The factory in Bangladesh. (A.M. Ahad/AP)

Kimberly Ann Elliott is a senior fellow at the Center for Global Development and an expert in international trade policy, with a particular focus on labor standards and trade as a tool for fighting global poverty. She served as chair of the National Advisory Committee on Labor Provisions of U.S. Free Trade Agreements at the Department of Labor from 2011 to 2012, and from 2009 to 2012 served on the USDA Consultative Group on the Elimination of Child Labor in U.S. Agricultural Imports.

In light of the tragic collapse of a factory in Bangladesh, killing at least 87 people, and the ensuing debate about labor standards in developing countries, we spoke on the phone Thursday. A lightly edited transcript follows.

Dylan Matthews: Just to start off, what's the state of labor standards in most developing countries doing these kind of export-based sectors? Are the laws there, are they enforced, etc.?

Kimberly Ann Elliott: A lot of developing countries have laws on the books that are up to international standards but enforcement is a problem. In this particular case, apparently it was so bad that inspectors had gone in and said, "This building is dangerous, you need to evacuate." Several other businesses in the building heard that and did, but the garment factory owners there said, "Oh, it's fine, you need to go to work." So they had both the laws and the enforcement, which at least was there in theory, but these guys ignored it with very tragic results.

 I'd imagine in a lot of countries you don't even have that level of inspection, whether or not it's ignored.

Again, I don't know the exact details of the inspection capability in Bangladesh. But look at our country, right? Take this fertilizer plant outside Waco. It's storing a known dangerous product, and it had barely been inspected. When things had been found wrong, it got very small fines and went along its merry way. And that's in a very rich country.

It's a combination of capacity, which is very low in a country like Bangladesh, and incentives to not inspect. A lot of things do slide through due to a lack of will because Bangladesh has its whole export economy riding on apparel, and it's clearly a strategy of the Bangladeshi government to succeed on low wage exports.

How expensive would it be to adopt basic improvements? The main argument against requiring them, of course, is that it'd make production too expensive and drive businesses out of the country.

In health and safety, just focusing on that area for a moment, there's a wide range of things that affect health and safety, and a lot of them would be really very cheap to fix. On the one hand, doing a building right is probably pretty expensive, but also pretty important. There are some costs you have to find a way to bear. But there are a lot of things that aren't like that. When these international inspectors go in, a lot of it is, "Is there a fire extinguisher and does it work?" Or the locking of doors. I don't know the numbers of how bad theft is in these factories, but unlocking the door isn't exactly expensive. There's a lot that can be done that doesn't add that much to costs.

Workers in a shoe factory outside Hanoi, Vietnam. (Kham/Reuters)

So why don't companies do it, if it's so cheap?

I think, at that level, the competition is so fierce that there is a collective action problem where the owners aren't willing to bear even the smallest costs. There's also a collective action problem on the other side, with the buyers, who are looking for the cheapest possible price for the product and aren't willing to raise that price a bit if their competitors aren't. It feeds into a vicious cycle. That's why I think you need some kind of external intervention in terms of unions, technical assistance for a stronger inspectorate, a stronger ILO, as, on their own, the companies aren't going to be able to overcome that competitive collective action problem.

Are there any countries that have adopted stronger standards and managed to maintain strong growth? Is there a precedent here?

There are some examples. Cambodia, back when we had the Multi Fibre Arrangement and the network of global quotas on apparel trade, agreed to raise its labor standards, and there was an ILO monitoring mechanism. There are still plenty of problems with Cambodia, but there are unions now, and there is some data coming out of the factory monitoring, which my colleague Raymond Robertson has written about, where companies have improved and are still doing fine.

In fact, there's a human resource management argument here for better conditions and labor standards improving productivity, and therefore improving the bottom line. There's a project just started up by the ILO in Vietnam where they're trying to be more rigorous in comparing better-work factories with similar factories not in the program, and get even better data than what's come out of Cambodia. The work that has been done does not find a negative correlation, and maybe even a positive one. It depends on the circumstances in which they're being done, but there is a business case for treating your workers well.

A lot of that case is built around the idea that healthier, more motivated workers are more productive workers. That's true in a lot of factory jobs. But in apparel it’s so low -skill, so low-wage, that they don't care much about retention. And that's a case where factory owners don't have an incentive to raise standards because when you fire a worker, there will be five applying the next day.

What's motivating this from Vietnam? Does the government just have a desire of its own to improve standards?

This program I'm talking about, the Better Work program, it could have been motivated by Vietnam knowing they were going to negotiate the Trans-Pacific Partnership. I would suspect that might have had something to do with them wanting to work with the ILO to help their labor conditions.

Again, the core of Better Work is this idea that improving labor conditions can be better for business. It's monitoring but it's also technical assistance and capacity building to help the businesses get more productive. The idea is for it to be a win win win.

There's a program that was started in Jordan, after the scandal there in the treatment of migrant workers, and there is one in Haiti, which was a condition of the increased preferential access they received, both before and after the earthquake. So there are a number of these programs.

In Cambodia, they had this kind of program before the global quota system expired in 2005, but they decided to continue with it even though there would be no more incentive from the United States in the form of increased market access. It was part of a marketing strategy. This was in the wake of the sweatshop scandals and protests in the late 1990s. They decided that they would try to stay in the program, and market themselves as being less of a risk for sweatshop scandals than others. That hasn't stopped buyers from going to Bangladesh, but it certainly hasn't hurt Cambodia.

