Paul Krugman, having apparently received another of his divine revelations, proclaims that if we demand (somewhat) better working conditions in Third World countries (backed up, presumably, with boycott threats), “we can achieve an improvement in workers’ lives … And we should go ahead and do it.”
Don’t ask how he knows; the ways of the Oracle are mysterious and beyond human ken.
Look. A well designed policy of boycotts and boycott threats can certainly improve working conditions in the Third World. It can also lower either wages, employment or both. Whether or not that package amounts to “an improvement in worker’s lives”, as Krugman puts it, is an interesting and important question, and well worth thinking about. But apparently the last thing Krugman wants you to do is think about it, since he’s already told you the answer, and seems to presume you won’t have the slightest interest in where it came from.
Now, among the many differences between me and Paul Krugman, there are probably some that redound to his credit. But his propensity to hide all of his reasoning (if any) is not one of them. Compare, for example, my blog post of a few years ago on working conditions in 1911 New York City, when the Triangle Shirtwaist fire claimed 146 lives, most of them young women, partly because the fire exits were blocked to prevent pilfering. Would workers in 1911 have wanted safer working conditions (including unblocked fire exits)? This was my answer:
I can’t be sure (and I’ve pointed out several reasons I might be wrong), but I’m guessing that no 1911 garment worker would have wanted to work in a factory with unlocked exit doors. If I’m right, they got the mix of risk and income they’d have chosen.
What’s important here is not my acknowledgement that I might be wrong. It’s that I took the reader through my logic and my guesstimates, step by step, and invited that reader to substitute his or her own guesstimates to see how they affect the conclusion. Along the way, I hope I taught some lessons about how to evaluate costs, benefits and policy choices.
I am blessed with a readership that would never let me get away with announcing a conclusion stripped of any reasoning. Krugman, with his considerably larger readership, can afford to ignore anyone who might be interested in learning something and preach instead to the lowest common denominator. If you don’t believe me, check out the quality of his comments section.
But the reason I try not to avoid “pulling a Krugman” is not just that my readers won’t accept it; it’s that I wouldn’t want to. In 1996, when Paul and I were the two economics columnists for the then-newborn Slate Magazine, I think we agreed that our mission was to show the world how economists think, and how to think like an economist, and why you might want to. The antithesis of that mission is to make policy pronouncements from on high, with no argument, no analysis, and not even an acknowledgement that arguments and analysis serve a purpose. When did Krugman lose his way?
“When did Krugman lose his way?”
My guess: November 2000, when Al Gore lost a statistically tied election. That was a devastating blow, to be sure.
November 2000 was the genesis of Bush Derangement Syndrome for many thousands of committed partisans. Some tried to fight their emotions; others reveled or wallowed in them. Krugman won the Nobel Prize by pandering to the latter category, which included the Nobel committee.
Steve, why would employment necessarily go down if we demanded improvements in working conditions and wages? Didn’t Krugman give an argument in the post that employment wouldn’t go down if we made the same demands on all the third world countries at once?
@Steve,
I agree with you that it was one of his less thoughtful posts as of late, but I think he very often includes very thorough analysis and explanations in his blog posts. Especially lately. I think his recent explanation of how he approaches research was extremely interesting. I don’t think he’s lost his way just because some of his blogposts aren’t the most content heavy. If Krugman produces five blogposts and you pick on the one that didn’t have good content, is that a fair analysis of his work?
I also agree that his readers comments are mostly me too’s. This is the reason I enjoy participating here. Having people agree with you constantly is not intellectually stimulating and can be slightly annoying. But don’t you think that is somewhat a function of posting on such a highly trafficked website? People that go looking for your website are probably people who are slightly more interested in economics and mathematics than the average reader of Krugman’s blogposts. If you go to Brad Delong’s website you’ll get slightly more intelligent readership than Krugman’s on average but I would still argue that you have a more healthy discussion going on here at TBQ. One thing I’ve noticed is that I rarely see libertarian economists putting effort into the comment sections of more progressive economics websites. One thing I’ve realized over the years is that you should never ignore an argument from someone just because you often disagree with them. The more each side stays in touch with and continues to listen to one another, hopefully the closer we’ll get to the truth.
@Keshav Srinivasm.
Yeah, but isn’t he ignoring the trade off between labor and capital in that analysis? Also, with so much unemployment around the world it’s a hard task to get everyone to somehow stay on board with that plan. It’s a bit of a prisoner’s dilemma in that each individual asian country has an incentive to bail on these reforms even if on net society would make out ahead.
Keshav:
Didn’t Krugman give an argument in the post that employment wouldn’t go down if we made the same demands on all the third world countries at once?
No, he argued that if you made the same demands on all the third world countries at once, then you wouldn’t destroy the apparel industry in any one of them. This, of course, is true.
But if you improve working conditions, you raise the marginal cost, and hence the price, of apparel. If you raise the price of apparel, people will buy less of it. If people buy less of it, fewer people will be employed producing it.
Steve, I agree with your critique of Krugman’s pontificating style, and your reasoning (above) is solid, but what is the price elasticity of the branded garments manufactured in poor countries, and in any case, won’t those garments still be cheaper than garments mades in developed countries, even with improved safety standards? Doesn’t basic human dignity and minimum safety standards count for something in one’s economic analysis
Improving working conditions is not an overnight phenomenon. As Paul Krugman states, “It remains true that given their low productivity, countries like Bangladesh can’t be competitive with advanced countries unless they pay their workers much less, and provide much worse working conditions too.” Should we infer that the workers will overwhelmingly support the types of processes, the mechanization and technology, that industrialized nations use?
