SciELO - Scientific Electronic Library Online
 
vol.15 número4Prevalencia y diversidad de parásitos intestinales zoonóticos en perros domésticos en un área urbano en el Caribe colombianoCompetitividad y ventaja comparativa de ganado bovino de carne en la Sierra Norte de Puebla, México índice de autoresíndice de materiabúsqueda de artículos 		Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

  • No hay artículos similaresSimilares en SciELO

Compartir

Revista mexicana de ciencias pecuarias

versión On-line ISSN 2448-6698versión impresa ISSN 2007-1124

Rev. mex. de cienc. pecuarias vol.15 no.4 Mérida oct./dic. 2024  Epub 21-Mar-2025

https://doi.org/10.22319/rmcp.v15i4.6609 

Articles

Spatial and vertical transmission of milk prices from the international market to Mexico’s regional and national markets

Benigna González Ortiza 

José Luis Jaramillo Villanuevaa  * 

Samuel Vargas Lópeza 

Ángel Bustamante Gonzáleza 

Juan de Dios Guerrero Rodrígueza 

Vinicio Horacio Santoyo Cortésb 

aColegio de Postgraduados, Campus Puebla. Boulevard Forjadores de Puebla No. 205, Santiago Momoxpan, Municipio de San Pedro Cholula, 72760. Puebla, México.

bCIESTAAM. Centro de Investigaciones Económicas, Sociales y Tecnológicas de la Agroindustria y la Agricultura Mundial. México.

Abstract

When a country imports a good or service, it is subject to prices determined by the world market; therefore, domestic market prices change when international prices do. This study aimed to estimate the degree of price transmission between the producer price of milk in Mexico at the national and regional levels and that of the United States (spatial transmission) and between the retail price of milk and the producer price in Mexico (vertical transmission). An econometric analysis of monthly time series of milk prices from January 1990 to December 2021 was performed, applying unit root tests, cointegration tests, and an error correction vector model. The results indicate that there is a long-term relationship between United States prices and producer prices at the national and regional levels, as well as between the retail price and the producer price. It was found that the spatial transmission of international prices to the producer price at the national level and in the regions of Jalisco and Veracruz is symmetrical and asymmetrical with the producer price in the state of Coahuila. There are differences between regions in the speed of adjustment when international prices increase and when they decrease. The vertical transmission was also symmetrical, unidirectional, from the producer to the retail market, and incomplete.

Keywords Price transmission; Milk market; Error correction model

Resumen

Cuando un país importa un bien o servicio se sujeta a los precios determinados por el mercado mundial, por lo tanto, los precios del mercado interno cambian cuando los precios internacionales lo hacen. El objetivo de este estudio fue estimar el grado de transmisión de precios entre el precio al productor de leche en México a nivel nacional y regional y el de Estados Unidos (transmisión espacial), así como entre el precio de la leche al menudeo y el precio al productor en México (transmisión vertical). Se realizó un análisis econométrico de series de tiempo mensuales de los precios de leche de enero de 1990 a diciembre de 2021, aplicando pruebas de raíz unitaria, de cointegración y un modelo de vectores de corrección de error. Los resultados indican que existe una relación a largo plazo de los precios de Estados Unidos con los precios al productor a nivel nacional y regional, así como entre el precio al menudeo y el precio al productor. Se encontró que la transmisión espacial de los precios internacionales al precio al productor a nivel nacional y regional en Jalisco y Veracruz es de forma simétrica, y asimétrica con el precio al productor en el estado de Coahuila. Existen diferencias entre regiones en la velocidad de ajuste cuando los precios internacionales aumentan que cuando disminuyen. La transmisión vertical también fue de forma simétrica, unidireccional del productor al mercado minorista e incompleta.

Palabras clave Transmisión de precios; Mercado de la leche; Modelo corrección de error

Introduction

Historically, Mexico has been an importer of powdered milk because domestic production is insufficient to cover domestic demand. In 2022, imports accounted for 23.3 % of national consumption1. In 2021, imports of powdered milk represented 40 % of the total value of dairy imports. Mexico’s main supplier of these imports is the United States2,3,4. For the national dairy subsector, imports historically exceed exports, i.e., the trade balance is in deficit2,3.

When a country imports a good, it is subject to prices determined by the world market; therefore, domestic market prices change when international prices do. In an import scenario, those who benefit are the consumers, and these benefits represent a loss for the producer because the imported products are obtained in the international market at lower prices than what is paid in the domestic market5,6. The United States can offer competitive prices of powdered milk to the international market since it has high productivity: the yield (liters of milk/cow/d) in Mexico in 2021 was 13.3, and for the United States in the same year, it was 29.787. In addition, both production and marketing in the United States receive high subsidies, which distorts product prices in world markets3,8.

