Following the latest round of drivel from Donald Trump, this might be a good time to review the standard textbook case for free trade. (You’ll also find this spelled out in The Big Questions .)
Suppose American manufacturers sell 1000 widgets a year to American consumers at a price of $9 each. Now, thanks to a new free trade agreement, foreign manufacturers can sell widgets to American consumers at $6 each. Let’s try to account for all the different ways that Americans are affected.
1. American manufacturers have two choices: They can match the foreign price of $6, or they can go do something else. If they match the foreign price, they lose $3 per widget (compared to what they were making before). If instead they go do something else, they lose at most $3 per widget. We know this, because they always have the option of matching the foreign price and therefore won’t choose any option worse than that. Therefore, the loss to American manufacturers is at most $3000. (In fact, under very mild assumptions, which almost always hold, the loss is surely less than $3000, but we won’t need to know that here.)
2. Existing American consumers — the ones who were going to buy those 1000 widgets anyway — pay $6 per widget instead of $9 per widget, and therefore collectively save $3000.
3. Some Americans who were unwilling to buy widgets at $9 will happily buy them at $6, and will be happy with their purchases. This is an additional gain to Americans.
Bottom line: American producers lose at most $3000. Existing American consumers gain $3000. New American consumers gain something too. Therefore the gains to Americans must exceed the losses to Americans.
Bottom line, restated: If you stand in the way of a free trade agreement, you will surely do more harm than good to American citizens.
Note that all of this completely ignores the effects on foreigners. If you care at all about those foreigners, their gains are icing on this cake.
This isn’t rocket science. It is Economics 101. Occasionally politicians pretend they’re too dense to understand it. I suspect Trump might be the rare exception who doesn’t need to pretend.
It would be useful if one mentioned comparative production costs. Those are a consideration for Ec101.
This omits the effects on those employed by the widget manufacturers. I know, understand, and appreciate the argument that they’ll do something else, but this ignores the primary argument people have against free trade and bolsters their claim that the educated class doesn’t give a fig about them, not even enough to consider how such policy affects them.
Re: “This omits the effects on those employed by the widget manufacturers. I know, understand, and appreciate the argument that they’ll do something else”
I think the argument is not that they’ll do something else, for they also have a choice: earn less (thereby helping their employer match the foreign price) or do something else.
It’s probably best to think of “American widget manufacturers” as referring to both the employers and the employees. From that point of view, the story doesn’t omit the effects on the employees. Instead, it omits the negotiation between the employees and employer about whether and how to match the foreign price.
“This omits the effects on those employed by the widget manufacturers.”
That’s accounted for in the cost of manufacturing a widget. That <$3000 loss is either coming out of the capitalist's pocket, or the workers' pockets, or both, but the net loss to both groups is not going to exceed $3000 in total.
>>I think the argument is not that they’ll do something else, for they also have a choice: earn less (thereby helping their employer match the foreign price) or do something else.
This is a more complete way of stating my point – thank you.
>>It’s probably best to think of “American widget manufacturers” as referring to both the employers and the employees. From that point of view, the story doesn’t omit the effects on the employees. Instead, it omits the negotiation between the employees and employer about whether and how to match the foreign price.
A big part of the argument is that the owners and workers are wrapped up together for the purpose of analysis. They do not necessarily benefit (or suffer) in tandem. Treating them as one entity for the purpose of analysis pushes aside the concerns of people who don’t have many options.
thomasblair: Daniel Grayson and Zachary Taylor are exactly right. As for this:
A big part of the argument is that the owners and workers are wrapped up together for the purpose of analysis. They do not necessarily benefit (or suffer) in tandem.
That’s fine. We’re adding up gains and we’re adding up losses. If we can add up all the losses to the owners and the workers in a single step, that saves us a little work.
On the first day of class, when I asked how we could measure the thrust generated by the booster assembly, this is exactly what Landsburg told me, too.
I share Landsburg’s conclusions from this exercise regarding social cost and benefit. But economics (and game theory) tell us that people aren’t generally motivated by social cost and benefit, but by individual cost and benefit. And people who have reviewed how productivity gains have been allocated since 1980 or so might be tempted to conclude that owners and workers seem to live in radically different societies.
You can tell a worker, “Yeah, you’re losing your job. But take satisfaction in the knowledge that society will be richer, and those riches will flow to your former boss. And because my model doesn’t distinguish between benefits that flow to you and benefits that flow to your former boss, that’s obviously not an important distinction.” But don’t be surprised to learn that many workers, lacking your superior education, are just too simple-minded to appreciate these sophisticated arguments.
Free movement of goods, capital, and labor arguably generate optimal social (aggregate) results. But we don’t have free movement of voting power. And within any given jurisdiction, the class of people who perceive themselves benefiting from free trade (including foreigners) may have fewer votes than the class of people who perceive themselves as losing. And workers may perceive themselves as having something akin to a property interest in their jobs. They may not be able to enforce that view in the courts, but they can try to enforce it in the voting booth.
Because the class of people who perceive themselves as losing wields the power to scuttle free trade deals, free trade advocates may need to develop a mechanism to buy them off. Otherwise, you get Brexit–and maybe Trump.
This is an argument that free trade is better than no trade. It does not really apply to the current situation, as Trump is criticizing deals that manage trade in particular ways, and he thinks that the trade could be done better.
The essential problem with adding up the costs and benefits of free trade is that they accrue to different groups of people in the US.
Broadly speaking, for the US, free trade (and immigration) have the consequence of greatly increasing the supply of low- and medium-skilled labor.
This changes the supply/demand profile in the US labor market – the result is that the pricing power of high-skilled labor goes up a lot compared with low-skilled labor.
And so large subgroups (regional and otherwise) within the US population have seen large declines in their real wages over the last 50 years, the middle stagnated and the elite has prospered.
Free trade and immigration have been very bad for tens of millions of Americans, even as (globally) they lifted several hundreds of millions out of grinding poverty.
This needn’t have been so bad for common Americans, if government had acted to re-distribute some of the gains from free trade, in particular in order to help people adjust.
Unfortunately, fiscal and other government policies evolved precisely into the opposite direction over the last 35 years, cutting services and increasing taxes for the middle and working class while cutting taxes for those at the top that benefitted the most from free trade and immigration.
To go further:
It is not at all clear to me that free trade has increased utility within the US.
