A model for the sector Jerez-Xèrés-Sherry and Manzanilla de Sanlucar de Barrameda

Maria C. Ferguson Amores* Jose A. Machuca** Miguel A. Machuca**
*Research Group EINADE

Depto. de Organización

de Empresa

Universidad de Cádiz

Porvera, 54

11403 Jerez de la Fra (Cádiz)

Fax. 34/56-345104

E-mail: concepcion.ferguson@uca.es

**Research Group GIDEAO

Depto. de Economía Financiera y Direcci-n

de Operaciones

Universidad de Sevilla

Avd. Ramón y Cajal 1

41018 Sevilla

Fax. 34/5-4557570

E-mail: jmachuca@cica.es

The fact that the period from 1985 to 1989 is widely recognised as one of the most critical in the entire history of the sector of "Jerez-Xérès-Sherry and Manzanilla Sanlúcar de Barrameda" first led us to propose the development of an aggregated economic model which would be representative of the companies belonging to the sector and which would achieve the following aims:

* Study the causes which brought about the crisis

* Obtain an educational tool which employs a systems approach

* Obtain a decision making tool which will help to avoid, as far as possible, such decisions as those taken in the previous decade which are now understood to have plunged the sector into the greatest reorganization of its entire history, producing a reduction of over 40% in the total area of planted vineyards, 30% of stock and 25% of employees.

We have chosen the maturing, shipping and export bodegas as being representative of the sector. These bodegas are the driving force behind the vine growers and cooperatives and, in turn, exert great influence on the sector as a whole. We decided to produce a model of the sector based on a study of the structure of these bodegas, which account for 95% of the sherry wine produced in the area and are the main force behind the sherry based economy.

System Dynamics was used in the modeling process and its conceptualization was based on a description of the system to be modeled (the sherry sector) and the way in which it operates. The origins of the imbalance between supply and demand were studied and we also considered the extent to which excess stock, prices and investment in advertising have affected the perceived image of the product. That is to say, at this stage we analysed and defined the different problems existing in each area and finally produced a causal diagram.

With regard to the first of these aspects, the decade of the seventies witnessed a spectacular increase in sales with the export figure surpassing 303,970 butts (approximately 150 million litres). This figure signified a increase in sales of 117.2% in 10 years. This increase was wrongly interpreted by the producers as signaling parallel increases in consumption, which generated enormous hopes in the companies involved and brought about an increase in production pressures. This led to costly investment in vineyards, deemed necessary in order to provide greater levels of stock, and in the four-year period from 1974 to 1977 the sector increased its total area of planted vineyards by over 100%.

A sharp fall in consumption took place in the main export markets as from 1979 and this provoked the beginnings of the problem of excess stock in the bodegas, a situation which, in turn, brought about the appearance of excess stock in both the cooperatives and the other producers of must. The imbalance between the supply and demand for must increased and this led relations in the sector to such a degree of tension that in 1990, and for the first time ever, all the affected parties joined together to implement a four-year restructurization plan which led to a reduction of both stock and vineyards in an attempt to comply with current levels of demand.

The formalization process was begun with the formulation of the production sub-system, this being where the first of the socioeconomic activities surrounding the sherry making process take place, and also where the imbalance suffered in previous years was made most clearly patent.

Once the formulation stage was finalized we moved on to the analysis and validation of the structure. We have attempted to analyse the capacity of the model to generate and reproduce past behavioral modes which closely resemble reality. The reproduction of behavior during the five-year period 1985-90 was taken as the main test of validation. The choice of this period was justified as it is prior to the restructuring plan of 1990. Once the initial levels corresponding to the campaign of 1985 were established, we went on to analysis the model. When the model was compared with historic data from the period prior to restructuring the results obtained showed a degree of concordance for the qualitative behavior of the sector and also the existence of specific decisions which were key to the origin of the aforementioned imbalance, such as:

- The decision to plant more vineyards, due to an erroneous forecast of demand, and their posterior maintenance for over 15 years which led to an exaggerated excess of capacity.

- A reduction in the percentage of the cupo, or quota, for each campaign in an attempt to control the market but which finally led to an excess of stock

- Erroneous decisions made concerning policy for purchasing must, in the sense that such purchases were carried out without taking into account existing warehouse stock.

If a model such as the one described here had been available, then the possible negative results of the situation outlined above could have been identified and, in turn, could have been avoided.

The results obtained have increased our confidence in the reliability of the model as a behavioral simulator and have allowed us to simulate different policies and situations, its application enabling us to analysis the long term consequences for the sector.

Besides helping to increase our knowledge of the sector as a whole, the model provides a simple, user friendly instrument which is flexible in its adaptation to possible environmental changes.

Finally, we shall state that both the simulation and the ensuing sensitivity analysis which we carried out have increased our confidence in the validity of the model as a representation of the system which forms the subject of our study. In this way the results of future simulations could be of great help to those responsible for decision making within the sector

ISDC '97 CD Sponsor