Migration, Trade and FDI in Mexico. Patricio Aroca Universidad Catolica del Norte, Antofagasta, Chile

Transcription

1 Unbelievably preliminary. Cite this and you re cactus. Migration, Trade and FDI in Mexico Patricio Aroca Universidad Catolica del Norte, Antofagasta, Chile W.F. Maloney Office of the Chief Economist for Latin America The World Bank April 15, 2002

2 Mexico wants to export goods, not people. Carlos Salinas de Gortari I. Introduction Mexican President Carlos Salinas de Gortari promoted NAFTA partly on the grounds that it would reduce the incentives for Mexicans to migrate north. Yet what limited evidence there is does not strongly support this claim. As Razin and Sadka (1997) note, theoretically trade and to a lesser degree capital flows, and migration can be complements. Markusen and Zahniser (1999), drawing on models by Feenstra and Hanson (1995) and Markusen and Venables (1995) study the effects of NAFTA on the convergence in the wages of unskilled workers between the two countries and they echo the widely held view that NAFTA will do little to achieve wage convergence and hence deter migration. This paper attempts to measure the direct impact of foreign direct investment and trade on migration. Ideally, we would answer the question using actual data on Mexican- US migration, but the illicit nature of these flows and generally poor quality of the data makes such a direct approach difficult. The paper instead asks whether these variables have had any impact on flows within Mexico where the census data permit careful analysis. We find that they do and that both FDI and exports are substitutes for labor flows- they deter migration. We then generate some tentative inferences about the impact on Mexico-US migration and find it to be of important magnitude. In the process, we also generate the first estimates of determinants of migration flows within Mexico. In line with recent advances in the industrialized country literature, we generate proxies for both the level of amenities and costs of living and find their influence statistically significant. Contrary to much of the literature, we also find very significant and very intuitively plausible signs on labor market variables. The signs on all these variables confirm the often postulated liquidity effect- it takes resources to move. 1

3 II. Methodology We assume that the potential migrant faces j possible destinations where i is the region of origin and k is the migration region chosen. The worker internal migration decision is reflected by the sign of the index function I * = V k V C i where V can be interpreted as a indirect utility function in the context of random utility theory (Domicech and McFadden, 1975 and Train, 1986), and C is a measure of costs. We assume that utility is a function of a linear combination of location characteristics X V j = X j β + ε j If the destination region is more desirable, measured along several dimensions, and if the migrant has sufficient resources to move, then we should observe migration. The probability that the indicator will be larger than zero is equal to the probability that the difference between V s is greater than transport costs: P( I * > 0) = P( V V C > 0) = P( ε ε X β X β C) k i i k k i This specification nests many standard estimated functions (See Greenwood) including Borjas (2001) where the only argument in the utility function is the wage. 1 The actual specification depends on the assumptions about the error term. The βs may be allowed to vary and in fact the literature tends to find a greater role for destination variable than for origin variables. This may be because of asymmetric information about locales (Gabriel et.al 1993), or that the individual variables are correlated with omitted variables that may have a greater impact on one end of the migration move. As an example, many variables could be correlated with unmeasured 1 * Borjas argues that I = max{ w } w C where I * is an indicator variable, w is the wage y C the j costs of transportation to the new locale. This function must satisfy all possible destinations, i the region of origin and k the region chosen. j i w = max{ w k j j }, where j represents 2

4 wealth or liquidity that would determine whether the worker has the savings to pay the fixed cost C of moving (See Vanderkamp 1972). The matrix X contains the variables capturing the relative expected incomes (Y) in the two areas(wages, unemployment, price indices), and the set of characteristics of the region (amenities) that may also affect the migration decision. It is through Y that we might expect the impact of FDI and trade. As the survey by Razin and Sadka (1997) argues, migration and trade, and to a lesser degree FDI may be substitutes or complements and hence there is no guarantee that NAFTA would necessarily, on purely theoretical grounds, lead to lower migration. Since we work with aggregate data, we follow Berkson s (1944) setup as elaborated by Ben-Akiva and Lerman (1985) and generalized in Gourieroux (2000). Here F 1 ( P( I * > 0)) = X β X β C k d i o where F is the probability function that is determined by the structure of the errors. III. Data Migration data: The 2000 census tabulates the question In which state did you reside five years ago? ( En qué estado vivía usted hace 5 años?) and then this response is compared with the present state. As with most data of this type, this has the drawback of obscuring migrants who may have left and returned in the five year period. Graph 1 shows the rates of Net Migration by states. The variable is calculated as net flows as a fraction of the population in the initial period. Moving Costs: Following the literature (again, see Greenwood), we approximate the costs of transportation as a quadratic function of distance. 2 This is a proxy for the costs of migration that consist of the moving costs themselves, the opportunity costs of moving 2 Though we assume that the indirect utility function is linear and the weight of each variable is similar in each region, this assumption can be easy relaxed to differentiate the origin (o) and destination (d) parameters. 3

