Introduction

In Mexico, sorghum is one of the most important feed grains in the livestock sector (^{Molina et al., 2012)}, also considered as an important source of raw materials in the manufacture balanced food for poultry, cattle and pigs (^{Financiera Rural, 2011)}.

With the entry into force of the Free Trade Agreement with North America (NAFTA), there was a change in the national production structure, which was not favorable for the domestic producer, as they lost competitiveness against increasing imports. During 1994 to 2000, production grew 57.83%, while imports increased 48.12%, this meant going from 3.70 to 5.84 million tonnes (t) for the first item and 3.47 to 5.14 million tons for the second (^{FAO, 2015}). These figures show that NAFTA came to reinforce the process of trade liberalization and increase the loss of food self-sufficiency initiated years earlier; coupled with in practice imports were made at the time of the national harvest (^{Rebollar et al., 2005)}. In 2008, domestic production of sorghum was 6.59 million tonnes, that volume was not enough to supply the domestic apparent consumption, which amounted to 8.14 million tons, so it had to import 1.55 million tons, mainly from United States (^{Financiera Rural, 2011)}.

Later in 2011, production recorded 6.42 million tons, it was insufficient to provide apparent domestic consumption, which dated back to 8.8 million t, so it was necessary to import 2.38 million tons (^{FAO, 2015)}. Moreover, from 1994 to 2012, the dynamics of sorghum production in the various regions of Mexico, showed clear differences, by registering agricultural years with reduced harvests and years with growth, which affected differently each producing regions. During 1994, the percentage share of entities such as Tamaulipas was 44.66%, Guanajuato 26.06%, Michoacan 8.41% and Sinaloa 3.19%, while for 2012, the same regions meant, 40.29%, 21.46%, 10.24% and 6.53% of national total.

With the above, it is evident that the behavior of sorghum production between different regions of Mexico, was different over time; hence the interest of an analysis on the regional dynamics of production of sorghum, as a crucial aspect to generate indicators to guide planners of public policy to implement appropriate strategies that encourage the production of that crop cultivation territorial and state level, which could be different for each region of the country.

Therefore, the objective of this study was to determine the growth of sorghum production and its degree of dynamism in eight regions of Mexico (Tamaulipas, Guanajuato, Michoacan, Sinaloa, Jalisco, Nayarit, Morelos and rest of the country), during the period 1994-2012, under the assumption that the various economic circumstances in the domestic and international environment, among other factors, affected in different ways to the growth of this feed grain in different regions of the country.

Materials and methods

To determine the regional dynamics of production of sorghum, Mexico was divided into eight producing regions; for purposes of this study, the main producing states were considered as regions; Tamaulipas (Tams), Guanajuato (Gto), Michoacán (Mich), Sinaloa (Sin), Jalisco (Jal), Nayarit (Nay) and Morelos (Mor); taking into account that these entities, on average, represent 85% of the national total (^{SIAP, 2012)}; he bunched the other states in one region, it was called, the rest of the country (Rp).

Sorghum grain production was analyzed for the period 1994-2012, which was considered the country as a whole and each of the producing regions that comprise it, then that period of time was divided into three sub-periods, 1994-2000, 2001-2006, 2007-2012. To determine the percentage of growth of sorghum production in each of the regions, the annual growth rate for each year was used from 1994-2012 and to determine the dynamics of sorghum production in each of regions, regional analysis technique (aRT), known as differential-structural method (MDE) was used (^{Boisier, 1980)}.

For which a SECRE (industry-region) matrix is a double entry table where rows represent sectors and columns was developed regions (^{Boisier, 1980)}; in this study were the year’s sectors and regions columns (^{Del Moral et al., 2008)}. The variable analysis was the volume of production of sorghum. The information is grouped information processing was performed using TAREA software (^{Lira and Quiroga, 2003)} and Microsoft Excel 2013.

The differential-structural method (known as a method of relative variations or model of change and participation or English as a shift-share analysis or industrial mix and share analysis), is to compare the observed change in a variable over a period, in the region and in the country. This change compared with what would have occurred in the region, if the variable in question, had behaved identically, both in the region and in the country. The method determines changes in the relative position of the regions and changes in the production structure of the territories over time (^{Boisier, 1980)}. This is useful to describe regional disparities, because from the results, you can build regional development policies (^{Del Moral et al., 2008)}.

