Info Service on WTO and Trade Issues (Nov14/06)
may lead to EU dis-integration, unemployment, instability - Part 1
Medford, MA, USA, 17 Nov (Jeronim Capaldo*) -- The European Union and the United States are currently negotiating the Trans-Atlantic Trade and Investment Partnership (TTIP), a major trade agreement intended to further integrate their economies.
In today's low-tariff reality, TTIP focuses on removing non-tariff trade barriers between countries, such as differing standards set in the EU and in the US for given consumer goods and services.
The underlying logic is the same as in traditional liberalizations: reducing the costs of trade - whether eliminating tariffs or other impediments - is supposed to lead to a higher trade volume and overall economic benefits.
Unfortunately, experience has shown that this appealing reasoning is often misleading.
As is common for trade agreements, TTIP negotiations have been accompanied by a series of econometric studies projecting net economic gains for all countries involved. In the EU, advocates have pointed to four main studies mostly projecting small and deferred net benefits alongside a gradual substitution of intra-EU trade with Trans-Atlantic trade.
This leads the European Commission, TTIP's main advocate in Europe, into a paradox: its proposed policy reform would favour economic dis-integration in the EU. TTIP might also lead to other serious consequences for the EU and its members. Recent literature has shown that the main studies of TTIP do not provide a reliable basis for policy decisions as they rely heavily on an unsuitable economic model.
Assessing TTIP with the United Nations Global Policy Model, a model based on more plausible assumptions on economic adjustment and policy trends, we found very different results. Evaluated with the UN model, TTIP would lead to net losses in terms of GDP, personal incomes and employment in the EU.
In particular, we project that labour incomes will decrease between 165 Euros and 5,000 Euros per worker depending on the country. We also project a loss of approximately 600,000 jobs, a continuing downward trend of the labour share and potentially destabilizing dynamics in asset prices.
Our projections point to bleak prospects for EU policymakers. Faced with higher vulnerability to any crises coming from the US and unable to coordinate a fiscal expansion, they would be left with few options to stimulate the economy: favouring an increase of private lending, with the risk of fuelling financial imbalances, seeking competitive devaluations or a combination of the two.
We draw two general conclusions. First, as suggested in recent literature, existing assessments of TTIP do not offer a suitable basis for important trade reforms. Indeed, when a well-reputed but different model is used, results change dramatically.
Second, seeking a higher trade volume is not a sustainable growth strategy for the EU. In the current context of austerity, high unemployment and low growth, requiring that economies become more competitive would further harm economic activity. Our results suggest that any viable strategy to rekindle economic growth in Europe would have to build on a strong policy effort in support of labour incomes.
Most assessments of TTIP predict gains in terms of trade and GDP for both the EU and US. Some also predict gains for non-TTIP countries, suggesting that the agreement would create no losers in the global economy. If this were the case, TTIP would be the key to a more efficient allocation of global resources, with some countries achieving higher welfare and all others enjoying at least the same welfare as before.
Unfortunately, as Raza and colleagues (2014) have shown, these desirable results rely on multiple unrealistic assumptions and on methods that have proven inadequate to assess the effects of trade reform.
Furthermore, once the calculations are reviewed, it appears that several of these studies share the same questionable economic model and database. The convergence of their results is, therefore, not surprising and should not be taken as providing independent confirmation of their predictions.
Quantitative arguments in favour of TTIP come mostly from four widely-cited econometric studies: Ecorys (2009), CEPR (2013), CEPII (2013) and Bertelsmann Stiftung (2013).
CEPR has been very influential: the European Commission has relied on it as the main analysis of the economic effects of TTIP going as far as presenting some of its findings as facts.
However, the EC's reference to CEPR as an "independent report" seems misleading since the study's cover page indicates the EC as the client for whom the study has been produced. Ecorys was also commissioned by the EC as part of a wider project encompassing economic, environmental and social assessments (Ecorys, 2014).
