Between the 1950s and 1980s, many developing countries engaged in limited trade policies aimed at achieving industrialization and reducing economic fragility. It was predicted on the basis of the belief that trade, through increasing specialization, would push countries into sectarian shocks, thus increasing instability. This is what we call the ‘fragility’ hypothesis. Decades go by and, with current globalization (such as the Financial Times 2022) and the backlash against trade, it is even more stressful to understand the evidence of how trade affects macroeconomic instability.
The way trade shapes instability is a more complex picture than the old macro-development argument. It will depend on what kind of impact the open economy has and how trade affects the exposure to these risks. This argument is a central aspect of Casselli et al’s earlier contributions. (2015) and Kramarz et al. (2019).1 Take a closed economy without any specialization as an extreme example. If the main sources of risk are sectoral shocks, then this economy will be better at dealing with risks and therefore reduce volatility like the argument of fragility. However, if the main sources of push are national macro-level policies, the lack of diversity would imply that the economy is more open to overall risk, as all sales are domestic. Similarly, a highly specialized open economy may become less versatile as it diversifies from the domestic market. The effect of trade on volatility depends on the complex interaction between the shocks, the symmetry between these shocks, the sectoral composition (specialization) of the output, the geographical expansion (diversification) of the sales of the sector and the co-movement between the sector and the destination market. .
In a recent study (Ardelean et al. 2022), we unpack these complex interactions using a multi-country, multi-field, multi-destination framework that we apply to a database of 34 countries, 19 sectors, and 85 destination markets. Since we can monitor the sales of each industry in a country in all its destination markets (including the home market), we can divide this sales growth into three types:2
- Push the destination: Push the destination market where the products are sold (including the home market) in the country in which they are produced independently.
- Push the origin: Shock which is independent of the specific sales destination for the producing country / industry.
- Idiosyncratic shock: The remaining shock is not explained by the above two.
Its growth rate Sum Gross output is a weighted average of the growth rate of industrial sales in each destination market. This weight will depend on two things: (1) the share of the industry in the total output and (2) the share of different destination markets for each industry. The former tells us about the degree of concentration level Specialization Economics, where the latter tells us about the degree Market diversity.
We then show that variability in the rate of growth of the overall output may be permeable between the destination risk, the source risk, the idiosyncratic risk, and the co-movement (covariance) between the sources of those risks. Each risk will be determined by the frequency and size of the shock as well as the weight. Since these weights depend on specialization and diversity, and are directly related to trade, we can analyze how changes in a country’s specialization and diversity structure will affect overall instability. That is, the sectoral and destination market structures determine the three types of risk exposure.
One advantage of this approach is that it allows us to see complex ways in which trade can affect overall instability. Consider destination risk: This risk does not depend solely on whether a country’s sales are concentrated in a market with high volatility. It also depends on whether the country’s output is concentrated in industries whose sales across the destination are subject to positively correlated shocks, and whether sales are centralized in the market with positively correlated shocks across the industry. That is, it is important to push the demand of the destination market only because you do not sell your output. It also depends on how these pushes create interactions across different destinations and across different industries. Similarly, the origin risk depends not only on whether the country specializes in volatile sectors, but also on how the sectors will co-operate. It turns out that some of these ‘indirect’ effects will predominate directly.
With this in hand, we can then run counterfactual situations about trade that allow us to measure how the structure of trade affects instability. Specifically, we run three counterfactuals:
- The opposite of total diversification: By stabilizing sectoral stocks, we change the level of destination market diversification and make it proportional to the GDP of the trading partner.
- Home Diversification Counterfactual: Same as before, but we only allow diversification of sales away from the home market.
- Specialization Counterfactual: By keeping the destination market share stable, we reduce the level of industrial specialization to make it compatible with the closed economy. The benchmark we use is the sectoral composition of the world economy.
With increased trade, we expect to increase destination market diversification and increase specialization. Thus, comparing real and counterfactual volatility allows us to measure the impact of potential volatility in trade.