There is a lot more involved that having the workers of the world unite. Has anyone asked why these workers worked in these factories? Could the incentives outweigh the risk? In a sense, this hearkens back to the early 19th century in America, where people went from farms to factories, from villages to cities. And it took time for an England, France, Germany, or America to become an industrialized powerhouse, a place with a high standard of living.
I don’t feel it correct to force our standard of living on a group that is unprepared for it; but then, what do I know?
Gee-whiz, where did my original post go?
He did link to the earlier Slate article to provide the background for anyone interested. Nonetheless, he concludes that the benefits would outweigh the costs to the workers without saying how he knows.
I am sure one can construct such an argument. The increase in price is tiny compared to incomes in developed world. Therefore the reduction in demand will be small. Still, we don’t know that this reduction in demand is worth it. We can conclude that it is if we can show that workers at subsistence wages undervalue long term gains, such as reduced risk of dying in a fire. I believe this is the case, and research into cognition supports that the sort of reasoning errors people tend to make are those that would lead to such an undervaluing.
We also saw in the minimum wage debate that increasing minimum wage did not reduce employment. The costs were absorbed elsewhere, sometimes in economically beneficial ways.
Kaushik Basu,a leading development economics theorist, and Homa Zarghamee have looked at this theoretically and found that “… consumer product boycotts can, in a wide class of situations, have a backlash that causes child labor to rise rather than fall. This happens under weak and plausible assumptions” (J Development Economics, 2009).
@9 Harold: It is wishful thinking to believe raising the minimum wage will not reduce unemployment (unless the employer has monopsony power, which is seldom true). The Card-Krueger finding has been found faulty. The big issue is statistical power: It is like we are trying to find bacteria using a kid’s microscope. We cannot see it, but it’s there.
Harold:
I am sure one can construct such an argument.
So am I. And I think it would be instructive and useful to write a blog post showing readers how that’s done.
But the message I get from Krugman’s blog post is that we don’t need arguments, because Paul Krugman is there to tell us what to believe.
@ Jack PQ,
“It is wishful thinking to believe raising the minimum wage will not reduce unemployment (unless the employer has monopsony power, which is seldom true)”
I think you meant employment rather than unemployment. Anyway, I’ve been thinking more about monopsony lately, and I kind of think that during high unemployment situations, companies do exhibit aspects of a monopsony because workers are afraid that the job they hold is the only job that is available for them. Fear prevents them from negotiating for higher wages or walking away when they’re not getting a fair deal. I think this might be one reason we see piles of corporate cash lying around right now.
Krugman claims he is not contradicting his earlier article. But now he emphasizes that sweatshops in Bangladesh are not competing against factories in Newark, but in China. The wage differential must be smaller now then, making his earlier argument apply a fortiori i would think. Instead he writes as if it were weakened. That needs explaining.
@ Ken B,
I think he was saying that it would be a problem if only one country out of Bangladesh and China adopted stricter rules since the wage differential is smaller. All of the competing countries would have to adopt the same rules to not reduce employment, was what he was saying. Before people jump down my throat, realize that I don’t agree with Krugman about this.
@15
The tighter the competition the more fragile the cartel, no?
@ 16,
Agreed.
It’s also not clear why the production would ever move back to advanced nations in any event (barring a return to highly protectionist policies)? I thought if I had learned anything back in the day from Krugman’s (deservedly) famed essays on “competitiveness”, it’s that 1) trade patterns are determined by relative productivity for different goods, not absolute levels, while 2) standards of living are determined by absolute growth in domestic productivity, “period”, not relative to other nations. Something about countries not competing like corporations.
#13 – Even if/when “the job they hold is the only job that is available”, what does it mean to say “they’re not getting a fair deal”? While I agree it can seem at times like labor market conditions swing in favor of *either* side, your case seems at best like a ‘temporary monopsony’ borne of weakness (inadequate demand).
#9 – Your story might also want to consider that when one’s primary concern is how one will provide one‘s family with their next meal, one may have good reason to discount other longer-term factors quite heavily. Perhaps such ‘cognitive errors’ are things one has the luxury of contemplating only when one is higher up the hierarchy of needs. As SL mentioned in his earlier piece, to the extent ignorance is involved it may be unclear in which direction – being “blissfully unaware” of safety conditions, or of their impact on wages?
@iceman 18,
“what does it mean to say “they’re not getting a fair deal”?”
By “fair deal”, I mean they’re not getting the wage that would prevail under competitive labor conditions (i.e. not a recession induced monopsony). I don’t think it’s an argument to permanently alter the minimum wage, just an observation that “monopsony” like conditions could occur more than rarely if we’ve entered a depression. In the same way that lot’s of companies have “brand monopolies” over certain segments of the population.
This is a bit off-topic.
@7
The problem is, the general (basic) Solow growth economic model says that growth patterns in different countries should be converging. We see this phenomenon in First World Countries, but in third world countries we see no such conversion.
If you can explain why this is, you win a Nobel Prize.
Even if Krugman does have a Prize (of many econ graduates at my school think he didn’t deserve since his work is a re-hash of ideas and explanations brought up by other people before him), I don’t think I would go with his reasoning.
