Introduction

Migration is a constant, continuous and dynamic phenomenon that has been present throughout history. Human migrations have been important factors in the consolidation of ethnic groups, serving as an expansion instrument for commerce, conquests and dominations (^{Alba, 2010}). At the end of the 19th century, migration was associated with lineal processes that characterized poor countries. People who migrate from one place to another justify their reasons in many different factors and causes that are the consequences of existing situations, both in the zones of origin and in the areas of attraction (^{Chimanikire, 2005}). Social phenomena such as poverty, unemployment and opportunity inequality act together as factors that prompt people to search other places for better life conditions.

The migratory process comprises internal displacements and, in particular, international movements. The distances and frontiers are shortened and reduced due to economic, political and global phenomena; these displacements can have negative or positive effects on the people that migrate to different parts of the world.

International migration has been the focus of many researchers that address origin theories, displacements, causes and their effects (Wallersteiteirn, 1985; ^{Wallerstein, 1987}; Durand, 2002; Taylor, 2002; ^{Arango, 2003}; Massey, 2003), as well as the economic impact of remittances (Moctezuma, 2010). Since migration plays an important role in the activities of organizations related to development, it has also been the focus of those entities that are involved in the decision making for the good of society.

International migration is also presented from a capitalist economic model perspective. In the transfer of people from a developing country to a developed nation, the increase in earnings prevails over the search for opportunities and employment for migrants in the cheap labor trade; so these circumstances are accepted in order to improve financial situations (Jacobo, 2010; ^{Yúnez, 2010}). Thus, international migration is justified by factors of attraction, migrant networks that import labor, migration policies, as well as the effects of remittances. Therefore, the increase in migration is a reality of the structural changes of the state, which entails social, political and cultural change.

Worldwide, migratory processes comprise an increasing number of migrants in the past few decades. It is estimated that from the world’s population there are around 175 million migrants, 14 million refugees, more than a million people requesting refuge and around 25 million internally displaced people (World Migration Report, 2005).

According to the Consejo Nacional de Población (^{CONAPO, 2011}), this increase is reflected by movements to different developed countries, especially European countries, with the exception of the United States which is the country that attracts the most migrants. In 2010, the number of migrants that the United States received increased to 13% of its total population (Table 1).

Migration is a phenomenon suffered by many countries. In Mexico it is particularly evident in the most vulnerable zones, where inhabitants look for better life conditions. In 2010 according to a survey carried by the United States Census (2011), it was estimated that 31.8 millions of Hispanics of Mexican origin lived in the United States; as such, Mexico is positioned as the country with the largest number of migrants. In 2012, migrants represented almost 2/3 of the Hispanic population (64%) and represented 11% of the total population of the United States (^{Pew Research Center, 2013}).

In this regard, the context that justifies these migratory flows can be seen since the end of the 19th century until present day. Four moments that characterize the migration phenomenon can be identified in Latin America: the first is linked to transoceanic migrations; the second to internal migrations between 1930 and 1940; the third to cross-border migration, consequence of economic, political and social conflicts; and the fourth as a product of globalization (^{Aruj, 2008}). The last two stages characterize the situation that forces the flow of migrants from Mexico to the United States.

The United States ranks in first place as host of migrants and its history is marked by constant flows in regards to the migratory movement of Mexicans. This case is quite particular as there has been no other migration flow that has lasted more than one hundred years and with such a large quantity as the one coming from Mexico. This phenomenon can be identified by three factors: history, massiveness and vicinity (^{Durand, 2000}).

The origin of Mexican migration to the United States can be defined in several stages. ^{Durand (2003)} classifies them in five periods during the 20th century, with a duration of approximately 22 years. The “attachment” stage (1900-1922); “deportation” stage (1921-1939), “bracero” stage (1942- 1964); “undocumented” stage (1965-1986); and the “Los Rodinos” stage, the last phase initiating in 1987. This author states that these processes are due to a pendulum movement marked by North American migration policy in accordance to political and social context.

Within this context, the objective of this study was to know the impact that migration has on the life quality and well-being of migrants in the municipalities of origin. It was asked if the social and economic problems of the people force them to migrate and cross the northern border in search of better life conditions.

Methodology

In order to verify the hypothesis, a cluster analysis was done using four variables: a) migration intensity index (MII), the index published by the Consejo Nacional de Población (^{CONAPO, 2010}) was used as summarized in Table 2; b) marginalization index, which lists the shortcomings of the population (^{CONAPO, 2010}) as summarized in Table 3; c) human development index, with health, education and income variables which measure the combination of abilities and freedom of the people to choose between alternative living standards, as calculated by the United Nations Development Programme (^{UNDP, 2010}) and summarized in Table 4; and, d) rural level according to the Organización para la Cooperación y Desarrollo Económico (OCDE, 2005) as summarized in Table 5.

