MILK PRODUCTION PROFITABILITY – MULTIPLE REGRESSION ANALYSIS

Sławomir Juszczyk
Institute of Agribusiness, Swietokrzyski University of Kielce, Poland

ABSTRACT

The aim of this research was to analyze the factors which contribute to the economic and organizational conditions of milk production profitability. Specialized farms in the Central Macroregion are featured. In the years 1997 – 2001 the costs were to achieve profitability in direct milk production. Apart from the direct costs and the average purchase price in the particular years of research, on average twice as high as the prices, other variables imparted the level of direct profitability. The greatest number of additional defining variables occurred in 2000. They were: complexity of technology, milk productivity per cow per year, replacement cow factor, value of milk sold, value of cattle production (without milk), milk cooler ownership, and summer pasture availability. In 2001, it was important to have a computer with software to analyze farm milk production and costs.

Key words: milk production, profitability, regression analysis, farm structure.

INTRODUCTION

Milk production is considered the most difficult, most complicated and most laborious activity on the farm. Milk production requires expensive capital assets and therefore it is very difficult to start from scratch. It is easier to improve upon one an existing facility. The producer must have a huge knowledge base, patience, and long-term emotional engagement. His family must be totally engaged because milk production needs agreeable, harmonious cooperation of all household members.

The main matter is that if a farmer wants to assure his family a high standard of living, he has to engage in milk production concentration on a large-scale. Meanwhile, usually this is very difficult for many reasons, but the most important are limited possibilities of buying land near the farm’s centre or no such possibilities at all and limited personal financial resources. Therefore, there are questions – how to raise and educate your children, how to maintain a farm without debt and high risk which enterprise mix to maximize income and how to modernize the farm.

A possible way to achieve these objectives is to specialize in production with the present farm structure and at the same time consider possible changes. The measure of success could be direct surplus of milk production, if the farm has no difficulties with access to financial resources. If there are such difficulties, the farmer’s economics enterprise goal could be achieving maximum milk production profitability and as a result achieve the highest potential multiplying of their own financial means. The evaluation of a farm’s effectiveness is so important because almost every farmer specialized in milk production is different. Therefore identifying economical, organisational, and technological mistakes made by the farmer himself is in general, difficult. Therefore determining the economical and organizational conditions for maximum milk production profitability on specialized farms seems economically important and is as fascinating goal. In addition, the research results could have theoretical as well as practical meaning.

THE GOAL, RANGE AND METHODS

The main goal of this researcher was to determine diversity/variability of direct profitability in the milk production enterprise and to identify the factors, which influence the level of profitability.

In the analysed issues, it is essential to recognize the simultaneous impacts of production scale, organization, and producing technology on production volume, costs and purchase prices. Price received depends not only on milk quality but also on the dairy balance of economic power where the farmer sells his milk. The research hypothesis follows – the most important factor, which influences milk production profitability, is production costs.

The geographic area of the research was limited. In 1996, for research purposes the Central Macro region was formed. It consists of the following provinces: ŁódŸ, Skierniewice, Piotrków Trybunalski, Sieradz i Płock. The Central Macro region is representative of central Poland in average climate conditions and average size of farms. The choice of farms to analize included – all farms in the Central Macro region which on 1. January 1997 had more then 10 milk cows. Subjectively, it was assumed that producers with ten or more cows are more engaged in production and more experienced than producers with fewer cows. Data was collected for the years 1997 – 2001. Data collected included:

• mass statistical data,

• survey (questionnaire) data from 157 farms that fit the research project framework identified by the author,

• technological cards for plant activity in cubic fodder production.

In this research, a very important relationship between potential cash value of milk production and direct costs was identified. This relationship was named the direct profitability index (Dpi).

Dpi = Potential cash value of milk product enterprise/Direct costs

The term direct profitability index was adopted from the work of Stańko [1973] and Klepacki [1997], because it describes methods of studying relationship such as economic competitiveness of alternative production activities on the farm or between farmers. A high direct profitability index means it is possible to choose to increase production, which leads to the maximum multiplying of financial means. In contrast, a shortage of financial means of a farm is indicated by a low direct profitability index. This situation arises when the sum of the farm‘s financial means is at a minimum. These circumstances arise when a farm has difficulties with following:

• maintenance of cash flow,

• access to credit,

• lack of cash flow of the milk receiver (mostly dairy co-operatives),

• adjustment to market requirements,

• decreasing market intervention, and so on.