5 which rise with the length of the journey, and rising communication costs with the family in the point of origin, including the increased costs of return visits. In general, the literature expects a negative impact on migration but with decreasing effect. Population: is the population by state in 1995 as reported by the Census. As Greenwood (1997) summarizes, population is often used as a measure of the availability of public goods. However, it is also true that larger states offer more connection points than small states and will hence, in a random reallocation, attract more migrants. Shultz (1982) is also correct in arguing that larger states may have smaller rates of outmigration simply by virtue of having more places to migrate to within. The 1995 value is used to eliminate any problems of simultaneity with migration flows that happen in subsequent years. Labor Market Variables: Expected Earnings. These correspond to the average of the rate of unemployment and wages in 1995, 1996 and 1997, the three years that are most likely to influence the decision to migrate. Cost of Living: To some degree, the potential migrant should be deflating expected wages. However, the effect may not be so straight forward. For instance, if a migrant plans to return to a low cost area to retire, he may generate real savings measured in his retirement destination faster by earning a lower real wage in a high cost area (See R.E.B. Lucas for a survey of this literature 1997). Further, high cost of living may point to a larger potential income over the long run even if not experienced (Spencer 1989, Pagano 1990). Certainly, in an intertemporal or intergenerational context, taking a lower real wage in the US is still likely to offer the migrants descendents far better options than would have been the case in Mexico. The literature provides mixed evidence. Cameron and Muellbauer (1997) studying UK migration find strong deflator effects that they argue dominate any expectation effects. On the other hand, Thomas (1993), also looking at the UK, finds no impact of regional house price difference on destination choice of any group except retirees. 4

6 Two indices were constructed and the methodologies are detailed in annex I. The first is a hedonically estimated housing price that is the analogue to those used elsewhere in the industrial country literature. Since food is likely comprise a larger share of the consumption basket in LDCs, we also generate the cost of a basic food basket by state. These were included both separately and as an average measure of the cost of living. Amenities: Price indices, however, may also simply reflect amenities available in the new area, implying a positive relation with migration decisions. Further, as Roback (1982) showed, they affect equilibrium wages as well and hence should be included as part of the net utility change of moving from one region to another. To attempt to control for this, we extract the principal component of a set of variables that include health, education, and infrastructure services (see Annex II). Trade and Investment Variables Foreign Direct Investment: This is aggregate per capita foreign direct investment from reported to the government by the firms investing and was provided by the Central Bank of Mexico. The Federal District (DF- Mexico City) shows vastly higher rates because much of the FDI destined for other states is registered at the headquarters of the firm in the DF. We include a dummy variable in the regressions to account for this measurement error. Graph 2 shows the incidence of FDI per capita by state. As is not surprising, FDI is highly, although not exclusively concentrated along the northern border with the US. Maquila Value Added: This variable may be seen purely as a proxy for FDI, but since the maquilas are primarily exporters, it can also be seen as a proxy for maquila exports per capita which brings it closer to being a trade variable. Since it is collected from industrial surveys, it does not suffer from the headquarters effect of the FDI variable. The two variables are moderately correlated (.64) and hence are it may be a better for FDI more generally. On the other hand, the government of Mexico tabulates 12 of the 5