But if you improve working conditions, you raise the marginal cost, and hence the price, of apparel. If you raise the price of apparel, people will buy less of it. If people buy less of it, fewer people will be employed producing it.
Doesn’t this argument hide a lot of the complexity of the industry? After all, there are several prices / costs here… there’s the price paid to a garment worker, the price paid to a factory operator, the price paid to an exporter, the price paid to an importer, the price paid to a label and the price paid to a retailer.
It’s not clear that all of these have the necessary elasticities of supply and demand to ensure that raising the “price paid to a garment worker” (eg, by paying them more or just improving their working conditions) would necessarily reduce demand at the retail store by enough to significantly lower the quantity flowing through the supply chain.
Doubling the benefits to a Bangldeshi garment worker will add 22c to the cost of a denim shirt made in Bangladesh. Do you really believe this will impact demand for shirts in any material* way?
Footnotes :
* [tee hee!]
JackPQ #11. It is not wishful thinking – in fact quite the opposite. It is the conclusion of an extensive report and evidence based, using research since 2000. I believe this conclusion is widely accepted. It may be wishful thinking to assume that the simple economic models hold in the face of contradictory evidence. However, I am not expert enough to say that the evidence is conclusive.
The report is titled “Why does the minmimum wage have no discernable effect on employment?”
http://www.cepr.net/documents/publications/min-wage-2013-02.pdf
21 – another good example of what constructing an argument looks like.
19 – I know you’re saying sometimes the balance of power tilts toward employers; I’m just pointing out that it’s reciprocal, and “fairness” is a loaded term in the context of voluntary, mutually beneficial contractual relationships. If employees don’t always get the “deal” that might prevail under optimal economic conditions, it’s because employers don’t want (need) to offer a better deal than anyone else is offering under other circumstances. “Monopsony” normally implies someone is actively buying something; in this case it seems a pretty weak show of force if nobody is hiring.
22 – at first glance I’d just say one troubling aspect of your report is that it considers the biggest shortcoming of the ‘seminal’ Card-Krueger study to be that it’s hard to generalize from a single example, while seemingly taking its findings at face value; my understanding is that the methodology of that study has been found questionable (i.e. based on phone surveys of store managers’ impressions of their hiring activity, whereas actually reviewing “hard” employment / tax documents gave a very different picture). Also note the CEPR calls itself a “progressive think tank” (doesn’t make them wrong, but not sure you’d be content simply to link to something by say Cato or Hoover).
@21:
Doesn’t YOUR question hide some assumptions? I make the franipulator for plogistrator, and essential component of the fogulator. I pay my workers .47 cents an hour, and a fogulator sells for $567.00, less on Amamzon, so you might thint I could pay my navvies a generous 2 cents an hour. Alas the markup on franipulators is so small I’d lose money paying then anything over .54 cents an hour. Comparing that imagined extravagant 1.53 cents raise to the price of the fogulator is mistaken.
@ iceman
““Monopsony” normally implies someone is actively buying something; in this case it seems a pretty weak show of force if nobody is hiring.”
Nope, the definition of monopsony is that there is a single buyer of a good. While this might not actually be true (there might be a market out there for people in a job) it can sure feel like to them in a depression that there’s not. So it satisfies all of the qualifications of a monopsony, since the seller of labor believes that this is the only buyer out there for their labor.
@Steve,
while you are certainly right that improving the conditions of textile workers raises the cost of textiles, the cost of providing for things such as fire escapes and non-collapsible buildings is minimal. Take the Triangle Shirt factory fire. There were 500 women working in that factory. An additional unlocked but GUARDED emergency exit, or a window that can be bashed in would have meant negligible additional costs. Such safety measures add up to only a tiny fraction of one percent of a products retail price and will not produce a statistically detectable decline in demand.
@Advo 26:
Wrong question. A better one is, would the women in that building, before the fire, have chosen a paid guard or no guard but a daily raffle for one of them to win his wages?
@25: “Nope, the definition of monopsony is that there is a single buyer of a good”
How does that not imply somebody is buying something?
@KenB 27:
I think you’re asking the wrong question. Wouldn’t it be “Would the women have accepted a 0.2% wage cut in exchange for an additional fire exit?”
An interesting aspect of the case is the fact that the factory had a single exit designed so only a single person could leave with a single watchman checking each woman’s bag. An arrangement such as this must have led to significant waiting times for workers at the end of a shift and was certainly quite inefficient. How could something like this exist in a competitive market?
@Henri Hein,
Um, there very clearly is a buyer of their labor, as in their current employer.
Advo:
An arrangement such as this must have led to significant waiting times for workers at the end of a shift and was certainly quite inefficient. How could something like this exist in a competitive market?
I share your puzzlement, but in my experience, the answer to the question “How could something like this exist in a competitive market?” is usually “For a very good reason that is quite obvious to the market participants and that nobody else would ever think of.”
@ Steve
“I share your puzzlement, but in my experience, the answer to the question “How could something like this exist in a competitive market?” is usually “For a very good reason that is quite obvious to the market participants and that nobody else would ever think of.””
Or they just didn’t think about how easy it would be to set up an emergency exit before the fire, and were motivated to come up with an innovative solution that ended up being more efficient than their previous plan of locking doors when the government made them think about the costs. You talk as if the market will automatically find solutions to problems before they even know they exist or try thinking about whether there may be an innovative solution.
