Bryan Mercurio on Estimating the Benefits of FTAs

Guest blogger Bryan Mercurio writes the following:

Governments often commission and rely on economic analysis to publicly justify FTA negotiations, but it seems to me that such analyses are becoming increasingly meaningless and even deceptive, at most serving only as a 'best case' scenario and often being used to justify a decision that has already been made.

The latest example of this is the recent release of the preliminary results of an Australia-Japan scoping study that finds a comprehensive FTA between Australia and Japan would be worth at least $38.7 billion to Australia over two decades and $27.4 billion to Japan (http://www.dfat.gov.au/geo/japan/tef-study/faq.html ). Based on this information (or so it has declared), the Australian government has expressed a readiness to begin negotiations.

But what appears deceptive to me is the fact that the scoping study uses a model based on almost complete liberalisation of all markets, including agriculture and services. This is mere fanciful thinking given Japan's long history of, among other things, keeping out agricultural imports by protecting sensitive products and maintaining 'food security'. Japan has already attempting to take such items off of the negotiating table. At best, Australia can only hope to match the deal Mexico negotiated (and I am not even sure that Australia's negotiating position will be as strong as Mexico's).

My second 'concern' regarding scoping studies/economic modelling is that they are conducted by groups hired by the governments with an implicit (and sometimes explicit) mandate of the finding benefits outweigh the detriments. Leaving aside the inherent difficulties with predicting gains in IP and services, the fact the governments go through the motions without an interest in actual gains/losses seems like a wasteful effort designed to justify a decision that has well and truly already been made. Making matters worse, the group which is always hired to conduct such studies in Australia are former diplomats with close ties to the government and, in what would seem to be a conflict of interest, have been or are prominent members of the pro-US, China, Japan and Malaysia FTA FTA lobbies.

I don't have a problem with FTAs or even this particular partnership. In fact, given the state of multilateral negotiations and meteoric rise in FTAs (and their consequences), Australia should negotiate this deal (to be sure, a comprehensive deal would benefit Australia. Japan is Australia's largest market for agricultural products, coal, LNG, aluminium, LPG and crude petroleum. It is the second largest source of tourists to Australia.). I do, however, have a problem with the process of this and many other FTAs.

Looking more broadly, have there been any studies looking at the predicted gains of FTAs (at both the start of the negotiations and conclusion) with the actual realised gains and analysing why the figures did not match?

Comments

[Joel P. Trachtman]

Thanks, Bryan. By the way, Ben Muse has a post referring to critiques of the use of CGE models in estimating the benefits or detriments of liberalization. See http://benmuse.typepad.com/ben_muse/2006/08/estimating_the_.html

[Peter Gallagher]

I think this criticism is a bit gratuitous, Bryan.

What would you consider a more 'realistic' basis for estimating the impact of a 'free trade' agreement? 'Partly free trade'? Would it really be more valuable to model some arbitrarily limited outcome? Where/how would you establish the limits in advance of the negotiations? And why would you expect the Australian government to publish such an estimate?

A 'partially free' outcome would be contrary to the Australian government's obligations under Article XXIV of GATT. Also, at the start of a negotiation with a trading partner an official scoping study based on a partial result might be considered to be 'leading with your chin'.

Before you launch into ad-hominem arguments about the bona fides of the economists who created the models, you might want to think a bit more carefully about what their clients (the Australian government) are trying to achieve with this 'joint study' and what role such models of ambitious outcomes might play, not only in revealing some things about the structure of a better 'integrated' Japan-Australia economy but also in the political economy of trade negotiation.

[Bryan Mercurio]

You missed the point, Peter.

My point is that the public should not rely on preliminary economic analysis based upon full liberalisation models as expected gains.

Of course, a government should undertake an economic analysis to see the possible economic benefits (and detriments) of an FTA. They should also start with a full liberalization model. They should also, as the negotiations proceed, at least privately revise the model to anticipate reasonable gains. This is undisputed.

My objection comes when the initial studies based on full liberalisation is sold to the public as the expected (emphasis) gains. All the newspapers have sold it in such a manner, as has the Minister for Trade. The Australian government also states on its website:

'The model with the most conservative estimate concluded that in the next twenty years:

• The present value of GDP gains to Australia would be $A38.7 billion, with consumption gains (as an indicator of welfare gains) of $A19.4 billion.

• Japan's GDP gains would be worth $A27.4 billion (around ¥2.2 trillion), with consumption gains of $A68.3 billion (around ¥5.6 trillion).

By 2020, Australia's GDP would be 0.7 per cent greater than would otherwise be the case, and Japan's would be 0.03 per cent larger.'

I also note that in the one page link I provided in my intial post, the government quotes these figures (by stating Australia/Japan’s economy will be ‘boosted’) three times. Three times in one page! It seems to me the government is selling these figures as the actual expected gains. I stand by my post that such statements are deceptive.

Additionally, in the process of packaging and marketing the scoping study as expected (emphasis) gains, the government also implicitly sells the study as unbiased and objective. It seems some countries do undertake objective analysis, the GAO in the US for instance. Peter seemingly admitted that some studies are not conducted at arms-length when he acknowledged that the individuals conducting the study take their ‘clients’ interests and direction into account. This, coupled with the fact that most studies are packaged as jointly authored by individuals from both parties, strengthens my point that these studies should not be relied upon as expected gains.

Finally, the point regarding Article XXIV is off-point and borderline ridiculous. What does Article XXIV have to do with a priminary scoping study when the final FTA will look nothing like the model used? Moreover, everyone reading this knows full well that Article XXIV is unclear (substantially all trade), unenforceable (zero condemned FTAs in the GATT?WTO due to consensus decision-making) and unstoppable regardless if FTAs technically comply (see all of the FTAs which exclude entire sectors, such as services).

Lastly, the question posed in my original post is a serious one and I would be keen to receive any information and tips.