7 states in an other other category and hence we lose substantial information. We run the regressions with the substantially reduced sample. Exports: This variable is provided by the Ministry of Finance (Hacienda) and is alleged to capture exports per state. However, it may be that this represents interpolations by the government based on employment in industry as measured by the IMSS (Pending verification). Imports: Provided by Bancomext. This variable could have multiple and conflicting effects. It could simply reflect the degree of integration of a state with external economies and hence proxy as well for exports. On the other hand, if it is seen as representing competition for import substituting firms, the short run labor market impact could be negative and hence it could conceivably lead to more migration. None of the variables are ideal, but they are complementary in the sense of being strong whether the other is weak. Together, we may get some reliable picture of the impact of trade/investment variables. IV. Results In preliminary regressions, we estimated a multinomial logit model for aggregate data. Though the results were consistent with the theory, the Fry and Harris test (1998) suggest that the data violated the Independence of Irrelevant Alternative (IIA). Therefore, we estimate a multinomial probit and follow Gourieroux s (2000) weighted least square procedure. 3 3 Given that the restriction on the pseudo indirect utility function that imposes the condition that all individuals in every state have an identical distribution of conditional probabilities may be strong, we follow the route suggested by Pudney (1989) that includes an alternative-specific additive constant in an otherwise invariant utility function. This constant can be interpreted as fixed effects associated with the average individual in each state and therefore measures the unobservable individual characteristics that individuals in each state use in making their decisions. As in Davies et al. (2001) estimating model problematic, most likely due to the large number of parameters and the likely correlation between origin characteristics and state fixed effects. 3 Where Davies et al tackled the problem by respecifying arbitrarily the variable associated between state, joining the states that 6

8 Column1 and la present the standard regression including the labor market variables. Overall, the specifications are very satisfactory with the coefficients on the core variables are all statistically different from zero and of predicted sign. [The distance measure is at the low end, but firmly within the usual range of -.02 to -.2 (Greenwood, see also Gabriel et al 1996, Fields 1982, and Schultz, 1982) and the population variables are also consistent with usual results. Preliminary regressions found both the origin unemployment and wage level to be insignificant as has been found frequently in the literature (Greenwood, Lucas 1997). However, we then attempted to isolate two countervailing effects, one a substitution effect among regions, and the other a wealth or liquidity effect that allows the worker to cover the fixed cost of moving. The latter effect has been found for unemployment by Goss and Schoening (1984) and Herzog et al. (1993) for the US who find that the probability of moving decreases with the duration of unemployment, and the literature on the wage effect is also extensive (See Stark and Taylor 1991 for a discussion of credit constraints). We generate relative wage (ln wj-ln wi) and relative unemployment (Uj/Ui) variables, and then allowing free standing initial wage and unemployment variables to capture credit constraint effects. In the complete sample, all but the free standing unemployment term enter significantly although it is of predicted sign and significant at the 10% level in the restricted regression. This overall strong performance suggests that the poor results of origin labor market variables in many previous studies arises precisely because they capture two contradictory tendencies. The cost of living variable enters very significantly and of important magnitude in the sector of origin. The strong positive coefficient on the destination is unexpected, share similar unobservable characteristics, we follow a different approach. We proceed to estimate stepwise based on the initial parameters of the model, and choosing the dummy variables according to their contribution to the model. 7

9 although perhaps consistent with the view of cost of living being a measure of expectations for future income growth as discussed earlier. The amenities variable also enters with predicted signs, but again in the relative/free standing format that suggests that amenities may be correlated with an omitted credit constraint variable. It is not significant in the restricted sample although the signs are as predicted. Integration Variables Columns 2a,3a,4a,5a drop the labor force variables and replaces them with the the trade and investment variables in Z. In all cases, we find evidence of a strong substitution effect: FDI and trade reduce migration. In most cases, simply entering the initial and final terms yielded symmetrical results. However, a free standing term was significant or borderline in half the cases suggesting again, a liquidity/wealth effect so the results are again reported in the constrained/free standing form. All other coefficients remain relatively unchanged with the exception of the destination price term which appears to almost double. Columns 2b, 3b, 4b, and 5b add the labor force variables and confirms that some, but not all of the effect of Z works through the labor market at the same time. The substitution effect diminishes by at least a factor of 2 in more cases and in the case of FDI becomes insignificant. The free standing terms also show propensity to flip sign suggesting that initial FDI was capturing the initial wealth/liquidity now capture by initial wages and unemployment and that, minus this, local Z has a more powerful deterent effect than destination an attractive effect. The fact that all Z retain an effect outside of the contemporaneous labor market variables may reflect that there is an independent disincentive effect to migrate, perhaps through expectations of future growth. 8