El structural-differential method is broken into three components: the total effect (ET), differential effect (ED) and structural effect (EE). The ET compares the final value (year t) of the variable under study, in the region j, the value that hypothetically would have had the variable, if the region had behaved like the country, in terms of growth. The value "expected or hypothetical" is obtained by applying the national variation ratio (SNR), the initial value of the variable in the year zero. ET expression is:

Where: Vij= V value of the variable corresponding to year i, j in the region; 0= zero or initial year (given the study period, the zero year, would be from 1994 to 2011); t= final year (given the study period, year t, it would be 1995 consecutively until 2012); rSR= coefficient of variation at national level rSR =

A positive ET, indicates a "hypothetical gain" of activity in the study region, explained by the higher growth in activity in the region, compared with growth of the same activity in the country. A negative ET, indicates a "hypothetical loss" of activity in the region, as the growth in activity is lower in the study region, relative to the country registered (^{Boisier, 1980)}. The ET is explained by the combination of two effects: differential effect and structural effect.

The differential effect, compares the final value (year t) of the variable under study, recorded in year i, j in the region, with the value that hypothetically would have had this variable during the same year; this effect, represents the dynamics of production each year i, j in the region, compared to the dynamics of the same year, nationwide. What it is expressed as:

Where: rSi = coefficient of annual variation, a national level

If a region, you get a positive differential effect indicates that annual production in the region grew above the national production recorded in the same year; therefore, the regions in which positive differential effects were obtained, were identified as dynamic (^{Boisier, 1980)}, competitive (^{Lira and Quiroga, 2003)} and with better production conditions (^{Del Moral et al., 2008)}. Which they obtained a negative ED were stragglers, not competitive and productive conditions worse.

The structural effect derives from the fact that nationally, some years’experience than other productive growth; hence regions with a productive structure specializing in years (sectors) of rapid growth (SRC) at the national level tend to show positive relative changes; while that regions whose productive structure shows a specialization in years (sectors) of slow growth (SLC) will present negative changes relative. EE ref lects the relative weight of the different years (sectors) at the regional level, compared to the relative weight of the same year at the national level (^{Boisier, 1980)}. The expression is as follows:

A positive EE, indicates that the region specializes in SRC at national level, while a negative EE, refers to the region specializes in SLC nationwide (^{Lira and Quiroga, 2003}). With the results of differential-structural method, a regional typology, where the rating component (^{Del Moral et al., 2008}) and a sign of the differential and the total effects, structural (^{Boisier, 1980}) was considered it was developed. The regions were classif ied into six types: three in total positive effect and a degree of dynamism very high, high and medium high (Type I, IIA and IIIA); and three overall negative effect and a very low degree of dynamism, medium low and low (Type IV, IIB and IIIB).

Results and discussion

With the onset of the economic crisis of December 1994 and the entry into force of NAFTA, domestic production of sorghum took a new direction. In 1995, this activity grew 12.66% over the previous year; likewise, Guanajuato, Michoacan, Sinaloa, Jalisco, Nayarit, Morelos and rest of the country recorded positive growth; but the entities that had higher growth rates were Morelos 163.01%, Sinaloa 104.49% and Michoacan 97.85% (Table 1). Meanwhile, Tamaulipas, reduced its production volume. The increase in national and regional production during that year, was explained in part by the establishment of the Program of Direct Support to the Countryside (PROCAMPO), finding a positive response from sorghum production and payment of PROCAMPO, which he indicated that the program favored the supply of feed grain (^{Molina et al., 2012)}.

Table 1 Annual growth rates of production of sorghum, by region, 1994-2012 (%). 

Elaboración con datos de ^{SIAP, 2012}.

The widespread decline in activity during 1997 was related to the effects caused by the climatic phenomenon "El Niño", which caused a decline in production in all regions and required increase 10.33% the volume of imports to supplement the total supply or domestic availability was 8 391.40 thousand tons (^{SAGARPA, 2003)}. The production was normalized in 1998, where growth activity nationally noted, because some regions such as Guanajuato that went from a production of 1 150.14 in 1997 to 1 501.35 thousand tons in 1998 while Michoacan increased from 582.45 to 723.21 thousand tons.