In Mexico, there is a high concentration of both production and industry. Fifty-three (53) percent of milk production is concentrated in four states, Jalisco (21 %), Coahuila (11.3 %), Durango (11.4 %), and Chihuahua (9.4 %)2; within the milk chain, in the industrialization link, there are 130 companies that process 86 % of the national production, with an employed staff of 42 thousand people9. The dairy industry is not only an important source of price change but also acts as a mediator of price signals originating from different parts of the food chain10.

The national milk industry is an oligopoly characterized by a high degree of concentration of firms, significant barriers to entry, and dynamic product differentiation11. In the latter, prices determine the allocation of resources and the production decisions of economic agents12.

Price transmission is the process through which information is transmitted between market participants10; its study allows to know if the markets are integrated. A theoretical framework used in the literature is the law of one price, which states that when there is a commercial exchange between two spatially separated regions, under conditions of perfect competition, price shocks in one market are transmitted completely and symmetrically to the other market and equilibrium prices differ only in transfer costs13-16. Markets that transmit price information quickly and comprehensively are said to be perfectly integrated and often efficient13.

Specifically, price transmission studies analyze the form and speed of adjustment of domestic prices in the face of changes in international prices14-17. The speed with which prices are transmitted, the magnitudes of the transmission, and the non-linear behavior in the transmission of prices are indicators of market inefficiency18.

The results obtained through a meta-analysis19 suggest that asymmetric price transmission in producer-retailer relationships is more likely in sectors in countries with a more fragmented agricultural structure and greater government support. In Mexico, studies have been carried out on the transmission of fluid milk prices from the international market to the national market and even to the regional and local markets14,15, but in a single geographical point, and there is no study of transmission in the largest producing regions, both spatial and vertical. This study aimed to estimate the degree of spatial transmission between United States milk import prices and the producer price of milk in Mexico at the national level and in the main producing regions and between the retail price of milk and the producer price at the national level (vertical transmission).

Material and methods

The econometric analysis performed used monthly time series of fluid milk prices from 1990:01 to 2022:12. The data for Mexico at both the national and regional levels are those paid to the producer (average rural price), obtained from the website of the Agri-Food and Fisheries Information Service of the Secretariat of Agriculture20. The regional prices correspond to those paid to the producer in the states of Jalisco, Coahuila, and Veracruz. The states that rank first in the production of bovine milk in their respective regions were selected; Jalisco is the first producing state in the Central-Western region and ranks first nationally (21 %), Coahuila is the first producing state in the Northeast region, ranking second nationally (11.8 %), and Veracruz is the first producing state in the South-Southeast region and ranks sixth nationally (6 %)21.

The international milk price corresponds to the export prices of skimmed milk powder from the United States to Mexico, transformed into milk equivalent, obtained from the USDA-AMS website22. Consumer prices correspond to the average consumer price per liter of pasteurized fluid milk provided by the National System of Information and Integration of Markets23 of the Secretariat of Economy and adjusted by the national consumer price index24. The price series were expressed in dollars per liter using the exchange rate published by the Bank of Mexico25. The data were transformed into natural logarithms in order to perform the econometric analysis and interpret the coefficients as elasticities.

The stationarity tests used were the Augmented Dickey-Fuller (ADF)26 and Phillips-Perron (PP)27 tests, with which the order of integration of each series was verified. The long-term relationship was then estimated using two-stage cointegration28 and confirmed with the Johansen test29. Finally, it was estimated an Asymmetric Vector Error Correction Model (AVECM), a test to select the order of lag for an AVECM, and an F-test for equality of ECT+ and ECT- coefficients (positive and negative changes in the error term, respectively). The null hypothesis (H0) of symmetry is rejected if β2+ and β2- (price adjustment coefficients) differ significantly.

Cointegration tests

The cointegration test was applied in both the spatial and vertical price transmission models. The cointegration between variables - once the existence of unit roots has been demonstrated - is a necessary condition for the existence of a long-term equilibrium relationship in the series. A vector of variables that have a unit root is cointegrated if a linear combination of these variables is stationary26. The two-step Engle-Granger cointegration test28 and the Johansen test29 were used to test the long-term relationship. The first approach is to estimate the cointegration regression (Equation 1) using OLS:

ptout=α+β1ptin+μt (1)

Where ptout is a firm’s production price in period t, ptin and is the price of inputs in period t. The residual Ût is obtained from the cointegration regression, and an ADF unit root test is applied to it. The failure to reject the null hypothesis of non-stationarity allows us to estimate Equation 2.

