Introduction

Egg consumption has been expanding and is increasing at the global level, due to its high protein content, and because of its broad availability in the market, easy preparation and competitive price with regards to other products that are rich in proteins of animal origin. From 1970 to 2012 the global average per capita consumption grew from 11.51 to 16.27 kilograms and showed an average annual growth rate (AAGR) of 0.83 %. In this period the per capita consumption in México increased from 7.11 to 20.8 kg; in China, from 1.89 to 20.4 kg; in Singapore, from 8.19 to 18.8 kg; in Colombia, from 4.79 to 14.6 kg; and in Argentina, from 7.58 to 12.6 kg. In the meantime, in the United States, it decreased from 19.72 to 15.5 kg; in Nueva Zealand, from 18.46 to 13.8 kg; and in Japan, from 16.92 to 16.3 kg (^{FAO, 2013}; ^{UNA, 2013)}. The growth in per capita egg consumption in México can be explained by its competitive price compared to other sources of protein; thus, with one minimum wage an average of 2.45 kilograms of eggs can be acquired, as well as 1.84, 0.69 and 0.70 kilograms of chicken, pork and beef meat (^{UNA, 2013)}.

The increase in egg consumption in the world and in México is due to the tastes and preferences of the population for this food, and to its energetic and protein contribution. In addition, in a study of 14 years in which 120 thousand people were included, ^{Hu et al. (1999}) found that there is not a higher cardiovascular risk among individuals who consume more than one egg per day, situation that is different among those who suffer from diabetes. On the other hand, when analyzing 30 thousand people during six years, ^{Scrafford et al. (2011}) found that among adult people who consume more than one egg daily, the risk of diabetes was reduced.

On the other hand, ^{Jillion et al. (2005)} have mentioned that a breakfast with egg produces greater satiation and reduces significantly the energy consumption from other sources that may induce greater obesity, so it is possible that ingesting it does not contribute to generating overweight.

^{Tomek and Kaiser (2014}) point out that the demand function for a food is determined by the variations in the consumer price of the product, the income available and its distribution, the prices of substitute and complementary products, and the tastes and preferences of the consumer. With historical series from 1960 to 1987 and from 1960 to 2003, and with minimum squares estimations in two stages, ^{López (1990)} and ^{Rojas (2005)} obtained elasticity coefficients for egg. For products related with it, the first author used ham and milk; the second, only milk; and in this study pork meat was used, as well as cheese, chili and bean. According to ^{Stamer (1969}), this modifies the order of magnitude of the coefficients of the demand elasticity estimated in the three studies.

According to data by ^{López (1990)} and ^{SIACON (2013)}, national egg production presented an accelerated growth, going from 365 745 tons in 1970 to 2 386 576 tons in 2012, showing an AAGR higher than that of the population (1.98 %). With a production of 108.5 million boxes of eggs obtained in 2012 (^{UNA, 2013)}, México was the sixth global producer, after China, USA, the European Union, India and Japan, which in the same order produced 1090, 218.4, 181.4, 178.3 and 114.3 million boxes. The national per capita egg production in México in 1970 and 2012 was 7.58 and 21.69 kilograms, amounts that supply the national consumption in 99.9 %, making the country self-sufficient in this food. The main producing states in México are Jalisco (55 %), Puebla (15 %), Sonora (8 %), the Comarca Lagunera region (Durango and Coahuila) (5 %), Yucatán (4 %), Sinaloa (3 %), Nuevo León (2 %) and Guanajuato (2 %) (^{UNA, 2013)}, where eleven of the principal producing companies are located, led by Proteína Animal, Bachoco, El Calvario, Empresas Guadalupe, Socorro Romero Sánchez, Granjas Avícolas Rancho Grande, José Asunción, Productora de Huevos Gigantes, Gena Agropecuaria, Avícola González y González and Impulsora Agropecuaria Santa Rosa, which create an oligopoly market that allows them to impose the sales price to its first buyer. The large egg producing companies increased their participation in production, from 29 % in 2006 to 45 % in 2012; the medium ones decreased from 50 % to 36 %, and the small ones from 21 % to 19 % (^{UNA, 2013)}. The large companies are becoming integrated vertically, because they maintain egg production and commercialization linked; this means that the prices are intertwined through direct property or through product contract. Oligopoly concentration of production into large companies and their vertical integration allows them to reduce unitary production and commercialization costs in what refers to egg packaging, transport, and distribution of the product wholesale, where a high percentage is executed by large producers, using the warehouses that they have in the country's wholesale supply centers. This decreases the trade margins and the prices that consumers pay, without harming the profit margins of producers and traders.