Subsequently in 1999; 11.65% domestic production shrank as a result of the decrease reported by most regions (except Michoacan); this decline is explained in part by the high production registered in the United States, which allowed increase the supply of grain, coupled with increased global production of feed grains, which pushed international prices down in 1998. This, it caused an oversupply and shrinking demand in external markets, preventing price recovery, closing the month of December 1998, with an average price of 91.25 USD/t (^{SAGARPA, 1999)}; situation which led to the actual price of sorghum, 12.93% decrease in Tamaulipas, 13.61% in Guanajuato, in Michoacan 17.77%, in Sinaloa 12.3%, in Jalisco 15.74%, in Nayarit 17.94% and 20.31% in Morelos.

During 2000 and 2001, domestic production increased favorably due to the increase in yield per hectare (h), recorded by all regions (except Tamaulipas), same as average increased 13.23% compared to 1999. The growth the experienced by most regions, production is partly explained by the increase in consumption, low international prices and low exchange rate (^{Rebollar et al., 2005)}. The decrease in domestic production during the following year (2002) was due to a decrease of 10.32% in harvested area and a slight decrease in output per hour, in some regions like Tamaulipas 25.68%, Guanajuato 4.19%, Michoacan 29.22%, Jalisco 13.70% and Morelos 57.39%. In 2003 and 2004, domestic production of sorghum increased by 29.83% and 3.62% compared to the immediately preceding years; and fell 21.12% and 0.10% during 2005 and 2006 showed similar declines most regions, this fall was partly explained by a decrease of 15.73% in apparent domestic consumption in 2005, compared to the previous year; from 10 163.7 to 8 544.9 thousand tons (^{Financiera Rural, 2011)}.

In 2007 and 2008, domestic production grew 12.40% and 6.28%, then 7.35% decrease in the following year and climbed back up 13.62% in 2010. The ups and downs in production was due to movements in harvested area, variations in the actual product price and changes in import volumes. In 2011, domestic production of grain showed decreases of 7.36%; due to a decrease of 34.99% in the volume of production recorded by Tamaulipas, regional percentage representing a decrease of 1 047.02 thousand tons in the previous year, caused by a decrease in harvested area of 57.08 thousand h (^{SIAP, 2012)}, caused by the presence of the meteorological phenomenon known as "hurricane Alex" which brought heavy rains, f looding and high winds, on 30 June and first of July 2010, causing damage to the production of sorghum and other crops in twenty municipalities in the region of Tamaulipas (^{SAGARPA, 2010)}, the main producer of this grain in the fall-winter cycle.

In 2012, domestic production recovered and grew 8.40%, reporting a volume of 6 969.50 thousand t. because some regions such as Tamaulipas grew 44.39%, Jalisco 38.35%, Michoacan 23.56% and Nayarit 20.69%, while Sinaloa decreased -64.57% and Morelos -2.74%. The largest decrease recorded by Sinaloa in 2012, was partly due to the drought that affected the region during may to november 2011 (^{DOF, 2012)}, a situation that led, drop in production of major grains, including sorghum (^{SAGARPA, 2013)}. The growth of domestic production in the last year of the review period was due to an increase in harvested area in the main producing regions, Tamaulipas reported an increase of 82.90 thousands of h, while Guanajuato achieved an increase of 49.99 thousands of h, in relation to 2011 (^{SIAP, 2012)}. The results of differential-structural method, showed that the dynamics of sorghum production in Mexico was different between regions and periods studied. During 1994-2000, the Michoacan and Jalisco regions had a positive overall effect (ET) as a result of a differential effect (ED) and structural effect (EE) also positive, while Sinaloa, Nayarit and Morelos, who also got a ET positive; as a result of increased contribution by the positive differential effect in the first two regions and greater involvement of positive EE, in the case of the third region.

For their part, Tamaulipas and Guanajuato, regions obtained a negative ET, due to a greater contribution of negative EE in the first region and a higher value of negative ED in the second one, sufficient values to influence negatively the ET, in the regions (Table 2).

Table 2 Coefficients obtained from the differential-structural method (thousands of t).  

Elaboración con datos de ^{SIAP, 2012}.

During 2001-2006, the Sinaloa, Nayarit and Morelos, regions were those which had a positive ET, derived from the major contribution that made the ED also tested positive. The other regions (except Tamaulipas), obtained a negative ET, due to the higher relative weight of the values of ED, which were also negative. In the case of Tamaulipas, the ET was negative, because both ED and EE were negative. During 2007-2012, the Tamaulipas, Guanajuato and Nayarit regions reported positive ET, as a result of the greater contribution made positive ED in the first two regions, and greater relative weight of EE, in the case of the third. The other regions, obtained a negative ET, derived from the higher share contributed negative ED.

