Every undergraduate student of economics will be exposed to the concept of a multiplier. Indeed, the use of the term has become part of standard jargon. A multiplier measures the factor by which one variable changes with respect to another. That sounds vague. There are multipliers and multipliers. Hence, let’s impart some precision to that vague statement by applying it to a specific type of multiplier: The fiscal multiplier. National income results from consumption, private investments, government expenditure or exports. An increase in expenditure on any of these four leads to an increase in national income. By how much is the question. If the increase in national income is more than the initial increase in expenditure, there is a multiplier or magnifying effect. That is, the multiplier is more than one. It is by no means axiomatic that the multiplier will be more than one. It can conceivably be less. When the multiplier is more than one, there is a chain reaction at work. The government builds national highways. Contractors are hired. There is employment. Wages/ salaries are spent. There is greater consumption, stimulating demand. When these are aggregated, the total increase in national income is greater than the initial spending by the government. This is an example of a fiscal multiplier, relating to government expenditure.
When an economy is not doing well and the government seeks to stimulate growth and employment, an expansionary fiscal policy comes into play. What is the value of the fiscal multiplier? Is it greater than one or less? What form should government expenditure take? Is the fiscal stimulus best ensured by increasing government expenditure or reducing taxes? Leading up to the Union Budget for 2022–23, many people expected tax cuts. There was none. Instead, there was an increase in government expenditure, of the capital investment and infrastructure variety. Were the expectations better-informed, or were government responses better-informed, judged in terms of multiplier effects? Is a government expenditure multiplier higher, or is a tax (reduction) multiplier higher? While one can build theoretical models, which are contingent on assumptions, the only satisfactory answer is to look at empirical estimations. The gut instinct of any economist will be that the government expenditure multiplier will be greater than a tax multiplier. That’s what models and empirical results both suggest. For instance, pre-pandemic, figures floated around when there was talk of a fiscal stimulus in the US in 2009. Broadly, tax rebate and temporary tax-cut multipliers were around 1.00, permanent tax-cut multipliers were around 0.5 and government expenditure multipliers were around 1.5, vindicating the economist’s gut instinct.
What of India? In 2013, Sukanya Bose and N R Bhanumurthy from NIPFP (National Institute of Public Finance and Policy) authored a paper on fiscal multipliers for India. To quote, “Whereas direct spending by the government fuels public consumption expenditure, transfers by the government result in higher disposable income for the private sector, particularly households, in turn, leading to higher private consumption expenditure. The dynamics of three kinds of expenditures (capital expenditure, transfer payments by the government to the rest of the economy and other revenue expenditure) in the economy are thus very different, and can be expected to generate different multiplier effects on the output.” They then used a NIPFP model and data for 1991–2012 to carry out some estimations. According to this paper, the capital expenditure multiplier is 2.45, the transfer payments multiplier is 0.98 and the “other revenue expenditure” multiplier is 0.99. What’s the difference between the last two? “Expenditure on the revenue account includes direct spending by the government on goods, services and wage payments, as well as transfer payments to the rest of the economy. Transfer payments involve transfers of purchasing power from the government to the rest of the economy and include major subsidies, pensions and other retirement benefits, relief on account of natural calamities, etc. Any increase in governments’ revenue expenditure can be either due to higher government consumption expenditure or on account of an increase in the transfers to the rest of the economy, or a combination of the two.” The tax multipliers were around 1, irrespective of whether it was GST, personal income tax or corporate tax. Sure, these numbers are for 1991–2012 and are therefore slightly dated. But numbers don’t change that fast, even if there is a pandemic. In other words, if you were the FM and had read the NIPFP paper, you would have done exactly what she has done—not reduce taxes and increase capital expenditure instead. In the long term, the capital expenditure multiplier is even greater.
There are some later figures from December 2020, when the RBI did a half-yearly review of government finances. (The original RBI estimates are from 2019.) To quote again, “In the Indian context, it is important to note that revenue expenditure multipliers for the government are less than unity. … In contrast, the capital expenditure multiplier is well above unity for the Central and state governments—for the Central government, it is 2.45 in period t and 3.14 in period t+1, while for state governments it is close to 2 .” The same endorsement: Capital expenditure is preferable to revenue expenditure. In addition, capital expenditure by the Union is superior to that by state governments from the multiplier perspective. Despite this, why do we prefer tax cuts? The primary reason is myopia. Tax cuts and their perceived benefits are immediately visible. The multiplier is less visible, working through a chain reaction. Myopia chains our perceptions.
(Views are personal)
Bibek Debroy
Chairman, Economic Advisory Council to the PM
(Tweets @bibekdebroy)