Your supposition in the old post was also that the workers were fully aware of their risks associated with locked doors. Do you think they were presented with a life expectancy table with and without locked doors before the fire occurred? This is the only way I can believe that a competitive labor market would come up with a solution by itself without motivation.
@Steve:
Most likely, it is a combination of information issues (for both employer and employees) and frictional costs.
The factory owner doesn’t realize that the wasted time of his workers has indirect costs for him.
I see this kind of thing a lot in real-life business. Especially in Anglo-Saxon countries, easily measurable “hard costs” receive far more attention than difficult or impossible to measure “soft costs”. Lowering employee morale through excessive work burdens, ill treatment, layoffs, etc. causes very high costs (due to lower productivity, waste, theft, high employee turnover) which however tend to be difficult to measure, and so quite often “hard costs” are saved at the expense of incurring much higher “soft costs”.
The workers, on the other hand, don’t know what conditions at alternative employers are like. A different employer will have its own set of advantages and disadvantages, largely unknown and unknowable to a prospective employee, and switching employers creates frictional costs.
#27 Wrong question. As mentioned by Daniel, a better question is would the women have chosen the extra exits / guard had they been fully informed.
However, we have no need to specify that a guard must be employed. Other mechanisms are available wghich may be cheaper. It is my belief that it is sensible to set a minimum exit time, or something like it, and let the company sort out how to do it.
#23. Why the focus on Card-Kreuger? The paper cites two meta studies which included 64 and 27 papers. Both concluded that the effect on employment was absent or too small to detect. There was an extensive review (not meta-study) that concluded there was negative effect, and a further review that concluded there was not.
I would not say this was necessasily conclusive, but it adds up to more than wishful thinking.
@Harold,
Because of leisure preference reservation wages, there are models in which the competitive wage will never be achieved by low skilled workers. Thus the reason minimum wage laws have a tiny at best effect on employment.
@34: “Fully informed”? Like AIDS sufferers in the 70s? You seem to assume that the risks were known and understood. They were not. We’re discussing the relative cost of unknown risk and you want to toss in prescience?
Steven – so did you meet PK when you were at Slate (or any other time)? If so, would be interested in how the interaction went.
@Ken B 36,
Yes exactly both my and Harold’s point. The entire idea that competitive labor markets would make the right decision in locking doors relies on both the laborers and the employers knowing the benefit and cost of that restriction. If they can’t know that than they would not arrive at the correct solution. As in, the laborers could be accepting one penny more in wages a week, for a restriction that actually costs them $1 a week. See Steve’s original post where knowing the cost of the restriction was one of his assumptions.
Daniel, the paper discusses the reasons for the lack of effect. They discuss 3 employment models – competitive, institutional and monopsony. The competitive model suggests that firms operate at maximum efficiency, and neglects costs of turnover. Adjustments may be made by reducing hours or raising prices. The institutional model assumes firms could become more efficient since management of efficiency is costly. Adjustments under this model include efficency savings and productivity increases as management tries to get more out of its workers. The dynamic monopsony model (friction model) takes into account the costs of turnover. Workers face costs and constraints in finding and keeping employment. A firm offering the “going rate” will take time to fill the vacancy, and will lose workers who find the costs are just higher than the value of the work (e.g. childcare or transport). The firm has a choice of operating with unfilled vacancies or raising the wages of the whole workforce. They may choose to operate with unfilled vacancies. Raising the minimum wage will make it easier to recruit and retain workers, lowering the costs of turnover, and paradoxically increasing employment. Is this the one you refer to with reservation wages?
The study concludes that the adjustments come from all three models – some price rises, some wage flattening, some efficiency gains and reduction in turnover.
@38
What I don’t get is why you would know better than the participants would, in advance. In hindsight we always know the inhabitants of Pompeii should have left town a couple days earlier, the passengers on the Titianic should have booked a slower boat, the building should have had 17 exits. Your claim, it seems, is that there’s a right answer, it’s obvious, and the competitive market won’t find it. That’s quite a parlay, and it can fail in several ways. First costs and benefits must be balanced, and markets do that better than commissars do, and second, when there’s a clearly correct answer that the participants do not see but we do it’s invariably because of hindsight.
Ken B 36. Fully informed does not necessarily require prescience. An understanding of the risks is more what I am thinking of. Prescience would mean that only those in a factory that actually burned down would need to invest in a fire door – all the rest could save themselves the cost. This could get into a discussion of predetermimation.
The risks can be better understood by evidence gathering and research than by personal experience. This means that some specialists will understand the risks better than most people. We can either use this specialist knowledge to set reasonable standards, or we can ignore it and let those with less knowledge and understanding make the choices – or we can try to inform everyone, but this will not be possible.
The key here is “reasonable”. We should not let the statistition, architect, economist or engineer set the standards based on their expertise alone. Ultimately they must inform the politician. Therein lies our problem.
Daniel (#38):
Yes exactly both my and Harold’s point. The entire idea that competitive labor markets would make the right decision in locking doors relies on both the laborers and the employers knowing the benefit and cost of that restriction.
Only if you define the “right” decision to be the one that would be made by an omniscient being. I don’t see any context in which this definition is a useful one.
Ken B (#40) and Harold (#41): Two excellent (and opposing) comments in close proximity. Thank you both!
Yeah, the third model was the one I was talking about. I also think it’s the one that’s closest to reality, but some combination of the models would probably best describe the entire US economy. I think figuring out where each model applies would be a topic worthy of a dissertation in labor economics.
