Statistics & Estimates / Projections of Protein Needs / Projections January 2013

Projections for oilcake demand for animal use in 2015, 2020 and 2025
Prepared by JSG Joubert (January 2013)

  1. Introduction

    1. The Protein Research Foundation (PRF) focuses mainly on increasing local production of protein to replace imported oilcake as far as possible. Replacing imported oilcake with locally produced oilcake leads to foreign exchange savings and reduces risks for the animal feed industry.

      In managing the promotion and research actions, the PRF must be aware of future oilcake needs and demand. The process requires regular preparation of projections that indicate future oilcake demand for animal use.

    2. The projections are prepared annually based on the actual oilcake consumption presented in the AFMA Annual Report. The methodology used to prepare the projections did not change over time, only the variables change. The AFMA Annual Reports do not distinguish between oilcake consumption per animal species and therefore the APRmodel was used, as in the past, to divide oilcake consumption for the basis year into consumption per animal species.

      This calculation is an essential element of the projections, because the future oilcake demand for animal use is based on and deduced from the future demand for animal products such as meat, eggs, dairy products etc. required by a growing population.

      The Nieuwoudt/McGuigan model is applied to prepare the projections for future oilcake demand, given the calculated future demand for animal products.

    3. Significant variables to prepare the projections include estimated population growth according to race. These assumptions also distinguish between urban and rural areas in terms of black population. In addition the figures include expected income growth according to race. Estimated income growth for low and high income scenarios are also taken into account and are shown in the projections. The projections differentiate between urban and rural black population income growth too.

      Another variable taken into account, is the income elasticity in terms of different animal products required per population group. Protective tariffs, like those that apply to various animal products, are also incorporated in the model.

      Population growth and income growth figures are obtained from the BMR Unisa.

    4. Variables mentioned under 1.3 were not amended, because of a lack of newer statistics and the variables are available for perusal in the December 2011 report.

      Projections mentioned in previous years were projected in terms of the 2010 year. This means it is possible to evaluate the accuracy of the projections calculated at the time. The actual 2010/2011 oilcake consumption was 1 857 490 tonnes, while 2010 projections for the following basis years were:

      Basis year Projections
      2004/05 1 703 735 tonnes
      2005/06 1 691 233 tonnes
      2006/07 2 068 342 tonnes
      2008/09 1 767 952 tonnes

      All oilcake consumption projections, excluding those for the 2006/07 basis year, were lower than the actual consumption in 2010/2011. The underestimates range from 4% to 9%, while the overestimate for one particular year was 11%.

    5. Oilcake consumption projections for 2015, 2020 and 2025 prepared in December 2011 (previous report), based on AFMA's 2010/2011 basis year are slightly higher than the projections in this report based on the AFMA 2011/2012 year.
      Projected oilcake consumption 2015 2020 2025
      From basis year   2010/2011 2 127 336 2 429 057 2 718 485
      From basis year   2011/2012 2 116 377 2 355 727 2 632 604

      The difference in projections for the two basis years used for 2015 projections, is 0,5%. The difference for 2020 is 3,1% and that for 2025 is 3,2%.


  2. Projected oilcake demand

    1. Table 1 shows the total oilcake consumption per animal species for 2011/2012. This is an important input for the model used for the respective projections from the 2011/2012 basis year. The oilcake consumption for the basis year is the actual consumption figures according to the AFMA Annual Report covering the period 1 April 2011 to 31 March 2012. The total oilcake consumption for the said period was 1 856 360 tonnes, marginally different from the actual oilcake consumption in the previous basis year (2010/2011: 1 857 391 tonnes). It is also important to mention that oilcake consumption figures include full-fat production offset as oilcake.