The direct profitability index is connected with rational management decision making. The Dpi could be used as an indicator of profitability in agriculture, per herd, hectare or work hour. It could help the farmer decide what one hour made to produce. Because indirect costs are not included, the direct profitability index level shows the impact of technology differences given the same product on different farms.

The value of potential cash production which is necessary to calculate the direct profitability index, was established by multiplying the purchase prices and quantities of milk in each quality class summarizing the results and adding to that the income from calf, heifer and cull cow sales, the value of home consumption, and the value of direct to consumer product sales.

Direct costs associated with milk production were collected. They included fodder (hay) and straw baling, filling fodder, buying fodder, internal transport, electricity, fuel, water, waste removal, medicine and vet services, artificial insemination, replacement heifers, services and hired labour, repairs, and insurance connected only with milk production. Costs for farm produced fodder was calculated on the base of market prices of direct inputs including: seeds, fertilizers, pesticides, services (including transport), hired labour, own mechanical force and own transport.

Statistical and mathematical methods were used to analyze the data Statistical methods used were simple and multiple regression and correlation analysis. In multiple regression models, the choice of independent variables was determined using Hellwig’s method, information from this research, and current literature on this subject matter [1].

RESULTS

There was considerable variability in the direct profitability index from one farm to another. The range of direct profitability index values in 1997 were from 0.91 to 4.89; in 1998 from 0.97 to 4.91; in 1999 from 1.09 to 4.94; in 2000 from 1.22 to 4,95 and in 2001 from 1.11 to 4.30. The percentage difference from the lowest to the highest farm profitability index was 100%, the greatest was in 1997 at 537.4%; followed by 1998 at 506.2%; 1999 at 453.2%; 2000 at 405.7% and 2001 at 387.4%. For the milk farms surveyed, a continual decrease in the difference between the lowest and the highest level of direct profitability index was observed. Also, note that all farm direct profitability indexes were greater than one beginning in 1999.

Five multiple regression models of the profitability index are presented. They represent the years 1997, 1998, 1999, 2000 and 2001. Independent variables were used that were strongly correlated with the dependent variable. Furthermore, independent variables, which were strongly correlated with other independent variables, were eliminated. For similar variables that were highly correlated (e.g. Big Piece and number of cows in the farm or direct costs and fodder costs, etc.) to the variable with the highest correlation coefficient was included as an independent variable. In the next approximations, variables for which the Student-t was insignificant were eliminated. Each model’s significance was tested using Snedecor’s F Test. The results are presented in table 1.

One research hypothesis was – the most important factor that causes differences in milk production profitability is production costs. In this research, the absolute values of net multiple regression coefficients for direct costs were the highest among all independent variables that explained profitability in milk production. This result supports the stated hypothesis. Over the research period, the growth of direct costs of about 10 grosz per milk litre was directly related to an average decrease of 0.42 to 0.77 in the direct profitability index.

The research results indicate a relationship between the direct profitability index and the purchase price of milk. Its growth of about 10 grosz per litre was connected with improvement in the direct profitability index of about 0.22 – 0.36. This means that the farmers have to try, if it is possible, to raise prices received for milk, e.g. by care to preserve milk quality or selling to a different milk buyer. From 1997 – 2001 absolute values of the regression coefficients for direct costs were rising. Except in 1998, the net regression coefficient connected with purchase price trended upward. It is important to note that the absolute values of the coefficients for direct costs were about two times as high as the coefficient for purchase prices. This demonstrates the significance of costs and purchase prices during the period analyzed. The importance of costs in determining the level of the direct profitability index was twice as high as for the purchase prices. These research results confirm that a necessary condition for improvement in efficiency in milk production is to decrease the direct costs. This is a very different task for milk producers to accomplish, but with determination and a huge effort, improving milk production profitability can be achieved.

In especially difficult period for milk producers was 1998 when new higher requirements connected with norm PN – A – 86002 “Row milk for purchase” started. Perhaps related is the net regression coefficient for purchase prices that decreased from 2.8098 in 1997 to 2.2352 in 1998. It indicates that at this time the farmers received an average, lower purchase prices for their milk. This was related to difficulties in achieving the quality parameters for milk in the newly introduced additional quality class.

