Are the Peterson Institute Studies Reliable Guides to Likely TPP Effects?

More on the controversy about the economic effects of the Trans-Pacific Partnership Agreement (TPPA). This commentary by guest blogger Jomo Kwame Sundaram continues the discussion begun by Jeronim Capaldo and Alex Izurieta here. The three are co-authors of the Global Development and Environment Institute (GDAE) working paper, “Trading Down: Unemployment, Inequality and Other Risks  of the Trans-Pacific Partnership Agreement,” available here. –Eds.

Are the Peterson Institute Studies Reliable Guides to Likely TPP effects?[1]

Jomo Kwame Sundaram

Professor Robert Lawrence[2] poses three seemingly reasonable questions to conclude that his colleagues’ trade policy work on the likely consequences of the Trans-Pacific Partnership Agreement (TPPA) is superior to our macroeconomic policy exercise using the United Nations Global Policy Model (GPM). In doing so, he misrepresents the nature of our work as well as the significance of the differences, thus confusing the issues.

Our study does not purport to be a comprehensive analysis of the TPPA and its effects, as his colleagues’ studies claim to be. We have not used the GPM as a trade policy model to substitute for trade projections made by Professor Lawrence’s colleagues. In fact, despite our reservations, we accepted the earlier projections by the Peterson Institute for International Economics (PIIE) on the likely trade outcomes of tariff reductions as our starting point, but proceeded to show that their claims about the TPP’s likely growth and employment effects were problematic and did not necessarily follow.

Our model incorporates macro-financial dynamics as well as distribution and employment variations, which more realistically represent the world economy rather than their self-equilibrating, full employment model. But in line with the limited scope and purpose of our paper, we do not claim to have provided reliable and definitive projections of the TPP’s likely effects.

As is well known, the United States Department of Agriculture’s Economic Research Service (USDA-ERS) computable general equilibrium (CGE) model[3] projections of the TPP are far more pessimistic than the PIIE’s. If the differences between our results and the findings of the PIIE were simply due to using different models, then one would have to explain how the USDA-ERS model came to its very different conclusions. Clearly, CGE models can come to very different findings using different, in this case, more realistic specifications.

Crucially, we noted that by assuming full employment, the PIIE model assumes away important macroeconomic implications of the TPP. We show that using the PIIE trade projections, but with more realistic modelling, assumptions and specifications, the TPP would result in more modest growth, net job losses, greater pressure on wages and a declining labour share of income. These effects may well be eventually offset by other developments having nothing to do with the TPP, but they nevertheless remain the likely effects of the TPP.

Professor Lawrence also makes much of the fact that the GTAP has made available a template for CGE modelling available to all. Undoubtedly, the well-resourced GTAP has done a great deal to promote CGE modelling in various ways. But the underlying equations and other details of the GPM are also readily available[4] to all those interested although they have hardly been promoted in the same way.

More importantly, it is misleading to give the impression that the PIIE studies are fully transparent, in contrast to ours. As Professor Lawrence knows, modelling exercises are not spreadsheet exercises. Inevitably, there are many details of all complex computable modelling exercises, involving myriad judgements of various types, which are never fully disclosed, owing to their nature.

How does the new PIIE modelling exercise come to even more optimistic findings than the earlier one? Total growth due to the TPP is projected to rise 0.5% after 15 years — compared to the earlier projection of 0.4% over ten years — by allowing more time for implementation and for the growth effects to set in. Adding data for 2008-2011 may also have driven up growth owing to the effects of the 2008 financial crisis and the ensuing stimulus efforts in those years. Conversely, the post-2011 withdrawal of various stimulus packages, which would limit growth effects, is not considered.

But the most important contributions to growth in the PIIE studies come from their treatment of so-called non-trade measures (NTMs), i.e. “growth through the back door,” as Professor Lawrence’s Harvard colleague, Dani Rodrik has put it. In essence, these are little more than contrived growth effects from other TPP features, not core results using established CGE modeling methods.

Thus, while even the PIIE exercises acknowledge modest growth gains from tariff reductions, most of the growth they project comes from their claims about the growth impacts of the many NTMs in the TPPA. Each of these NTMs are treated as equivalent to further tariff reductions using methodologies which have no bases in either theory, evidence or past experience.

For example, strengthening intellectual property rights (IPRs) is projected to promote growth by enhancing the trade in services. The cost implications of stricter enforcement and more expensive medicines for consumer welfare, public health, government budgets, etc. are hardly considered. It is also presumed that additional rents to the firms concerned and the shift of income from labour to capital will be reinvested for further growth in TPP economies.

Thus, the PIIE studies correctly draw attention to the implications of the TPP’s NTMs, which have not been subject to careful analysis anywhere. Even though the PIIE studies’ treatment of NTMs remains unacceptable to me, such matters, not its relatively paltry trade effects, should be central to analysis of the TPP. Our work only sought to inject a greater dose of realism into understanding some likely macroeconomic implications of their TPP trade growth projections.

Professor Lawrence may well find the PIIE studies’ novel innovations to boost the TPP’s growth effects acceptable and satisfactory, but this involves an expression of faith, or perhaps, of colleagial solidarity or loyalty. This should be made explicit, instead of comparing apples with oranges, inadvertently misrepresenting the nature and significance of our differences.

[1] This commentary complements one by my coauthors, Jeronim Capaldo and Alex Izurieta. See “Modeling TPP: A Response to Robert Z. Lawrence”. GDAE Globalization Commentaries, February 3, 2016; http://ase.tufts.edu/gdae/Pubs/rp/GC96Feb16CapaldoIzurieta.pdf

[2]  Robert Z. Lawrence, “Studies of TPP: Which is Credible?” http://blogs.piie.com/trade/?p=553

[3] There are well-known critiques of CGE models as well as of their underlying general equilibrium premises. See Lance Taylor and Rudi von Arnim, 2006. Modeling the Impact of Trade Liberalization. Oxfam International, Oxford.

[4] See Francis Cripps and Alex Izurieta, 2014. The UN Global Policy Model: Technical Description. United Nations Conference on Trade and Development, Geneva.

Triple Crisis welcomes your comments. Please share your thoughts below.

Triple Crisis is published by