The factory collapse is hardly a one-time occurrence. One hundred twelve people died in a factory fire in Bangladesh in November, prompting the above protest. (AP)

How do bilateral agreements like NAFTA fit in here? My understanding was that our agreement with Jordan had pretty strong protections.

All of the bilateral agreements going back to NAFTA have had labor standards. In order to deal with the U.S. domestic politics around labor standards, they haven't been very strong. One case was just settled with Guatemala under DR-CAFTA. I'm not sure it even made it to the dispute settlement panel, but it was the first one where the U.S. requested a dispute settlement panel be appointed.

They have a varying degrees of enforceability, and they're still being tested. Even in the Jordan case, though it had a labor chapter and there was a complaint filed under the trade agreement, the Bush administration decided to come to this agreement on the Better Work program and to deal with it outside the trade agreement, since that was an agreement negotiated by the Clinton administration and they weren't wild about the labor chapter.

Who brings complaints in those agreements, normally?

Usually the AFL-CIO, it's usually the unions or human rights groups. There's also a labor condition in our generalized system of preferences and there as well, it's typically a complaint from unions or human rights groups concerned about labor issues.

 Is there a difference in the character of complaints depending of who's making them? Are the unions more concerned with protecting U.S. industries?

Obviously it's hard to draw out very much statistically, but we tried to do an assessment as to whether there was a positive outcome in each of these cases. There might have been somewhat more likelihood of getting a positive outcome if the petitioner was a human rights group than a union, but we also had a second finding which we think explains that: technical standards around wages or safety were more likely to see improvement than cases around freedom of association, which were more likely to be brought by unions but also politically more difficult for developing companies to address. By contrast, they can raise the minimum wage pretty easily to get the U.S. off their back.

 What are the unions in these countries like? Are they basically indigenous or linked into the broader labor movement?

Depends on where you are and what the situation is. In Cambodia they were really starting from scratch. There wasn't a very strong civil society, as you're coming out of a long period of civil war and there wasn't a very strong concept of unions. They're locally organized, not having any experience, and they didn't really know how to run a union. They were going on strike every other day, which isn't good for anybody.

So the Solidarity Center at AFL-CIO sent people over to train the union leaders, and teach them about when they're allowed to strike, when it's in your interest to strike and when it's not in your interest. So they did grow up indigenously, but they also had this advice and technical assistance from outside. Other countries have stronger civil societies and more experience with unions.

A Bangladeshi police officer inspects the damage after November's factory fire. (Khurshed Rinku/AP)

Just to circle back, you said it's easy for these countries to raise their minimum wage. What level do those minimums tend to be at?

Bangladesh is barely over a dollar a day. The number I saw was $37 a month, but it's a very poor country. It's going to vary very widely.

The thing about it is in other export sectors, the wages tend to be higher than the minimum, if there is a legal minimum, and than the prevailing wage. It really is in apparel, where labor is such a big part of the costs, that they tend to be very very low. There's a paper by Anne Harrison on Indonesia, which was one of the countries back in the 90s that was targeted by the sweatshop activists. It was targeted by one of these GSP cases and raised their minimum wage, and Harrison found this external pressure did appear to contribute to raising conditions and did not appear to increase unemployment or decrease exports.

What exactly do the ILO standards expect of countries? It's not like they're asking for a $9 an hour minimum wage or something in Bangladesh, but what are they asking for?

That's one of the points that I wanted to make having read the Slate piece. It's a little bit tricky having talked about health and safety, but there's a distinction between core and cash standards. The idea is that core standards like freedom of association, nondiscrimination, child labor, or forced labor are both fundamental rights, and they're also framework rights in terms of having a well-functioning rule of law system in place for your economy. Those rights they can vary in the details but, and this is what the 1998 ILO declaration said, all countries, regardless of level of development, should respect these core rights.

Then you have all these other standards like health and safety, like wages, that will necessarily differ by a country’s level of development and, as Matt Yglesias says, by their choices. I wouldn't go so far as Yglesias to say that therefore it's only up to them. In a lot of these cases the workers aren't making a fully informed choice to take these risks. They don't know the chemicals are toxic. They don't know that the building's unsafe. These still need to be addressed and the question is "How do you do that?"

 And how can you enforce those core standards?

It depends on whether you have a country that's willing to improve, recognizes that it has an interest in having better paid and better treated workers and just needs some assistance in making those commitments, or whether there's a problem with political will, and you run into that with freedom of association, as that has political implications. You have options up to trade sanctions if it is an egregious set of violations, I'd prefer those sanctions to be on the factory level rather than the country level. You can target individual factory owners who are telling people to go into the building that's going to collapse.

So standards have to be appropriate for the level of development of a country, but there is a role for the ILO in setting a floor. The question is how you do it so that you get around the collective action problem.

The other thing is the idea, which has kind of faded, but when Richard Freeman and I wrote our book it was a big deal, and that is that consumers can help by paying attention to what they're buying, and from where. Again, the minimum wage, the labor part, even in apparel, it's not a big piece of the final price when we buy it here in the United States.

In terms of improving conditions, freedom of association is both the most politically contested and the most important. It's pretty clear that the corporate codes aren't enough. The monitors under these corporate social responsibility codes are there maybe once a year, and it's usually a pre-announced visit. The workers are there every day. You need to find some way, whether it's a union or a  hot line to a credible office of someone they trust so workers can protect themselves where there are otherwise minimum levels of enforcement.