On the relationship between public and private spending in developing and developed countries

Abstract

Using annual data, the paper studies the time-series evidence regarding the effectiveness of government spending. The emphasis is on the relationship between public spending and private spending. The objective is to identify whether the effects of public spending on macro variables are reinforced or mitigated through the spillover effects on private spending. The evidence attests to the importance of stimulating private spending to maximize the positive effect of an increase in public spending on real growth. Concerns about the crowding out effects of higher public spending on private demand are more dominant in developing countries. Moreover, the scope for government spending to determine aggregate uncertainty is much larger in developing countries. Overall, the evidence attests to the importance of managing trends and variability of government spending towards maximizing the fiscal multiplier. The paper's evidence spells out potential to maximize the fiscal multiplier via private spending and concerns about the ineffectiveness of fiscal policy where crowding out concerns dominate.

Keywords:

• fiscal stimulus
• crowding out
• supply constraints
• developing and advanced countries

JEL Classifications:

• E61
• E62
• E63
• E32
• E21

Acknowledgements

The author wishes to thank the Editor and Associate Editor for helpful comments on previous revisions.

Disclosure statement

The views in the paper are those of the author and should not be interpreted as those of the CBUAE or its policy.

Notes

1. Indeed, the unfolding debt crisis in Europe provides a clear illustration of how the increase in government spending financed by higher cost of borrowing results in a ballooning deficit that demands ever increasing risk premium. The end result is unsustainable public finances that undermine the effectiveness of government spending to stimulate growth and crowds out private activity.

2. This explanation was advocated in view of the evidence of expansionary fiscal contractions, see, e.g., Giavazzi and Pagano (1990), and Alesina and Perotti (1995). For evidence of asymmetry in interest-rate adjustment to government spending shocks, see Kandil (2001).

3. See, e.g., Barro (1989): “The substitution of a budget deficit for current taxes has no impact on the aggregate demand for goods. In this sense, budget deficit and taxation have equivalent effects on the economy – hence the term “Ricardian equivalence theorem.” For a coherent theoretical illustration of the equivalence theorem, see Barro (1974).

4. For more details, see, e.g., Feldstein (1976). Tax payers may face binding liquidity constraints if consumption absorbs disposable income entirely. With these constraints, households are likely to increase savings in anticipation of future tax liability. In contrast, households are constrained from increasing consumption (decreasing savings) in anticipation of future tax reduction.

5. See, e.g., Kandil (2002b).

6. See, e.g., Ball, Mankiw, and Romer (1988).

7. See, e.g., Ball and Mankiw (1994).

8. The results, following the suggestions of Nelson and Plosser (1982), verify the empirical validity of non-stationarity. Based on tabulation provided by Dickey and Fuller (1981), the dependent variables in the empirical model are non-stationary in level and stationary in first-difference. Nonetheless, there is no evidence of joint co-integration between the non-stationary dependent variable and explanatory variables that include decomposed variables into anticipated and unanticipated components. Hence, the empirical model does not account for an error correction term on the right-hand side of the equation.

9. For a detailed theoretical illustration, see Kandil and Mirzaie (2002).

10. Quarterly data are not adequately available to estimate the model using distributed lags.

11. The time-series evidence is available upon request.

12. Due to space limitations, correlation coefficients in Tables A3 and A4 are for the time-series responses to government-consumption and private consumption only. Other details are available upon request.