Skip to content

The Cyclicality of Monetary and Fiscal Policy in South Africa since 1994

Literature › Works

Stan du Plessis; Ben Smit; Federico Sturzenegger (2008). The Cyclicality of Monetary and Fiscal Policy in South Africa since 1994. CID Working Paper No. 163, Center for International Development, Harvard University. Center for International Development, Harvard University.
Type working paper
Scope Post-1994 South African economy
Method econometric
Period studied 1960–2006
Regions South Africa
Themes macro fiscal finance debt, transition post apartheid, methods data sources
Citation key plessis2008b

Argument

The paper estimates a three-variable structural VAR with long-run restrictions (Blanchard-Quah tradition). It identifies supply shocks and separate fiscal and monetary policy shocks, and uses them to judge whether policy has been pro- or counter-cyclical since 1994. It argues that earlier South African studies depended on contestable base years, elasticities and cycle measures. Monetary policy is found to be largely counter-cyclical since 1994, with pro-cyclicality only at the very end of the sample. Fiscal policy shows pro-cyclicality, mainly after 2002, but counterfactual simulations suggest only a small destabilising effect on output.

Evidence and method

Seasonally adjusted quarterly SARB series, 1960Q2-2006Q4: real GDP growth, government consumption to GDP, and a real interest rate built from the discount rate and CPI. Also uses an IFS cross-country comparison of inflation and output variability, plus results from earlier studies (structural budget balances, fiscal impulse measures, Du Plessis and Boshoff 2007). Estimated on a long sample (1960+) and a shorter one (1983+).

Findings

  • The authors identify a post-1994 'great moderation' in stability, with a favourable position relative to other emerging markets. (p. 5)

    Indeed a ‘great moderation’ – as the improvement in economic stability has been called in the literature on developed economies

  • Earlier fiscal-cyclicality studies disagree because they depend on contestable assumptions, and this motivates the SVAR. (p. 18)

    Many of the results were heavily dependent on contestable assumptions about base years, elasticities, measures of the business cycle, a narrow view of the dynamics of macroeconomic relationships

  • Du Plessis and Boshoff's revenue elasticities are more than twice Swanepoel and Schoeman's, so automatic stabilisers are stronger than previously thought. (p. 10)

    the output elasticities of government revenue as estimated by Du Plessis and Boshoff (2007) are more than twice as high as the estimates by Swanepoel and Schoeman (2003)

  • Fiscal policy shocks show little counter-cyclical impact on GDP, being counter-cyclical in 1993-96 but pro-cyclical in 1998-99 and during the recent expansion. (p. 24)

    since 1994 fiscal policy moved counter-cyclically during the expansion of 1993 to 1996 but pro-cyclically during the subsequent contraction of 1998 to 1999 and also pro-cyclically during the present expansion.

  • Monetary policy was pro-cyclical in 1984-94 but counter-cyclical after 1994 in the correlation analysis. (p. 26)

    this pro-cyclicality of monetary policy is mainly a feature of the sub-sample 1984 to 1994, for both of the models reported in Table 3. Since 1994 both models show a negative (i.e. counter-cyclical) correlation.

  • Pro-cyclical fiscal correlation with supply shocks is almost entirely a post-2002 phenomenon. (p. 26)

    the positive correlation between the output effect of fiscal policy and the cumulative supply shock is almost entirely due to the post-2002 sub-sample.

  • Counterfactual: removing fiscal shocks lowers post-1994 output volatility modestly (e.g. longer model 0.49 to 0.42). (p. 26)

    the standard deviation of the changes drops for the period 1994 through 2006 from 0.49 to 0.42 in the short sample, and from 0.45 to 0.38 in the longer sample).

  • Counterfactuals show monetary shocks either lowered output variability or were neutral, consistent with earlier work attributing 55% of improved stability to more systematic monetary policy. (p. 26)

    They found that 55% of the improved stability in terms of output and inflation variability, could be attributed to a more systematic monetary policy over this period

  • The paper links the pre-1994 fall in potential output to opposition to apartheid and to political instability, a decline that reversed after 1994. (p. 7)

    potential output had declined precipitously relative to trend in the two decades prior to 1994, a process associated with the internal and external opposition to the system of Apartheid

  • Fiscal policy was rarely used explicitly as a counter-cyclical tool after 1994. (p. 8)

    there is little evidence of fiscal policy being used intentionally as a counter-cyclical policy tool since 1994