Blending Social Protection Theory with Heckscher-Olin in Trade Agreement Analysis

In light of the recent trade deal controversies, I decided to post the following paper I wrote regarding how social protection affects free trade politics within various countries:

In the study of trade and market integration, two theories are very relevant: the Heckscher-Ohlin international trade theory and the Estevez-Abe, Iversen and Soskice theory about skill formation and social protection regimes. Specifically, an implication of the Heckscher- Olin trade theory is that in trade negotiations those who own a scarce factor will be against free trade, and those holding abundant factors will favor free trade[1]. The theory advanced by Estevez-Abe, Iversen and Soskice draws a connection between the types of skills formed in a country and the social protection regimes that are in place—specifically employment protections and unemployment insurance. While these theories are powerful on their own, I will investigate the possibility that together they can help explain some of the anomalies of the politics that surround free trade efforts in different countries.

My idea is only relevant when examining countries with similar factor endowments. For the purposes of this paper, I will only analyze countries with a relative scarcity of labor but an abundance of capital, land or both. Most OECD countries fall into this category, and this is convenient because Estevez-Abe et al only examine OECD countries in their skill formation theory[2].  It is important to note, however, that if my hypothesis is true in this initial phase of analysis, it should produce many testable implications for countries not in the OECD.

I have two possible hypothesized casual mechanisms. One is as follows: countries with high employment and unemployment protection have populations with more specific skill sets, while countries with low protections have populations with more general skills. This abstracts away the four categories outlined by Estevez-Abe et al, and combines them into two.[3] I assume that the scarce factor, which is labor, will generally oppose free trade. However, in countries where laborers have very firm or industry specific skill sets, this opposition will be more forceful because free trade will cause their industries to be exposed to potentially devastating competition from countries with abundant labor. These laborers will find it harder to transition to a new firm or industry. In countries with more general skill populations, labor will better be able to shift to a different industry or firm, and as a result will not be as forcefully opposed.

The second is as follows: I again assume that countries with high employment protection and unemployment protection have highly specialized labor forces. In this pathway unskilled workers will still oppose free trade to some degree, but they will not be as vehemently opposed, because they are insured against dramatic shifts in the market through unemployment and employment insurance. This story ignores skill sets and only examines the fact that risk is mitigated and removed from the unskilled worker as a function of social insurance. The largest implication of this is that in high protection societies, it should be domestic companies or employers that are marginally more anti-free trade then the average unskilled worker.

An interesting concept I will mention throughout this paper is something that I call “hiring (or employment) elasticity.” In countries with social insurance regimes that are based primarily on employment protection, firms will have low hiring elasticity, meaning that they will be unable to easily cut jobs or adjust their level of labor when the market shifts from something like a free trade agreement. Countries with high levels of unemployment insurance result in laborers with high “employment” elasticity: they can more credibly threaten or risk leaving their job because unemployment insurance in these countries pays out a higher replacement ratio.

In order to look at empirical evidence, I again keep with the Iversen theory and use their casual pathway—that employment and unemployment  protections cause a nation to be either more general or more specific in its skill sets. Specifically, that more protections result in more workers investing in specific skills, and less protections result in workers “self-insuring” by pursuing general skills. Because I accept this pathway, I combine their employment and unemployment protection metrics into one metric, which I call the protection variable. I combine them because of my earlier decision to abstract the groupings from four into two—one group of high protection specific skill countries, and one group of low protection general skill countries.

In an initial OLS regression analysis examining the effect of my protections metric on tariff levels in the early 1990s, I found an insignificant effect. I also found no significant effect from employment or unemployment protections taken alone or together in separate models.  However, this analysis used only 15 OECD countries and as a result did not fully control for all relevant factors. If more extensive country data for the rest of the OECD countries or perhaps other developing countries in general can be found, then more detailed large-N analysis can be undertaken. For now, I will explore some critical test cases to see if my hypothesis has any predictive power.