Sure it has boosted US GDP and average incomes, but it has done so in a very inequitable fashion. Not only have the benefits largely accrued to the very top of the income pyramid, the bottom has actually seen a considerable decline in income.
From an overall utility perspective, an additional $10,000 in income for Bill Gates certainly does not outweigh an income loss of $5,000 for Joe Sixpack.
(globally, however, it is a different issue).
Nobody really has a good point. They can match the foreign price of $6, or they can go do something else. There is nothing in Econ101 that says everyone will gain. If that something else is drawing the dole that individual will not be happy. There will be some constituencies that lose. Unfortunately there will be even greater numbers of people that think or fear they will lose. There will be others who will be convinced they have lost even if they have gained, as other unrelated factors will adversely affect them. There is lot of scapegoating.
Is there an argument that one person losing $3000 has a greater effect than 3000 people gaining $1? Or one region losing all its jobs causes damage out of proportion to the dollars gained across the nation?
This analysis leaves out the fact that if the negative impacts of free trade are concentrated in one place (while the benefits are dispersed across the whole population), then the negative impacts can have a snowball effect, and grow to the point where they end up exceeding the benefits. If you lose your job, you might be able to borrow a bit from friends and neighbors while you look for another one, but not if they all lost their jobs too. Crime can increase, which has costs beyond the value of stolen property. General morale can spiral downward in an area hit by heavy job losses, which could lead to increased alcoholism and drug use. The example of Detroit seems to suggest that something like this does happen in practice.
That’s not an argument for restricting trade, so much as it’s an argument for redistributing the benefits, to prevent the harmful impacts from snowballing in one place. Maybe give the affected workers subsidized loans to pay for re-training in another industry.
Steve, how would you respond to Gene Callahan’s short dialogue about the textbook case for free trade?
http://gene-callahan.blogspot.com/2016/03/a-dialogue-between-economist-and.html
Gene’s point is that the reason we care about whether social benefits exceed social costs is that it allows us to turn a policy into a Pareto improvement. But if a free trade is done without redistribution from the winners to the losers, then you don’t have a Pareto improvement and thus it does not matter whether social benefits exceed social costs. So Gene argues that in practice, free trade agreements are sold under false pretenses.
Advo:
Not only have the benefits largely accrued to the very top of the income pyramid…
This strikes me as ridiculous. The big beneficiaries of free trade are shopping at your local Wal-Mart. Go have a look around there. Do those people look particularly rich to you?
Steve,
I meant the *net* benefits. The relatively low Wal-Mart prices* only partially offset the substantial decline in income large population groups of formerly highly paid manufacturing workers in the US have suffered.
*(Wal-Mart prices are actually very high if you compare them to Aldi, which has a way more efficient business model that will completely upend US grocery retail over the next 10-20 years)
To the people who hold as a reason to be against free trade due to differences in cost of living/wages between countries, are you also against people in, say, New York City or San Francisco trading freely with people in, say, Mississippi? If not, why not? Is it because someone in Mississippi can more easily move to San Francisco than they can to China? But wouldn’t that then be a case for more open borders instead of tariffs? Or at least could you admit that relativity open borders (with simplified immigration law) would be better for the world, even if you were forced to be against it happening for your own sake?
We also have to take into account what those who collectively are saving $3000 from being able to buy widgets cheaper do with those savings. Are they piling them up and setting fire to the dollars? I think not. They are spending them either now, or saving them and spending them later. That increased spending creates income and jobs elsewhere in the economy.
It is easy to point out job losses when the US widget factory closes due to cheaper foreign competition. That is what is seen, but what isn’t seen but still exists is the number of people hired by companies elsewhere in the economy as widget buyers have more to spend. It is hard to point out, say, a yoga instructor, or a carpenter, and say with certainty that they owe their jobs to cheap imports, but if that is where the the widget savings got spent, then they do.
Or put another way, if a protectionist government puts in a $3/widget tariff, that may or may not cause a closed widget factor to open, but there will be yoga instructors and carpenters somewhere in the country who lose their jobs as a result.
Josh@22: it is an excellent argument for open borders instead of tariffs, but it is salient here that we do not have generally open national borders nor is there any plausible scenario in which we will get them in our lifetimes or our grandchildrens’ lifetimes. And even presuming that we had them, there would still be strong practical disincentives — cash cost, family and social ties, language barriers, local resistance at the destination — against workers relocating, as the EU countries are rapidly re-learning.
In the meantime, I would strongly prefer that the polities inside those borders (especially the one I happen to live in) not tip over into totalitarianism (a scenario that our leaders apparently think is unlikely to happen based on a relatively short run of good luck), so some consideration for the human cost of free(er) trade without open borders might be a wise policy. Consider it guillotine insurance.
Our esteemed host once asked me why we should attempt to fight recessions. I believe he intended it as a rhetorical question: natural cycle of business, creative destruction, re-pricing of bad assets and all of that. I think the Trump candidacy sums up my answer to the question: because long recessions provide an opening to politicians who’s success depends on making them permanent.
The people I see shopping at my local Walmart tend to look a lot like Warren Buffett. Hmmm….
I can very well add up the cost of my own unemployment, as well as have a moderately accurate guess at that of my family, friends, etc.
I have no idea how much I save from trade. Gas, diapers, clothing. I imagine a smartphone would be unfathomably expensive.
But an economist certainly has covered those benefits. What about my cars? Both my cars were assembled in Japan, though probably with plenty of parts from America, Europe, and Korea. Technically, it didn’t save me money compared to buying a comparable American version of each, but would those American models be anywhere near as good if not from pressure from Toyota and VW?
I probably don’t save money at all in my ability to buy foreign produce, but it sure is nice to have fruit and vegetables in the winter (I HATE grapefruit).
And, @Aldo, thanks to foreign grocery stores, I have the option of shopping for groceries at Aldi, if I so choose. I don’t, by the way. You’re right that it’s cheaper than Wal-Mart, but Wal-Mart’s quality is barable in a pinch, Aldi makes me angry every time I’ve tried. But my Godmother loves it, and I am glad she has the ability to go there and isn’t limited to grocery stores from Austin or Bentonville.
@Advo
If trade has increased increased U.S. GDP and average income, then it is a good thing for the U.S.
