 15 minuwn yn ymarfer, mae nid yn fwy oherwydd. Rwy'n wen weld eich bod, cwrs iawn i allu'r peir, wrth ynglyn â'r cwestiwn, mae'n cael ei wneud o waith y myfarturi? O wnaeth yn cael ei wneud i'r peir i'w myfarturi a'r rôl o'r peir i'w myfarturi? Yma yna'n dda'r peir i'w myfarturi yn South Africa around the minerals energy complex and what that's meant for the specific pattern of manufacturing and the implications for employment because of the domination of a narrow group of capital intensive sectors. So that's the sort of framing. We're then going to look at why the Houghton City region and we're going to present our preliminary findings about how these sectors have evolved, particularly around employment, but not only about employment. It's work in progress, which we're doing for a GCRO here at VITS, but we do have some conclusions, we think already, on the question of linkages and employment, which are worth sharing. Firstly, there's a very big theoretical question here and there's a big theoretical debate about why manufacturing matters, which goes down to the question of is growth sector-neutral or is growth sector-specific? The mainstream approach, as I'm sure a learned audience that we get at tips, will emphasise how growth is sector-neutral, whereas classical and structural perspectives have a history of arguing that growth is sector-specific and that a change in sectoral structure affects economic growth. We don't have time to go into it, but obviously there's a lot of body of literature in classical development economics. Why does manufacturing matter? What do the arguments revolve around? Manufacturing has greater scope for learning by doing, for increasing returns to scale, for cumulative productivity increases, for strong growth pulling linkages with the rest of the domestic economy and for tendencies to technological progress to boil it down. Manufacturing may act as an engine of growth through both output and employment channels and that you can see the growth pulling effects of manufacturing through backward and forward linkages with the rest of the domestic economy and you can see demand multiplier effects through wages paid. Though obviously that's only true if wages are higher in manufacturing in other sectors. There's a whole body of literature summarised there in one slide, but we can obviously come back to it. Clearly, very closely tied to the question of why manufacturing matters is the question of whether de-industrialisation matters and Fiona Tregana, her colleague of mine at UJ, has done very, very good work on this. What you clearly see is advanced economies undergoing de-industrialisation since the 1970s, but also, with the exception of East Asia, many developing economies beginning to de-industrialise at relatively low levels of industrial development, which is really quite a significant thing. What you see in conditions of what you might call premature de-industrialisation, premature and inverted commons, is that can services, which are very likely to be relatively low skilled, low productivity and non-tradable activities in retail and personal services, do these really have the capacity to serve as dynamic replacements, which I think is very, very questionable. What does all this mean in the context of South Africa, which has got a very specific history? It's a very long room and I'm sure at the back your eyes are struggling already. But if we look on the left, the left third, the MEC sectors, this is capital stock, MEC sectors at the core of the South African economy, mining, energy, rubber, petroleum, chemicals, you know, the historic core of the South African economy, clearly rooted in mining in English capital, but then Africana speaking capital through state-owned enterprises in steel, in escom, in electricity, in sassol, in liquid fuels. So these are the core sectors of the economy historically. The right third, that's government services, finance, retail and so forth. The middle third, is really what I want you to look at, is non-MEC manufacturing. Actually, and we're going to show, these are actually the key jobs generation. But if you look at capital stock in these sectors, it's negative in many important sectors. This is leather, wool, paper, printing and so forth. Here, if you look, this is work we did for Houghton in 2007, if you look at what in terms of employment multipliers and backward linkages, if you look at what fall in the top right quadrant, you see wood, paper, print, food, all the sectors where there isn't the capital stock are actually the jobs generators. That's what that shows. Similarly, if you look at the question of linkages, this shows the very, very close linkages between the sectors in the MEC core and their relative lack of linkages with the rest of manufacturing, particularly if you look at the bottom line, share of inputs from MEC, share of output to MEC for non-manufacturing sectors. That shows you've got an MEC core which has tight linkages between it, between, within it, between its and relatively weak linkages with the rest of manufacturing. That's the historic path of accumulation in South Africa. The term is obviously Benfin and Zav Rustumges, Zav who's sitting at the back. So if you look, again, this is just GVA across manufacturing South Africa as a whole, look clearly petroleum products, chemicals, rubber and plastic, metals, metal products, machinery and equipment. In a way, still, still way out in the lead, as it were. And then again if you look at GVA by province, this is for 2011, Hauteng was still the historic centre really of the economy. So I'm going to hand over to Susan in a sec, but as I said, we're doing this work for GCRO, Hauteng City Region, and they have quite a good concept. They call it Hauteng City Region because it's bigger than actually just the borders of Hauteng. So it spreads further to the east to include some of the coal belt, it tells further to the south to include Sassol, further to the north west to Rustenberg, Mining, Platinum and so forth. This is an area encapsulates the historic heartland of the MEC. So what we've been looking at and what Susan's going to now take you through is what's the picture of how it's evolved. I don't know if you want to add anything about the map. Thanks, Sam. Okay, I'm just going to take us through some of the sort of broad data in Hauteng City Region first. And the first thing, this is a sub-sector share of manufacturing GVA in the city region in 2011. And just to show not surprisingly, given the centrality of Hauteng in manufacturing in the country, is that this pattern is basically the same pattern that Sam showed earlier in the pie chart. So if we look at GVA, more than 50% of GVA in the city region in manufacturing is made up of petroleum products, chemicals, rubber and plastic, metals, metal products, machine and equipment. And so these two sort of very central MEC manufacturing sub-sectors. There are two other, if we add in the two next largest sectors by GVA in manufacturing in the city region, we've got food beverages and tobacco and transport equipment. So together these four sectors make up just short of three quarters of all GVA in manufacturing in the city region. This is just to show this extremely skewed pattern of