Implications of immigration policies for the U.S. farm sector and workforce.

Devadoss, Stephen ; Luckstead, Jeff

I. BACKGROUND

In 1942, the United States and Mexico signed the Bracero program, a formal agreement to draw seasonal labor to work in U.S. agriculture, which provided employment opportunities to poor rural Mexicans. (1) However, this program was abolished in 1964, ending the legal seasonal labor supply to U.S. agriculture in the southwestern states. Concurrently, the United States passed the Immigration Act of 1965, establishing a quota of 120,000 legal immigrants per year. The abandonment of the Bracero program, the new legal-immigrants' quota limits, and economic inequality between the United States and Mexico paved the way for steady entry of illegal laborers and led to eventual insurmountable illegal immigration population in the United States.

Because illegal immigration was not a serious problem in the 1960s and 1970s, legislation addressed only the number of legal immigrants allowed to enter the United States. But in the 1980s, illegal immigration began to emerge as a national problem, and extensive debates entrenched around issues such as preventing the entry of unauthorized workers, providing public services to illegal immigrants, and even legalizing these workers. Consequently, the U.S. Congress attempted to address the immigration problems by enacting the 1986 Immigration Reform and Control Act (IRCA). The goals of IRCA were to eliminate the stock of undocumented workers through amnesty (2) and domestic enforcement of employer sanctions and curb the influx of illegal immigrants by increasing the border surveillance. Amnesty failed to eliminate the stock of illegal immigrants because only about half of the illegal immigrants filed for citizenship, and it created future expectation of amnesty and more illegal unauthorized entry. Furthermore, domestic sanctions on employers of undocumented workers and deportation of these workers were scantly enforced. To stop the influx of immigrants, IRCA focused heavily on tightening border control. The IRCA also legislated the H-2A program, which allowed agricultural employers to bring in guest workers during seasonal operations (ERS 2007). However, farmers complained that the cumbersome paperwork of H-2A and bureaucratic delay were not conducive to procure seasonal laborers at the time of peak farm operations such as vegetable and fruit picking. (3)

In spite of IRCA's amnesty provision and strengthened control measures, illegal immigration continued to rise--about 12 million unauthorized immigrants resided in the United States in 2007 (Martin 2007) which is reaffirmed by many popular press reports--leading to an extended congressional debate that began at the start of this decade to solve the illegal immigration problem. Several bills were proposed by the House of Representatives, the Senate, and the White House, addressing issues related to increased domestic and border enforcements, (4) paths to citizenship, and guest-worker programs (Montgomery 2006). These bills were not passed because of major disagreements among lawmakers over providing citizenship and guest-worker programs.

As a result of the failed legislations and the September 11 attack, the government primarily focused on border security. Accordingly, funding for border enforcement has steadily increased, (5) and resources were diverted from domestic to border enforcement. However, Boucher and Taylor (2007) documented that increased funding to secure the border did not deter undocumented workers from crossing the border because determined immigrants eventually find a way to enter the country by repeated attempts. Following September 11, 2001, the U.S. Immigration and Customs Enforcement (ICE) further decreased the number of human hours devoted to worksite inspection because monitoring critical infrastructure took priority (GAO 2005). For example, from 1999 to 2003, the number of human hours for domestic enforcement decreased from 480,000 to 18,000. (6,7) But, by late 2005, the U.S. government started to intensify domestic surveillance. For example, only 25 criminal arrests relating to illegal immigration occurred in 2002, but increased to 716 by 2006 and 1,103 by 2008 (U.S. Department of Homeland Security 2008c). Domestic surveillance has further intensified under the current administration (Meyer and Gorman 2009).

According to Passel (2008), a decreasing trend in the unauthorized immigrant population is recently occurring. (8) This is largely due to worksite and border enforcements and the recent U.S. economic recession. These enforcements have exacerbated U.S. agricultural labor shortages before the 2008/2009 economic crisis. According to the National Agricultural Worker Survey, 80% of the newly hired farm labor force is from Mexico, of which 96% are unauthorized (U.S. Department of Labor 2005). Therefore, as border and domestic enforcements intensified, entry of undocumented immigrants into the U.S. farm labor force was thwarted, which led to an acute labor scarcity. For example, the Wall Street Journal (2007) reported that in 2006, about 20% of agricultural products were not harvested nationwide. Furthermore, the Rural Migration News (2007) provides a detailed and specific list of these shortages and the adverse effect on crucial cultivational operations which resulted in heavy losses. As a result, farm groups are one of the strongest allies of overhauling the current guest-worker program to bring immigrants to legally work in U.S. agriculture.

For the last several decades, immigrants played a crucial role in the development and competitiveness of U.S. agricultural production (Torok and Huffman 1986). For example, Devadoss and Luckstead (2008) provide evidence of the importance of immigrant farm workers to vegetable production which is highly labor intensive. The United States has a great land endowment and ideal growing conditions; however, without immigrant labor who perform the back-breaking labor-intensive operations that U.S. low-skilled workers are unwilling to perform, agricultural productivity and total production would decline. Consequently, costs to U.S. consumers of agricultural products would increase and net exports would also decrease. In recent years, Mexican immigrant labor contributed significantly to the expansion of U.S. agricultural exports, particularly between the United States and Mexico. For example, between 1994 and 2008, net U.S. exports to the world and to Mexico increased by 82% and 200%, respectively (U.S. Department of Agriculture 2008f). Devoid of these laborers, this dramatic increase would not have been possible.

Although domestic and border enforcements address only the symptoms of illegal immigration, they are an important part of the immigration policy and the U.S. government devotes vast resources to prevent illegal entry and the employment of illegal aliens. The specific objectives of this paper are to: (1)analyze theoretically through illegal immigration and trade theory the effects of domestic and border enforcements on the illegal farm wage rate, commodity prices, unauthorized entry, and commodity trade between the United States and Mexico and (2) empirically implement the theoretical model through econometric estimation and simulation analysis and quantify the impacts of immigration policies on farm labor and commodity markets.

II. MODEL AND ANALYTICAL RESULTS

The model consists of two countries (United States and Mexico) that are linked through agricultural commodity trade and cross-border migration. (9) Over the past 15 yr, the integration of the United States and Mexico through trade has strengthened because of the naissance of the North American Free Trade Agreement (NAFTA), and also illegal labor migration from Mexico to the United States have burgeoned during this period, (10) Labor is used in the production of agricultural commodities in both countries. However, Mexican laborers are faced with a decision of working in Mexico or illegally immigrating to the United States. If they decide to immigrate, they could be apprehended at the border and sent back to Mexico, where these laborers can re-enter the Mexican labor market or reattempt to enter the United States. (11) The probability of getting apprehended is influenced by the U.S. government's resource allocation to border enforcement. Immigrants that successfully enter the U.S. labor market could also be caught--the probability of this apprehension depends on the U.S. government's budget allocation for domestic enforcements--by worksite raids and sent back to Mexico, where the labor cycle restarts.

The illegal laborers seeking employment in U.S. agriculture, combined with the U.S. domestic farm laborers, comprise the U.S. agricultural labor force which is an important input in U.S. commodity production. The commodity markets are influenced by U.S. subsidies and trade barriers, which also indirectly impact the labor markets. (12) The U.S. government provides subsidies to increase U.S. agricultural production and the resulting excess supply floods the Mexican market. Unable to compete, Mexican farmers are forced out of business and enter the labor migration cycle described above. In both countries, the economic conditions influence both the labor and commodity markets. If the U.S. economy is expanding, then more immigrants leave Mexico to the United States; however, if the economy is contracting, fewer illegal laborers enter the United States.

Model

The theoretical model captures the above described vertical link between labor and agricultural production, the horizontal link between the two countries through commodity trade and restricted labor flow, (13) and policy variables surrounding these interconnections.

A. Labor Market

The labor market specifications for the United States are:

(1a) [L.sup.S.sub.U] = [L.sup.S.sub.U] ([W.sub.U]);

(1b) [L.sup.D.sub.U] = [L.sup.D.sub.U] ([W.sub.U], [P.sup.S.sub.U])

where [L.sup.S.sub.U] is the U.S. farm labor supply, Wu the U.S. farm wage rate, [L.sup.d.sub.U] the U.S. demand for farm labor, and [p.sup.S.sub.U] the U.S. agricultural price support. The excess labor demand is expressed as:

(1c) [L.sup.ED.sub.U] = [L.sup.D.sub.U]([W.sub.U], [P.sub.S.sub.U]) - ([W.sub.U]).

The model incorporates U.S. employment of immigrant labor from Mexico to produce agricultural commodities. However, domestic enforcement policy, which punishes employers for utilizing this labor, limits the use of illegal labor in agricultural production. The agricultural employers pay lower wages for illegal workers than for domestic workers because worksite enforcements result in government sanctions such as fines and also imprisonment for hiring unauthorized workers. This wage discrepancy is modeled using the approach of Bond and Chen (1987) (14):

(2) [W.sub.U] = [W.sub.I] + [beta](E)c

where [W.sub.I] is the wage rate for an illegal worker, [beta] the probability an employer is caught hiring an illegal laborer, which depends on government expenditures (E) allocated to domestic enforcement, and c the monetary cost of the sanction. Thus, the U.S. wage rate for unskilled workers is equal to the illegal wage rate plus probability of getting caught times the penalty per illegal workers.

The labor market specifications for Mexico are:

(3) [L.sup.D.sub.M] = [L.sup.D.sub.M] ([W.sub.M], [P.sub.M])

where [L.sup.D.sub.M] is the demand for unskilled labor in Mexico, [W.sub.M] the Mexican unskilled wage rate, and [P.sub.M] the Mexican commodity prices. The total supply of unskilled labor ([bar.L]) is taken as exogenous as immigrants from Latin America also enter the United States via Mexico.

