Journal Issue

4. Potential Growth and Output Gap in the Aftermath of the Financial Crisis: An International Perspective

Nir Klein
Published Date:
August 2011
  • ShareShare
Show Summary Details

The sharp deceleration in the potential output growth is well reflected in the developments in labor and capital inputs. As shown in Figures 11 and 12, South Africa is among the few countries that experienced a non-negligible contraction in both employment and fixed capital formation in 2009–10. In particular, South Africa experienced the largest loss of employment following the financial crisis (2010 average compared to the 2008 average), among its comparable group of emerging markets even though the level of economic activity expanded in real terms since the crisis unfolded, while gross fixed domestic investment continued to decline in both 2009 and 2010 (in constant prices).

Figure 11.The change in gross fixed capital formation in selected economics, 2008-2010

Source: World Economic Outlook database and staff calculations.

Figure 12.The change in employment in selected economies, 2008-2010

Source: World Economic Outlook database and staff calculations.

An international comparison reveals that the financial crisis has led to a deceleration in potential growth in advanced economies, as a group, as well as in a number of emerging economies. The magnitude of the deceleration seems to vary, reflecting in part the various degrees of financial and trade openness and other country-specific factors such as the rigidities in the labor market (Figure 13). The deceleration in South Africa seems to be among the highest in this group, with a deceleration of about 3 percent—second only to Russia (nearly 4 percent).13 The confluence of the considerable deceleration of potential output growth and the relatively shallow trough in 2009 places South Africa’s output gap in the postcrisis period close to the sample’s median, despite the anemic recovery that was registered in 2010 (Figure 14).

Figure 13.Potential output real growth in selected economies

Source: WEO database.

Figure 14.Output gap in selected countries (in percent of potential output)

Source: WEO database.

The Impact of Global Economic Fluctuations on South African Output Gap

Given the impact of the global financial crisis on the South African output gap, this section aims to quantify the extent to which the South African output gap is sensitive to external shocks. This is done by regressing the global business cycles (the deviations of the world output from its HP filter) together with other variables on the estimated output gap (average of all the methods). The other explanatory variables include the yield on 3-month Treasury bills adjusted for inflation (real rate), the global inflation cycles (world’s consumer price index, weighted by trade imports from advanced economies) to control for sharp movements in the prices of commodities, and D(G_Y), which reflects the change in public consumption as a share of GDP. The explanatory variables also include the output gap (GAP) with lags. The various specifications of the estimations appear in Table 4.

Table 4.Dependent Variable: South African Estimated Output Gap, 1985Q1–2010Q41
Reg. 1Reg. 2Reg. 3Reg. 4
Gap (-1)1.286*1.298*1.265*1.221*
Gap (-2)−0.208−0.227−0.245−0.244
Gap (-3)−0.228**−0.217**−0.224**−0.183
Real Rate(-8)−0.026***−0.025***−0.032**−0.023
Global business cycles0.079***0.103***
Global inflation cycles−0.158*
Significance level: * Significant at 1 percent.** Significant at 5 percent.*** Significant at 10 percent.

Figure 15.Rolling regression, the coefficient of global business cycles

Although the estimated output gap is highly correlated with its level in the previous quarter, the estimations reveal that other external and internal factors do play a role. As expected, an increase in the short-term real rate contributes to the contraction of output gap with a lag of eight quarters, while an increase in public consumption (as a share in GDP) has a positive impact on the output gap with a lag of three quarters. Additionally, the estimations show that the output gap is indeed sensitive to external shocks measured by both fluctuations in the global output and inflation.

Although the estimation indicates that the size of the coefficient of global business cycles was on average around 0.1 in the sample period, a rolling regression, which is based on the Reg. 4 specification, shows that it varied significantly over time.14 In particular, this exercise shows that the coefficient has become significantly different from zero at around 2000 and its size has increased steadily since then (Figure 13). This result points to South Africa’s increasing integration into the global economy in recent years, and as a corollary, the growing sensitivity of the economy to global economic shocks.

Given that every estimation technique has its pros and cons, this section uses a simple average of the estimated output gap and potential output growth.

The rolling regression is based on a fixed window of 30 observations, in which all explanatory variables were allowed to change.

    Other Resources Citing This Publication