Did the South African economy really contract by 51%?

Thu, 10 Sep 2020 21:13:14 -0500

Disclaimer - The opinions expressed here are solely my own and do not reflect the views of anyone else.

The build up to the latest South African quarterly GDP release was captivating to say the least, as analysts awaited eagerly to confirm their projections on just how bad the second quarter (Q2) was set to turn out. The catastrophic economic performance recorded over Q2 was largely in line with most expectations, but by all means still managed to dominate headlines across the country. Stats SA indicated that an economic contraction of 51% was recorded over Q2 of this year - but does this really mean that half of South Africa’s GDP was wiped out in as little as three months? This article offers some insight into why this number might mislead you to think so.

What is the difference between annual and annualised growth rates?

When Stats SA publish economic growth statistics on a quarterly basis, they present these figures in both annual and annualised terms. Here is a quick breakdown of how these two measures differ.

An annual growth rate measures the change in GDP between a given quarter versus the same period a year earlier (in this case, this would be the change between Q2 of 2019 and 2020 respectively). Analysts tend to focus on this metric as it is useful to draw conclusions on whether households and businesses are better (or worse) off today compared to where they were a year ago. Under this approach, the South African economy contracted by 17.2% over Q2 of 2020 (see Figure 1).
In contrast, an annualised growth rate takes the percentage change in GDP between two consecutive quarters (in this case, Q1 and Q2 of 2020) and estimates what GDP growth would be if it that calculated rate persisted for an entire year. In actual fact, GDP only contracted by 16.3% between these two periods - the 51% figure (that’s received widespread attention) was calculated on the assumption that the economy will contract by 16.3% during each quarter over the next year.

There is such a sharp contrast between these two growth rates, so which one provides a more accurate view given the current economic context? Well, the main issue with using annualised figures to measure changes in GDP (particularly in a year as unprecedented as this one) is that outliers will distort how you interpret the magnitude of this shock. To illustrate this point, suppose you receive a one-time bonus of R1000 with your salary this month. It is unlikely that you’ll think this windfall translates into an annual raise of R12000 given your expectation that receiving this bonus is a one-time event. The GDP contraction recorded over Q2 should be viewed in the same light - as devastating as the lockdown measures were on the economy, it is critical to note that this was an anomaly (because policymakers are unlikely to resort to these drastic measures again throughout the remainder of the year). Therefore, it’s unrealistic to infer a long term trend based off the disaster that transpired during these three months.

Figure 1: Annual vs Annualised GDP Growth Rates for South Africa (Q1 2019 - Q2 2020).

Should we expect production to normalise over the next year?

To answer this question, one needs to understand how production has fared since the gradual reversal of lockdown measures has been initiated. An industrial production index can be a useful tool to analyse this, seeing as this indicator is essentially an informal proxy for monthly GDP. The annual growth rate in the economy’s manufacturing sector is shown in Figure 2, which indicates that there has been a partial recovery in production since the historic low recorded in April. A smaller annual contraction recorded over the July already represents some improvement compared to how the second quarter started out, though it its important to note that production levels remain at substantially lower levels compared to where they were prior to the pandemic’s onset.

Figure 2: Annual Growth Rates for Monthly Manufacturing Production in South Africa

Purely on the premise that the economy will be coming off the low base created by the destructive events during the second quarter, it is only practical to expect further (though limited) improvement in production over the next few months as lockdown measures are eased. The unfortunate part about this is that the media’s reaction once the next quarter’s figures are released in a few months will most likely be as overblown as what we saw earlier this week. We can expect a sharp rise in annualised growth at the end of Q3 due to this “low base effect” from Q2, which will create the false impression that South Africa’s economic renaissance has taken full effect.

Right now, the economy is going through extreme short-term changes. This doesn’t mean that the scars from the lockdown period won’t remain for years to come - consumption and investment will take some time to fully normalise, and it is unlikely that the job losses induced by the pandemic will be reversed anytime soon either. Additionally, high levels of government debt coupled with depleted tax revenues have left fiscal policymakers with limited buffers to cushion the effects of the shock. With that being said, it makes little sense to suggest that the remainder of the year will resemble anything similar to the unprecedented troughs experienced in Q2. If you share the same sentiment, then you shouldn’t focus too much on annualised growth rates this year.

Gross Domestic Product | Tangeni Shatiwa

Did the South African economy really contract by 51%?

What is the difference between annual and annualised growth rates?

Should we expect production to normalise over the next year?