Thread: Revisiting NAFTA

Due to the legal characteristics of the maquiladora industry, its activity does not depend on the trade opening resulting from NAFTA, as the sector has its own rules. So it has been argued that the increment in maquiladora activity is due more to the devaluation subsequent to 1994 than to NAFTA itself (Gruben 2001).

The maquiladora industry primarily produces metal and equipment products, electronics and textiles, as well as steel, paper and printing, clothing, and plastic products. For example, in 2002, of the 47.9% of total industrial exports generated by the maquiladora sector, metal and machinery products account for 39.8 percentage points of the total and textiles and garments represent 4.3 percentage points. The rest (approximately 3.8% of total exports) is shared by the remaining industries.

Agriculture and mining have a reduced presence in trade (currently, they do not account for more than 20% of non-maquiladora exports, whereas in 1991, they accounted for 35% of this category). In contrast, the proportion of manufactured goods in the total of non-maquiladora exports grew to reach 78.8% in 2002. These exports were principally metal products followed by textiles and garments, which represented, cumulatively, 66% and 68% of the exports of non-maquiladora manufactured goods. Outside of metal products, textiles and garments, and the food and beverage industry, the percentage of non-maquiladora exports of other industries—chemical, petrochemical, metallurgic products and steel production—shrank as a share of total manufactured exports.

The manufactured goods sector has grown, but the basic problem is that the specific type of productive specialization occurring in Mexico is product assembly based on imported inputs with little to no link to the rest of the nation's productive apparatus (Aroche 2001). This process does not ensure sustained industrial development in the framework of markets with high value-added products.

In fact, the location of export manufacturing zones is not determined by competitive factors such as training and knowledge, but rather by low wages. As Palley (2004) shows, there is a race to the bottom related to labor norms. Foreign companies are more interested in locating themselves so as to benefit from the national content clauses of NAFTA, always when labor or regulatory costs do not surpass the advantages of being able to sell to the U.S. market.

Despite apparently counting on the advantage of NAFTA to stimulate exports to the United States, between 2000 and 2003, the evolution of the export sector was very weak. This contrasts with the performance of Chinese manufactured goods, which increased rapidly after China joined the World Trade Organization (WTO) in 2001. This evolution is shown in Figure 2-H, together with the Mexican exports to the United States. The difference in export promotion policies is very evident in the results of these last years in the case of China, while in Mexico the weak evolution of exports is attributed to the slow down of the US economy. In 1987, Mexico's share of U.S. exports was more than triple that of China (1.6% versus 5%). By 2004, China's exports to the U.S. were 26% larger than Mexico's.

Figure 2-H

The evolution of foreign direct investment
After 1994, foreign direct investment (FDI)—a significant portion of which has been directed towards the purchase of existing assets—accounted for most of Mexico's net financial inflows (Blecker 2003).

Throughout the period of time that NAFTA has been in force, FDI flows have been relatively stable, lacking large, episodic swings. In fact, the majority of foreign investment has entered Mexico as foreign direct investment and not into money market or stock market funds.

The majority of FDI is composed of "new investments" (Figure 2-I), funds that have been used mostly for the purchase of existing companies (as is shown by the enormous flow in 2001, much of which was derived from the purchase of BANAMEX by Citigroup).

Figure 2-I

These "new investments" have followed an irregular pattern. In contrast, the investments in maquiladora and the flows of accounts between firms have grown in a sustained manner. The problem with both types of flows is that they correspond to account balances between firms that do not translate into real technology transfer. Additionally, the flow of FDI toward industrial activities has diminished since 1980 and has been directed increasingly toward services. In 1980, 80% of FDI went toward manufacturing, while in 2004 this percentage had fallen to 52%.

Therefore, the general growth driven by exports appears to be more a mirage than a reality. On the one hand, the only benefits resulting from maquiladora activity are the direct wages and salaries that it pays because it uses relatively few inputs from other Mexican firms or industries. On the other hand, the flow of FDI toward services rarely results in technology transfer. As has already been shown, FDI translates into the acquisition of existing firms as part of foreign firms' consolidation or their introduction into the Mexican market (Mattar et al. 2003).

The evolution of employment, earnings, and the distribution of income
One of the elements used most often to affirm the export-led growth model, and NAFTA in particular, is Mexico's low unemployment rate, in both absolute and relative terms. However, the following question always hangs in the air: Why is the country's unemployment rate so low? To respond to this question, we began by analyzing the characteristics of those who are currently unemployed. The majority of Mexico's unemployed are young people (over 50% of the unemployed are under 25 years of age), with slightly higher academic preparation than the national average (over 50% have at least some college studies). Most are not heads of households (80%). While the unemployment rate has grown throughout the 2000-04 period, it has not achieved the record levels observed following the 1995-96 crisis.

Nevertheless, Table 2-1 reveals a disturbing fact. Between the second quarter of 2000 and the second quarter of 2003, the total number of unemployed increased 50% and the average period of time unemployed also increased.4 The data also show that both layoffs and the termination of temporary work positions are increasing.

Table 2-1

The average duration of unemployment was fewer than five weeks in 2000, which demonstrates the frictional nature of open unemployment in Mexico. It has been shown that the majority of those who gain employment do so via the micro-business sector, meaning economic entities with five or fewer workers, including one person operations (Salas 2003). (This theme of micro-businesses will be addressed in the sub-section, Open Employment, on p. 39.)

Job creation and job loss
Beginning with the agricultural sector, agricultural employment in Mexico increased slightly at the end of the 1980s, achieving employment for 8.1 million Mexicans at the end of 1993, barely before NAFTA entered into force. Thereafter, employment in the sector began a constant reduction, falling to 6.8 million employed workers by the end of 2004. In fact, the population dedicated to agricultural activities fell from 26.8% in 1991 to 16.4% in 2004, a significant decrease. The principal affected parties are corn producers, with a total loss of 1.013 million jobs (Table 2-2). Additionally, 142,000 jobs were lost in the cultivation of flowers and fruits, which have been the primary products of agricultural exports (USDA 2003). This job loss leads Polaski to declare, "Therefore, the liberalization of agricultural trade linked to NAFTA is the most important factor in the loss of agricultural employment in Mexico" (Polaski 2003, 20).