^{López (1990)} included as explicative variables of the offer the price of balanced meal, which he forced to depend on the price of sorghum and soy; the producer price of egg, which he related with the official price; and the technology, which he represented with a trend variable. ^{Rojas (2005)} incorporated directly as explicative variables of the offer the price of balanced meal, technology -which he approximates to the food conversion to kilogram of egg -, the producer price of egg, a variable of trend, and one of classification with which he separates the periods of closed (1960-1985) and open (1986-2010) economy. In addition to the producer price of egg, this study includes the producer price of chicken as product that competes with egg, and balanced meal, technology, a classification variable, and one trend variable.

It should be highlighted that balanced meal constituted 69 % of the total production cost of one kilogram of egg in 2012 (^{UNA, 2013)}. From 1970 to 2012, the requirements of balanced meal to obtain one of egg went from 2.8 to 2 kilograms, indicating that the rate of conversion increased from 0.36 to 0.50 kilograms of egg per one of meal. The technological progress caused the decrease in the real price of balanced meal to an AAGR of 2.08 % per year and the production grew at an AAGR of4.57 %, despite the real producer prices of egg decreasing each year by 3.42 %.

Egg production is an important source of employment. Between 1994 and 2012, it generated 775 thousand and 194 thousand direct jobs, and 115 thousand and 972 thousand indirect jobs. Likewise, it is an activity that is highly demanding of agricultural inputs (fodder grains and oleaginous pastes). With regards to the consumption of fodder grains, in 1994 and 2010 egg-producing poultry farming consumed 2 516 028 and 4 155 546 tons of sorghum and maize.

From 1970 to 2012, the real sorghum prices per ton decreased from $3095.24 to $1958.64 (36.72 %); those of balanced meal went from $7142.86 to $3540.78 (50.42 %), and the real producer price

of egg fell from $24 615.38 to $5730.64 per ton (76.72o%). The real producer price of egg compared to the real value of balanced meal, which is required to produce one kilogram of egg, went from 1.524 to 1.596 in the corresponding years. This indicates that, relatively, the price of egg increased more than that of the balanced meal (^{UNA, 2013)}.

The objective of this research is to explain the functioning of the egg market in México; that is, to estimate the functions of offer and demand, and to determine the variables that condition them. The hypothesis is that the demand of this product in México is determined by real consumer prices of egg, pork meat, cheese, chili, bean and real income per capita, while the offer was determined by the real producer prices of egg, chicken and balanced meal, as well as technology, in addition to the effect of the import price of egg which is reduced in the national market of the product, because the country is self-sustainable in this food.

Methodology

^{Stamer (1969}) indicates that the amount of an agricultural product offered in the market depends first on the farmers' expectations of benefit (B); if these estimate high profits for the next economic period, the production and the amount offered will increase, and vice versa. Since the benefit B= Q(P-CMT) depends on the amount of product sold (Q), the product price (P), and the total average production costs (APC), then the future offer is determined by the expected price of the product and the evolution of costs. ^{Stamer (1969)} and ^{Tomek and Kaiser (2014}) point out that the offer of an agricultural product is determined, in addition to variations in the price expected by the producer of the good under study, the prices of inputs or production factors, the state of the technique that is given by the shape of the production function, the prices of competitive and coupled products, and the institutional restrictions.

The population in México consumes egg combined with green chili, onion, tomato, cheese and beans; they also eat it with chorizo, ham and bacon. In the estimation, these products had the sign expected and significant coefficients, as complementary goods, green chili, cheese and beans, the three latter were captured with the consumer price of pork meat, which was complementary to egg.

The production cost of egg approached the price of balanced meal which, as such, includes the prices of fodder grains, sorghum and maize, soy paste, labor, and technology used for its manufacture. Technology is the factor that has driven most the growth of egg production in México, which through genetic engineering has originated a hen that produces a higher number of eggs with the same amount of food and other inputs and, therefore, with the same total cost; that is, productivity per bird has increased and the production costs have reduced.

An econometric model of simultaneous equations of the egg market in México was formulated and estimated, in which the main economic variables that determine the offer and demand functions are involved, taking into consideration the transmission of the import price into those of wholesale, producer and consumer, and, as consequence, on the trade balance. The model specified represents the functioning of the internal market, taking into account the influence of the external market during the 1960 to 2012 period. Estimating the coefficients was done through the minimum square method in two stages, with the SYSLIN procedure of the ^{SAS software package (Statistical Analysis System, v.9.0.)}. Its statistical coherence was determined through a coefficient of determination (R²), which indicates the goodness of fit in each one of the equations estimated; the global significance of the coefficients of each equation was observed with the F test and that of each coefficient with the t Student or t frequency. The economic aspect of the model was validated according to the signs expected indicated by the economic theory of the coefficients of each equation, and by the magnitude of the elasticities (^{García et al., 2004}; ^{Pérez et al., 2010}; ^{García et al., 2013}).

