Regional eXplorer (ReX) update
March 2010

Introduction

Only a few months into the year, and we have already seen some great improvements to the ReX suite of models. For starters, we have rebased our GDP figures from base year 2000 to base year 2005. Also, due to the improved provincial labour figures that are now being released under the Quarterly Labour Force Survey, we have built a strong provincial component into our labour model.

In this newsletter, we take a look at the implications of rebasing GDP, give an outlook on the South African economy for the year ahead and provide a list of changes taking effect in this release.

GDP rebasing implications

Gross Domestic Product, or GDP, is a measure of the total production of a country. It is calculated by adding together the final prices of all goods and services produced in the country for a specific period. This provides a figure that is called nominal GDP and is useful in a number of applications. However, one of the applications in which this measure of GDP is not useful is in the comparison of production from one year to the next: In other words, in calculating GDP growth rates.

This is because the final prices for goods and services change every year. These prices generally tend to rise over time, therefore increasing the nominal GDP for each year that passes - even if no additional production has taken place. There is no need for concern however, because ‘stripping out' this effect of rising prices is simple. The process for stripping out the inflation effect boils down to counting all of the goods and services in the economy, and then multiplying them by prices from a ‘base' year. The prices in the base year are always the same, and thus the price changing effect on nominal GDP is removed. This new figure is known as real GDP.

However, every couple of years, the base year needs to be updated, i.e. moved forward. This is because over time, new and different types of goods and services are released, structural changes in production techniques take place and so on. Therefore, when the base year is changed, we say that GDP has been ‘rebased.' What this really means is that the underlying final prices of goods have been ‘updated' to the prices of a more recent period. If GDP had never been rebased since 1980, for example, There would be no base price for goods like DVDs, flat screen TVs, cell phones and so on.

StatsSA typically rebases GDP every five years, which coincides roughly with the years of the Income and Expenditure of households survey and the resultant reweighting of the inflation basket. This means that, up until recently, the base for GDP was the year 2000. However, near the end of last year, the base year was changed to reflect prices in the year 2005.

The implications of such rebasing are a matter of some academic study, but suffice it to say that rebasing does have an effect on past GDP figures and even past GDP growth rates. In other words; after a rebasing exercise, past GDP growth rates may be shown to have changed slightly and are often revised by the national statistics authority. In the most recent rebasing exercise in South Africa, StatsSA revised GDP data as far back as 1995.

In Regional eXplorer therefore, we have revised all real (constant) GDP to the start of our time series datasets, 1996. The actual real GDP figures have all been increased substantially, but the growth rates have remained similar to (but not the same as) their base year 2000 estimates.

GDP Outlook

Fourth quarter 2009 GDP figures allude to a moderate improvement for economic growth in 2010. This follows the 2009 economic contraction. However, the 2010 recovery is expected to be slow and below potential. This on the back of a gradual global recovery, a strong rand, delayed private investment growth and reluctant consumer spending over the short-to-medium term. However, by 2011, the delayed effect of loosening monetary-policy, a more pronounced global recovery, and supportive government spending on infrastructure should help to bring the growth path back on track.

The above chart depicts GVA per sector over time and includes the IHS Global Insight expectations for 2010 and 2011, showing a stepped recovery to 2011.

The primary sector may be starting to benefit from increased global demand. However, in fourth quarter 2009, the primary sector only contributed 0.1 percentage point to growth as the agricultural sector contracted 7.6% while mining posted only a small gain. Going forward, the relative strength of the currency is bound to restrict a recovery in these sectors as rising global prices are watered down.

However, it is really the tertiary sector and government spending that dominates the local economy. In fact, this sector contributed 1.4 percentage points (out of 3.2%) to overall growth in the fourth quarter. Sustained government spending led to growth of 7% in the government services sector in the fourth quarter. Financial services, accounting for 21.0% of overall GDP, surprised on the upside by growing 1.1% in 2009 quarter four.

Going forward, IHS Global Insight expects GDP growth to come in at 2.6% in 2010, reach 3.5% by 2011, and rest at around 4.6% by 2013. Unfortunately, this is still below the targeted 6% growth rate.

IHS Global Insight includes below a chart of expected Metro and non-Metro area economic growth rates for 2010, broken into sector contributions. Clearly, the tertiary sector dominates in the metro areas, with the primary sector making an important contribution to the non-metro areas. We expect the City of Tshwane Metropolitan Municipality to show the fastest GVA growth in 2010, with Ekurhuleni coming in at the lowest, slower even than non-metro areas. However, across the country, we expect that most local municipalities (although certainly not all) will show some economic growth in 2010.  

List of additions, improvements, revisions and fixes

The following changes have been made to the ReX data in this release.

• Updated GDP forecasts including the latest estimates from the IHS Global Insight Macroeconomic model. This had an effect on the forecasts from 2009 up to 2013.

• Rebasing from the base year 2000 to 2005. This has obviously led to a large change in real GDP for all years in the Regional eXplorer. This adjustment has also had some marginal effects on the growth rates of certain areas.

• We have built a provincial labour model and used that as an intermediate balance point between the national and local municipal labour data levels. This has recently been made possible by the release of better quality provincial labour figures coming out of the Quarterly Labour Force Survey and represents an improvement to the labour model. There have been some revisions to unemployment and EAP figures for all years, with the largest revisions made in the Kwazulu-Natal and the North West provinces.

• Minor bug fixes:
o Very specific extractions of literacy figures were occasionally adding literacy numbers, instead of applying a weighted average. This problem has been resolved.
o Household infrastructure figures for individual local municipalities did not add up to the national total number of households. This has also been resolved.