Table 2-2

Considering disaggregated data from 30 economic sub-sectors, one aspect that stands out is that, while the largest number of the (economically) active population at the beginning of the 1990s was in agriculture, by the beginning of the 21st century, the largest sector was retail trade (16.2% in 2003). This process is framed by a light recovery of the manufacturing sector (between 1991 and 2003, it grew from 15.7% to 17.3%) and accelerated growth of manual labor in the services sector (from 33.6% a 39.1%).

In the least urbanized zones (those with fewer than 100,000 residents), the percentage of the population active in the agriculture sector during the 2000-03 period fluctuated around 28%, but at the beginning of the 1990s that figure was greater than 44%. The largest drop in the sector is in male workers, which fell from 53.4% to 36.3% of the employed population, but the decrease of females was also appreciable (from 20.5% to 9.1%).

Next we examine the population engaged in non-agricultural work with a detailed focus on their occupations, considering the varying outcomes for employers, wage-earning workers, self-employed workers, and workers receiving no remuneration.5 The proportion of wage-earning workers in the total share of workers active in this sector fell from 74% in 1991 to a minimum of 67% in 1998, to later recover slowly to 68% in 2004. The positions for wage-earning workers represented 65% of the new jobs created between 1991 and 1998 in the most urbanized areas, while this category represented 64% of the positions created between 1998 and 2004. Wage-earning work is not accessible to all people. As people age, they are resigned from duty (they are encouraged to resign voluntarily, but sometimes they are laid off) in such a way that the proportion of wage-earning workers falls as age increases, i.e., there are fewer wage-earning workers in older age groups.

Among young people, the proportion of wage-earning women by age group is greater than that of men.

Self-employed workers represent another important group of those working in the non-agricultural sector. The self-employed share oscillates around 24%, while the rest of the population is split evenly between employers and workers without remuneration, each group accounting for 5% of the total.

Between the second quarter of 2000 and the second quarter of 2004 2,788,851 jobs were created, of which 54% were wage-earning jobs, 4% were employers, and 43% were jobs created through self-employment. Next we examine the characteristics of the wage-earning positions that were created during the period in question.

To begin with, 23% of the new wage-earning positions generated between the second quarter of 2000 and the second quarter of 2004 have no social benefits, while only 37% of the new jobs have full social security benefits. These data suggest that the process of making employment more precarious may have been accentuated.6 Further, in the second quarter of 2004, 43% of the total of wage-earning workers labored under a verbal contract, of which 86% received no social benefits. Of the wage-earning workers laboring under permanent contracts, 3% do not receive social benefits. Thus, lack of social protection is quite extensive in Mexico.

Upon investigating wage distribution patterns (where positions were created according to the size of the economic entity), another facet of precarious employment emerges: 65% of all new jobs were created in micro-businesses (economic entities with up to five employees), and 52% of new wage-earning jobs were found in such entities, which are characterized by low wages, low productivity, and a low level of technology.

In summary, the creation of jobs between 2000 and 2004 was relatively dynamic, given that, on average, approximately 700,000 job positions were created annually. Nevertheless, this rate is inferior to that of the decade of the 1990s when approximately 1 million new positions were created each year. Furthermore, as shown above, a significant share of these new positions were precarious jobs.

Maquiladoras

Now the discussion turns to the major components of the non-agricultural economy. Between 1980 and 1993, the manufacturing sector as a whole grew by fewer than 100,000 jobs, of which 40,000 were in maquiladora activities. Between 1991 and 2000, manufacturing grew by 2.7 million jobs, a significant number of which—800,000 jobs—resulted from maquiladora activities. But as some have pointed out (Polaski 2003; Gruben 2001), the maquiladora industry grew due to trade and not due to NAFTA. In fact, as Polaski (2003) shows, while it is not possible to know precisely how many jobs were created by the non-maquiladora export industry, it can be estimated that between 1994 and 1999, this sector grew by 500,000 jobs. Starting with the stagnation of 2000, total manufacturing employment began to decline, especially in the maquiladora sector. In fact, although manufacturing employment recovered slightly in 2004, there were still 180,000 fewer jobs in this sector than there were in the peak year of 2000 as shown in Figure 2-J.

Figure 2-J

An important series of questions arises here concerning the type of employment created in manufacturing in general and in the maquiladora sector in particular. Wages in the maquiladora sector are almost 40% lower than those paid in heavy non-maquila manufacturing (Salas and Zepeda 2003a). In fact, a recent study by Bendesky et al. (2004) shows that productivity in the maquiladora sector is stagnant, and its average technological base is weak. From this it can be inferred that the maquiladora sector is stuck in a trap of low productivity growth, reduced skills, and sustained by low wages. In fact, Figure 2-J shows that the number of maquiladora companies has diminished since 2000, which is the result of various companies leaving the country to go to other countries with wages even lower than those in Mexico.

The options for the majority of the working-age population are concentrated in service activities. In fact, as was shown earlier, the share of unemployed people who find employment within one month or less is 59% and a majority of those who find employment do so in very small scale activities. These activities are found in the trade and services sectors, which account for 70% of the non-agricultural work force. Sixty-seven percent of trade-based operations and 47% of service entities employ five workers or fewer. The working conditions, income, and productivity in these operations are very precarious, and yet they represent an earning opportunity for large groups of the population.

Open employment

Now we are able to respond to the question posed earlier, related to the reduced rate of open unemployment.

The mechanism is the following: because the labor force is growing much faster than employment in larger companies, self-employment or wage-earning employment in micro-businesses provides the only job opportunity for an important number of workers. Faced with the alternative of not finding any job, people take jobs in the micro-business sector where they generally are paid a low wages.

In this way, the micro-business sector acts as a full-employment buffer, absorbing and retaining a large share of workers as GDP growth slows and accelerates, as seen in Figure 2-K, which compares the rate of growth of GDP with the proportion of people engaged in very small scale activities. The share of workers in this sector has trended up over time, rising from 40% in 1990 to 45% in 2005, at similar stages of the business cycle. Furthermore, the share of micro-employment is counter-cyclical, rising during recessions and falling during periods of recovery, thus confirming the buffer role of micro-business activity.

