Modelling Longterm Commodities: the Development of a Simulation Model for the South African Wine Ind

1
Modelling Long-term Commodities the Development
of a Simulation Model for the South African Wine
Industry within a Partial Equilibrium Framework

• Presenters Michela Cutts
• Sanri Reynolds
• Other authors Ferdi Meyer
• Nick Vink

2
From Regulated to Competing in the International
Arena

• The dynamic, recursive partial equilibrium model
  built to assist the industry by answering WHAT
  IF questions

3
What If
Sin Tax
MELBOURNE (Reuters) - Australia's drought could
cut the 2008 wine grape vintage by more than half.
Climate change risks devastating South Africa's
wine and fruit industries
Australian warning on oversupply - Worldbeat
Price and Exchange Rate Changes
4
Data

• Domestic data South African Wine Industry
  Councils information unit, South African Wine
  Information and Systems (SAWIS).
• International Prices Compendium of Wine
  Statistics
• Competing Crops Abstract of Agricultural
  Statistics
• Macroeconomic variables South African Reserve
  Bank and Statistics South Africa websites

5
Methodology

• Methodology developed by the Food and
  Agricultural Policy Research Institute (FAPRI) at
  the University of Missouri
• Adapted to suit the nature of a long-term
  commodity

6
Wine Grape Supply Block

• Divided into 8 production regions
• South Africas top 10 varietals by volume
• Given different slopes, trellising practices etc,
  vine numbers are used
• Grape production per varietal vine numbers x
  yield

7
Farmers Decision Making Process
Plant vines or alternative fruits/crops based on
expected real gross return
Fruits/crops
vines
Fruit/crop sector level model
Choice of varietal -based on weighted sum of
expected real gross returns
allows for determination of cross price
elasticities with competing crops and different
varietals
8
From Grapes to Wine

• Juice (litres) 0.85 x Grape production (tons)

Juice
Non-alc.
Good wine
Rebate wine
Distilling wine
9
Wine Demand Block

• Domestic Demand
• Estimated for rebate, distilling wine and good
  wine
• No quality attributes considered for wine.
• Per capita consumption f(real wine price,
  GDP/capita)

• Export Demand
• Disaggregated into red and white wine
• Individual equations estimated for 10 country
  groupings
• Exp dem f(exch, lagged exports, SA price, new
  world wine price)

10
Linking Grape and Wine

• Price equations create the link between
  grapes and wine
• Noble varietals price f(lagged variety price,
  real wine price, and production of the variety)
• Non-Noble varietals price f(lagged variety
  price, real rebate and distilling wine price, and
  production of the variety)
• Wine pricef(producer wine sales, wine
  production, exchange rate)
• Rebate price f(wine price, rebate wine
  production)
• Distillate pricef(Rebate price, distillate
  production)

11
The Closing Block

• For equilibrium to be reached total demand
  total supply
• System is closed using change in stock
• wine production plus wine imports,
• less exports and domestic consumption

12
The Baseline A Possible Outlook

• A benchmark of likely trends and levels of
  prices, production, consumption and trade under a
  particular set of assumptions
• Does not constitute a forecast

13
Macroeconomic Assumptions
Sources Global Insight
14
Red Vines
15
White Vines
16
Domestic and Total Demand
17
What if

• the Rand appreciates relative to the baseline?
• Baseline Rand depreciates gradually from
  R7.18/USD in 2007 to R10.40/USD in 2014?
• Scenario What if Rand stays constant over the
  baseline at R7.18/USD?

18
What if (cont.)
19
What if (cont.)
20
What if (cont.)
21
What if (cont.)
22
Conclusion

• So how do you use these results?
• Evaluate results critically and decide whether
  you agree
• It teaches us something about the system, market
  structures and price formation
• How exogenous factors influence the business and
  policy environment