Methodologically, the similarities among the four studies are striking. While all use World Bank-style Computable General Equilibrium (CGE) models, the first two studies also use exactly the same CGE.
The specific CGE they use is called the Global Trade Analysis Project (GTAP), developed by researchers at Purdue University. All but Bertelsmann use a version of the same database (again from GTAP).
The limitations of CGE models as tools for assessments of trade reforms emerged during the liberalizations of the 1980s and 1990s. The main problem with these models is their assumption on the process leading to a new macroeconomic equilibrium after trade is liberalized.
Typically, as tariffs or trade costs are cut and all sectors become exposed to stronger international competition, these models assume that the more competitive sectors of the economy will absorb all the resources, including labour, released by the shrinking sectors (those that lose business to international competitors).
However, for this to happen, the competitive sectors must expand enough to actually need all those resources. Moreover, these resources are assumed to lack sector-specific features, so they can be re-employed in a different sector.
Under these assumptions, an assembly-line employee of an automobile factory can instantly take up a new job at a software company as long as her salary is low enough. Supposedly, this process is driven by speedy price changes that allow an appropriate decrease of labour costs and, consequently, the necessary expansion of the competitive sectors.
In practice, however, this "full employment" mechanism has rarely operated. In many cases, less competitive sectors have contracted quickly while more competitive ones have expanded slowly or insufficiently, leaving large numbers of workers unemployed. One need only look at the experience of Europe in the last decade to see that full employment does not re-establish itself even if job seekers are willing to work informally and at relatively low pay.
A critical point is that the distribution of gains and losses is rarely uniform within economies. If workers in competitive sectors may benefit from higher salaries, while those in shrinking sectors lose, the economy as a whole may be worse off.
This is because in some countries domestic demand is mostly supported by the incomes earned in traditional occupations. In practice, aside from their high social costs, these transitions have led to a drop of domestic demand that CGE-based calculations have often overlooked.
Moreover, most CGEs rely on misleading assumptions on the pattern of international trade, imposing a fixed structure on the market share that each country has in its export markets, and on a static analysis that does not explain how economies reach a new equilibrium.
For example, when Country A expands trade with Country B, the rest of the world's economies do not simply stand still. Countries C, D and E will find that they are more or less competitive in these markets as a result of the A-and-B trade changes. This effect is known as "trade diversion", and has been a significant by-product of recent trade integration initiatives.
Finally, the strategy chosen to simulate a "TTIP future" has a strong impact on the results. Ecorys assumes that so-called "Non-Trade Barriers" impose a given cost on trade and that TTIP can remove up to one half of them.
CEPR and CEPII borrow this approach, but assume a lower share. These barriers can include what other stakeholders refer to as consumer and environmental regulations. Phasing them out may be difficult and could impose important adjustment costs not captured by the models.
All assessments project large increases in bilateral US and EU exports. In CEPR and CEPII, US bilateral exports increase by 36.6 percent and 52 percent respectively in the long term, compared to 28 percent and 48 percent for the EU. According to CEPR, the net increase in total exports will be 8 percent in the US and 5.9 percent in the EU.
However, in all cases, these increases in trans-Atlantic trade are achieved at the expense of intra-EU trade.
Implicitly, this means that imports from the US and imports from non-TTIP countries through the US will replace a large portion of current trade among EU countries.
If these projections were true, higher trans-Atlantic interdependence would heighten the EU's exposure to fluctuations in US import demand. This is an under-examined consequence of certain patterns of trade liberalization.
Even if higher exports were to bring higher demand and economic activity (a link that doesn't always work in practice, as discussed), more reliance on the US as an export market would also make the EU vulnerable to macroeconomic conditions in North America.
If Europe could effectively implement countercyclical policies, this greater interdependence would not necessarily be a problem.
However, the EU's current institutional structure lacks a central fiscal authority while in practice preventing national governments, through the Maastricht treaty, from implementing any fiscal expansion. This constellation of factors indicates that the TTIP might usher in a period of higher instability in Europe.