It’s up to the U.S. to decide what to do with that increased wealth. We could be using that money to improve infrastructure which would generate construction jobs which would benefit Joe Sixpack’s brother who has construction skills.
As long as the hit to GDP from doing increased infrastructure spending is less than the gain from trade, we have even more winners.
Lets see if I have this right, after following some of the links.
International free trade results in Kaldor-Hicks improvement. That is, it is possible in principle to redistribute the gains so there is a Pareto gain – nobody is worse off and at leat one person is better off. That unemployed steel worker can in principle get a new job working for (perhaps) the direct beneficiary of shutting down the factory, or somebody else that has more money to spend due to lower prices. He will be at least as well off as before, once you include the price reductions.
In practice we know that this re-distribution will not happen perfectly because of various frictions and market failures. The kinetics are too slow for everybody to get their share of the cake, as at least one steel worker will die before they can get a new job.
So during this adjustment period between a Kaldor-Hicks and a Pareto improvement, we know some people will be worse off.
Does this matter? It seems it does, as pointed out above, the effect of concentration of losses can have an amplification effect that may fix in some of the losses, so a Pareto improvement will not be achieved even at equilibrium. An example may be that steel worker dying because he lost his job. We now have one less productive unit, so the new equilibrium is lower than Pareto improvement.
It is a while since I looked at Pareto improvement, so I will have missed something.
On another tack, LK discusses comparative vs absolute advantage. If capital is mobile but labor is not we no longer have comparitive advantage, he says. It is then not clear that Kaldor-Hicks improvement will result in a particular country, although it will of course over the globe. The prevention of movement is simply another way to prevent the Pareto gain from being realised.
Hi Steve,
I thought I remembered something about this, so I took down my copy of The Armchair Economist. Here’s the part I thought was relevant (Chapter 7, page 64-65):
So, has your opinion on this changed in the >20 years since you wrote that, to the point where you’re now mocking people who raise the same questions you thought were “troubling” then?
@Brett It would be a much shorter timeframe than 20 years, Steve kept the line in his 2012 revised edition:
https://books.google.com/books?id=R3ckZvTAZgQC&lpg=PT94&ots=e3qX5gb3Kw&dq=%22And%20if%20this%20eliminates%20the%20argument%20against%20inefficiency%2C%20what%20argument%20for%20efficiency%20remains%3F%22&pg=PT94#v=onepage&q&f=false
I have yet to see an argument for why, just because free trade plus redistribution leads to a Pareto improvement, we should favor free trade agreements that do not include redistribution.
@Brett That was the point of the dialogue by Gene Callahan I linked to in comment 14 above; free trade agreements are sold by reference to Pareto improvements, but then the redistribution doesn’t actually happen in practice.
@Keshav: Thanks for the link. My hard copy is from 1993, so it’s certainly interesting to hear that the same commentary is in the revised version.
Steve: I like your books. My copy of The Armchair Economist has been sitting on that shelf for most of that 20 years since I bought it in college, and it’s getting pretty battered given the number of times I’ve reread it. One of the things I like best about them is the attitude of respectful curiosity about the world and why people behave the way they do. In your own words, that
It seems every time specific trade restrictions are studied they’re estimated to cost so much per job saved it would clearly be more efficient to just give each worker the money, “proving” these are bad deals.
The next question (that everyone is now asking here) is are we actually compelled to offer compensation? I’m sympathetic to arguments for retraining etc., at least from a ‘political economy’ standpoint. But it also seems fair to ask, if protecting those good ol’ manufacturing jobs by restricting voluntary exchange is essentially a tax on consumers, who has that kind of a “right” to a particular job in the first place?
I have no idea if Trump understands any of this, however politically it is probably quite clever of him to completely “co-opt” this issue.
This isn’t rocket science. It is Economics 101. Occasionally politicians pretend they’re too dense to understand it. I suspect Trump might be the rare exception who doesn’t need to pretend.
Most of us educated even a little in econ people are convinced that Hillary Clinton is just pretending that she does not know this (she talks anti-trade), and so we support her because we are confident that she is lying. I also think Bernie Sanders knows that but many people seem to think that he is sincere, that is hard to believe. Trump having got in to Penn and graduated from Penn could know this and should know that, but we hate him.
All of this is very strange from a logical perspective.
There are many issues like that, another example is a CO2 tax verses all the strange crap that we do like ethanol, CAFE and subsidies. http://www.nytimes.com/2013/02/22/opinion/the-case-for-a-higher-gasoline-tax.html?partner=rss&emc=rss&_r=2&
Very odd.
thomasblair wrote:
“This omits the effects on those employed by the widget manufacturers. I know, understand, and appreciate the argument that they’ll do something else, but this ignores the primary argument people have against free trade and bolsters their claim that the educated class doesn’t give a fig about them, not even enough to consider how such policy affects them.”
Considering the huge welfare state (tax funded schools, SNAP, SS, Medicare, TANF, EITC) it is hard to argue that the educated class doesn’t give a fig about them.
@nobody.really
One of the interesting things to think about is that not so many people ever had high paying manufacturing jobs or even wanted to have a high paying manufacturing job. Even today manufacturing jobs are not so easy to fill. It is more about voters sympathetic to those with manufacturing jobs. I think it is that we hate to see losses even to others. (Loss aversion)
On a related issue why is it worse to deport someone who has had a chance to work at 1st world wages than to prevent someone who has never had that chance from coming here? http://un-thought.blogspot.com/2016/01/immigration.html I say that is odd logically except that we do not want to have to sad loss stories about deportees.
@Keshav #25
The U.S. currently has redistribution as part of it functioning. Unemployment insurance, social security, medicare, food stamps, pell grants, Earned Income Credits, Section 8 housing vouchers, to name just a few all redistribute wealth from winners to losers of any agreement/action anyone enters into/takes.
Why would we need to have redistribution as part of trade agreements?
In fact what you propose is worse than just having a trade agreement. If the U.S. ever wanted to change the way it redistributed trade wealth/income, it would need to renegotiate the agreements with other countries.
It seems what you want is kind of what we have now. A system of redistribution which the U.S. maintains and can change without needing agreement from any other country.
Trade agreements specific to the countries involved and related to trade only.
If Trade Agreements aren’t profitable for the U.S., then renegotiate the Trade Agreement or start a war or something.
