By Francois Fouche
Income, Welfare and Trade Effects
The welfare effects of preferential trade arrangements are theoretically ambiguous (Krugman 1991; Limao 2016).
This is the case even when the arrangements do not affect members’ terms of trade and when there are no other distortions.
- Lower trade barriers allow countries to expand trade within membership (trade creation), which increases welfare.
- At the same time, the removal of trade barriers among members may re- duce trade with more efficient countries outside the membership (trade diversion), thus reducing welfare.
The impact of reducing trade barriers on efficiency and welfare depends on whether trade diversion or trade creation dominates.
When considering terms-of-trade effects and second-best scenarios (that is, includ- ing additional distortions) the assessment is further complicated — both terms-of- trade movements and the presence of other (non-trade-related) distortions can reinforce or offset welfare gains from in-
creased efficiency owing to trade liberalization. Whether preferential trade ar- rangements increase or decrease welfare is therefore an empirical question.
General-equilibrium-based empirical studies typically find that preferential trade ar- rangements have a positive effect on welfare, with the size of the gains depending largely on initial conditions (for example, level of trade barriers and the extent of trade liberalization).
Empirical studies find that the AfCFTA would increase overall income and welfare for the majority of African countries.
These studies are based on ex ante simulations using multi-sector, multi-country computable general equilibrium (CGE) trade models. They consider both import- tariff elimination and NTB reduction. The size of income and welfare gains varies, depending on the specific features of the model: some studies (Chauvin, Ramos, and Porto 2016) find income gains of up to 5%.
Virtually all studies find relatively small gains from the elimination of import tariffsand significantly larger gains from reduction of NTBs.
These results are intuitive since, as previously discussed, intraregional tariffs in Africa are low, while NTBs are large. These studies also show that gains vary greatly across countries, but most coun- tries see their welfare rise as a result of the reduction in trade barriers, especially NTBs.
Recent work at the IMF has used newer, quantitative trade models to estimate the AfCFTA’s potential welfare effects.
Abrego and others (2019) find very small welfare gains from import tariff elimina- tion and much larger gains from a moderate reduction in NTBs.
In their baseline simulations (perfect competition, with a 35% reduction in NTBs, there is an overall welfare gain of 2.6% for sub-Saharan Africa (SSA) and 2.1% for the continent as a whole. As is stan- dard in general equilibrium trade models, the reduction in tariffs and NTBs affects welfare both through consumption and output (by reducing distortions) and thereby improves efficiency. A notable feature of the model used by Abrego and others (2019) is that international prices are endogenous, allowing for changes in the terms of trade. Therefore, in addition to changes in efficiency, terms-of-trade movements also affect welfare outcomes, and this has a material impact on the distribution of welfare gains across countries. The authors also find that the size of over- all welfare gains for the continent is very similar under imperfect competition, sug- gesting that, overall, scale effects are not very strong in Africa. Global welfare also increases, albeit modestly, under the various scenarios considered. Importantly, the restrictive features of the model used in Abrego and others (2019) likely underestimate the welfare benefits from the re- duction in trade barriers. In particular, the model is static and therefore does not include capital accumulation, innovation, or knowledge diffusion effects (which can be large over time) that may arise from in- creased trade openness. The model does not consider intermediate inputs in pro- duction either.
This study also reveals that while all African countries experience increased welfare from a reduction in NTBs, these benefits are unequally distributed.
By far, the largest proportional gains go to countries with the most open economies, which also tend to be the smallest economies. In contrast, the larger economies are less open and tend to ben- efit less, because of adverse terms-of- trade movements, which offset efficiency gains, among other reasons. Unsurprisingly, welfare gains also tend to rise with the extent of trade-barrier reduction. From a sectoral perspective, the study also reveals that tradeable sectors, partic- ularly manufacturing, which accounts for over 60% of welfare gains, and agricul- ture – 16% – are the key drivers of estimated welfare changes for the vast majority of countries, particularly for the smallest economies. The distribution of welfare gains differs across world regions, with some regions seeing their welfare decline, albeit very modestly, suggesting that they are adversely affected by trade diversion or terms-of-trade changes.
The AfCFTA would have a strong impact on intra-African trade volumes, but its effects on overall trade would be limited. Abrego and others (2019) estimate intra- African trade growth of more than 80% owing to the AfCFTA, which represents an increase of about US$60 billion in African exports. Trade growth is similar under the different model structures (perfect and imperfect competition) considered in the International trade has often been seen as a key factor behind the increase in inequality in recent decades, although this is not necessarily supported by empirical studies.
It is well-known from trade theory that, in addition to the level of income, trade can also affect income distribution, notably among factors of production. While in principle undesired distributional effects from trade could be offset by compensating losers, compensation is difficult to implement in practice, and it rarely happens. The recent international backlash against free trade reflects to a large extent the perceived key role of trade in increased inequality. Empirical work conducted over the past decades has often focused on the impact of trade policy changes on the distribution of income between skilled and unskilled labour, or the so-called skill premium (defined as the difference in the wages of skilled and unskilled workers).
However, there is broad consensus among economists, arising from this work, that the increase in wage inequality seen in many countries in recent decades has been driven mainly by skill-rewarding technical change, with trade playing only a small role (see, for example, Pavcnik 2017; Helpman 2016).
Some of the empirical literature finds that the contribution of trade liberalisation to inequality is context specific. Pavcnik (2017) shows that the nature of trade policy changes, trade patterns, and the level of mobility of workers and capital are key determinants that may affect inequality. Some empirical studies find that the negative effect on inequality may dissipate in the long run, but the adjustment process can take a long time and feature strong localised and sectoral effects (Autor and others 2014; Dix, Carneiro and Kovak 2017).
Studies on the impact of the AfCFTA on inequality are inconclusive.
Using a traditional CGE model and a sample of 6 African countries, Chauvin, Ramos, and Porto (2016) find large heterogeneity across countries. For instance, the AfCFTA appears to benefit richer households in some countries and poorer households in others, and rural households gain more than their urban counterparts in certain countries. Model-based simulations in the Regional Economic Outlook (IMF 2019) also show a heterogeneous impact of the AfCFTA on inequality across countries — income inequality decreases in poor countries that export agricultural goods and increases in countries that export manufacturing goods because of a higher skill premium study. However, given that the initial level of these trade flows is modest, the conti- nent’s total trade grows only by 8%. The study reports considerable variation of results among African countries, with those facing initially higher trade barriers generally showing stronger trade growth. Econometric analysis at the firm level for African countries also shows that the reduction in trade barriers has a positive effect on firms’ decisions to export.
*****
Source: The African Continental Free Trade Area: Potential Economic Impact and Challenges: IMF Staff Discussion Note No. SDN/20/04; May 2020