packages announced by the finance minister so far mostly target the supply side. It will take a while to yield results by way of turning around growth. 
 
Having seen these premises and the estimates of tax buoyancy, what policy options do we have now to arrest the slowdown, revive the economy and moderate the fiscal slippage? With the general consensus that a fall in aggregate demand is the main culprit, steps can be initiated to shore up aggregate demand. These interventions can be on both revenue and expenditure sides. On the revenue side, a reduction in taxes that will benefit the relatively poorer sections and rationalisation of GST will definitely have a high multiplier effect. Expenditure on infrastructure and upscaling programmes like MGNREGA will also have a higher multiplier effect, leading to revival of growth.
 
More detailed analysis of buoyancies of individual taxes including GST (where we have only a short-time series) is essential, which is a subject matter for our future research. Although we have incorporated the optimal parameterisation in the models by choosing the apt lag lengths, the estimates can be refined further by incorporating variables like inflation, structural variables, political factors and business cycles in the tax buoyancy estimation models. At a disaggregated level analysis, it is also important to see whether the buoyancy of divisible pool taxes is greater than states’ own taxes. Along with these, an understanding of how tax buoyancies behave in different phases of business cycle (output gap) will throw more light on the effectiveness of such polices.

The authors are Professor, NIPFP, and Doctoral  Fellow, CESP, JNU, New Delhi 

 
The views expressed in the post are those of the author only. No responsibility for them should be attributed to NIPFP.