If Americans want to change their redistribution system, then change that.
Floccina: “Trump having got in to Penn and graduated from Penn could know this and should know that, but we hate him.”
Donald Trump and I both obtained degrees from Wharton. As I understand it, his 1960s B.S. program was highly concentrated in real estate. My 1980s Finance MBA program was likely much broader. It’s possible that Trump learned little about international trade of goods. I don’t think international trade was that important for business schools in the 1960s, and probably even less so for real estate programs.
@Will A (#32) “The U.S. currently has redistribution as part of it functioning. Unemployment insurance, social security, medicare, food stamps, pell grants, Earned Income Credits, Section 8 housing vouchers, to name just a few all redistribute wealth from winners to losers of any agreement/action anyone enters into/takes. Why would we need to have redistribution as part of trade agreements?” Well, but that redistribution you already in place. In deciding to approve a new free trade agreement, the question before us is to evaluate what effects it will have relative to existing policy. The textbook case for free trade is that a free trade agreement will lead to a Pareto improvement relative to existing policy, if it is combined with a redistribution relative to existing policy, i.e. on top of whatever redistribution is already going on.
“In fact what you propose is worse than just having a trade agreement. If the U.S. ever wanted to change the way it redistributed trade wealth/income, it would need to renegotiate the agreements with other countries.” But I’m not proposing that the redistribution be specified in the trade agreement itself. I’m just suggesting that alongside approving the trade agreement, Congress also passes a redistribution bill designed to turn the effects of the free trade agreement into a Pareto improvement. Otherwise, passing the trade agreement is arguably not justified, at least by the textbook case for free trade.
@nobody.really#7. Thanks for giving me the first laugh of the day! As for #20, do I understand your statement correctly by emphasizing *looking*like*?
@Bennett Haselton#13. @Keshav Srinivasan#14". @Harold#23.
All of you seem to be making the point that trade liberalization has a distributional impact, beyond merely the positive impact on total income (which our gracious host’s argument demonstrated with characteristically simple and irrefutable logic and which I do no read you as denying).
And you are right. Any economic innovation is likely to have distributional effects which are not captured by the total economic impact. From this you conclude that, total positive be whatever it may, we should not take this free money unless we also at the same time impose additional redistributive policies that mitigate the impact on all the losers.
The first response to that is that redistribution has a negative effect on efficiency at the both the source and recipient side. For example, redistribution by increasing high-end tax rates and creating new means-tested benefits have dead-weight losses at both ends. In fact, the nature of the dead-weight loss is the same for high marginal taxes and means-tested benefits which are, in effect, just lump-sum payments (which have no incentive effect) combined with often extremely-high marginal tax rates (which have a very harmful incentive effect). It is quite possible that the dead-weight losses of more redistribution would consume all of the free lunches that trade provides and more. In that case, would you favor the trade liberalization simpliciter?
The second, deeper response is that all of you are normatively privileging the status quo ante as an obviously just distribution of income which we need to preserve. That is a curiously conservative position to take. If another economic change led to net increases in income, but tended to redistribute income from the rich to the poor, would you demand that it barred unless accompanied with taxes on the lower orders to compensate the losers among the upper classes? If not, why not?
Finally, why do these distributional concerns override any consideration of a general improvement in efficiency and income only in the case of trade? Why not for technological innovation?
Imagine that instead of a trade agreement allowing a $6/widget producer to come to market, I invented a new machine that allows me to profitably make and sell widgets at $6 each. The impact of allowing me to operate this new machine on the current widget makers (companies and workers, collectively) is *exactly* the same as allowing imports of $6 widgets.
Yet, would any of you argue that I should be prohibited from deploying my new machine *only*if* there is in place a system to tax all those widget consumers benefiting from the lower price (or me?) to pay compensation to all the current widget manufacturers I am hurting?
I doubt any of the protectionists would embrace that argument because it so obviously would have prevented or greatly delayed all the economic disruptions over recent centuries which collectively have made the West so rich. But logically there is no difference between dislocating economic innovations, be they trade liberalization or invention. The reason that only barring the former has a large public constituency is that it can somehow be linked to the nefarious machinations of nasty foreigners. (n.b. that is not a motivation I attribute to any of the above cited commenters.)
@ Keshav #34
Can you give me an example of what such a parallel redistribution bill might look like?
Does Congress guess at the future impact? Does the bill say,
”
If you are laid off and workers in China in your field make less than you, you will received the difference between the average wage in China and the wage from when you retired in perpetuity.
The funds for paying the above will come from 1.5% tax on profits from companies buying good manufactured in China.
”
And yes, I know this is basically a tariff which basically means it really isn’t free trade. But perhaps you can give me an example that actually works.
Sub Specie: As is so often the case, your comment is pretty much perfect throughout. But this paragraph is especially worth repeating:
Sub Specie Æternitatis@35:
Uh … no, the transfer payments are designed to incentivize people who fear being laid off to support free trade. It’s a pretty straightforward quid pro quo.
And yes, taxation/redistribution alters (“distorts”) people’s behavior relative to a fictional world in which we might get similar outcomes without taxation/redistribution: we call that distortion “dead weight losses.” In contrast, the taxation/redistribution “distorts” the behavior of people who benefit from the transfers. Those changes result in “deadweight gains.”
For example, in any recession, people lose their incomes temporarily. Society incurs transition costs as people get disconnected from utility services, get evicted, have cars repossessed, etc.—only to get these things restored when the economy recovers. Econ 101 tells students about the deadweight cost that arises from the tax burdens to finance the social safety net. They never hear about the deadweight gains when people can use the social safety net to avoid those transition costs. Go figure.
In other words, it’s quite possible that the cost to convert a Kaldor–Hicks efficient outcome to a Pareto outcome would exceed the value of moving to the Pareto outcome. And if it turns out that the cost of fixing the harm cause by free trade is greater than the benefits of free trade, then this would seem to be a strong argument for opposing free trade.
People own their votes. People exercise their own discretion to promote their own self-interest. They get to vote for or against Trump/Brexit/whathaveyou. And they will likely look upon those votes from an incremental, rather than systematic, perspective: What specific change should I expect relative to the status quo if one side wins relative to if the other side wins? (Ok, this is a stylized argument, but good enough for our current discussion.)