@Steve 38,
Let me put it another way. Do you think workers at the shirt waste factory were underestimating or overestimating the cost of having a locked door before the fire? I would guess underestimating but I suppose if you believe they were overestimating than we are at an impasse.
If you believe like me that they were underestimating, than doesn’t some regulation to correct the market make sense. If not, why not?
I believe research in behavioral economics has established that: “individuals tend to preference certain short term rewards over uncertain long term risks and losses”.
It is therefore not reasonable to expect employer and employees to arrive at an efficient arrangement with regard to fire safety.
http://www.thefreelibrary.com/Public+perception+of+the+data+environment+and+information…-a0312290660
@Harold 41:
It still seems to me you are assuming too much. ” An understanding of the risks is more what I am thinking of.” me too, but I am talking about the understanding ante the fire and you are talking about the understanding post fire. It’s like AIDS; the risk was not well undertood by *anyone* in 1978. Disasters are the clearest example of why experience is the harshet teacher: she gives the test first and the lesson later.
@ Ken B 40,
The Triangle Shirtwaist factory event, was not a one off. Unlike Pompeii, accidents and injuries in the workplace are a predictable events that exist on well known probability distributions. Anyone that has information about that can run models and predict the potential costs. The question is, does the average worker at a factory have access to this information?
#47. I see the problem. The extent of the culpability of the factory owners depends on the extent of knowledge at the time. I am sure there were people who knew more about the risks than the women, and the factory owner. There had been fires before this one, so to some extent it was a forseeable incident.
EEK. 47 got posted by a fat finger before I finished so I take up where I left off.
Now we see the fire risk clearly. But the people at the time, who had more experience with fire than we do I note, did not. They probably extrapolated from other types and sizes of building. (Note I am speculating. They really DID know fire risks first hand which means I could be underselling my argument below.)
So you have a problem. Either the right course is gobsmackingly obvious or it isn’t. If it isn’t, shouldn’t we let those involved make the decisions? And won’t markets be the best general mechanism, absent prescience? If it IS obvious then you have to explain why it is not obvious to the participants. And I bet what you’ll invariably find is hindsight: something we know now that no-one knew then.
If you look at smoking ads you will see its health benefits extolled. We laugh but shouldn’t. Smoking really DOES have a lot of health benefits. Quite a lot in fact. But it has one or two health costs that are not immediately obvious, short term. And they dominate in the longer term. That was not widely known until about the 1950s. But our laughter at the old ads shows just how hard it is to overcome hindsight.
@48
As I noted, people then had a LOT more experience with fire than we have. (I suspect that they had less experience with large buildings though.)
@49
I don’t think culpability is the issue BUT let’s say it is. Then we still differ. I say the workers would be happier to have a choice between an extra door or a lottery to spread those costs amongst themselves. So to me the owner would be culpable only for hiding what he knew. You say they would be better off if the rich boss made the decision for them. So to you he would be culpable for giving them options.
Advo (#46):
I believe research in behavioral economics has established that: “individuals tend to preference certain short term rewards over uncertain long term risks and losses”.
It is therefore not reasonable to expect employer and employees to arrive at an efficient arrangement with regard to fire safety.
If individuals have conflicting preferences in the short run and the long run, how do you define efficiency?
@Steve:
I don’t think this is a problem in this context, because the preference hinges not so much on “short run” and “long run” as on “certain” and “uncertain”.
Individuals treat certain long-term risks far more rationally (as in, similar to short-term risks) than uncertain long-term risks.
So you would define efficiency based on how people would treat certain long-term risks.
This thread illustrates a fatal weaknesses of economic science – there is an ongoing argument that works from the assumption that the economic actors approach this situation rationally even though it’s empirically observable that in reality, they don’t.
54 – “certain risk” – do you mean quantifiable? I think some simply call that “risk” and the other kind “uncertainty”. Certainly seems easier to identify a ‘rational’ response to the former. There also does seem to be a time element here to the extent people are weighing current income with events they consider unlikely to happen i.e. in a more ‘distant’ sense. Which seems to be the stronger case for claiming they “over-discount” such events, but at least in this case we also have to sort out whether such a preference is not so irrational for people working at a subsistence level.
@56:
If I remember correctly, people act irrational when the risk is known (as in, a 5% chance of a 1000 dollar loss is not discounted appropriately, while a 100% chance of a 50 dollar loss is) and also tend to underestimate the likelihood of risks they have no personal experience with (as in, they think a risk is 1% when it’s really 5% or 10%).
Advo:
This thread illustrates a fatal weaknesses of economic science – there is an ongoing argument that works from the assumption that the economic actors approach this situation rationally even though it’s empirically observable that in reality, they don’t.
That would indeed be a serious weakness. So, of course, would be the assumption that regulators and other political actors always approach such situations with beneficient intentions, even though it’s emprically observable that in reality, they don’t.
@ Steve,
“That would indeed be a serious weakness. So, of course, would be the assumption that regulators and other political actors always approach such situations with beneficient intentions, even though it’s emprically observable that in reality, they don’t.”
A weakness of this would be to assume that each instance of government actions did not have beneficent intentions. Let’s boil it down to the example you use here. Do you think that creating safety standards did not have beneficent intentions?
