Representatives of 12 Pacific Rim countries, including the United States, on Wednesday signed the sweeping Trans-Pacific Partnership trade deal, a treaty that covers 40 percent of the global economy. While it’s famously difficult to project the effect a trade deal has on jobs, that hasn’t stopped some researchers from trying.

Depending on which school of thought you prefer, the TPP might have no effect on employment levels in the United States — or it could lead to a staggering loss of 448,000 American jobs over a period of a decade.

Peter Petri, a professor at the Brandeis International Business School and visiting fellow at the Peterson Institute for International Economics, a prominent Washington-based think tank, belongs to the former camp. Petri co-authored a paper last month that found, in addition to modest wage gains overall, the agreement will not alter U.S. employment levels.

Indeed, that’s what classical economic theory says about trade deals. Competitive industries flourish, less competitive industries struggle, and in the end the job gains and losses even out.

“Some industries grow, and some industries don’t grow as fast they would otherwise under a trade agreement,” Petri said. In that process, “some people have to leave jobs and find other jobs.”

In the case of TPP, the churn amounts to more than 150,000 jobs, by Petri’s estimates.

“The numbers aren’t very large, but the effects on individuals and possibly the communities around them that sort of don’t make it, that don’t have the skills or are geographically constrained ... there the costs can be pretty serious,” he said.

Under TPP, Petri projects the U.S. manufacturing sector will see a decline in employment growth. That, in turn, could translate into incredibly painful times for some: A laid-off steelworker without a college degree, for example, is going to find it much more difficult to land a job in one of the faster-growing sectors such as IT than, say, a recent college graduate who already lives close to an industry hub.

In other words, some workers won’t be able to make the transition. But others will rush to fill the new, higher-paying jobs and, ultimately, employment levels will balance out.

Another recent paper tackled the TPP jobs question from an entirely different perspective  and came to wildly different conclusions. In research published last month by the Global Development and Environment Institute at Tufts University, authors Jeronim Capaldo and Alex Izurieta projected the TPP would lead to job losses of 448,000 in the United States, from 2015 to 2025.

The authors used what’s known as the United Nations Global Policy Model, which unlike the classical model does not assume TPP trading partners are at full employment.

“From the starting gate, they’re beginning from two very different assumptions,” said Josh Bivens, research and policy director at the Economic Policy Institute, a left-leaning think tank, also based in Washington.

Under Capaldo’s model, TPP countries face demand constraints. By redistributing demand from countries that are far from full employment to countries that are close to full employment, the trade deal can lead to job loss. According to the research, the agreement translates into 771,000 overall job losses among signatories a group that includes Canada, Japan, Australia, Mexico and others.

“The TPP favors competition on labor costs and remuneration of capital,” Capaldo and Izurieta concluded. “Depending on the policy choices in non-TPP countries, this may accelerate the global race to the bottom, increasing downward pressure on labor incomes in a quest for ever more elusive trade gains.”

Some economists have recoiled at the findings. In response to similar research that measured the jobs impact of the proposed United States-European Union free trade agreement, critics charged the model used by Capaldo isn’t designed to measure things like trade deals.

Still, Bivens stressed any such employment projections are an uphill battle from the start. It’s hard to isolate the effects of trade deals from the broader state of the economy. “Any attempt to model a trade agreement is a heroically ambitious thing to do,” he said.

While it's nearly impossible to measure its direct effects on employment levels, the last major trade deal inked by the U.S. soured many on the notion of free trade. Although it did not specifically isolate the North American Free Trade Agreement (NAFTA) — a treaty that took effect in 1994 — recent research found the United States’ growing trade deficit with Mexico has displaced nearly 700,000 American jobs.

Critics have pointed to American automakers' recent pivot to Mexico as a prime example of the deal's downside. 

Petri said increased productivity among foreign competitors and technological advances help account for the “very difficult period of adjustment” experienced by workers in the United States over the last 20 years. Decreased trade barriers have probably played some role too, he acknowledged.

“That has to have an effect,” he said. “I’m not sure what we can do about it. I think the theory tells us if we react to that with protectionism, we’ll just make ourselves worse off.”

The U.S. Congress still has to ratify the Trans-Pacific Partnership. Last year, it agreed by a small margin to grant the Obama administration so-called fast-track authority. As a result, federal legislators are now limited to a simple yes-or-note vote and cannot amend the treaty. The vote, which business leaders say might not happen until next year, is expected to be close.