An interesting situation occurred in 2001 when a milk purchase price reduction was observed, but these farmers, which could rationally decrease costs, achieved greater effectiveness of production. In this period, absolute values of net regression coefficients for direct costs and price decreased to: -5.4072 and 2.8244. It helps confirm a decreasing price and production cost level at this time. Important also is the fact that the ratio of absolute coefficients decreased to less than two. This demonstrates relative growth in the importance of price for achieving a high level of direct profitability in milk production.

In addition to direct costs and average milk purchase price the other variables contributing to the explanation of the level of a direct profitability index were:

• in 1997 additionally three variables – complexity of technology index, share of farm heifer sales in sale of milk and other goods connected only with milk production, and milk yield per cow;

• in 1998 additionally five variables – complexity of technology index, share of farm heifer sales in the sale of milk and other goods connected only with milk production, milk yield per cow, seasonality index of production, and number of cows in herd;

• in 1999 additionally six variables – complexity of technology index, milk yield per cows, seasonality index of production, number of cattle in all ages, cull cow index and share of milk in sale of all cattle products;

• in 2000 additionally seven variables – complexity of technology index, milk yield per cow, cull cow index, value of milk sales, value of cattle production (without milk), possessing of milk chilling container and pasture raising in summer;

• in 2001 additionally four variables – milk yield per cow, cull cow index, share of milk in sale of all cattle products and possessing of a computer.

Taking all of this above into consideration it is necessary to answer the question – why in the individual years of research was there instability in the variables explaining the level of each direct profitability index. For example – in 2001, why did the complexity of technology index not have an influence on effectiveness of production distinct from earlier years? In 2001 milk purchase prices were decreasing and producers looked for ways to improve production profitability through more extensive production and rational reductions in costs. This means that with the application of extensive production, complexity of technology and keeping a strong technology regime has less influence than in the case of intensification of production. Nevertheless, it should be stressed that in the first four years of research keeping a strong technology regime was a very important feature, which had an influence on profitability in milk production. It indicates that keeping up with technology is a condition for development of farms specialized in milk production. It is interesting that the strength of the influence of technology complexity in succeeding years was lower and lower. It could be connected with the fact that farmers specialized in milk production realized many technological advances early in the study period, made fewer mistakes, and that the technological adoption level of all groups became more and more similar over time. Keeping the technological regime updated was just in the best interest of farmers that set high standards and achieved average milk’s purchase prices close to the price of the additional higher priced class.

The growth in share of farm heifer sales in total sales of milk and other goods connected with milk production positively and significantly influenced the model results in the first two years of research. It could be presumed that in years 1999 – 2001 the specialization of production by farmers’ achieving the highest class of milk quality required their attention and concentration on the main product. Distraction from production and sale of heifers was reduced and had smaller influence on the effectiveness of milk production.

Important also is the influence of milk yield per cow on the effectiveness of milk production. This is a reflection of production intensity and in fact its effect. Optimal levels of production intensity for each product, including milk, were a result of the relationship between input prices and agricultural products prices. As can be seen from the models outcomes in the first years of this study these relationships in milk production were profitable for farmers. The higher level of the outlays was connected with improving economic results, but in the last two years of research, these relationships diminished. Furthermore, for years 2000 and 2001, the multiple regression results increases the probability of recognizing that a significant increase and then a quick dropping of prices intercepts organizational and technological progress in milk production. In addition, in the case of a decreasing price, that was observed in 2001, the intensification of milk yield per cow was not sufficient for profitability of milk production. It confirms the need for the stabilization in the purchase price of milk because continual fluctuation in milk prices could be, in the final analysis economically unfavourable.

In 2000, the net regression coefficient for milk sales value was not large, nevertheless, the presence of this factor in the model shows that the quick increase in purchase price of milk encourages special care to maintain milk quality and in this way add to milk sales value. In 2000, it was important to posses a refrigerated milk tank. At this time, not every farm had invested in one but it created the differential in milk quality that lead to an increase in milk price received by the farmer. Summer pasture grazing appears as a significant variable in the year 2000 regression model indicating a growth in importance of decreasing feed costs at this time. It is interesting that in 2000 computer possessing with milk production software appeared in the model. We could presume that with milk price, decreasing it was a rational decision to decrease costs, especially in feeding, and this model was easier with computer analysis. Increasing farming profitability in difficult times requires information and the ability to process and use it in decision-making. Indirectly, this confirms the growth in significance of knowledge, based on farm decision making especially during times of difficult external economic conditions.

