Most people will have a view on the contribution (if any) of globalisation and trade to inequality. There are two main theories that suggest that trade increases inequality. The first theory is the Stolper–Samuelson theorem, which in turn comes from the Heckscher–Ohlin model of international trade, which suggests that trade increases inequality in high-income countries, but decreases inequality in low-income countries.
The explanation works like this. The Heckscher-Ohlin model says that when a country opens to trade, the returns to relatively abundant factors of production in that country will increase, while the returns to relatively scarce factors of production will decrease. Now, in a high-income country, skilled labour is relatively abundant, while unskilled labour is relatively scarce. So, in a high-income country trade increases the wages of skilled workers, but decreases the wages of unskilled workers, increasing inequality. For low-income countries though, unskilled labour is relatively abundant, while skilled labour is relatively scare. So, in a low-income country trade increases the wages of unskilled workers, but decreases the wages of skilled workers, decreasing inequality. At least, that's what the theory says. In practice, trade has increased alongside improvements in technology that have increased the productivity of skilled workers (what economists call skills-biased technological change), which means that trade has been associated with higher inequality within all countries.
The second theory for why trade increases inequality comes from a very influential 2023 paper by March Melitz. In this theory, more productive firms are more likely to export than less productive firms. That means that the wages of workers in exporting firms will increase more than those of workers in non-exporting firms, increasing inequality.
Now, if trade increases inequality (at the least for high-income countries), does it matter what countries they trade with? Does trading more with a high-inequality country have a different effect than trading more with a low-inequality country? In other words, is higher inequality transmitted through trade?
Those are the questions that this recent article by Sergey Nigai (University of Colorado Boulder), published in the American Economic Journal: Economic Policy (sorry, I don't see an ungated version online), sets out to answer. The paper is quite complex and not for the faint-hearted. However, once you realise what Nigai is doing, it is quite elegant. Nigai proposes an interesting mechanism for why inequality in the trade partner country should matter, where firms:
...target specific segments in the distribution of consumers, thereby creating connections between consumer income inequality and the distribution of firm profits. Targeting rich population segments is costly such that there is a positive assortative matching between high-productivity firms and rich consumers. Hence, more unequal income distributions in export markets raise the profits of high-productivity firms relatively more, which, in turn, leads to higher incomes of individuals associated with these firms––ultimately increasing domestic income inequality.
Notice that this is a slightly more nuanced version of the Melitz theory. In this case, the more productive exporting firms sell to the highest income foreign consumers, increasing their profits more than the less productive exporting firms, which sell to lower income foreign consumers. This then leads to higher wages at the more productive exporting firms relative to wages at the less productive exporting firms, which leads to higher inequality. Nigai illustrates the mechanism in Figure 2 in the paper:
On this figure, Nigai explains that:
There are three relevant firm-level relationships:
(R1) Higher income inequality in the importing country generates higher dispersion of profits/export revenues across exporter firms in the exporting country.
(R2) Higher dispersion of export profits/export revenues is associated with higher dispersion of worker incomes across exporter firms such that wage inequality in the exporting country also increases.
(R3) Higher income inequality across workers employed in exporter firms increases overall income inequality in the exporting country.
In terms of the aggregate outcomes, R1–R3 result in R4, indicating a positive effect of higher income inequality in the importing market on inequality at home such that:
(R4) Higher exports to high-inequality countries are associated with increasing inequality in the domestic country.
Next, Nigai notes that R2 and R3 are easily established because:
substantial evidence based on aggregate and micro-level data shows that there is a robust, strong, and positive relationship between exporters’ profit and wages... Given this evidence, the relationship in R2 must hold mechanically. Second, the relationship in R3 also holds mechanically, as higher wage inequality among exporter firms that pay higher wages relative to pure domestic producers must have a positive effect on overall income inequality in the exporter country.
Nigai then goes about showing empirical evidence for R1 and R4. I won't get into the weeds of the methods here, because they are fairly complex, but suffice to say that for R1, firm-level evidence suggests that income inequality in the export destination is associated with higher dispersion of profits in the exporting country. Specifically, income inequality has no effect on profits for most firms, but has a significant effect on the profits of the most-productive firms. On this, Nigai notes that:
...the effects of income inequality on the dispersion of profits operate mainly through large exporters in the right tail.
On R4, Nigai uses country-level data and an OLS panel regression model, and finds that:
...the Gini coefficient in country i would increase by approximately 1 percent if the Gini coefficients in all export markets for country i increased by 10 percent.
The effects for an instrumental variable regression are smaller, but still statistically significant. Overall, the results support both R1 and R4.
Nigai then turns to estimating how important this channel is as an explanation for inequality. He parameterises a general equilibrium economic model based on data from 40 mostly OECD countries, and finds that:
...for an average country, consumer targeting and inequality effects transmitted through international trade explain 4.4 percent of the observed levels of the Gini coefficients and 4.8 percent of the observed levels of income shares of the top 1 percent of population.
That doesn't sound like a lot, but 4.4 percent of the Gini coefficient is the difference in Gini coefficient between Australia (ranked 10th highest inequality in the OECD, with a Gini Index of 34.3), and South Korea (ranked 17th, with a Gini Index of 32.8). So, it's reasonable substantial.
Overall, it appears that not only does trade contribute to inequality, but the specific trade partners also matter. Is there a policy implication from this? Should countries start imposing export tariffs or export controls on exports to more unequal countries? Answering those questions would require a more careful consideration of the welfare impacts of trade, since it suggests a trade-off between the gains from trade and the inequality effects of trade.
No comments:
Post a Comment