Mexican farm workers while immigrating spend a considerable amount of time seeking the assistance of coyotes (smugglers who bring illegal immigrants into the United States) and dangerously trekking to cross the border. Following Bandyopadhyay and Bandyopadhyay (1998), this model utilizes the illegal migration theory to model the Mexicans' illegal entry into the United States. The labor wasted ([L.sub.W]) in migration is:

[L.sub.W] = r[L.sub.I]

where r is the time wasted in crossing the border and [L.sub.I] the total labor attempting to migrate. The probability of getting caught at the border is denoted by d. Then, the total supply of illegal labor to the United States is:

[L.sup.S.sub.I] = (1 - d)(1 - r)[L.sub.I].

Combining the above two identities yields:

(4) [L.sub.W] = [r/((l - d)(1 - r))] [L.sub.S.sub.I].

The illegal labor supply to the United States is total supply of labor minus the labor demand and labor wasted:

(Sa) [L.sup.S.sub.I] = [bar.L] - [L.sup.D.sub.M] ([W.sub.M], [P.sub.M]) - [L.sub.W].

Substituting Equation (4) into Equation (5a) and solving for [L.sup.S.sub.I] yields:

(5b) [L.sup.S.sub.I] = [psi][[bar.L] - [L.sup.D.sub.M] ([W.sub.M], [P.sub.M)]]

where [psi] = ((1-d)(1-r))/(1-d(1-r)) is the porosity coefficient, which incorporates the labor wasted coefficient and the probability of getting apprehended crossing the border. As government expenditures increase for border surveillance, the probability of an illegal immigrants getting caught (d) increases, which leads to a lower [psi]([partial derivative][psi]/[partial derivative]d < 0).

In their decision to immigrate, they consider the Mexican wage rate and the U.S. illegal wage rate. We would expect that at the equilibrium:

(6a) [W.sub.M] = (1 - d)(1 - r)[W.sub.I] + d(1 - r)[W.sub.M].

That is, the Mexican wage rate is defined by weighing the U.S. illegal wage rate with the probability of successfully crossing the border and the Mexican wage rate with the probability of getting caught and re-entering the Mexican labor market. Solving for [W.sub.M] yields the wage linkage equation between Mexico's wage rate and the U.S. illegal wage rate:

(6b) [W.sub.M] = [psi][W.sub.I].

B. Commodity Market The specifications for the agricultural commodity market in the United States are:

(7a) [A.sup.S.sub.U] = [A.sup.S.sub.U] ([P.sup.S.sub.U], [W.sub.U]);

(7b) [A.sup.D.sub.U] = [A.sup.DS.sub.U] ([P.sub.U], [Z.sub.U])

where [A.sup.S.sub.U] is the U.S. commodity supply, [A.sup.DS.sub.U] the U.S. commodity demand, and [Z.sub.U] a vector of macroeconomic variables affecting the commodity demand. Producers receive an output subsidy for their commodity production which is captured by the U.S. price wedge equation:

[P.sup.S.sub.U] = [P.sub.U] + [S.sub.U]

where the producer support price ([P.sup.S.sub.U]) is equal to the consumer price ([P.sub.U]) plus the subsidy ([S.sub.U]) the government provides to producers. Because the United States is a net exporter of output, the excess supply is expressed as:

(7c) [A.sup.ES.sub.U] = [A.sup.S.sub.U] ([P.sup.S.sub.U], [W.sub.U]) - [A.sup.D.sub.U] ([P.sub.U], [Z.sub.U]).

The specifications for the commodity market in Mexico are:

(8a) [A.sup.S.sub.M] = [A.sup.S.sub.M] ([P.sub.M], [W.sub.M])

(8b) [A.sup.D.sub.M] = [A.sup.D.sub.M] ([P.sub.M], [Z.sub.M])

where [A.sup.S.sub.M] is the Mexican commodity supply, [P.sub.M] the Mexican commodity price, [A.sup.D.sub.M] the Mexican commodity demand, and [Z.sub.M] is a vector of macroeconomics variables that influence the commodity demand. Mexico is a net importer of the agricultural commodity and imposes an advalorem tariff (T) on agricultural imports. The price linkage equation is expressed as:

[P.sub.M] = [P.sub.U] (1 + T).

Mexico's excess demand is expressed as:

(9) [A.sup.ED.sub.M] = [A.sup.D.sub.M] ([P.sub.M], [Z.sub.M]) - [A.sup.S.sub.M] ([P.sub.M], [W.sub.M]).

C. Effects of Domestic and Border Enforcement on the Illegal Wage Rate and Commodity Price

The output price in the labor demand (Equations [1b] and [3]) and wage rates in the commodity supply equations (Equations [7a] and [8a]) capture the interrelationship between these markets. The equilibrium conditions require that the Mexican excess supply of farm labor that enters the United States illegally equal the U.S. excess demand for farm labor, and that the Mexican commodity excess demand equal the U.S. commodity excess supply. After equating these equations and substituting for the wage and price linkage identities, these equilibrium conditions are written as:

(10a) [L.sup.D.sub.U] ([W.sub.I] + [beta](E)c, [P.sub.U] + [S.sub.U]) - [L.sup.S.sub.U] ([W.sub.I] + [beta](E)c) [psi] [[bar]L- [L.sup.D.sub.M] ([psi][W.sub.I], [P.sub.U](1 + T))] = 0;

(10b) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Equations (10a) and (10b) with endogenous variables [W.sub.I] and [P.sub.U] are the core equations used for the analysis below. By totally differentiating Equations (10a) and (10b) and using Cramer's rule, we can examine the effects of a marginal change in domestic and border enforcements on the endogenous variables d[W.sub.I] and d[P.sub.U]. (15)

Because the Congress failed to pass any meaningful immigration reforms since IRCA in 1986, the Department of Homeland Security intensified its domestic raids in recent years to curtail the stock of unauthorized workers in the United States. For this stepped-up enforcement, the Department of Homeland Security increased its expenditures on domestic control. Hence, it is worth examining the impact of this higher domestic spending on the illegal wage rate. (16)

The effect of an increase in domestic enforcement on the illegal wage rate is (17):

(11a) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The first set of terms on the right hand side of Equation (1 la) reflects the direct effect in that increased spending on domestic control reduces the excess demand for unauthorized labor and the illegal wage rate. The magnitude of this wage decline depends on the probability of catching undocumented workers, the amount of sanctions, and the responsiveness of U.S. excess supply and Mexican excess demand in the commodity market. The second set of terms traces the indirect effect of the enforcement through the commodity market. Explicitly, higher domestic spending causes the wedge between the U.S. illegal and legal wage rates to be greater, leading to a higher U.S. legal wage rate. This higher legal wage rate reduces the U.S. commodity supply, causing U.S. and Mexican commodity prices to rise. This higher price increases labor demand and puts upward pressure on the illegal wage rate. Because domestic enforcement directly influences the wage rate, this direct effect will dominate the indirect effect working through the commodity market.

The impact of higher domestic spending on the commodity price is expressed as (18):

(11b) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The first set of terms describes how an increase in domestic enforcement affects the price through the U.S. commodity market. The higher domestic enforcement spending, by curtailing the demand for undocumented worker, augments the demand for domestic farm workers and increases the U.S. legal wage rate, forcing the commodity supply to decrease and thus increasing the U.S. commodity price. The level of price increase depends on how responsive the Mexican labor demand is to the change in the Mexican wage rate, the probability of an undocumented worker getting caught, the fine for employing illegal labor, and the porous nature of the border. The second set of terms captures the effect of high domestic expenditures on the commodity price through the Mexican commodity market. Specifically, the increase in expenditures reduces the demand for undocumented labor and the illegal wage rate. This leads to immigrants returning to Mexico, which lowers the Mexican wage rate. Consequently, the Mexican commodity supply expands which leads to a lower commodity price both in Mexico and the United States. The extent of the price decline depends on how responsive the U.S. excess demand for labor is to the wage rate, the probability of an illegal worker caught, the fine for employing illegal labor, and the porous nature of the border. The net effect of domestic enforcement on the U.S. commodity price will be positive because the direct effect through illegal labor demand in the United States (the set of first terms) is likely to dominate the indirect effect through the commodity market in Mexico (the second set of terms).

In Equations (11a) and (11b), the term c[beta]' offers an insight on the U.S. government's commitment to enforcing its domestic surveillance policy. Between the start of IRCA in 1986 and the start of the recent congressional debate on illegal immigration, the United States did not strictly implement its domestic enforcement policy. During this period, the values of c[beta]' are low, and in 2006 the United States stepped up the enforcement of its domestic policies, and consequently, for this period the value of c[beta]' is high.

The Department of Homeland Security has drastically heightened the border enforcement measures in recent years. The effect of this increased surveillance on the illegal wage rate is:

(12a) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

As explained below Equation (5b), greater border enforcement causes [psi] to decrease, indicating a tighter border control. Both sets of terms in Equation (12a) reinforce each other of how heightened border surveillance raises the illegal wage rate by curbing the illegal labor flow. The first set of terms demonstrates the Mexican labor-market effect. As border enforcement is tightened, more immigrants are apprehended and returned back to Mexico. Because fewer laborers enter the U.S. labor market, the illegal wage rate rises. The magnitude of the illegal wage reduction depends on how responsive the U.S. commodity excess supply and the Mexican excess demand are to a change in the price. The second set of terms reflects the Mexican commodity-market effect. Specifically, as a result of heightened border security, more workers return to Mexico, causing the wage rate to decline and the commodity production to expand. As more laborers are absorbed in Mexican agriculture, fewer laborers enter the United States, driving the illegal wage rate up. The degree to which the wage rate increases depends on how responsive the labor demand is to a price increase. Thus, both effects raise the illegal wage rate.