@59:
Can I answer? Then the answer is NO. I can think of things masquerading as safety regulations that had no benificent intent. A recent example is beef labelling. Lobbyists fo American beef farmers have pusehd for mandatory labelling of the country of origin of beef that has passed inspection. This ios no small matter as it means that large slaughterhouses cannot use economies of scale. Estimates are it can drive up the cost of imported beef by about 20%. This is pure protectionism masquerading.
“Oh” you cry “But I meant real safety regulations!” Well let’s forget the outmoded rules that are still in force, driving up costs, and the foolish rules that preculed better measures, and get back to what the debate is really about: sustainable social mechanisms for making these decisions. An omnipotent benevolent dictator — me for instance — always does best in the short run, but once I am gone what ya gonna do? Markets are your best answer.
@Steve
That is true, however, I’d argue that things such as fire safety regulations have created a situation superior to that which would exist if things were left to the market (even if you disregard externalities).
I think that the problem with malicious regulation could be much ameliorated if the political system in the US wasn’t designed to institutionalize the purchase of political influence by large donors. Getting at the root of political corruption in the US may be a better target to aim for than simply trying to cut back the size of a system of government which is corrupt by design.
But then – what regulators (and especially US regulators) would do or not do is not of primary importance with regard to the question to what degree a discussion by economists about the behavior private parties relates to reality.
61 – I believe someone we know once wrote that campaign finance reform mistakes symptom for cause. The fewer thing politicans have influence over, the less people care about buying their vote.
59 – It sounds like you’re asking as though rhetorically? I’d hope we agree the most direct incentive politicians have for anything they do is to get votes (or cash), for which all that matters is that people *perceive* them as acting out of good intentions (or they serve special interests). This can include demagoguing beyond an optimal level of things like reductions in pollution or workplace risk. To me this post highlights that *if* people really are as poorly informed / incapable of making decisions under uncertainty as some are suggesting here, that makes them particularly vulnerable to the allure of a free lunch such as: “I’ll improve your working conditions, and pretend this will have no resulting impact on your wages [safely assuming you won’t notice, and if you do I’ll just have a second opportunity to bash the greedy employers].”
This ties in with #41 which also frames up the discussion nicely: “we should not let the statistition, architect, economist or engineer set the standards based on their expertise alone. Ultimately they must inform the politician. Therein lies our problem.” How about “they could inform the *workers* who can then prioritize for themselves”? It seems we reach a crossroads where, even if we assign a legitimate role to such public “specialists”, we either trust people or we don’t. (Note no externalities involved here – unless of course you “feel” the employer is a monopsonist).
Having said that, as Ken B suggests having boots on the ground allows one to become self-informed through a process of logic, intuition, trial and error etc. that is often difficult to replicate by parachuting in. As David Friedman put it, “If you had to understand something intellectually in order to do it, none of us would be able to walk – as became clear when people started trying to program robots.”
@63:
I note a huge difference (as in >10:1) between malicious regulation in Germany (my home country) compared to malicious regulation the US.
We have a problem with overregulation in many areas, but cases where the intent of regulation is something other than to serve the perceived public interest is relatively rare – in the US it seems to be more the rule than the exception.
You said in 62:
“The fewer thing politicans have influence over, the less people care about buying their vote.”
is fallacious. In the long term, government to a large degree determines ITSELF how many things it has influence over.
And so if government is for sale, it is ALWAYS worth buying.
Corrupt politicians make themselves useful.
You can have clean large, invasive government and clean small, hands-off government, and you can have large, invasive corrupt government, but you can’t really have corrupt small government. Corrupt government will always be very invasive in many ways because that is how it earns its bribes.
@Steve,
Your phrase “…I try not to avoid ‘pulling a Krugman’” looks like it has a double negative in it that reverses the intended meaning. That is, if “pulling a Krugman” means something like “presenting conclusions as if they were facts.”
Can too much knights-and-knaves reasoning cause this type of misngegation?
Jeff S: You caught me. I’m glad to know there are people reading this carefully.
SL: “That would indeed be a serious weakness. So, of course, would be the assumption that regulators and other political actors always approach such situations with beneficient intentions, even though it’s emprically observable that in reality, they don’t.”
Both are serious weaknesses. The important point is that if weakness 1 exists, then regulation can in principle improve matters. Bad regulation can also make things worse.
We pretty much know the market is imperfect. We also know that regulation will be imperfect. We should be striving to identify when and how intervention will actually be beneficial.
Iceman #63 “I’d hope we agree the most direct incentive politicians have for anything they do is to get votes (or cash)” I don’t think we know this for politicians generally. Many other motivations are possible, even a genuine desire to help others.
I think the problem comes with successful politicians. I have a suspicion that those who retain their idealistic desire to help tend to fall by the wayside. It is those that are motivated by Icemans catagories who become the political leaders. It is a form of natural selection.
@Harold 67,
it is a form of natural selection which is PARTICULARLY PRONOUNCED in the US system, where political campaigns cost a great deal of money which must in turn be procured by selling political influence.
68- So it is a sort of market for politicians? Should this give us an efficient result?
The political “market” is characterized by rational ignorance, hence the outcomes are driven by superficial policies and special interests. However I think you’ve made a nice case for term limits.