CONCLUSIONS

Instability of variables, which explain the level of the direct profitability index for the individual years of this research, results from dynamic production conditions, both internal and external. It was surprising that the largest number of variables, which explain the level of the direct profitability index, occurred in 2000 when the milk purchase price obviously increased. The multiple regression models indicate that in the circumstance of price increases for goods produced, farmers rationally should increase production and pay close attention to find weak points in the production process. Increasing production without maintaining the highest milk quality, and purchase price could lead to higher costs and not achieve the level of expected financial profits. And with the future decline in economic conditions, the farm could experience deeper technological, organisational and economic difficulties. Taking the need of milk production stabilization into consideration, the above issue could have more and more importance both in reference to the farm and in milk quota organization, which may occur in the future.

Finally, in years 1997 – 2001 of this research specialized milk producing farms in the Central Macro Region of Poland the net regression coefficients for direct costs were the highest amongst all of the significant variables, which explained the level of direct profitability in milk production. The most important factor in explaining direct profitability in milk production was direct costs.

REFERENCES

1. Borkowski B., Dudek H., Szczęsny W., 2003: Ekonometria, wybrane zagadnienia [Econometrics, selected issues]. PWN, Warsaw [in Polish].

2. Juszczyk S., 2003: Ekonomiczne podstawy wyboru srodków higieny dla specjalistycznych gospodarstw mlecznych [Economic bases of choosing the hygienic means for specialized milk farms]. Roczniki Nauk Rolniczych, seria G Ekonomika Rolnictwa [Agriculture Science Annals, series G Agricultural Economics] vol. 90, b. 1, Warsaw [in Polish].

3. Klepacki B., 1997: Produkcyjne i ekonomiczne przystosowania gospodarstw prywatnych do zmian warunków gospodarowania. [Production and economical adjustments of private farms to the chances of farming condition]. SGGW, Warsaw [in Polish].

4. Klepacki B., 2003: Gospodarka oparta na wiedzy jako szansa rozwojowa Polski. [Economy based on the knowledge as a chance of Poland’s development]. Wies Jutra No 1 (54), Warsaw [in Polish].

5. Majewski J., 2002: Zmiany jakosciowe i sezonowosc skupu mleka. [Quality changes and seasonality of milk purchase]. RN SERiA, vol. IV, issue 3, Warsaw [in Polish].

6. Okularczyk S., 2004: Opłacalnosc produkcji mleka krowiego w latach 1991 – 2003. [Profitability of caw milk production in years 1991 – 2003]. Wies Jutra No 11-12, Warsaw [in Polish].

7. Parzonko A., Ziętara W., 1999: Organizacja gospodarstw mlecznych w Polsce i w wybranych krajach [Organisation of milk farms in Poland and in chosen countries] Przeglšd Hodowlany No 2, Warsaw [in Polish].

8. Reklewski Z., Runowski H., 2004: Poradnik producenta mleka. [Milk Producer Guide]. Ministry of Agriculture and Rural Development, FAPA, SAPARD, Warsaw [in Polish].

9. Sawicka J., 2003: Programowanie rozwoju regionalnego według standardów UE [Regional development programming in adjustment to the EU standards]. Zeszyty Naukowe SGGW Ekonomika i Organizacja Gospodarki Żywnosciowej No 48. SGGW, Warsaw [in Polish].

10. Stańko S., 1973: Efektywnosc chowu bydła w gospodarstwach indywidualnych. Ocena w oparciu o koszty bezposrednie, dochód bezposredni i koszty specjalne na przykładzie województwa białostockiego. [Cattle breeding effectiveness in private farms. The evaluation based on direct costs, direct income and special costs as a Białostockie voivodship example] Rozprawa doktorska, SGGW [PhD thesis, Warsaw Agricultural University], Warsaw [in Polish].

11. Sznajder M., 2004: Globalne tendencje w rozwoju gospodarstw mlecznych [Global tendencies in milk farms development]. Wies Jutra No. 11-12, Warsaw [in Polish].

12. Wilkin J., 2004: Dlaczego potrzebujemy długofalowej strategii zintegrowanego rozwoju wsi i rolnictwa w Polsce [Why do we need the long-term strategy of integrated country and agriculture development in Poland.] Wieœ i Rolnictwo No. 2, Warsaw [in Polish].

13. Ziętara W., 2003: Przemiany w rolnictwie polskim w latach 1990 – 2001 [Changes in Polish agriculture in years 1990 – 2001]. RNR, series G Ekonomika Rolnictwa vol. IV, b. 1, Warsaw [in Polish].

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.