Figure 2-K



Migration

Another element that explains the low unemployment rate is illegal migration to the United States. Between 1990-94, the average annual flow of illegal migrants has been estimated to have been 260,000 people (Passel 2005). After 1994, the rate of immigration increased significantly: between 2000-04, illegal migration is estimated to have totaled approximately 485,000 persons per year (Passel 2005). In this way, migration serves as an escape valve that reduces the demand for new jobs.

Earnings from work
In the case of agriculture, wage-earning women worked fewer hours per week than men (29 and 41 hours, respectively) in 2003, but they received better real hourly wages (3.4 pesos compared to the 2.7 pesos paid to men). The difference reflects the fact that rural wage-earning female workers are generally employed by larger productive entities (with 16 or more workers). In contrast, women landowners (of whatever size plot of land) work longer days than men yet earn less—female landowners work 55 hours a week while male landowners only work 35 hours. The value of this work for women is the equivalent of 2.9 pesos per hour while for men, the value equivalent is 7.8 pesos per hour.

The uneven evolution of wages and earnings in rural areas has favored landowners. Between 1991 and 2003, remuneration paid to day laborers in the agricultural sector fell significantly from 535 to 483 pesos per month (unpublished tables from the Agricultural Module of the Encuesta Nacional de Empleo, Instituto Nacional de EstadÃ*stica GeografÃ*a e Informática (INEGI)); earnings by self-employed field workers collapsed from 1,959 pesos in 1991 to 228 pesos in 2003, an 88% decline. In the same period, landowners increased their earnings from 626 to 1,625 pesos.7

Table 2-3 shows the global evolution of earnings from work between 2000-04. Earnings from work is another element that has received considerable attention, given that it is widely claimed that such wages have increased significantly. As can be seen, only wages for mobile/street vendors increased significantly, at 2.8 percent per year over six years. However, these levels do not even manage to recover the cumulative losses dating from 1990, as shown in Table 2-3 (Salas y Zepeda, 2003a, 6. Small wage gains in the maquiladora sector were more than offset by losses of 1.8% per year for employees in 205 heavy manufacturing industries, which were more than twice as large as wages in the maquila industries.

Table 2-3

As shown in Figure 2-K and Table 2-3, not even the relative stability of prices, which characterized the country beginning in 1996, has lent itself to the recovery of purchasing power of earnings from work.

Note that Table 2-3 only reports average earnings, but says nothing about the dispersion of wages within each sector. The benefits of income growth are not uniformly distributed across the population; other research has shown that income dispersion in general and wage dispersion in particular is relatively large (Salas and Zepeda 2003a, 73).

Two additional problems with the information presented in Table 2-3 are that the coverage of each group within the series varies over time, and they do not provide information on changes in average compensation levels over time. Figure 2-L was constructed using the same set of 16 cities between 1994 and 2004, so comparison problems do not arise. It shows the weak performance of the real income growth process. From the last quarter of 1999 to the corresponding quarter of 2004, the total income increased only 7%. Furthermore, average household labor income in 2004 (over the four quarters) was 15% lower than incomes in 1994.

Figure 2-L

Income distribution
This section begins with the manner in which income is distributed in rural areas, where, in response to lowered earnings, government programs were put into place to offset these earning losses. Between 1992 and 2000, the proportion of monetary transfers in the income of rural zones increased from 10% to 18%. During this same period, the percentage of rural homes that received transfers swelled from 25% to 60% (INEGI, Encuesta Nacional de Ingresos y Gastos de los Hogares, several years). By 2002, transfers had increased to 19.4% of total income, and the percentage of dependent homes rose to almost 70%.

Such transfers were most often focused on the poorest peasants. For the poorest 10% of rural households, the situation is as follows: in 1992, 25% of the poorest 10% of households depended on these transfers to obtain 15% of their total income. By 2000, 65% of these households used this method to acquire 37% of their income. This situation worsened in 2002, when 74% of the poorest peasants obtained 38% of their income from this source.

Rather than designing support programs aimed at generating employment and raising productivity, the government is satisfied to transfer resources, in addition to the remittances that Mexican workers in the United States send to Mexico, which total as high as $15 billion (Banco de México 2005).

Income distribution improved between 2000 and 2002, above all for families in the 20% poorest (lowest quintile) of the population (Figure 2-M), the lowest four quintiles all gained income shares at the expense of the top in 2002. Nevertheless, inequality is lower now than it was at any time since in 1984. The improvement for the middle quintile groups can be explained by a diminished earnings gap between owners and wage-earning workers (Figure 2-N) and a modest increase in wages since 2000. However, the promise of greatly improved living conditions for the majority remains largely unfulfilled.

Figure 2-M

Figure 2-N

Conclusion
The first section showed how the export-oriented model with reduced state participation in directing the economy and unrestricted support for an unregulated market economy led to a period of unstable growth. NAFTA, which is only the most recent expression of this model, bound the country to a model proven to be inefficient in fulfilling a promise essential to every successful development model: an improvement in the living conditions of the majority. Expressed in another way, the current model is exclusionary and is inefficient even in achieving its own objectives. The trade balance continues in deficit, and production levels depend on increasing imports over time. Foreign investment has grown, but mostly in the purchase of existing assets, which neither creates improved conditions in the productive stock nor achieves greater integration of manufacturing into the national economy.

As such, job creation has been left to fate; there is no employment policy other than that of low wages. Additionally, one-sixth of the population that worked in agricultural activities in the beginning of the 1990s has been displaced from the field, literally. This population migrates searching for any place to work, be it in other states of the republic or outside of Mexico.

With respect to generating non-agricultural employment, most recent growth has been concentrated in jobs without social benefits, in small-scale and low-productivity activities. We are witnessing a systematic process of destabilization of labor markets, which will be exacerbated if the labor reform proposed by the party in power is approved. Additionally, the evidence presented indicates the need to consider an integrated U.S.-Mexico labor market, not only due to the presence of Mexican workers in the United States, but also to the impact that low-wage policies have in Mexico on the working conditions in the neighboring country. In other words, when the relationship between the two countries is examined, the analysis must include both employees and employers of Mexico as well as the United States. Neither the workers nor the nations can be mutually exclusive.