      Table 1

      Total oilcake consumption per animal species for 2011/2012
        Sunflowers Peanuts Soybeans Cotton Canola Lupins Other Total
      Species Tonnes Tonnes Tonnes Tonnes Tonnes Tonnes Tonnes Tonnes
      Broiler chickens 0 0 635 523 0 0 0 0 635 523
      Broiler breeding hens 24 064 0 109 621 0 0 0 0 133 685
      Laying hens 29 620 0 215 486 0 0 0 0 245 106
      Sheep 1 492 0 73 0 0 0 0 1 565
      Cattle 157 003 236 40 871 54 199 0 0 14 181 266 490
      Pigs 41 195 0 128 306 0 9 482 3 985 0 182 968
      Dairy 114 360 0 38 924 65 254 2 677 13 475 14 181 248 871
      Ostriches 1 855 0 21 553 433 12 041 0 0 35 882
      Horses 4 602 0 7 774 210 0 0 0 12 586
      Aquaculture 0 0 727 0 0 0 0 727
      Pets 19 934 0 72 483 0 0 540 0 92 957
      Total 394 125 236 1 271 341 120 096 24 200 18 000 28 362 1 856 360

      Note: Information from AFMA Annual Report 2011/2012. Distribution per animal species according to APR model.

    2. As in the past, the 2015, 2020 and 2025 projections are based on four scenarios:
      • High income growth, keeping protective tariffs in place;
      • High income growth, phasing out protective tariffs;
      • Low income growth, keeping protective tariffs in place; and
      • Low income growth, phasing out protective tariffs.

      Table 2

      Oilcake demand projections for 2015, 2020 and 2025 for four (4) different scenarios ¹
          Annual growth from base year Projection 2015 Annual growth from base year Projection 2020 Annual growth from base year Projection 2025
      Scenario Base year 2011/2012 % Tonnes % Tonnes % Tonnes
      Scenario 1 1 856 360 3.33 2 116 377 2.68 2 355 727 2.53 2 632 604
      Scenario 2 1 856 360 0.01 1 857 407 0.89 2 010 967 1.20 2 193 724
      Scenario 3 1 856 360 2.07 2 014 754 1.77 2 173 073 1.65 2 333 630
      Scenario 4 1 856 360 -0.77 1 800 031 0.28 1 903 335 0.57 2 010 109

      ¹   Scenario 1: High income growth keeping protective tariffs in place.
      Scenario 2: High income growth phasing out protective tariffs.
      Scenario 3: Low income growth keeping protective tariffs in place.
      Scenario 4: Low income growth phasing out protective tariffs.


  3. Oilcake demand and supply

    1. For further calculations, the projected scenario 1 figures are used as the assumption. Since the start of the projections, scenario 1 figures were used as the figures best reflect the realities.

      The next step is to compare local oilcake production to local consumption and future projected demand. Table 3 expresses the locally produced oilcake as a percentage of total consumption for the past four seasons.

      Table 3

      Locally produced oilcake as percentage of total consumption
        Locally produced oilcake Total oilcake consumption Percentage of total consumption
      Period Tonnes Tonnes %
      2008/2009 565 181 1 664 916 33.9
      2009/2010 701 030 1 743 137 40.2
      2010/2011 624 912 1 857 391 33.6
      2011/2012 766 927 1 856 360 41.3
      Average 664 512 1 780 451 37.3
    2. In mitigating the effects of, inter alia, climatic variances on production, as well as price ratios relating to short-term production decisions, the projections are based on the average figures for four seasons to express the local production in terms of total consumption. Table 3 illustrates clearly that there remains a large gap between total oilcake consumption and local production.
  4. Rates at which local oilcake production should increase to provide for projected consumption

    1. An important PRF objective is to replace imported oilcake with locally produced oilcake. Therefore, it is important to know the annual rates at which local oilcake production should increase to reduce the gap between local production and projected consumption. Lastly, it could also help to determine the required percentage of locally produced oilcake in terms of the projected consumption in 2015, 2020 and 2025, if local production were to grow at the historic annual growth rate (see table 7).