We compiled a database of manufacturing and bilateral trade for 19 sectors in 34 countries by selling in 85 destination markets for the period 1981-2011.3 Figure 1 shows the total risk across 34 countries and the contribution of each risk component in 2011. Clearly, for most countries the destination risk predominates, followed by the idiosyncratic risk. Source risk is small. Importantly, coexistence between pushes is negative.
This aggregate result hides some details. For example, the most important driver of destination risk is the term related to the shock of destination shock in markets across the industry. In other words, the risk of destination is greater because countries sell intensively in the market with positively related destination shocks across the industry. This is consistent with the predominance of country-specific shocks found in empirical international business cycle literature (e.g. Kos et al. 2003). For core shocks, the largest component is the output co-movement between sectors.
Figure 1 Risk of rot in 2011
Note:: The figure shows the contribution of total output volatility and various risks in 34 countries. DR, OR, and IDIOR are destination, source, and idiosyncratic risk, respectively. Cov terms capture the symmetry between destination and source, destination and idiosyncratic, and source and idiosyncratic shocks, respectively. Each box contains the middle, inter-quadrant range and minimum and maximum without the outlier.
Figure 2 presents the total effect arising from the three counterfactuals. Diversity will reduce overall instability. Medium effects appear smaller. However, it hides many differences. For 26 of the 34 countries in the sample, diversity risks were significantly reduced. These are precisely countries with higher output volatility. A large part of this risk reduction arises from the reduction of destination market risk. As Figure 3 shows, this is highly correlated with diversification from the home market. Complete diversification adds an additional hedging process, but the diversification effect is largely driven by the home market.
An interesting feature of Figure 2 is this Reduce Specialization seems to increase the total risk. This is contrary to the conventional wisdom of the ‘fragility’ argument. The reason for the increased risk is that specialization increases co-movement between declining sectors. In other words, countries tend to specialize in low-level coordination with the rest of the economy. When this specialization is removed, the total instability increases.
Figure 2 Total risk under various counterfeitual, 2011
Note:: The figure shows the actual total average output volatility across 34 countries and the total volatility obtained under the three opposite conditions. Each box contains the middle, inter-quadrant range and minimum and maximum without the outlier.
Figure 3 Home vs. Complete Diversity, 2011
Note:: The image plots the percentage change of the home diversification view (y-axis) versus the full variation view (x-axis) by country and 45-degree line.
The comment is final
Talk about our results Possible The effects of trade instability through exposure to various sources of risk. Since sales are highly concentrated in a few markets and are dominated by the home market, increased diversification may have the effect of reducing strong volatility in countries where macroeconomic instability is high. Moreover, the effects of specialization instability do not seem to be consistent with the conventional view that specialization increases fragility.
Ardelean, A, M Leon-Ledesma and L Puzzello (2022), “Growth Volatility and Trade: Market Diversification vs. Production Specialization”, CEPR Discussion Paper 17330.
De Giovanni, J. and AA Levchenko (2009), “Trade Openness and Instability”, Economics and Statistics Review 91 (3): 558-585.
Caselli, F, M Koren, M Lisicky and S Tenreyro (2015), “Macro-diversification through trade”, VoxEU.org, 14 October.
Financial times (2022), “The WTO’s Solo Struggle to Protect World Trade”, 13 June.
Koren, M and S Tenrero (2007), “Instability and Development”, Quarterly Journal of Economics 122: 243-287.
Kose, MA, C Otrok and CH Whiteman (2003), “International Business Cycles: World, Region, and Country-Specific Factors”, American Economic Review 93 (4): 1216-1239.
Kramarz, F, J Martin and I Mejean (2019), “Idiosyncratic risks and the volatility of trade”, VoxEU.org, 11 December.
Loayza, N and CE Raddatz (2006), “Structural Determinants of External Weakness”, World Bank Research Working Paper 4089.
Rodrik, D (1998), “Why is there more government in a more open economy?”, Journal of Political Economy 106 (5): 997-1032.
1 For contributions to the role of trade in instability, see Rodrik (1998), Loayza and Raddatz (2006), and di Giovanni and Levchenko (2009), among others.
2 This digestion extends to Koren and Tenreyro (2007) because we observe not only industries and countries, but also markets.
3 For details and cross-check see: https://l-puzzello.github.io/indstat-TPP/