Mexico's experience should serve as a warning concerning the dangers of any trade agreement, bilateral or multilateral, which is similar to NAFTA. As the poet John Donne wrote, "Therefore, never send to know for whom the bell tolls, it tolls for thee."

Endnotes
1. Refer to the article by Boltvinik y Hernández Laos (1981) for a discussion of the exhaustion of the domestic market based development model.

2. For a long time, capital goods never accounted for more than 8% of total exports. Beginning in 1997, this percentage began to grow, especially the share of those capital goods produced by the maquiladora industry. Nevertheless, capital goods continue to account for a low percentage of total exports. (Source: Bank of Mexico, Balance of Payments at http://www.banxico.org.mx/eInfoFinancie ... ciera.html.

3. Maquiladora activities flourished via the use of the Code of Customs Tariffs in the United States (rule HTS 9802), through which the companies of that country may send domestic manufactured inputs abroad and then import finished and semi-finished products back into the United States by paying a customs tariff based only on the value added in the foreign country.

4. The share unemployed for five to eight weeks fell by 2 percentage points, while the share unemployed for nine weeks or longer increased by the same amount, thus increasing the average duration of unemployment

5. In Mexican labor statistics, hourly workers are known as "trabajos a salariados," or salaried workers, to distinguish them from self-employed and informal-sector workers. We refer to them in this report as "wage earning."

6. Precarious employment is defined as a worker not under the protection of labor laws (even if he's entitled to the protection), has no permanent contract, has low wages, and in general, works under bad labor conditions (Rodgers 1989).

7. This situation may in part result from problems comparing data from National Employment Surveys conducted between 1991 and 2003, yet even taking this into account does not eliminate the evidence of a large benefit for rural land owners who employ wage-earning workers.

References
Altimir, Oscar .1983. "La distribución del ingreso en México, 1950-1977." Distribución del ingreso en México. Ensayos, Serie de Análisis Estructural, Banco de México, Cuaderno 2, tomo 1, México, 1983.

Aroche, Fidel. 2002. "Structural transformations and important coefficients in the North American economies." Economic Systems Research. Vol. 14, p. 257-73.

Aroche, Fidel. 2001. "Vertical Integration and Comparative Advantages," in Martin Puchet and Lionello Punzo, eds., Mexico Beyond NAFTA. London: Routledge.

Aspe, Pedro. 1993. El camino mexicano de la transformación económica. México: Fondo de Cultura Económica.

Banco de México. 2005. Informe Anual 2005. México.

Bendesky, León, Enrique de la Garza, Javier Melgoza and Carlos Salas. 2004. "La industria maquiladora en México: mitos, realidades y crisis" Estudios Sociológicos.Vol.22, No. 65, May-August.

Blecker, R.A. 1996. "NAFTA, the peso crisis, and the contradictions of the Mexican economic growth strategy." Center for Economic Policy Analysis, Working Paper 1996-2004. New York, N.Y.: New School University. http://www.newschool.edu/cepa/papers/index.htm.

Blecker, R.A. 2003. "The North American economies after NAFTA: A critical appraisal." International Journal of Political Economy. Vol. 33, No. 3 (Fall 2003 issue, published 2005), pp. 5-27.

Boltvinik, Julio and Enrique Hernández Laos. 1981. "Origen de la crisis industrial: El agotamiento del modelo de sustitución de importaciones. Un análisis preliminar" in Cordera, Rolando (comp.) Desarrollo y crisis de la economÃ*a mexicana. Colección Lecturas del Trimestre Económico, No. 39, Fondo de Cultura Económica, México.

Calva, José Luis. 2000. México más allá del neoliberalismo. Plaza y Janés, México.

Comisión Económica para América Latina y el Caribe. 2000 La Inversión Extranjera Directa en América Latina y el Caribe. Informe 1999. Santiago,Chile: CEPAL.

Chávez, Marcos. 2002. "El fracaso de las polÃ*ticas de estabilización en México. Retos y opciones de polÃ*tica económica." in José Luis Calva, ed, PolÃ*tica económica para el desarrollo sostenido con equidad. Casa Juan Pablos, Instituto de Investigaciones Económicas, Universidad Nacional Autónoma de México.

Dussel Peters, Enrique. 1997. La economÃ*a de la polarización. Universidad Nacional Autónoma de México.

Gruben, C. William. 2001. "Was NAFTA behind Mexico's high maquiladora growth?" Economic & Financial Review. Third quarter. pp. 11-21.

Hernández Laos, E. 1999. "Evolución de la distribución del ingreso de los hogares (1963-1989)," in Boltvinik, J. and E. Hernández Laos, Pobreza y distribución del ingreso en México. Siglo XXI, México. pp. 154-90.

Mariña Flores, Abelardo and Fred Moseley. 2001. "La tasa general de ganancia y sus determinantes en México: 1950-1999." EconomÃ*a, TeorÃ*a y Práctica. Nueva Época, No. 15.

Insituto Nacional de EstadÃ*stica, GeografÃ*a e Informática. Several years. Encuesta nacional de empleo. Mexico: INEGI.

Insituto Nacional de EstadÃ*stica, GeografÃ*a e Informática. Several years. Encuesta Nacional de Ingresos y Gastos de los Hogares. Mexico: INEGI.

International Monetary Fund. 2005 World Economic Outlook, 2005. Washington, D.C: IMF.

Máttar, J., J. C. Moreno-Brid, and W. Peres. 2003. "Foreign Investment in Mexico after Economic Reform." In Confronting Development: Assessing Mexico's Economic and Social Policy Challenges, ed. K. J. Middlebrook and E. Zepeda. Stanford, Calif: Stanford University Press.

Palley, Thomas. 2004. "The economic case for international labour standards." Cambridge Journal of Economics. Vol. 26, No. 1. p. 21-36. January.

Passel, Jeffrey. 2005. Estimates of the Size and Characteristics of the Undocumented Population. Pew Hispanic Center. March 21.

Polaski, S. 2003. "Jobs, Wages, and Household Income." In NAFTA's Promise and Reality: Lessons From Mexico for the Hemisphere, ed. J. J. Audley et al. Washington, D.C.: Carnegie Endowment for International Peace.