The information of each variable was considered for 2 411 municipalities of the Mexican Republic. The study analysis unit was the municipality in order to understand migration on a territory level.

With the aforementioned variables, the multivariate cluster technique was used as it allows to perform several measurements on the municipalities in order to characterize the territorial development, and with this to also analyze the relation that the municipalities have with migration.

The cluster analysis technique was done in order to search certain characteristics and behavior patterns regarding migration. 2 411 municipalities were grouped in clusters or similar groups, for each municipality, the two variables were included. According to ^{Rencher (2002)}, the matrix of information is represented by:

Where: 'i y is a row (observation vector of each municipality) with i= 1, 2… 2411; and ( ) ' j y is a column of variables with j= migration intensity, marginalization, human development with education, health, income variables and rural level.

For the realization of this work, Ward’s Hierarchical Method was used. This method is an algorithm that allows finding good groups within the data. In this case, the municipalities are the object of analysis and were analyzed using the software package Statistical Analysis System (SAS).

Ward’s technique is a cluster type; in each step, an observation or cluster of observations was fused with another cluster. It starts with n groups (individual municipalities) and is finalized with a simple cluster that contains the set of complete data or groups of municipalities.

With the agglomerative process, the close groups are fused together, using a similarity or dissimilarity measure between the groups. For this, the incremental sum of squares method (Ward’s technique) was used, which uses the quadratic distances of intra-conglomerates and between groups.

Results and discussions

Migration distribution: main results

Based on the MII, MI, and HDI variables of the 2 411 municipalities in Mexico, 15 groups (Table 6) were formed, obtained through the cluster analysis. It can be observed that 11 groups (1,2,3,4,5,6,7,8,9,10,11) comprise 77% of the municipalities and they are characterized for having medium, high and very high marginalization (Figure 1). More than half of the municipalities of the country have different types of shortcomings. In regards to the MII, 7 groups (1, 2, 3, 4, 5, 8, 9, 11) comprise the municipalities with a medium and high migration index, which represents 54% of the total municipalities with movement and displacement outside their zones of origin. Therefore, if Mexico is the main country that sends migrants to the United States, the largest quantity corresponds to the municipalities that comprise these groups (Figure 2).

Table 6. Groups obtained from the cluster analysis. 

Grupos	Municipios	IIM2010	IM	IDH	Ruralidad
1	251	0.03	-0.49	0.67	2
2	113	0.29	0.42	0.61	1
3	281	-0.25	0.21	0.64	2
4	157	1.91	-0.33	0.65	2
5	132	-0.06	1.41	0.52	1
6	180	-0.56	-1.47	0.77	3
7	327	-0.31	-0.9	0.71	2
8	428	-0.04	0.43	0.61	2
9	177	0.09	-0.42	0.67	3
10	179	-0.44	1.48	0.55	1
11	44	2.11	0.79	0.58	1
12	42	-0.26	1.44	0.49	1
13	37	-0.73	-1.68	0.79	4
14	51	-0.59	0.29	0.65	3
15	12	-0.69	-1.29	0.74	4
Total	2411

Figure 1. Municipalities with medium, high and very high poverty (77%). 

Figure 2. Municipalities with IIM medium and high (54%). 

According to the OCDE, 60% of Mexico’s municipalities are considered semi-urban rural level, which are concentrated in five groups. The municipalities with a disperse rural index are clustered in five groups as well, and represent 21% of the total number of municipalities. In contrast, the municipalities that are characterized for having an index considered urban metropolitan represent 2% of the total number of municipalities. The rest correspond to the index considered urban intermediate.

There is a relation of the groups in regards to their indexes; when the groups have a very low MI, their MIII is also low and their human development is therefore high. Another characteristic of the groups is that they have a rural urban index, which could indicate that when the municipalities have the necessary infrastructure and allow the access of goods and services, they prevent a large percentage of people from leaving their zones of origin.

In this sense, group 8, which comprises 17.7% of the municipalities, has a MI mean of (-.04) m and the majority of the group belongs to the state of Puebla, with a lower presence in the states of Oaxaca, Yucatán and Michoacán. Group 7 agglomerates a total of 327 municipalities and this group has a MII value of (-.31); they have a low migration level as well as an MI value of (-0.90). The municipalities that comprise this group are mainly in the states of Chihuahua, Jalisco, State of Mexico, Sonora, Veracruz and Yucatán.