      Table 4

      Percentage growth required for annual local oilcake production to achieve certain targets in terms of projected 2015 consumption
      Projected consumption 2015 Required annual growth rate for local production from 664 512 Oilcake quantity Percentage of projected consumption
      Tonnes % Tonnes %
      2 116 377 4.4 789 409 37.3
      2 116 377 12.3 1 058 188 50
      2 116 377 26.3 1 693 101 80
      2 116 377 33.6 2 116 377 100
      Local growth at historic rate
      2 116 377 6.34 849 747 40.2

      Table 5

      Percentage growth required for annual local oilcake production to achieve certain targets in terms of projected 2020 consumption
      Projected consumption 2020 Required annual growth rate for local production from 664 512 Oilcake quantity Percentage of projected consumption
      Tonnes % Tonnes %
      2 355 727 3.2 878 686 37.3
      2 355 727 6.6 1 177 864 50
      2 355 727 12.3 1 884 582 80
      2 355 727 15.1 2 355 727 100
      Local growth at historic rate
      2 355 727 6.34 1 155 507 49

      Table 6

      Percentage growth required for annual local oilcake production to achieve certain targets in terms of projected 2025 consumption
      Projected consumption 2025 Required annual growth rate for local production from 664 512 Oilcake quantity Percentage of projected consumption
      Tonnes % Tonnes %
      2 632 604 2.8 981 961 37.3
      2 632 604 5.0 1 316 302 50
      2 632 604 8.6 2 106 083 80
      2 632 604 10.3 2 632 604 100
      Local growth at historic rate
      2 632 604 6.34 1 571 289 59.7
    2. According to the projection report prepared in December 2011, based on the 2010/2011 basis year figures published in AFMA's actual consumption figures, the historic annual growth rate at which the locally oilcake production took place was 5,16%. The rate was calculated over 14 seasons.

      Local oilcake production reported in this report covers 15 seasons and the growth rate was 6,34% per year. Based on the figures, certain targets will be reached earlier.

      The higher trend in the annual growth rate for the local production of oilcake is based mainly on the events of the past few seasons. If one were to calculate the growth rate for the past five seasons, the annual growth rate is 12,5%. This represents a double annual growth rate over the past 15 seasons.

      If the trend is sustainable, we could produce 50,3% of the total oilcake demand in 2015, compared to 40,2% shown in table 4. In 2020 we could produce 81,4% of the total local demand, compare to the figure of 49,0% shown in table 5. By 2025 we could produce in excess of 31,3% compared to the 59,7% shown in table 6, if we could continue with the oilcake production rate.

      Until recently, sufficient press capacity for soybeans was considered an obstacle, but given the growth in press capacity currently taking place, this aspect is no longer considered a problem.

      Table 7

      Production and consumer trends for protein commodities
      Commodities
        Area planted (Ha) Production (Tonnes)
      Fish meal con­sumption Fish meal pro­duction Fish meal imports Oil­cake con­sumption Oil­cake pro­duction Oilcake imports Soy­beans Sweet lupines Canola Soy­beans Sweet lupines Canola
      Tonnes Tonnes Tonnes Tonnes Tonnes Tonnes Ha Ha Ha Tonnes Tonnes Tonnes
      1996/1997
      196 039 78 430 117 809 832 600 400 675 431 925 87 000 n/a n/a 76 806 n/a 8 734
      1997/1998
      187 374 100 000 87 374 785 401 319 006 466 395 125 000 n/a n/a 120 000 n/a 11 000
      1998/1999
      96 267 88 340 7 927 1 080 354 493 581 586 773 130 500 16 300 17 000 200 900 9 000 21 000
      1999/2000
      127 386 95 000 32 386 1 063 338 554 903 508 435 93 787 19 000 25 000 188 367 13 000 23 000
      2000/2001
      142 848 110 000 32 848 1 021 862 514 020 507 842 134 150 14 705 19 145 148 720 17 360 26 549
      2001/2002
      118 414 128 000 0 1 149 224 482 448 666 776 124 150 14 785 27 000 209 705 16 338 25 750
      2002/2003
      82 988 123 000 0 1 210 396 472 311 738 085 100 150 11 000 33 000 216 000 11 700 37 975
      2003/2004
      127 000 122 000 0 1 121 460 489 413 632 047 135 000 10 100 44 200 136 500 4 040 40 770
      2004/2005
      127 000 132 000 0 121 593 416 736 795 857 150 000 7 100 44 250 220 000 3 950 32 000
      2005/2006
      80 000 101 000 0 1 414 338 572 231 842 107 240 570 14 100 40 200 272 500 14 100 44 200
      2006/2007
      40 240 97 500 0 1 635 525 608 370 1 027 155 183 000 16 000 32 000 424 000 14 400 38 050
      2007/2008
      50 000 102 200 0 1 758 185 494 557 1 263 628 165 800 14 000 33 260 205 000 13 300 39 840
      2008/2009
      75 000 90 000 0 1 664 927 565 181 1 099 746 237 750   34 000 282 000   30 800
      2009/2010
      58 000 111 500 0 1 743 137 701 031 1 042 106 311 450   35 060 516 000   40 350
      2010/2011
      37 000 94 000 0 1 857 490 624 912 1 232 578 418 000   34 820 560 950   36 900
      2011/2012
      45 000 105 000 0 1 856 360 766 927 1 089 433 472 000   43 510 710 000   57 340
      2012/2013
                      44 100 650 000   79 650
      Average annual growth %
      -3.48 2.99 48.30 6.04 6.34 7.54 15.38 3.40 9.02 21.25 25.27 18.44