∆μt=α+β1μt-1+β2∆μt-1 (2)

A negative coefficient of the error term (between -2 and zero) confirms a long-run relationship between the milk price paid to the producer and the international milk price. On the other hand, the Johansen test derived the distribution of two test statistics for the null hypothesis of non-cointegration: the tests of eigenvalues and trace29. Once the cointegration between prices was verified, an Error Correction Model (ECM) was applied to capture the short- and long-run effect of ptin on ptout, and the rate of adjustment at which ptout returns to equilibrium after a change in ptin.

Asymmetric spatial transmission of prices between international and domestic prices

Considering that producer and international prices are cointegrated, an Asymmetric Vector Error Correction Model (AVECM) was estimated to investigate the possible interdependence of import prices on domestic prices (spatial transmission). The division of the Error Correction Term (ECT) into positive and negative components allowed to verify the existence of asymmetric price transmission; following this approach, Equation 3 was used to study spatial asymmetric price transmission12.

∆ptfarm=α+∑j=1k(βj+ D+ ∆pt-j+1int)+ ∑j=1Lβj- D- ∆pt-j+1 int+ ∅+ ECTt-1 ++ ∅-ECTt-1  -+γt (3)

Where Δ is the first difference of the operator; Ptfarm = producer price; Ptint =import price, D+ and D- are fictitious variables by means of which the price of inputs is divided into a variable that includes only increasing prices and another that includes only decreasing prices; βj+ and βj- are price adjustment coefficients; ø+ and ø- are constant parameters; ECT is the error correction term. An F-test was used to verify the null hypothesis of symmetry.

Asymmetric vertical price transmission

Equation 4 was used to estimate asymmetric vertical transmission and an F-test was used to verify the null hypothesis of symmetry12.

∆ptfarm=α+∑j=1k(βj ∆pt-j+1int)+ ∅+ ECTt-1 ++ ∅-ECTt-1  -+γt (4)

Results and discussion

According to the results of the ADF and PP unit root tests on the price series, the statistical value of t does not allow to reject the null hypothesis of unit root with a confidence level of 95 %, i.e., the price series are non-stationary (Table 1). Recent studies of milk price transmission obtained similar results of non-stationarity between time series15,17,30. The non-stationarity result of the time series justifies the use of cointegration tests. Cointegration allows a combination of non-stationary variables to be stationary. It can be seen as a long-term equilibrium relationship between variables despite the fact that in the short term, they go through situations of disequilibrium31. To determine if the series are cointegrated (long-term equilibrium), the residuals (ut) of the cointegration regression must be stationary; this is achieved by applying the ADF test to determine the stationarity of the time series.

Table 1 Results of ADF and PP tests on the milk price series corresponding to imports, domestic producer, regional producer, and consumer 

Price series ADF test 5%
critical value		 PP test 5% critical
value
Import price -2.328 -3.425 -20.865 -21.406
Consumer price -2.032 -3.425 -13.350 -21.406
Production price, National -3.409 -3.425 -37.672 -21.406
Production price, Jalisco -3.339 -3.425 -24.447 -21.402
Production price, Coahuila -2.830 -3.425 -15.919 -21.402
Production price, Veracruz -3.301 -3.425 -22.482 -21.402

Long-term co-integration of the national and regional spatial model

The results of the ADF test of the error term indicate that the null hypothesis of non-stationarity is rejected (Table 2), which means that the import price series is cointegrated in the long run with both the national and regional price series.

Table 2 Results of the ADF test of the error term 

Price pairs ADF test of the error 5% critical value
PImp-Pnational prod -4.174 -2.875
PImp-PJal -3.450 -2.875
PImp-PCoah -3.344 -2.875
PImp- PVer -3.598 -2.875

Engle and Granger28 confirmed a long-term relationship between the producer milk price at the national level and the milk import price (Table 3), as well as between the producer price in Jalisco, Coahuila, and Veracruz and the international price (Table 4).

Table 3 Results of the two-step Engle-Granger cointegration test of the spatial model between the import price and the domestic producer price 

Variable Coefficient Standard error t-value P>|t|
μt-1 -0.118624 0.0179831 -6.60 0.000
∆μt-1 .4825031 0.044505 10.84 0.000
Constant 0.0000821 0.0021071 0.04 0.969
F-test 69.26
R-squared 0.2677

Table 4 Results of the two-stage Engle-Granger cointegration test of the spatial model between the import price and the regional producer price 

Jalisco Coahuila Veracruz
Variable Coef. SE t value P>|t| Coef. SE t value P>|t| Coef. SE tvalue P>|t|
μt-1 -0.0657 0.0169 -3.87 0.000 -.0640 .0169 -3.77 0.000 -.0692 .018 -3.82 0.00
∆μt-1 .1426 0.0510 2.79 0.006 .1366 .0512 2.67 0.008 .0762 .051 1.48 0.13
Constant -0.0005 0.0024 -0.21 0.831 -.0004 .0021 -0.21 0.832 -.0007 .002 -0.29 0.77
F-test 9.94 9.27 7.66
R-squared 0.506 0.42 0.39

SE= standard error.