Rodgers, Gerry. 1989. "Precarious employment in Western Europe: The state of the debate." In Precarious Jobs in Labour Market Regulation: The growth of atypical employment in Western Europe. Geneva: International Institute for Labour Studies.

Salas, Carlos. 2001. "The impact of NAFTA on wages and incomes in Mexico." in NAFTA at Seven, Briefing Paper. Economic Policy Institute, Washington, D.C.: EPI.

Salas, Carlos. 2003. "Trayectorias laborales entre el empleo, el desempleo y las microunidades en México." Papeles de Población. Vol. 9, No. 38, Oct-Dec. p. 159-94.

Salas, Carlos and Eduardo Zepeda. 2003a. "Empleo y salarios en el México contemporáneo." De la Garza. Enrique and Carlos Salas, eds., La situación del trabajo en México. Plaza y Valdéz, México.

Salas, Carlos and Eduardo Zepeda. 2003b. "Employment and Wages: Enduring the Cost of Liberalization and Economic Reform." in Kevin J. Middlebrook and Eduardo Zepeda, eds., Confronting Development: Assessing Mexico's Economic and Social Policy Challenges. California: Stanford University Press and Center for U.S.-Mexican Studies,UCSD.

Salas, Carlos and George Callaghan. 2004. "Labour and free trade: Mexico within NAFTA." in Bromley, Simon, Maureen Mackintosh, William Brown and Marc Wuyts, eds., A World of Whose Making? Making the Internacional: Economic Interdependence and Political Order. London: Pluto Press and Open University.

SolÃ*s, Leopoldo. 1981. "La realidad económica de México." Retrovisión y Perspectivas. Siglo XXI, México.

Zapata, Francisco. 1997. "The Paradox of Flexibility and Rigidity: The Mexican Labour Market in the 1990s," in Edward J. Amadeo and Susan Horton, eds., Labour Productivity and Flexibility, London: Macmillan.

[top of page]

PART 3: CANADA

Backsliding
The impact of NAFTA on Canadian workers1

by Bruce Campbell, Canadian Centre for Policy Alternatives

This section argues that the impact of the Canada-U.S. FTA and NAFTA, together with its neo-conservative policy siblings, has been adverse when measured against the standard that ultimately counts when evaluating public policy: has it bettered the lives of people affected by it? Not only has NAFTA failed to deliver the goods it promised, its effect on the well-being of a large majority of Canadians and on the social cohesion of society has been negative. Some sectors of the economy and some income groups have benefited, but the overall effect has been negative. While average income growth under free trade has registered its worst performance of any comparable period since World War II, income inequality (after tax and transfers) has grown for the first time since the 1920s.

The most striking feature of this growing inequality has been the massive gains of the richest 1% of income earners at the expense of most of the population. The growth of precarious employment, the undermining of unions as a countervailing power to transnational capital, the erosion of the Canadian social state, and heightened economic dependence on the United States are the hallmarks of the free trade era in Canada.

Parameters and promises
Any Canadian analysis of the effects of "free trade" begins not January 1, 1994, but five years earlier on January 1, 1989, when the Canada-U.S. Free Trade Agreement (CUFTA) was implemented. NAFTA extended and deepened the CUFTA; and NAFTA has been the template for other trade deals including the U.S.-Central American Free Trade Agreement (CAFTA), and the indefinitely stalled Free Trade Area of the Americas negotiation (FTAA).

Second, this analysis must recognize the difficulty of isolating the impacts of NAFTA from those of other neo-conservative or market-centered policies: monetary austerity, tax cuts, public sector cuts, privatization, deregulation, etc. Different components of this policy package may be dominant at different times. What is important is that they reinforce each other and their effects are cumulative. NAFTA is both an integral component of this policy package and also a mechanism for locking it in.

Third, NAFTA is about much more than deregulating trade. It is about removing restrictions on the mobility of capital. It goes way behind the border to the heart of domestic policy making. It is an economic constitution, conferring enforceable rights on investors, limiting the powers of government, and making it extremely difficult for future governments to change. At its core, NAFTA is about shifting the power in the economy from government to corporations, from workers to corporations.

Finally, impacts must be evaluated against claims made by Canadian free trade proponents. Among the promised benefits were the following:

* Increased economic growth, income, and employment—rising living standards that would be widely shared across all sectors, regions, and income groups.
* A closing of the longstanding productivity gap with the United States, and the creation of a more diversified, more efficient, and more knowledge-based economy.
* The ability, with the promised stronger economy, to maintain and strengthen the unique features of the more generous Canadian social model.

The economic record
Economic integration has deepened in the wake of the CUFTA and NAFTA. Two-way trade and investment flows have grown immensely. Exports as a share of Canada's GDP grew from 25% to about 40%. Canadian manufactured exports grew from one-third to over one-half of total output. Conversely, almost one-half of the Canadian market for manufactures is now met through imports.

The share of Canadian merchandise exports going to the United States grew from 73% in 1989 to 84% in 2005. The share of imports coming from the United States remained steady at about 68% until the late 1990s but since 2000 dropped steadily to 57% by 2005. One-half of all bilateral trade is intra-firm and is much higher in the manufacturing sector.

Expectations that Canada would become a magnet for foreign direct investment (FDI) from companies wanting to export into the U.S. market have not materialized. Canada's share of inward FDI flows to North America dropped from 17% to 13% during 1993-2004. Indeed, the outflow of Canadian direct investment abroad (including to the United States and Mexico) exceeded FDI inflows by one-third during this period.

Much has been made of Canada's NAFTA-driven trade success, but the reality does not live up to the hype. Canada's merchandise trade surplus with the United States—which grew from $48.6 billion in 1996 to $124.6 billion in 2005—is less than meets the eye.2 (It should be noted that deficits on the services and investment income accounts reduce the merchandise trade surplus by about one-third.)

Canadian merchandise exports to the United States grew by $138 billion from 1996 to 2005. Imports from the United States grew by $63 billion during this period. Exports peaked in 2000, fell off in 2001 and 2002, and then—spurred by the commodities boom—rose again over the last three years. Imports from the United States also peaked in 2000, fell off in 2001 and 2002, and then rose, but not as rapidly as exports.

