Exporting to insecure markets

Matthieu Crozet, Pamina Koenig , Vincent Rebeyrol, 14 January 2009

a

A

The quality of institutions is a very important matter for the world economy. Countries with stable economic, social and judiciary institutions favour long-term growth. Among the various aspects of international economic relations that are affected by the quality of institutions is world trade. A large body of work already documents that the poor quality of institutions and the subsequent unreliable environment for international exchanges participate in the realm of formal and informal trade barriers making international trade relations more difficult. Among others, Anderson (2000), Anderson and Marcouiller (2002), Dollar and Kraay (2002), François and Manchin (2006) and Levchenko (2007) show that countries with better institutions trade more. Hence, political risk and institutional failures can be assimilated to an ad-valorem trade barrier. Anderson and Marcouiller (2002) assert that "predation by thieves or by corrupt officials generates a price mark-up equivalent to a hidden tax or tariff", and Blomberg and Hess (2006) estimate that the impact of terrorism and wars is equivalent to a 30% tariff.

On aggregate, institutional failures act like a simple trade barrier. But one might wonder about the microeconomic mechanisms generating this result: Does insecurity discourage new exporters from entering foreign markets, and which exporters in particular? How do poor institutions affect the amount exported by firms to insecure markets? This information is of central importance in the debate concerning the effects of economic and political integration on trade and welfare. Indeed the market structure of imported products determines the level of prices and the gains from trade for consumers.

The recent academic publication of trade models with heterogeneous firms (Melitz, 2003; Chaney, 2008) describes the behaviour of exporting firms and the subsequent patterns of bilateral trade in a framework that supports recent empirical analyses. These models consider that firms willing to export have to pay an additional fixed cost; consequently, exporting is profitable only for firms that are competitive enough to earn a sufficiently large market share abroad. Two different firm-level mechanisms govern aggregate international trade flows: changes in the number of exporters (the extensive margin) and changes in the volume exported by each firm (the intensive margin).

In this setting, a decrease in trade barriers has two effects on firm-level trade: it increases the number of exporters by letting less productive firms sell on foreign markets, and it increases the volume exported by existing exporters. However, the case of institutional quality may have a slightly different impact on international trade. In a formal trade model with heterogeneous firms, we investigate the role of institutional failures on trade, and emphasise a major difference between tariffs and insecurity (Crozet, Koenig and Rebeyrol, 2009).

Insecurity in the export market is not a simple trade barrier

While formal trade barriers like tariffs affect all potential exporting firms, insecurity does not. We describe insecurity as an exogenous probability for firms to be directly hurt by a negative event when trying to enter the export market. Unlucky exporters have to pay an extra fixed cost to sell on the foreign market: they can lose their shipment because of hijacking, have to pay a bribe, or face particular obstacles related to poor governmental regulations. Consequently, the level of the export fixed cost which determines whether a firm decides to export or not becomes uncertain.

A crucial assumption of the model is that these troubles only affect a random subset of potential exporters. As a result, two major predictions emerge.

First, insecurity decreases the volume of bilateral exports by reducing the number of exporters. However, in contrast with the existing literature, a higher level of insecurity may dissuade unlucky productive firms from exporting, while some lucky unproductive ones may succeed. Hence, the selection pattern linking exporting firms to their productivity is weakened when the foreign country exhibits institutional failures, and the firms that are evicted from exporting due to fiercer insecurity are not necessarily the least productive exporters.

The second prediction relates to the respective importance of the intensive versus extensive margins increase following a security improvement in the importing country. In a country with initially low insecurity, a lowering of insecurity increases trade mostly through an increase in the extensive margin, hence by letting a large number of small firms enter the export market. In contrast, in a country with initially a high level of insecurity, a decrease in the level of institutional failures will more significantly increase the average incoming shipment and less importantly increase the number of exporters.

This difference means that there are very different trade-related incentives to improve the quality of institutions. The impact in terms of increasing the number of firms selling on the market (and hence the degree of local competition) is higher in the case of an initially relatively secure country.

Empirical evidence

Let us visualise the effect of the quality of institutions on trade through a glance at the data. We use individual French firm-level export data to more than 100 destinations for the period 1986-1992, together with data provided by IRCG (International Country Risk Guide) as a proxy for insecurity. ICRG provides annual indices of political stability (measured by socioeconomic conditions, democracy, and ethnic tensions or military conflicts), and determinants of the business climate, such as the contract viability and payments delays, corruption, efficiency of the bureaucracy, and legal system.

We first estimate the role of insecurity on the probability that a firm exports to a given country (110 in our sample). Results follow the theoretical predictions: the political insecurity variable has the expected negative sign: a 10% increase in the ICRG index reduces the probability of exporting to this destination by 0.7% to 2.8%, depending on the estimation method. Hence better institutional conditions in the importing country increase the number of firms exporting to that country. From this point of view, insecurity acts as a trade barrier, in the sense that it decreases the number of firms selling to the insecure country.