The effect of tighter border security on the commodity price is determined by:

(12b) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The first set of terms expounds the impact of border control on the U.S. commodity price through the repercussions in the Mexican commodity market. As border enforcement increases, more laborers remain in Mexico which reduces the Mexican wage rate. This wage decrease expands the Mexican commodity supply, decreasing Mexico's excess demand, causing the commodity price to fall. The level of price decline depends on how responsive the U.S. excess demand for labor is to a change in the wage rate. The second term reflects the change in the price through U.S. commodity supply adjustments. As border enforcement tightens, the illegal labor supply in the United States contracts. This causes U.S. producers to augment their demand for domestic labor, which increases the legal wage rate. The wage rate increase reduces the U.S. commodity supply and raises the commodity price. The magnitude of this increase depends on the excess supply of labor in Mexico, the porous nature of the border, and the responsiveness of labor demand in Mexico to the change in the wage rate. The net effect of tighter border control is to raise the commodity price because the labor shortage effect is likely to dominate the commodity market effect.

D. Effects of Domestic and Border Enforcement on Labor Flow and Trade

This subsection examines the impacts of U.S. immigration policies on unauthorized entry and agricultural trade. Because the labor excess supply and excess demand are equal (see Equation [10a]) and the commodity excess supply and excess demand are equal (see Equation [10b]), the excess supply of labor (Equation [5b]) and excess demand for commodity (Equation [9]) can be totally differentiated to determine changes in the illegal labor flow into the United States and U.S. commodity exports to Mexico:

(13a) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

(13b) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

To examine the effect of an increase in the domestic enforcement in the United States on the illegal labor flow, d[W.sub.I] and d[P.sub.U] from Equations (11a) and (11b) are substituted into Equation (13a). Holding all other exogenous variables, except for domestic enforcement, constant in Equation (13a), the change in illegal labor flow in response to tighter domestic control is stated as:

(14a) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The first set of terms shows the wage effect. Because tighter domestic control is likely to reduce the illegal wage rate, the unauthorized entry will contract. The second set of terms demonstrates the price effect. Increased domestic enforcement will make production more expensive, leading to a higher commodity price and increased demand for labor. The net effect of domestic enforcement is to reduce the cross-border migration because the wage effect is likely to dominate the price effect.

To study the effect of an increase in the domestic enforcement on commodity trade, d[W.sub.I] and d[P.sub.U] from Equations (11a) and (11b) are substituted into Equation (13b). Holding all other exogenous variables, except for domestic enforcement, constant in Equation (13b), the change in U.S. exports to Mexico is expressed as:

(14b) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The first set of terms illustrates the wage effect. Tighter domestic surveillance forces undocumented workers to return to Mexico, which will increase the Mexican workforce and reduce the Mexican wage rate. This lower wage rate will increase Mexican commodity production, leading to lower imports. The second set of terms demonstrates the price effect. Increased domestic enforcement will cause the U.S. wage rate to rise and make production more expensive, leading to a lower U.S. commodity supply. This will result in a higher commodity price and lower imports by Mexico. Thus, both effects reinforce each other in reducing U.S. exports to Mexico.

To analyze the effect of heightened U.S. border enforcement on the illegal labor flow, d[W.sub.I] and d[P.sub.U] from Equations (12a) and (12b) are substituted into Equation (13a). Holding all other exogenous variables, except for border security, constant in Equation (13a), the change in illegal labor flow in response to a change in border control is written as:

(15a) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

In Equation (15a), the first set of terms shows the wage effect. Strengthened border surveillance reduces the illegal entry and thus raises the illegal wage rate. Even though immigrants apprehended at the border are sent back to Mexico, the higher illegal wage rate in the United States lures them to cross the border repeatedly. The second set of terms illustrates the price effect. As the United States implements additional measures to secure its borders, fewer illegal workers enter the United States, causing the cost of production and commodity price to rise. This also results in a higher Mexican commodity price. The higher Mexican price draws would be immigrant laborers back into Mexican production, which contracts the illegal labor supply. The third set of terms represents the direct effect of an increase in U.S. border security on the illegal labor supply. As a result of the tightened border control, fewer laborers successfully cross the U.S. border, which reduces the supply of unauthorized labor. The combined effect of the three terms should result in fewer illegal laborers entering the U.S. labor market from Mexico.

To examine the effect of heightened U.S. border security on the commodity trade, d[W.sub.I] and d[P.sub.U] from Equations (12a) and (12b) are substituted into Equation (13b). Holding all other exogenous variables, except for border surveillance, constant in Equation (13b), the change in commodity trade resulting from tighter border security is expressed as:

(15b) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

The first set of terms in Equation (15b) demonstrates the wage effect. Increased border surveillance reduces the illegal entry and thus drives up the illegal wage rate in the United States, but augments the Mexican workforce and lowers the Mexican wage rate, which expands Mexican commodity production, leading to lower imports. The second set of terms represents the price effect. An increase in border enforcement curtails the illegal workforce in U.S. agricultural production, leading to an increase in U.S. production cost, reduced supply, and lowers exports to Mexico. The third set of terms conveys the indirect effect of heightened border enforcement on commodity trade through changes in the wage rate. This effect also reduces U.S. exports to Mexico. Thus, the combined effects of all three terms reinforce each other in reducing U.S. exports to Mexico.

III. MODEL ESTIMATION

This section describes the estimation procedure, data, and the results of the empirical model. Table 1 presents the estimated equations and identities, and Table 2 provides the variable definitions. The system of equations was estimated using the three-stage least square technique, which uses generalized least squares and corrects for cross equation correlation to achieve efficient, consistent, and unbiased estimates. The system weighted [R.sup.2] is 0.97, and the direction

of influence for all of the estimated parameters is consistent with the theoretical predictions.

The data used for estimating U.S. farm labor and commodity markets, Mexican farm labor and commodity markets, price and wage linkage equations are obtained from several sources. The agricultural labor data for the United States is collected from the U.S. Department of Agriculture (2008a, 2008c, 2008i) and the U.S. Department of Labor (2008a, 2008b). The farm labor data for Mexico are obtained from the Banco de Mexico (2008), Comision Nacional de los Salarios Minimos (2008), Food and Agricultural Organization (2008a), and Organization for Economic Cooperation and Development (2008). The agricultural commodity market data for the United States and Mexico are collected from the Food and Agricultural Organization (2008b, 2008c), International Monetary Fund (2008), U.S. Department of Agriculture (2008e, 2008f, 2008h), and U.S. Department of Commerce (2008a, 2008b, 2008c). The data for the price and wage linkage equations are gathered from the U.S. Congress (1993), U.S. Department of Agriculture (2008b, 2008g), and U.S. Department of Homeland Security (2007a, 2007b, 2008a, 2008b). The data were compiled for the period 1989-2007.

The U.S. agricultural labor supply is estimated as a function of the real agricultural wage rate, real manufacturing wage rate, and total non-farm employment. The estimated coefficients for these variables are significant at the 1% level. The real wage rate is an important determinant of labor supply, and the elasticity estimate is elastic at 1.12. The manufacturing wage negatively influences the farm labor supply as higher wages in the manufacturing lures labor from agriculture, which is consistent with illegal labor moving from agriculture to construction during the U.S. housing boom in the early part of this decade. The elasticity estimates for the manufacturing wage rate are elastic at -1.16. The estimated coefficient for the total non-farm employment is negative, indicating that the agricultural sectors' labor supply is diverted to the manufacturing and service sectors, where employment is typically located in urban setting, steady, and non-seasonal.

The explanatory variables used to estimate the U.S. agricultural labor demand are the U.S. real wage rate, real subsidized producer price, number of U.S. farms, and a binary variable. The signs of the estimated coefficients of the real wage rate and output price are consistent with the theoretical predictions and significant at the 5% and 10% levels, respectively. The estimated elasticities for these variables, as one would expect, are relatively inelastic at -0.41 and 0.30, respectively. The estimated coefficient for the number of farms is positive. A dichotomous variable is created, one for the year 1999 and zero otherwise, to account for the sharp increase in labor use in agriculture in that year because of a rise in unauthorized workers.

The wedge between the U.S. agricultural legal and illegal wage rates accounts for the probability of an undocumented worker being detained (0.0003) and employer sanctions for hiring an undocumented worker ($6,073). Thus, undocumented workers accept lower wage rates because of the risk they impose on their employers.

The U.S. commodity supply is estimated using the essential supply determinants: prices received including farm supports and the real wage rates. The estimated coefficients for both these variables are significant at the 1% and 5% levels. The price elasticity of supply is 0.80 and wage elasticity of supply is relatively inelastic at -0.16. The U.S. agricultural commodity demand is estimated by utilizing real food and beverage CPI and real personal income as regressors. The price and income elasticity of demand are inelastic at -0.43 and 0.37, respectively.

In the empirical analysis, we differentiate U.S. producer and consumer prices by incorporating the transportation cost, processing costs, and market margins. This difference is determined by regressing the food and beverage CPI on the supply price. The estimated coefficient for the intercept and supply price are positive and significant at the 5% and 10% levels, respectively, implying that there is a significant market margin between the producer and consumer prices. The next equation presents U.S. exports to the rest of the world. The commodity market equilibrium identity entails that U.S. production minus demand minus exports to the rest of the world (excluding Mexico) equal Mexican demand plus net exports to the rest of the world (excluding the United States) minus production.