As I say, people seem to feel a proprietary interest in their jobs. They expect to keep their jobs and the resulting income, and feel great loss when those expectations are thwarted. From the vantage point of “creative destruction,” this may seem like an unjustified and curiously conservative perspective on their part—but I sense it’s a rather pervasive one.
Indeed, why not?
Historically people owed services—especially military services—to their liege. Providing these services whenever the liege demanded could be quite disruptive, and frankly, their military prowess wasn’t always up to snuff anyway. So eventually this arrangement evolved so that people paid their liege money in lieu of services. That is, they bought him off.
I expect that Uber drivers are on average less affluent than these taxi drivers who, at least in New York, hold medallions worth hundreds of thousands of dollars. And in various cities, taxi drivers have launched efforts to keep out the Uber drivers. If Big Taxi can wield sufficient clout, it may behoove Uber to buy them off—in effect, causing the agent of the poor to pay off the rich.
Throughout the nation, people are installing solar panels on their roofs, or are investing in solar gardens/wind farms, displacing some of the sales that would otherwise be made by electric utilities. These utilities are pushing back, arguing that these customers are no longer bearing their fair share of the system’s costs, and demanding that these customers pay up. If utilities can implement these new “make-whole” fees, customers then face the choice of whether to pay off the big utility for the privilege of entering into the electricity generating business, or not.
I wonder if we’re confusing discussions of what *is* for what *ought.* The point is not whether anyone likes these dynamics. The point is that these dynamics exist wherever one social faction has the means to push back against another. We’re discussing might, not right.
Arguably, if we believe that free trade generates optimal social outcomes but harms individuals, we might refrain from burdening the trade directly, and rather burden society (via progressive taxation funding a sufficiently generous social safety net). Society gets the benefits of relatively free trade–trade burdened the the deadweight losses of taxation, offset by the deadweight gains arising from the wealth transfers.
nobody.really@38:
C’mon, admit that when I said that the high marginal tax rates associated with means-tested benefits have a negative incentive effect, you knew I was talking about the incentives in the actual economy (e.g., whether, how, and how much to work). In return I will admit that welfare payments may serve as effective incentives for certain voting behavior.
Given that there is only one real world, any comparison about what the state of affairs would be in different worlds, i.e., any meaningful statement about policy, requires at least one of the compared worlds to be hypothetical. Calling it “fictional” seems rather dismissive of any attempt at analysis.
You are adopting an extremely conservative position here: any social change–no matter how beneficial on net–must deemed overall “harmful” if it is not accompanied by full compensation to anybody who sees their future expectation of benefit reduced as a consequence.
Is this a principle you are willing to defend in principle?
For example, wide-spread social changes over the last fifty to hundred years have opened up large parts of US society to women, African-Americans, and others. Most economists would argue that the elimination of de jure discrimination against these groups was a positive and liberalizing effect because it allowed, e.g., many positions to be filled by more able women or African Americans. This better matching of talent to positon has undoubtedly increased social welfare and therefore provided a net benefit to Americans collectively. In short, it was a Kaldor-Hicks improvement.
But it was not a Pareto improvement. Nobody ever talks about the losers of this social change. Think of all the marginally competent white male lawyers who were displaced by more able female ones, once they were legally and practically permitted to enter the profession. Or the marginal white foreman of a work crew who–thanks to the civil rights revolution–finds himself displaced by one of his smarter black crew and is demoted to the crew himself. All of these people suffered real harm due the aforementioned social changes. And that is not even counting the strong preferences of the many bigots and chauvinists whose self-esteem was deeply hurt by blacks and women not keeping in their place any more, even if they suffered no direct economic harm.
So I take it to be nobody.really’s position that the legal changes won by the civil rights and feminist movements were harmful because they did not include redistributive taxes on blacks and women combined with special government benefits available only to white men. And if such a redistributive policy was practically impossible, that just means that equal rights for African Americans and women just are a bad idea.
Or perhaps not? Do you have a list of individuals or groups whose harms economists should disregard when doing welfare analysis properly? Could you share it?
Pervasive, no doubt. But then so are many demonstrable fallacies. Once error is amply demonstrated, why should a philosopher or any seeker of the truth give it a moment’s consideration just because it is widely held?
That is a fair description of what happened. But do you endorse the normative position that any net-benefitial social change should be opposed unless it fully compensates the losers?
Medallions are expensive, but I believe they are generally not owned by the actual drivers, but by their bosses or even investment funds. The intersection between (a) people who have hundreds of thousands or a million of dollars at hand (or who could even reasonably raise that sum) and (b) people who want to spend their days driving around strangers is vanishingly small.
Maybe. And, if such a deal was struck, perhaps the net benefits would be worth it. But if it couldn’t be, would you oppose Uber?
Wholesale and retail electric markets actually being my area of professional expertise, I feel tempted to write a few more pages to explain the actual intricacies, but anybody who hasn’t stopped reading this comment already, likely will do so at the in-depth discussion of retail feed-in tariff policy.
Actually, the original post was purely about what is right from a welfare economics perspective. Despite the title, it was not about politics or who is likely to prevail in the presidential race. The latter is an interesting issue too. But, as you rightly note, it is extremely important to keep the two issues separately.
@Sub Specie Æternitatis (#35) “All of you seem to be making the point that trade liberalization has a distributional impact, beyond merely the positive impact on total income (which our gracious host’s argument demonstrated with characteristically simple and irrefutable logic and which I do no read you as denying).” I’m saying more than that: I’m that the only reason why “total income” is even a number that anyone would care about is that because of its implications for Pareto improvements. That is to say, the reason why we would care that the social benefit of policy exceeds its social costs is that that it implies that the policy is a Kaldor-Hicks improvement, and any Kaldor-Hicks improvement can be turned into a Pareto by enacting an associated redistribution policy. So by not enacting said redistribution policy, the supporters of free trade are selling trade deals under false pretenses if they are relying on the textbook case for free trade.