The first case is the EU-South Korea Free Trade Agreement that went into effect in 2010. This agreement would substantially lower tariff barriers however it is important to note that an initial review of the documents and press releases of the time reveals no significant opposition movement. The only noted opposition came in the form of French and Italian automakers, which feared Korean car imports[4]. The fact that there was little opposition to the FTA in general seems consistent with Hekler-Olsen, as the average salaries in South Korea and the European Union are almost identical, as is the relative level of capital. Because of this, European workers would not need to fear a fall in wages. The opposition coming from France and Italy also makes sense, as both countries have high levels of employment protection, and thus employers have low hiring elasticity. Their profits will bear the brunt of the cost of adapting to more automobile competition, as they cannot fire laborers. Italy also has extremely low unemployment insurance levels—and yet there was no visible opposition at the worker or grass roots level that I could see from a brief review of the reports. This case is thus mildly supportive of the second pathway.

In another instance, the United States became embroiled over whether or not to renew China’s Most Favored Nation privileges. Removal of such privileges would result in a change in tariff rates. This situation is interesting because it is clear that unskilled labor is the abundant factor in China but the scarce factor in the United States. Although the issue has clear implications for trade and the various factors in the countries, the point of contention was mainly China’s crackdown on protesters following the incident in Tiananmen Square[5]. This makes the case interesting, because it was clearly an exogenous shock to the free trade dynamic in the United States. As expected, one of the largest collective action groups representing labor causes, the AFL-CIO, issued a strong statement against granting China a renewal of status[6]. Based on a review of all press releases issued by the AFL-CIO in 1996, 1997 and 1998, I found that there were no other press releases regarding free trade issues. Taking this as a crude measure of importance, it seems clear that the AFL-CIO, and labor at large, took a firm stand on an issue that actually split Congress along nonparty lines[7]. Because the United States has the lowest level of employment and unemployment protection in the OECD sample of the Estevez-Abe paper, I take this firm stance as evidence that the second pathway is a better mechanism for the effects of social protections on free trade debates.

With these two cases in mind, I conclude that most anecdotal evidence points to the second casual pathway. To be precise, this pathway is that less social employment or unemployment protections will tend to make unskilled labor more vulnerable to market shifts due to free trade agreements. As a result, in countries with sparser protections, like the United States, Ireland, etc[8] the labor movement will be more actively opposed to free trade agreements. The empirical data that I reviewed, while not yielding anything conclusive, does tell the story that free trade agreements are complicated, and the relative strength of the factions involved will depend on population size, type of domestic industrial production, and institutional design.

The case of the South Korea-EU agreement also sheds light on the fact that there may be cause for more exploration as to the different effects caused by employment vs unemployment protections. The individual case does seem to suggest that perhaps employment protections cause firms to more actively oppose free trade because of reduced flexibility. This would be similar to how rigid employment regimes create higher natural levels of unemployment in many European countries. Finally, the Heckscher-Ohlin trade theory seems to hold true, especially in cases like the one observed in the United States. As one might expect, higher disparities in unskilled labor scarcity do seem to determine the level of activism on the part of the AFL-CIO.




Works Cited

Estevez-Abe, Margarita, Torben Iversen, and David Soskice. “Social Protection and the Formation of Skills: A Reinterpretation of the Welfare State.” Social Welfare in Global Context (1997): 134-56. The London School of Economics. Web.

HECKSCHER–OHLIN TRADE THEORY (n.d.): n. pag. Rochester Economics. Rochester University. Web. <;.

Lee, Ho-Jin. “The EU-Korea Free Trade Agreement.” Brookings. The Brookings Institution, 12 Oct. 2010. Web. 02 June 2015. <;.

“Most Favored Nation (MFN) Trading Status for China.” AFL-CIO. AFL-CIO, 22 May 1997. Web. 02 June 2015. <;.

Simon, Roger. “Clinton Pleads For China’s Trade Status.” Tribune Digital. Chicago Tribune, 04 June 1998. Web. 02 June 2015. <;.


[1] Heckscher-Olin, page 1.

[2] Estevez-Abe et al page 145.

[3] Ibid.

[4] Lee.

[5] Simon.

[6] AFL-CIO.

[7] Simon, lower part of page.

[8] Estevez-Abe, page 168.