10 In short, Salinas was correct to suppose that NAFTA, through various channels, would lead to reduced migration, at least within Mexico. The next section attempts to quantify how large these effects might be in the context of US/Mexican migration. V. Simulations With the estimate above, we attempt to make some inferences about the impact of foreign investment or Trade on migration to the US. The key assumption is that we can treat the US as a 33 rd Mexican State. It is probably not too far out of sample to make predictions. As table 3 shows, ratio of the per capita income of Mexico s richest state, the D.F. relative to its poorest is about 6.4 in nominal or 5.6 in real terms while the ratio of the average for the US relative to the D.F. is only about 1.9 or 2.3 PPP adjusted. That is, in terms of development the US is closer to Mexico City than Mexico City is to Chiapas. Nor are wage differentials radically different. The ratio of the Hispanic real wage in the US to the mean wage of the DF adjusted for PPP is roughly the same as the ratio of the real DF wage to that in Chiapas. There may be some concern about the importance of the border representing fundamentally different transport costs. In fact, the evidence is strong that this is more a difference in magnitude than kind. Donato, Durand and Massey (1992) argue that our data from Mexico reveal a fairly high probability of apprehension by INS combined with a near-certain probability of ultimately entering the United States. (p 152) and that every migrant who attempted a border crossing, whether apprehended or not, eventually gained entry p 155 italics theirs. This suggests that border control serves more as a tariff than a quota. The costs of movement are substantially different, but perhaps not so much as we might think at first look. The present (2002) cost of a second class bus from Quintana Roo to Coahuila, one of longer trips in our sample, was roughly $US100 compared to 9

11 very little between Mexico State and DF. 4 Anecdotal evidence about the cost of direct transport across the border in the 1980s was $150 (244 $US 2002 CHECK; Conover 1987 cited in Hanson and Spilimbergo 1999), again, not so far out of sample. 5 That is the Mex/US cost is roughly 2.5 time the ratio Max/Min within Mexico- not too far out of sample. The cost of a Coyote or smuggler/guide appears to have held steady in real terms since the 1960s at around $350 ( $2075 $US2002; Donato, Durand and Massey 1992) 6 although Crane et al (1990 cited in Hanson and Spilimbergo 1999) suggest that in 1993 only 8.3 percent of those apprehended by the INS had employed one. Clearly, the premium for risk involved in crossing the border and for the multiple trips will widen the gap substantially. Even if these gaps are wide, eq (1) suggests that the coefficients on the components of the indirect utility functions should be separable from those of the cost of travel. That is, the travel cost elasticity may not be suitably estimated for forecasting the impact of a change in Mex/US travel costs, but the response to wages, FDI or exports should not be affected. Under these assumptions, we can think of the estimates above in two ways. First, as corresponding to those of the aggregate average Mexican state vis a vis any other state including the US. The implicit elasticities therefore capture the reduction in push to the US and the increased attractiveness from the US of and increase in trade or investment. Alternatively, and with the potential for greater richness, we assume that total emigration from any state is the sum of migration to other states plus the US: 32 M i = mij + m ius j= 1 for all i = 1,..., M = Pop * p + m ius i i ij j= 1 4 White Star Bus line 5 Inflated from 1985 figure with CPI. 6 This showed little change with the Immigration Reform and Control Act of 1986 suggesting that it is a fairly robust number. Figure was reached by inflating the 1960 figure by the CPI. Coyotes only get paid for successful crossings. 10

12 and that further total emigration is fixed: M 32 = M. This implies that we are only concerned about the redistribution of emigrants between the Mexican states and the US in response to a change in the X variables. This is consistent with models which postulate a two step decision on the part of migrants- first to migrate, and then which destination to choose. It is also only problematic to the degree that we believe that the impact on total flows are not second order and/or that they are radically different in distribution among states from the substitution effects. We can write i= 1 i M = m + Pop p = InM + Pop p ius i ij US i ij i= 1 j= 1 i= 1 j= 1 where InM US is the total migration from México to the US. From above, we know that the probability of migrating from state i to state j depends on the characteristics in X, p =Φ ( X β ) and hence that ij ˆ 1 Φ ( pij ) = zij = Xb. The percentage change in immigration to the US we can obtain as: ln 1 ln InM US Φ( Xb) = Popi X k InMUS i= 1 k= 1 ln X k para k = {i,j} Tables 4a and 4b present simulations of the impact of a 10% and 100 million dollar increase in a given state on migration. The former is perhaps more interesting as an elasticity while the other can answer if we were to locate a plant in Mexico, what state would show the greatest fall in migration. The first column presents reduced migration, the second increased return migration and column 4 presents the total effect. Column 3 shows whether the effect occurred largely to retention of workers, or to attraction. 11