“The first response to that is that redistribution has a negative effect on efficiency at the both the source and recipient side. For example, redistribution by increasing high-end tax rates and creating new means-tested benefits have dead-weight losses at both ends. In fact, the nature of the dead-weight loss is the same for high marginal taxes and means-tested benefits which are, in effect, just lump-sum payments (which have no incentive effect) combined with often extremely-high marginal tax rates (which have a very harmful incentive effect). It is quite possible that the dead-weight losses of more redistribution would consume all of the free lunches that trade provides and more. In that case, would you favor the trade liberalization simpliciter?” No, I wouldn’t, at least on the basis of the textbook case for free trade. The whole reason we would care about something being a Kaldor-Hicks improvement is that it can turned into a Pareto improvement. But if no possible redistribution can turn it into a Pareto improvement, then the fact that something is a Kaldor-Hicks improvement does not really matter. Now of course, you could still try to argue for free trade in some other way, for instance on libertarian grounds. I’m just saying that the textbook case for free trade does not make sense without redistribution.
“The second, deeper response is that all of you are normatively privileging the status quo ante as an obviously just distribution of income which we need to preserve. That is a curiously conservative position to take. If another economic change led to net increases in income, but tended to redistribute income from the rich to the poor, would you demand that it barred unless accompanied with taxes on the lower orders to compensate the losers among the upper classes? If not, why not?” No, because my argument is not that all policies which have distributional impacts must be opposed. It is rather that “net increases in income” as you call it, i.e. Kaldor-Hicks improvements, do not suffice on their own to justify a policy. Which is not to say that all policies which happen to be Kaldor-Hicks improvements but not Paretto improvements are bad; they may well be good for reasons entirely unconnected to the Kaldor-Hicks criterion. For instance a lot of people might favor the hypothetical policy you describe specifically because it a redistributes from the rich to the poor.
“Yet, would any of you argue that I should be prohibited from deploying my new machine *only*if* there is in place a system to tax all those widget consumers benefiting from the lower price (or me?) to pay compensation to all the current widget manufacturers I am hurting? I doubt any of the protectionists would embrace that argument because it so obviously would have prevented or greatly delayed all the economic disruptions over recent centuries which collectively have made the West so rich. But logically there is no difference between dislocating economic innovations, be they trade liberalization or invention.” Well, two things. First of all, I am not saying that there’s anything inherently wrong with a policy that has a distributional impact. You can certainly make a libertarian case for why you should deploy your machine. But if you were justifying your deployment of the machine by comparing social benefits and social losses, I think it would be fair to ask that there be a redistribution policy enacted which compensates the current widget manufacturers.
By the way, on a side note I’m not really a protectionist; I don’t have firm opinions on trade or other economic issues. But when I do think about economic policies, I often find it helpful to use as a benchmark the left-libertarian “Grab What You Can World” described in many posts of Matt Bruenig, such as this one:
http://www.demos.org/blog/6/3/14/lesson-grab-what-you-can
@Keshav Srinivasan#40
Thank you for your thoughtful response. I think I understand your point of view better now.
I do understand that you do not consider the fact that a policy is a Kaldor-Hicks improvement a decisive argument for the policy. I might agree and could certainly imagine certain extreme hypothetical examples of Kaldor-Hicks improvements that I would have to struggle to endorse.
But surely you must concede that a sufficiently large and certain Kaldor-Hicks improvement is very strong evidence in support of a policy? Or would you hesitate to endorse a policy change that increased the real income of everybody on earth by a $1m/year but reduced that of an unfortunate, random, unidentifiable, but wealthy person by $1/year?
On the whole Kaldor-Hicks v. Pareto standard for permitting economic changes, I strongly suspect (thought this is of course an empirical question) that turning each individual instance of the former into the latter would involve such large transaction costs (identifying winners and losers, political negotiations on how to split the surplus, incentive effects typically associated with such arrangements) that a great many free lunches would remain uneaten to such an extent that (virtually) everybody would be made worse off in the long run.
In other words, switching from a meta-policy of only permitting Pareto improvements to one of permitting virtually all Kaldor-Hicks improvements is *itself* a (near-)Pareto improvement.
“But surely you must concede that a sufficiently large and certain Kaldor-Hicks improvement is very strong evidence in support of a policy? Or would you hesitate to endorse a policy change that increased the real income of everybody on earth by a $1m/year but reduced that of an unfortunate, random, unidentifiable, but wealthy person by $1/year?” The policy you describe obvuous sounds like a great policy. I don’t really have a problem with arguments that a policy should be adopted because it has an extremely large number of winners and a tiny number of losers, and moreover that the losses the losers suffer are minuscule. I just don’t think that adding up the dollar amounts of the wins and losses and comparIng them is a good way to think about things, unless you are planning to use that information to do a redistribution.
My comment #42 above was addressed to @Sub Specie Æternitatis
Sub Specie Æternitatis @39:
Well, YES, I knew what you meant. But I wasn’t just being persnickety; I’m making a point: Citizens have an OWNERSHIP RIGHT in their votes, and so it should not shock us that people who want those votes should have to pay for them in one form or another.
If proponents of globalism want the support of the working classes, they need to make a persuasive case that the working classes have more to gain than to lose from protectionism. And typically, this is done by cutting them in for a slice of the benefits. And if that doesn’t work, cut them in for an ever BIGGER slice, until the cost of buying their cooperation exceeds the benefit.
What’s the difference between paying for services and paying extortion? Social convention, mostly.
You may be right about there being only one real world. But according to Landsburg, that’s the world of mathematics.
Anyway, I again fear you miss my point: The models demonstrating the deadweight losses of taxation show the results of an efficient market, and then compare it to the results of a market influenced by taxation—that is, the need to accrue funds to pay for government services. The moral is that taxation leads to suboptimal results compared to a world without taxation.
Yet when we go to worlds that have no taxation—say, parts of Syria, Afghanistan, or pretty much anywhere where the government as effectively collapsed—do we observe especially a booming economy? No. We observe places that are economic shitholes because they lack government services: law and order, property and contract rights, police and courts, roads, uniform weights and measures, a stable currency, antitrust laws, etc. You could scarcely design a more burdened market than a market operating without government services.
Consequently, I argue that a graph comparing a market burdened by taxation to a market that enjoys all the efficiencies that come from government services but none of the costs is a graph that compares the real world to a fiction. Because it compares things on an unequal basis, treating as exogenous factors that should be endogenous, it leads to some distorted conclusions.
(And sure, even distorted models can serve pedagogical purposes. But we need to bear the distortions in mind.)
Ok, you raise quite valid arguments. So I’m gonna walk this one back a bit. “[F]ixing the harm caused by free trade [or other policy]” may set the bar too high.