The Mexican labor demand is estimated as a function of real wage rate, real agricultural producer price index, and real textile producer price index as explanatory variables. The real wage rate and the textile producer price index are significant at the 1% level. The labor demand elasticities with respect to the wage rate and producer price are very inelastic. The estimated coefficient for textile prices is positive, indicating an increase in these prices augments cotton production and thus raising the agricultural labor demand.

The linkage equation for the Mexican unskilled wage rate and illegal wage rate in the United States explains the relationship between these two wage rates by accounting for the probability an illegal immigrant is detained while attempting to cross the border and the time an illegal immigrant wastes during migration. Since late 2001, the U.S. government has drastically increased its border enforcement efforts for national security reasons, particularly on the southern border. The computation based on the data show that the probability an undocumented worker being apprehended is 61%, and an immigrant's total work time wasted while attempting to emigrate from Mexico to the United States is 21%. These values are plugged into Equation (6a) to obtain the Mexican and U.S. illegal wage linkage equation, which is further simplified as in Equation (6b) to obtain the porosity coefficient of 0.59. The next equation is the labor market equilibrium between the two countries, which states that U.S. excess demand for farm workers is equal to the Mexican excess supply of labor that cross the border successfully, which is captured by the porosity coefficient.

The explanatory variables used in the estimation of the Mexican agricultural commodity supply are the lagged dependent variable, agricultural producer price index, wage rate, and manufacturing producer price index as a proxy for input prices. The agricultural commodity price, wage rate, and input prices are important determinants of the commodity supply. The 1-yr lagged dependent variable is included as a regressor to account for the rigidity in Mexican production, and the estimated coefficient for this variable is highly significant and less than one. The elasticity of agricultural commodity supply with respect to the producer price and wage rate are 0.40 and -0.11, respectively. The Mexican agricultural commodity demand estimation utilizes the agricultural CPI and gross domestic product (GDP) as regressors, and the estimated coefficients are significant at the 1% level. The commodity demand elasticities with respect to the agricultural CPI and GDP are -2.46 and 1.09, respectively.

The empirical model incorporates the wedge between the Mexican producer and consumer prices by taking into account transportation costs, processing costs, and market margins. The estimated coefficient for the consumer price is positive and significant at the 5% level. The estimated intercept is significant and positive, demonstrating that producer and consumer prices do differ. The relationship between the U.S. and Mexican agricultural CPI is described by the U.S. Mexican price linkage equation by accounting for the exchange rate, ad valorem tariff, and transportation cost between the countries.

IV. SIMULATION RESULTS

The estimated system of equations is used to conduct simulation analysis to examine the effects of a change in the border and domestic enforcement policies on the endogenous variables over the period 1994-2007. A benchmark simulation is run using the historical values of the explanatory variables and two alternate scenarios are run to analyze the impacts of changes in domestic and border enforcement policies. Because of the vertical linkages through the labor and commodity production and horizontal linkages through commodity trade and labor migration, changes in these policies impact both countries' labor and commodity markets.

A. Domestic Enforcement Effect

The baseline is run using the actual domestic enforcement expenditure as the gauge for the level of worksite surveillance. The alternate scenario is run with a 10% increase in the enforcement budget over the baseline (Table 3). Comparison of the alternate scenario with the baseline offers insight into the impacts of tighter domestic enforcement on the endogenous variables. Greater spending on domestic enforcement reduces the illegal labor employment, and causes labor shortages, because undocumented workers are concerned about getting caught and deported and employers are concerned about government sanctions, including heavy fines and jail time. The results show that the decline in labor use is more pronounced in the recent years, which is consistent with current events, that is, stricter workplace raids have not only led to curtailing of illegal employment but also undocumented workers are returning to Mexico and fewer illegal immigrants are entering the United States. Specifically, as the demand for illegal labor declines, the illegal wage rate for farm workers also declines. The simulation results show that a 10% increase in domestic expenditure reduces the illegal wage rate on average by 11.49%, but the reduction is more pronounced in the later years. For example, the illegal wage rate falls by 36.11% in 2006 and 55.23% in 2007, which is significantly larger than the 0.18% decline in 1994. The lower labor demand and depressed illegal wage rate reduces the illegal agricultural labor flow by 9.12% or about 42,000 illegal immigrants in 2007.

The lower labor demand reduces U.S. agricultural production. As U.S. production declines, fewer farm products are exported to Mexico. As shown in Table 3, the agricultural trade flow is reduced by 0.04% in 1994 and 22.67% in 2007. (19) This is consistent with the larger decline of illegal labor employment in recent years. The decline in the U.S. supply and lower exports to Mexico causes the commodity prices to increase in both countries. Because fewer immigrants are entering the United States and undocumented workers are returning to Mexican agricultural employment, the Mexican farm supply increases.

These results confirm the theoretical findings that an increase in the worksite enforcement will deter employers from hiring illegal immigrants. However, this policy reduces trade between the countries. Thus, domestic enforcement has a tradeoff between curbing the undocumented workers versus lower agricultural production and exports.

B. Border Enforcement Effect

The impact of border control can be analyzed using enforcement spending or human hours patrolling the border. Both these variables have an impact on border apprehensions. Consequently, in this study, the probability of border apprehension is used as the policy variable to stem the flow of unauthorized immigrants. The baseline is run using a border apprehension probability (d) increasing from 0.30 in 1994 to 0.40 in 2001, which reflects a fairly porous border. To account for the increase in border security after September 11, d is increased from 0.50 in 2002 to 0.60 in 2007. The value of 0.60 for d in 2007 was calculated based on the actual data. The alternate scenario is run by considering heightened border enforcement because of increased illegal immigration and national security concerns, which are evident from the larger budget allocation for border fencing and also minuteman operations monitoring the border. Thus, d is set at 0.60 over the entire simulation (19.) It is worth noting that baseline net exports are small in 2005 because of the Mexican peso crisis which lead to a 50% devaluation. The small value of baseline net exports magnifies the impacts.

period in the alternate scenario. Thus, comparison of the alternate scenario to the baseline offers insight into the effect of heightened border security on the endogenous variables from the beginning of the simulation period (Table 4).

As the probability of apprehension of illegal immigrants at the border increases, the flow of undocumented workers into the United States lessens. The results show that if the border security level in 1994 is the same as that in 2007, then the inflow of undocumented workers would have declined by 2.96% or about 8,147 in 1994. This large decline in illegal immigrant flow holds during much of the 1990s. These results are consistent with the real world occurrences. For instance, the National Statistics and Geography Institute survey reports that tightened border surveillance drastically increased the apprehension rate (Olson 2008). The contraction of the illegal labor supply to U.S. agriculture drives up the U.S. illegal wage rate, while reducing the Mexican wage rate as more workers are turned back to Mexico. The U.S. illegal wage rate increases by an average 1.27% in the 1990s; whereas the Mexican wage rate decreases by an average of 1.45% during this period. As the number of illegal farm workers dwindles, the demand for U.S. domestic workers increases, and consequently, the U.S. legal wage rate rises. In response to this increase in the wage rate, the domestic farm labor supply increases; however, the total labor use declines because of fewer unauthorized farm workers entering the country.

The increase in the illegal and legal wage rates drives up the cost of production in U.S. agriculture. As a result, the U.S. agricultural production decreases, which reduces the U.S. exports to Mexico. The U.S. exports to Mexico fall by about 5.0% over the simulation period (Table 4). Because of the decline in U.S. production and fewer exports entering Mexico, the price of agricultural products increases in both countries.

The results of an increase in border security corroborate the findings of the theoretical analysis in that both the illegal labor flow and commodity trade decrease. These finding show a similar tradeoff between immigration and commodity trade as in the domestic enforcement policy.

V. CONCLUSION

Because of the failure of the 1986 IRCA legislation to eliminate the stock and reduce the inflow of illegal immigrants and the heightened national security concern after the September 11 attack, the U.S. government stepped up the border enforcement beginning in 2002. Because of the legislative impasse to pass immigration reform, the U.S. government also tightened the domestic enforcement starting in 2005. This study develops a trade and illegal-immigration model to analyze the impacts of domestic and border enforcements on U.S. farm labor and commodity markets. The theoretical results show that increased resource allocation to domestic enforcement and border control contracts the flow of illegal labor but also reduces commodity trade. The empirical analysis utilizes a three-stage least square model to estimate the structural equations and simulation runs to quantify the impacts of domestic and border controls. The results of the simulation analysis are consistent with the theoretical results. A 10% increase in spending on domestic enforcement curbs illegal labor use by an average of 8,947 workers to U.S. agriculture and reduces the commodity trade by an average of $180 million. If the recent tighter border security were enforced from 1994, the illegal labor force to U.S. agriculture would have declined by 8,147. Heightened border enforcements reduce the employment of undocumented workers and commodity production, which causes U.S. agricultural exports to Mexico to decline by an average of 5%. The results for these two scenarios show a distinct tradeoff between a reduction in illegal labor flow and commodity trade.

In response to the economic downturns, heightened border surveillance, and new national- and state-level workplace enforcements, undocumented workers are returning to Mexico and also fewer immigrants are attempting to enter the United States, which reduce the farm labor supply. As a result, the U.S. agricultural sector is facing labor shortages in labor-intensive operations such as vegetable production, food processing, and manufacturing (Gans 2007; Wall Street Journal 2007). (20) Despite the economic slowdown, farmers still need labor to carry out crucial farm operations such as planting and harvesting. Producers in several states have been beset with labor scarcity and are experiencing devastating effects on farm production and profitability.