13 States are ranked by order of effect. It is not surprising that DF would have the largest impact in percentage terms since it has by far the largest FDI and Durango the least given it has virtually none. However, locating a plant in Baja California Sur, will have the largest effects with the DF ranked only in 9. The top five seem to gain their power primarily through raising return migration from the US. Those with the greatest potential to retain future migrants are generally further south the DF, Sn Luis Potosi and Yucatan heading the list. Somewhat depressingly, the poorest states of the South, Chiapas, Gerrero, Oaxaca where we might like for reasons of poverty reduction to encourage investment are not the ones that will most reduce migration. In fact, in the top 10, only Tlaxcala, Morelos and Hidalgo have below average per capita incomes. Since the simulations depend on the magnitude of migration to the US which is not well known, Figures XX present the range of values of percent reduction for a range of values. The calculated elasticity of reduction in response to a 10% rise in investment averages at the most likely level roughly 1.5% and ranges from 6% for the DF to around.5 for Campeche. The impact of a $US 100 million project is highest in Baja California Sur, reducing net migration by 3%, and lowest in Chiapas, with again, roughly.5%. The average, at around 1.25% is implicitly above that of the previous exercise since the mean level of FDI is roughly double that used in the simulation. Hence, we might expect around a 2.5% reduction in the case of a 10% increase in this simulation. VI. Conclusions The paper first presents nice specification where, contrary to the literature, labor market variables enter as predicted. We argue that part of the traditional poor performance of the origin state variables is that they capture a mix of deterrent effects, and credit constraints. Higher wages reduce the incentive to leave, but make it more possible at the same time. Structuring the specification to capture these too effects yields well behaved coefficients. 12

14 Second, it created variables for costs of living and amenities that have generally the predicted effects. Third, it established that FDI, maquila value added, exports, and imports all have the effect of reducing outmigration/encouraging inmigration. This suggests that Salinas, in principal, was correct that increased trade and investment flows would reduce labor flows to the US. Fourth, it simulated the magnitude of the response to an increase in FDI flows (will later do exports etc.). We find the impacts to be non-negligible, although not immense (must check). Further, the impact varies greatly by the state where the investment is located. Unfortunately, it is not always the case that investment in the poorest states will yield the greatest fall in migration to the US. 13

15 REFERENCES Ben-Akiva, M. and S.R. Lerman (1985). Discrete Choice Analysis. Theory and Application to Travel Demand. MIT Press, Cambridge, MA, USA Beyer, H., P. Rojas and R. Vergara (1999). Trade Liberalization and Wage Inequality. Journal of Development Economics, 59: Borjas, George (2001). Does Immigration Grease the Wheels of the Labor Market?. Brooking Panel on Economics Activity, March Available on Borjas, George (1994 The Economics of Immigration, Journal of Economic Literature, 32: Cameron, Gavin, and John Muellbauer (1998) The Housing Market and Regional Commuting and Migration Choices. Centere for Economic Policy Research, Discussion Paper no Conover, Ted (1987) Coyotes: A Journey through the Secret World of America s Illegal Aliens. New York: Vintage. Crane, Keith, W., Beth J. Asch, Joanna Z Heilbrunn, and Danielle C. Cullinane, (1990) The Effect of Employer Sanctions on the Flow of Undocumented Immigrants to the United States. Urban Institute Report No. 90-8, Washington, DC, Davies, P.S., M.J. Greenwood and H. Li (2001). A Conditional Logit Approach to U.S. State-To-State Migration. Journal of Regional Science, Vol. 41, No. 2, pp Domecich, T.A., and D. McFadden (1975). Urban Travel Demand. A Behavioral Analysis. North Holland Publishing Co., NY, USA. Donato, Katharine M. Jorge Druand and Douglas S. Massey (1992), Stemming the Tide? Assessing the Deterrent Effects of the Immigration Reform and Control Act Demography, 29: Faini, R., J-M. Grether and J. De Melo (1999). Globalisation, and migratory pressures from developing countries: a simulation analysis. Published in Migration. The Controversies and the Evidence, Ed. By R. Faini, J. De Melo and K.F. Zimmermann, Cambridge University Press, Cambridge, UK. Feenstra, R.C. and G.H. Hanson (1995). Foreign Direct Investment and Realtive Wages: Evidence from Mexico s Maquiladoras. NBER wp 5122, Cambridge, MA, USA. 14