According to Statistics Canada researchers (Cross and Ghanem 2005, 3.1), much of the growth in gross exports over the last decade reflected the markedly elevated use by Canadian-based companies of imported inputs in their production, significantly overstating the employment impact of the growth of manufactured exports. (For example, more than one-half of auto inputs are now imported.)

Furthermore, oil and gas exports alone accounted for close to 40% of the rise in exports to the United States over the last 10 years, and during the current resources boom (2003-05) accounted for 62% of the increase in exports to the United States.

The commodities boom (energy, forest and agricultural, and minerals) has boosted the share of resources in Canada's overall exports, from 40% to 50% over the last three years. Stripping out the higher import content of manufactured exports, the share of resources has risen to over 60% of total value-added Canadian exports (Cross and Ghanem 2005, 3.1). Although these sectors have experienced significant job growth, their contribution to employment overall is small.

Several other inconvenient facts contradict the claims of NAFTA-driven trade success. First, there is no evidence of Canada gaining special U.S. market advantage under NAFTA. In fact, Canada's share of U.S. imports actually fell after 1994. Second, a federal Industry Department study found that by far the largest factor—accounting for 90% of the 1990s export surge—was the low Canadian dollar (Ram et al. 2001). Finally, another Industry Department study found that the import content of Canadian exports increased to the point where, by 1997, more jobs were being destroyed by imports than created by exports (Dungan and Murphy 1999).

Contrary to the promise of free trade proponents, diversification of Canada's industrial base has been disappointing. Although there was an increase in some high-tech sectors—notably telecommunications (until the 2001 meltdown) and aerospace—the trade deficit in high-tech products remains high, the capital goods sector remains weak, and Canada's poor record in private sector R&D persists. Relative to GDP, Canada's exports of higher value-added products—including autos, machinery and equipment ,and consumer goods—have fallen by one-quarter since 1999 (Stanford 2004, 9). Ironically, NAFTA eliminated many of the policy tools that could help shift competitive advantage to more knowledge-intensive activities.

Restructuring in the Canadian manufacturing sector has been far-reaching. By 1997, 47% of the plants in existence in 1988, accounting for 28% of the jobs, had closed. On the other hand, 39% of all plants in 1997, accounting for 21% of all jobs, did not exist in 1988 (Baldwin and Gu 2003). The plants that closed tended to be larger, higher productivity plants, and those that opened were smaller, lower productivity establishments. This helps to explain the continuing manufacturing productivity gap.

Big business, by and large, has done well under free trade. A study of 40 non-financial member companies of the Canada's main big business lobby, the Canadian Council of Chief Executives, found that their combined revenues jumped 105% between 1988-2002, while their overall workforce shrank by 15% (Campbell and MacDonald 2003).

Canadian manufacturing employment, which suffered major losses—nearly 400,000 jobs—in the first four years of free trade, grew steadily thereafter, and by 2001 had returned to its 1989 level. However, its vulnerability to exchange rate movements was evident as the Canadian dollar in late 2002 began to climb. Manufacturing employment has dropped by 8.5% or 198,000 jobs (to March 2006). According to Stanford (2004), based on historical experience, manufacturing job loss could reach 400,000 jobs by 2007 if the Canadian dollar stays in the 85 cent U.S. range.

CUFTA was sold as a solution to Canada's persistent unemployment problem. Though there are other factors at play, the record does not bear this out. Average unemployment during the last 15 years has remained about the same as the average rate during the previous 15 years. Canada's unemployment rate was 6.8% in 2005, modestly lower than the 7.6% in 1989 (1.2 million workers are currently looking for work). This compares with U.S. unemployment, which was 5.1% in 2005, slightly below the 5.3% level in 1989.

Nor has promise of increased employment quality—high-skill, high-wage jobs—under free trade materialized. On the contrary, displaced workers in the trade sectors have moved to the lower-skill, lower-wage jobs in the services sector. Precarious forms of employment (part-time, temporary, and self-employment) have also increased, disproportionately impacting women and workers of color.

Furthermore, the productivity gap with the United States that was, according to proponents, supposed to narrow under free trade, has in fact widened. Canadian labor productivity (GDP per hour worked) rose steadily in relation to U.S. productivity during the 1960s and 1970s, peaking at 92% of the U.S. level in 1984. Thereafter, it slid to 89% in 1989 and by 2005 had fallen to just 82% of U.S. productivity—below where it was in 1961.3

Despite slower (almost flat) wage growth in Canada, labor cost competitiveness (unit labor costs) expressed in Canadian dollars deteriorated significantly compared to U.S. costs. It was only the depreciation of the Canadian dollar that preserved cost competitiveness. Unit labor costs expressed in U.S. dollars fell 19.7% in Canada compared to 7.2% in the United States from 1992-2002. Since then, this advantage has been eliminated by the 40% appreciation in the Canadian dollar.

If free trade was supposed to usher in a new era of rising living standards, thus reversing the sluggishness of the 1980s, the record reveals quite the opposite. Annual growth in average personal income per capita fell to a plodding 1.55% per annum in the 1980s, from the rapid 3.9% annual average gain during the 1960s and 1970s. From 1989-2005, personal income per capita growth continued its slide to a snail's pace of 0.63% yearly.4 What is particularly striking is that GDP per capita was growing almost three times faster—1.57% annually—than personal income. While U.S. GDP per capita grew at an annual rate of 1.80%, slightly faster than the Canadian rate, U.S. personal income per capita grew at an annual rate of 1.05%, almost twice as fast as the Canadian rate from 1989-2005.

Compared to American performance, Canadian GDP per capita fell sharply—from 86% of the U.S. level in 1989 to 81% in 1992—in the wake of the free trade recession (see Figure 3-A). From 1997 to 2002, a period of economic recovery driven in large part by a low dollar and strong U.S. demand for Canadian exports, GDP rose to 87% of the U.S. level. However, personal income per capita experienced no such recovery. It fell precipitously from 89% in 1990 to 78% of the U.S. level in 2000 where it has remained to the present. The growing divergence between the ability of GDP and personal income per capita to keep pace with American performance after 1996 is explained by the massive cuts to social programs, the increased share of the national income pie appropriated by profits and interest income, and the stagnation of wage income during this period. Only those at the top of the income scale saw significant growth in their earnings. It is dramatic evidence of how NAFTA-driven integration had altered relations of power between labor and capital, between state and market in the Canadian economy. This "structural adjustment" should come as no surprise. It is what NAFTA was designed to do.