Consequently, consumers have also incurred higher costs for fruits and vegetables at the grocery stores. Any reduction of the immigrant workforce, by deporting undocumented workers. and scuttling the guest-worker program, has several adverse implications for U.S. agriculture.

The current IRCA guest-worker program has come under sharp attack by the U.S. businesses, particularly by farmers, because it requires cumbersome paperwork and unduly bureaucratic delays, which are not conducive to procuring seasonal laborers at the time of peak farm operations such as vegetable and fruit picking. Therefore, U.S. government policies aimed at deporting unauthorized workers--without taking adequate measures to supply farm laborers through guest-worker programs--will adversely affect the supply of farm laborers to crop production. If immigration reform allows a well-functioning guest-worker program, it can increase the availability of the farm workforce and will have a positive impact on agricultural production and profitability, as well as reducing the cost of agricultural products to consumers.

ABBREVIATIONS

CPI: Consumer Price Index

GDP: Gross Domestic Product

ICE: Immigration and Customs Enforcement

IRCA: Immigration Reform and Control Act

NAFTA: North American Free Trade Agreement

PPI: Producer Price Index

REFERENCES

Banco de Mexcio. "Statistics". 2008. Accessed 2 October 2008. http://www.banxico.org.mx/sitioingles/ PortalesEspecializados/inflacion/inflacion.html.

Bandyopadhyay, S. "Illegal Immigration and Second-best Import Tariffs." Review of International Economics, 14(1), 2006, 93-103.

Bandyopadhyay, S., and S.C. Bandyopadhyay. "Illegal Immigration: A Supply Side Analysis." Journal o/' Development Economics, 57(2), 1998, 343-60.

Bond, E. W., and T.-J. Chen. "The Welfare Effects of Illegal Immigration." Journal of International Economics, 23(3-4) 1987, 315-28.

Boucher, S. R., and J. E. Taylor. "Policy Shocks and the Supply for Mexican Labor to U.S. Farms." Choices, 21, 1st quarter, 2007, 37-42.

Calavita, K. Inside the State: The Bracero Program, Immigration, and the I.N.S. New York: Routledge, 1992.

Comision National De Los Salarios Minimos. "Secretaria Del Trabajo Y Prevision Social." 2008. Accessed 13 October 2008. http://www.conasami.gob.mx/.

Devadoss, S. "An Evaluation of Canadian and U.S. Policies of Log and Lumber Markets." Journal of Agricultural and Applied Economics, 40(1), 2008, 171-84.

Devadoss, S., and J. Luckstead. "Contributions of Immigrant Farmworkers to California Vegetable Production." Journal of Agriculturul and Applied Economies, 40(3), 2008, 879-94.

Djajic, S. "Illegal Aliens. Unemployment and Immigration Policy." Journal of Development Economics, 25(1), 1987, 235-49.

--. "Illegal Immigration and Resource Allocation." International Economic Review, 38(1), 1997, 97-117.

ERS (Economic Research Service). "Immigration Reform and Control Act of 1986." Washington DC, 2007. Accessed 24 May 2007. http://www.ers.usda.gov/publications/ah719/ah719f.pdf.

Ethier, W. J. "Illegal Immigration: The Host-Country Problem." American Economic Review, 76(1), 1986a, 56-71.

--. "Illegal Immigration." American Economic Review, 76(2), 1986b, 258-62.

Food and Agricultural Organization. "FAOStat." United Nations. 2008a. Accessed 20 October 2008. http://faostat.fao.org/site/613/DesktopDefault.aspx?PageID=613#ancor.

--. "TRADEStat." United Nations. 2008b. Accessed 20 October 2008. http://faostat.fao.org/site/406/default.aspx.

--. "PopSTAT." United Nations. 2008c. Accessed 23 October 2008. http://faostat.fao.org/site/550/DesktopDefault.aspx?PagelD=550#ancor.

Gans, J. The Economic Impacts of Immigrants in Arizona. Tucson: Udall Center for Studies in Public Policy, University of Arizona, 2007.

GAO (General Accounting Office). 2005. "Immigration Enforcement: Preliminary Observations on Employment Verification and Worksite Enforcement Efforts." GAO-05-822T, 2005.

Gaytan-Fregoso, H., and S. Lahiri. "Foreign Aid and Illegal Immigration." Journal of Development Economics, 63(2), 2000, 515-27.

Hanson, G. H. "'Illegal Migration from Mexico to the United States." Journal of Economic Literature, 44(4), 2006, 869-924.

Hanson, G. H., and A. Spilimbergo. "Illegal Immigration, Border Enforcement, and Relative Wages: Evidence from Apprehensions at the U.S.-Mexico Border." The American Economic Review, 89(5), 1999, 1337-57.

International Monetary Fund. "World Economic and Financial Surveys." World Economic Outlook Database. 2008. Accessed 15 October 2008. http://www.imf.org/ external/pubs/ft/weo/2008/01/weodata/index.aspx.

Martin, P. "Immigration Reform, Agriculture, and Rural Communities." Choices, 21, 1st quarter, 2007, 43-47.

Meyer, J., and A. Gorman. "Napolitano Targets U.S. Employers in Immigration Effort." Los Angeles Times, March 30, 2009, p. A3.

Montgomery, D. 2006. "Senate Passes Immigration Reform." The Spokesman-Review, 2006, p. A6.

Olson, A. 2008. "Mexican Emigration Drops Sharply." The Spokesman-Review. 21 November, A4.

Organization for Economic Cooperation and Development. "Trade and Agricultural Directory." 2008. Accessed 17 June 2008. http://www.oecd.org/document/59/.

Passel, J. "Trends in Unauthorized Immigration: Undocumented Inflow Now Trails Legal Inflow." Pew Hispanic Center. 2008. Accessed 9 February 2009. http://pewhispanic.org/reports/report.php?ReportID=94.

Rural Migration News. "Farm Labor Shortage." 14, no. 1. 2007. Accessed 15 June 2007. http://migration.ucdavis.edu/rmn/more.php?id=1182_0_4_0.

The Wall Street Journal. "Immigration Non-Harvest." 20 July, 2007. p. A12.

Torok, S. J., and W. E. Huffman. "U.S. Mexican Trade in Winter Vegetables and Illegal Immigration." American Journal of Agricultural Economics, 68(2), 1986, 246-60.

U.S. Congress, 1st Session, November 4, 1993. Message from the President of the United States Transmitting North American Free Trade Agreement, Texts of Agreement. Implementing Bill, Statement of Administrative Action and Required Supporting. Washington D.C.: Statements House Document 103-159, Volume 2.

U.S. Department of Agriculture. "Farm Labor." Washington, DC. Accessed 1 October 2008. http:// usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.do?documentID=1063.

--. "Quick Star (Agricultural Statistics Data Base)." Washington DC. 2008b. Accessed 29 September 2008. http://www.nass.usda.gov/Data and Statistics/Quick_Stats/index.asp.

--. "Agricultural Prices." Washington DC. 2008c. Accessed 2 October 2008. http:// usda.mannlib.cornell.edu/MannUsda/viewDocumentInfo.do?documentID=1002.

--. "Data Sets Farm Income: Data Files." Washington DC. 2008d. Accessed 3 October 2008. http://www.ers.usda.gov/data/farmincome/

--, "Foreign Agricultural Service. U.S. Trade Exports--BICO Commodity Aggregations." Washington DC. 2008e. Accessed 5 October 2008. http://www.fas.usda.gov/ustrade/USTExBICO.asp?QI=.

--. "Economic Research Service. Data." Washington DC. 2008f. Accessed 8 October 2008. http://www.ers.usda.gov/Data/ExchangeRates/.

--. "Farm and Commodity Policy: Government Payments and the Farm Sector.'" Washington DC. 2008g. Accessed September 9 2008. http://www.ers.usda.gov/Briefing/FarmPolicy/gov-pay.htm.

--. "Publications: Statistical Bulletins." Washington DC. 2008h. Accessed 2 October 2008. http://www.nass.usda.gov/Publications/ Statistical_Bulletins/Final_Estimates_19871993.

U.S. Department of Commerce. "Implicit Price Deflators for Gross Domestic Product." Bureau of Economic Analysis. Washington DC. 2008a. Accessed 30 September 2008. http://www.bea.gov/national/index.htm.

--. "National Income and Product Accounts Table.". Bureau of Economic Analysis. Washington DC. 2008b. Accessed 3 October 2008. http://www.bea.gov/national/nipaweb/ TableView.asp?SelectedTable=58&Freq=Qtr&FirstYear=2006&LastYear=2008.

--. "CPI Detailed Report, Monthly." Bureau of Economic Analysis. Washington DC. 2008c. Accessed 12 October 2008. http://www.bea.gov/national/nipaweb/TableView.

U.S. Department of Homeland Security. "Yearbook on Immigration Statistics. Enforcement.'" Washington DC. 2007a. Accessed 30 October 2008. http://www.dbs.gov/ximgtn/statistics/publications/YrBk07En.shtm.

--. "Estimates of the Unauthorized Immigrant Population Residing in the United States." January 2006 Population Estimates. Washington DC. 2007b. Accessed 30 October 2008. http://www.dhs.gov/xlibrary/assets/statistics/ publications/ois_ill_pe_2007.pdf. "Frequently Asked Questions about Worksite Enforcement." Washington DC. 2008a. Accessed 10 November 2008. http://www.ice.gov/pi/news/factsheets/worksite.htm.