16 Field, G. (1982). Place-to-Place Migration in Colombia. Economic Development and Cultural Change, 31, pp Fry, T.R.L. and M.N. Harris (1998). Testing for Independence of Irrelevant Alternatives. Some Empirical Results. Sociological Methods & Research. Vol 26 No 3, pp Gabriel, S.A., J. Shack-Marquez and W.L. Wascher (1996). Does Migration Arbitrage Regional Labor Market Differentials? Regional Science and Urban Economics, 23, pp Goss, E. P and N.C. Schoening (1984), Search time, Unemployment and the Migration Decision, Journal of Human Resources, 19: Gourieroux, Christian (2000). Econometrics of Qualitative Dependent Variables. Cambridge University Press, Cambridge, UK. Greenwood, M. J. (1997). Internal Migration in Developed Countries. In Handbook of Families and Population Economics. Rosenzweig and Stark (Eds.). North-Holland, Amsterdam. Hanson, Gordon and Antonio Spillembergo (1999) Illegal Immigration, Border Enforcemtn, and Relative Wages: Evidence from Apprehensions at the U.S.-Mexico Border American Economic Review 89: Herzog, H.W. Jr., A.M. Schlottmann and T.P Boehm (1993), Migration as Spatial JobSearch: A Survey of Empirical Findinggs, Regional Studies 27: Leonard, J.S. and R. McCulloch (1991). Foreign-Owned Businesses in the United State. Published in Immigration, Trade and the Labor Market, Ed. By J.M. Abowd and R.B. Freeman, NBER Project Report, The University of Chicago Press, Chicago, USA. Lucas, R.E.B. (1997) Internal Migration in Developing Countries, In Handbook of Families and Population Economics. Rosenzweig and Stark (Eds.). North-Holland, Amsterdam. Markusen, J.R. and S. Zahniser (1999). Liberalisation and incentives for labour migration: theory with applications to NAFTA. Published in Migration. The Controversies and the Evidence, Ed. By R. Faini, J. De Melo and K.F. Zimmermann, Cambridge University Press, Cambridge, UK. Markusen, J.R. and A.J. Venables (1997) The Role of the Multinational Firm in the Wage-Gap Debate, Review of International Economics,5: Pagano, M. (1990 Discussion of Muellbauer and Murpher (1990) Is the Balance of Payments Sustainable? Economic Policy 5: Cited in Cameron and Muellbauer (1998) 15

17 Pudney, S. (1989). Modeling Individual Choice. The Econometrics of Corners, Kinks and Holes. Basil Blackwell, Oxford, Uk. Roback, Jennifer (1982), Wages, Rents and the Quality of Life Journal of Political Economy 6: Sjaastad, Larry A. (1962) The Costs and Returns of Human Migration. Journal of Political Economy 70: Shultz, T.P. (1982). Lifetime Migration Within Educational Strata in Venezuela: Estimates of a Logistic Model. Economic Development and Cultural Change, 31, pp Spencer, P.D. (1989), Comments on Bover et al (1989), Oxford Bulletin of Economics and Statistics, 51: Cited in Cameron and Muellbauer (1998). Stark, Oded, and J. Edward Taylor(1991) Migration Incentives, Migration Types: The Role of Relative Deprivation, The Economic Journal, 101: Thomas, Alun (1993) The Influence of Wages and House Prices on British Interregional Migration Decisions, Applied Economics 25: Train, Kenneth (1986). Qualitative Choice Analysis. Theory, Econometrics, and an Application to the Automobile Demand. The MIT Press, Cambridge, MA, USA. Vanderkamp, J. (1971) Migration Flows, their Determinants and the Effects of Return Migration, Journal of Political Economy, 79:

18 Annex I. Amenities Indexes A measure of amenities across Mexican states was created by the National Institute of Economic and Geographical Information using the 2000 General Census of Population and Housing. However, it includes measures of labor market tightness as well as migration variables(percentage of resident population who were born in other states and percentage residing in other states) whose influence we are trying to separate out. We create a new index extracting principal component from : the percentage of the population living in urban areas, mortality rates, health and education infrastructure, and percentage of houses with electricity and without water s drainage. The analysis derived four significant factor loadings presented in table I.1. We select the first one, with the strongest interpretation of amenities. There urbanization, houses characteristics, houses characteristics and mortality rates are correlated in the expected way, that is, a higher value of the factor is more desireable. The correlation with wages for this measure is 0.37, while the correlation with the INEGI index is Figure # plots both indexes. In both cases Federal District has the highest values. However, although Mexico State has the second value for INEGI, for our index is under the mean. Table I.1 Factor Loadings Variable Uniqueness % Urbanization Mortality Mortality Mortality Mortality Mortality No.Hosp.per cap Beds in Hosp Drs. per cap Nurses per cap Prim.Sch.(1) High Sch.(1) No Drainage Electricity (1) Students per teachers. 17

19 6 Created Factor INEGI Welfare Index Amenities measures 18

20 Annex II. Cost of living across states (Pato, more detail please) The existing price indexes for Mexican states do not allow to make comparisons of cost of living across states. For this reason we create a price index using the Expenses Section of the ENIGHs 1992/8 (Encuesta Nacional de Ingresos y Gastos de los Hogares). The disposable information given by the survey allows us to create two indexes: Food and Housing 7. Price Indexes (relative to average of states) Housing Price Index Food Price Index Aguascalientes Baja California Norte Baja California Sur Campeche Coahuila Colima Chiapas Chihuahua Distrito Federal Durango Guanajuato Guerrero Hidalgo Jalisco Mexico Michoacan Morelos Nayarit Nuevo Leon Oaxaca Puebla Queretario Quintana Roo San Luis Potosi Sinaloa Sonora Tabasco Tamaulipas Tlaxcala Veracruz Yucatan Zacatecas The first one is a Laspeyres Price Index where the price and basket (200 items) of reference is the national average in The Housing Price Index was created using Hedonic Prices for rented houses only. The houses characteristics include : community s size; number of rooms; kitchen; bathroom; electricity; telephone; water drainage and potable water; and type of walls, floors and ceilings. Both indexes are plotted in Figure ##. 7 The ENIGH gives information on expenses for all items, but only unit price and quantity consumed for food items. Nevertheless the survey also includes House Characteristics which allow us to estimate also Housing. 19

21 Figure 1: Net Migration 2000 N W E S Percentage Miles 20

22 Figure 2 Per Capita FDI (From 1994 to 2000) Millions of Dollars N W E S Miles Graph 2 21

23 Percent Reduction in Imigration to USA if FDI is increased by 10% in a Mx State Most Likely Sit uat ion DF CAMP MEAN Alternatives Current Inmigration to USA Percent Reduction in Inmigration to USA if US$ 100 Million Are Invested in a Mx State Most Likely Sit uat ion BCS CHIS MEAN Alternatives Current Inmigration to USA 22

24 Table 1: Summary Statistics of Data Statistics Obs. Mean St. Dev. Min. Max. Emigrants from i to j 992 3,890 17, ,693 Population 992 2,848,697 2,440, ,494 11,170,796 Distance Distance sq Prices log Prices Unemployment Nominal wages log Nominal wages Amenities FDI log FDI Exports log Exports Imports log Imports Maquila , ,480 1, ,211 log Maquila

25 Table 2: Determinants of Migration (log formulation) Basic 1a Reduced sample 1b FDI 2a 2b Maquila (reduced sample) 3a 3b Exports 4a 4b Imports 5a 5b Distance (17.72) -(11.21) -(16.13) -(18.64) -(11.61) -(11.72) -(15.7) -(17.7) -(16.3) -(18.26) Distance sq (10.86) (7.28) (10.84) (11.74) (8.22) (7.77) (9.88) (10.68) (10.55) (11.19) Population i (5.15) (.78) (5.67) (4.15) -(.25) (1.13) (6.33) (5.34) (7.97) (7.98) Population j (22.97) (11.02) (20.35) (21.12) (9.85) (11.92) (20.61) (22.24) (20.46) (22.35) Log prices i (4.09) (3.82) (4.11) (4.86) (3.2) -(.42) (2.56) (3.98) (3.38) (2.17) Log prices j (5.37) -(1.09) (9.83) (5.13) (6.88) (.73) (13.72) (5.91) (11.5) (5.35) Relative Amenities (3.92) (1.2) (2.69) (2.25) (.23) -(.53) (3.21) (2.57) (2.66) (1.68) Amenities i (9.24) (.86) (11.54) (10.7) (4.57) (3.24) (10.01) (8.98) (12.79) (11.82) Labor market Relative unemp (3.64) -(2.21) -(4.45) -(2.16) -(3.94) -(3.79) Unemployment i (1.19) -(1.89) -(2.49) -(2.) -(1.29) -(.2) Relative nominal wage (11.73) (7.45) (9.41) (5.17) (8.9) (7.94) Log nominal wage i (6.) (4.98) (6.88) (5.66) (5.17) (7.27) Investment and trade Relative FDI/Maquila/Exports/Imports (log Zj-log Zi) (8.01) (.81) (5.69) (1.79) (8.76) (3.49) (10.34) (3.12) log of FDI i / Maquila i /Exports i / Imports i (log Zi) (1.66) -(3.6) (1.54) -(1.9) (3.9) (.37) (.54) -(4.3) DF and Mexico State (12.64) -(4.48) -(12.32) -(10.03) -(5.57) -(4.97) -(12.75) -(12.34) -(13.96) -(13.45) 24