Let me propose this test instead: 1) Government should maintain a social safety that [insert appropriate criteria here]. And 2) government should refrain from adopting a policy if that policy would cause more people to have to rely on the social safety net, and if the added costs of this reliance to government of this reliance would exceed the added benefits to government of the policy.
Under this test, government could adopt policies willy-nilly that harmed people, just so long as it did not reduce those people to relying on the social safety net. And government could adopt policies that reduced some people to relying on the social safety net, provided that government also derived sufficient benefit (say, via added taxes or gifts) to offset the added costs.
Indeed, nobody.really ever talks about this. I’d guess I talk about it every week. This dynamic formed the basis of Nixon’s Southern Strategy that switched the Solid South from Democratic to Republican control. And there’s been much speculation that this dynamic has fueled Donald Trump’s rise. It’s a hot topic of conversation around here.
That said, the question remains: Did civil rights actually make white guys worse off in an absolute sense? Or did white guys merely feel a loss of relative advantage (status) because, while everyone’s living standards were rising, the living standards of minorities and women rose faster than the standards of men?
Landburg suggests that the displaced worker benefits from lower-cost consumer goods, and that would suffice to offset the worker’s loss. That is, the benefits that flow to a person’s status as a consumer swamp the harm that flowed to a person’s status as a producer. And who knows–maybe he’s right.
But the analogous argument also applies to civil rights: The marginal white male worker may have lost out to better qualified black and/or female workers—but that white worker could now get goods and services from those better-qualified workers. After all, the primary benefit of civil rights is not that black women can benefit from becoming surgeons. Rather, the primary benefit is that society now picks its surgeons on the basis of criteria other than race and gender, and we as consumers all get the benefit of this better sorting of the talent pool.
Maybe. Still, that greeter at Walmart looks an awful lot like Warren Buffett….
I’m conflicted about this. I don’t know what specific legal rights a medallion allegedly bestows. But if a medallion reflects a contract between government and a private citizen to shield the citizen from competition from anyone other than another medallion-holder, and now government is reneging merely because a would-be competitor is popular, yeah, I’d have problems with that. That would look like taking of property without due process of law.
And even without medallions, I don’t understand the merit of having one set of policies for taxis and a separate set of rules (or neglecting to enforce the existing rules) for Uber. Shooting from the hip, states should either eliminate taxi regulations, or should enforce those regulations against taxies AND Uber uniformly.
[A]nybody who hasn’t stopped reading this comment already, likely will do so at the in-depth discussion of retail feed-in tariff policy.
(…weenie….)
Yeah, I don’t actually embrace the argument that utilities are bad actors when they seek to adopt new policies designed to help offset certain fixed costs of providing back-up service. But I know others do—and it fits the pattern of relatively poor people (consumers) compensating the relatively rich (utilities) to buy their cooperation.
Sub Specie Æternitatis @ 39, Part 1:
Well, yes, I knew what you meant.
But I wasn’t just being persnickety; I’m making a point: Citizens act as if they have an OWNERSHIP RIGHT in their votes, so people who want the benefits of those votes should expect to have to compensate the owners in some fashion.
If proponents of globalism want the support of the working classes, they need to make a persuasive case that the working classes have more to gain than to lose from protectionism. And typically, this is done by cutting them in for a slice of the benefits. And if that doesn’t work, cut them in for an ever BIGGER slice, until the cost of buying their cooperation exceeds the benefit.
What’s the difference between paying for services and paying extortion? Social convention, mostly.
You may be right about there being only one real world. But according to Landsburg, that’s the world of mathematics.
Anyway, I again fear you miss my point: The models demonstrating the deadweight losses of taxation show the results of an efficient market, and then compare it to the results of a market influenced by taxation—that is, the need to accrue funds to pay for government services. The moral is that taxation leads to suboptimal results compared to a world without taxation.
Yet when we go to worlds that have no taxation—say, parts of Syria, Afghanistan, or pretty much anywhere where the government as effectively collapsed—do we observe especially a booming economy? No. We observe places that are economic shitholes because they lack government services: law and order, property and contract rights, police and courts, roads, uniform weights and measures, a stable currency, antitrust laws, etc. You could scarcely design a more burdened market than a market operating without government services.
Consequently, I argue that a graph comparing a market burdened by taxation to a market that enjoys all the efficiencies that come from government services but none of the costs is a graph that compares the real world to a fiction. Because it compares things on an unequal basis, treating as exogenous factors that should be endogenous, it leads to some distorted conclusions.
(And sure, even distorted models can serve pedagogical purposes. But we need to bear the distortions in mind.)
Maybe. Still, that greeter at Walmart looks an awful lot like Warren Buffett….
I’m conflicted about this. I don’t know what specific legal rights a medallion allegedly bestows. But if a medallion reflects a contract between government and a private citizen to shield the citizen from competition from anyone other than another medallion-holder, and now government is reneging merely because a would-be competitor is popular, yeah, I’d have problems with that. That would look like taking of property without due process of law.
And even without medallions, I don’t understand the merit of having one set of policies for taxis and a separate set of rules (or neglecting to enforce the existing rules) for Uber. Shooting from the hip, states should either eliminate taxi regulations, or should enforce those regulations against taxies AND Uber uniformly.
(…weenie….)
Yeah, I don’t actually embrace the argument that utilities are bad actors when they seek to adopt new policies designed to help offset certain fixed costs of providing back-up service. But I know others do—and it fits the pattern of relatively poor people (consumers) compensating the relatively rich (utilities) to buy their cooperation.
Sub Specie Æternitatis @ 39, Part Deux:
Ok, Uncle; you (and Landsburg) raise valid arguments. “[F]ixing the harm caused by free trade [or other policy]” may set the bar too high. So I’m gonna walk this one back a bit.
Let me propose this test instead: 1) Government should maintain a social safety net that maintains a minimum living standard of [insert appropriate criteria here]. And 2) government should refrain from adopting a policy if that policy would cause more people to have to rely on the social safety net, and if the added costs of this reliance to government of this reliance would exceed the added benefits to government of the policy.
Under this test, government could adopt policies willy-nilly that harmed people, just so long as the harm was not sufficient to cause those people to have to rely on the social safety net. And government could also adopt policies that DID reduced some people to relying on the social safety net, provided that government also derived sufficient benefit (say, via added taxes or gifts) to offset the added costs. How’s that?