--. "U.S. Immigration and Customs Enforcement." NEWS. Washington DC. 2008b. Accessed 11 November 2008. http://www.ice.gov/pi/investigations/worksite/newsreleases.htm.

--. "U.S. Immigration and Customs Enforcement." NEWS. 2008c. Accessed 11 November 2008. http://www.ice.gov/pi/news/factsbeets/worksite.htm.

U.S. Department of Labor. National Agricultural Workers Survey. "Findings from the National Agricultural Workers Survey 2001-2002." 2005. Accessed 21 May 2007. www.doleta.gov/agworker/naws.cfm.

--. "Employment and Training Administration. The National Agricultural Workers Survey." Washington DC. 2008a. Accessed 29 September 2008. http://www.doleta.gov/agworker/naws.cfm.

--. Bureau of Labor Statistics. "Employment, Hours, and Earnings from the Current Employment Statistics Survey (National).". Washington DC. 2008b. Accessed 27 September 2008. http://data.bls.gov/PDQ/servlet/SurveyOutputServlet.

(1.) Calavita (1992) and Hanson and Spilimbergo (1999) provide more details about the Bracero program.

(2.) This provision provided a one-time amnesty to unauthorized workers to apply for citizenship.

(3.) IRCA also introduced the special agricultural workers (SAW) program, which legalized an additional 1.2 million agricultural workers, and the replenishment agricultural worker (RAW) program, which permitted hiring of additional immigrants from foreign countries if the shortages occurred as a result of SAW moving to other industries. However, RAW was never implemented because the Department of Labor determined that a labor shortage did not occur right after IRCA was passed (ERS 2007).

(4.) These include employer lines for hiring illegal immigrants: employee verification system for employers: border fences: and felony charges for anyone assisting, encouraging, directing, inducing illegal entrance, and for anyone attempting to enter or remaining illegally in the country (Montgomery 2006).

(5.) For example, in 2005 expenditures on border enforcement increased to $2.2 billion, a six-fold increase over the past 25 yr (Hanson 2006).

(6.) The number of employers paying fines of $5,000 or more fell from 15 in 1990 to 0 in 2004 (U.S. General Accounting Office 2005).

(7.) Under IRCA, employer fines range from $250 to $10.000 per illegal workers and up to 6 yr in jail for repeat offenders.

(8.) Passell (2008) observes illegal immigrants peaked in 2007 at 12.4 million and dropped to 11.9 million in 2008.

(9.) This theoretical model is expanded to include the rest of the world in the empirical analysis.

(10.) For instance, U.S. (Mexican) agricultural imports from Mexico (United States) increased by almost 280% (250%) since the implementation of NAFTA (U.S. Department of Agriculture 20081). Also, because of the congruent location and employment opportunities in the United States, Mexican immigrants account for about 31% of the U.S. foreign-born population (Hanson 2006). According to the National Agricultural Workers Survey, in 2001-2002,

immigrant labor accounted for 78% of all U.S. agricultural employment, and of these laborers about 75% were from Mexico (U.S. Department of Labor 2005). These statistics illustrate the importance of Mexican laborers in U.S. agricultural production and exports.

(11.) Gaytan-Fregoso and Lahiri (2000) extend the twocountry general equilibrium analysis of illegal immigration by examining the allocation effects of border enforcement and domestic controls.

(12.) Bandyopadhyay (2006) addresses the effects of changes in tariff policies on illegal immigration by employing a small-union Meade model.

(13.) Also see Devadoss (2008) for a model that captures the vertical link between log and lumber production and the horizontal link between Canada and the United States through lumber and log trade.

(14.) Bond and Chen 0987) draw on the seminal work of Ethier (1986a, 1986b) on illegal immigration.

(15.) We hold all other exogenous variables (c, T, SU, ZM, ZU, and [bar.L]) constant.

(16.) Djajic (1987) uses a two-country, one-good model to analyze the effect of a decline in the host country's minimum wage, an increase in the source country's labor endowment, an increase in host country's capital endowment, and an increase in the discrimination rate toward illegal immigrants on the level of illegal immigration and enforcement rate in a dynamic setting.

(17.) Totally differentiating Equations (10a) and (10b) and rewriting in the form Ax = d and solving for the determinant of A yields

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Comparing the similar terms and using plausible coefficients of demand and supply function, one can ascertain that the determinant is positive, and thus, the matrix A is non-singular.

(18.) Djajic (1997) examines the effect an exogenous increase in illegal immigrants has on commodity prices in the short- and long-run, wages of three classes of workers, and resource allocation.

20. However, this labor shortage is mitigated by the current financial crisis and layoff in the construction industry.

STEPHEN DEVADOSS and JEFF LUCKSTEAD *

* The authors gratefully acknowledge the helpful comments of the coeditor Vincenzo Quadrini and two anonymous reviewers. We also thank Victoria Seever for editorial work. This research was partially supported by Grant No. BD2056 from the Student Grant Program at the University of Idaho, and by the Edward J. and Maude R. Iddings Research Fellowship, College of Agriculture and Life Science, University of Idaho.

Devadoss: Department of Agricultural Economics and Rural Sociology, University of Idaho, Moscow, ID 838442334. Phone 1-208-885-6806, Fax 1-208-885-5759, E-mail [email protected]

Luckstead: Department of Agricultural Economics and Rural Sociology, University of Idaho, Moscow, ID 838442334. Phone 1-208-885-6806, Fax 1-208-885-5759, E-mail [email protected]

doi:10.1111/j.1465-7295.2010.00300.x

TABLE 1
Empirical Results for the United States and Mexico (a)

U.S. Labor Supply

USALS = 1,387.47 + 96.88 USRLWR - 54.52 USRWW
 (6.99) *** (3.51) *** (-3.59) ***
 [1.12] [-1.16]
 - 6.35 USTNFE
 (-2.44) ***

U.S. Labor Demand

USALD = 194.26 - 61.02 USRLWR + 2.50 USRPRS
 (0.22) (-2.16) ** (1.93) *
 [-0.41] [0.301
 + 0.43USF + 90.911399
 (1.40) (2.48) **

U.S. Real Legal and Illegal Wage Rate Identity

USRLWR = USRIWR + 0.0003 (6,072.87)

U.S. Agricultural Supply

USRVAP = 79.35 + 1.30 USNPRS - 4.86 USRLWR
 (3.98) *** (7.37) *** (-2.00) **
 [0.80] [-0.161

U.S. Agricultural Demand

USRVDAP = 200.11 - 0.49 USRFBCPI + 0.01 USRPI
 (1.35) (-0.48) (2.69) **
 [-0.43] [0.37]

U.S. Consumer and Producer Price Linkage
USRFBCPI = 139.70 + 0.25 USPR
 (20.04) ** (3.78) *

U.S. Net Exports to the Rest of the World
USNEROW = 37.75 - 0.14 USRFBCPI

Commodity Market Equilibrium
USRVAP - USRVDAP USNEROW = (ER) MRVDAP
+ MREREUS - MRIREUS - (ER)MRVAP

Mexican Labor Demand

MALD = 8,316.76 - 18.77 MRW + 0.39 MRAPPI
 (59.92) *** (-5.34) *** (0.23)
 [-0.10] [0.004]
 + 10.32 MRAPPI
 (5.96) ***
 [0.13]

Mexican Wage and U.S. Illegal Wage Linkage Identity
MRW = (1-0.61)(1-0.21) USRIWR + 0.61(1-0.21) MRW
MRW = 0.59 USRIWR

Labor Market Equilibrium

USALD - USALS = [psi] (MALS-MALD)

Mexican Agricultural Supply

MRVAP = 375.97 + 0.77 LMRVAP + 2.05 MRAPPI
 (1.69) (18.18) *** (1.80) *
 [0.40]
 - 1.97 MW - 4.87 MRAPPI
 (-1.57) (-2.97) ***
 [-0.11]

Mexican Agricultural Demand

MRVDAP = 1,227.66 - 17.87 MACPI + 0.13 MGDP
 (23.24) *** (-9.24) *** (5.38) ***
 [-2.46] [1.09]

Mexican Producer and Consumer Price Linkage

MRAPPI = 21.74 + 0.72 MRACPI
 (1.79) * (6.54) ***

U.S. and Mexican Price Linkage Identity
MRACPI = USRFBCPI (ER) (I + T) + TC

(a) Values in parenthesis are t-ratios and values in brackets are
elasticities.

*, **, *** Significant at 10%, 5%, and 1% level, respectively.

TABLE 2
Variable Definitions

Name Description (Unit)

USALS U.S. July legal farm employment (1000s)

USRLWR U.S. real legal wage rate ($)

USRMW U.S. real manufacturing wage ($)

USTNFE Total non-farm employment (1000s)

USALD U.S. July farm employment (1000s)

USRPRS Real price received by farmers including
 the PSE (index $)

USF Number of farms in the United States
 (1000s)

D99 Binary variable for the year 1999

USRIWR U.S. real illegal wage rate ($)

USRVAP U.S. real value of agricultural production
 (billion $)

USNPRS U.S. nominal price received by farmers
 including PSE (index $)

URVDAP U.S. real value of agricultural product
 demand (billion $)

USRFBCPI U.S. real food and beverage consumer price
 index (CPI) (index $)

USRPI U.S. real personal income (billion $)

USPR U.S. price received (index $)

USNEROW U.S. net exports to the rest of the world
 excluding Mexico (billion $)

ER Exchange rate between the United States
 and Mexico (NP/$) (a)

MRVDAP Mexican real value of agricultural production
 demand (billion NP)

MREREUS Mexican real exports to the rest of the world
 excluding United States (billion NP)

MRIREUS Mexican real imports from the rest of the
 world excluding United States (billion NP)

MRVAP Mexican real value of agricultural production
 (billion NP)

MALD Mexican total employment in agriculture
 (1000s)

MRW Mexican real wage rate (NP)

MRAPPI Mexican real agricultural producer price
 index (PPI) (index NP)

MRTPPI Mexican real textile PPI (index NP)

MALS Mexican agricultural labor supply (1000s)

LMRVAP Lagged Mexican real value of agricultural
 production (billion NP)

MW Mexican nominal wage rate (NP)

MRMPPI Mexican real manufacturing PPI (index NP)

MACPI Mexican agricultural CPI (index NP)

MGDP Mexican GDP (billion NP)

MRACPI Mexican real agricultural CPI (index NP)

T Mexican tariff rate (%)

TC Transportation costs (NP)

(a) New Pesos.