manufacturing. That is a legacy in the remnants of the particular nature of industrial development in South Africa. Rostam G and Ben Fine refer to the mineral energy complex. If we look at fixed capital stock, this is again just zooming in a bit on that slightly harder to see pie chart. The slightly harder to see bar chart earlier. If we look at this in the region, petroleum products make up over 55% of all capital stock and then that together with metals and metal products is almost three quarters. So this just says something about the capital intensity and the sort of skewed capital bias of industry in South Africa and Cal Teng in particular. The reason we want to bring our attention to this is because we are interested in employment generation and how the industrial structure of the economy determines and conditions the nature of employment. So here is employment again. So much slightly different picture. We see that those two large sectors again make up most of employment in manufacturing but metals here overtakes petroleum products because of the difference in capital intensity but we see some other sectors coming in now as really important employment generators in manufacturing, food, beverages and tobacco, furniture and other manufacturing and transport equipment. I want to draw your attention back to that scatter plot that Sam showed where she showed in the top right quadrant were the sectors that we identified back in our work in 2010 as sectors with very strong backlit languages and strong employment multipliers. So potential sectors for growth pulling effects and employment generation. So the first thing to notice is that these three sectors here are food, beverages, furniture, manufacturing and transport equipment are in that top right hand quadrant and I'll come back to that in a moment when we discuss what's been happening in terms of these sectors in terms of investment and employment since 1995. The first thing to say is that investment has been very closely tied to the success of these sectors both in terms of GVA and employment generation. This is not a pretty picture sadly, it's not the most positive of pictures. If we look from 1995 to 2013, the first five to ten years since 94 there's been a decline in fixed investment in manufacturing in Xauteng. It recovered a bit in 2000s only to plummet again with the financial crisis and it's been creeping up. Now this picture, this aggregate picture, masks enormous diversity across industrial subsectors. So this graph is not very clear, I'm afraid given how far away everyone is, but I just want to see if I can use this. Just wanted to show you, so if we remember the last graph and these are the four largest manufacturing subsectors in Xauteng according to GVA, we see that sectors like food and beverages tobacco follow largely the patterns we saw in aggregate and similarly with transport equipment and we get the same across a number of other subsectors. But what's very different, what really bucks the trend is petroleum products. We see this enormous increase here probably related to Sassol, but what's really sad is that sector which has the largest employment in manufacturing in the province has really suffered in terms of productive investment. So one of our preliminary conclusions is if we're interested in a manufacturer, we're interested in employment, this is a space where we really need to start to increase investment and capacity building in metals, metal products, machinery and equipment. So that's investment and what we found in some of our data analysis is a pretty close relationship between investment and employment generation. Again here we show employment generation in manufacturing across the city region and there was a bit of a decline from 1995 to 2003. Part of that might be due to firm restructuring and outsourcing of things like cleaning, but that recovers slightly only to fall again since the crisis and it has not recovered since. This is just to bring out some of that data. So I'm looking at the changes in employment here across manufacturing subsectors and what we've highlighted and read here, so first of all overall decrease since 1995 has been 13% of employment in manufacturing in Haltang city region. And if we look at the sectors that have suffered the most, these are again those sectors in that top right quadrant of having strong backward linkages and strong employment multipliers and have been largely neglected by policy. So things like textile clothing and leather traditionally a very strong employment generator has declined in terms of employment numbers by 31% in these two decades or so. Another dimension of this has been if we look again at the, there's been a bit of a skills bias in terms of this, in this decrease in employment across manufacturing subsectors. So this first column is fixed capital stock, the second column is total manufacturing across subsectors, the changes I showed you in the last graph, but then we separate them across skills level, highly skilled, skilled and semi skilled and the red numbers are all those where employment numbers are registered to decline and we see again, we see also that the job losses are concentrated and skilled, semi and unskilled with highly skilled jobs relatively safe or actually even registering an increase in numbers. This challenges one of the really strong arguments in South Africa that the problem of unemployment lies in labour markets themselves, the supply and demand within the labour markets and the shortage of supply of skills. And we just like to challenge that a little bit to show that actually there's a structural basis for this employment that's not about the lacking of skills in the labour market itself but the lack of demand from industry for these skilled and semi skilled work. So just a couple of conclusions and I think Sam has discussed already why manufacturing matters on a theoretical level both as the engine of growth through growth pull effects and linkages as well as employment multipliers and also we want to show that the MEC continues to matter, it continues to shape the way in which the sort of bias nature of manufacturing in the South African economy and it conditions the nature and scope of linkage development and employment and what we've seen is you know with since 1995 it's those manufacturing sub sectors not linked to the MEC that have seen massive decline. We have two sort of more policy related conclusions as well and the first is if policy you know in terms of employment generation our position is that employment policy or employment generation needs to be closely tied to industrial policy and to understand and cannot sort of look only at the labour market as disconnected from the industrial structure of the South African economy and this is we've got structure here with a big S so this is not just competition or market concentration this is structure as in the vertical linkages and the horizontal diversity, vertical and horizontal linkages of manufacturing itself and so our position is that industrial policy needs to focus critically on these linkages in order to develop a broad, independent and diversified industrial structure and only this can lead to a significant dent in the employment problem. Thank you.