Like the Engle-Granger test, the Johansen test determines the existence of a stable and long-term equilibrium relationship. The results of the Johansen test of cointegration of the import and domestic producer price series yielded a trace statistical value (5.9169) higher than the critical value of 5 % (3.76), as did the values for the series of prices in Jalisco, Coahuila, and Veracruz (6.8901, 4.7077, 5.0788, respectively); therefore, the null hypothesis of non-cointegration between prices is rejected, i.e., import prices influence the behavior of producer prices at the national level and in the states of Jalisco, Coahuila, and Veracruz in the long term.

Similar studies14,15 conducted in Mexico with milk price data confirmed the long-term cointegration between the import price and the price paid to the producer, suggesting that the import price influences the behavior of producer milk prices in the long run. In addition to the long-term relationship of Mexican prices with prices in the United States, this relationship with prices in Oceania and the European Union was also verified15. This relationship of cointegration of import prices with domestic prices does not always follow a long-term relationship, especially in countries where exports are higher than milk imports16.

Given the confirmation of cointegration between the import price and the producer price of milk at the national and regional levels, an Error Correction Model (ECM) was estimated, which relates the changes in Ptint with the changes Ptfarm in the case of the spatial model and the so-called error correction term (ECT), the lagging residuals of the cointegration equation were calculated.

Spatial vector error correction model: national

The values of ECTt-1+ and ECTt-1- reflect that when the import price changes, the producer price at the national level changes in different proportions when it increases compared to when it decreases. When import prices increase, producer prices increase by 8 %, but when import prices decrease, producer prices decrease by 14 % (Table 5).

Table 5 Results of the spatial vector error correction model (VECM) of the import price series and producer price series at the national level 

Independent variable Symmetric spatial model Asymmetric spatial model
Coef. SE t p>|t| Coef. SE t p>|t|
Pint 0.1140 0.0365 3.12 0.002 --- --- ---
Pintt- --- --- --- 0.2438 0.1004 2.43 0.016
Pintt+ --- --- --- 0.2493 0.0949 2.63 0.009
Pfarmt-1 0.5741 0.0486 11.82 0.000 0.5685 0.0486 11.68 0.000
Pfarmt-2 0.1047 0.0503 -2.08 0.038 -0.097 0.0504 -1.94 0.053
0.0105 0.0402 0.26 0.793 0.0149 0.0459 0.33 0.745
Pintt-2 -0.0428 0.0376 -1.14 0.255 -0.0385 0.0376 -1.03 0.306
ECTt-1 -0.1135 0.0182 -6.22 0.000 --- --- ---
ECTt-1+ --- --- --- -0.0861 0.0235 -3.66 0.000
ECTt-1- --- --- --- -0.1487 0.0262 -5.66 0.000
Constant 0.00031 0.0019 0.16 0.875 -0.0003 0.0019 -0.16 0.876
Normality: (Prob>z) 0.963 0.96389
LM test (Prob>Ji2) 0.863 0.049
DW test 2.012 1.9148
R-squared 0.3937 0.3985
Test: H0=β1+ = β1- --- F(1,374) = 2.35
Test: H0=β2+ = β2- --- F(1, 372) = 3.41

SE= standard error.

Similar results are reported in Chile32 in a study of the spatial transmission of international prices to domestic prices, where negative effects or shocks are passed on more quickly than positive effects, which could be explained by the oligopsonic structure in the reception of fluid milk. As reported in previous studies, for the 1990-2016 period14, in this study, the contemporaneous exchange coefficients are significantly less than one in both equations, indicating that producer prices do not fully react in one month to changes in international prices. The F-test indicates that the null hypothesis of symmetry H0=β2+ = β2- is accepted at a significance level of 5 % (Prob ˃ F=0.065).

Studies of milk price transmission in Mexico are scarce14,15 and have yielded different results in error correction models. In the present study, the speed of adjustment of the national price in the event of deviations from equilibrium shows a lower value than in the preliminary studies (-0.113). The evidence of symmetry in the spatial transmission of prices at the national level differs from that found by previous studies in Mexico for the period 1990-201614. This is probably because the period of analysis is different. Nonetheless, the findings of the present research coincide with other recent research in Mexico15 for the period 2001-2019. On the other hand, in a similar study carried out in Chile32, they found evidence of asymmetry in the spatial price transmission between markets.

