Cost of strikes to South African economy grossly exaggerated
THE RECENT spate of strikes in South Africa has generated some stock responses from both the media and foreign investors: that the country, through its politically strong trade union movement, is held to ransom by high wage demands, or the belief that the strikes are almost always violent and economically destructive.
But, by analysing strike data, a more objective and informative understanding is made possible.
On the number of formal sector workers involved in strikes, the raw data from the Department of Labour’s Annual Report on Industrial Action shows in 2003, 83 533 workers were on strike, while in 2007 the estimate was 608 000 and in 2011 it stood at 203 000. These numbers however, must be appropriately normalised. For the same years, these strikers represented only 1.1 percent of the workforce in 2003, 7 percent in 2007 and only 2 percent of all workers in 2011.
In other words, over the three years reviewed, less than 7 percent of South Africa’s workforce was actually on strike.
To estimate the economic costs of strikes, the standard approach is to simply calculate the number of strikers multiplied by the number of days of the strike. Hence, if 10 workers strike for four days, the days lost through striking would be 40 days.
In 2003 then the total number of strike days lost stood at about 920 000, in 2007 the total was 952 000, increasing rapidly to 2.8 million working days lost in 2011. Again, though, it is crucial to estimate lost working days as a share of the total number of working days in a given year. When the proportionate number of working days lost was calculated, it was actually extremely low in 2003, 2007 and in 2011. The number exceeds 1 percent only once – reflecting the public sector strike in 2011. The share of working days lost as a share of total working days in 2003 was 0.05 percent, in 2007 it was 0.45 percent and by 2011 it was 0.13 percent.
It may be argued though that despite this, we seem to be a particularly strike-prone economy when international comparisons are made.
Yet data suggest this is not true at all. Among a sample of BRICS economies (excluding China due to the lack of data), South Africa ranks just third in terms of workdays lost due to strike activity (with an average of only 3.77 percent of workdays lost per year).
Brazil ranks first with a significantly higher 46.4 percent, India is second with about 8 percent and Russia fourth with 2.5 percent. In fact, the average percentage of workdays lost per year in these countries from 1999-2008, is 12.26 percent. Ultimately, evidence for South Africa’s reputation as an economy with a high strike rate is very thin.
Beyond the BRICS and in relative terms – compared to seven countries with similar economies, such as Nigeria, the US and Turkey – South Africa ranks only sixth.
With Turkey (19.77 percent of workdays were lost per year between 1999 and 2008), Nigeria (10.94 percent) and the US (9.51 percent), all exceeding South Africa’s 3.77 percent.
The results of these comparisons show strike action, and the implied costs in terms of workdays lost, are common in any economy with a functional industrial environment where workers and employers are free to engage and negotiate the terms of labour costs.
In the media, South Africa often appears as a strike-prone country. Reporters frequently focus on the apparently large number of workers involved in industrial action. In fact, in terms of strike intensity (or the number of strikers per 1 000 workers) South Africa’s average for the period between 1998 and 2009 is mod- est at best. On average, only 28 out of 1 000 workers were involved in strike action for the period reviewed.
In this, South Africa has a similar strike intensity to economies such as Denmark (25 strikers per 1 000 workers), Australia (24/1 000) and Iceland (31/1 000). And, in turn, South Africa’s economy is less strike-intense than Italy’s (66/1 000), Spain’s (73/1 000) and particularly Argentina’s economy with its average of 307 out of 1 000 workers involved in strike action.
This suggests there are three key lessons to be learned from the strikes that have taken place in South Africa in recent years.
First, such strikes are part and parcel of any normal functioning industrial relations environment, and are no different from the strikes in both developed and developing countries.
Second, for the three years we reviewed, less than 7 percent of the country’s workforce was involved in strike action. In some years, this figure was less than 2 percent, while the cost to the economy in terms of days lost is extremely small.
And, lastly, real strike data suggest that compared with other countries – relative to other middle- and high-income economies – the incidence, intensity and cost of strikes in South Africa is drastically overstated.
Bhorat and Tseng are researchers in the Development Policy Research Unit at the University of Cape Town