TABLE 3
Baseline Estimates and Domestic Enforcement's Impact on the U.S. and
Mexican Agricultural and Labor Markets

Year 1994 1995 1996

U.S. labor demand (1000s) 1,095.90 1,062.68 1,015.18
 Impact (%) -0.01 -0.01 -0.02
U.S. labor supply (1000s) 658.37 646.74 571.27
 Impact (%) 0.01 0.03 0.07
Mexican labor demand (1000s) 8,650.46 8,634.45 8,653.15
 Impact (%) 0.00 0.00 0.01
U.S. legal wage rate ($) 5.62 5.96 5.69
 Impact (%) 0.02 0.04 0.07
U.S. illegal wage rate ($) 5.51 5.84 5.56
 Impact (%) -0.18 -0.40 -0.75
Mexican wage rate (NP) 14.18 17.08 22.29
 Impact (%) -0.02 -0.04 -0.07
Illegal immigration (1000) 437.53 415.93 443.91
 Impact (%) -0.03 -0.08 -0.14
U.S. agricultural demand 161.93 168.09 178.46
 (billion $)
 Impact (%) 0.00 0.00 0.00
U.S. agricultural supply 182.35 198.43 206.57
 (billion $)
 Impact (%) 0.00 0.00 0.00
Mexican agricultural demand 216.26 232.15 235.76
 (billion NP)
 Impact (%) 0.00 0.00 0.00
Mexican agricultural supply 210.14 219.67 223.99
 (billion NP)
 Impact (%) 0.00 0.00 0.01
U.S. food CPI (index $) 146.85 147.63 151.97
 Impact (%) 0.00 0.00 0.00
Mexican agricultural CPI 30.48 39.83 56.45
 (index NP)
 Impact (%) 0.00 0.00 0.00
U.S. net exports to ROW 15.78 25.62 24.01
 (billion $)
 Impact (%) 0.00 0.00 0.00
Net exports to Mexico (billion $) 4.64 4.72 4.09
 Impact (%) -0.04 -0.08 -0.17

Year 1997 1998 1999

U.S. labor demand (1000s) 1,086.94 1,058.47 1,095.01
 Impact (%) -0.03 -0.05 -0.06
U.S. labor supply (1000s) 542.21 562.44 570.37
 Impact (%) 0.11 0.15 0.20
Mexican labor demand (1000s) 8,648.83 8,615.21 8,571.25
 Impact (%) 0.01 0.02 0.02
U.S. legal wage rate ($) 5.89 6.53 7.11
 Impact (%) 0.10 0.13 0.17
U.S. illegal wage rate ($) 5.74 6.36 6.63
 Impact (%) -1.14 -1.50 -2.03
Mexican wage rate (NP) 24.99 30.66 32.61
 Impact (%) -0.12 -0.16 -0.24
Illegal immigration (1000) 544.73 496.03 524.64
 Impact (%) -0.17 -0.27 -0.35
U.S. agricultural demand 185.63 182.73 176.37
 (billion $)
 Impact (%) 0.00 0.00 0.00
U.S. agricultural supply 206.63 201.44 188.59
 (billion $)
 Impact (%) -0.01 -0.01 -0.02
Mexican agricultural demand 240.52 244.96 254.01
 (billion NP)
 Impact (%) -0.01 -0.02 -0.03
Mexican agricultural supply 234.59 226.47 244.04
 (billion NP)
 Impact (%) 0.02 0.03 0.05
U.S. food CPI (index $) 159.05 159.37 165.17
 Impact (%) 0.00 0.00 0.00
Mexican agricultural CPI 69.29 77.56 89.65
 (index NP)
 Impact (%) 0.00 0.00 0.01
U.S. net exports to ROW 19.02 14.60 10.42
 (billion $)
 Impact (%) 0.00 0.00 -0.01
Net exports to Mexico (billion $) 1.98 4.11 1.80
 Impact (%) -0.61 -0.44 -1.50

Year 2000 2001 2002

U.S. labor demand (1000s) 1,036.01 1,019.41 975.14
 Impact (%) -0.09 -0.14 -0.22
U.S. labor supply (1000s) 514.67 480.17 474.07
 Impact (%) 0.30 0.51 0.75
Mexican labor demand (1000s) 8,630.80 8,612.62 8,623.00
 Impact (%) 0.03 0.04 0.07
U.S. legal wage rate ($) 7.64 7.17 7.43
 Impact (%) 0.21 0.36 0.52
U.S. illegal wage rate ($) 6.93 6.89 7.06
 Impact (%) -2.69 -4.34 -6.69
Mexican wage rate (NP) 35.95 38.12 40.17
 Impact (%) -0.33 -0.51 -0.75
Illegal immigration (1000) 521.34 539.24 501.07
 Impact (%) -0.48 -0.72 -1.13
U.S. agricultural demand 184.15 193.90 189.77
 (billion $)
 Impact (%) 0.00 0.00 0.00
U.S. agricultural supply 195.94 208.98 203.20
 (billion $)
 Impact (%) -0.02 -0.03 -0.05
Mexican agricultural demand 250.53 263.01 267.78
 (billion NP)
 Impact (%) -0.05 -0.07 -0.11
Mexican agricultural supply 249.66 257.75 255.31
 (billion NP)
 Impact (%) 0.09 0.14 0.24
U.S. food CPI (index $) 169.29 174.65 178.25
 Impact (%) 0.00 0.01 0.01
Mexican agricultural CPI 92.96 97.16 101.88
 (index NP)
 Impact (%) 0.01 0.01 0.02
U.S. net exports to ROW 10.58 12.70 10.76
 (billion $)
 Impact (%) -0.01 -0.01 -0.02
Net exports to Mexico (billion $) 1.21 2.38 2.67
 Impact (%) -3.31 -2.77 -3.78

Year 2003 2004 2005

U.S. labor demand (1000s) 967.90 966.38 942.79
 Impact (%) -0.33 -0.49 -0.75
U.S. labor supply (1000s) 467.69 494.81 466.28
 Impact (%) 1.15 1.62 2.53
Mexican labor demand (1000s) 8,586.92 8,576.75 8,537.29
 Impact (%) 0.10 0.15 0.22
U.S. legal wage rate ($) 7.75 8.05 8.27
 Impact (%) 0.76 1.12 1.66
U.S. illegal wage rate ($) 7.25 7.39 7.37
 Impact (%) -10.15 -15.51 -24.10
Mexican wage rate (NP) 42.07 43.70 45.33
 Impact (%) -1.17 -1.81 -2.72
Illegal immigration (1000) 500.21 471.57 476.51
 Impact (%) -1.71 -2.70 -3.95
U.S. agricultural demand 202.92 222.42 232.95
 (billion $)
 Impact (%) -0.01 -0.01 -0.01
U.S. agricultural supply 219.40 235.96 237.05
 (billion $)
 Impact (%) -0.08 -0.11 -0.17
Mexican agricultural demand 276.82 285.17 277.65
 (billion NP)
 Impact (%) -0.21 -0.35 -0.52
Mexican agricultural supply 256.39 245.92 276.57
 (billion NP)
 Impact (%) 0.43 0.77 1.12
U.S. food CPI (index $) 180.41 185.68 189.29
 Impact (%) 0.01 0.02 0.03
Mexican agricultural CPI 106.76 114.07 119.36
 (index NP)
 Impact (%) 0.03 0.04 0.06
U.S. net exports to ROW 13.35 9.34 3.20
 (billion $)
 Impact (%) -0.02 -0.05 -0.21
Net exports to Mexico (billion $) 3.13 4.21 0.90
 Impact (%) -4.93 -5.72 -42.35

Year 2006 2007

U.S. labor demand (1000s) 884.78 841.28
 Impact (%) -1.17 -1.80
U.S. labor supply (1000s) 410.14 399.45
 Impact (%) 4.21 6.30
Mexican labor demand (1000s) 8,470.13 8,518.32
 Impact (%) 0.33 0.47
U.S. legal wage rate ($) 8.79 9.17
 Impact (%) 2.37 3.40
U.S. illegal wage rate ($) 7.58 7.56
 Impact (%) -36.11 -55.23
Mexican wage rate (NP) 47.31 48.95
 Impact (%) -3.99 -5.81
Illegal immigration (1000) 474.64 441.83
 Impact (%) -5.82 -9.12
U.S. agricultural demand 234.30 265.73
 (billion $)
 Impact (%) -0.02 -0.02
U.S. agricultural supply 242.19 287.35
 (billion $)
 Impact (%) -0.26 -0.33
Mexican agricultural demand 274.03 285.39
 (billion NP)
 Impact (%) -0.78 -1.11
Mexican agricultural supply 264.58 271.91
 (billion NP)
 Impact (%) 1.86 2.80
U.S. food CPI (index $) 196.28 201.31
 Impact (%) 0.04 0.05
Mexican agricultural CPI 126.65 130.55
 (index NP)
 Impact (%) 0.07 0.10
U.S. net exports to ROW 5.51 17.68
 (billion $)
 Impact (%) -0.18 -0.08
Net exports to Mexico (billion $) 2.37 3.94
 Impact (%) -24.89 -22.67