      Figure 1

      Oilcake consumption, production and imports

      Graph showing oilcake consumption, production and imports

      Figure 2

      Areas and production of protein commodities

      Graph showing areas and production of protein commodities
  5. A few remarks and conclusions

    1. Variables used in the model to estimate the human demand for animal products have not been amended since the 2011 report. According to BMR (Unisa) there were no later information available and the data is still considered reliable.

      Projections calculated annually from basis year 2004/2005, relating to the consumption of oilcake in 2010, indicate that the model's predictions may be considered reliable if the actual oilcake consumption for 2010 is taken into account. BFAP's prediction of actual oilcake consumption of about 1,8 million tonnes in 2020 also does not differ materially from the Nieuwoudt/McGuigan and APR models used by the PRF. The assumption is that soy oilcake in 2020 will represent 80% of the total oilcake consumption.

    2. According to table 1, the poultry industry consumption represented 54,64% of oilcake consumption. For the previous year (2010/2011), the poultry industry consumed 48,1% of the total oilcake. Any change in oilcake consumption between animal species may also impact future oilcake consumption, especially if there is a change in the human demand for the particular animal product.
    3. From table 3 it is clear that there remains a large gap between locally produced oilcake and the total oilcake consumption. During the four seasons from 2008/2009 to 2011/2012, the average local production of oilcake represented only 37,3% of total consumption. It is possible that this gap could be reduced very quickly if the accelerating growth in local production of oilcake during the past five seasons could be maintained. The annual growth percentage over the past five seasons accelerated to 12,5% compared to the 6,34% over the previous 15 seasons.
    4. From table 7 it is clear that, over time, the soy and canola yields showed even more rapid growth than growth in areas of the particular crops. In case of soy, the areas grew by 15,4% per annum, although the yield over the same period grew by 21,25%. In case of canola, the annual area growth grew by 9,0%, while yield grew by 18,4%.

      Improved management practices and cultivars are largely responsible for the trend.

    5. The proportion of soy oilcake as part of the total oilcake consumption, increased over time. Currently the proportion is 69% of total oilcake consumption. If it is assumed that it will represent 75% in 2015, it implies a demand for 1 587 282 tonnes soy oilcake in 2015. The soy oilcake demand for 2020 and 2025 will be 1 766 795 and 1 974 453 respectively. Translated into soybeans, it will represent 1 984 103 tonnes in 2015, but 2 208 494 tonnes and 2 468 066 tonnes for 2020 and 2025 respectively.

      These figures do not include soybeans for human consumption and imports.

    6. Protective tariffs that apply to animal products for human consumption remain the single most important factor that affects the demand for protein for animal consumption. A reduction in the protective tariffs will decrease the demand for protein for animal consumption and vice versa. Another aspect that has been identified now, is the relatively larger quantities of poultry products that are imported. This will also have an impact on the demand for protein for animal consumption. It will also affect job opportunities and may lead to a loss in terms of foreign exchange.