Page 83 - ES 2020-21_Volume-1-2 [28-01-21]

66 Economic Survey 2020-21 Volume 1

income growth. Sandri (2014) examines 62 episodes of growth spurts from 1960 to 2011
among non-OECD countries and shows that productivity growth across these episodes is
combined with not only a rapidly rising investment rate but an even more steeply increasing
savings rate. Carroll and Weil (1994), Attanasio, Picci and Scorcu (2000) and Rodrik (2000)
show that savings and growth are not only positively correlated but their positive correlation
is even stronger than that between growth and investment. Using a VAR framework, Kulkarni
and Erickson (1995) find no statistically significant evidence of crowding out in India. Due
to dynamic interdependencies between public investment and GDP, the literature has also
resorted to using vector auto-regressions (VARs) to estimate the crowding out phenomenon.
Mitra (2005) uses a structural VAR, and finds evidence that the impact of public investment
on crowding out of private investment is less than one for one. These results broadly support
the static, unconditional estimates provided below.

2.26 For a country such as India with an extremely young population, the role of demographics
in fostering savings becomes crucial to understand possible crowding out due to government
spending. Bosworth and Chowdorow-Reich (2007) show for Asia that both savings and
investment rise with the proportion of the working population. Curtis, Lugauer and Mark
(2011) find that jobs that pay meaningful wages drive savings rate in the economy. Lee,
Mason and Miller (2000) and Bloom et al. (2007) show that savings increases as average life
expectancy increases in a country. Thus, in an economy operating below full capacity, the
supply of savings may grow from greater government spending through demand creation and
thereby greater employment. This is because, as highlighted by recent research, favourable
demographics – in the form of a large population of working age – would enhance savings
through meaningful jobs.
2.27 Consistent with these arguments against crowding out, studies find no evidence of
crowding out of private investment due to public investment for developing economies.
Erden and Holcombe (2005) analyse the public and private investment in developing and
developed economies, and conclude that while public investment is complementary to
private investment in developing countries, the opposite holds for developed countries.
Eisner (1994) argues that whether an increase in Government expenditure for goods and
services ‘crowds out’ domestic private investment, may depend upon how close the economy
is to full employment. Bahal et al. (2015) find no evidence of crowding out in India over the
period 1980-2012.

2.28 We analyse the relationship between changes in public investment and changes in private
investment for the period FY 1991- FY 2019 and find the correlation to be insignificant (Figure
13b). Thus, consistent with the results in Bahal et al. (2015), we find no evidence of crowding
out over the last three decades post liberalization. However, during the pre-liberalisation period
of FY 1951-FY 1990, a negative correlation between changes in public investment and changes
in private investment provides evidence consistent with the rationale of fixed loanable funds and
possible crowding out (Figure 13a).