Figure 3-A

The social record
Canada's social model differs significantly from the United States. Canada has a more equal distribution of earnings reflecting higher unionization rates, higher minimum wages, and a smaller pay gap between the middle and the top of the earnings spectrum. It has a more progressive tax system and a more generous system of social transfers.

Thus, while the average disposable income in the United States is higher than Canada, the bottom third of Canadians are much better off than their U.S. counterparts. The gap between middle-income Canadians and Americans is small, particularly if adjusted for out-of-pocket health care costs. It is only among the richest third where the disposable income of Americans is much greater than that of their Canadian counterparts. The after-tax-and-transfer income gap between the top and bottom 10% of families is 4-to-1 in Canada compared to 6.5-to-1 in the United States. The poverty rate (defined as less than two-thirds of the median income) is 10% in Canada compared to 17% in the United States.

That said, growing wealth and income inequality and a shrinking Canadian social state have been hallmarks of the free trade era. NAFTA, while adding pressure, does not mandate this kind of harmonization downward to the U.S. social model. Nor is it inevitable. But NAFTA competitiveness considerations have provided a pretext for the tax-cut and "smaller government" agendas of neo-conservative provincial and federal governments.

It has already been noted above that average earnings hardly grew despite steady if unspectacular productivity growth. That most of the productivity gains went to profits in the free trade era is reflected in the rise of profit income as a share of GDP at the expense of labor income—from 10.6% in 1988 to a record 14.2% in 2005.

After four decades of declining inequality, after-tax-and-transfer family income inequality widened during the free trade era. The bottom 20% of families saw their incomes fall by 7.6% during 1989-2004, while the incomes of the top 20% of families rose 16.8% (Table 3-1). During the 1980s, the incomes of the bottom 20% increased 12.8%, while those at the top stayed roughly the same. After declining during the 1980s, the incomes of top 20% of families grabbed an unprecedented extra share of the income pie during 1989-2004—41% to 44%—at the expense of the other 90% of Canadian families.

Table 3-1

A study by Saez and Veall (2003) highlights how concentrated at the very top inequality growth has been. The top 1% of Canadian taxpayers—similar to their American counterparts—increased their share of total taxable income from 9.3% to 13.6% during the first free trade decade, 1990-2000. The top 0.1% increased their share even more sharply—from 3.0% to 5.2%. The authors attribute this development in large part to pressure from deepening integration with the United States, where income inequality is much greater, and where Canadian senior executives can move more freely across the border. Subsequent U.S. tax cuts under the Bush Administration for the highest income earners have likely aggravated this situation.

The wage data compiled by Saez and Veall (2003) are even starker. While the average Canadian wage increased 8% between 1990-2000, the average wage of the top 1% of wage earners jumped 64%. Wages of the top 0.1% soared by 100%. This latter group's wages—which were 23 times greater than those of the average wage earner in 1990—had almost doubled to 43 times greater by the end of the first free trade decade.

Recent research on inequality by Frenette et al. (2006), using census data for the first time, confirms that while increases in market income inequality during the 1980s were—in contrast to the United States—fully offset by the tax and transfer system, in the 1990s, large increases in market income inequality were not offset to nearly the same degree. Transfers had no effect on reducing market inequality growth, and taxes had only a small effect in reducing the increase. As a result, the first free trade decade saw overall income inequality increase for the first time since the 1920s. This inequality reflects the shrinking social safety net discussed below.

Tracking wealth trends in the free trade era is difficult because infrequent Statistics Canada wealth surveys preclude precise benchmarks. The latest was in 1999 and before that 1984. Nevertheless, the changes from 1984-99 contrast starkly with the 1970-84 period. For example, the bottom 10% of families increased their average wealth 28% from 1970-84, but their average wealth fell 78% in the 1984-99 period. Meanwhile, the average wealth of the richest 10% of families rose 51% during 1970-84, and continued to rise, 47% during 1984-99 (Kerstetter 2002). To the extent that wealth trends mirror trends in income distribution, the free trade portion of the period would likely have seen the steepest rise in wealth inequality.

Unionization rates also fell, another sign of the eroding bargaining power of workers in the free trade era. The most trade-exposed manufacturing sector experienced the steepest decline, from 45.5% in 1988 to 32.6% in 2003 (Jackson 2005, 170). This reflects disproportionate closures of unionized plants and a disproportionate concentration of new hiring in non-union plants; it also reflects legislative attacks in key jurisdictions on organizing capability. The decline in union density overall, though not as steep—39.5% to 32.4%—is ongoing and is evident across all sectors of the economy.

The shrinking social state
If the first half of the 1990s saw the most intense restructuring of the corporate sector, the second half of the decade saw the "structural adjustment" of the public sector. Federal non-military program spending cuts were the largest in Canadian history, bringing spending down to the level of the late 1940s. Canadian governments collectively reduced their program spending from 41% to 32% of GDP from 1992 to 2005. Governments reduced transfers to persons from 11.5% to 7.8% of GDP during this period (Mackenzie 2006). The cuts were accompanied by a major re-engineering of government—privatization, deregulation, and decentralization.

Reversing its pre-CUFTA promise, the big business lobby pushed hard for personal and corporate tax cuts on the grounds that they were necessary to maintain competitiveness, attract investment, and fuel growth. The federal government complied with major tax cuts, which shrunk federal revenue as a share of GDP from 17.2% in 1997-98 to 15.4% in 2004-05 (Finance Canada 2005), representing a loss to the federal treasury of $C20 billion in the latter year. Provinces also cut taxes, the combined effect of which was a loss to provincial treasuries of $C30 billion in 2005 (Lee 2006). The benefits of the tax cuts were tilted to high-income groups and to the corporate sector despite the fact that lower-income groups had borne the brunt of the program cuts. Canada has dropped in the Organization for Economic Co-operation and Development (OECD) ranks from a middle-level taxation country to the bottom third of OECD countries in terms overall taxation level.

