Displaced Workers and Unemployment Insurance in Mexico: Preparing for the Next Crisis
David S. Kaplan and Raymond Robertson
Inter-American Development Bank / Macalester College
May 1, 2011
The recent global economic crisis highlighted the fact that Mexico does not have a safety net that is sufficient for dealing with transitory but serious shocks to employment. We believe that a system of unemployment insurance should be a priority for the country. We hope that in the near future, before the memory of the 2008 crisis fades, policymakers will consider adopting such a system. We especially hope that an unemployment insurance system is in place when the next crisis occurs.
Unemployment in Mexico’s northern border region jumped from under 2% in the third quarter of 2007 to over 7.5% in the third quarter of 2009. Unemployment rates were even higher in Mexico City. Like many other developing countries, the global financial crisis brought social protection programs in Mexico under renewed scrutiny because, while the economic crisis may have been temporary, the effects of displacement may be permanent.
Since the publication of Jacobson et al. (1993), the topic of displaced workers has been actively researched in the literature. A displaced worker is defined to be a worker who is forced to leave a job for reasons that are unrelated to his or her job performance. By focusing on workers who separated from their firms as part of mass layoffs, Jacobson et al. were able to study the effects of displacement (that is, of job loss) on subsequent labor market performance.
The results of Jacobson et al. (1993) were striking. They found that the negative effects of displacement for U.S. workers began prior to displacement, were large, and were long lasting. Interestingly, subsequent papers found either no effects or even positive effects of displacement. For example, Abbring et al. (2002) found no change in wages in the United States and Bender et al. (2002) found positive wage changes following displacement in France and Germany. These differences suggest that that both time and location of displacement may significantly affect how workers fare after losing their jobs. Further research was needed to understand why the effects of displacement could vary to such a large extent.
It was in this context that in 2005 we wrote, together with Gabriel Martínez, a paper for Economía—LACEA’s policy journal (see Kaplan et al. 2005). We applied a methodology similar to that of Jacobson et al. (1993) to study displaced workers in Mexico. Our paper was the first such study to use data from Latin America. We exploited important geographical and time differences to show that, even when studying the same country with the same dataset and the same methodology, it is possible to find negative, zero, or positive effects of displacement on subsequent wagesdepending on the time or region that is analyzed.
Perhaps unsurprisingly, we found that the effects of displacement were worse in the poorer regions of Mexico and that the effects of displacement during the economic crisis in 1995 were worse than the effects of displacement during the subsequent recovery. These findings led us to conclude that “…targeting aid to displaced workers during recessions and in less economically active areas has potentially significant efficiency gains. These workers tend to suffer larger and more lasting adverse effects from displacement than other workers, which suggests that targeted aid may be especially valuable.”
Unfortunately, when the global economic crisis hit Mexico in 2008, little had been done in terms of social protection programs that targeted displaced workers. Although significant progress had been made in the area of conditional cash transfers, these transfers are poorly suited to mitigate transitory shocks because eligibility depends mainly on more permanent aspects such as the conditions of the family’s residence. As unemployment rose and the formal sector contracted, there was no comprehensive mechanism in place to protect displaced workers from transitory but quite serious shocks.
Protection for displaced workers has become increasingly important as Mexico becomes more integrated into the world economy. Mexico is a leader among developing countries in pursing trade agreements and free-trade policies. As a result, trade volumes have increased significantly over the last 20 years. Nevertheless, this increase in trade is correlated with an increased susceptibility to international shocks. For example, Bergin et al. (2009) describe how Mexico's maquiladoras have generated significant employment opportunities for Mexican workers, but these jobs are in an especially volatile part of the production chain.
The significant benefits of higher employment therefore come with the potential cost of being increasingly subject to international shocks, and being in the volatile part of the production chain implies that adjustment in Mexico might be greater than in countries that are upstream in the value chain. Indeed, the U.S. financial crisis was transmitted to Mexico as a trade shock: the decline in Mexican trade was highly correlated with the decline in U.S. GDP (Robertson 2009). Between the local peak (around April 2008) and the trough (January 2009), Mexican trade fell about 43% in real terms. The drop nearly erased a decade of trade growth in a few months.