Advo – I find your comments contradictory. My impression is govt is some degree larger (i.e. more invasive) in Germany, and you say malicious regulation is much greater there, I presume in spite of more restrictive campaign finance laws since you suggest the corruption problem should be particularly pronounced here. (A possible answer is suggested by those who call campaign finance reform “incumbency protection acts”.) Here’s the kicker: “government to a large degree determines ITSELF how many things it has influence over…[versus] you can’t really have corrupt small government.” Once upon a time the US tried to institutionalize limits on the size and scope of govt. I guess if we let that slip (by electing legislators who push the envelope and appoint justices that look the other way) we only have ourselves to blame.
Harold – “We should be striving to identify when and how intervention will actually be beneficial.” Agreed of course. I believe the point of this post was “where’s the striving / analysis here?” I also think we ought to agree it’s just an objective fact that a politician’s most direct incentive is to do things that get them votes (or cash to get votes). Like a business person’s most direct incentive is profits. Neither case is always true. But the general point was in our striving for analysis, we don’t get to weigh the raw realities of market outcomes against a utopian vision of govt.
And even if everyone has the best of intentions, another consideration is that bureaucrats (or their hired specialists) are inherently incented to weigh the benefits of regulation more heavily than the costs as well, which is not a formula for efficient outcomes. This is the key difference in the thought experiment here of letting *informed* workers decide; seriously, what exactly is the objection to that, that they’re simply not capable of deciding what’s most important to them?
“that they’re simply not capable of deciding what’s most important to them?” Precisely. Humans are not good at that. Ask any self-help coach. People have goals and then take actions that do not move them towards their goals. Either they don’t know what their goals really are, or they don’t know how to get there.
@Iceman,
I unfortunately switched the numbers around accidentally. I wanted to say that it’s Germany 1, US 10.
Regulation with malicious intent is incredibly rare in Germany, in the US, it seems to be very common.
With regard to “informing workers”, that is in general a good idea, but only for a limited number of issues. In today’s complex society, if people had to make informed decisions on 10% of what’s currently being decided for them by government, they would collapse from information overload. Or, in economic terms, requiring the workers to get informed and make such a decision causes costs to the workers.
71 – so in a nutshell, humans can’t make decisions for themselves, only for others (even when equally informed)? Regarding goals, I think we like to *say* we want to do a lot of things, but we tend to figure out how to do the things we really want to do and we don’t do the things we don’t really want to do. (Revealed preferences.) I also think we don’t like to spend time thinking about things we don’t want to if we don’t have to, e.g. if the govt is going to do it for us anyway, but that doesn’t necessarily speak to capability.
72 – 1) teasing out “malicious intent” may be a dead-end here (and while I’d like to understand the German system of campaign finance that may be too off-topic), but how about the ‘adverse consequences’ of just one big issue, say, more restrictive labor laws in Germany creating much higher structural unemployment vs. the US for many decades now?
2) I could ask how humans remained upright for so long before govts started deciding so many things for them (what exactly has become so much more complex, or is this circular?), but how about starting with the specific example at hand – is asking workers whether they’d prefer a pay cut of $X in exchange for say 2 more guarded safety exits something would that overload them? We allowed that some experts could inform them, but the issue remains that someone else making the decision (e.g. by saying “if only one life is saved”) imposes costs on them too, and possibly not in a way they would have preferred.
@73:
Adverse consequences of socialist overregulation are a different issue. In Germany, campaigns are almost entirely financed by public money (in accordance with the number of votes a party gets) and party membership fees. Also, all parties running in national elections get a significant amount of free TV air time.
Also – are you sure Germany has had higher structural unemployment for the last couple of decades IF YOU STRIP OUT THE EFFECTS OF INTEGRATING COMMUNIST EAST GERMANY? Caution – unemployment numbers until recently are not directly comparable with US numbers, IIRC the differing methodology had an effect of about +2 percent (on the German side).
With regard to your other question: “I could ask how humans remained upright for so long before govts started deciding so many things for them”?
That should be pretty obvious. Until the age of industrialization, life was relatively simple compared to today. A donkey cart is easier to understand and evaluate than a car. Plus, the damage you can do riding a donkey cart is lower. There weren’t a million chemicals that could be added to your food. There weren’t any financial products. The vast majority of people never took out any kind of loan (aside perhaps from getting credit at the butcher shop). Prior to industrialization, buildings were much smaller so you could usually get out without requiring dedicated fire exits.
If you only look at the number of legal contracts people enter into in the modern world, that number is magnitudes higher than the number of legal contracts people entered into when they were mostly subsistence farmers.
The modern regulatory state developed in response to this. It arose because technological progress has created a world so complex that no human being could possibly keep pace with the countless new opportunities and dangers that arise each and every day.
It arose because in its absence, the prosperous complex society wouldn’t exist, its development arrested by frictional costs.
@73 With regard to your specific question:
“is asking workers whether they’d prefer a pay cut of $X in exchange for say 2 more guarded safety exits something would that overload them?”
Let’s forget the idea of “overload” and talk about cost. The question is – how expensive is it for them to get informed and make that decision? And how well CAN you inform them?
I suppose it would be ideal if you put it in the job advert: “$ 5 dollars an hour and a 1% chance of burning to death horribly each year.”
While the competing employer would then have a job advert: “$ 4.98 per hour and no chance of burning to death horribly”.
Each worker could then decide which option he prefers. Costs are minimal.
