Wednesday 1 May 2013

Safety preferences

Economists know a few things. Among the things I think we know:
  • People generally require compensation to take on a known risk of death or injury: safety is a good for most people;
  • As people get richer, they require more compensation to take on a risk of given size: safety is a superior good;
  • Risk-tolerance is heterogeneous across individuals, and not simply because of differences in income: some people are better able to manage risks or incur fewer psychic costs in thinking about downside costs of risk.
  • Observations that some individuals engage in a risky activity for little compensation could be evidence that they are poorly informed, or it could be evidence that they have a high tolerance for risk.
Matt Yglesias argues that we should avoid policy overreactions to the recent and tragic factory collapse in Bangladesh. We expect more on-the-job deaths in poorer countries even in cases where there is perfect knowledge among workers about risks; poorer workers prefer total compensation bundles with higher monetary income and fewer on-the-job amenities like safety.

It may look on the surface like Yglesias is being all ‘realist’ and ‘sensible’, but in fact he gets the economics wrong. He forgets three things:
  • preferences are only half the story. The other half is the choice space in which preference can be expressed. It is the combination of preferences and available options that lead to the choices made. Ascribing the choices to preferences alone gets the theory wrong; one can just as legitimately point to the limited options
While that's true, surely the options available are endogenous to local effective preferences. If effective demand among local workers were for a total compensation bundle that included more safety and lower wages, and if it were no more expensive for the firm to provide that kind of bundle than to provide the less-safety-higher-wages bundle, they'd wind up providing it. It could take a while - if safety is largely contingent on prior fixed plant investments and if plants have a relatively long life, it would be hard for workers to change things at existing factories. But it would also be hard for anybody else to effect that change.
  • the market theory that Yglesias uses to underpin his ideas — that there are market transactions deciding the prices of garments and safety — assumes freely available and perfect information. A large economic literature then explores the impact of relaxing that assumption. But that’s the post-grad course, and Yglesias is stuck in 101. Here’s the thing: we could make it perfectly obvious to Western consumers how their garments were made, what the working conditions were. Then we could talk about a market solution. Let me put it another way: is Burger King going to launch a horse-burger because people were buying them before they found out what was in them?
Let's take the case of imperfect information among workers first as I think it's the more serious one. Suppose that workers trust in that the existing government regulations around safety form a binding floor on the level of safety that a firm can provide, but that the owners chisel by bribing officials in ways not noticed by workers. Then, the workers are effectively accepting a lower total compensation bundle than they thought that they were being provided. Bill links to plenty of evidence that the factory owners in this case were up to all kinds of shenanigans.

What do we expect might work in this state of the world?
  • Asking Bangladesh to increase its building standards is unlikely to have much effect where the quality of governance is too poor to enforce the current standards; asking Bangladesh to improve its generalised quality of governance may not be far from asking for unicorns. Improved governance would be great, but we have no good general handle on how to achieve that.
  • Local workers will discount existing building codes and government regulations and perhaps demand more direct and credible signs that the expected level of safety is being provided. I'm not here suggesting that a local worker can go up to the company owner and demand things; rather, a factory opening up next door showing with credible evidence of stronger standards will hire away workers where workers have effective demand for safety. 
    • This could yield a market in external certification, but it could be tough to establish. You need somebody that the workers trust to provide the certification, who is competent to do it, and whose officials are less corruptible than state workers. It's far from impossible, but I'm not sure it's easy.
  • If local certification options don't pan out, this is actually a spot where Western retailers could help out. Western retailers can enforce standards among manufacturers through spot-checks and the like, and they're credible. But what should they enforce?
    • If you think that the main problem is that owners lie to workers about existing safety standards but that workers are otherwise competent to choose safety-salary bundles, you want the retailers not to be enforcing minimum safety floors but rather to be providing accurate information to workers about real safety risks and about safety conditions in other potential places of employment. Putting in minimum safety standards risks worsening the lot of workers who legitimately would choose less safety and higher salaries.
    • If you think that the main problem is that poor foreign workers cannot adequately judge safety risks, you want the retailers to put in a minimum safety standard using their best estimate of what well-informed poor workers would choose. I worry that these kinds of standards wind up instead embedding a lot of Western wishful thinking; I also worry that rich country lobbyists have strong incentive to push for standards that work to raise their rivals' costs rather than to improve the lot of workers. 
I'm not sure that there's any particular net market failure caused by imperfect information among Western consumers. My basic model has two offsetting effects: consumers don't really know what goes into the sausage, but they're also prone to believe that banning sweatshops leads to kids in schools rather than kids in malaria-ridden fields or in garbage dumps. It is way way easier to get Western consumers outraged about working conditions in third world countries than it is to get them to think hard about policies that might do more good than harm. And first-world manufacturers and unions are quick to jump in with entrepreneurial policy proposals that look nice while raising rivals' costs.

And so I worry about the risk of making workers in third world factories worse off as they see things. I'm sure Bill worries about that too; we may just be disagreeing on how far above the ideal floor any minimum-standards campaign would wind up pushing things. 


  1. If rising urbanisation brings a lot of workers to cities in Bangladesh (as in China and other developing nations) and sweatshop work is one of the best available jobs then surely the pool of willing workers is greater than the number of positions (hence the low wages). Given all of this, is it really fair to assume that worker preferences have much influence over the job market?

  2. Are you hanging the case on that people make errors in moving to the city in the first place and word not getting back to others in the countryside?

  3. I'm saying that if garment industry work is their best option then surely they aren't in a position to negotiate better conditions. This really hinges on how scarce workers are. I looked into the statistics briefly but the grey market obfuscates accurate employment numbers.

    Most of the garment workers are young migrants from rural areas (90% are women, who have limited job options in rural areas). They send remittances back to their families so it follows that they must be able to communicate too. So I don't think they are moving in error.

  4. I'm assuming zero negotiating ability on the part of the workers. Say that every company makes take-it-or-leave-it offers and that no worker has huge expectations of that he'll get lots of offers. Suppose that workers will be happy taking a $1 pay cut in order to get another unit of safety. If it takes less than $1 to provide that unit of safety, then the company that offers the bundle with lower pay and more safety gets its pick of workers. And once we're at equilibrium, people have sorted across firms offering different safety/wage bundles. As you've noted, migrants expect to be better off by moving to the city; they could choose to stay in the country. The bundle they're then choosing has to be one that's preferred to the reversion option.

    What breaks it is if employers are chiseling on the level of safety that they're actually providing.

  5. "you want the retailers not to be enforcing minimum safety floors but rather to be providing accurate information to workers about real safety risks and about safety conditions in other potential places of employment. "

    You may not want them enforcing global safety floors. But I would think that enforcing the floor provided by local regulation would be relatively independent of the Western-wishful-thinking problem, so there's not an obvious problem with Western retailers acting to (help) enforce the relevant local safety standards.

  6. Yeah, but you can also make the argument that if everybody's fully informed about the floor, then they should be able to contract around it.