The sharp drop in trade could have affected wages, employment, or both. In a recent paper we wrote with Daniel Lederman (see Kaplan et al. 2011), we found that the 2008 recession was particularly serious in northern Mexico due to their very strong linkages with the United States. Our econometric analyses indicate that, despite the fact that wages respond somewhat to trade shocks, the labor market mainly adjusts to adverse trade shocks by displacing workers. Specifically, we find that formal employment in the trade-intensive northern states fell more than 9% from September 2008 to March 2009, but the drop in wages was closer to 1%. These results are consistent with those of Robertson and Dutkowsky (2002) who find that firms’ de facto costs of adjusting employment in Mexico are quite low. From the perspective of a Mexican worker during a crisis, however, the increased flexibility of employment that results from low adjustment costs translates into high probabilities of losing a job, quite possibly with no safety net to help smooth out the negative shock.
Our econometric analyses also find that lower wage workers are particularly likely to lose their jobs during a crisis, which once again is consistent with the results of Robertson and Dutkowsky (2002) who find that the firms’ costs of adjusting the employment of low-skilled workers are lower than those for high-skilled workers, which implies that the employment of low-skilled workers will fall dramatically in bad economic times and recover dramatically in good times. All evidence suggests that the lower-skilled workers, those who are least likely to be able to mitigate the effects of adverse events by tapping into their savings, are most likely to be displaced during an economic crisis.
One might argue that existing employment protection legislation should help these displaced workers cope with unemployment, at least for a while. Mexican labor law specifies that workers who are dismissed without cause are entitled to at least 12 days’ worth of wages for each year of tenure accumulated at the firm. Unfortunately for these workers, however, firms do not always comply with their legal obligations regarding severance payments.
Dávila (1997), for example, finds that around 30% of workers in Mexico who consider themselves unjustly dismissed do not go to court, mainly because of high uncertainty about outcomes and the long duration of the processes. More than two-thirds of those who actually initiate legal action end up settling out of court for a rather low fraction of what the law dictates. Kaplan and Sadka (2008) find that less than half of workers who are awarded a severance payment by a judge in a labor court in Mexico are successfully able to collect their award. Kaplan and Sadka (2010) analyze these uncollected awards in greater detail and find that the process of enforcing a favorable judgment in a labor court is time consuming and quite costly for a worker.
The results of these papers make it quite clear that workers cannot count on the severance payments that are specified in Mexican labor law as a safety net in case of displacement. These results also explain why Robertson and Dutkowsky (2002) find that the de facto labor market appears so flexible although the de jure adjustment costs are quite high.
One might therefore be tempted to suggest that a return to a more protective trade policy would be an effective way to protect workers from increased volatility. As a result of Mexico’s intensive efforts in pursuing free trade policies, trade has both grown and diversified. In our opinion, the benefits of the resulting job growth -especially in the maquiladora sector- as well as the gains to consumers in the form of lower prices suggest that using more protective trade policies to increase employment stability would not bring net gains to the country.
Instead, we advocate for the introduction of an unemployment insurance system to protect displaced workers. Several countries in the region have already implemented an unemployment insurance system or system of individual accounts to protect unemployed workers, most notably Argentina, Brazil, Chile, Colombia, and Uruguay. Mexico might be well advised to have an unemployment insurance system in place before the next, possibly trade-induced, economic crisis occurs.
Some evidence exists that unemployment insurance schemes in Latin America are achieving their objectives. Kugler (2005), for example, finds evidence that Colombian workers place much higher value on money in an individual account that can be accessed in case of unemployment than in money that is “promised” in case of unemployment in the form of severance payments. She points out that, in order to avoid severance payments, firms can declare bankruptcy and “take the money and run.” Amarante et al. (2010), show that unemployment benefits in Uruguay have been effective in smoothing out income shocks for displaced workers.
It is also worthwhile noting that Martínez (2010) offers a specific proposal for an unemployment insurance system in Mexico. He proposes that formal-sector workers receive contributions into individual accounts that would be accessible during periods of unemployment. He also proposes an additional subsidy for low-wage workers that would be financed through government funds, which would impose only a small fiscal cost for the government. We believe that the Martínez proposal is an excellent step forward in promoting a debate on the best way to implement unemployment insurance in Mexico and we encourage the elaboration of alternative proposals.