On the other hand, groups 6, 13 and 15 are comprised of 229 municipalities; they are characterized by the fact that they send out less migrants and have a low MI and a very high development. In these three groups, the 16 delegations of Mexico City are found as well as municipalities from the states of Sonora and Nuevo León who have similar characteristics.

According to ^{CONAPO (2010)}, Mexico has been the main country that receives remittances in Latin America. This is primarily due to their characteristics and proximity. In contrast with this information, it can be observed that the two groups that send more migrants (groups 4 and 11 as seen in Figure 3) are not necessarily the ones that receive the most remittances (Table 7).

Figure 3. Groups with high IIM. 

Table 7. Migration and remittances. 

Grupos	Municipios	IIM 2010	Viviendas con remesas
6	180	-0.56	7 120 171
13	37	-0.73	6 825 746
15	12	-0.69	3 473 243
9	177	0.09	2 622 496
7	327	-0.31	1 674 634
14	51	-0.59	1 469 774
8	428	-0.04	1 381 477
3	281	-0.25	1 296 515
1	251	0.03	1 212 721
4	157	1.91	615 189
10	179	-0.44	590 521
5	132	-0.06	103 000
12	42	-0.26	100 264
11	44	2.11	77 311
2	113	0.29	63 547
Total	2411		28 626 609

Therefore, within these two groups, it can be observed that the municipalities in the states of Oaxaca and Zacatecas are the ones that send more migrants; however, in regards to remittances, they are positioned in the sixth and tenth places of states that receive the most money (CONAPO, 2013). These groups are also characterized by a medium and high MI; however, they still have shortcomings. In this regard, ^{Arroyo (2003)} states that smaller municipalities with high migration lack the necessary infrastructure, and the money sent from abroad does not directly help the receiving community as they have to pay for products and services from other cities and regions.

Furthermore, groups 6 and 13 (Figure 4) comprise 9% of the total number of municipalities and are characterized for having the lowest MII of all the groups, receiving the most remittances from the total number of groups analyzed. The municipalities of Jalisco, State of Mexico and Mexico City concentrate a great number of homes that receive remittances.

Figure 4. Groups that receive the most remittances. 

In the state of Jalisco, Guadalajara is the municipality with more homes that receive remittances on a national level and it has a medium MII. Nationally, it ranks in third place in regards to the entities that receive remittances (CONAPO, 2013). On the other hand, Mexico City has a very low MII and according to the ^{Bank of Mexico (2014)}, it generated $1 393 1524 millions of dollars in familiar remittances. For the second trimester of 2014, it has accumulated $437 8609 millions of dollars, surpassing Oaxaca who has the most municipalities that send migrants.

Meanwhile, the State of Mexico has a medium MII and it can be observed that there are six municipalities in group 13 with a considerable number of homes that receive remittances; Naucalpan, Tlalnepantla, Atizapán de Zaragoza, Tultitlan and Nicolás Romero are the municipalities considered urban which receive the most remittances in the entire state. This further confirms that several small municipalities with very high migration do not directly help the receiving community as they do not have the necessary infrastructure.

Even if these municipalities have a low MII, the context and the conditions in which they live could prompt their inhabitants to migrate in order to fight these conditions. As indicated by ^{Aruj (2008)}, the decisions of the people to migrate can be understood by looking at factors such as the lack of work alternatives and insecurity in their financial future, as well as the fact that they cannot meet basic needs.

Conclusions

The migration effect is related to the marginalization suffered by thousands of people that by not having the capacities and opportunities tend to move to other zones where their necessities are met. Thus, the migration phenomenon has an important role in decision making, both for the individuals in search of a better life as for the entities that procure better living conditions.

Mexico being the country that sends the most migrants to the United States throughout a history of more than 100 years should regard migration not as a social phenomenon, but as one with several consequences. It is therefore important that the entities of the state take notice of the migration phenomenon and try to reduce it, making decisions based on the factors that prevent the exclusion of people from goods and services and generating the means to strengthen their acquisition capabilities.

It is important that the state’s institutions prioritize those municipalities that form the groups with a high MI and a medium or low MII. Even if there is no tendency to migrate, they have the factors that prompt the movement of people and that promote the idea of migration as a solution to the marginalization conditions under which they live. The institutions should take care of the 353 municipalities that comprise these groups and that belong primarily to the states of Oaxaca, Chiapas and Veracruz.