Spatial vector error correction model: regional

The values of the coefficients associated with the ECT of the regional models were negative and significant. The result of the F-test indicates that the null hypothesis of symmetry: H0=β2+ = β2- is accepted at the significance level of 5 % for the response of producer prices in Jalisco (F=1.01) and Veracruz (F=0.02); however, it is rejected for the case of Coahuila (F=7.71; P˃F= 0.0058), suggesting evidence of asymmetry in the price response in this state (Table 6). The symmetry results for the transmission of international prices to the producer in Jalisco coincide with what was reported by a recent study carried out in the same region15, in which they analyzed the transmission of prices from three international markets to the national market, regional market in Jalisco and local market in Chicontepec, Veracruz.

Table 6 Results of the Regional Asymmetric Spatial Model 

Independent variable Jalisco Coahuila Veracruz
Coef. SE t p>|t| Coef. SE t p>|t| Coef. SE t p>|t|
Pintt- -0.0326 0.126 -0.26 0.796 -0.0028 0.105 -0.03 0.978 -0.0258 0.134 -0.19 0.847
Pintt+ -0.0284 0.119 -0.24 0.796 0.0029 0.099 0.03 0.976 -0.0318 0.126 -0.25 0.801
Pfarmt-1 0.1563 0.052 2.99 0.003 0.1086 0.052 2.07 0.039 0.0838 0.053 1.59 0.114
Pfarmt-2 -0.0337 0.052 -0.65 0.516 -0.0182 0.051 -0.36 0.723 -0.0029 0.052 -0.06 0.954
Pintt-1 0.0648 0.057 1.13 0.259 0.0768 0.047 1.62 0.107 0.0762 0.061 1.25 0.210
Pintt-2 -0.0220 0.046 -0.48 0.631 -0.0158 0.038 -0.42 0.678 -0.0470 0.048 -0.97 0.334
ECTt-1+ -0.0497 0.023 -2.14 0.033 -0.1037 0.022 -4.70 0.000 -0.0737 0.024 -3.10 0.002
ECTt-1- -0.0847 0.027 -3.17 0.002 -0.0105 0.025 -0.42 0.674 -0.0690 0.028 -2.47 0.014
Constant -0.0008 0.002 -0.33 0.743 -0.0008 0.002 0.39 0.697 -0.0004 0.003 -0.17 0.867
Normality test (Prob>z) 0.8032 0.8554 0.7778
LM test (Prob>Ji2) 1.546 1.706 0.303
DW test 1.511 1.668 0.296
R-squared 0.637 0.855 0.550
Test: H0:β2+=β2- F(1, 366) = 1.01 F(1, 366) = 7.71 F(1, 366) = 0.02

SE= standard error.

The present study also found differences in the transmission of producer prices from one region to another; the values of ECTt-1+ and ECTt-1- reflect that when the import price changes, the producer price in Jalisco, Coahuila, and Veracruz changes in different proportions when it increases compared to when it decreases.

Unlike national and regional (Jalisco) prices, import prices affect prices in the state of Coahuila in different ways; when import prices increase, producer prices in Coahuila increase by 10 %, but when import prices decrease, producer prices decrease by 1 %, i.e., the speed of adjustment is significantly greater when prices rise than when they decrease. The speed of adjustment for producer prices in Veracruz did not show significant differences when prices rise or when they fall (Table 6).

These variations in price transmission and in the response of producer prices in the three regions to changes in import prices may be associated with the market structure within each region. Coahuila is part of La Comarca Lagunera, which together with the state of Durango contribute 22.5 % of the national production21; in this region, the specialized system predominates and the market structure is of the oligopsony type; here are the two companies with the largest share in the dairy market in Mexico, LALA and Alpura. Both companies maintain a close relationship with their partners from whom they buy milk at a comparatively high price33, which could explain why positive changes in import prices are reflected faster than negative changes in producer prices. The importance of the study of market structure as part of the analysis of price transmission in the milk market has already been addressed in other studies34. Studies carried out in Mexico35 and in the milk market between countries that make up a sector at the international level36 have emphasized the need to consider the analysis of price transmission at the regional level since differences in the response to price transmission between regions can be identified.

Long-term cointegration of the vertical model

For the vertical transmission model between the producer price of milk (Pprod) and the retail price (Pcon), the hypothesis that the retail price is caused by the producer price was tested. Since the time series were non-stationary, was necessary to proceed to perform the long-term cointegration tests using Equation (1).

The results of the estimation of Equation (1) showed an R2 of 0.14, a statistical value of t of 16.72, and a statistical value of F of 279.58. The ADF test of the error term showed a test statistic of -3.646, compared to the 5 % critical value of -2.875, indicating that the null hypothesis of non-stationarity is rejected. The results of the two-stage Engle-Granger cointegration test show a negative coefficient of error, confirming the long-term relationship between prices (Table 7).