With the results of differential-structural method, a regional typology according to (positive or negative) sign of each of the effects (ET, ED, EE) was prepared. Also, according to the rating of the component obtained from each coefficient, the degree of spatial dynamism of sorghum production was determined for each of the regions during the different sub-periods (Table 3).

Table 3 Typology of regions according to their spatial dynamism of sorghum production. 

Fuente: elaboración con base en (^{Boisier, 1980)}, (^{Lira y Quiroga, 2003)} y (^{Del Moral et al., 2008)}.

During 1994-2000, in the regions of Michoacan and Jalisco, sorghum production obtained a very high degree of dynamism; however, for the second and third period, the regional activity changed its dynamics and positioned at a medium level low. Meanwhile, in the Morelos region, said high dynamic activity recorded in the first period, decreased in the second and positioned at a medium level and in the third ended with a low average dynamics. In the regions of Sinaloa and Nayarit, production earned a median dynamism for the first period; however, during the next period, the first retained its dynamics, while the second is increased and positioned at a very high level. Finally in the third quarter, both recorded a decrease in its dynamics and stood at an average level low and high.

Meanwhile, in the Guanajuato region, he recorded a low average activity dynamics for the first two periods and increased dynamism to a middle position at the end of the study. In the region of Tamaulipas, sorghum production reported a sluggishness in the first period; however, for the second it increased its dynamics and positioned at a medium level low and refined the study at an average level. From 1994 to 2000, the highest degree of dynamism of sorghum production in the regions of Michoacan and Jalisco, was associated with the magnitude of their hypothetical earnings for the first region was 109.42 and for the second was 67.66 thousand tons. These figures indicate better production conditions, greater interregional competitiveness and specialization of production during the years that production was increasing.

Meanwhile, activity in the region Morelos, experienced a lower degree of dynamism compared to Michoacan and Jalisco, although obtained a hypothetical gain of 20.10 thousand t, production majored in years when domestic production shrank. The average dynamism of sorghum production, reported by Sinaloa and Nayarit, during that six-year period, was because in some years the referred period, annual production in these regions grew above the national production recorded for the same years. Instead the sluggishness of production, recorded by the regions of Guanajuato and Tamaulipas, during that six-year period, was due to regional growth lower than the country as a whole, which resulted in hypothetical losses in production volumes and, consequently loss of interregional competitiveness.

On the other hand, the degree of dynamism of production experienced by the region of Nayarit and Morelos, during 2001-2006, said hypothetical gains, increases in interregional competitiveness and better production conditions. While the degree of sluggishness corresponding to Guanajuato, Michoacán and Jalisco explained by the negative annual growth recorded during this period. This ref lected losses in Jalisco represented hypothetical 170.35, Guanajuato 65.67 and Michoacan 3.25 thousand t; because growth in these regions was lower than the country as a whole.

During 2007-2012, the regional dynamic is explained by the magnitude of the total positive effect meant hypothetical profits that Tamaulipas was 234.55, Guanajuato 64.71 and Nayarit 7.13 thousand t. This behavior was because these regions had a greater than for the country as a whole, which indicated greater intra-regional competitiveness, greater specialization of the productive activity and better conditions for growth such territories. For its part, the lag experienced by the Michoacan, Sinaloa, Jalisco and Morelos regions during that six-year period, was due to the lower growth recorded for the same compared to that reported by the country; therefore, these regions obtained hypothetical losses of various magnitudes and a decrease in intra-regional competitiveness, coupled with growth and specialization in the years when domestic production shrank sorghum.

Conclusions

During the study period sorghum production in Mexico, he showed clear differences concerning growth and dynamism among regions. Nayarit was the region that received the highest positive growth rates than the rest of the regions analyzed, reflected in a greater degree of dynamism during the three sub-periods, and meant higher hypothetical gains, better production conditions, greater interregional competitiveness and specialization of activity during the years when domestic production was increasing.

Meanwhile, Michoacan and Jalisco, showed greater decreases in production volumes sorghum, growth was lower than the rest of the regions and the country as a whole, causing losses hypothetical various magnitudes and decreased competitiveness intraregional. This is due to the lack of diversification of activity throughout the period and its growth and specialization in years sorghum production shrank, which positioned itself to such areas as stragglers.