So if a trade deal would put some people on the dole, it had better generate enough tax revenues to compensate the public purse. Or, if not, someone else (presumably a beneficiary) had better be willing to write a check to cover those costs.
In other words, this system does not grant any individual a property interest in his or her economic status (other than the status of living at the minimal acceptable level provided by the social safety net.) But it would recognize government’s interest in adopting policies that would, in effect, transfer public funds into private pockets.
Indeed, nobody.really ever talks about this. I’d guess I talk about it every week. This dynamic formed the basis of Nixon’s Southern Strategy that switched the Solid South from Democratic to Republican control. And there’s been much speculation that this dynamic has fueled Donald Trump’s rise. It’s a hot topic of conversation around here.
That said, the question remains: Did civil rights actually make white guys worse off in an absolute sense? Or did white guys merely feel a loss of relative advantage (status) because, while everyone’s living standards were rising, the living standards of minorities and women rose faster than the standards of men?
Landburg suggests that the displaced worker benefits from lower-cost consumer goods, and that would suffice to offset the worker’s loss. That is, the benefits that flow to a person’s status as a consumer swamp the harm that flowed to a person’s status as a producer. And who knows–maybe he’s right.
But the analogous argument also applies to civil rights: The marginal white male worker may have lost out to better qualified black and/or female workers—but that white worker could now get goods and services from those better-qualified workers. After all, the primary benefit of civil rights is not that black women can benefit from becoming surgeons. Rather, the primary benefit is that society now picks its surgeons on the basis of criteria other than race and gender, and we as consumers all get the benefit of this better sorting of the talent pool.
@Keshav Srinivasan#42:
I understand that you find Kaldor-Hicks efficiency to be an imperfect guide (for sufficiently small values of , I might agree), but I do not understand with what more-perfect criterion you would replace it. The advantages of Kaldor-Hicks effiency criterion are that (a) it can readily be determined (at least directionally), as the original post demonstrated, and (b) many not-perfectly-correlated (an empircal question) Kaldor-Hick improvements will add up to near-Pareto improvements.
PS: As I learned from nobody.really, we can use the blockquote tag in our comments which makes responses much more readable.
@nobody.really#44, #45:
I think your point is better understood as citizens feeling that they have a property right in current policies and vote accordingly. That is probably true. So it is probably also true that in some cases this means that as long as some people vote against policies which hurt their future prospects, while an often larger number whose future prospects benefit (and collectively more greatly), it may be a political necessity to pay off these groups in order to get the benefits of policy changes.
In so far as you take that to be a purely descriptive point, I don’t disagree. But if you elevate this to some sort of a normative principle, I have to point to the hypocrisy which rejects this normative principle in almost all cases other than trade.
Yeah, Tegmark’s book (recommended here http://www.thebigquestions.com/2014/04/01/many-many-worlds/), is a pretty good read, both accessible to the intelligent layman and of interest to people with some professional experience in theoretical physics. The only sour note was the section at the end where Tegmark called for all good scientists to join together against the anti-scientific forces of Big Business and Christians, joined at the hip.
Personally, I find that debating my atheism with a friend, a C.S. Lewis Foundation trained Apologist, tends to be a great deal more rational than the response set of one of Tegmark’s colleagues to an hypothesis she disagreed with: feint, vomit, or run out of the room. http://archive.boston.com/news/education/higher/articles/2005/01/17/summers_remarks_on_women_draw_fire/?page=full
This is a deep rabbit hole to go down in because the line between what is a property or contractual right and what is merely a governmental policy is a little bit fuzzy. (Remember Goldberg v. Kelly and the New Property?).
But clearly such a line must be drawn. Or would you demand compensation for abolition of the marital rape privilege to be paid to rapy husbands (and presumably paid for a special tax on wives?). Or, if you buy the standard story line about recent financial regulation that greedy bankers were exploiting the populus until they were stopped by the virtuous new Dodd-Frank law, does that mean that all those greedy bankers must be compensated for the loss of the prospect of future exploitation?
And if such a line must be drawn, it seems almost as clear—as one says when one doesn’t want to tax one’s readers patience even more—that taxi medallions and protectionist policies fall on the non-property side of the line.
I agree. But I suspect that much of the support for these alleged health-and-safety regulation would evaporate once they no longer serve as profitable barriers to entry. One cannot help but be amused by progressive city councils across the nation simultaneous embracing the principles that companies, on pain of criminal penalty, (1) must perform background checks on potential contractors (as in Uber) and (2) must not even ask about criminal background of potential employees (as in the Ban-the-Box-ers).
Interesting. Let me think about it and perhaps respond some day in a thread somebody is still reading.
So you were self-taught?
Cute. But as far as I know, Ban-the-Box does not bar employers from asking about the criminal backgrounds of employees; rather, it delays that inquiry until after an employer conducts an initial interview.
Keshav
[quote]The policy you describe obvuous sounds like a great policy. I don’t really have a problem with arguments that a policy should be adopted because it has an extremely large number of winners and a tiny number of losers, and moreover that the losses the losers suffer are minuscule. I just don’t think that adding up the dollar amounts of the wins and losses and comparIng them is a good way to think about things, unless you are planning to use that information to do a redistribution.
[/quote]
First, can we agree that it does at least make it sensible to compare the outcomes and argue that one is objectively better? Many people deny you can make interpersonal comparisons at all. Maybe in some cases you can’t but KH isolates examples where you can, since all the participants have cexpressed their relative preferences in a fungible measure, money.
So while conversion of wins and losses to dollars may not always make sense, it surely does in these situations. You agree to that?
So let’s look beyond one KH improvement to a whole range of them. Now while some lose in one case they win in almost every other. The number of losers is zero in the limit, right?
So now we seek a comprehensible, defensible and not particularly subjective or arbitrary rule for assessing changes by weighing gains and losses. Here we have one which fits the bill, and which in the long run has produces Pareto improvement.
@SSA 35
I regularly argue with people who *would* hobble innovation, and use your argument in support of Luddism, as a consequence of the iniquity of free trade.
@Roger 9
Isn’t it an argument that increased trade is good? The status quo ante needn’t be “no trade” for this to work. In which case doesn’t the argument apply in the cases Trump talks about?