26 Constant (14.57) -(8.62) -(15.42) -(14.61) -(10.53) -(9.24) -(16.37) -(14.25) -(15.96) -(15.08) Obs Note: All Zi expressed per capita 25

27 Table 3 Monthly wages and GDP per capita, Mexico and United States GDP per capita Current $US U.S. (3) Average 5,393 8,349 29,451 Min. 2,368 3,617 20,856 Max. 15,226 20,268 40,870 Max./Min Average US / Max. Mexico Monthly wage Current $US Mexico (1) Mexico (2) PPP Adjusted by State CPI PPP Adjusted by State CPI Current $US U.S. (4) Current $US Average ,716 Min Max Max./Min Average US / Max. Mexico Notes (1) INEGI; (2) Author's estimation for male urban workers using ENEU (3) Bureau of Economic Analysis ( ; (4) Bureau of Labor Statistics ( median earnings of Hispanic origin workers. 26

28 27

29 Table 4a: Impact of a 10% rise in FDI on Migration Reduce Emigration from (RE) Increase Inmigration to (IM) RE + IM Total Effect IM - RE Code Mexican States DF 09 DF MEX 15 MÉXICO VER 30 VERACRUZ JAL 14 JALISCO GTO 11 GUANAJUATO PUE 21 PUEBLA MICH 16 MICHOACÁN MOR 17 MORELOS NL 19 NUEVO LEÓN OAX 20 OAXACA GRO 12 GUERRERO HGO 13 HIDALGO TAMPS 28 TAMAULIPAS QRO 22 QUERÉTARO COAH 05 COAHUILA BC 02 BCN SIN 25 SINALOA CHIS 07 CHIAPAS TAB 27 TABASCO SON 26 SONORA SLP 24 SAN LUIS POTOSÍ AGS 01 AGUASCALIENTES BCS 03 BCS CHIH 08 CHIHUAHUA TLAX 29 TLAXCALA QROO 23 QUINTANA ROO COL 06 COLIMA ZAC 32 ZACATECAS YUC 31 YUCATÁN NAY 18 NAYARIT CAMP 04 CAMPECHE DGO 10 DURANGO

30 Table 4b: Impact of a 100 million investment in a Mexican state Reduce Emigration from (RE) Increase Immigration to (IM) RE + IM Total Effect IM - RE Code Mexican States BCS 03 BCS MOR 17 MORELOS COL 06 COLIMA QRO 22 QUERÉTARO QROO 23 QUINTANA ROO TLAX 29 TLAXCALA AGS 01 AGUASCALIENTES DF 09 DF CAMP 04 CAMPECHE HGO 13 HIDALGO NAY 18 NAYARIT TAB 27 TABASCO ZAC 32 ZACATECAS BC 02 BCN GRO 12 GUERRERO GTO 11 GUANAJUATO MICH 16 MICHOACÁN PUE 21 PUEBLA SLP 24 SAN LUIS POTOSÍ TAMPS 28 TAMAULIPAS OAX 20 OAXACA COAH 05 COAHUILA MEX 15 MÉXICO JAL 14 JALISCO SIN 25 SINALOA DGO 10 DURANGO VER 30 VERACRUZ SON 26 SONORA YUC 31 YUCATÁN NL 19 NUEVO LEÓN CHIS 07 CHIAPAS CHIH 08 CHIHUAHUA

31 30