In early February, trade ministers from the dozen countries that make up the Trans-Pacific Partnership will gather in New Zealand to sign the massive trade pact, which covers approximately one-third of global trade. It’s then up to the legislatures of member countries to decide whether to ratify the agreement.
But parsing exactly what the treaty will or won’t do is a difficult task given how much it covers – there are 30 chapters spanning thousands of pages, covering labor, the environment, e-commerce, sanitary and phytosanitary measures, among others.
Peter Petri hopes his work will help answer that question. On a recent afternoon, he sat in front of a laptop in his office at Brandeis University in the suburbs of Boston and pulled up an economic model that estimates the impact of the TPP.
Petri, a professor of international finance, spent the better part of last year working with a handful of colleagues to update a 2012 version of the model based on estimates of what trade ministers might agree to in the TPP. They added new information and data from the agreement once its final terms were made public in early November.
Turning to his laptop, Petri tapped a red small button to start the model and demonstrate its process. A mess of characters and symbols scrolled along the screen, the computer's fan began to whine.
“It tells you that it has 108,000 rows and columns, which are equations and variables,” he explained excitedly. “It has 1.18 million data points that are not zero in that column. And it’s now solving it.”
To do that, the model draws on data about the economies of the 12 TPP countries.
“For example, that in order to make an automobile in Japan, you need this much steel, this much energy, this much rubber,” Petri said.
It establishes a baseline of economic projections through 2030 and then estimates the impact of changes the TPP will make if it takes effect, including reductions in the taxes or tariffs charged on goods as they crosses borders.
Petri scrolled to the end of a spreadsheet that contained a line for each change to the tariff schedule.
“All together it’s 5 million tariff lines,” he said.
The model also contains data on changes to so-called non-tariff barriers, the rules and regulations that can make it difficult and costly for companies to operate in foreign countries. Trying to calculate the impact of reducing those barriers is squishier, but estimates are also in the model.
About eight minutes after Petri started the model, it finished its calculations, creating a series of new files. Petri opened a spreadsheet that showed how much the TPP will add to the size of each country’s GDP and annual real incomes by 2030.
It showed Vietnam gets the largest boost to its GDP, relative to the size of its economy.
U.S. annual real incomes would rise $131 billion by 2030 and every year thereafter.
“If you look at incomes in 2030, they’re half a percent higher than they would be otherwise,” Petri explained.
While that might not sound like a significant amount, Petri believes the TPP would be worthwhile.
“Getting a half percent in a huge economy like the U.S., at modest investment, in other words, without spending $1.5 trillion dollars that one would normally spend to get that much extra income, is not bad,” he said.
However, the agreement will come with some costs.
“Workers have to move,” Petri said. “Some people will lose their jobs and find places elsewhere in the economy. Overall, the places should be there, but that doesn’t necessarily solve any one person’s problem.”
Job losses, roughly 50,000 in any given year, will mostly be concentrated in labor-intensive manufacturing, such as textiles, apparel, and some types of machine building. People who lose their jobs may have a tough time getting new ones, perhaps they’ll have to retire early or take a wage cut.
Over time, Petri expected a roughly equal number jobs created in areas like high-skilled manufacturing, financial services or food processing.
Additionally, he stressed there are elements of the TPP the model can’t accommodate. For example, its proponents hope it will fundamentally rewrite rules for global trade. While the model can try to quantify the individual impact of different pieces of these rules, that mega-impact should it succeed, Petri said the model can’t capture that.
It also doesn’t try to predict what else might change or impact the global economy in the next 15 years.
“The world will always unfold differently than what models assume will take place,” said Dani Rodrik, professor of international political economy at the Harvard Kennedy School. “So in a way, the modelers have a perfect cover.”
Rodrik’s recent book, “Economics Rules: The Rights and Wrongs of the Dismal Science,” looks at the strength and weaknesses of economic models. He said the strength for these types of model is that they can look at many different interactions across many markets.
However, “they give us a sense of precision that really doesn’t belong there,” he added. “There’s going to be a lot of uncertainty about these numbers, even for things like the effects of tariff reductions, which we ought to have a better handle on.”
Moreover, he said the different ways researchers perceive trade can shape the models they use or build.
In fact, a team at Tufts University released their own study based on their own modeling they estimate the U.S. GDP would be half a percent lower ten years after TPP takes effect.