The remaining two studies raise similar concerns. In Bertelsmann, aggregate figures for bilateral export increase and net increase are not readily available but results exhibit the same pattern as in other studies.
While bilateral exports are predicted to increase by more than 60 percent for the EU and more than 80 percent for the US, intra-EU exports are expected to decrease between 25 and 41 percent. This implication raises the same concerns about vulnerability to US economic shocks as the other studies.
Finally, as noted above, the rest of the world does not stand still when two economies integrate. Applying Bertelsmann's percentages to recorded trade data with EU exports to the world as a whole, Raza et al. (2014) calculate that the overall impact of TTIP on EU global exports, including those to non-TTIP countries, would be negative.
Furthermore, Felbermayr and Larch (2013) find that TTIP will have a negative effect on non-TTIP countries' exports, in a pattern observed after other trade agreements. In other words, both exports and imports of non-TTIP countries are projected to decrease, with uncertain or negative net effects.
CEPR and CEPII do not find negative effects on non-TTIP countries assuming ad hoc effects (spill-overs) that allow exports in the rest of the world to grow.
GDP AND PERSONAL INCOMES
Given the small net effects on exports, most assessments predict small increases in TTIP countries' GDP.
In Ecorys, CEPR and CEPII, GDP increases less than 0.5 percent in both the EU and US. This means that, at the end of the simulation period in 2027, GDP would be 0.5 percent higher in a TTIP scenario than the baseline, non-TTIP scenario, implying negligible effects on annual GDP growth rates.
This is a defining aspect of the results: Ecorys, CEPR and CEPII point to a one-time increase in the level of GDP, not to an increase in the growth rate of GDP. Furthermore, this one-time increase is small and projected to occur only over the course of 13 years.
Bertelsmann reports higher figures (5.3 percent for the EU and 13.9 percent for the US), but provides little detail on the study's methodology. It is, therefore, unclear how the results compare to those of other studies.
Furthermore, given the assumptions on spill-over effects, CEPR estimates that all regions of the world would benefit from long-term GDP increases. However, Felbermayr and Larch (2013) indicate that this expectation contradicts previous experiences of trade agreements such as CUSFTA, NAFTA and MERCOSUR since these agreements typically affect the relative trade prices between members and non-members.
Despite the small projected increases in GDP, some studies suggest that TTIP might lead to large increases in personal incomes in the long term. In often-cited examples, Ecorys estimates that the average EU household would gain 12,300 Euros over the work life of household members, while CEPR estimates that the same household would earn 545 Euros more every year.
However, as noted above, these estimates are misleading since the studies provide no indication of the distribution of income gains: they are simply averages. With EU wages falling as a share of GDP since the mid-nineties, it is far from certain that any aggregate gains will translate into income increases for households living on income from wages (as opposed to capital).
Finally, most studies are not informative on the potential consequences of TTIP on employment. While CEPII does not discuss employment effects, CEPR and Ecorys (2013) assume a fixed supply of labour. This amounts to excluding by assumption any consequences of TTIP on employment - wages are assumed to fall or rise enough to ensure that all workers remain employed regardless of the level of economic activity.
On the other hand, Bertelsmann predicts that TTIP will lead to the creation, in the long term, of approximately one million jobs in the US and 1.3 million jobs in the EU. However, these positive figures are strongly dependent on the period chosen in the estimation.
Using data up to 2010, the authors estimate that economies where labour and labour income are more protected (for example, by higher unemployment benefits) suffer from higher unemployment, concluding that any cost reductions introduced by TTIP would lead to positive employment effects in those countries.
When more recent data is taken into account, this conclusion ceases to hold since all countries - not just those with stronger labour protection - appear to have experienced higher and persistent unemployment.
[* Jeronim Capaldo, Research Fellow at the Global Development and Environment Institute (GDAE) of Tufts University, contributed this first part of a two-part article. The second part will appear in the next issue of SUNS.] +