An addition to the Martínez proposal that offered some help to informal workers would be particularly useful. The existing federal and state employment services, which offer job-finding assistance, training, and monetary support to unemployed people regardless of whether they have formal-sector experience or not, may provide the necessary infrastructure to offer some support to workers who are displaced from the informal sector. The public employment service has the operational experience and the local presence to take on the responsibility of screening displaced workers for eligibility while also providing job search support through its existing programs.
In conclusion, the recent global economic crisis once again highlighted the fact that Mexico does not have a safety net that is appropriate for dealing with transitory but serious shocks to displaced workers. We believe that a system of unemployment insurance should be a priority for the country. We hope that in the near future, before the memory of the 2008 crisis fades, policymakers will consider adopting such a system. We especially hope that an unemployment insurance system is in place when the next crisis occurs.
Disclaimer: The views expressed in this document are those of the authors, and do not necessarily represent the views of the institutions that they represent.
Abbring, Jaap H. et al. 2002. "Displaced Workers in the United States and the Netherlands." Losing Work, Moving on: International Perspectives on Worker Displacement. Ed. Peter J. Kuhn ed. Kalamazoo, Mich.: W. E. Upjohn Institute for Employment Research, 105-194.
Amarante, Verónica, Rodrigo Arim, and Andrés Dean. 2002. “The effects of being out of the labor market on subsequent wages: evidence for Uruguay.” Mimeo.
Bender, Stefan, Christian Dustmann , David Margolis , and Costas Meghir. 2002. "Worker Displacement in France and Germany." Losing Work, Moving on: International Perspectives on Worker Displacement. Ed. Peter J. Kuhn ed. Kalamazoo, Mich.: W. E. Upjohn Institute for Employment Research, 375-470.
Bergin, Paul; Robert Feestra, and Gordon H. Hanson. 2009. “Offshoring and Volatility: Evidence from Mexico’s Maquiladora Indsutry,” American Economic Review, September (99): 1664-1671.
Dávila, Enrique R. 1997. “Mexico: The Evolution and Reform of the Labor Market.” in S. Edwards and N. Lustig (eds.) Labour Markets in Latin America. Ed. Sebastian Edwards and Nora Lustig. Brookings Institution Press, Washington D.C.
Jacobson, Louis S., Robert J. LaLonde, and Daniel G. Sullivan. 1993. "Earnings Losses of Displaced Workers." American Economic Review 83.4: 685-709.
Kaplan, David S., Daniel Lederman, and Raymond Robertson. 2011. “Employment and Wage Responses to Trade Shocks: Evidence from Mexico during the 2008-09 U.S. Recession.” Mimeo.
Kaplan, David S., Gabriel Martínez, and Raymond Robertson. 2005. “What Happens to Wages after Displacement?” Economia: Journal of the Latin American and Caribbean Economic Association 5.2: 197-234.
Kaplan, David S. and Joyce Sadka. 2008. “Enforceability of Labor Law: Evidence from a Labor Court in Mexico.” World Bank Policy Research Working Paper no. 4483.
Kaplan, David S. and Joyce Sadka. 2010. “The Plaintiff’s Role in Enforcing a Court Ruling: Evidence from a Labor Court in Mexico.” Mimeo.
Kugler, Adriana. 2005."Wage-Shifting Effects of Severance Payments Savings Accounts in Colombia." Journal of Public Economics 89.2-3: 487-500.
Martínez, Gabriel. 2010. “Unemployment in Mexico: Policy Options, Unemployment Insurance and a Comparison with EPL Rules.” Mimeo.
Robertson, Raymond, and Donald H. Dutkowsky. 2002. "Labor Adjustment Costs in a Destination Country: The Case of Mexico." Journal of Development Economics 67.1: 29-54.
Robertson, Raymond. 2009. " Mexico and the Great Trade Collapse" in The Great Trade Collapse: Causes, Consequences and Prospects. Ed. Richard Baldwin. A VoxEU.org. Ebook available at http://voxeu.org/index.php?q=node/4297.
For Membership information and other activities of LACEA click here.