Unfortunately, that’s not what a job advert would look like in a libertarian world. What you would have instead is a job advert that looks like this:
“$ 5.00 per hour, a 1% chance of a fiery death, a 0.3% chance of dying in a building collapse, a 3% chance of being maimed by unsafe machinery, a risk of between 1 and 10% of developing lung cancer 15-20 years in the future due to asbestos contamination, and long-term exposure to varying amounts of novel chemicals which may – or may not – lead to acne, brain damage and sterility. We think they’re relatively safe, but we’re not entirely sure.”
The advert of other employers would look similar, except with different numbers and different hazards, with the list of hazards and chemicals possibly going on for a few pages.
I don’t think this is an efficient solution.
hi prof,
but isnt this an example of prisoners dilemma type games? if one country enacts “some” safeguards, its inefficient, but if all countries enacts safeguards, then it *might* be efficient, or definitely more efficient that the regular, unilateral, equilibrium?
this is a standard/”usual” result right?
Just one point on the knowledge possessed by workers – while individual workers may be ignorant, they will inevitably hire lawyers to represent them if something happens from a safety standpoint. Firms have HR departments and legal departments and set policies in place as if workers were fully-informed legal specialists trying to sue them.
So I think in the long run, we don’t get incremental workplace safety because workers are somehow making a tradeoff between saftey and compensation. As a worker, my personal experience when I have taken a job is to make sure the place looks decently safe and clean relative to other places I have worked, which happens about 5 seconds after walking into the door. I have never (nor seen anyone) attempt to inspect a workplace for hidden dangers when taking a job.
A better explanation would be that over time, what society determines to be “safe and clean” expands, so what constitutes legal “negligence” ratchets up. My guess is that the law follows these definitions. Someone sues an employer in some novel way and the legal definition of negligence expands. Incremental changes in legal definitions is a more plausible explanation than incremental changes in worker’s cost benefit analysis.
So it’s not about a worker saying “x% chance of death is worth $yy salary”, rather it’s about firms saying “I better get rid of that open elevator shaft so the families of my uninformed employees don’t sue me”
GabbyO:
Lots of things *might* happen, which is why it’s worth thinking about what *will* happen. Krigman’s prescription, apparently, is to subdtitute a wild ass guess for thought, and to sacrifice an opportunity to show his readers how to think about such things.
73 “so in a nutshell, humans can’t make decisions for themselves, only for others (even when equally informed)?” Humans can’t make perfect decisions. They can obviously make decisions for themselves, just not necessarily good ones – i.e. ones that would lead to an efficient outcome. In some circumstances, choosing for others may allow a dispassionate appraisal of the information and lead to a better outcome for everybody. We assume this is the case for children as amatter of routine. Do we suddenly become perfect at age 18?
Following thre children ananlogy, we allow greaeter control ofer choices as they mature, and we arbitrarily cut them loose at a particular point. We know full well that they still get it wrong form time to time. 18 year olds probably make some choices that are not in their best intersts, but we generally have to let them make them anyway. Likewise 50 yr olds also make choices that are not in their best interests, and we generally let them make them anyway. There are some aspects where we can predict that most people will make a similar type of mistake, and force them not to choose it through regulation. Where to draw the line is a tricky question. I would guess most people think speed limits are sometimes a good idea. SOme sort of food standards are probably a good idea. Requiring every shop to have a wheelchair ramp is probably not. Requiring every shop to think about disabled access may be a good idea.
Perhaps we can all agree at that at least today a better, workable solution for the factory is alarmed exits. I agree with Ben K that litigation will often get you to a similar place as regulation (I guess we convince ourselves regulation is more proactive). However it seems both will tend to err on the side of over-regulating (vs. what people might have chosen, e.g. I don’t really need to have been willing to pay for something, just to convince a jury I would have). I also agree with Harold & Advo that sometimes people may tend to under-regulate themselves, hence the need to weigh both sides which was the point here. IMO this is not primarily because people are helpless (even in buying a car??), but because *there are some risks they can’t reasonably anticipate*. E.g. on food standards, when I enter a restaurant there’s an implicit contract if you will that my meal will not poison me (and it’s reasonable to assume there’s no relevant discount I would accept to take such risk). People also may tend to over-discount low probability events, but 1) in those cases I find it hard (presumptuous) to go much farther than informed consent, and 2) that’s a harder case to make when people are at subsistence level as in the example here. In general I believe policymaking should be based on the rule and deal with the exceptions, rather than acting as though *most* people need a life coach for most decisions.
Advo – the catch with public campaign funding is that *someone has to decide* who’s worthy of funding in the first place. Those who already have the most votes need a money advantage the least. This seems less conducive to diversity of thought. New voices sometimes require patronage. Plus there’s the whole free speech thing (if I have the right to say something, I should have the right to hire someone to say it for me). And on the complexity of life, I think you’re overestimating quite how simple things used to be, and underestimating the potential for market solutions like advisors and consumer research groups. If you’re arguing we have too many lawyers, I agree :)
@Iceman:
With regard to diversity of thought, we certainly have no shortage of that in the German political system.
The two major parties (SPD and CDU) have shrunk in size over the recent two decades, and new parties have mounted successful challenges in regional and local elections.
E.g:
http://en.wikipedia.org/wiki/Pirate_Party_Germany
Compared with a US-style donor financed system, the German campaign financing system GREATLY favors new parties.
Note that you can of course get additional donor financing in Germany, it’s just that its impact is greatly muted because of the free airtime you already have (additional spending suffers from diminishing returns).
Interesting…but how different are they from each other really? Run the gamut from moderate left to ultra-left?