Business also changed its tune around social programs once CUFTA was passed, arguing that cuts—especially welfare and unemployment insurance—were necessary to create a level playing field of competition. The largest of the unemployment insurance cuts were made by the Liberal government under cover of deficit elimination, but were also part of a strategy to increase labor market "flexibility." This was done by reducing the eligibility criteria and by reducing the duration and amount of benefits. Thus, the proportion of unemployed people who qualified for unemployment insurance dropped from 75% in 1989 to 38% by 2002, about the same level as in the United States. Hardest hit were the most vulnerable workers—part time, casual, and seasonal—mainly women.

The federal government also slashed welfare transfers to the provinces, breaking its 50-50 cost-sharing commitments under the Canada Assistance Program. Most provinces in turn slashed welfare support payments and bumped hundreds of thousands of people off the welfare rolls altogether.

While Canadian governments still spend significantly more on social programs and public services than their American counterparts, the difference has been shrinking rapidly. A federal Finance Department study found that Canadian government (non-military) program spending fell from 42.9% of GDP in 1992 to 33.6% of GDP in 2001. This compares with United States (non-military) program spending, which increased marginally from 27.7 to 27.9% of GDP in this period. The 2001 gap in non-military spending between the two countries—5.7 percentage points of GDP—is down dramatically from a 1992 gap of 15.2 points of GDP (Table 3-2). Thus, if Canadian governments were still spending at 1992 levels, they would have spent an additional $C103 billion on programs and public services in 2001 alone.

Canada now spends proportionately less than the United States on public education, only slightly more on health care (though more efficiently because of not-for-profit delivery and a single-payer Medicare insurance system.) It continues to spend substantially more on income security and, though the gap has shrunk by half, more on housing and community services.

Table 3-2

Conclusion
Economic and political elites promised that free trade would usher in a golden era of prosperity for Canada. It clearly has not delivered the goods. Nevertheless, these elites simply disregard the "inconvenient facts" presented here as they push for even deeper forms of continental free market integration. NAFTA, they say, has greatly increased exports and investment; Canada's trade surplus is up, unemployment is down, inflation is low, wages are flat, business is experiencing record profits, growth is steady. Therefore NAFTA has been a success. What is there to re-examine? Let's just move forward, they say, and build on our success.

Instead of continuing down this road, it is time to look back at the road already traveled. We should undertake a comprehensive assessment of NAFTA's costs and benefits, and take a hardheaded look at the advantages and disadvantages of withdrawing from NAFTA. It is time to stand back and ask: is NAFTA working for us? Do the benefits outweigh the costs? Is it serving our needs? It is time to reconsider whether NAFTA in its current form, is contrary to the well being of Canadian workers (and indeed of workers in all three NAFTA countries) as the overarching framework for managing North American economic relations.

Endnotes
1. I am greatly indebted to andrew Jackson, chief economist at the canadian labour congress, from whose work i draw heavily for this paper.

2. All figures are in U.S. dollars unless specified otherwise.

3. Statistics canada data posted at www.csls.ca.

4. Statistics canada data posted at www.csls.ca.

References
Baldwin, John and Wulong gu. 2003 Plant Turnover and Productivity Growth in Canadian Manufacturing. Statistics Canada. http://www.statcan.ca.

Campbell, Bruce. 2000. "False Promise: Canada in the free trade era" in NAFTA at Seven: Its Impact on Workers in All Three Countries. Washington, D.C.: Economic Policy Institute.

Campbell, bruce and David macDonald. 2003. Straight Talk: Big Business and the Canada-U.S. Free Trade Agreement Fifteen Years Later. ottawa, ontario: Canadian Centre for Policy Alternatives.

Cross, P. and Z. Ghanem. 2005. "Canada's natural resource exports." Canadian Economic Observer. Statistics Canada. May. http://www.statcan.ca.

Department of Finance Canada. 2005. Economic and Fiscal Update. October. http://www.fin.gc.ca/fin-eng.html.

Dungan, P. and S. Murphy. 1999. "The changing industry and skill mix of Canada's international trade." Perspectives on North American Free Trade. Paper no.4. Ottawa, Ontario: Ondustry Canada.

Frenette, Marc, David Green, and Kevin Milligan. 2006. Revisiting Recent Trends in Canadian After-Tax Income Inequality Using Census Data. Statistics Canada Research Paper, family and labour studies division. February.

Jackson, Andrew. 2003. From Leaps of Faith to Hard Landings Fifteen Years of "Free Trade." Ottawa, Ontario: Canadian Centre for Policy Alternatives.

Jackson, Andrew. 2005. Work and Labour in Canada: Critical Issues. Toronto, Ontario: Scholars Press.

Kerstetter, stephen. 2002. Rags and Riches: Wealth Inequality in Canada. Ottawa, Ontario: Canadian Centre for Policy Alternatives.

Lee, Marc. 2006. Tax Cuts and the Fiscal Imbalance. Ottawa, Ontario: Canadian Centre for Policy Alternatives, May.

Mackenzie, Hugh. 2006. The Art of the Impossible Fiscal Federalism and Fiscal Balance in Canada. Ottawa, Ontario: Canadian Centre for Policy Alternatives. June.

Ram, C. Acharya, Prakesh Sharma, and Someshwar Rao. 2001. Canada's Trade and Foreign Direct Investment Patterns with the United States. Paper presented to North American Linkages Conferences. Calgary, Alberta.

Saez, E and M. Veall. 2003. The Evolution of High Incomes in Canada, 1920-2000. Working Paper. National Bureau of Economic Research. June.

Stanford, Jim. 2003. Canada's Economic Role in North America: Evaluating the Impacts of Integration and Considering the Next Steps. presentation to the Policy research networks roundtable.

Stanford, Jim. 2004. Dollar'' Rise Will Exact Painful Toll on Canadian Manufacturing, December. Canadian Auto Workers union. www.caw.ca.

Statistics Canada. 2004. Historical statistical supplement 2003/04. Canadian Economic Observer. Ottawa, Ontario: Statistics Canada. July. http://www.statcan.ca/english/freepub/1 ... 10-xib.pdf.