However the results show an additional important feature: we split the sample into two sub-groups, above-median and below-median insecurity destinations. We estimate the role of firm productivity as a determinant of firm export performance in the two groups. Our goal is to assess whether productivity impacts differently the decision to export in secure and insecure markets. The results corroborate the link between poor-quality institutions and the weakening of the role of firm productivity as a determinant of exports: productivity has less influence on the probability that a firm exports to insecure markets.

To summarise, more insecurity penalises consumers in the foreign market in two dimensions: First it reduces the number of available imported varieties, which means fewer consumption possibilities. Second, it implies a less efficient market structure of imports, as firms that export to insecure markets are not necessarily the most productive.

The second interesting feature lies in the comparison of the effect of a decrease in insecurity on the extensive and intensive margins in different types of countries. To illustrate this result, imagine that Italy and Pakistan succeeded in reducing their insecurity levels to that of their safest neighbouring country, Switzerland and India respectively. This would represent a reduction of the insecurity index of 19.3% for Italy and 16.6% for Pakistan. Relying on our results, such an improvement would increase Italian imports by 13.5%. This increase is completely channelled by the extensive margin – improving Italian security would drastically increase the number of firms exporting to Italy. In contrast, the impact on Pakistani trade would mainly be driven by the intensive margin – the mean value of a shipment would rise by 31% while the number of varieties would rise by only 12%.

Although it is a theoretically difficult exercise to provide a detailed computation of the impact of better institutions on a country's overall welfare, it is possible to assess the impact of institutions on the level of available varieties in the importing countries, as well as on the degree of local competition between firms. Improving a country’s security may improve its market structure for imports by attracting the most productive exporters and increasing the number of competitors.

References

Anderson J. E., 2000, “Why Do Nations Trade (So Little)?Pacific Economic Review 5: 115-134.
Anderson J. E. and D. S. J. Marcouiller, 2002 “Insecurity and the Pattern of Trade: An Empirical InvestigationReview of Economics and Statistics, 84: 342-352.
Blomberg S. B. and G. D. Hess, 2006, “How much does violence tax trade?”, The Review of Economics and Statistics, 88(4): 599612.
Chaney T., 2008, “Distorted gravity: Heterogeneous Firms, Market Structure and the Geography of International Trade”, American Economic Review, Vol. 98, No. 4..
Dollar D. and A. Kraay, 2002, “Institutions, Trade, and Growth”, Journal of Monetary Economics, 50: 133-162.
Francois J. and M. Manchin, 2006, “Institutions, Infrastructure, and Trade”, mimeo.
Levchenko A., 2007, “Institutional Quality and International Trade”, Review of Economic Studies, 74(3): 791-819.
Melitz M., 2003, “The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity, Econometrica, 71(6): 1695-1725.

Topics: International trade
Tags: heterogeneous firms, insecurity, institutions

Comments

implications of your crucial assumption

As you notice, the crucial assumption behind your result is that insecurity in a given export market affects potential exporters idiosyncratically ("these troubles only affect a random subset of potential exporters"). If insecurity equally affected all potential exporters, it would just be an additional market entry fixed costs, and the model would be uninteresting. There is however a third possibility, namely that insecurity affects exporters in non-random ways. Assume that insecurity is essentially corruption, you may easily have a situation in which only big, productive exporters can afford to pay corrupt officials and start exporting, while small, less productive would-be exporters cannot. The extensive margin would be reduced, but the self-selection of the most productive firms into exporting would be reinforced, rather than weakened as in your first prediction. I didn't read the research paper, but I feel that you need a very good story to convince the reader that the insecurity "troubles" affect potential exporters randomly.

Cosimo Beverelli
Graduate Institute, Geneva

the crucial assumption

As you notice, the crucial assumption behind your result is that insecurity in a given export market affects potential exporters idiosyncratically ("these troubles only affect a random subset of potential exporters"). If insecurity equally affected all potential exporters, it would just be an additional market entry fixed costs, and the model would be uninteresting. There is however a third possibility, namely that insecurity affects exporters in non-random ways. Assume that insecurity is essentially corruption, you may easily have a situation in which only big, productive exporters can afford to pay corrupt officials and start exporting, while small, less productive would-be exporters cannot. The extensive margin would be reduced, but the self-selection of the most productive firms into exporting would be reinforced, rather than weakened as in your first prediction.

Matthieu Crozet
Professor at the University of Reims and scientific advisor at CEPII.
Pamela Koenig
Assistant professor at the University of Paris X and a research affiliate at the Paris School of Economics
Vincent Rebeyrol
Max Weber fellow at the European University Institute