Table 7 Results of the two-stage Engle-Granger cointegration test for the vertical price transmission model 

Variable Coefficient Standard error t-value P>|t|
μt-1 -0.0963809 0.0155752 -6.19 0.000
∆μt-1 0.5261937 0.0435508 12.08 0.000
Constant 0.000011 0.0021777 0.01 0.996
F-test 81.75
R-squared 0.3014

A study of vertical price transmission in the milk market in Russia37 found that there is no long-term cointegration relationship between producer prices and retail prices; nevertheless, a change in retail price has a significant effect on producer price and vice versa, i.e., there is a bidirectional effect.

The value of the trace statistic (3.3776) of the Johansen test was less than the 5 % critical value (3.76), which does not allow us to reject the null hypothesis of cointegration, i.e., it confirms that the price series are cointegrated.

Vertical vector error correction model

Once the cointegration of retail and producer milk prices was verified, a Vector Error Correction Model12 was estimated and an F-test was used to test the null hypothesis of symmetry. Error correction models allow to quantify what proportion of the price is transmitted throughout the marketing chain and the speed with which this occurs12. The result of the F-test indicates that the null hypothesis of symmetry (H0: β2+ =β2-) is accepted at the significance level of 5 % (P˃F= 0.5534); this result differs from that found in a previous study in Mexico14 for the 1990-2016 period, in which they identified evidence of asymmetry in the transmission response of producer prices to the retail market. In contrast to this study, the ECTt-1- values obtained induce a slightly greater change in the retail price than the ECTt-1+ (Table 8).

Table 8 Results of the error correction model: symmetrical and asymmetric vertical 

Independent variable Symmetric model Asymmetric model
Coef. SE t p>|t| Coef. SE t p>|t|
Pprodt 0.0326 0.0593 0.55 0.583 --- --- ---
Pprodt-1 0.6234 0.0489 12.75 0.000 0.6248 0.0489 12.75 0.000
Pprodt-2 -.1419 0.0511 -2.77 0.006 -0.1441 0.0513 -2.81 0.005
Pcont-1 0.0572 0.0596 0.96 0.338 0.0454 0.0642 0.71 0.480
Pcont-2 -0.0222 0.0592 -0.38 0.708 -0.0222 0.0593 -0.38 0.708
ECTt-1 -0.0780 0.0153 -5.08 0.000 --- --- ---
ECTt-1+ --- --- --- -0.0692 0.0211 -3.28 0.001
ECTt-1- --- --- --- -0.0865 0.0213 -4.06 0.000
Constant 0.0002 0.0020 0.13 0.899 0.00007 0.0020 0.04 0.969
Normality test (Prob>z) 0.99330 0.9933
LM test (Prob>Ji2) 1.011 0.996
DW test 2.013912 2.0144
R-squared 0.3440 0.3451
Test:H0: β2+ =β2- --- F (1,372) = 0.35

SE= standard error.

Studies of the vertical transmission of milk prices in other countries such as Slovakia38,39, Hungary40, and Uruguay30 found evidence of asymmetry in the transmission of prices in different links of the chain. One of the factors causing asymmetry, common in these studies, is the market power of the industry; however, the fact that producers are more integrated into the production chain (being part of the industry through cooperatives, for example) makes them react more quickly to changes in prices30.

Conclusions and implications

There is a long-term cointegration relationship between import prices of powdered milk and the producer price at the national level and in the regions of Jalisco, Coahuila, and Veracruz, and between retail and domestic producer prices. Import prices of powdered milk are transmitted symmetrically to the producer at the national level and in the regions of Jalisco and Veracruz, indicating that there are no significant differences in the response of the producer price whether import prices increase or decrease. Nevertheless, evidence of asymmetry in the transmission of international prices to producer prices in the state of Coahuila was found, where an increase is transmitted more quickly than a decrease. The speed of adjustment to deviations in long-run equilibrium behaved differently between regions. No evidence of asymmetry was found in the vertical price transmission between the retail price and the domestic producer price; the adjustment speed shows that the response of retail prices is faster when producer prices decrease than when they increase. Understanding the dynamics of the spatial and vertical transmission of prices can guide public policy designers to design more comprehensive and regionally differentiated dairy support programs, thereby ensuring a better distribution of welfare and income throughout the chain. This study contributes to the literature on the transmission of milk prices in Mexico; likewise, it also identifies possible differences in the spatial transmission of milk import prices at the regional level, highlighting the importance for public policy designers to consider regional differences when formulating strategies to serve the sector to ensure a better distribution of welfare and income along the chain. Finally, given the constraints of the model used, it is suggested, in subsequent studies, to extend the linear VECM model to a threshold VECM incorporating the Momentum-Threshold Autoregressive (M-TAR) model since they allow the identification of profound changes in the price series, in addition to the fact that asymmetries in price adjustments can be obtained in the face of positive or negative deviations.