TABLE 4
Baseline Estimates and Border Enforcement's Impact on the U.S. and
Mexican Agricultural and Labor Markets

Year 1994 1995 1996

U.S. labor demand (1000s) 1,095.90 1,062.68 1,015.18
 Impact (%) -0.43 -0.44 -0.42
U.S. labor supply (1000s) 658.37 646.74 571.27
 Impact (%) 1.25 1.28 1.30
Mexican labor demand (1000s) 8,650.46 8,634.45 8,653.15
 Impact (%) 0.15 0.15 0.14
U.S. legal wage rate ($) 5.62 5.96 5.69
 Impact (%) 1.37 1.32 1.27
U.S. illegal wage rate ($) 5.51 5.84 5.56
 Impact (%) 1.39 1.35 1.30
Mexican wage rate (NP) 14.18 17.08 22.29
 Impact (%) -1.36 -1.56 -1.41
Illegal immigration (1000) 437.53 415.93 443.91
 Impact (%) -2.96 -3.12 -2.64
U.S. agricultural demand 161.93 168.09 178.46
 (billion $)
 Impact (%) -0.01 -0.01 -0.01
U.S. agricultural supply 182.35 198.43 206.57
 (billion $)
 Impact (%) -0.09 -0.08 -0.08
Mexican agricultural demand 216.26 232.15 235.76
 (billion NP)
 Impact (%) -0.02 -0.05 -0.09
Mexican agricultural supply 210.14 219.67 223.99
 (billion NP)
 Impact (%) 0.06 0.10 0.15
U.S. food CPI (index $) 146.85 147.63 151.97
 Impact (%) 0.03 0.03 0.02
Mexican agricultural CPI 30.48 39.83 56.45
 (index NP)
 Impact (%) 0.02 0.04 0.04
U.S. net exports to ROW 15.78 25.62 24.01
 (billion $)
 Impact (%) -0.03 -0.02 -0.02
Net exports to Mexico (billion $) 4.64 4.72 4.09
 Impact (%) -3.14 -2.88 -3.34

Year 1997 1998 1999

U.S. labor demand (1000s) 1,086.94 1,058.47 1,095.01
 Impact (%) -0.39 -0.42 -0.40
U.S. labor supply (1000s) 542.21 562.44 570.37
 Impact (%) 1.34 1.36 1.31
Mexican labor demand (1000s) 8,648.83 8,615.21 8,571.25
 Impact (%) 0.13 0.14 0.14
U.S. legal wage rate ($) 5.89 6.53 7.11
 Impact (%) 1.21 1.16 1.06
U.S. illegal wage rate ($) 5.74 6.36 6.63
 Impact (%) 1.25 1.19 1.14
Mexican wage rate (NP) 24.99 30.66 32.61
 Impact (%) -1.45 -1.43 -1.52
Illegal immigration (1000) 544.73 496.03 524.64
 Impact (%) -2.10 -2.43 -2.26
U.S. agricultural demand 185.63 182.73 176.37
 (billion $)
 Impact (%) -0.01 -0.01 -0.01
U.S. agricultural supply 206.63 201.44 188.59
 (billion $)
 Impact (%) -0.08 -0.09 -0.10
Mexican agricultural demand 240.52 244.96 254.01
 (billion NP)
 Impact (%) -0.11 -0.16 -0.20
Mexican agricultural supply 234.59 226.47 244.04
 (billion NP)
 Impact (%) 0.20 0.28 0.34
U.S. food CPI (index $) 159.05 159.37 165.17
 Impact (%) 0.02 0.02 0.02
Mexican agricultural CPI 69.29 77.56 89.65
 (index NP)
 Impact (%) 0.04 0.04 0.04
U.S. net exports to ROW 19.02 14.60 10.42
 (billion $)
 Impact (%) -0.02 -0.03 -0.04
Net exports to Mexico (billion $) 1.98 4.11 1.80
 Impact (%) -7.51 -3.98 -9.60

Year 2000 2001 2002

U.S. labor demand (1000s) 1,036.01 1,019.41 975.14
 Impact (%) -0.41 -0.39 -0.23
U.S. labor supply (1000s) 514.67 480.17 474.07
 Impact (%) 1.41 1.39 0.80
Mexican labor demand (1000s) 8,630.80 8,612.62 8,623.00
 Impact (%) 0.13 0.12 0.07
U.S. legal wage rate ($) 7.64 7.17 7.43
 Impact (%) 0.98 0.99 0.55
U.S. illegal wage rate ($) 6.93 6.89 7.06
 Impact (%) 1.08 1.03 0.58
Mexican wage rate (NP) 35.95 38.12 40.17
 Impact (%) -1.51 -1.39 -0.80
Illegal immigration (1000) 521.34 539.24 501.07
 Impact (%) -2.21 -1.97 -1.21
U.S. agricultural demand 184.15 193.90 189.77
 (billion $)
 Impact (%) -0.01 -0.01 0.00
U.S. agricultural supply 195.94 208.98 203.20
 (billion $)
 Impact (%) -0.10 -0.10 -0.06
Mexican agricultural demand 250.53 263.01 267.78
 (billion NP)
 Impact (%) -0.22 -0.20 -0.12
Mexican agricultural supply 249.66 257.75 255.31
 (billion NP)
 Impact (%) 0.40 0.40 0.26
U.S. food CPI (index $) 169.29 174.65 178.25
 Impact (%) 0.02 0.02 0.01
Mexican agricultural CPI 92.96 97.16 101.88
 (index NP)
 Impact (%) 0.04 0.03 0.02
U.S. net exports to ROW 10.58 12.70 10.76
 (billion $)
 Impact (%) -0.04 -0.03 -0.02
Net exports to Mexico (billion $) 1.21 2.38 2.67
 Impact (%) -15.35 -7.62 -4.05

Year 2003 2004 2005

U.S. labor demand (1000s) 967.90 966.38 942.79
 Impact (%) -0.19 -0.15 -0.10
U.S. labor supply (1000s) 467.69 494.81 466.28
 Impact (%) 0.67 0.49 0.34
Mexican labor demand (1000s) 8,586.92 8,576.75 8,537.29
 Impact (%) 0.06 0.04 0.03
U.S. legal wage rate ($) 7.75 8.05 8.27
 Impact (%) 0.45 0.34 0.22
U.S. illegal wage rate ($) 7.25 7.39 7.37
 Impact (%) 0.48 0.37 0.25
Mexican wage rate (NP) 42.07 43.70 45.33
 Impact (%) -0.69 -0.54 -0.37
Illegal immigration (1000) 500.21 471.57 476.51
 Impact (%) -1.00 -0.81 -0.53
U.S. agricultural demand 202.92 222.42 232.95
 (billion $)
 Impact (%) 0.00 0.00 0.00
U.S. agricultural supply 219.40 235.96 237.05
 (billion $)
 Impact (%) -0.05 -0.03 -0.02
Mexican agricultural demand 276.82 285.17 277.65
 (billion NP)
 Impact (%) -0.12 -0.11 -0.07
Mexican agricultural supply 256.39 245.92 276.57
 (billion NP)
 Impact (%) 0.25 0.23 0.15
U.S. food CPI (index $) 180.41 185.68 189.29
 Impact (%) 0.01 0.01 0.00
Mexican agricultural CPI 106.76 114.07 119.36
 (index NP)
 Impact (%) 0.02 0.01 0.01
U.S. net exports to ROW 13.35 9.34 3.20
 (billion $)
 Impact (%) -0.01 -0.01 -0.03
Net exports to Mexico (billion $) 3.13 4.21 0.90
 Impact (%) -2.90 -1.72 -5.72

Year 2006 2007

U.S. labor demand (1000s) 884.78 841.28
 Impact (%) -0.06 0.00
U.S. labor supply (1000s) 410.14 399.45
 Impact (%) 0.20 0.00
Mexican labor demand (1000s) 8,470.13 8,518.32
 Impact (%) 0.02 0.00
U.S. legal wage rate ($) 8.79 9.17
 Impact (%) 0.11 0.00
U.S. illegal wage rate ($) 7.58 7.56
 Impact (%) 0.13 0.00
Mexican wage rate (NP) 47.31 48.95
 Impact (%) -0.19 0.00
Illegal immigration (1000) 474.64 441.83
 Impact (%) -0.27 0.00
U.S. agricultural demand 234.30 265.73
 (billion $)
 Impact (%) 0.00 0.00
U.S. agricultural supply 242.19 287.35
 (billion $)
 Impact (%) -0.01 0.00
Mexican agricultural demand 274.03 285.39
 (billion NP)
 Impact (%) -0.04 0.00
Mexican agricultural supply 264.58 271.91
 (billion NP)
 Impact (%) 0.09 0.00
U.S. food CPI (index $) 196.28 201.31
 Impact (%) 0.00 0.00
Mexican agricultural CPI 126.65 130.55
 (index NP)
 Impact (%) 0.00 0.00
U.S. net exports to ROW 5.51 17.68
 (billion $)
 Impact (%) -0.01 0.00
Net exports to Mexico (billion $) 2.37 3.94
 Impact (%) -1.17 0.00