Information for the period of 1970 to 2012 was obtained from ^{López (1990)} and ^{Rojas (2005)}, the National Poultry Producers' Union (Unión Nacional de Avicultores, ^{UNA, 2011, 2012 and 2013)}. Other statistics were obtained from the United Nations Food and Agriculture Organization (^{FAO, 2013}), National Information and Market Integration System (Sistema Nacional de Información e Integración de Mercados, ^{SNIIM, 2013)}, National Institute for Statistics and Geography (Instituto Nacional de Estadística y Geografía, ^{INEGI, 2013}), México's Central Bank (Banco de México, ^{Banxico, 2013}), and the Agrifood Information System for Consult (Sistema de Información Agroalimentaria de Consulta, ^{SIACON, 2013)}.

The econometric model was constituted by five functional relations and one identity, as indicated next:

QPH_t = α ₁₁ + α ₁₂PRHP_t + α ₁₃PPCRP_t + α ₁₄PRALIBA_t + α ₁₅TECN_t + α ₁₆TE_t + α ₁₆D_t + α _1t

PRHP_t = α ₂₁ + α ₂₂PRHMAY_t + α ₂₃CTHRT_t + α _2tPRHMAY_t + α ₃₁ + α ₃₂PIMPHR_t + α ₃₃PRHMM_t + α _3t

PRHC_t = α ₄₁ + α ₄₂PRHMAY_t + α _4t

QDPER_t = α ₅₁ + α ₅₂PRHC_t + α ₅₃YNDRPER_t + α ₅₄PCCCR_t + α ₅₅PCQR_t + α ₅₆PCHISR_t + α ₅₇ PCFR_t+ α _5t

SCE_t = QDH_t - QPH_t

The endogenous variables of the model are: amount of egg produced (QPH_t) (t), real producer prices of egg (PRHP_t) ($/t), wholesale prices (PRHMAY_t) ($/t), consumer prices (PRHC_t) ($/kg), amount of egg demanded per capita (QDPER_t) (kg); and foreign trade balance (SCE_t) (t). The predetermined variables of the model are: real import price of egg (PIMPHR_t) ($/kg), for medium-wholesale (PRHMM_t) ($/kg), for chicken meat producer (PPCPR_t) ($/t), of balanced meal (PRALIBA_t) ($/t), for consumer of pork meat (PCCCR_t) ($/kg), of cheese (PCQR_t) ($/kg), of chili (PCHISR_t) ($/kg), of beans (PCFR_t) ($/kg), real cost of egg transport (CTHR_t) ($/t), technology (TECN_t) (food conversion: grams of egg per one of meal); and real per capita income (YNDRPER_t) (thousands of $/ año). The nominal values were transformed into real values with the appropriate price indexes.

Results and Discussion

The five regression equations that make up the estimated model showed a goodness of fit acceptable in terms of the coefficient of determination (R²), which is above 0.92 for all the relationships. The results estimated in their structural and reduced form are presented in Tables 1 and 2. The global F test was significant at for all the functional relations. According to the individual significance of each parameter from the five equations, all the explicative variables were significant ("t value" higher than one).

The signs of the coefficients of the variables of the five equations that make up the structural model estimated agree with the economic theory. The elasticities were calculated with the partial derivatives of each equation and the average values of the variables for the 1960-2012 period, with the aim of quantifying the effects of the explicative variables on the endogenous ones. The analysis of elasticities implicitly includes the concept of ceteris paribus, that is, that the effect of one variable in particular is such, insofar as the other variables remain constant (Table 3).

Conclusions

The price elasticity of the offer indicates a direct and inelastic relationship between the amount offered and the real producer price of egg. This is why the continuous fall in the producer price of this food has had a small effect on the decrease of the amount offered. On the other hand, the demand-price of egg has an inverse and inelastic relation; because of this, the continuous decrease in the consumer price of this product has caused for the amount demanded to increase slightly.

The variables that determine the egg offer in México are the producer prices of egg, chicken, balanced meal, and technology, which were inelastic, with technology being the variable that stimulates most the growth of egg production in México.

The egg demand in México is determined primarily by the income and the real consumer prices of egg, pork meat, cheese, chili and beans, products which were complementary to the consumption of this food.

The continuous decrease in the consumer price and the increase in income are the factors that have driven the growth in egg demand in México, to the greatest extent.

The increase in the consumer price of egg of 73 %, which was seen from July to August 2012 as a result of the impact of the avian flu outbreak, should have been 45.78 %; the additional increase to the price was due, to a great extent, to speculative issues.