Acknowledgements

C. Faustino Zaragoza López of the National System of Information and Integration of Markets (SNIIM, for its acronym in Spanish) of the Secretariat of Economy is thanked for providing historical information on consumer prices of milk. We declare that there is no conflict of interest.

REFERENCES

1. FIRA. Fideicomisos Instituidos en relación con la Agricultura (2021). Panorama Agroalimentario. https://www.fira.gob.mx . Consultado 20 Ago, 2022. [ Links ]

2. CANILEC. Cámara Nacional de Industriales de la Leche (2022). Estadísticas del Sector lácteo 2011-2021. https://www.canilec.org.mx/estadisticas.htm . Consultado 05 Ago, 2022. [ Links ]

3. SIAP. Servicio de Información Agroalimentaria y Pesquera. (2021). Panorama Agroalimentario. https://nube.siap.gob.mx/panorama_siap/pag/2021/Panorama-Agroalimentario-2021 . Consultado 10 de Ago, 2022. [ Links ]

4. FAO. Food and Agriculture Organization of the United Nations (2022). Dairy Market Review. Overview of global dairy market and policy developments in 2021. Rome. [ Links ]

5. Parkin M, Loría DE. Microeconomía. Versión para Latinoamérica. 9a ed. México: PEARSON EDUCACIÓN; 2010. [ Links ]

6. Robledo PR. Producción de leche en México y su comercio de lácteos con países del APEC. In: Teoría, impactos externos y políticas públicas para el desarrollo regional. Universidad Nacional Autónoma de México y Asociación Mexicana de Ciencias para el Desarrollo Regional A.C, Coeditores, México: 2018. [ Links ]

7. FAOSTAT . Food and Agriculture Organization of the United Nations (FAO). https://www.fao.org/faostat/es/#home(1997) . Consultado 29 Jul, 2023. [ Links ]

8. SE. Secretaría de Economía . Análisis del sector lácteo en México. Dirección General de Industrias Básicas. 2012. https://www.economia.gob.mx/files/comunidad_negocios/industria_comercio/informacionSectorial/analisis_sector_lacteo.pdf. [ Links ]

9. SAGARPA. Secretaría de Agricultura, Ganadería, Pesca y Alimentación. https://www.gob.mx/agricultura/colima/articulos/crece-la-produccion-de-leche-en-mexico-sagarpa-158944?idiom=es#:~:text=Dentro%20de%20la%20cadena%20de,380%20mil%20-millones%-20de%20pesos . Consultado 05 Ene, 2023. [ Links ]

10. Lloyd T. Forty years of price transmission research in the food industry: Insights, challenges and prospects. J Agr Econ 2017;68(1):3-21. [ Links ]

11. Caballero GMA. La industria de la leche en México: conductas y tendencias. ENT#091;Tesis doctoradoENT#093;. Texcoco, Estado de México: Colegio de Postgraduados, Campus Montecillos; 2010. Disponible en: http://colposdigital.colpos.mx:8080/jspui/handle/10521/83. [ Links ]

12. Meyer J, Crammon-Taubadel SV. Asymmetric Price transmission: A survey. J Agr Econ 2004;55(3):581-611. [ Links ]

13. Bakucs Z, Benedek Z, Fertö I. Spatial Price Transmission and Trade in the European Dairy Sector. AGRIS on-line Papers Econ Inf 2019;11(2):13-20. [ Links ]

14. Jaramillo-Villanueva JL, Palacios-Orozco A. Transmisión de precios vertical y espacial en el mercado mexicano e internacional de leche. Rev Mex Cienc Pecu 2019;10(3):623-642. [ Links ]

15. Martínez-Marcos D, Barrios-Puente G, Pérez-Soto Francisco, Ruiz-Hernández JA, Lara-Reimers EA. Transmisión espacial de precios desde el mercado internacional al mercado mexicano de leche. Ecosist Recur Agropecu 2023;10(1):e3022. [ Links ]

16. Arcos CA. Transmisión de precios internacionales en el mercado lácteo chileno. Chilean J Agric Anim Sci 2020;36(1):86-94. [ Links ]

17. Jaramillo-Villanueva JL, Benítez-García E. Trasmisión de precios en el mercado mexicano e internacional de café (Coffea arabica L.): un análisis de cointegración. Agrociencia 2016;50(7):931-944. [ Links ]

18. Ramsey AF, Goodwin BK, Hahn WF, Holt MT. Impacts of COVID-19 and Price transmission in U.S. Meat Markets. Agr Econ 2021;52(3):441-458. [ Links ]

19. Bakucs Z, Fałkowski J, Fertő I. Does market structure influence price transmission in the Agro-food sector? A Meta-analysis perspective. J Agric Econ 2013;65(1):1-25. [ Links ]

20. SIAP- SADER . Servicio de Información Agroalimentaria y Pesquera. Indicadores económicos. https://www.gob.mx/siap/acciones-y-programas/produccion-pecuaria . Consultado 15 Abr, 2022. [ Links ]

21. SIAP- SADER . Servicio de Información Agroalimentaria y Pesquera (2022). Panorama Agroalimentario. https://www.gob.mx/siap/acciones-y-programas/panorama-agroalimentario-258035 . Consultado 01 de Ago, 2023. [ Links ]

22. USDA- AMS. United States Department of Agriculture. http://www.fas.usda.gov/ustrade/ . Consultado 10 Mar, 2022. [ Links ]

23. SNIIM. Sistema Nacional de Información de Mercados, Secretaría de Economía (SE). http://www.economia-sniim.gob.mx/ . Consultado 18 Feb, 2022. [ Links ]

24. INEGI. Instituto Nacional de Estadística y Geografía. Índice Nacional de Precios al Consumidor y al Productor. http://www.inegi.org.mx/ . Consultado 15 Mar, 2022. [ Links ]

25. Banco de México . Indicadores económicos. http://.banxico.org.mx . Consultado 20 Mar, 2022. [ Links ]

26. Dickey DA, Fuller WA. Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica 1981;49(4):1057-1072. [ Links ]

27. Phillips PCB, Perron P. Testing for unit root in time series regression. Biometrika 1988;75(2):335-346. [ Links ]

28. Engle RF, Granger CWJ. Cointegration and error correction: Representation, estimation and testing. Econometrica 1987;55(2):251-276. [ Links ]

29. Johansen S. Estimation and hypothesis testing of cointegration vectors in Gaussian vector autoregressive models. Econometrica 1991;59(6):1551-1580. [ Links ]

30. Bertamini F, Carriquiry M. Incidencia del mercado internacional y de la estructura del mercado doméstico en la transmisión de precios en la cadena láctea: evidencia desde Uruguay. Udelar. FCEA. Lecon 2021. Serie Documentos de Trabajo; 12/21. [ Links ]

31. Jiménez TR. Orígenes y evolución de la teoría de la cointegración y del análisis de la transmisión de los precios en mercados. Historia de la probabilidad y la estadística (V) 2011:249-262. [ Links ]

32. Labra-Hernández J, Cabas-Monje J, Velasco-Fuenmayor J. Efectos del precio internacional de la leche sobre el precio pagado a productor chileno: un análisis de cointegración. Rev Científica 2017;XXVII(6):385-392. [ Links ]

33. García HLA. Estrategias de las agroindustrias lecheras latinoamericanas. Estudio comparativo ante el proceso de globalización económica. Rev Mex Agron 2001;5(9):263-273. [ Links ]

34. Acosta A, Ihle R, Cramon-Taubadel SV. Combining market structure and econometric methods for Price transmission analysis. Food Sec 2019;11:941-951. [ Links ]

35. Rojas-Juárez LA, Jaramillo-Villanueva JL, Vargas-López S, Bustamante-González Á, Guerrero-Rodríguez JDD. Transmisión vertical y horizontal de precios en el subsector ganado carne de bovino en México. Ecosist Recur Agropecu 2022;9(3):e3307. [ Links ]

36. Bakucs Z, Benedek Z, Fertö I. Spatial price transmission and trade in the European dairy sector. AGRIS on-line Papers in Economics and Informatics 2019;11(2):13-20. [ Links ]

37. Kharin S. Price transmission analysis: The case of milk products in Russia. AGRIS on-line Papers in Econ and Inf 2018;10(1):15-23. [ Links ]

38. Lajdová Z, Bielik P. Vertical price transmission analysis: The case of milk in the Slovak dairy sector. APSTRACT: Applied Studies in Agribusiness and Commerce 2013;7(4-5):89-96. [ Links ]

39. Weldesenbet T. Asymmetric price transmission in the Slovak liquid milk market. Agric Econ Czech 2013;59(11):512-524. [ Links ]

40. Ben-Abdallah M, Fekete-Farkas M, Lakner Z. Analysis of dairy product price transmission in Hungary: A nonlinear ARDL model. Agriculture 2020;10(6):217. [ Links ]

Received: December 07, 2023; Accepted: July 20, 2024

Creative Commons License Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons

 
CENID-Microbiología Animal, Km. 15.5 Carretera México-Toluca, Colonia Palo Alto, D.F., D.F., MX, 05110, 01 (55) 38 71 87 00